The Origins of Inorganic Acidity in Fogs Citation Jacob, Daniel James (1985) The Origins of Inorganic Acidity in Fogs. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/54GV-ZQ97. https://resolver.caltech.edu/CaltechETD:etd-01112005-133508 Abstract A rotating arm device to collect large samples of fogwater for chemical analysis was designed and fully characterized. This instrument was used to determine the chemical composition of fogwater at a large number of sites in California. Fogwater at both urban and non-urban sites was frequently found to contain important inorganic acidity. Chemical models were developed to interpret the origins of this acidity in terms of aqueous-phase S(IV) oxidation reactions and acid/base titration processes in the fog and the precursor atmosphere. These models were found to successfully interpret the inorganic acidity in fogwater and aerosol during wintertime stagnation episodes in the San Joaquin Valley of California. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: Environmental Engineering Science Degree Grantor: California Institute of Technology Division: Engineering and Applied Science Major Option: Environmental Science and Engineering Thesis Availability: Public (worldwide access) Research Advisor(s): Hoffmann, Michael R. Thesis Committee: Hoffmann, Michael R. (chair) Flagan, Richard C. Morgan, James J. Seinfeld, John H. Shair, Fredrick H. Defense Date: 15 January 1985 Funders: Funding Agency Grant Number California Air Resources Board UNSPECIFIED Record Number: CaltechETD:etd-01112005-133508 Persistent URL: https://resolver.caltech.edu/CaltechETD:etd-01112005-133508 DOI: 10.7907/54GV-ZQ97 Related URLs: URL URL Type Description https://doi.org/10.1021/es00129a005 DOI Article adapted for Chapter I. https://doi.org/10.1063/1.1137995 DOI Article adapted for Chapter II. https://doi.org/10.1021/es00138a013 DOI Article adapted for Chapter III. https://doi.org/10.1029/jc088ic11p06611 DOI Article adapted for Chapter V. https://doi.org/10.3402/tellusb.v36i4.14909 DOI Article adapted for Chapter VI. https://doi.org/10.1029/jd091id01p01073 DOI Article adapted for Chapter VII. https://doi.org/10.1029/jd091id01p01089 DOI Article adapted for Chapter VIII. ORCID: Author ORCID Jacob, Daniel James 0000-0002-6373-3100 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 121 Collection: CaltechTHESIS Deposited By: Imported from ETD-db Deposited On: 14 Jan 2005 Last Modified: 21 Dec 2019 02:48 Thesis Files Preview PDF (Jacob_dj_1985.pdf) - Final Version See Usage Policy. 17MB Repository Staff Only: item control page

THE ORIGINS OF INORGANIC ACIDITY IN FOGS Thesis by Daniel James Jacob In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 1985 (submitted 15 January, 1985) ii TABLE OF CONTENTS ACKNOWLEDGEMENTS ABSTRACT LIST OF TABLES LIST OF FIGURES OVERVIEW CHAPTER I - FOGWATER COLLECTOR DESIGN AND CHARACTERIZATION CHAPTER II - AN INSTRUMENT TO COLLECT FOGWATER FOR CHEMICAL ANALYSIS CHAPTER III - CHEMICAL COMPOSITION OF FOGWATER COLLECTED ALONG THE CALIFORNIA COAST CHAPTER IV ~ KINETICS AND MECHANISMS OF THE CATALYTIC OXIDATION OF DISSOLVED SULFUR DIOXIDE IN AQUEOUS SOLUTION: AN APPLICATION TO NIGHTTIME FOGWATER CHEMISTRY CHAPTER V - A DYNAMIC MODEL FOR THE PRODUCTION OF H*, NO”, AND SO,” IN URBAN FOG CHAPTER VI - A FIELD INVESTIGATION OF PHYSICAL AND CHEMICAL MECHANISMS AFFECTING POLLUTANT CONCENTRATIONS IN FOG DROPLETS CHAPTER VII - THE HoSO, - HNO, - NH, SYSTEM AT HIGH HUMIDITIES AND IN FOGS: A FIELD STUDY OF WINTERTIME STAGNATION IN THE SAN JOAQUIN VALLEY OF CALIFORNIA CHAPTER VIII - COMPARISON OF FIELD DATA WITH THERMODYNAMIC CALCULATIONS FOR THE HoSO, - HNO, - NH, SYSTEM AT HIGH HUMIDITIES AND IN FOGS CHAPTER IX - FUTURE RESEARCH X1V 23 48 121 133 148 209 236 iii . TABLE OF CONTENTS (continued) Page APPENDIX A - INVENTORY OF AMMONIA EMISSIONS IN THE 240 SOUTHERN SAN JOAQUIN VALLEY OF CALIFORNIA APPENDIX B - CALCULATION OF SULFATE PRODUCTION IN FOGWATER 244 FROM SUCCESSIVE SAMPLES COLLECTED AT ONE SITE APPENDIX C - FOGWATER CHEMICAL COMPOSITION - COASTAL 250 CALIFORNIA, 1981- 1983 APPENDIX D - FOGWATER CHEMICAL COMPOSITION - BAKERSFIELD, 256 CALIFORNIA, DECEMBER 1982 - JANUARY 1983 APPENDIX E - FOGWATER AND AEROSOL CHEMICAL COMPOSITION, 286 HNO,(g) AND NH3(g) CONCENTRATIONS - SAN JOAQUIN VALLEY, DECEMBER 1983 - JANUARY 1984 iv ACKNOWLEDGEMENTS Completion of this thesis owes considerably to my "fog squad" coworkers, Jed Waldman and Bill Munger. They have been great friends and colleagues. I am deeply indebted to my adviser, Michael Hoffmann, who was always sensitive, enthusiastic and supportive, let me have a lot of freedom, and put me back on the right track when needed. I am also deeply indebted to my first adviser, Rick Flagan; in addition to providing great technical advice, he was sensitive and supportive in dealing with my first-year crises - research, courses, army. I am grateful to Jim Morgan, John Seinfeld, and Fred Shair for serving on my examining committee. My research was supported by the California Air Resources Board. I thank the many faculty, students, and staff who helped me out at some point or the other. Special thanks go to Ted Russell, Glen Cass, and Greg McRae; they were full of great advice, and took the time to provide it. Don Buchholz did great work for the fog squad in the field. Thanks to Sandy Brooks and Elaine Granger for their help with paperwork. Thanks to Theresa Fal] and coworkers for putting up with my deadlines and changes of mind. Thanks to Rayma Harrison and Gunilla Hastrup for their help in the library. Thanks to Elton Daly, Joe Fontana, Rich Eastvedt, and Leonard Montenegro for their great advice and for building worthwhile equipment from my vague designs. Finally, I dedicate this thesis to my parents who loved me and supported me the whole way. ABSTRACT A rotating arm device to collect large samples of fogwater for chemical analysis was designed and fully characterized. This instrument was used to determine the chemical composition of fogwater at a large number of sites in California. Fogwater at both urban and non-urban sites was frequently found to contain important inorganic acidity. Chemical models were developed to interpret the origins of this acidity in terms of aqueous-phase S(IV) oxidation reactions and acid/base titration processes in the fog and the precursor atmosphere. These models were found to succesfully interpret the inorganic acidity in fogwater and aerosol during wintertime stagnation episodes in the San Joaquin Valley of California. Table Ii, 1 Ill. 1 III. 2 III, 3 III. 4 III. 5 III. 6 IV. 1 IV. 2 IV. 3 IV. 4 IV. 5 IV. 6 vi LIST OF TABLES Fogwater collectors reported in the literature. Comparison of fogwater concentrations in samples collected simultaneously with the screen collector (SC) and the rotating arm collector (RAC) set side by side. Description of sampling sites. | Source concentrations of particulate matter. Liquid water-weighted average fogwater concentrations. Source contributions to fogwater loading. Equivalent ratios in coastal fogs. Chemical equilibria of the NaC1-H,S0,-HNO2-H50 system. Fog water composition in Southern California: observed concentration ranges. Theoretical rate expressions obtained from the Backstrom mechanism for various termination steps. Empirical rate laws reported by various investigators for the metal-catalyzed autoxidation of S05- Theoretical rate expressions obtained from the Hayon et al. and Schmittkunz mechanisms. Rate expressions for limiting cases of the Bassett and Parker mechanism. Empirical rate laws for Mn2* catalyzed autoxidation of S(IV) in aqueous solution. Page 19 38 39 40 42 43 44 50 55 56 57 60 62 © ‘Table IV. 7 IV. 8 IV. 9 IV. 10 IV. 11 IV. 12 IV. 13 Iv 14 vii LIST OF TABLES (continued) Page Empirical rate laws for Fe>* catalyzed autoxidation of 63 S(IV) in aqueous solution. Empirical rate laws for Cu2* catalyzed autoxidation of 68 S(IV) in aqueous solution. Empirical rate laws for Co(III)/Co(11)-catalyzed 69 autoxidation of S(IV). Empirical rate laws for the oxidation of S(IV) by hydrogen 83 peroxide. Empirical rate laws for the oxidation of S(IV) by ozone. 87 Comparison of stability constants for metal-sulfito and 94 -sulfato complexes at 25.0 C. Equilibria and thermodynamic data considered in fog water 99 model. Atmospheric conditions for the fog events simulated by 102 the model. Henry's law and aqueous-phase equilibria relevant to 123 the droplet chemistry. | Kinetic expressions for the aqueous phase oxidation 124 reactions of S(IV) to S(VI) and N(III) to N(V). Reaction stoichiometries for reactions in Table 2a. 124 Composition of the air mass (trace gases and 125 condensation nuclei) prior to fog formation at a polluted site. Scavenging of gases by fog droplets under conditions 126 of Table 3. viii LIST OF TABLES (continued) Table Vv. 5 Scavenging of gases by fog droplets under conditions of Table 3. VI. 1 Fogwater concentrations at Bakersfield, California. VI. 2 | S09 oxidation rates in the atmosphere. VII. 1 S0,, NO» and NH. emission inventories for the southern San Joaquin Valley (SSJV; see Figure 1 for definition). VII. 2 Aerosol, HNO3(g), and NH3(g) data, 31 December 1983 - 14 January 1984, VII, 3. Liquid water-weighted average fogwater concentrations, 31 December 1983 - 14 January 1984. VII. 4a Average alkalinities at each site. VII. 4b Contributions of different species to fogwater alkalinity. VII. 5 Parameters for st irred-tank model simulation. VIII. 1 Dissociation and vapor pressure equilibria for HNO, and NH, VIII. 2 Bakersfield data for aerosol and gaseous HS0,-HNO3-NH, species. VIII. 3a Chemical composition of fogwater samples used in comparison. VIII. 3b Comparison of fogwater data with aerosol, HNO3(g) and NH3(g) data. A. 1 Inventory of ammonia emissions in the southern San Joaquin Valley of California. B. 1 Rates of sulfate production in fogwater. 137 143 188 189 190 191 191 192 228 229 230 231 . 241 249 Figure II. 1 II. 2 It. 1 III. 2a III. 2b IV. 1 ix LIST OF FIGURES (Top) Typical urban aerosol size distribution profile. (Bottom) Expected shift in the size distribution profile as a result of fog formation. Caltech rotating arm collector. Model rotating arm collector used in calibration experiment. Calibration experiment setup. Collection efficiency of model collector vs. the generalized Stokes number, for greased (in bold) and ungreased impaction surfaces. Collected water per cubic meter of air sampled vs. liquid water content. Screen collector to sample fogwater for chemical analysis. Collection efficiency vs. droplet diameter. Fogwater sampling sites. Fogwater concentrations at Del Mar. Source contributions to the fogwater loading at Del Mar. Schematic representation of the proposed catalytic reaction mechanism starting with the Co(II)-TSP complex as the active catalytic center. 21 22 45 46 47 71 Figure IV. IV. IV. IV. IV. IV. IV. IV. IV. 2 10 Xx LIST OF FIGURES (continued) Speciation of Fe(III) vs. pH in aqueous solution as a function of the stability constant of FeSO,” as determined with SURFEQL. Speciation of S(IV) vs. pH for the conditions on the far right of the figure as determined by SURFEQL. Speciation of Fe(III) perturbed by iron(III)oxalato complexes as a function of pH and stability constant of the FeS0,". Speciation of Fe(III), S(1V), Cu(II) and Mn(II) in fog water as a function of the pH. Sulfate formed, pH and total S(IV) for the fog water as a function of time after fog formation. Sulfate formed, pH and total S(IV) for the fog water as a function of time after fog formation. Sulfate formed, pH and total S(IV) for the fog water as a function of time after fog formation. Sulfate formed, pH and total S(IV) for the fog water as a function of time after fog formation. Sulfate formed, pH and nitrate for the fog water as a function of time after fog formation. (a) Profile versus time of total sulfate in the fogwater and of the individual contributions to the total sulfate of sulfate aerosol and different S(IV) oxidants. (b) Profile of pH versus time. Page 92 93 94 95 104 105 106 107 108 126 xi LIST OF FIGURES (continued) * Figure Page Vv. 2 Speciation of Fe(III) and Mn(II) in the foqgwater as a 126 function of time. Vv. 3 Concentrations of S(IV) species in the fogwater as a 127 function of time. Vv. 4 (a) Profile versus time of total sulfate in the fogwater 127 and of the individual contributions to the total sulfate of sulfate aerosol and different S(IV) oxidants. (b) Profile of pH versus time. Vv. 5 Concentrations of S(IV) species in the fogwater as a 128 function of time. Vv. 6 (a) Profile versus time of total sulfate in the fogwater 128 and of the individual contributions to the total sulfate of sulfate aerosol and different S(IV) oxidants. (b) Profile of pH versus time. Vv. 7 Speciation of Fe(III) in the fogwater as a 128 function of time. Vv. 8 Concentrations of S(IV) species in the fogwater as a 129 function of time. V. 9 ~~ Liquid water profile versus time chosen for the 129 simulation of a fog event in the Los Angeles basin. V. 10 Profile versus time of the concentrations of major ions 130 in fogwater under conditions of Table 3, with liquid water content as given in Figure V. 9. | V. 11 Speciation of Fe(III) in the fogwater as a 131 function of time. Figure V. 12 VI. 1 VI. 2 VI. 3 VI. 4 VI. 5 VI. 6 VI. 7 VI. 8 VII. 1 VII. 2 VII. 3 x1i LIST OF FIGURES (continued) Concentrations of S(IV) species in the fogwater as a function of time. The Central Valley of California. Evolution of fogwater concentrations over the course of the 7 January 1983 fog event at Bakersfield. Fogwater and aerosol loadings measured at Bakersfield during the period 30 December 1982 - 15 January 1983. Evolution of fogwater ionic loadings over the course of the 7 January fog event. Frequency distribution of the pseudo first-order rate constant for S(VI) in-fog production. Evolution of the sulfate equivalent fraction in the fogwater and in the aerosol at Bakersfield during the period 30 December 1982 - 15 January 1983. pH frequency distribution for fogwater samples collected at Bakersfield during the period 30 December 1982-15 January 1983. Frequency distribution of the non-neutralized fraction of the acidity. Sampling sites in the San Joaquin Valley of California. Concentrations of S055 NOs and CO at Kernridge, mixing heights at Kernridge and stratus cloud bases over Bakersfield NWS. Flow patterns during the sampling program. page 131 135 138 139 140 143 143 144 145 202 203 204 "Figure VII. 4 VII. 5 xiii LIST OF FIGURES (continued) Fogwater composition determined simultaneously at four sites. Aerosol and gas-phase concentrations, and atmospheric alkalinities, at the six valley sites. VII. 6 VII. 7 VIIT. 1 VIII. 2 Concentrations of 50, and NO, at McKittrick. Stirred-tank simulations of aerosol accumulation over the course of a stagnation episode. Fraction of gaseous ammonia scavenged by fogwater. Comparison of Bakersfield field data to the Bassett and Seinfeld (1983) equilibrium model. (a) External mixture assumption. (b) Internal mixture assumption. Page 205 206 207 208 233 234 235 XiV OVERVIEW. The principal object of this thesis is to document the chemical composition of fogwater in California, explain the frequent occurrence of high-acidity fog events, and characterize the main processes determining fogwater acidity. Few attempts have been made prior to this work to collect fogwater for chemical analysis. Houghton (1955) observed pH values in the range 3.5 - 7.4 in coastal fogs of New England and Canada, and large concentrations of chloride and sulfate. Mrose (1960) sampled coastal and urban fogs in East Germany, and reported pH values in the range 3.8 - 4.2. Mack et al. (1977) found high sulfate and nitrate concentrations in fogwater collected off the shore of California. A few other investigators have collected cloudwater intercepting mountain slopes (Okita, 1968; Lazrus, 1970). The sparseness of data on the chemical composition of fogwater is rather surprising, since fogs have been associated with major health-threatening air pollution episodes: notorious examples are the fogs of London (Commins and Waller, 1967), the Meuse valley (Firket, 1936), and Donora, PA (Schrenk et al., 1949). Further, deposition of acid fogwater has been shown to damage crops (Thomas et al., 1952; Granett and Musselman, 1984) and forests (Scherbatskoy and Klein, 1983). A likely reason for the sparseness of data is that collecting fogwater is not a trivial matter. Fog droplets range in size from 1 to 100 um, with a mass median diameter typically in the 10 - 30 pm range (Pruppacher and Klett, 1978). Droplets of this size are difficult to collect efficiently because of their inertia. A XV ‘fogwater sampling device must collect efficiently droplets up to at Jeast 100 um; at the same time, it must exclude non-activated submicron’ particles which have physical and chemical properties different from those of the fog droplets. Further, large amounts of water must be provided for wet chemical analysis, while minimizing evaporation and condensation during the sampling process. These requirements cannot be satisfied by standard techniques for aerosol sampling. The first step in this project was to develop a reliable instrument for fogwater sampling. The rotating arm collector designed in our laboratory (Chapter 1) has proven to be an effective means for collecting representative fogwater samples (Hering and Blumenthal, 1984). The instrument collects fog droplets by impaction on a slotted rod rotating at high velocity; impacted droplets are driven by centrifugal force into bottles mounted at the ends of the 1 are typically rod. In dense fog, collection rates of 1 ml min” achieved. Droplet sizes are preserved at all stages of collection; i.e., no significant evaporation or condensation occurs over the course of the sampling process. The instrument was calibrated by releasing a single plume of SF ¢ tracer and chemically tagged monodisperse aerosol in the atmosphere upwind of a scaled-down version of the instrument. By comparing the SF¢ concentration measured in the plane of collection to the amount of aerosol collected on the rod, the collection efficiency vs. particle size was determined. The lower size cut (50% collection efficiency) was found to be 20 pm diameter. The rotating arm collector has drawbacks in that it does not xvi collect small fog droplets efficiently, it is not easily automatable, ‘rand it presents safety problems. Chapter 2 describes a screen collector which we recently developed to provide a simple. reliable instrument for routine monitoring of fogwater chemical composition. The rotating arm collector was used in fogwater sampling programs at several locations in California. Fogwater samples collected along the California coast were found to be consistently acidic, even in rural areas (Chapter 3). Extremely high acidities (pH down to 1.69) were observed at sites in the Los Angeles basin and downwind. The main contributors to strong acidity were sulfuric and nitric acids; coastal urban activities and oi] production operations are important sources of SQ, and NO,» precursors of these acids. Coastal atmospheres were found to have low acid-neutralizing capacities, and thus to be very sensitive to acid inputs. A dynamic equilibrium model of fogwater chemistry was developed to study the effect on fogwater acidity of SO, oxidation in the fog droplet phase. Chapter 4 presents a review of the role of trace metals in catalyzing S(IV) aqueous-phase oxidation processes. It is shown with a simple model that the autoxidation reaction catalyzed by Fe? * and Mne* in solution can be an important source of sulfuric acid production in fog. A model is developed in Chapter 5 which includes a large number of aqueous-phase equilibria and oxidation pathways, although it makes some crude assumptions regarding physical processes (e.g., transport, deposition, size-composition relationships are ignored). From the S(IV) oxidation rates derived in Chapter 4, it is shown that a fog droplet initially forming on non-acidic nuclei under conditions typical of the Los Angeles coastline may experience a xvii rapid drop in pH to below pH 3 as a result of S0, Scavenging and aqueous-phase oxidation of S(IV) to S(VI). The main aqueous-phase 2+) ‘oxidants are found to be H,0, and 0, (catalyzed by Fe°* and Mn@*). - Ozone is an important oxidant at high pH, but becomes ineffective at pH < 4, The stable S(1V)-formaldehyde adduct is found to be the major form of S(IV) in the fogwater, but it forms too slowly to compete with S(IV) oxidation. The extent of S(IV) oxidation is mostly limited by the availability of H05 and the low solubility of S0, at low pH. When fog droplets form on acidic nuclei, it is found that the low initial pH prevents oxidation of S(IV) other than by H,0,. Aqueous-phase conversion of N(III) to N(V) in fog is found to be very slow. The model of Chapter 5 was used to study the scavenging of trace gases in fog. Nitric acid is 100% scavenged under all conditions because of the large amounts of suspended liquid water. Ammonia is 100% scavenged at pH < 5, but at higher pH it mostly remains in the gas phase. Only a small fraction of $0, is initially scavenged, but this fraction increases as aqueous-phase S(IV) is oxidized to S(VI). Still, most of the S05 remains in the gas phase after 4 hours of fog. A small fraction of formaldehyde is scavenged in fog by formation of the S(IV)-formaldehyde adduct. Of the oxidants, 03 is poorly soluble but H0. is highly soluble. Because H,0, rapidly oxidizes S(IV) in the droplet, it is quickly depleted from the atmosphere. Fogwater acidity may be neutralized if a sufficient amount of base is present in the atmosphere. Neutralization of fogwater acidity by ammonia was investigated in two field experiments xviti conducted in the San Joaquin Valley of California, where large inputs of strong acids are provided by oi] and gas production (SO,,-NO,), ‘and large quantities of NH are released by industrial farming and livestock feeding operations. Chapters 6 and 7 present the results of these field experiments. The H5S0,-HNO,-NH, system contributed over 90% of the inorganic aerosol and fogwater ionic loading, and the fogwater alkalinity/mineral acidity could be correctly predicted from the local balance of sulfuric acid, nitric acid and ammonia in the atmosphere. The fogwater acidity directly reflected the relative proximities of oil fields and confined feeding operations. Most of the fogs sampled in the San Joaquin Valley formed over the course of extended stagnation episodes. Progressive pollutant accumulation under stagnant conditions was documented under non-foggy stagnant conditions, and decreases in aerosol concentrations were observed following fogs. It is suggested that deposition of fog droplets provides a rapid sink of suspended pollutants, so that the occurrence of fogs limits pollutant accumulation in a stagnant atmosphere. Accumulation profiles observed under stagnant conditions are discussed in detail in Chapter 7. Deposition was the principal means of removal for S(VI), N(V), and N(-III); the patterns of pollutant accumulation were consistent with deposition velocities < 0.1 cm sec for secondary (S0,°", NO,”, NH,” H*) aerosol, and deposition velocities > 1 cm sec for HNO3(g). Slow scavenging of SO, by fog was documented, and an average S05 conversion rate of 2 % hv) was determined in an overcast boundary layer. Production of HNO. was observed at the beginning of the stagnation episode, but was limited Xiv by the availability of gas-phase oxidants. NO. was found not to be significantly scavenged by fogwater. Overall, accumulation of strong acids HS0, and HNO, over the course of a stagnation episode resulted in a decrease of alkalinities at all sites in the southern San Joaquin Valley, and production of mineral acidities at the sites most distant from NH, emissions. The San Joaquin Valley experiment provided simu] taneous measurements of S(VI}, N(V), and N(-III) concentrations in the gas phase, the aerosol, and the fogwater. These data were used in Chapter 8 to test the applicability of thermodynamic models for the H5S04-HNO,-NH,-H.0 system at high humidities and in fogs. Under non-foggy alkaline conditions the aerosol was a neutralized mixture, alkalinity remained in the gas phase as NH3(g)> and HNO3(g) concentrations were very low. Under non-foggy acidic conditions NH3(g) concentrations were very low, and the acidity was present both in the gas phase as HNO.(g) and in the aerosol phase as partly neutralized sulfuric acid. These findings are in qualitative agreement with the model results of Bassett and Seinfeld (1983). The dissociation constants K=P wy 3PuNO3 predicted by Bassett and Seinfeld agreed to within one order of magnitude with the products of concentrations observed in the field, although they were systematically too low. Several possible explanations for this discrepancy were discussed. In fog, concentrations of HNO,(g) were at or below the 3 under non-acidic conditions, and were detection limit of 4 neq m very low (although generally nonzero) under acidic conditions. The observation of detectable HNO, (g) in acidic fog was attributed to the XX presence of unsaturated air parcels within the fog, and to the slow “rate of diffusion of HNO3(g) to the fog droplets. No detectable NH3(g) was found under foggy acidic conditions, but substantial amounts were found under foggy non-acidic conditions. The observed NH3(g) concentrations were of the same magnitude as those predicted at equilibrium with fogwater. REFERENCES Bassett, M., and Seinfeld, J.H. 1983. Atmos.Environ. 17,2237-2252. Commins, B.T., and Waller, R.E. 1967. Atmos.Environ. 1,49-68. Firket, J. 1936. Trans.Faraday Soc. 32,1192-1197. Granett, A.L., and Musselman, R.C. 1984, Atmos.Environ. 18,887-891. Hering, S.V., and Blumenthal, D.L. 1984, "Fog Sampler Intercomparison Study: Final Report to the Coordinating Research Council." Available from Coordinating Research Council, 219 Perimeter Center Parkway, Atlanta, GA. Houghton, H.G. 1955. J.Met, 12,355-357. xxi Lazrus, A.L., Baynton, H.W., and Lodge, J.P. 1970. Tellus 22,106-114, Mack, E.J., Katz, U., Rogers, C.W., Gaucher, D.W., Piech, K.R., Akers, C.K., and Pilie, R.J. 1977, Report CJ-6017-M-1, Calspan Corp., Buffalo, NY. Mrose, H. 1960. Tellus 18, 256-270. Okita, T. 1968. J.Met.Soc.Japan 46,120-126. Pruppacher, H.R., and Klett, J.D. 1978. Microphysics of Clouds and Precipitation. Reidel, Amsterdam, 418-421. Scherbatskoy, T., and Klein, R.M. 1983. J.Environ.Quality 12,189-195, Schrenk, H.H., Heiman, H., Clayton, G.D., Gafefer, W.M., and Wexler, H, 1949, "Air Pollution in Donora, PA: Epidemiology of the Unusual Smog Episode of October 1948, Preliminary Report." Public Health Bulletin 306,pp 1-73, Public Health Science: Washington D.C. Thomas,-M.D., Hendricks, R.H., and Hil], G.R. 1952. in Air Pollution, L. McCabe ed., McGraw-Hill, New York, p.41-47. CHAPTER I FOGWATER COLLECTOR DESIGN AND CHARACTERIZATION by Daniel J. Jacob, Rueen-Fang T. Wang, and Richard C. Flagan Environmental Science and Technology 18, 827-833 (1984) m ENVIRONMENTAL SCIENCE & TECHNOLOGY, Vol. 18, Page 827, November, 1984 rinted fro ower © 1984 by the American Chemi ical Society and reprinted by permission of the copyright owner 2 Fogwater Collector Design and Characterization Daniel J. Jacob, Rueen-Fang T. Wang, and Richard C, Flagan’ Environmental Engineering Pasadena, California 91125 - @ The detailed characterization of a rotating arm collector to sample ambient fog droplets for chemical analysis is presented. Because of the large volume of sample required, and because fog droplets are of supermicron size and are sensitive to local thermodynamic disturbances, conven- tional methods for atmospheric aerosol sampling are not suitable for fogwater sampling. Design samplers are outlined. Devices used in previous investi- gations are evaluated in light of these criteria. The design Of a rotating arm collector is discussed, and it a shown that this instrument performs adequately in preserving the physical and chemical integrity of the sample at al] stages of collection. Limitations in the design due to mechanical constraints are discussed. Results of an in situ calibration experiment using a chemically tagged monodisperse aerosol indicate a size cut of 20-um diameter. Introduction Supermicron particles contribute significantly to the total mass of a dry aerosol (J). This contribution increases considerably when the aerosol is wetted, especially under nuclei grow rapidly to form cloud or fog droplets. Whereas the mass loading of an urban aerosol under nonsaturated conditions is of the order of 10“ g m%, the liquid water content in a cloud or fog ranges from 0.01 to 1 gm", with supermicron droplets constituting the bulk of the aerosol mass. In urban environments fogwater has been found to contain extremely high pollutant concentrations, often associated with high acidities (2, 3). Supermicron particles are difficult to collect efficiently because their inertia may prevent them from following the air streamlines converging toward the inlet of the sampler (4). Furthermore, water droplets in the atmosphere are in a fragile thermodynamic balance with the ambient hu- midity which is very sensitive to perturbations by a sam- pling device. Conventional methods for collecting samples of total particulate matter may, therefore, | lead to sampling _ biases due to anisokinetic sampling conditions or changes in temperature or pressure. Fog sampling is further com- plicated by the relatively large sample volume required for the study of the detailed aquatic chemistry. At least 10 ma of sample is needed for the standard inorganic analysis routinely carried out in our fog program (3). If the vari- ation in chemical composition throughout the fog event ia to be studied, sampling intervals should be short. For a typical liquid water content of 0.1 g m™, a sampling rate of over 1.7 m* min” is required to collect 10 mL for analysis in an hour. Larger sampling rates are required for the study of light fogs. . A rotating arm collector based on the principle of inertial impaction has been developed and used in our intensive fog sampling program. Major advances in the under- standing of fog chemistry and the role of fogs in acid de- position have been made by using this device (3, 5). In a recent field intercomparison study of fogwater collectors (6), samples were collected simultaneously with instru- ments from five different research groups and analyzed for major ions. Ionic concentrations in samples collected by the rotating arm collector and a jet impactor (7) agreed 00 13-936X/84/08 18-0827$01.50/0 © 1984 American Chemical Society criteria for fogwater Science, Keok Engineering Laboratories, California Institute of Technology, within 5%; other collectors gave systematically either higher or lower concentrations. The rotating arm collector was found to collect water efficiently in both light and heavy fogs. To date, the sampling characteristics of the rotating arm collector bave only been qualitatively explored by meas- uring the change in the droplet size distribution in a cloud chamber which resulte from its operation. These mea- surements indicated that the minimum size of particles collected was at least 8 um but did not provide sufficient resolution to determine the size-dependent collection ef- ficiency (8). In this paper we first elaborate on the design criteria relevant to fogwater collection and then present a detailed examination of the design and operation of our rotating arm collector. Constrainta on the design due to power requirements and possible sample biases due to aerodynamic heating are explored. Measurements of the collection efficiency as a function of particle size are presented. Design Criteria for Fogwater Collectors Size Cut. Fog droplete form by activation of atmos- pheric particles (condensation nuclei) under supersaturated conditions. At the levels of supersaturation found in the atmoephere, the lower size limit for particles to be acti- vated is of the order of 0.1 wm (9). Figure 1 shows how fog formation can shift the size distribution of an urban aerosol; particles in the first mode (below 0.1 um) are rather unaffected by the condensation process, but most particles in the two higher modes grow by condensation to much larger sizes. Therefore, two types of particles coexist in a fog: {1) supermicron fog droplets and (2) nonactivated, primarily submicron, particles. Being dilute aqueous solutions, fog droplete do not interact with their environment in the same way as the solid or concentrated submicron particles (10, 11). It is, therefore, important that a fogwater sampler differentiate between the two types of particles. Fog droplets range in size from 1 to 100 um, with a mass median diameter usually in the range 10-40 wm (12-15). The dependence of the fog droplet chemical composition on droplet size has not been rigorously investigated to date; general predictions from droplet growth theory (J6) are difficult to make because humidities in fogs fluctuate rapidly in a manner that is still poorly understood (17, 18). Large droplets are not necessarily more dilute than smaller droplets because they generally result from condensation on larger nuclei. If the total pollutant burden associated with fogs is to be determined, droplets of all sizes should be collected with the same efficiency. Nonactivated sub- micron particles represent a very smal! fraction of the total aerosol mass but they could, if collected, contribute a sizable amount of solutes to the sample and result in a serious bias. A sharp lower size cut in the range 1-10-um diameter is, therefore, desired. Furthermore, since most of the fog mass is associated with large droplets, droplets up t to about 100-um diameter must be collected without Three methods are available to collect the large particles while excluding the smaller particles: sedimentation, in- ertia] separation, and removal of smaller particles by Environ. Sci. Technol., Vol. 18, No. 11, 1984 827 3 Table I. Fogwater Collectors Reported in the Literature characteristic width’ ower size inlet Stokes impaction velocity, asempling rate, inlet, impaction no. for 100-4m reference type ens? wo? min? em = surface, cm cut, . droplets passive Mrose (20) eloth surface ambient wind variable w variable Okita (21) gid ambient wind variable 0.01 variable Lazrus et al. (22) = screen ambient wind variable ? variable Sedasivan (25) screen ambient wind variable ? variable Falconer and grid ambient wind variable 0.02 variable Falconer (24) active Houghton and screen 600 102 30 0.01 bad 0.6 Radford (12) May (25) gid 450 il 10 0.05 r 14 May (25) jet impactor 1700 0.05 0.85 04 ? 13 Okita (21) ecreen o4 ‘1 16 ? ? 0.3 Mack and rotating arm 1500-5000 7 045 ? Pilie (26) Katz (7) jet impactor 2000 12 ? 0.2 5¢ ? Brewer et al. (27) screen 320 15 5 0.02* 5° 2.0 this paper rotating arm 3800-5600 5 0.48 20 * Characteristic width of inlet: radius of circular inlet (12, 21, 256, 27), half-width of square inlet (25a). Characteristic width of impaction sorfaces: radius of wires (12, 21, 24, 25a, 27), half-width of jet (7, 255), radius of rod (26, this paper). *Diameter of droplets collected with 50% efficiency. "Calculated from impaction theory for cylinders (19). ‘Calculated by Katz from theory and confirmed by experiment. “Obtained by personal communication from R. L. Brewer. ‘? = not reported, or cannot be computed from available data. eo! . Rr of \ NG a wf J \ } 4 ro / “ee . old. ! LN ot cy Diam) © $s C) ” £ a "4 AL i ° w 100 © tym) Figure 1. (Top) Typical urban sercsol size distribution profile (1). @ottom) Expected shift in the size distribution profile as a result of fog diffusion. Of these, inertial separation, particularly im- pection, most readily achieves sharp size cuta of a few microns and has been the method favored by past inves- tigators. The efficiency of collection by impaction is a function of the Stokes number St (79): ppD?U St = ee0 q@ where pp is the droplet density, D is the droplet diameter, U is the velocity of approach, y is the viscosity of air, and a is a characteristic width of the impaction surface. The size cut of a sampler is defined as the diameter Dy of dropleta collected with 50% efficiency, and the cutoff Stokes number St, is the corresponding Stokes number. Fogwater samplers used in past investigations (Table I) can be divided into two general groups: (1) passive col- 828 Environ. Sci. Technol., Vol. 18, No. 11, 1984 lectors, where droplets impact on the collecting surface with the velocity of the ambient wind, and (2) active collectors, where droplets are accelerated to a certain ve- locity as they approach the collection surface. Passive collectors have the obvious drawback that the impaction velocity is not well-defined but instead fluctuates with the ambient wind. As a result the size cut varies with time, and the sample is not necessarily representative of the actual fogwater. Furthermore, at low wind speeds the minimum size of particles which will impact becomes large enough that the rate of sample collection becomes unac- ceptable, thereby precluding the use of these collectors in many locations. In active collectors, the size-dependent collection efficiency can be characterized for a given ge- ometry of the impaction surface. However, modification of the ambient velocity of the droplets introduces two potentially important problems: perturbation of the am- bient thermodynamic equilibrium and anisokinetic sam- pling. Perturbation of the Ambient Thermodynamic Equilibrium. Modification of the flow field in active collectors may produce significant evaporation or con- densation of the droplets as they approach the impaction surface. The total loss of water droplets has been reported past the first two stages of a cascade impactor (25). The extent to which droplet sizes are modified during approach ean be estimated by simple mass transfer calculations, as shown in the next section. Unfortunately, no such calcu- lations have been reported in the references of Table I. Evaporation may also occur following collection if the droplets are not sheltered immediately from the flow of air past the impaction surface, because the air mass in a fog is thermodynamically inhomogeneous (29) and contains pockets of unsaturated air. Aside from evaporation, con- tact of the collected droplets with changing air masses may alter their compositions. It is therefore important that collected droplets be removed rapidly from the air flow. In the jet impactor designed by Katz (7), this is achieved with a rotating impaction surface; in the rotating arm collector, centrifugal force rapidly drives the impacted droplets into collection bottles where the air is stagnant. The screen collectors of Table I rely on gravity to draw the impacted droplets into storage bottles at the bottom of the . tothe walls within a few inlet di 4 screen, but this is a slow process Ceepecially in light fogs). Anisokinetic Sampling. Recent studies of anisokinetic -gampling biases (30, 31) have indicated that a sampler oriented into the wind can lose over 30% of particles with inlet Stokes’ numbers larger than 0.5. In a crosswind, . hosses are considerably greater. Losses are found to be maximum for sampling velocities 2-5 times the ambient ’ wind velocity. In addition, deposition of most particles from the inlet has been reported for Re > 5000 (31). ' Although moderate sampling velocities have been used in the ducted devices listed in Table I, inlet Stokes’ num bers are still high in most cases. In screen and grid col- ‘Jectors, the moderate sampling velocities used introduce the additional problem that very thin wires are required for impaction. Because the larger droplets in the fog are then comparable in size to the radius of the impaction cylinder, interception and aerodynamic interactions be- tween the dropleta and the wires become important. The impaction characteristics are then quite different from those of the point particles considered in most theoretical treatments of impaction on cylinders. The apparent ge- ometry of a very thin wire also changes significantly as fog droplets collect on it, introducing uncertainty in the size cut. Proper scaling of the inlet could allow the use of high impaction velocities with little anisokinetic sampling bias. However, droplets tend to fly off collection strings at ve- locities higher than about 8 m s™! (12). The anisokineticity effect can be suppressed by elimi- nating the inlet and, instead, generating an apparent ve- locity by moving the impaction surface at high speed through the ambient air. Such devices have been used in airborne sampling of clouds (32, 33), where the speed of the aircraft constituted the apparent velocity. In ground-based collectors, high velocities can be generated by using a rotating systex. Design of a Rotating Arm Collector parm collectors have been used for many years to collect biological particles on adhesive coated surfaces (34-36). Recently, a rotating arm virtual impactor has been developed to sample isokinetically giant atmospheric particles (37). In another device, the solid arms have been modified to collect fogwater by introducing a slot in the leading edges of a hollow arm (26). For the purpose of our field programs, we have developed an improved version of the latter collector. Principle. The design of the rotating arm collector used in the Caltech fog sampling program is shown in Figure 2. A motor drives at high speed (1700 rpm) a type 304-L stainless steel solid rod of length 22 = 63 cm. Each end of the rod has a alot milled into its leading edge. Standard 30-mL VWR narrow mouth bottles are mounted at the ends of the arm to collect the water which impacta in the slots and flows outward by centrifugal force. Threaded Tefion tubes screwed onto the end of the arm and ex- . tending inside the collection bottles prevent the collected fogwater from running out after the instrument is stopped. Deflectors prevent water which impacts on the solid part of the arm from entering the slot since the collection of this water would bias the sample toward large droplet sizes. Small fins are welded to the back of the arm for extra strength. The entire arm is Teflon coated. In the absence of previous calibration data, the design was based on a cutoff Stokes number of unity. The cutoff size is weakly dependent on Sts; i.e., Dyp = Styo'/?, a0 even though the ‘actual St, may differ substantially from unity, the effect on Dy was not expected to be too large. ‘cellection viel ; Ik 1S \ SECTION AA’ N lentorged) OD. 23 00.4 (A OD, OD. wa OD.18 5 — V 19 i\ B A 30 mi to 155———»4 - 195 te — 100 eb S04 40 o} — 385 4 All dimensions are in millimeters Figure 2. Caltech rotating arm collector. Full welds attaching the fin ® the rod are shown in dark. Mechanical Constraints. The size cut is determined by the choice of (U/w)'/?, where w is the width of the slotted portion of the arm. Because reducing w leads to @ proportional reduction in the sampling rate achieved with the instrument, a preferable way to obtain an acceptably - low Dy is to operate at a high approach velocity. However, the resulting aerodynamic drag force F limits the extent to which U can be increased. The arm consists of cylindrical sections and a half- cylinder cup. The drag coefficients C, of these shapes at Reynolds numbers in the range 1010" are approximately 1 and 2, reepectively (38). The power P required to rotate the arm at a rotating speed w (rps) is given by Pe 2( — U(D dl (2) This may be separated into the components for the solid rod (! < L,), the slot (ZL, < I < L,), the bottle cap (L, < 1 < L,), and the bottle (L; <1 <L). Noting that the drag force is given by F = CyAp,(U?/2) (3) where A is the projected area of the shape and p, is the density of air, we find P = 2pyteS CywlL - Les" «) is] where the w;’s are the widths of the corresponding portions of the arm, L, = 0, and L, = L. The power required to drive the arm is a strong function of the dimensions of the arm and the rotation frequency. To facilitate operation of the arm using readily available electrical circuits, the size and speed must be limited. Our rotating arm was designed to achieve an acceptable sampling rate and size cut, while not requiring more than a 110-V, 15-A circuit to drive the sampler. The power required when the col- lector is rotating at 1700 rpm is calculated from eq 4 to be 900 W. A 1.5-hp (1120-W) motor is sufficient to drive the collector, and this is etill compatible with electrical circuits likely to be available in the field. The 30-mL ical collection bottles are major contributors to the drag (450 W) because of their size, shape, and velocity. . Collection vials are very emall in the instrument described by Mack and Pilie (26), but the resulting samples are then too amall for chemical determination, and the sampler is Environ. Sci. Technol., Vol. 18, No. 11, 1984 829 inconvenient because of rapid overloading of the vials. Streamlined bottles or bottle casings could be adapted to the present design to minimize the drag while keeping the sample volume large enough to make the instrument practical 8 The slot velocities range from 38 to 56 mo", and the Reynolds numbers at the slots range from 94000 to 35000. Laboratory tests under zero wind con- ditions indicate that air is drawn through both faces of the collector at a velocity of 1.5 ms! roughly uniform across the plane of collection (measured 25 cm away from the plane of collection). The sampled air is expelied radially (velocity of 4 m a"! measured 25 cm away from the tip). This induced flow ensuree that the sampled air is entirely renewed at every half-rotation of the arm, so that the sampling rate is 5 m* of air/min. Assuming 100% effi- ciency, a collection rate of 0.5 mL min™ would be achieved in a fog of 0.1 g m™ liquid water content. Collection rates of up to 2 mL min" have been obtained in the field. Droplet Evaporation. Droplet evaporation may occur at the three stages of collection: (1) as the droplet ap- proaches the impaction surface, (2) in the collection slota, and (3) in the collection bottles. Evaporation during approach is most likely to occur as the droplet approaches the slot at high apparent velocity. Let us consider, as a worst case, the stagnation streamline; as air approaches the slot its velocity decreases from U. in the free stream to 0 at the stagnation point. This de- celeration leads to aerodynamic heating. We write the appropriate equations for mass transfer (16) and droplet trajectory: because we are concerned only with activated supermicron droplets, we neglect the effect of solutes on the physical properties of the droplets. dr s- (T) ra (5) PwRT + SHow(l SHM, 1 @u(T)DM, kT \ RT 20M, 9(T) = RToar Ya a. -Up (6) dUp 3f a \_ (Up- U a -3( oo a ris the droplet radius; s = (e/e,.,(T)) — 1 is the ambient supersaturation, where e is the water vapor pressure over the droplet and e,,,(T) is the saturation water vapor pressure over a plane water surface at temperature 7; is the surface tension of water against air; R is the universal gas constant; py and p, are the densities of water and air, respectively; D is the diffusivity of water vapor in air; AH is the latent heat of vaporization of water, k is thermal heat conductivity of air, M, is the molecular weight of water; z is the distance of the droplet to the collector; U is the velocity of the air flow; Up is the velocity of the ap- proaching droplet. The local Reynolds number Re, of the droplet is _ Up - Uv td Re, (8) where » is the kinematic viscosity of air. The drag coef- ficient Cy can be approximated for Re, < 10° by (39) 24 G= Re," + 0.158Re,/*) (9) Under the assumption that the compression proceeds is- 830 Environ. Sci. Technol., Vol. 18,No. 11, 1984 entropically, the temperature and pressure fields for the approaching droplet are given by U- U T=T + (10) 2cy . ard . par (2) an where c, is the specific heat of air at constant pressure and + is the ratio of specific heats. The water vapor pressure is calculated by assuming that the mixing ratio of water vapor remains constant as the air approaches the collector and that droplets in the free stream are at equilibrium with water vapor: eT) = Feud To) + AT (12) For a given U(z) along the stagnation streamline, the coupled eq 5-7 can readily be solved numerically. Un- fortunately, the complicated potential flow around a cup has not yet been characterized to the authors’ knowledge. It is assumed here that the flow (at least to within some distance of the impaction surface) should be comparable to that around a cylinder of diameter w, for which the potential flow solution along the stagnation streamline - U= v.| 1- (2) (13) The system of eq 5-7 is then integrated with a fourth- order Runge-Kutta routine. Under the conditions U. = 66 m s!, T.. = 283 K, and P. = 1 atm, the percentage of loss in droplet mass by the time the droplet reaches the stagnation point (actually the time at which (2z - w)/w < 0.001) is less than 0.1% for activated supermicron dro- plets. Increasing U. to 100 m 8” does not significantly increase thia loss; therefore, evaporation during approach to the slots is not a constraint in choosing higher approach velocities. A similar mass transfer calculation was conducted to evaluate the effect of thermodynamic modifications as air is drawn through the sampler with a velocity U = 1.5 m a"). It was found that no significant change in droplet size occurs for supermicron droplets during that stage of ap- proach. Evaporation in the slot could also occur after impaction from the thin film of water in the slot to the unsaturated air above. Because the flow patterns in the slot originate mostly from turbulent eddies, an accurate description of the transfer phenomena would be very complicated. By using an oversimplified model of tangential flow of un- saturated air over a flat plate, we estimated that under the worst conditions the evaporation rate could be no more than 0.01 g min“. Even in this case, evaporation is still very small at typical collection rates (0.1-1 g min”). Finally, evaporation in the collection bottles must be considered. As a test, the bottles were filled halfway with water (15 mL in each), mounted on the collector, and spun for 30 min in a dry atmosphere. This did not affect the volume of water in the bottles, which shows that evapo- ration in the bottles is negligible. Bafety. Safety is a primordial concern for a large di- ameter device rotating in the open. The collector must be carefully balanced to prevent vibrations and securely mounted to a rigid stand. The mechaniams for mounting the bottles must be able to withstand the loads due to high acceleration (1200g). Stainless stee! caps with close tol- 6 erance threads were found to be satisfactory in laboratory tests. The operating site must be carefully selected and supervised to minimize the hazard. The arm must be inspected to detect any mechanical flaws or stress concentration points which could lead to failure by fatigue. In the original design of the collector (8), the back fin was attached to a 304 steel rod with tack welds. This method of attachment produced a sharp angle with the rod, which served as an incipient crack. Stress concentration at that point led to one occasion of failure by fatigue. To prevent this from reoccurring we have modified the design of the fins; the points of contact with the rod are now smooth and fully welded. We have switched to 304-L steel, which is lese prone to weakening at the weld points. Also, the instrument is now stress- Calibration Mack and Pilie (26) used theoretical results for impac- tion on cylinders to predict the collection efficiency of their rotating arm collector. However, they presented no jus- tification for doing so. Because of the complicated flow inside the cavity, impaction in the slot may differ from. impaction on a cylinder. Experimental calibration is necessary. Scaling Considerations. At high flow Reynolds num- bers and low interception numbers, the collection effi- ciency, », of an impaction surface can be ‘satisfactorily reduced to a function of the three dimensionless groups St, Rep, and Mach number (79). Rep is the droplet Rey- nolds number based on the droplet diameter and the free-stream velocity. When the Mach number based on the local sonic speed is smaller than 0.4, it has a negligible effect on the collection efficiency (40). As a good ap- proximation 9 = f(St,Rep) . (14) In the range 10° < Rep < 10°, which is the range at which the droplets are collected in the slots, 7 is a strong function of both St and Rep (19). To reduce the dependence of 7 to one dimensionless group, Israel and Rosner (28) have proposed a generalized Stokes number which accounts for the variation in the droplet drag coefficient at higher particle Reynolds numbers: ra A(\(D) O° _aRe_ st «(2)(2) 5 CyReyRe =) The generalized Stokes number reduces to the Stokes number for Rep « 1. For higher particle Reynolds num- bers St’ is related to St by St’ = Sty(Rep) (16) where 2 (™ dRe’ vRep) fas J, CaReyRe =.) By substitution of St’ for St, the dependence of 7 on Rep is reduced. (28), so that the collection efficiency is a function of St’ alone, i.e. » = fiSt) (18) Calibration Method. Calibration of an aerosol sam- pling device involves dispersal of a calibration aerosol in a large volume of air under well-controlled conditions and would generally be conducted in a wind tunnel. Since a _ wind tunnel suitable for the rotating arm collector, or even ascaled-down version of the instrument, was not available, an alternate approach was taken. This method, which has { [om ———t ae Blom ss OL 95mm 1 Figure 3. Mode! rotating arm collector used in calibration experiment. recently been used to test sampler inlets (47), allows evaluation of the performance of the instrument under ambient conditions. A tracer gas is used to follow an aerosol plume as it is advected in the atmosphere from an aerosol generator to the collector to be tested. A chemically tagged monodis- perse aerosol is released at a constant rate together with a tracer gas. The collection device is set one meter downwind of the point of release and samples the diluted aeroeol plume. Prior to testing, the concentration of aerosol at the release point is measured. During testing, tracer gas samples are taken at the release point and next to the impaction surface of the sampler. Automatic con- tinuous gas samplers give integrated measurements of tracer gas concentrations at both locations throughout the testing interval. If sedimentation of the particles is in- significant over the distance from the release point to the sampler, the trajectories of the particles and the gas are nearly identical over that distance. The average concen- tration C of aerosol crossing the collector path is then given by c* C# o( =) (19) where the subscript 0 refers to the release point and the asterisk refers to the tracer gas. If a mass m of aerosol is collected by the sampler over a time interval At with a sampling rate Q, the collection efficiency is m y= Ceat (20) By running several tests over # range of St’ values, one obtains the dependency of » on St’. In our tests a monodisperse sodium fluorescein solid aerosol was generated with a Berglund—Liu Mode) 3050 vibrating orifice generator. Sulfur hexafluoride (SF,) was used as a tracer gas and connected to the upstream of the dilution air system of the aerosol generator (for which the flow rate was about 40 L min“) to ensure complete mixing. The aerosol production rate ranged from 10 to 107 g min7, and the flow rate of SF, ranged from 0.43 to 0.68 em? min, Due to the slow rate of aerosol release, the prototype collector had to be scaled down so that sufficient amounts of aerosol could be conveniently collected for analysis. The model collector (Figure 3) consists of two slotted ends screwed onto a sclid rod. It was spun with a variable speed 0.5-hp 10000 rpm motor. Similarity of the Reynolds numbers was maintained, and the aspect ratio was kept as close as possible to that of the prototype. For structural] reasons the width /length ratio for the slots is 0.15 in the model, whereas it is 0.095 in the prototype. The experimental setup is shown in Figure 4. A fan generating a wind speed of about 5 m s' was used to advect the aerosol toward the collector. The experiments were conducted under conditions of low ambient winds. Dilution of the plume from the point of release to the Environ. Sci. Technol., Vol. 18, No. 11, 1964 631 qos SF, eercec! Figure 4. Calbration experiment setup. sampling point, as measured by C*/C*,, ranged from a factor of 30 to 1000 depending on the meteorological conditions. Particles ranging in diameter from 3 to 20 um were generated to cover a wide range of Stokes numbers. The settling velocity of 20 um particles is less than 1 cm s}, so that sedimentation is negligible over the distances of concern. Tests were run over 30-min intervals, which allowed collection of a sufficient amount of aerosol for accurate determination of the collected mass. Two automatic continuous gas samplers, sampling over 5-min intervals, and one syringe pump, sampling over 30-min intervals, were used to obtain SF, samples. Com- perison of the 5- and 30-min integrated measurements provided a check on the SF, levels and raised attention to any major changes in the ambient wind speed or tur- bulence over the course of testing. Analysis was done by gas chromatography using an electron capture detector (42), Samples were diluted to reach the optimal SF, de- tection range (100-300 ppt (parts per trillion)). The aerosol concentration at the release point was measured prior to testing by running a known volume of air through a giass-fiber filter. To restrict the aerosol loading on the mode] arm to the collecting slots, all other parts of the collector were covered with tape, which was removed prior to analysis. A coat of Vaseline grease was applied on one of the two slots to teat against possible particle bounce-off. After testing, the sodium fluorescein was extracted in an ultrasonic bath for $0 min with a 0.1 N NH,OH solution. For the greased slots extraction waa first carried out with toluene for 30 min and then with the 0.1 N NH,OH so- lution for another 30 min. The aqueous phase was then separated by centrifugation. The sodium fluorescein w analyzed by spectrofluorometry. oo Results and Discussion. Figure 5 shows the observed dependence of the collection efficiency on St’ for the scaled-down model. The scale below the abscissa gives the droplet diameter as a function of St’ for a droplet im- pacting at the middle of the slot in the full-scale arm. Although there is some scatter in the data around n = 50%, St'y) (based on the data from the greased slota) is about 5, which corresponds to Dy = 20 um. Ungreased surfaces collect particles with lower efficiency, which indicates that some particle bounce-off occurs. Each collection efficiency measurement corresponds to a range of Stokes numbers because velocities of approach vary from one end of the alot to the other. The spread in the Stokes numbers in the scaled-down model is 17% in each direction off the middle of the slot. In the prototype this spread is 19%, which leads to a 10% uncertainty on the size cut of the instrument. A way to eliminate this uncertainty would be to use a tapered slot, Le., narrow close to the shaft and widening toward the tip, to maintain St’ constant over the whole length of the slot. In ref 28, it is claimed that the use of St’ in interpreting inertial impaction on a surface greatly reduces the de- $32 Environ. Sci. Technol., Vol. 18, No. 11, 1984 T TYTT TT ao - ie a ao “ -_ “ 7 ay - ” a e 7 o~ > “ m or “ 4 ¢ 4 — / — oe “ ” 4 3 / =o c _ ve / _—, ’ —~— | 4 FI er 7 = / ” — ra — -_-..W.. ” om, —_— Oh eT ran 4 oo ” ry wt GENERALIZED STOKES NUMBER s o 8 wBaRK so - Ol pm) Figure 8. Collection efficiency of model colector vs. the generalized Stokes number, for greased (In bold) and ungreased impaction sur- faces. The scale below the abscissa converts the generalized Stokes funder to the diameter of droplets impacting in the middie of the slots of the prototype collector. The dashed curve represents theoretical collection efficiencies for cylinders and spheres (28). pendence of n on the geometry of that surface. Comparison of our calibration data to the curve obtained in ref 28 from previous calculations for cylinders and spheres shows similarity in shape, but our results are shifted toward the higher Stokes numbers (Figure 5). Because the largest discrepancies occur at low St’ values, this is not likely due to particle bounce. Recent studies of virtual impactors (43) have suggested an explanation for the larger St’, in the rotating arm collector. In order to be collected, the par- ticles must not only pass the virtual surface through which there is no mean air flow but also penetrate through the air in the cavity to reach the wall of the collector. Those particles which do not impact on the walls have a signif- icant probability of being reentrained, particularly in the rotating arm collector where there is an induced flow along the length of the collector alot. This may explain the large Stokes numbers required for efficient collection. Comparison with Field Results. Some liquid water content data were obtained at Bakersfield, CA, in Jan 1983 by drawing air with an open-faced Hi-Vol sampler through a paper filter. By measuring the difference in weight of the filters after a certain volume of air has been drawn through the Hi-Vol, one obtains an estimate of the liquid water content. At another location (Albany, NY, Oct 1982) the liquid water content was determined by infrared scattering using a laser transmissometer (44). It must be noted that uncertainties as large as 50% are commonly associated with these two liquid water content measure- ment methods. Figure 6 compares the amount of water collected per cubic meter of air sampled (assuming a sampling rate of 5 m* min=') to the actual liquid water content measured independently. A best fit to the data indicates an overall collection efficiency of 60%. The apparent decrease in collection efficiency when the liquid water content is high may be due to overloading of the bottles before the collector was stopped. To prevent premature overloading due to air lock in the bottles, we have since then added a smal! hole (0.36-mm diameter) on the upper part of the Teflon tube extending inside the bottle. This allows exchange of air in and out of the bottle during sampling and has been found to improve the col- lection characteristica in dense fog. Conclusion The performance of instruments to collect fogwater for chemical analysis can be assessed in the light of the fol- 5 » . ra * a © GAKERSFIELD, CA 7 * 7 eo = ALBANY, BY va ~ a2 7 a ef : gor / ° e ce ° « / e e c a . = /. oe ° oe e EB aw /? 7 e J ‘@ . 8 c ee e es a eo 8 - A 0 ores ee ee * . ° 1 rn n al. 4 L 1 ° aos cx) os e200) as 0.30 oss | Oto LIQUID WATER CONTENT OF FOG {@m=?) Figure 6. Collected water per cubic meter of alr sampled vs. quid water content. Liqui water content was measured with an oper-faced HEVol sampler (Bakersfield, CA) or with an infrared iaser tranemis- someter (Albany, NY). (-—) Linear best fit to the data. (---) 100% efficiency line. lowing design criteria: (a) high collection efficiency for droplets in the size window 1-100 um, (b) collection rate high enough to supply the sample volume required for chemical analysis, (c) conservation of droplet size during approach to the collection surface, and (d) rapid removal of the collected droplets away from the air flow into a quiescent environment. We discussed in depth the design and characteristics of the rotating arm collector used in our ongoing fogwater chemistry field program and concluded that it performs well in preserving the chemical integrity of the collected droplets while providing large sample volumes. These results have been confirmed in the field (6). An in situ calibration indicated a size cut of 20 um diameter, which is significantly higher than that desired. Because our sampler and the jet impactor designed by Katz (7) give consistent determinations of ionic concentrations in fog- water, there appears to be no obvious dependence of the composition on the size of droplets collected. The rotating arm to collect samples representative of the fogwater chemistry in spite of its high lower size cut. Our laboratory is currently investigating several prom- ising designs of fogwater samplers that will collect effi- ciently droplets in the 1~-10-um size range while minimizing - potential sample contamination. The objective is to build a reliable, fully automated instrument suitable for routine monitoring under a wide variety of field conditions. Re- sults will be presented in a future report. Acknowledgments E. F. Daly and J. J. Fontana constructed the collectors and provided many valuable practical suggestions. Com- ments from D. S. Wood, J. M. Waldman, J. W. Munger, and M. R. Hoffmann are gratefully acknowledged. Literature Cited - (1) Whitby, K. T. Atmos. Environ. 1978, 12, 135-159. (2) Waldman, J. M.; Munger, J. W.; Jacob, D. J.; Flagan, R. C.; Morgan, J. J.; Hoffmann, M. R. Science (Washington, - D.C.) 1982, 218, 677-680. (3) Munger, J. W.; Jacob, D. J.; Waldman, J. M.; Hoffmann, M. R. J. Geophys. Res. 1983, 88, 5109-5123. (4) Watson, H. H. Am. ind. Hyg. Assoc., Q. 1954, 15, 21-25. (5) Jacob, D. J.; Waldman, J. M.; Munger, J. W.; Hoffmann, M. R. Tellus, in press. (6) Hering, S. V.; Blumenthal, D. L. “Fog Sampler Intercom- parison Study: Final Report’. Prepared for Coordinating Research Council, Atlanta, GA, 1984, Project ST] 11 90063. (7) Katz, U. “Communications de la 8eme Conference sur la Physique des Nuages”. Clermont-Ferrand, France, July 18-19, 1980. (8) Jacob, D. J.; Flagan, R. C.; Waldman, J. M.; Hoffmann, M. R. Proc. Int. Conf. Precipitation Scavenging, Dry Depos- ition, Resuspension, 4th 1983, 125-136. (9) Junge, C.; McLaren, E. J. Atmos. Sci. 1971, 28, 382-390. (10) Jacob, D. J.; Hoffmann, M. R. J. Geophys. Res. 1983, 88, $611-6621. (11) Bassett, M.; Seinfeld, J. H. Atmos. Environ. 1983, 17, 2237-2263. (12) Houghton, J. G.; Radford, W. H. Pap. Phys. Ocean. Met. Mass. Inst. Technol. Woods Hole Ocean. Inst. 1938, 6 (4). (13) Garland, J. A. Q. J. R. Meteorol. Soc. 1971, 97, 483-494. (14) Mallow, J. V. J. Atmos. Sci. 1975, 32, 440-443. (15) Goodman, J. J. Appl. Meteorol. 1977, 16, 1056-1067. (16) Pruppecher, H. R.; Klett, J. D. “Microphysics of Clouds and Precipitation”; Reidel: Dordrecht, Netherlands, 1978; ’ pp 141-142, 418-421. (17) Roach, W. T. Q. J. R. Meteorol. Soc. 1976, 102, 355-359. (18) Gerber, H. E. J. Atmos. Sci. 1981, 38, 454-458. (19) Friedlander, S. K. “Smoke, Dust and Haze”; Wiley: New York, 1977; pp 95-109. (20) Mrose, H. Telius 1966, 18, 266-270. (21) Okita, T. J. Meteorol. Soc. Jpn. 1968, 46, 120-126. (22) Lazrus, A. L.; Baynton, H. W.; Lodge, J. P. Tellus 1970, 22, 106-114. (23) Sadasivan, 8. Atmos. Environ. 1980, 14, 33-38. (24) Falconer, R. E.; Falconer, P. D. J. Geophys. Res. 1980, 85, 7465-7470. (25) May, K. R. Q. J. R. Meteorol. Soc. 1961, 87, 535-548. (26) Mack, E.; Pilie, R. U.S. Patent 3889532, 1975. (27) Brewer, R. L.; Ellis, E. C.; Gordon, R. J.; Shephard, L. S. Atmos. Environ. 1983, 17, 2267-2271. (28) Israel, R., Rosner, D. E. Aerosol Sci. Technol. 1983, 2, 45-51. (29) Roach, W. oy Brown, R.; Caughey, S. J.; Garland, J. A.; Readings, C. J. Qua.R. Meteorol. Soc. 1976, 102, 313-333. (30) Durham, M. D.; Lundgren, D. A. J. Aerosol Sci. 1980, 11, 179-188. {31) Davies, C. N.; Subari, M. J. Aerosol Sci. 1982, 13, 59-71. (32) Petrenchuk, O. P.; Aleksandrov, N. N. Glavnoi Geofiz. Obs., Leningr. Tr. 1966, 185, 126-132. (83) Mohnen, V. A. Atmos. Technol. 1980, 12, 20-25. (34) Durham, O. C. J. Allergy 1947, 18, 231-238. (35) Perkins, W. A. Stanford University, Palo Alto, CA, 1957, second semiannual! report. (36) Asai, G. N. Phytopathology 1960, 50, 535-541. (87) Hameed, R.; McMurry, P. H.; Whitby, K. T. Aerosol Sci. Technol. 1983, 2, 69-78. (38) Hoerner, S. “Fluid-Dynamic Drag”, Hoerner Fluid Dy- namics: Brick Town, NJ, 1965; pp 3/9, 3/17. (89) Serafini, J. S. 1954, NACA Report 1159. (40) Biswas, P.; Flagan, R C. Environ. Sci. Technol. 1984, 18, 611-616. {41) Hofschreuder, P.; Vrins, E.; van Bozel, J. J. Aerosol Sci. 1983, 14, 65-68. (42) Drivas, P. J.; Shair, F. H. Atmos. Environ. 1974, 8, 1155-1163. (43) Biswas, P., California Institute of Technology, Pasadena, CA, personal communication, 1983. (44) Jiusto, J. E.; Lale, G. G. State University of New York at Albany, Albany, NY, 1983, ASRC-SUNY Publication 869. Received for review August 8, 1983. Accepted May 15, 1984. This research was funded by the California Air Resources Board (A2-048-32) and the President's Fund of the Cakfornia Institute of Technology. Environ. Sci. Technol., Vol. 18, No. 11, 1984 833 CHAPTER II AN INSTRUMENT TO COLLECT FOGWATER FOR CHEMICAL ANALYSIS by Daniel J. Jacob, Jed M. Waldman, M. Haghi, Michael R. Hoffmann, and Richard C. Flagan Submitted to Review of Scientific Instruments (December 1984) 10 ABSTRACT An instrument is presented which collects large samples of ambient fogwater by impaction of droplets on screens. The collection efficiency of the instrument is determined as a function of droplet size, and it is shown that fog droplets in the range 3 - 100 um are efficiently collected. No significant evaporation or condensation occurs at any stage of the collection process. Field testing indicates that samples collected are representative of the ambient fogwater. The instrument may easily be automated, and is suitable for use in routine air quality monitoring programs. 11 INTRODUCTION Under supersaturated conditions in the atmosphere, fog droplets form by activation of condensation nuclei! and rapidly grow to sufficiently large diameters (1 - 100 ,m) that the deposition to vegetation canopies is considerably enhanced”. Recent reports of extremely high acidities in fogs, and clouds intercepting mountain slopes, have raised concern regarding related environmental consequences”. Air quality control agencies in areas exposed to acidic fog have expressed the need to establish networks of sites monitoring the chemical composition of fog on a routine basis. A fogwater sampler to be used in such programs must meet four basic requirements: (1) collect efficiently fog droplets in the 1-100 1m size window while avoiding collection of the submicron non-activated aerosol, (2) preserve the size and chemical composition of fog droplets through all stages of collection, (3) rapidly collect large amounts of liquid water for wet chemical analysis, (4) be automated, inexpensive to construct, and require minimal maintenance. A recent paper from our group“ has elaborated on design criteria for fogwater collectors, and reported that none of the currently available instruments is satisfactory in all respects. A rotating arm collector developed in our laboratory has been shown to provide samples representative of the ambient fogwater”, but it is not suitable for automation and presents safety problems. We describe herein a collector that we have recently developed explicitly to provide a simple, reliable, and inexpensive instrument for routine monitoring. 12 DESIGN The instrument (shown in Figure 1) collects fogwater by inertial impaction of fog droplets on a series of screens. The screens are placed in a duct, through which a blower samples ambient fog-laden air at a velocity U, =9m sec !, The screens consist of 330 pm diameter Teflon monofilament strung vertically between two Teflon-coated threaded rods. The pressure drop across one screen is 87 Pa. The screens slide into position along grooves carved in the wall of the duct, inclined at an angle 6 = 35° (Figure 1b). We have found that inclining the strings considerably reduces the residence time of droplets on the strings (as compared to strings set vertically), because aerodynamic drag pushes the impacted droplets to the bottom of the screen. Further, inclining the strings at 35° prevents resuspension of the impacted droplets into the air flow; such resuspension would occur if the strings were set vertically’. Accumulation of water on the strings proceeds by formation of large droplets spaced at intervals along the strings; these droplets rapidly grow large enough (about 1-2 mm diameter) to flow down. A Teflon funnel drains the fogwater collected at the bottom of the ‘screen into a storage bottle directly below. The instrument is set on a platform surmounted by a wind vane so that the inlet is oriented into the wind at all times. Automation of instrument start-up may be provided by a number of methods. 3 1 The flow of air through the instrument is 22 m~ min » but only a fraction of that air is actually sampled because of the spacing 13 between strings . Based on the dry diameter of the strings, each screen would sample 20 % of the air flowing through; in practice, ‘accumulation of water on the strings will slightly increase the impaction surface area and therefore the amount of air sampled. With four screens, the instrument samples about (1 - (0.8)*) = 60% of the air flowing through the duct. Thus, in a fog of typical liquid water content 0.1 g n> ; 80 ml of sample is collected in an hour. This is sufficient for most analytical purposes. The principle of operation is that inertia prevents droplets approaching the strings from following the curved flow streamlines around the strings. The deviations of droplet trajectories increase with droplet inertia, and droplets above a certain size deviate euf ficiently to collect on the strings. The efficiency of this process is characterized by the Stokes number’, St: 2 oD U,cos 6 St = ie pk (1) where p is the droplet density, D is the droplet diameter, li is the viscosity of air, and R is the radius of the string. The Reynolds number of the flow through the duct is sufficiently high (10°), and the Mach number is sufficiently low (0.03), that the collection efficiency can be satisfactorily described as solely a function of St and the droplet Reynolds number’ ?®, Experimental data are available for the collection efficiency of particles on cylinders as a function of these two dimensionless groups’. When U, differs from the ambient wind velocity Us large droplets cannot follow the bending of the streamlines during approach 14 to the inlet. A collection bias is introduced at the upper end of the size distribution, which depends on the relative magnitudes and orientations of UL and Us and on the inlet Stokes number (defined by R= half-width of the inlet). Inlet biases have been documented in detail’; they are considerably greater in a crosswind than when the inlet is oriented into the wind. In the latter case, the following pattern is found: (a) no sampling bias for UL = O0or U, = U,> (b) negative biases for UL < U,> which go through a maximum for u,/U,* 0.3, (c) positive biases for UL > U;,> which approach an asymptotic limit equal to u,/U, at large inlet Stokes numbers. The collection efficiency of our instrument was determined as a function of droplet size from the data of Refs. 8 and 9, for various ambient wind velocities and an inlet oriented into the wind (Figure 2). The impaction Stokes number was calculated by taking R equal to the radius of the dry strings. Droplets in the size window 3 - 100 um diameter are efficiently collected, while collection of submicron aerosol is entirely avoided. Fog droplets below 3 um are poorly collected, but represent only a very small fraction of the total suspended liquid water. The anisokinetic sampling bias for large droplet sizes depends considerably on the ambient wind speed. Although wind speeds in radiation fogs are in general less than 2m sec, we have frequently found higher velocities (up to 10m sec!) in coastal advection fogs. Positive collection biases require even higher ambient wind velocities, which could be found in clouds intercepting mountain slopes but are unlikely in other situations. Perturbation of the ambient relative humidity during the sampling process may change the size of the fog droplets by 15 condensation or evaporation, and therefore affect chemical concentrations in the sample. Aerodynamic cooling as the droplets are accelerated at the inlet leads to condensation, and aerodynamic heating as the droplets are decelerated during approach to the strings leads to evaporation. To determine the extent of the resulting droplet size modification, we simultaneously solved the equations for flow field, relative humidity, droplet trajectory, and mass transfer, from equations previously derived’. These calculations showed that no significant droplet growth or evaporation occurs for droplets above 1 um. Therefore, droplet sizes are preserved at all stages of approach. Evaporation on the strings must also be investigated. Aerodynamic heating near the strings leads to a localized subsaturation. Assuming that the compression is adiabatic, the maximum temperature gradient at the surface of the string is: AT = > (2) where c, is the heat capacity of air. We find AT = 0.03°K, which corresponds to a decrease in relative humidity of about 0.2 %. More important as sources of thermodynamic modification are the fluctuations of the ambient relative humidity in fog: rapid oscillations of up to 1% relative humidity in amplitude have been reported in a radiation fog!, Mass transfer calculations for a flow i indicate of air at 99 % relative humidity past a bank of cylinders an overall rate of evaporation from the screens of 10°? g min-1. This is small compared to the collection rates achieved. 16 FIELD TESTING ‘The collector, with only one screen installed, was first tested during a stratus cloud sampling program conducted on the mountain slopes above the Los Angeles basin!?, Collection rates were in the range 0.5 - 1.5 ml min !, Four samples were collected simultaneously with the rotating arm collector, and the chemical compositions of the samples were compared (Table 1). Concentrations were not significantly different between the two instruments, and usually within the errors previously documented between two rotating arm collectors set side by side, Exceptions were Na’, ca2t, and Mg2*, which were present at significantly higher concentrations in samples collected with the rotating arm collector. These cations may be substantially contributed by large insoluble soil dust particles collected along with the fog droplets; apparently, the rotating arm collector collects these dust particles more efficiently than the screen collector. After some time of operation, a black deposit was observed on the screens, which suggests that impacted dust particles stick to the strings and are not entrained by the water flow. The good agreement found for water-soluble constituents demonstrates that the screen collector provides representative samples of the fogwater itself. i7 ACKNOWLEDGEMENTS Elton F. Daly and Joseph J. Fontana constructed the collector and offered valuable practical suggestions. This work was funded by the California Air Resources Board. REF ERENCES 1. H. R. Pruppacher and J. D. Klett, Microphysics of Clouds and Precipitation (Reidel, Amsterdam, 1978), p. 141-142, 418-421. 2. G. M. Lovett, Atmos.Environ. 18, 361 (1984). 3. B. Hileman, Environ.Sci.Technol. 17, 117A (1983). 4. D. J. Jacob, R.-F. T. Wang, and R. C. Flagan, Environ.Sci.Technol. 18, 827 (1984). 5. S&S. V. Hering and D. L. Blumenthal, Fog sampler intercomparison study: Final Report to Coordinating Research Council. Available from Coordinating Research Council, 219 Perimeter Center Parkway, Atlanta, GA (1984). 6. J. G. Houghton and W. H. Radford, Pap. Phys. Ocean. Met. Mass. Inst. Technol. and Woods Hole Ocean. Instn. 6, N°4 (1938). 18 7. S. K. Friedlander, Smoke, Dust and Haze (Wiley, New York, 1977), p. 95-109. 8. R. Israel and D. E. Rosner, Aerosol Sci.Technol. 2, 45 (1983). 9. C. N. Davies and M. Subari, J.Aerosol Sci. 13, 59 (1982). 10. H. E. Gerber, J.Atmos.Sci. 38, 454 (1981). ll. R. B. Bird, W. E. Stewart, and E. N. Lightfoot, Transport Phenomena (Wiley, New York, 1960), p. 646. 12. J. M. Waldman, Ph.D. Thesis, California Institute of Technology, Pasadena, California (1985). 19 *__s38[q Jesuyuueq 32 ¥gEl euUNEG ZI UO paqoeT[Oo saTdues 4 jo uosfieduoD (F) ral HZ°OTVE’Z 80°04+40°T ZI°0+66°0 «LZ*OFSO°T §«8Z°OG8°O BT°OZI'T §YyE°OFTI'T ova/os aqea o°HO A TN uy qa aa uot I09T [OD 1a emMBoj 10°0+66°O 90°O+S8°O OT°OFE6°O YI°OFHT'T O°OFIL'O OT*OFOL'O YO°OFTL°O II°O+80°T dVva/9S 7 € 4 -z 08 _’ON _19 + +250 +2 + . 2)°PTs kq apts 398 (OVY) LOJETTOO Wae ZuTIeIOA |YyA puB (DS) AOJDVT TOO usae190s £tano$te3 Tnuzs HN 4) eN H + 20 FIGURE CAPTIONS Figure 1. Screen collector to sample fogwater for chemical analysis. Air is aspirated through a leucite duct at a rate of 22 n° min! (blower: model 80A, 1/3 HP 115V motor, Central Blower Co., City of Industry, California). Fog droplets impact on Teflon FEP monofilament screens (13 mils, Dupont de Nemours Corp.) strung on threaded Teflon-coated brass frames. Fogwater collected on the screens flows down to a Teflon funnel and a polyethylene bottle directly underneath. (a) perspective view of the instrument, (b) side view (section), (c) detail of the screen frames. Figure 2. Collection efficiency vs. droplet diameter. Anisokinetic sampling biases are considered for an inlet oriented into the wind and various ambient wind velocities (m sec!) indicated on the curves. No inlet bias occurs for U, = 0 or UL = 9m sec}. Maximum negative bias occurs for US = 3m sec}. Collection efficiencies in excess of 100% indicate positive sampling biases (U5 < U,)- 21 T aan3ty (9) wri Ogg = sbuiuys jo ew ipuy sed sposy) 2¢ as COLLECTION EFFICIENCY (%) 22 160 -- 140 -- 120 100 ;- 80/-- 60 i al [ao i. a | di ol. | 10° 10° 10! DROPLET DIAMETER (yum) Figure 2. 23 CHAPTER III CHEMICAL COMPOSITION OF FOGWATER COLLECTED ALONG THE CALIFORNIA COAST by Daniel J. Jacob, Jed M. Waldman, J. William Munger, and Michael R. Hoffmann Submitted to Environmental Science and Technology (September 1984) 24 ABSTRACT Fogwater collected at both urban and non-urban coastal sites in California was found to be consistently acidic. The highest acidities (pH values down to 2) occurred at sites in the Los Angeles basin or downwind. Concentrations of acid-neutralizing constituents in coastal air were found to be very low, so that fogs were acidic even under relatively clean conditions. Chloride loss relative to its sea salt contribution was observed at sites furthest from anthropogenic sources. 25 INTRODUCTION Recent investigations of fogwater chemical composition in the Los Angeles basin (1, 2) have revealed high ionic concentrations and, often, very high acidities (pH values typically in the range of 2 to 4). Comparable acidities have been observed in low stratus clouds collected by aircraft over the basin (3) and sampled on the slopes of the surrounding mountains (4). These high acidities have raised concern regarding potential damage to materials, vegetation (5), crops (6), and public health (7). Laboratory data indicate that conversion of SO. to sulfate can proceed rapidly at the concentrations found in urban fog droplets (8) and in the precursor haze aerosol (9). Field studies have proposed evidence for aqueous-phase sulfate production in the atmosphere (10,11,12,13). Fogs are frequent seasonal occurrences along the California coast. During the summer, coastal stations may report over 50% foggy days (14). These fogs are often coupled with land breeze/sea breeze systems, which recirculate the same air parcels several times across the shoreline (13). Tracer studies in the Santa Barbara Channel (15) have shown that emissions from offshore and coastal sources may reside several days along the coast. These humid, poorly ventilated conditions favor pollutant build-up and sulfate production. As part of an extensive fog sampling program in California, we have collected fogwater at a number of coastal sites. Coastal fogs may be a major cause of sulfate pollution episodes in Southern California (16). Further, because impaction of fog droplets can be a significant source of water input to the coastal vegetation (17), the 26 chemical composition of the fogwater has important local implications for acid deposition. Recent interest in these problems has been stirred by federal plans to encourage oil exploration and production in the outer continental shelf off California (18). SAMPLING SITES AND METHODS Fogwater was sampled with a rotating arm collector (19) at eight coastal sites and one island site (Figure 1). The sites, and conditions during sampling, are described in Table 1. Samples were collected over intervals ranging from 30 minutes to 2 hours, and were analyzed for major ions and metals following the protocol described in Ref. 1. Concentrations of Na‘, Kr, Ca2* and Mg?* were determined by atomic absorption spectroscopy on filtered aliquots; at some sites (Del Mar, Long Beach, Lennox, Morro Bay) we analyzed both filtered and unfiltered aliquots and did not observe significant differences in concentrations. Determination of anions by ion chromatography revealed unidentified peaks eluting just before chloride in fogwater samples collected at some urban sites. When chloride concentrations were low, a positive interference from these peaks was noted; chloride concentrations reported are then an upper bound of true chloride ‘concentrations and are indicated as such. The unidentified peaks were likely due to organic acids (20,21). We have measured formate and acetate at concentrations around 10-°M in fog samples (J. W. Munger, unpublished results). The contributions of different source types to the fogwater 27 chemical composition were determined from an emission matrix for primary California aerosol (Table 2). Concentrations of fly ash and automobile exhaust were calculated from the concentrations of vanadium and lead, respectively. Contributions from sea salt, soil dust, and cement dust were obtained by a balance on sodium, iron, and calcium, after subtraction of fly ash and automobile exhaust contributions. Because concentrations of the different elements analyzed in a sample spread over a wide range, we found that this apportionment method gave more reliable results than a least-squares fit of all elemental concentrations to the emission matrix (22). Secondary sulfate was calculated by subtraction of primary contributions from the total sulfate concentration. All nitrate was assumed to be of secondary origin. RESULTS AND DISCUSSION Fogwater concentrations. Table 3 gives liquid water-weighted average fogwater concentrations for each event; the detailed data set is available elsewhere (25). Contributions from the different source types are given in Table 4 in terms of fogwater loadings, which we define as the mass of material in fogwater per mn of air. Liquid water contents were estimated from the collection rate of our rotating arm collector, assuming a 60% overall collection efficiency (19). A number of methods have been used in field investigations to measure liquid water contents in fogs, but none has been reliably calibrated, and a field intercomparison has revealed systematic biases between different methods (4). Although the absolute error on 28 our liquid water content estimates cannot be ascertained at this time, we are satisfied that differences in the collection rate _ closely follow actual differences in the liquid water content (4). In Table 5, elemental ratios in fogs are compared to those for sea salt. The observed Na/Mg ratios were close to that for sea salt; exceptions were the Long Beach, Lennox, and San Marcos sites, where sea salt constituted only a small fraction of the total loading and significant soil dust contributions of Na and Mg were apparent. Calcium and sulfate were partly of marine origin, but usually had larger contributions from dust (calcium) and secondary production (sulfate). The K/Na ratios were close to that for sea salt at Del Mar, San Nicholas Island, and Pt. Reyes; soil dust was an important source of potassium at other sites. The sites in and around the Los Angeles basin were by far the most affected by anthropogenic sources (Table 4). The large contribution from automobile exhaust in the fog samples collected at Lennox can be attributed to the nearby freeway and airport traffic. The sulfate-to-nitrate equivalent ratio at Long Beach Harbor was much higher than that usually observed in the Los Angeles basin (0.2-0.5); non-sea salt (NSS) sulfate concentrations in the basin are known to peak in the Long Beach Harbor because of oil burning by ships (10). The fogs on 7 and 18 December 1981 at Lennox and 7 December 1983 at Corona del Mar occurred during severe sulfate pollution episodes in the Los Angeles basin. Average fogwater pH values below 3 were observed on those nights. On the day following the Corona del Mar event, all coastal stations of the South Coast Air Quality Management District recorded their highest 24-hour sulfate 29 concentrations for 1982, ranging from 23 to 37 pg m>. The prevailing southward transport of pollutants on the night 6~-7 December, 1982 (26) explains the extremely high acidities observed at Corona del Mar; a second sample (1 ml) collected as the fog dissipated had a pH = 1.69 (a value which was later confirmed in the laboratory). High fonic concentrations and acidities were also found in samples collected at Del Mar on 8 January 1983 (Figure 2). The NW winds blowing over Del Mar at the beginning of the sampling period carried pollutants from Los Angeles which had been transported offshore by weak NE winds the previous morning (27). A developing land breeze led to flow reversal between 1900 and 2000 PST and advection over the site of inland air from suburban San Diego County. Sea salt concentrations dropped considerably as the land breeze developed; contributions from anthropogenic sources also decreased, although less, as the Los Angeles air was replaced by cleaner air. Fogwater collected at San Marcos Pass (low stratus), Morro Bay, Mt. Sutro, and Pt. Reyes contained much less nitrate , NSS sulfate, and automobile exhaust than fogwater collected in the Los Angeles basin or downwind. However, except at Morro Bay, all samples were acidic; fogwater pH ranged from 4.21 to 4.69 at San Marcos Pass, and from 3.60 to 5.00 at Pt. Reyes. Significant concentrations of metals and NSS calcium at all sites show that the air sampled was of partly continental origin even under onshore wind conditions. This has previously been observed in aerosol collected elsewhere along the California coast (28). Fogwater collected at Pt. Reyes under offshore wind conditions (12 Aug. 1983) contained less sea salt, more 30 soil dust, and more automobile exhaust than fogwater collected on other nights under the more usual N-NW wind conditions. Nitrate and NSS sulfate loadings at San Nicholas Island were higher than those at other non-urban sites, even though impact of Los Angeles pollutants is very unlikely under the wind conditions observed on that night. High concentrations of metals and NSS calcium indicated that the air over the island was of mixed marine/continental origin. The relatively high fly ash and low automobile exhaust loadings suggest that the high acidity could be due to advection by NNW winds of a plume from oil drilling operations off Pt. Conception or emissions from the Morro Bay power plant; dilution of the plume would be limited by the low mixing heights and slow horizontal dispersion over the ocean (15). Transport from the Pt. Conception area over San Nicholas Island has been previously documented (29). The acidity of coastal fogs in California. Fogwater was often extremely acidic in and around the Los Angeles basin; elsewhere, relatively high “background" nitrate and NSS sulfate concentrations apparently led to ubiquitous acidic conditions along most of the California coastline. Pollutant transport with slow dispersion along the coast has been well documented (13, 15), and has been attributed to the dominant NW wind, the sea-breeze/land-breeze circulation system, the low surface roughness of the ocean, and the persistent strong inversions. Acidities in non-urban coastal fogs were comparable to those found in fogs sampled in the inland San Joaquin Valley of California, 31 where much higher nitrate and sulfate loadings were observed (30). The non-neutralized fraction of the acidity, [a*]/([N0,"]+1(NSs)S0,7"}}, was typically less than 10% in San ‘Joaquin Valley fogs (30); in comparison, the non-neutralized fraction of the acidity was on the average 25% at San Marcos Pass, 312% at Mt. Sutro, and 48% at Pt. Reyes. Obviously, there is little alkalinity available in the coastal air to neutralize acid inputs. Ammonia was found to be the main acid-neutralizing component in San Joaquin Valley fogs (30). An excess of ammonia (H = 50 M atm! 5 > RK = 1.7x10_ M at 10°C) maintains fogwater pH above 5. Fogwater below pH 5, as found along the coast, cannot support gaseous ammonia at equilibrium; ammonia is then expected to be nearly 100 2 scavenged by the fog droplets. The relatively low ammonium fogwater concentrations (as compared to nitrate and sulfate concentrations) show that alkalinity from ammonia is lacking in coastal air. Soil dust is an alternate source of alkalinity, but the small amount present in the droplets (Table 4) limits the extent of H’-neutralizing ion-exchange surface reactions. These reactions would mostly involve the cations Ca2t and Mg?" , but as mentioned previously we found these ions to be mostly dissolved. Alkalinity from scavenged soil dust was therefore exhausted. The low nitrate concentrations observed at Pt. Reyes and Mt. Sutro on 10-13 August, 1982 suggest a source of acidity especially rich in H,S0,. Two likely sources are emission from ships and oxidation of dimethylsulfide volatilized from the ocean surface. Non-sea salt sulfate concentrations of 0.1 - 1 ug n> over the Pacific Ocean have been attributed to the oxidation of 32 dimethylsulfide (31), and oceanic dimethylsulfide production has been found to be maximum near coasts (32). Rapid photo-oxidation of alkyl sulfides in a sunlight-irradiated chamber has been reported (33). “This source of sulfate could therefore possibly be a major contributor to the acidity at non-urban coastal sites in California. Volatilization of HC1(g) from sea salt nuclei. A number of investigators of marine aerosols have observed chloride loss relative to its expected sea salt contribution (34,35). These investigators report chloride losses ranging from 0% (no loss) up to 100%. Chloride loss proceeds by incorporation of a strong acid in sea salt-containing aerosol, resulting in pH-lowering and volatilization of HCl(g). The strong acids can be HNO, (34) or H,S0, (35,36). Displacement by NO, (g) on NaCl(s) has been found to be effective at ppm concentrations of NO, (37). Table 6 is a summary of thermodynamic data regarding the main reactions involved. If the sea salt particle is solid, the reaction may proceed by adsorption of HNO, (g) and sublimation of HCl(g) from surface NaCl crystals, or by coagulation of sea salt aerosol with sulfuric acid or bisulfate droplets. Above the deliquescence point, volatilization of HCl(g) is given by the position of equilibrium (3). Both H,SO, and HNO, added to a NaCl(aq) aerosol will displace HCl, _ but ENO, is less efficient than H,SO, because it is only slightly less volatile than HCl. Hitchcock et al. (35) infer from field data that HSO,— does not displace Cl; however, equilibria (3) and (5) indicate that a substantial fraction of chloride could be released to the gas phase at the liquid water contents typical of haze cio73 Z 33 m>). Although volatilization of HCl(g) proceeds effectively in acid haze, reaction (3) indicates that fog, even acidic, cannot support _HC1(g). This is because of the higher liquid water contents, and because acidities in fogs are not as high as those reached in haze. If chloride was lost in the precursor aerosol, it could still be recovered in the fog by scavenging of HCl({g). Chloride deficiency with respect to its sea salt contribution in the fog therefore requires loss of chloride from the precursor aerosol followed by removal of the resulting HCl(g) from the air parcel during transport prior to droplet activation. The fogs sampled in our study were acidic, and chloride loss in the precursor aerosol would be expected. However, measured chloride concentrations were in excess of the sea salt contribution at most of our sites; this would be due to (1) anthropogenic sources of chloride, (ii) organic acids interfering with chloride in analysis. Significant chloride loss was observed in 1 sample from Del Mar (18%), all samples from San Nicholas Island (12%-35%), and 4 samples from Pt Reyes (10%-28%). Therefore, chloride loss was observed in the fogwater only at those sites where long transport of acids over the ocean was involved. Transport may have led to separation of HCl(g) from the nuclei, possibly by removal of HCl(g) to the ocean surface. Because of its high reactivity with surfaces, HCl(g) should deposit faster than NaCl aerosol. CONCLUSION 34 Fogwater samples collected at both urban and non-urban sites along the coast of California were consistently acidic. Apart from sea salt, the main constituents were sulfuric and nitric acids (partially neutralized by ammonia). Extremely high acidities (pH values down to 2) were observed at sites in the Los Angeles basin or downwind. The acid-neutralizing capacities were found to be much lower in the coastal air than in inland areas of Southern California, and insufficient to neutralize even low acid inputs at non-urban sites. Chloride loss in fogwater was observed at sites affected by acids advected from distant sources, even though hydrochloric acid volatilized by pH-lowering of the precursor sea salt aerosol would be scavenged upon fog formation. A possible explanation is that HCl(g) deposits faster than NaCl over ocean surfaces. ACKNOWLEDGEMENTS . We thank the organizations who provided us with sampling sites: the U.S.Navy, the National Park Service, the South Coast Air Quality Management District, the San Luis Obispo and San Diego Air Pollution Control Districts. This research was funded by the California Air Resources Board (contract No. A2-048-32). LITERATURE CITED ' (1) Munger,J.W.; Jacob,D.J.; Waldman,J.M.; Hoffmann,M.R. | J.Geophys.Res. 1983,88,5109-5123. (2) Brewer,R.L.; Ellis,E.C.;Gordon,R.J.;Shepard,L.S. Atmos.Environ. 1983,17, 2267-2271. (3) Richards ,L.W.; Anderson,J.A.; Blumenthal,D.L.; McDonald,J.A.; 35 Kok,G.L.; Lazrus,A.L. Atmos.Environ. 1983,17,911-914. (4) Waldman,J.M.; Munger,J.W.; Jacob,D.J.; Hoffmann,M.R. submitted to Tellus. (5) Scherbatskoy,T.; Klein,R.M. J.Environ. Quality. 1983,12,189-195, (6) Granett,A.L.3; Musselman,R.C. Atmos.Environ. 1984,18, 887-891, (7) Hoffmann,M.R. Environ.Sci.Technol. 1984,18,61-63. (8) Martin, L.R. in “Acid Precipitation," J.G. Calvert ed., Butterworth, Boston, 1984,p.63-100- (9) Crump,J.G.; Flagan,R.C.; Seinfeld,J.H. Atmos.Environ. 1983,17,1277-1289. (10) Cass,G.R. Atmos.Environ. 1981,15,1227-1249. (11) Hegg,D.A.; Hobbs,P.V. Atmos.Environ. 1982, 16, 2663-2668. (12) Daum,P.H.; Schwartz,S.E.; Newman,L. in "Precipitation Scavenging, Dry Deposition, and Resuspension ," H.R. Pruppacher et al. eds,.,Elsevier, New York 1983,pp.31-44- (13) Cass,G.R.; Shair,F.H. J.Geophys.Res. 1984,89, 1429-1438. (14) deViolini,R. 1974 Techn.Publ. TP/74/1, Pacific Missiles Range, Pt. Mugu, CA 1974. (15) Reible,D.D.; Shair,F.H.; Lehrman,D.E. manuscript in preparation. (16) Zeldin,M.D. et al. Evaluation and Planning rpt. 76-1, Southern California Air Pollution Control District, El Monte, CA, Aug. 1976. (17) Azevedo,J.; Morgan,D.L. Ecology .1974,55,1135-1141. (18) California Air Resources Board 1982 “Air quality aspects of the development of offshore oil and gas resources," Sacramento, CA. (19) Jacob,D.J.; Wang ,R.-F.T; Flagan,R.C. Environ.Sci.Technol. 1984 (in press). 36 (20) Nagamoto,C.T.; Parungo,F.; Reinking,R.; Pueschel,R.; Gerish,T. Atmos.Environ. 1983,17,1073-1082, (21) Keene, W.C.; Galloway,J.N. Atmos. Environ. 1984 (in press). (22) Friedlander ,S.K. Environ.Sci.Technol. 1973,7,235-240, (23) Cooper,J.A.; Watson,J.G.,Jr. J. Air Poll. Control Assoc. 1980,30,1116-1125, (24) Weast,R.C. ed. 1975 “Handbook of Chemistry and Physics ,'"56¢" ed., p.F-199. (25) Jacob,D.J. 1984 Ph.D. thesis, California Institute of Technology, Pasadena, CA. (26) Unger,C.D. “An analysis of meteorological and air quality data associated with the occurrence of fogwater with high acidity in the Los Angeles basin," July 18, 1984 memorandum, California Air Resources Board, Sacramento, CA, (27) Brown,H.W. 1983 “An analysis of air flow over the Southern California bight and northern San Diego county on January 8,1983," San Diego Air Pollution Control District, San Diego, CA. (28) Whitby,K.T.; Sverdrup,G.M. Adv.Environ.Sci.Technol. 1980,9,477-517. ~ (29) Rosenthal,J.; Battalino,T.E.; Hendon,H.; Noonkester,V.R. Techn.Publ. TP/79/33, Pacific Missiles Test Center, Pt. Mugu, CA, 1979. (30) Jacob,D.J.; Waldman,J.M.; Munger,J.W.; Hoffmann,M.R. Tellus (in press), (31) Saltzman,E.S.; Savoie,D.L.; Zika,R.G.; Prospero,J.M. J.Geophys.Res. 1983,88, 10897-10902. (32) Andreae,M.O.; Raemdonck.H. Science (Washington, D.C.) 37 1983,221,744-747. (33) Grosjean,D. Environ.Sci.Technol. 1984,18,460-468. (34) Martens,C.S.; Welosowski,J.J.; Harriss,R.C.; Raifer,R. J.Geophys.Res. 1973,78,8778-8792. (35) Hitchcock,D.R.; Spiller,L.L.; Wilson,W.&. Atmos.Environ. 1980,14,165-182. (36) Eriksson,E. Tellus,1960,12,63-109. (37) Finlayson-Pitts,B.J. Nature 1983,306,676-677. (38) Latimer,W.D. "The oxidation states of the elements and their nd potentials in aqueous solutions,"2 ed.,Prentice-Hall, New York, 1952. 38 Table 1. Description of sampling sites. Site Conditions during sampling Inversion base)tamp.) — gurface wind > (a) Cc) @s”) Del Mar Coastal lagoon, residential surface ll moderate W shifting surroundings. Collector set to moderate E on ground, 800 m from shore. Corona Residential area. Collector 260 ll calm del Mar set on pier. Long Beach Industrial area, ships. Collector 220 10 calm Harbor set on dock, 10 m from water. Lennox Industrial and residential surface 12 INE area. los Angeles Int'l Airport is (7 Dec. 1981) 2 km WW, major freeway is 500 m E. surface 14 1SE Ocean is 4 km W. Collector set (18 Dec. 1981) on roof of l-story building. 220 10 calm ( 7 Jan. 1983) San U. S. Navy base, 100 km offshore. 150 14 5 NNW Nicholas No impact from local sources. Collector Island set at 50 m elevation, 400 m from shore. San Marcos Mountain pass in coastal range, 1800 815 2s Pass 700n elevation. No nearby sources. Morro Bay Rural tow at the base of mijor power 450 ll 0-2 WW ‘plant. Agriculture, ranches. Som local traffic. Collector set 500 m fram shore, on roof of l-story building. Mt. Sutro 250 m elevation hill in San 570 12 l1W Francisco. Radio towers, no local traffic. Ocean is 5 kn W. Pt. Reyes National Seashore, no nearby 600 12 10-15 N sources. Collector set at tip (9 Ag. 1982) of peninsula, 10 m elevation, 420 11-12 10-15 N 50 m from shore. (10 Aug. 1982) 1800 12-14 10-15 N (11 Ag. 1982) 240 11-14 2 SE (12 Aug. 1982) (a) measured by the National Weather Service at San Diego, Los Angeles, Vandenberg AFB, or Oakland. (b) measured at the site. 39 yz °494 (>) €2 °s284 (4) "payloads as msayyo ssajlun zz °4Jad WOuy e3ep (e) Ov 40°0 0 20°0 (2)g-8l qd ¢°0 9 60°T 2°e 0 a4 0 t 0 900°0 (2) ,-01%6 A (209 a ad G*T 9T°T * . TO 9°2 {eyo (Vo (4) 4 (e.g 9°OE PN ysneyxa use Als 3snp ysnp 31es a, iqowoyne [LO Lan quawas LLOS eas ween eee e eww eww een enn eene SSB Y cern n ere rerrreoere +, dazqew azepnapqued JO suo} zRUQUaDUOD aduNOS *Z alge (2) Site Date(*) Del Mar 9 Jan. & 1840-2300 Corona 7 Dec. & del Mar 2100-2300 Long Beach «= 6 dan. 8 0400-0500 Lennox) 7 Dec. 81 205-0840 18 Dec. 81 2315-0045 6 Bn. B 0000-0430 San Nicholas 26 Aug. 82 Tsland 2115-0755 San 2 Aug. & Pass 2A0-1200 “Morro Bay 14 wl. 8&8 0600-0900 Mt. Sutro 13 Aug. & 2125-2225 Pt. Reyes 9 Aug. & 2200-0000 10 Aug. &2 0250-1115 11 Aug. & 0200-1155 12 Aug. @ 0340-0615 (a) date is that of the a.m. samples, or that of the moming following the fog. Time is local tim. 40 Table 3. Liquid water-weighted average fogwater concentrations. 5 nao “WD eM (b) averages for S(IV) and (c) average liquid water cont (f) stratus cloud. pH 2.85 2.16 4.9 2.96 2.66 3.63 3.86 4.49 6.17 3.99 3.60 4,48 3.88 4.60 may be -1 Ht fet 1410 511 20 725 2.7 = & ho 66 2190 131 37 138 6060 2.1 10 0.67 746 102 648 21 B20 3.4 3146 ie 498 a 4 x 148 72 2 1.6 ~ueg | mi, ca mgt cm 1,” 781 452 97 107 183 327 95 L] 8B 49 197 Cs) ili 127 450 242 153 27 3 19 18 2i In 880 782 118 10 on incomplete data sets. 614'9) 1860 1060 7900 z2i'9) 26 178'9)2910 1509) 27@0 63(9) 365 5330 1580 yo 74 im 114 81 = 87 DO 8 4576 3 45 sy) 6 469 1290 487 214 319 1280 463 208 Ee) (g m~), calailated from the tota? volute collected. (d) [C17] may be overestimated due to interference from organic acids during analysis. At Del Mar, this uncertainty is significant only for the last three samples (see text). (e) 1981 events at Lennox have been previously reported by Munger et al. (1983). <M 1 6 NA z pe 4 5 5 78 NA n 196 178 15 0.24 0.11 0.25 0.30 0.14 0.17 0.052 0.43 0.14 0.056 0.054 0.081 0.13 0.19 4l Table 3 (continued) anwar ug 1 ween nwseaan Site = on Fe Mn Pb Cu Ni Y Del Mar 4 354 36 310 NA 111 6 Long Beach 2 96 22. #152 NA 17 1 Lennox (a) 7 Dec. 1981 7 1440‘) 37 1180 34 10 6 18 Dec. 1981 3 1330/, 51 NA NA NA 33 6 Jan. 1983 § 315'P)197 gaz NA 25 (C)5 San Nicholas 5 431 93 49 49 99 8 Island San Marcos 9 22 3 23 12 6 NA Pass Morro Bay 1 192 2. 11 #57 21 3 Mt. Sutro 1 160 10 24 50 28 NA Pt. Reyes 9 Aug. 1982 1 484 67 67 87 209 NA 10 Aug. 1982 3 276 12 21 36 66 NA 11 Aug. 1982 4 301 10 38 151 59 NA 12 Aug. 1982 5 265 7 26 48 17 NA NA: not analyzed. Metal concentrations were not determined in the Corona del Mar fog. (a) average from 6 samples (b) average from 2 samples (c) average from 4 samples ‘42 *7SMP JUAUBI OF PEyNg{J77e SOM &) SN LE “BulSsuW sem eyeP a4 UoLM *Buysspu sem (qy4*A‘a4) SJ80Rs} BA} DedSay J13Yy2 JO SU0L3eJ9UG0U0D ay} UEYM PaUWUAzap OU auaM ySNeyXe Jo ‘yse ALJ *3SMP {10S Woy SUD} INGLUIWO) *qUBAa YRS JO SebeyaAe aue UaAIH SJaquTy “ule JO i! Jad Jazembo} up (eLJa7zeW JO ssa ayy Se peussap sBuypeo, ssey e10°0 6t00°0 9°T lv'0 —-vL0°0 #5°0 zest *6ny zt p10°0 020°0 9°I Ut 82°0 2°S 2e6t “Gry TT £00°0 8£0°0 62°0 ¢6°0 91°0 rai 2a61 “Sry of 0600°0 12°0 $1°0 L°2 8°1 tT Ze6T “Gry 6 Sahay °d ££00°0 91°0 22°0 tl°0 = 0 “2 013NS °W ¢900°0 9800°0 6S°0 46°0 21 It GL Keg 0.104 £20°0 /v0°0 ‘7'0 a 6°I v€°0 SSq SOMEW URS #900°0 £00°0 2v°0 cs°0 ae) 2°S 74 PUR|S] SPLOUPIN UeS 020 z10°0 92°0 oo a O°T 6°E 15°0 £R6T “uer g £20°0 oZ°0 LS 0°6 Gz 2°1 1861 °290 St 29°0 610°0 Ut 9°8 Or V9 08°0 186T °280 Z XOUUT] 60°0 9£00°0 cb’0 95°0 9°5 6°E 21 weed uo} 08°0 9°9 5 19 Jey [ap eud0) 12°0 £20°0 £2°0 8°2 6° 92 0°6 J 120 ysnepe ss yse Aig JSMP quad «= SMP [10S ayeJiNS ByIzJU «= «ABS aS BS Bliquuoyne = tO fang Asepuosas + Ayepuooes ( pall! Bin) Bujpeo, ssau- *Burpeoy Jazemboy 03 SUOLINGL!yUOD BOUNDS *y |aLqGey 43 T20°0 6£0°0 §20°0 €20°0 920°0 080°0 980°0 $92°0 20°0 T2°0 €2°0 61°0 02°0 860°0 810°0 4PN/ 2T°0 62°T cv"0 ST°0 9€°0 6v°0 62°0 6°L 8°1 26°0 y 22N/_2"0S *sHoy [eyseoo ut solqeu qua_eatnb9 £v0°0 £90°0 vS0°0 6v0°0 690°0 vt°O 9T°0 c€°0 pl0°0 G6°0 96°0 “°t €4°0 £2°0 960°0 4PN/, 289 LU°T Le*t O€°T Sv'l 98°0 Te*T 19°T 96°T tm ° oO e Ni ete 9°€ vt 2°T 4 PN/19 €2°0 v2°0 ovo S2°0 £v°0 9€°0 s9°0 20°0 92°0 §2°0 +°N/ 4250 *g 91qel 318s eas ze6t “BGny zt 2861 °Bny 11 2861 “Bry oT 2861 °Gny 6. sahay *I4d ou3ns * WW Keg ousow SSq SO2U0W US pues] SPLOYDIN URS €86T “uer 9 T86T °9°0 8ST T86T °2290 Z xouuay yoeeg 6u07 Jey [ap eu0u0). Jey 190 ats 44 *(8€) uoTAeWIOZ JO sazydfeyque |aaaz WOAZ yQKGZ YS PaIe[NOTed squeqsuod wnysAaqr¢Tinby pol x BL (be) _"osH = (be)_,"os + (be) A "s (Ol K€ (3)fonn = (be)_*on + (be) H *y pO X 9°S (3)1oH = (be)_TD + (be) 8 *€ Ol X 22 (8)°onen + (3)TOON = (8)19eN + (3)? oNz Zz pol x ve (8)°oneN + (8)TOH = (8)toeN + (3)°oNH T queqjsuos wmnpfaqfty[ Faby vo, Ioesy € *worsks OCH-ONH-’0S°H-TOEN 243 JO BFIGELFNbe [eaymayy +9 aTqey 45 pence e e / 1. Conception Senta Barbora * gore Los t RG y e S DEL MAR SAN NICHOLAS 4 ISLAND $on DiegdB_-2 Figure 1. Fogwater sampling sites (MB). 46 uy ve €- °4eq Bap Yoesa Jo doj uo payedzpuyT are (sasayjquosed w 3) squajuod 1ajem4 pynby]T pue sentea yd *1eW [aq 3e sUOCTIeIQUaDUOD 1azeMBO (1Sd SUNOH) SWIL €z 22 Iz 02 “BZ aan3ty 8h YY $8°2 (10) (1/bew) NOLLVYLNSONOD 47 ‘atTdwes 3S8®] |9y3 IOJ paupwaaqap Jou v1am ysneyxe pue ‘yse ATJ ‘4snp TOs WOAZ suoFyngzaquod taeW [aq Je Buppeo, AsIeMZoy ay 02 sucTANGTAQUOD BdANOg ‘qz saAN3Ty (1Sd SYNOH) AWIL v2 ee ce be 02 61 Bt HTT n Oo - Oe 2 Raaannnnnnaeee > AAA m 3aivs ins 9) ma AYVONOD3S — Op rr SIERRA RARER e) SLVULIN bag AYVONODSS > = Livs v3s ES OA +09 ° isna tos NNN. | Ke eee E OO > rLSNG LNANWSD xO} SAaRnnaes -—08 o ARRAS “" ASNMVHX3 xO} 2. MOO ~ — HSVATS x00} SSS + 001 ! i 1 1 a 48 CHAPTER IV KINETICS AND MECHANISMS OF THE CATALYTIC OXIDATION OF DISSOLVED SULFUR DIOXIDE IN AQUEOUS SOLUTION: AN APPLICATION TO NIGHTTIME FOGWATER CHEMISTRY by Michael R. Hoffmann and Daniel J. Jacob Acid Precipitation: SO,, NO, and NO... Oxidation Mechanisms: Atmospheric Considerations. J.G. Calvert, ed., Butterworth, Boston, 101-172 (1984) 49 © Kinetics and Mechanisms of the Catalytic Oxidation of Dissolved Sulfur Dioxide in Aqueous Solution: An Application to Nighttime Fog Water Chemistry Michael R. Hoffmann Daniel J. Jacob Sulfur dioxide can be oxidized to sulfate aerosols either homogeneously in the gas phase or heterogeneously in atmospheric microdroplets [1-3]. Field studies indicate that the relative importance of homogeneous and heterogeneous processes depends on a variety of climatological factors, such as relative humidity and the intensity of incident solar radiation [4-9]. Cass [5] has shown that the worst sulfate pollution episodes in Los Angeles occur during periods of high relative humidity and when the day begins with low clouds or fog in coastal areas, while Cass and Shair [4] have reported that nighttime conversion rates (5.8%/h) for SO2 in the Los Angeles sea breeze/land breeze circulation system are statistically indistinguishable from typical daytime conversion rates (5.7%/h) for the month of July. Liljestrand and Morgan [10], in their study of rainfall in the Los Angeles basin, have reported that light, misting precipitation events resulted in low pH values (e.g., pH 2.9) and correspondingly high SO2- and NO; concentrations. Results of these investigations along with the results of other investigators [6-9,11--13] indicate that aqueous-phase oxidation of SOz2 is a significant pathway for the total transformation of SO». Waldman et al. [14] characterized the chemistry of winter fogs in the Los Angeles basin; they reported that nighttime fog water has extremely low pH values (e.g., pH 2.2) and extremely high concentrations of sulfate, nitrate, ammonium ion and trace metals. Observed ranges reported by Waldman et al. [14] are summa- rized in Table I. Of special interest are the high values observed for SO2-, NO3, SCV), CH2O, and Fe, Mn, Pb and Cu ions in the fog water droplets. These values 101 50 8S7—¥S sz OS7-0E SEC-ISI (:-T-bavl) _20s 91-10 ard | vI-1'9 9 L-6'E PETE (1-3) OFHD 007-01 SIs 06S-SZ1 S1S-€°7 JEI-OL G-1-37/) IN 691-77 OIS‘I-€Z rel OOr-Or OSI-6 SOI-88 (:- 7-371) np 88-Z L61-01 Ors‘z 99€-1vz OOr‘7-878 OPS‘7-O1 EF (i-1-8) qa a da | 6S-S'1 Z18 18-Sz 9S rE G-~T-37/) uw ZEIGT 097'I-+1 00L'EZ 009°9-0rz 080°7-0Z0'I = OL L‘ 1-076 (-1-8n) a4 SUIT LLY-PE Oz7S‘01 OzS‘OI-OLE‘S ~=——- OLS T-0S6 O8£‘Z-062'T G-T-bav) THN StI-1l€ PrS-EIT 090's 000's-0Sz‘Z —- OL6‘I-O1S ¥b6-18b G_T-bar) 208 lLv-+'l £6€‘I-001 000'71 OFI‘s-Orl‘E 069°€-0L0°% —OZS‘*E-O77'T (:-T-bev!) £ON 1Z'S-L9'€ O8'E-S67% =—OTZ L0'€-06°7 18°%-7S'Z $8'b-767 Hd 7861-1861 18-€2-IT sausaatxg 78-bI-1 18-81-ZI I8-EZ-1 ulDy spnoj) Playsaayog xouuaT bDuapospg sajasup soy uO0NDIOT [pI-ZI] se8uey vonesjueou0D poasasgg ‘BIWIO;TED UIIYINOY Ul VOINISOdWOD JaIeA 34 “1 F981 51 CATALYTIC SOg OXIDATION IN NIGHTTIME FOG = 103 and their time-dependent changes [14] indicate that nighttime fogs provide a very reactive and complex environment for the incorporation and transformation of SO, and NO, to their acidic products, H2SO, and HNO. Concomitant incorpora- tion of NHs gas and calcereous dust into the droplet phase results in the partial neutralization of the generated acidity. Later sections of this chapter will discuss models for the detailed thermodynamic speciation and possible reaction kinetics that lead to sulfate and nitrate production in these droplet systems. Because of their similarity to clouds with respect to physical characteristics, fogs are likely to reflect the same chemical processes occurring in clouds and, to some degree, in aqueous microdroplets. Cloud and fog water droplets are in the size range 2-100 xm, whereas aqueous microdroplets will be in the range 0.01-1 pum. On the other hand, raindrops are approximately 100 times larger than cloud and fog water droplets (e.g., 0.1-3 mm). In the Los Angeles study, Waldman et al. [14] found that fog water was more concentrated in the primary constituents than was the overlying cloud water, which was in turn more concentrated than rain water during overlapping periods of time. These results suggest that fog and low-lying clouds may play an important role in the diurnal production of sulfate and nitrate in the Los Angeles basin during certain times of the year when the meteorological conditions are propitious for fog and cloud formation. Furthermore, Hegg and Hobbs [15] have observed sulfate production rates in cloud water over western Washington that ranged from 4.0 to 300%-h™! and pH values from 4.3 to 5.9. The sulfate production rate appeared to increase with an increase in pH. Similar pH values and sulfate levels were observed in stratus clouds over the Los Angeles basin [15]. Historically, fog events have been correlated with severe pollution episodes in which elevated concentrations of SO, and particulate aerosol have been observed [16]. During many of these extended fog periods excess deaths were recorded {17-21]. The London fog of 1952 was particularly bad in this regard. For example, during the London fog of 1952 the daily mass emission rate of SO2 has been estimated to be 1.82 X 10° g, the affected area to be 1.3 X 10° m?, the height of the inversion layer to be 150 m and the liquid water content at 1.23 x 10! g (22]. Given the daily observed increase in the gas-phase SO2 concentration of 0.18 ppm and an established droplet residence time of 0.25 days [18], the sulfate concentration of the fog water can be estimated to be approximately 11.0 meq- L~ with an apparent conversion rate of 12.5%-h7!. This estimated value can be compared to the sulfate ranges observed in the Los Angeles fog water, which were 0.8-3 meq-L~!. However, the London fog lasted for five days while the Los Angeles fogs of 1981-1982 persisted for no more than eight to ten successive hours. AS pointed out in Chapter 1, gas-phase reactions involving hydroxy radical oxidation of SO2 are too slow to account for the very high transformation rates. Alternatively, the catalytic autoxidation of SO, in aqueous microdroplets has been suggested as a nonphotolytic pathway for rapid production of sulfuric acid in humid atmospheres [22-30]. In addition, HzO. and O3 have been given serious consideration as the major oxidants of dissolved SO. as discussed in Chapters 2 and 4. Oxidation by H,O2 seems to be most favorable because of its extremely 52 104 SOs, NO AND NO, OXIDATION MECHANISMS high rate of reaction [31-34] and its pH dependence which favors the reaction at low pH. In comparison, metal-catalyzed autoxidations tend to exhibit decreasing reaction rates with a decrease in pH [27]. Similarly, the reaction rate of O3 with S(IV) decreases with a decrease in pH. Limiting factors in the metal-catalyzed pathways will be the total concentra- tion of the active metal catalyst and its speciation as a function of pH. As shown in Table 1, Los Angeles fog water has high concentrations of Fe, Mn, Cu, Ni and Pb ions. Of these metals, Fe, Mn and Cu are expected to be the most effective catalysts for autoxidation of S(TV) [27,34]. The highest observed concentrations for Fe and Mn were 424 and 14.8 uM, respectively, while the average Fe and Mn concentrations in Los Angeles fog water were observed to be 51.6 and 1.7 pM, respectively. For comparison, Thornton [35] has reported that the worldwide mean concentrations for Fe in urban, rural and remote rainfall are 4.5, 3.1 and 0.13 4M, respectively. In many calculations of the droplet-phase SO, oxidation rate [2,36-39], the assumed Fe concentrations range from 1 to 360 »M and the assumed Mn concentrations range from 1 to 37 uM. At the upper range of the concentration scales, metal-catalyzed reactions may play an important role in the overall SO, oxidation rate in light of the rate laws discussed in Chapter 2. The role of metal ion speciation and kinetic models for fog water chemistry will be discussed later in this chapter. To translate laboratory results on the kinetics of various pathways for SOz oxidation in aqueous systems to atmospheric droplet systems, detailed rate laws, mechanisms, activation energies and ionic strength dependencies should be deter- mined. In only very few cases has this complete information been assembled. How- ever, in the case of metal-catalyzed systems, there are numerous discrepancies among investigators concerning the detailed kinetic information [15]. Metal-cata- lyzed autoxidations can proceed via four distinctly different mechanistic pathways [40]. These include a thermally initiated free-radical chain reaction involving a series of one-electron transfer steps, an inner-sphere metal-sulfite complexation pathway involving a series of two-electron transfers, a surface complexation pathway involving metal oxides and oxyhydroxides in suspension, and photoassisted path- ways in which the oxidation is initiated by absorption of light by S(IV), metal ions, metal oxide surfaces or a specific metal-sulfite complex. The details of these different reaction pathways will be discussed in the next section of this chapter. The mechanism of the oxidation of S(IV) by hydrogen peroxide is fairly well understood [3 1-33,41,42]. This reaction proceeds via nucleophilic displacement of H,O, on bisulfite (HSO3) ion and is catalyzed by specific and general acid catalysis. The significance of this latter feature for open atmospheric systems has been discussed in Chapters 2 and 4. On the other hand, the mechanism of the oxidation of sulfite by ozone [32,43] and its various catalytic influences are less well understood. Most likely, the reaction with ozone proceeds via a free radical mechanism involving the sulfite radical and peroxymonosulfite radical species [32]. The details of the polar pathway involving H2O, and the radical pathway involving Os will be presented in the next section. This discussion will be followed by presenta- tion of a fog water chemistry model that will incorporate and test the various reaction pathways for an open nonphotolytic system. 23 CATALYTIC SO2 OXIDATION IN NIGHTTIME FOG 105 §(IV) OXIDATION MECHANISMS Metal-Catalyzed Pathways for Oxidation Reactions of the triplet ground state of molecular oxygen with singlet ground state reductants such as SO, proceed slowly because they involve changes in spin multiplicity and a large degree of bond deformation or alteration in the formation of products. In many cases, the reactions of O2 with organic and inorganic reduc- tants can be accelerated in the presence of transition metal ions and their complexes. In particular, the first-row transition metal ions have been shown to be particularly effective as catalysts for autoxidation [44]. Catalysis by first-row transition metals may occur homogeneously or heterogeneously in the liquid phase. If the catalytic center involves a soluble metal ion or complex, the reaction can be classified as a homogeneous process; however, if the catalytic center involves a metal surface such as an oxide, oxyhydroxide or sulfide solid, then the reaction can be classified as a heterogeneous process in the liquid phase. Transition metal ions such as Co(II), Co(IID, Cu(II), Fe(II), Fe(II), Mn(I, Ni(ID, VV) (as VO?*) and their soluble complexes have been shown to be effective homogeneous catalysts [44], while solid surfaces such as Fe,O; [45], TiOz [45], CdS [45], ZnO [45], FeOOH [37], Co(OH)s [46], V20; [47], MnO, [48] and MnOOH [48] have been shown to be effective heterogeneous catalysts for certain redox reactions. Numerous attempts have been made to characterize the kinetics and mecha- nisms of metal-catalyzed autoxidation of S(IV) in aqueous solution; however, these studies have been characterized by inconsistent reaction rates, rate laws and pH dependencies [27,40,44,49]. In this section, postulated mechanisms and their result- ing theoretical rate expressions wil! be examined critically. In general, the homoge- neous reaction mechanisms can be broken down into three general categories: (1) free radical chain mechanisms involving a sequence of one-electron transfer steps following a thermal initiation; (2) polar mechanisms involving inner-sphere complexation and two-electron transfer steps; and (3) photoassisted mechanisms. These mechanisms, their rate expressions and the empirical rate laws will be com- pared for consistency. Free Radical Chain Reactions Oxidation of S(TV) in aqueous solution proceeds according to the following simple stoichiometry: H,OSO?-»” + 0.5 02 > H,SO2-” (1) - where n varies from 0 to 2 depending on pH. Backstrém [50] proposed that the reaction of Equation ! involves a chain mechanism in which a metal ion, M"*, of appropriate oxidation state initiates the radical sequence as follows: , k _ initiation M™* + SO2- —— M®-)+ + $03 Q) ; 3 34 106 SO2z, NO AND NO2 OXIDATION MECHANISMS propagation SO; +0,—2 SO; (3) propagation SO; + SO2- 2+ SO2- + SO} (4) oxidation SO?" + SO2- “+ 2802- (5) ‘4 . ks 2803 —*+ $,02- (6) : : = > ke _ termination <~ SO3 + SO; —— 8,02" + O. (7) SO; + SO] —+ $,02- + 20, (8) where the initiation step involves a one-electron reducible metal ion as the catalytic initiator [for a truly closed catalytic sequence, M‘"-))* must be oxidized back to M"*], the propagation steps involve SO3, SO; and SO2- (peroxymonosulfite ion) as reactive intermediates, the oxidation step is invoked for stoichiometric reasons, and the termination steps result in the production of dithionate, a frequently ob- served but low-yield product. Theoretical rate expressions for the overall stoichiometric reaction (Equation 1) G.e., the summation of Equations 2-5) can be derived readily from the appropriate differential equations describing the total rate of appearance or of disappearance of each species involved in the mechanism provided that the steady-state approxima- tion is used. For example, assuming that the chain mechanism involves Equations 2-5 and Equation 6 as the termination step the following rate expression can be derived [51]. —d[SO2-]/dt = d[SO2-]/dt = k,[SO2-}[M"*] (9) + ko(2k:/ks)°5[M"*]°5[SO32-]5[O.] Since the sulfite chain reaction has a chain length of ~50,000, according to Backstrom [50], the long chain length approximation can be invoked (i.e., vp/ vi; >> 1 or the ratio of the rate of propagation to the rate of initiation is very large) and Equation 9 can be rewritten as: —d[SO2-]/dt = k'[M"*]°5[SO2-]°5[O.] (10) where k' = ko(2ki/ks)°* Different theoretical rate expressions can be obtained if assumptions are made about the nature of the rate-determining propagation step and the corresponding termination step. In addition, the steady-state and the long chain length approxima- 55 CATALYTIC SQ2 OXIDATION IN NIGHTTIME FOG 107 Table II. Theoretical Rate Expressions Obtained from the Backstrém [50] Mechanism for Various Termination Steps* ; Rate Termination Step Coefficient, k a B Y ks ko(2ki/ks)}/2 1/2 1/2 1 ke (2k: kok3/ke)?/2 1/2 1 1/2. ky ks(2k,/k7}' ? 1/2 3/2 0 *a, B, and y represent the reaction orders for the reaction rate given by the generalized rate expression —d[SO2-]/dt = k[M"*}*(SO2-]4[O,]’. tions for a free radical, closed-sequence catalytic [51,52] process can be used to simplify the algebraic manipulations. Theoretical rate expressions obtained with these procedures are listed in Table IT [25,26,37,46,53-60]. The theoretically derived rate expressions can be compared to the empirical rate laws reported by other investigators (Table III). Perusal of this list indicates a considerable lack of agreement among these investigators as to the specific reaction orders; in particular, few empirical rate laws agree with the derived rate expressions shown in Table II. The empirical rate laws observed by Barron and O’Hern [66], Chen and Barron [58] and Bengtsson and Bjerile [60] at constant pH are in general agreement with the third rate expression of Table II. However, later results reported by Sawicki and Barron [59] using a thin-film reactor system contradicted earlier results reported by Barron and co-workers [58,66]. Most often the various investiga- tors concur that the reaction order in oxygen is zero. However, it should be pointed out that an oxygen dependence in many cases was overlooked by using pseudo- order conditions in oxygen or, in other cases, mass-transfer limitations may have resulted in an apparent zero-order oxygen dependency. The first and second rate expressions of Table I] show a nonzero reaction order in oxygen and a half-order metal ion dependence, whereas most investigators report a first-order metal ion dependence and a zero-order oxygen dependence. Nonetheless, many of these inves- tigators cite the Backstrém {50] mechanism as a logical sequence of elementary reactions that is adequate to explain their empirical observations. Clearly, in many cases it appears to be inadequate. Finally, a caveat to this comparison is that few investigators used similar analytical and kinetic methodologies and fewer used identical concentration and pH ranges. As a result the lack of agreement is some- what understandable and may suggest that parallel mechanistic pathways are fol- lowed which will result in complicated multiterm rate laws. In most kinetic studies of the metal catalyzed autoxidation of S(IV), SO2- has been identified as the reactive form of S(IV). If this is indeed the case, the apparent pH dependencies in part can be expressed as follows: {SdV)] = [H:0- SO] + [HSO3] + [SO27} (1) 56 108 SOs, NO AND NO2 OXIDATION MECHANISMS Table III. Empirical Rate Laws Reported by Various Investigators for the Metal Catalyzed Autoxidation of SO2°* Mn+ a B Y A Reference Mn?* 2 0 0 53 2 0 0 —1 Chapter 2> 1 1 0 1 Chapter 2° 1 1 0 1 : 54 Sl <i 0 55 1: 1 0 56 Feit 1 1 0 1 37 1 2 0 1 25 1 i-2 ? 2 26 1 1 0 -1 Chapter 2 2 2 0 i/n S7e 1 1 0 I/n 57¢ Co?* 1/2 3/2 0 58 1/2 ? 2 59 Co?* 1/2 3/2 0 —1 60 1 1 1 61 2 1 1 —1 62 1/2 1 0 l/n 63¢ 1/2 3/2 0 64 Cu?* 1 1 0 1/2 65 1/2 3/2 0 66 1/2 3/2 0 46 1 1 0 56 ; *a, 8, y, and A represent the reaction orders for the reaction rate given by the generalized rate expression —d[SO2-]/dt = k[M™*}*[SO2-]4[O.)[H*}4. > Multiterm rate law. © Multiterm rate law where n = 0, 1 or 2. [H20-SO,] 2% [HSOs] + [H*] [HSO5] =: [H*] + [SO] isavy]= (sop) (+ BI +1) 2-1) _ KaiKae [SO3-] = [SV)] (az + KaofH*] + Kaka) Substitution into Equation 10 gives —d{SdV)]_ os a 7 KM tsavyistoa ( KaiKae2 ) [H*}? + K,2[H*] + Ka:Kao (12) (13) (14) (15) (16) 37 CATALYTIC SO2 OXIDATION tN NIGHTTIME FOG 109 Alternative free-radical mechanisms have been postulated [67,68]. Hayon et al. [67] observed the formation of the sulfate radical ion, SO, during the flash photolysis and pulse radiolysis of oxygenated sulfite solutions, and they found no firm evidence for SO5. If the flash photolysis experiments employed by Hayon et al. [67] truly give insight into more moderate energetic conditions anticipated in thermal reactions, the following mechanism can be postulated: k. - initiation Mt + S02? —>M®-)++S03; (17) - ky ~ SO; + O. —> SO; (18) a kh propagation SO; + S02 —> SO; + SO2- (19) S07 + SO 4 s02- + S03 (20) aa - k SO; + SOF —> S,02- + 20, (21) termination 407 + S07 2+ $,02- + O, (22) - - k |SO7 + S07 —- S,02- (23) This mechanism is similar in many respects to the Backstrém mechanism with the exception of the inclusion of the sulfate radical ion as a reactive propagation intermediate. Once again, theoretical rate expressions can be derived from the mechanism above using the steady-state approximation, the long chain length assumption, and the assumption that the rate of initiation is equal to rate of termination at Steady-state (i.e., vi = v, when d[SO3]/dt = d[SO7]/dt = d[SO3]/dt = 0). The resulting equations are listed in Table IV. Regardless of the basic assumptions, Table IV. Theoretical Rate Expressions Obtained from the Hayon et al. [67] and Schmittkunz [68] Mechanisms? ; Termination Rate Mechanism Step Coefficient, k a B Y Hayon et ai. ky k4(2k,/kg}!/? 1/2 3/2 0 Hayon et al. kg or kg k3(2k,/k7)/2 1/2 3/2 0 Schmittkunz kie ko(2ki/ky2)!/? 1/2 1 1/2 *a, B, and y represent the reaction orders for the reaction rate given by the generalized rate expression —d[SO2~]/dt = k{M"*]*[SO2-]4[O,). 58 110 SOs, NO AND NQgz OXIDATION MECHANISMS the mechanism of Hayon et al. [67] yields a single rate expression in which there is a half-order dependence on the metal ion concentration and a three-halves- ' order dependence on sulfite. The same result is obtained if the kz step is assumed to be the rate limiting propagation step. The mechanism proposed by Schmittkunz [68] postulates two separate chain propagation sequences resulting from two different initiation steps. The first se- quence is similar in form to the Backstrém mechanism and the second involves O32, HO,,, -OH and H- as chain carriers. The postulated mechanism is: MtLE- + Op is M*L-D- + 03 (24) initiation - K;? | O37 + Ht == HO: (25) [ O2H- + SO2- , SO2- + -OH (26) k. propagation + -OH +SO2- —> S02" +H: (27) ku k termination 2HO.- ——~H,O.+ O, (29) where M**L2- is a metal(M)}-ligand(L) complex of appropriate charge, x-n, and K, is the acid dissociation constant for the hydroperoxyl radical/superoxide acid- base equilibrium. The second radical pathway involves SO3, SO and SO; as chain carriers accordingly: ky : M**L3 + SO2- ——> M*™"*L3" + $05 (30) - k - SO; + 0. —> SO3 G1) SO; + SO2- 2+ SO2- + SO7 (32) Tv - ki vv — SO7 + SO2- 4+ $0} + SO? (33) ~ ks - 250; —K+ $02 + Os (34) This two-part overall mechanism indicates that there are two transition states and, consequently, a two-term rate law. The rate expression obtained with the . methods used above is 39 CATALYTIC SO2 OXIDATION IN NIGHTTIME FoG” 111 d[SOf"]/dt = 2ko(ki/2k12)°5[M**L3~1J0.5(0,J05[SO2-] (35) + 2k (ki /2ks)°5[M**L3-!os[go2-}'5 In this case, pH dependencies in terms of the acid base chemistry of SO2(aq) have been ignored. A complete analysis valid over broad pH ranges requires consid- eration of the roles of SO2.H2O, HSO3, SOz-H,O* and HSO;- as reactive species. Larson et al. [29] used a hybrid mechanism involving two distinct chain propagation sequences with superimposed acid-base equilibria to obtain a rate expression that agreed moderately well with their empirical rate law. Polar Mechanisms Involving Inner Sphere Complexation Nonradical, polar mechanisms for metal-catalyzed autoxidation of SO2(aq) have been proposed [25,40,55,56,69]. The common features of the mechanisms include inner-sphere complexation of the catalytic metal by sulfite as a prelude to electron transfer, subsequent binding of dioxygen by the resulting metal-sulfito complex, and finally electron transfer via successive two-electron transfers as opposed to a series of one-electron transfer, chain propagation steps. Catalytic Effect of Mn?*, Bassett and Parker [69] found that Mn?* salts had a pronounced catalytic effect on the rate of sulfite autoxidation with almost complete suppression of dithionate formation; that the nature of the Mn?* counterion (i.e., SO2- or Cl-) strongly influenced the reaction rate due to competitive complexation; and that a similar catalytic sequence was observed for Co(II) and Ni(IJ) salts. Based on these observations they proposed the following polar mechanism: Mn?* + $O2- = MnSOs (36) MnSO8 + s02- =e Mn(SO;)2- 7) Mn(SO;)3- + O2 = Mn(SOs),02- (38) Mn(SO3),02- Mn?* + 2S02- (39) A theoretical rate expression can be derived from this sequence of elementary reactions using the vector method of King and Altman [70]. In this procedure the mechanism is written in a cyclic form that represents the closed nature of the catalytic cycle for the overall reaction. The number of intermediate complexes is readily identified and the concentration of each form of the catalyst can. be shown to be proportional to the sums of the terms that are obtained from the 60 112 SOs, NO AND NO: OXIDATION MECHANISMS elementary reaction steps which individually or in sequence lead to the specific form of the catalyst. The number of rate constants (or products of individual rate constants and concentrations of species in excess) must be one less than the number of species in a particular cycle. The steady-state approximation is applied to the resultant equations for each reactive intermediate and the final rate expression is obtained by substitution into a designated rate-limiting step. Using this procedure, the following rate expression is obtained: d{SO3] _ dt 2- (40) k,[Mn?*}.[SO3 ]?[O2] Ka + Kp[O2] + Kc[SO2-] + Kp[SO2-][O2] + Ke[SO2-}? + [SO2-}*[02] where K, = Kj!K3*(Kj! + ke/ks) Kc = K;? (K;! + ki/ks) Kp = ki/ko + K,/ki Ke = Kj! 4+ ki/kg [Mn?*], = initial manganese concentration K,, K2, Ks = equilibrium constants for reactions 1, 2 and 3, respectively This cumbersome expression can be.simplified for certain limiting cases if assump- tions are made about the magnitude of individual terms, as shown in Table V. It Table V. Rate Expressions for Limiting Cases of the Bassett and Parker [69] Mechanism* ee dt k,[Mn?*]o[SO3"]?[02} K, + Kg[Oz] + Kc[SO2-] + Kp[SO2-][O2] + Ke[SO3-?? + [SO}-}7[02] Dominant Term in Denominator a B Y A 1 2 1 B | 2 0 Cc 1 1 1 D i 1 0 E j 0 1 [SO2-}*[02] 1 0 0 "a, B, and y represent the reaction orders for the reaction rate given. by the generalized fate expression = d[SO2-]/dt = k[M"*]*[SO2-}4[0,]”. 61 CATALYTIC SOz OXIDATION IN NIGHTTIME FOG 113 should be emphasized, however, that these limiting cases are idealizations and _ that intermediate cases generally would be expected. From laboratory studies of S(IV) oxidation in aqueous aerosols, Matteson et al. [55] reported that the following rate law was adequate to account for their environmental observations under certain conditions: __ (9[H20-S0.]) ___k{Mn?*]o[H20-SOz]o Vo = ( , Je 7 (41) 1[H20-SOo]}o + k’[Mn?*]o + K Here the concentration of S(IV) is expressed as [H20-SO,] to refiect the condition of high acidity employed in the study. Equation 41 can be reduced to Vo = k"[Mn?*]o[SOz]o (42) for the initial rate when K >> k,[H2O-SOo]o + k'[Mn?*]o. Indication of such behavior based on aerosol conversion experiments has been observed [22,28]. Similar kinetic behavior has been reported by Ibusuki [54] and in Chapter 2 for low concen- trations of S({V). These results have been summarized in Table VI. Complexation by Fe(I/l). Freiberg [71] has proposed a polar catalytic cycle involv- ing inner-sphere complexation of sulfite by Fe(III) as a prelude to electron transfer. His hypothetical mechanism, which was postulated in an effort to obtain an empiri- cal rate law consistent with the experimental data of other investigators (Table VII) is as follows: H,0-SO a2 H* + HSO3 (43) HSOs 42: Ht + $02 (44) Fe(OH)* + H* 22 FeS+ + HO (45) Fe?* + HSOz = Fe(HSO;)?* (46) Fe(HSO;)?* + SO2- = Fe(SO,)sH° (47) ks 4 Fe(SO.)sH* + Oz = Fe(S03):HO3 or Fe(OH)? +2HSOz; (48) -3 2' . k, Fe(SO3)2H® + Fe?+ + H,O —> 2Fe?* + HSO; + HSO; + H* (49) Equation 49 represents direct oxidation of S(IV) by 2Fe(III) to produce S(VI). 62 AW 1 < [+09] W »-01 = L424] W »-01 < [(ADs] Way aysiz1eudg (ues +901 x 1€°9 x etl+zUW}eOl x 2° + t) €-0 Hd 1-8 78'0 = 74 1-L-HI[ADS][+c24]#4 + Z aaideyD OSst=l rsLe= 1-LHiel.cUw)4 =4 W +-01 X9 = [ADS] oO7v-+ 1 Hd bh D0l=L j-UIW- FW 901 X 8 = 4 1-L-HI[-cuna = 4 €¢€-+7 Hd €L Ose=L 1-51-W 201 xX 8—-S = z~etl+zt WA = 4 w4o1 ~ Fos] 1-S1-W 9-01 X TE =? eel?OSI4 + cL cUWI _ 0, ZL WwW OOI-1 = 24] 1-S1-A e-Ol X CET = 9 22Fos)e.24W] j-UNU OQ] = °¥ SUINJOA JOSOJoe = U ;-WIW 770 = 74 Apnys yajdoap __W-,Wd-U = y+ 24 + LUN) + [408] _a ss z> Hd e-W1D-,_UI-;_JOWW OL x oe =F [°"-os]L.24W]'474 Dst=L W »-01 > (ADSI Z sadeyp 7-0 Hd SL =4 1-LHd+c4WH = 4 Ost=1 ¥s ¢i-¢ Hd 1-5-W Ol XS =4 [£0SH]L.24Whi =4 Dost=L1 W +-01> [Caps] z wideyD 70-0 Hd 1-SS7=4 -L-HICADSI[.<4Ww)a = 4 aouasafoy suonipuoD JUuDISUuOD ajDy Op/[(ADS\p— = 4) avy WOIINjOg snoanby ut (APS JO UOHpIKOINY pezyeyed-,,Uj Joy SMET aEY JeonidwY “fA FQUL 63 W -01 > [CADSI W ¢-01 > [Q1D24] £ a.q]e cerlHd «SW sOlxcp="4 ; Cosi eA) + AT1=a ,SW.i0lxst=7 [-zosl4 + °L-cosl(iDes] LS DOl=L SWI XE =O : zl_£Os}LQ1D24)"4 90~Hd [ADs] << [(pea] wsIuByooul poiejnisod poviodai SL wold poaLagd syuejsuoo ON (?0)*4Z/[429d #4 — 1D) & 1-H 402A) 4 = 4 W »-01 <[(ADs] W »-01 < {(1D?a] DOl=L tL €-S' I Hd 1-51-W OF = 4 1-[+H]e(ADS][GiD244 = 4 aposyoa]a 7S 207=1 97 ¢zuHd a(aDsSIHGm24}4 = 4 (¢ Hd) aposzoa|2 205 1-51-W 08 = 4 D07=L (y Hd) 9% pSHd SWI xK8=4 = G-LH)=(-H)y = 9(ADsIGIped4 =4 wut ~ [(ADs] wi | ~ (pe) Le 8-- Hd 1-81-W 001 = (CaDsi[Gipea)a = 4 Wy-01 < [(CADS] Zz soydeyD €-0 Hd 1-8 780 = 4 1-L-HICADSI(G1Dea}4 = 4 ajuadafoy Suoipuoy JupjsuoD any GP/[ADS]p— = 4) avy uouNjOS snoanby ut (ADS Jo UoEprxoyny pozAyEyEd ,,9J JOJ SMUT ayey feodur "TTA A148 1 64 116 SOs, NO AND NOz OXIDATION MECHANISMS Freiberg [71] assumed that Equation 47, the formation of an iron-sulfito complex of 1:2 stoichiometry, was rate-limiting in the catalytic sequence of Equa- tions 43-48. With this assumption, and use of the steady-state hypothesis, Freiberg [71] derived the following overall rate expression: A[SOF] _ krko(2kok«[Oz] + kelFe**][HSOs][Fe?*][SO3] Since [SO2-] = K,2[HSO3]/[H*] = KaiK,2[SO.(aq)]/[H*}? (51) Equation 50 was then expressed as d[{SO7] _ KrK2,[Fe?*][SO.(aq)}? (52) dt [H*}° where Kr= kikoK g2(2ksk,[Oa]/ks + [Fe#*}) T= (53) k-,[(k-2 + k3[O2])/k,) + [Fe*] Equation 50 can be recast in terms of [SO2-] as follows: 2— oO) — pre y[s07- PH) (54) where k'=Ky7/K2. Since the rate-limiting step (Equation 47) presumably involves the slow substi- tution of a second SO2- to the coordination sphere of Fe(H2O0);HSO3*, there should be no observed kinetic dependence on [O,]. In addition to a sequence of two- electron transfers involving oxygen as the oxidant, oxidation of S(IV) directly by Fe* via successive one-electron transfers is predicted thermodynamically (Equation 49) in acidic solution: HO + 2Fe(III) + SO.-H,O === SO2° + 4H* + 2Fe(II) (55) AG° = —115.97 kJ-mol7} However, due to the extremely slow rate of oxidation of Fe(II) to Fe(III) at low pH [48] a loss of apparent catalytic activity would be predicted if Equation 49 were a predominant pathway in the Fe(III)/O2/SO,2 system, unless Fe(II) also exhibits catalytic activity in a fashion similar to Mn(II), Co(II), Ni(iD and Cu(II). 65 CATALYTIC SO2 OXIDATION IN NIGHTTIME FOG 117 In addition to direct oxidation of S(IV) by Fe(III) to produce SO2- {S(VI)}, dithionate S,O2- {S(V)} can be produced according to the following stoichiometry: 2SO0.°H,0 + 2Fe(IID — $,02- + 4H* + 2Fe?*(ID (56) AG° = —38.6 kJ-mol7! The kinetics of this reaction have been studied by Pollard et al. [76]; they have proposed the following mechanism: Fe(III) + SO,-H,O we Fe(II) + HSOs- + H* (57) k, 2HSO,- —> H.S.0¢, (58) HSO,: + Fe(III) “2+ Fe(Il) + $0, + H* (59) SO; + H,0 “++ H,S0, (60) where k,=2.7 X 1074+ M-!=5s71 Likewise, Karraker studied direct oxidation of S(TV) by Fe(II) under the conditions of [Fe(II], > [S(TV)]. and proposed the following mechanism involving initial complexation of Fe(III) by HSO;: Fe(III) + HSOs <= Fe(HSOs)?* (61) Fe(HSO,)?* + Fe(I1) + HSOs- (62) HSO,: + Fe(Il) 2+ Fe(III) + HSOz (63) H,O + HSO,- + Fe(II) 2+ S07 + Fe(II) + 3H* (64) where k, and k2/kg were estimated from kinetic data to be 7 min~! and 22, respec- tively. Further evidence for the formation of discrete iron-sulfito complexes as a prelude to electron transfer has been cited [57,77-80]. Additional work by Carlyle and Zeck [81] suggests that the following noncatalytic mechanism provides a path- way for direct oxidation of S(IV): Fett + H,O <M: FeOH* + Ht (65) 66 118 SOs, NO AND NOzg OXIDATION MECHANISMS Fe? + HSO; == FeSOf + Ht - (66) FeSO} a2 Fe?* + S03 (67) FeOH?* + S03 — Fe?* + HSOz (68) H,0 + FeSO¢ + SO} — Fe?* + HSO;z + HSO3 (69) where | en = 2k’ [Fe(II]? (70) k’ = 1.7 X 10-°fHSO5]/[H*}[Fe2*] + (5.7 x 10-5[HSOs]?/[H*][Fe?*] X (1/0.19 + [SO,]) (7) Catalysis By Cu(/!). Direct oxidation of S(IV) by Fe(III) is sensitive to catalysis by Cu(II) [76,81,82]. In general, the catalytic cycle would proceed accordingly: Fe(III) + S(IV) == Fe(II) + S(V) (72) 2S(iV) —— S,02- (73) Fe(III) + S(V) —— Fe(II) + SO2- (74) Cu(II + S(V) —> Cu(D + $O2- (75) Cu(I) + Fe(HI) st, Cu(II) + Fe(II) (76) Equations 72-76 may constitute an important alternative catalytic cycle for oxida- tion of S(IV) to S(VI). As shown in Table I, fog and cloud water droplets have significant levels of Fe and Cu. At these levels, Cu(II) catalysis of the Fe(IH)- S(IV) reaction may contribute to the overall conversion of SO, in the droplet phase. According to Carlyle and Zeck [81], the catalytic rate of oxidation of S(IV) is given by: —d[S(IV)] at = 2k" {Fe(1I])] (77) where k” = [Cu?*][HSO3][H*] X 1/(14.1[Fe?*] + 2.2[Cu?*] + 0.8[Cu?*}?/[H*]) (78) 67 CATALYTIC SO OXIDATION IN NIGHTTIME FOG 119 The reaction sequence of Equations 72-76 is an example of a redox reaction that is accelerated by a catalyst that exists in more than one oxidation state. Catalysis in this case is due to the fact that Equations 75 and 76, which involve the Cu(ID/ Cu() couple, are fast relative to Equations 72 and 74. The net result is the apparent catalysis of the overall reaction given in Equation 55. In the case of catalysis by Cut, Lunak et al. [56] proposed that autoxidation of S(IV) proceeds via a two-electron polar pathway as follows: Cut + S02- + Cusoz (79) CuSO; + S02 2+ Cu(SO,)3- (80) Cu(SO,)3- + O2 2+ Cu(SO3)203- (81) Cu(SO0,),03- > Cut + 2802- (82) where K, = 10755 The rate law (Table VIII) that results from this mechanism is identical in form to that presented in Equation 40 for catalysis by Mn?*. In the scheme proposed by Lunak et al. [56], Cu?* is reduced rapidly to Cu* by $O2°; Cut then acts as the active catalytic center when complexed by excess S(IV) in solution. These investigators established a relative order of catalytic activity as follows: Fe(II) ~ Co(II) > Mn) ~ Cu(I) ~ CudD >> Fe(II) ~ Co(II) >> Nid) (83) Unfortunately, Lunak et al. [56] failed to report the pH at which these studies were performed. The mechanistic interpretation of Lunak et al. [56] is in direct opposition with the kinetic observations and mechanisms proposed by other investigators [46,65,66]. These investigators favor the free radical pathway of Equations 2-8 in which the Cu?*+/Cut* redox couple acts as an one-electron acceptor Cu? + SO2- —— Cut + $03 (84) However, using the estimated value of E° (—0.89 V) for the SO2-/SO3 redox ‘couple and the E° (0.15 V) for the Cu(ID}/Cu(i) couple, it can be shown that Equation 84 is thermodynamically unfavorable, since Ey° for this reaction is —0.74 V (AG° = 71 kJ-mol-!) [85]. This fact tends to support the hypothesis that the Cu(il)-catalyzed reaction proceeds via the formation of copper-sulfito complexes, 68 120 SQ, NO AND NOg OXIDATION MECHANISMS Table VIII. Empirical Rate Laws for Cu?+-Catalyzed Autoxidation of S(IV) ‘in Aqueous Solution Rate (v= —d[SUV)]/dt) Rate Constant Conditions Reference v= (ki + ks[Cu*)[SO2-] k= 1.3 x 10-73 pH. = 8.7 65 ks = 2.5 x 108 M-4s- [Cu?*], > 10°° M Cu(OH)a(s) T=25C v=0.5 k,[Cu?t]'/2[SO2-}3/2 ok, = 12 < 103 Mots"? pH 7-9 66 E, = 76 kJ-mol"! T=25C v=0.5 k[Cu?*}/[SO2-/2 pH 8.5 46 E, = 78.2 kJ-mol7! T=25C [Cu2*] = 1 .M v= k[Cu?*][SO2-J/(INHIB] k=7.3* 10¢M-+-s7?) INHIB= 56 2-propanal T=20C [Cu2*] = 10 uM pH 8.5 v = k[Cu?*][SO2-] k=2.3 Mos) pH +12 83 (pH > 12) T=25C 1-10 M (NH.OH) v= k[Cu?*}[SO27] 84 which react subsequently with oxygen as suggested originally by Titoff [83] and confirmed later by the results of Reinder and Vies [82]. Metal lan Redox Couples. Metal ion redox couples that would be suitable for 1 @ oxidation of S(IV) to S(V) (i-e., Equation 2) and for initiation of a chain reaction (Equations 2-8), are Co(IID/Co(II) (Table IX), Mn(II)/Mn(il) and V(V)/V(V) with E° values of 1.82, 1.5 and 1.0 V, respectively [85]. The Fe(III)/ Fe(II) couple with an E° of 0.77 V would be insufficient for one-electron production of SO3. Likewise, initiation of a chain reaction by Equation 24, involving a first- row transition metal redox couple, will be thermodynamically unfavorable as shown below. For example, two appropriate half-reactions: O, + H* + lé === HO,- (E° = —0.13) (85) Fe(iI) —= Fe(il}) + 16 (E° = —0.77) (86) can be combined to give an initiation step as envisioned by Schmittkunz [68]: 69 CATALYTIC SOQ. OXIDATION IN NIGHTTIME FOG 121 Table IX. Empirical Rate Laws for Co({III)/Co(I1)-Catalyzed Autoxidation of S(IV) Rate (v= ~—d[SUV)]/dt) Rate Constant Conditions ' Reference v = k{Co(ID}][SO2-][0,] k = 1.5 X 107 M-2-s73 T=20C 61 pH 8-9.2 {Co(iD] < 0.1 pM [S(V)] < 0.01 M _ ki[Co?*].[$037][Oz] k, = 1.96 X 109 M-2-s72 T=20C 86 ve + k»[SO2-] ko = 4.74 X 10? s7! pH 8-8.3 v = k[{Co(H.0)2*}!/2[SO2-]/2 No constant reported E, = 73.15 kJ-mol7} 58 pH 9-9.7 v = k[Co?*})/2(S02-}/4[H*}-2 k~1X 107357! T = 30-60 C 60 kfH*]"? = 103-104 M-)-s"! pH_ 6-7.5 , [Co?*}o <3 pM [SQV)]o < 1 mM v = k[Co(IIL,O2}[SO27] k=1.3 X10! M-1-571 T=22C 87 E, = 112.4 kJ-mol-! v=k,[(CoQDLs)2Oo][SUV)[H*}"! = ki = 6.2 * 104 Mo-}-s71 T=25C 62 p=0.5M pH 0.5-1.3 Fe(II) + O2 + Ht == Fe(III) + HO,- (87) E°=-0.9 AG° = 86.7 kJ-mol7! However, from AG° it is evident that this is an unfavorable process. Similarly, the other one-electron couples involving the soluble metal ions of Co, Mn, Cu and V are unfavorable for the above free radical initiation step. Catalysis by Specific Metal-Ligand Complexes Hoffmann and Boyce [40] have studied kinetics and mechanisms of autoxidation of dissolved sulfur dioxide over the pH range of 4.5-11.0 in the presence of Co(II), Fe(II), Mn(II), Cu(II), Ni(ID and V(IV)-4,4’ ,4”,4 ” -tetrasulfophthalocyanine com- plexes. Phthalocyanine ligands are macrocyclic tetrapyrrole compounds similar in structure to porphyrins that readily form square planar complexes in which the metal atoms are located in the plane of the phthalocyanine ring. Certain metal- phthalocyanine complexes, such as cobalt tetrasulfophthalocyanine [Co(II)-TSP] _ are known to actively bind dioxygen and to serve as reversible oxygen carriers {88,89}. Because of their relationship to naturally occurring organic macromole- 70 122 SO2z, NO AND NOzg OXIDATION MECHANISMS cules, coupled with their high stability, catalytic specificity and oxidase-like activity, metal-phthalocyanine complexes have been shown to be suitable models for studying the catalytic effects of trace metals in aqueous systems [90]. The rate law reported by Hoffmann and Boyce [40] for Co({II)-TSP at pH 6.7 was: —d[S(IV)]/dt = k’[Co(II)-TSP]!/2[S(Iv)] (89) and at pH 9.4: —d[S(IV)]/dt = k"[Co(I])-TSP][SUV)] (90) where k‘! = 1.62 M71'/2.5-1 "= 1.83 X 108 M-1-s71 Based on additional experimental observations and kinetic data reported by Boyce et al. [91], a two-electron transfer, bisubstrate complexation pathway was postulated. The mechanism, which can be classified as an ordered ternary-complex mechanism in the terminology of Laidler and Bunting [92], is as follows: (Co(II)-TSP)§- ets 2Co(I1)-TSP2- (91) Co(II)-TSP2- + SO2- = SO3Co(II)-TSP*- (92) SO3Co(II)-TSP* + On =i SO;Co(II1)-TSPO2- (93) $02 + SOsCo(III)-TSPO.:*- = SOsCo(II)-TSPO,-SOS (94) H,O + SO,Co(II1)-TSPO,-SOS- 2+ SO0,Co(II)-TSPOS- + HxS0, (95) SOsCo(II)-TSPO$- + 2H* —*> SO,Co(II)-TSP4- + HzO, (96) HO, + $02 2+ SO2- + H,O (97) This general reaction scheme is depicted graphically in Figure 1. Equation 92 represents the formation of the reactive monomeric catalytic center, SO;Co(II)-TSP‘-, from the dominant dimeric form of the catalyst in solu- tion. This sequence of events is consistent with observed spectral changes that 71 CATALYTIC SO. OXIDATION IN NIGHTTIME FOG 123 { r bisubstrate oe 01H active catalytic center complex é ‘i f , . meSurQ——| Q, a) a o7 \o% N N N - N Ko le Se / Co(IIT) / —— / co (MI) N N N t N s—o- of / 0 Ht H20 so; + Ht fo) 4 VE “H N N ka Co(H) — CoTSP(OSO," ) + Hp, { N N oo ~sQo- Figure 1. Schematic representation of the proposed catalytic reaction mechanism starting with the Co{II)-TSP sulfito complex as the active catalytic center. Reaction procedes via the formation of a ternary dioxygen adduct and results in the formation of hydrogen peroxide as an intermediate reduction product of oxygen. indicate a shift in the dimer-monomer equilibrium (Equation 91) on complexation of the added substrate in an axial coordination position on Co(II) either above or below the plane of the phthalocyanine complex. Complexation of Co(II)-TSP by SO2- in an axial coordination site enhances the subsequent complexation of dioxygen as written in Equation 93. Carter et al. {93] studied reversible binding of dioxygen by Co(II) complexes of the general form, Co(II)-(L)B, where L is a quadridentate planar ligand and B is an axial ligand, and they concluded that as the z-electron donating ability of B increased the electron density on the cobalt atom would. be enhanced; this in turn results in a greater m-bonding electron flow from cobalt to oxygen. Rollman and Chan [94] reported that the imidazole complex of Co(II)-TSP rapidly complexed O2, whereas the pyridine complex did not. Similar enhancements of the binding of dioxygen by Co(II)-TSP when com- plexed by an appropriate axia! ligand have been reported [90,95,96]. In addition, studies of the catalytic properties of heme have shown that an axial ligand trans to dioxygen affects both the formation constant and the reversibility of oxygen transfer [89,97]. 72 124 SO2, NO AND NOz OXIDATION MECHANISMS Electron spin resonance (ESR) studies [94,95] on Co(II)-TSP/O, complexes have been shown to be consistent with the formulation of the complex as Co(IID- TSPO3. The intermediate complex given in Equation 93 is considered to be a . mixed-ligand Co(III) complex with a superoxide ion (O02) and a sulfite ion bound trans to one another in axial coordination positions. In the proposed mechanism, the O2 adduct reacts with an additional SO2- ion to form a ternary complex as indicated in Equation 94. This complex, in turn, undergoes a rate-limiting two- electron transfer from the second bound sulfite to the coordinated dioxygen-Co(IT) system to form a SO; bound to a Co(II)-peroxide complex. The attached SO; hydrolyzes rapidly to give a coordinated H2SO,, which readily dissociates from the complex. After protonation, the coordinated O2- is released as H2O2. This intermediate HO. reacts with another molecule of SO2- to form SO?- via an additional two-electron transfer in the final step. Hydrogen peroxide was identified positively as a reaction intermediate in this system using three different analytical procedures [98-100]. Schutten and Beelen [98} have observed directly the formation and accumula- tion of H,Oz as an intermediate reduction product of O2 in Co(II)-TSP-catalyzed autoxidation of 2-mercaptoethanol in water. Davies et al. [62] have reported, on the basis of !8O tracer experiments, that approximately half of the oxygen transferred to sulfite comes from a superoxo- complex, (NH3)5 Co(III)-O2-Co(II)NH3*. The other half presumably comes from water. A compatible conclusion as to the origin of oxygen in the oxidized sulfite was reported by Holt et al. [101], who have shown conclusively that the 1*O content of the product sulfate for metal-catalyzed autoxidations of SO2- was linearly related to the '*O content of water, and that at least three of the four oxygen atoms in the sulfate products were isotopically controlled by the solvent water (the remaining atom of oxygen coming from Oz). Finally, Yatsimirskii et al. [87] found strong evidence for complexation of sulfite by the Co(II)-dioxygen complex, Coo{L-histidine)Oz2, as a prelude to electron trans- fer from S(IV) to Oo. In total, these results are consistent with the mechanism postulated in Equations 91-97. The theoretical rate expression, that results from the above mechanism, was obtained by standard procedures [92]; it has the form: v= —d[SCIV)}] ___k'[SOsCo(II)-TSP*“}n[SO3- [02] dt K, + Kg[O2] + Kc[SO2-] + [O2][SO27] (98) where k’ = (kak,)/(ks + ky) Kg = [ka(k-2 + ks)]/[ko(ks + k,)] Keo = (ksk4)/[ki(ks + ka)] When the monomer-dimer equilibrium of Equation 91 is considered, Equation 98 can be rewritten as: 73 CATALYTIC SO2 OXIDATION IN NIGHTTIME FOG 125 ___ kK’ [(Co(Il)-TSP)$-}/2[SO2-][03] K, + Ks[O2] + Kc[SOF] + [O2][SO37] (99) where K’= BK3}/[S502-] when [SO2-], >> [(Co(II)-TSP)$"]. Equation 99 can be simplified for the experimental conditions [02] >> [SO37] such that Ks[O.] >> Ka, Kc[SO37] yw KK UColt)-TSP) SP [80%7] Ks + [SOF] o Two extremes can be considered for Equation 100. If Kp >> [SO27}, v = k'BK4?[(Co(II)-TSP)3"}*/7[SO3-]K5* (101) and if [SO2-] >> Kg, v = k’BK4/2[(CO(MD-TSP)$-}'/2[SO2-} (102) The theoretical rate expressions for the limiting cases given in Equations 101 and 102 can be compared with the experimentally observed rate laws for pH 6.7 (Equation 89) and pH 9.4 (Equation 90). At pH 6.7 a half-order dependence on the catalyst concentration is observed, whereas Equation 101 predicts a half- order dependence on the dimeric form of Co({II)-TSP. Since, at pH 6.7, the dimer is the dominant catalyst species in solution, a half-order dependence on total added catalyst is consistent with the assumption that the monomer is actually the active form. Nonintegral reaction orders arise frequently in polar reactions when the principal reactive species is derived from the dissociation of a dimer [102]. At pH 9.4, a first-order dependence on the catalyst concentration is observed. This result is consistent with the kinetic formulation above if the dimeric form of the catalyst is no longer the dominant species. Cookson et al. [95] see evidence for a shift in the monomer-dimer equilibrium toward the monomer with an increase in pH. If this is the situation, the prior equilibrium can be neglected. Consequently, the theoretical rate expression would show a first-order dependence in catalysts as indicated in Equation 98. Davies et al. [62} and Yatsimirskii et al. [87] have studied the oxidation of S(IV) by the z-superoxo dimers, (NH3)sCo(III)O2Co(II)(NH3)3* and (L-histidine)2 Co(II])O.Co(II)(L-histidine)z. Two mechanisms are possible for these systems. In these mechanisms the 1-superoxo complex, which can be symbolically written as (CoL,)202, is the active catalytic center and it reacts either via a two-electron transfer sequence: . (CoLn)202 + S03- = 02 + Co(L)Ly + Co(IV)LnO (103) 74 126 SOs, NO AND NO» OXIDATION MECHANISMS. Co(IV)L,O + S02- — SO2- + Co(I)La (104) Co(IDL, + O2 = Co(MIDL,Oz (105) Co(ML,OF + Co(I)Ln = (CoLn)sO2 (106) or via a one-electron transfer sequence: (CoLn)2O2 + $027 — (CoLn)202 + SO3 (107) SO; + H,0 = 2H* + SO?- 2H* + CoL,02° = CoL, + H202 (109) Anast and Margerum [103] have reported that copper(III)-tetraglycine reacts rapidly with sulfite in two reversible one electron steps to give copper(II)-tetragly- cine and aquated SO;, which hydrolyzes to sulfate as follows: k Cu(II)(H-sG.)" + SOF = CudINH-sGo* + SOx (110) 1 Cu(III)(H-3G,)- + 803 = Cu(I)H_3G,)?- + SO; (it) -2 SO; + H,O — 2H* + $02" (112) In the presence of oxygen, Anast and Margerum [113] have proposed that the sulfite radical anion reacts with oxygen to give the peroxymonosulfite radical, SO;, which in turn oxidizes the Cu(II) back to Cu(IID) SO; + 02-7 SO; (113) Cu(II)(H-3G,)?> + SO; — CuGID(H-3G.)- + $027 (14) H2O + 2Cu(II\(H-3G,)?" + 802° — 2Cu(III)(H-3G,)~ + S027 + OH™ (115) This mechanism is an example of an induced catalytic reaction in which the Cu(III)/ _ Cu(Il)-tetraglycine couple catalyzes the autoxidation of sulfite to sulfate, which _ can be described by the simple overall stoichiometry of: Cu(lll(H-.G,)" 116) - 2SO2- + Oy Os a502- (116) 75 CATALYTIC SO2 OXIDATION IN NIGHTTIME FOG 127 Values for ki, ko/k-, and k-2/kg were reported to be 3.7 X 10 M~-1-s7}, 1.66 and 177 M~}, respectively. In this reaction system, the oxidation of SO2- by O2 ‘promotes an induced oxidation of Cu(II) to Cu(II) via an autocatalytic process in which there is a net gain in Cu(III)(H-3G,)- when Cu(II)(H-3G,)?-, O2 and SO2- are present initially. . An induced oxidation mechanism of this type represents a conceptual alterna- tive to a free radical chain sequence (Equations 2-8), even though SO3 and SO; are postulated as reactive intermediates. Induced reactions occur when a relatively fast reaction between two substances (A + B — P) forces an otherwise slow reaction (A + C — P’) [103]. According to Edwards [104], the phenomenon of induced reactions provides a method for identifying unstable intermediates in reaction mech- anisms; in this particular case, the unstable intermediate would be SO3. Although similar to the free radical scheme of Equations 2~8, the above mechanism involves the regeneration of the initiator, Cu(III, whereas the Backstrém scheme does not. Synergistic Catalysis Barrie and Georgii [39] and Chapter 2 reported that an apparent synergism exists between Fe5+ and Mn?* in the catalytic autoxidation of S(IV). Barrie and Georgii [39] noted that the addition of either Fe®+ or Fe2* to an equimolar solution of Mn?* in the presence of S(IV) and oxygen resulted in nonlinear increase in the absorption rate of SOz. The net effect of adding iron was to increase the reaction rate tenfold. Chapter 2 examines this apparent synergism in greater detail, and finds that the mixed catalyst system exhibited an overall oxidation rate that was greater than the linear sum of the individual metal-catalyzed rates. This effect was quantitatively described in Chapter 2 with the following multiterm rate law obtained by an empirical force-fit of experimenta! data: —a[SdV)] _ +72) +J]~-1 ' +]-1 ap KIMn®+PIH*] + + k' [FeQISQV)ITH"] ai +k" [Fe(IID][Mn2*][S(IV)][H*]-2/(K + [Fe(IID) Existence of a multiple-term rate law indicates that two or more transition states of different composition are involved in parallel in the overall reaction [102,105,106]. Furthermore, each term in the rate law gives the empirical composition and electric charge of the transition state for the rate-determining step for that particular parallel pathway (with the exception of chain mechanisms) [105,106]. With these fundamental rules of kinetics as guidelines, mechanisms for each term in the above rate can be proposed. Neither Chapter 2 nor Barrie and Georgii {39] proposed a hypothetical mechanism for the Fe-Mn synergism; therefore, mecha- nisms are suggested here for each term. At the concentration levels used in Chapter 2, catalysis by Mn?* alone exhibits a simple second-order rate law in [Mn?*] with zero-order dependencies on [S(1V)] 76 128 SOs, NO AND NOz OXIDATION MECHANISMS and [O.] as indicated in Equation 117. A mechanism consistent with the above term would be: prior equilibrium Mn+ + H,O = MnOH* + H+ (118) rate determining MnOH* + Mn?+ — MnOHMn** (119) Mn,OH** + HSO; =< Mn,OHSO? + Ht (120) Mn,OHSO$ + HSOs == Mn,OH(S0,)7 + H* (121) Mn,OH(SO;)s + O2 == Mn,OH(SO,)203 (122) Mn,OH(SO;),05 2+ Mn,OH** + 2S02- (123) where = ki /k-; = KX (= 107!04 at p = 0.5 M and T = 25 C) [107] k2/k-2 = 104 (at p = 0.5 M and T = 25 C) [107] With the assumption that Equation 119 constitutes the rate-limiting step in the mechanism and that kz > k-2, the following rate expression can be derived: _ ~afSCV)I Sp Keun PIT (124) 17 This rate expression agrees well with the first term of the empirical rate law. For the second term of the rate law, the following mechanism can be postu- lated: eee eaten @ Fe(Iil) + HSOs <2 FeSO3 + H* (125) ee Fe(II)S(1V)O3 ais Fel)s(vyo3 (126) Fe(IDS(V)OF + HSO;Z ate Fe(1I)(S(V)O3)(SO3)" + H* (427) Fe(ID(S(V)O3)(SO3)- + O, = Fe(III)(S(V)O3)(SO3)O3 (128) Ku Fe(1I1)(S(V)O3)(SO3)03 —— Fe(V)(SO3)O* + SO2- (129) 77 CATALYTIC SO. OXIDATION IN NIGHTTIME FOG 129 -Fe(V)($0,)0* 8 Fe(II) + $02- (130) In this mechanism, the first step, which involves the slow complexation of Fe?* by HSO; is rate determining; consequently, the theoretical rate expression that re- sults is as follows if it is assumed that [H*] = Kai (.e., ~ pH 2) and k; >> k-z: vex —d{[S(IV)] _ k7Ka: ; dt 2 [Fe**][SQV)][H*}> (131) where —_[S(IV)] = [HSO3] + [H20-SO.] Ka = [H*][HSO5}/[SO2(aq)] _ The overall kinetic coefficient kop; = k7Ka)/2 can be estimated from known kinetic data on ligand substitution [108,109]. Using the water exchange rate constant [110] for Fe**+, key = 10? s"! and a calculated ion-pairing constant [111], K. = 1 M~'! for the Fe’+:HSO3; ion pair, a value for k, (~10? M~}-s~1) can be obtained. At p = 0, Ka, = 0.0123. Using these values, the estimated value of kops is 0.62 s! at T = 25 C, pH 2 and » = 0. This value can be compared to the value of 0.82 s~! reported in Chapter 2. However, if k7 is estimated from the value of the overall second-order rate constant [112] for the complexation of Fe?*+ by SO2- (k; =~ 1 X 105), the value of kon; would be predicted to be 6.2 s~! at 25 C, pH 1.8 and 1 = 0. These estimates can be made in this fashion because the second- order rate of ligand substitution has been shown to be controlled most often by the water exchange rate of the central metal and to be relatively independent of the nature of the ligand [113]. The third term of the rate law indicates that Fe(III), Mn(II) and S(IV) are brought together in the transition state. One possible mechanism that would result in such a combination involves the formation of a reactive Mn(IV) intermediate from the interaction of the ferry] ion—-sulfite complex [114] with Mn?" as follows: 2H* + Fe(V)(SO3)O* + Mn(I) mtn Fe(III)SO3 + Mn(IV)+H,O (132) H.O + Mn(IV) + HSO3; a Mn(II) + HSO; + 2H* (133) where the rate-limiting step is the two-electron oxidation of Mn(II) by Fe(V). The activation energy for this mechanism should be significantly less than the two previous pathways. This conclusion is consistent with the low apparent activa- tion energy reported by Barrie and Georgii [39] for the synergistic effect of Fe?* on the catalysis by Mn?*. Alternatively, a one-electron oxidation of Mn?* to Mn°* could be envisioned: Ky - Fe(V)SO,0* + Mn2* == Fe(IV)SO,0 + Mn3* (134) 78 130 SOs, NO AND NO2 OXIDATION MECHANISMS. This would be followed by a chain sequence (Equations 2-8) initiated by Mn**: Mn?* + HSO3; aie Mn?* + HSO3; (135) The redox potential of the Mn?*/Mn?* couple (E° < 1.5 V) is sufficient to promote the one-electron oxidation of S(IV) to S(V)(E° = 0.9 V) and truly initiate a radical chain sequence. The rate law for this process would be one of the general forms given in Tables II or IV. This latter mechanistic possibility could readily explain the nonlinear concentration effects observed in Chapter 2 and by Barrie and Georgii [39]. Finally, a third possibility, involving the formation of mixed hydroxy-bridged complexes of Fe(III) and Mn(II), can be considered. FeOH?* + Mn?* a Fe(OH)Mn‘** (136) Fe(OH)Mn‘** + 2HSO; = Fe(OH)Mn(SOs3)2 + 2H* (137) Fe(OH)Mn(SOs3)2 + O2 = Fe(OH)Mn(SOs3)202 (138) Fe(OH)Mn(SO3)202 — Fe(OH)Mn‘* + 2502- (139) The rate expression obtained from this mechanism would have the following form if complexation of HSO3 by the hydroxy-bridged species is assumed to be rate limiting: v =k [Fe(IID)][Mn(ipD][S(V)][H*}? (140) Similar expressions involving the concentrations of Fe(III), Mn(II) and S(IV) are obtained for the other mechanisms given above. The rate expression of Equation 140 is not exactly of the form shown in Equation 117, but it does reflect some of the essential features. Photoassisted Catalysis Homogeneous Processes The catalytic effect of light [50,62,115-125] on autoxidation of dissolved SO2 has _ been acknowledged for years as an alternative mode of initiation of the chain reaction sequence given in Equations 2-8. UV irradiation of sulfite/bisulfite systems is thought to produce an aquated electron [62,121] and a sulfite radical ion: h - + SO2- —— SO} + ena 79 CATALYTIC SO. OXIDATION IN NIGHTTIME FOG 131 A transient electronic absorption maximum at A = 700 nm has been assigned to the solvated electron and a similar transient maximum at A = 260 nm has been assigned to the SO; radical ion. The aquated electron decays by reaction with O. or HSO; to produce a superoxide ion or sulfite ion and a hydrogen atom, respectively. However, only Lunak and co-workers [126-128] have systematically explored the possibility of metal-catalyzed photooxidation of sulfite, even though the phenomenon of homogeneous metal-assisted photooxidation/reduction pro- cesses has been reported previously for Fe(III) and Cu(II)-ligand systems [128- 135]. , ' The preliminary work of Lunak and Veprek-Siska [126] has shown a distinct photoassisted catalysis by Fe(III at wavelengths higher than the cutoff for absorp- tion by sulfite and bisulfite ion in aqueous solution. Quantum yields were shown to depend on the wavelength of irradiation and the concentration of Fe(II) in the system. They have proposed that a quantum of light is absorbed by a ferric- sulfito complex, thereby initiating the catalytic cycle through photoreduction of Fe(III) to Fe(II). In the absence of added Fe(II], the photochemical autoxidation of sulfite above A = 300 nm did not proceed to a noticeable extent. A mechanism consistent with their findings would be: Fe(OH)S-™* + HSO; = Fe(OH),SO?-)* + H* (141) Fe(OH),SO?-")* + HSO; = Fe(OH),(SO3)"*)~ + Ht (142) Fe(IID)(OH)n(SO3)*P- , *Fe(II)(OH)n(SO3)(SOg: ot - (143) *Fe(II)(OH)n(SO3s)(SO3°)"t)~ = Fe(IID(OH)nSO8- + SO3 (144) or - Fe(III)(OH)n(SOs)"*)- + O2 = Fe(IID)(OH)n(SO3),0"*9- (145) Fe(I11)(OH),(SO3)209*+- + *Fe(11I)(OH)a(SO.)$"*?- (146) 2H* + *Fe(IID(OH)n(SO,)*)- = Fe(HII(OH)S-™* + 2HSOZ (147) After photoinitiation, autoxidation would proceed via a free radical chain after Equation 86 or by inner-sphere electron transfer as shown in Equations 145- 147. Lunak and Veprek-Siska [127] also report that photoinitiated autoxidation of sulfite is influenced by the presence of Mn(II) at trace levels (10-&-107!9 M) and that Cu(II) [125] does not exhibit a photocatalytic effect. On the other hand, Hoffmann and Boyce [40] have reported that cobalt(IDtetrasulfophthalocyanine catalyzed autoxidation of S(IV) is strongly influenced by visible light in the range 80 132 SOs, NO AND NO, OXIDATION MECHANISMS of 300-400 nm. They have interpreted their results in terms of formation of singlet oxygen due to a ligand-to-metal charge transfer absorption. Heterogeneous Processes A primary source of iron, manganese, zinc, titanium and other metals found in aerosol systems is from combustion of coal. These metals will be released to the atmosphere in their highest oxidation states in the form of metal oxides such as Fe,03, MnO, ZnO and TiO. Taylor and Flagan [135] and Ouimette and Flagan [136] found direct evidence for significant mass fractions of FeO; in submicrometer particle size fractions of fly ash from a coal-fired combustor and from ambient aerosol. Barrie and Georgii [39] have used average values of Mn and Fe in urban aerosols of 0.2 and 1 jzg/m, respectively, and an average liquid water content of 0.1 g/m? to estimate that the soluble concentration in the aqueous phase is on the order of 10 »M. In making these estimates they assumed that 25% of the total Mn was soluble and 10% of the total iron was soluble. The remaining solid phases are likely to exhibit some catalytic activity [137-141]. Frank and Bard [45] reported that Fe.O, (hematite) surfaces catalyze photoas- sisted autoxidation of HSO; to HSOj. Irradiation at 400 nm of an oxygenated bisulfite solution with suspended FeO; particles resulted in a rapid conversion of SCV) and S(VI), whereas in the absence of light no conversion of sulfite to sulfate was observed. A sequence of stoichiometric reactions that offers one possible explanation for photoassisted catalysis on a metal oxide (semiconductor) surface is: Fe,03 + 2hv SUT8e*, (Fe,O,) + 2p* + 2e7 (148) H,O + HSO; + 2p* “"*, Hso; + 2H* (149) O + 2e- + 2H* 5, 4,0, (150) H,0, + HSO; 5 Ho; + H,O (51) where p* = charge vacancy created by the photoassisted excitation of an electron from the valence band of the metal oxide to the conduction band Excited electrons in the conduction band and positive holes in the valence band migrate to the surface (presumably without recombination), where they react, re- spectively, with oxygen and bisulfite to produce hydrogen peroxide as the reduced product and bisulfate as the oxidized product. Similar, but less effective, photocatalytic effects have been reported by Frank and Bard [45] for other semiconductor materials such as TiO, CdS and ZnO. 81 CATALYTIC SO2 OXIDATION IN NIGHTTIME FOG 133. However, results for CdS and ZnO are obscured by the fact that at low pH (pH ~3) these solids dissolve to a certain extent; thus, any apparent photocatalytic effects may be due to dissolved metal species rather than due to inherent properties of the semiconductor material. Faust and Hoffmann [142] studied photocatalytic autoxidation of HSOz on synthetic Fe,0; suspensions at pH 2.5; they report that the presence of Fe,0;3 particles results in a slight rate enhancement as compared to the corresponding homogeneous photocatalytic system. Childs [143] and Childs and Ollis [144] have addressed some of the fundamen- tal considerations that are required to ascertain whether a heterogeneous photoreac- tion is truly photocatalytic according to Equations 148-151 or stoichiometrically photochemical. In a stoichiometric photochemical reaction, the observed conversion of sulfite to sulfate in irradiated systems may result from (1) direct photoactivation of an adsorbed sulfite on the surface; (2) a surface-prompted reaction of sulfite photoactivated in solution; (3) a noncatalytic reaction of the FegO3 surface with the adsorbed sulfite; or (4) homogeneous catalysis by Fe(III)/Fe({I) leached from the FeO surface. These questions remain to be resolved in the case of iron-catalyzed systems. It is apparent [37,58,65,83,86] that the rate of autoxidation of S(IV) is most rapid in the concentration-pH domain where metal hydroxides are the dominant species such as Ni(OH)2z, Co(OH)s, Co;04, FeOOH, Fe(OH)3 and Cu(OH)e. For example, Brimblecombe and Spedding [37] have proposed that the oxyhydroxides FeQOH (a-FeQOH, 8-FeQOH, y-FeOOH) would be effective initiators for the production of SO3 (since E°(FeQOH) = 0.91 V) as follows: Fe(OH)as) + SO5 — FeOQOH(.) + SO?" + H* (153) Soot particles have been shown to be effective catalytic surfaces for autoxida- tion of S(IV) by Brodzinsky et al. [138], Chang et al. {139] and Benner et al. [145] although Cohen et al. [140] reported that the apparent catalysis by ambient fly ash particles was due to Fe(III) leached into aqueous solution. Furthermore, soot and powdered activated carbon surfaces provide moderately effective sites for adsorption of trace metals and these adsorbed metals may in turn be the active catalytic centers. Ambient soot contains a significant level of polynuclear aromatic hydrocar- bons (PAH). Certain PAH have been shown to be photosensitizers for the produc- tion of singlet oxygen, which will in turn react electrophilically with an oxidizable substrate rather than through a free radical pathway [146]. A sensitizer, such as _ PAH, is a molecule that can absorb a quantum of light and then transfer this - energy to a second substrate. For example, benzophenone is an efficient producer of 102 (' Ag) [141]: 1(CoHs)aC = 0 "> (CyHs):C = O* (154) 82 134 SOs, NO AND NO» OXIDATION MECHANISMS 1(CgHs)2C = O* — 3(CgHs)2C = O* (155) 3(CgHs)2C = O* + 302 > (CgHs)oC = O + 102 (156) Mn(SO3)2- + #02 -— Mn?* + 2802- (157) which in turn could oxidize S(IV) to S(VI) by a sequence of two-electron transfers. Energy transfer from an excited triplet state sensitizer to 30, (Equation 156) to give 4Oz2 is a spin-allowed process [146]. Fridovich and Handler [148] have shown that photosensitized dye molecules such as methylene blue and flavin mononucleotide are effective catalysts for autoxi- dation of S(IV); however, their early work was interpreted in terms of a free radical sequence rather than through production of 102 as follows: 158 dye , dye* (158) dye* + HSO; — dyeH + SO3 (159) dyeH + O, — dye + HO,: (160) Later work by McCord and Fridovich [149] showed that the autoxidation of S(IV) as catalyzed by dimethyl sulfoxide was inhibited by superoxide dismustase. This result indicates that HO. was a likely intermediate. However, in metal-catalyzed reactions, superoxide dismutase had no effect. In light of more recent work [150], the reaction observed by Fridovich and Handler [148] most likely involved the production of Oo. Srinivasan et al. [150] have shown that irradiated solutions of the dye rose bengal produce both Oz and 'O2, which are quickly consumed by SO? at pH 7.8. They have estimated that 77% of the active oxygen generated is 10, and 23% Oj. Additional work by Srinivasan et al. [150] has shown that irradiated solutions of Ru(2,2’ bipyridyl)?+ produce Oz, which leads to rapid oxidation of S(IV) Ru(bipyy* —> *Rubipy)?* (161) *Ru(bipy)?* + O02. —— Ru(bipy)*t + OF (162) Oz + H* = HO; (163) HO,- + HSO; = SO3 + H202 (164) Much work remains to be done on photocatalytic systems in aqueous systems to ascertain the role of condensed-phase photochemistry on total S(IV) conversion rates. If photoassisted pathways are shown to be important under ambient condi- 83 CATALYTIC SO. OXIDATION IN NIGHTTIME FOG 135 tions, differences between nighttime and daytime conversion rates may have a plausible explanation. Oxidation by Hydrogen Peroxide, Ozone and Nitrous Acid: Catalytic Influences As indicated in Chapters 2 and 4, hydrogen peroxide appears to be a potentially important oxidant in atmospheric droplets because of its favorable Henry’s law constant (Ky = 1 X 105 M-atm7)), its rapid oxidation of S(IV), and its relative insensitivity to pH changes in an open two-phase system. Hydrogen peroxide is generated in the gas phase by the recombination of hydroperoxy] radicals, in the aqueous phase as an intermediate in metal-catalyzed autoxidations, and at air- water interfaces due to photoinduced redox processes [151]. Measurements by Kok and co-workers {152-—155] indicate that steady-state aqueous-phase concentra- tions can be as high as 50 pM. The kinetics of oxidation of S(IV) by H,O2 have been studied [31--33,41]. Mader [41] determined that over the pH range of 8-13 the rate of oxidation of S(IV) was characterized by a multiterm rate expression as shown in Table X. The first term accounts for uncatalyzed oxidation, the second term for general acid catalysis, and the third nominally for specific acid and other forms of unspeci- Table X. Empirical Rate Laws for the Oxidation of S(IV) by Hydrogen Peroxide Rate (v = —d[(SUV)} /dt) Rate Constant Conditions Reference v = ko[H20,][SO2-] ko = 0.2 M-1-s7! T=25C 41 + k,[H*][H2PO7]{H202][SO2-] kp =1.7X 10° M-3-s-? p=1.0M +r, E,(k,) = 63.5 kJ-mo!7! pH 7.9-12.8 v =k{H*}[H.0,][S(IV)] k=2.7X 10° M-2-s7? = =T=12C 31 {([H*)/({H*] + Kas)} k'=1.3X 10° M251 p=1.0M + kTH2PO;7][H202][S(IV)] pH 4-8 ((H*)/((H*] + Kaz)} v = k[H,O2][HSO3]{H*}°? k= 2.46 X 10° M-l-s-) = T=20C 32 + k,[H20.][HSO;][CH;CO.H] (pH 4.6) p=0.2M k, = k[H.O.]{H*}°? + k,{H.O.J[HA] k,=1.9 10¢M-2-s-1 pH48 E,(k:) = 24.9 kJ-mol7! (pH 4.6) y = K[H202][HS05][H*] k=5.6X 106 M-t-s-2 - T=25C 33 K+[H*] K=01M pH 0-3 p (variable) py = K{H20z][H20 S02] k=8.3X10¢M-4-s"! = T=25C 33 K + [H*] K=0.1M pH 0-3 E, = 28 kJ-mol™? p (variable) 84 136 SOz, NO AND NOz OXIDATION MECHANISMS fied catalysis. In addition to phosphate buffer catalysis, Mader observed carbonate .and arsenate were effective as general acid catalysts. Mader offered no mechanistic interpretation of his kinetic results. Earlier, Halperin and Taube [42] had suggested that the oxidation proceeded through the formation of a peroxymonosulfurous acid intermediate, which was followed by an intramolecular rearrangement to give sulfate and water. Their con- clusions were based on double-isotope labeling experiments that showed that both of the labeled oxygens of H!8O!8OH were incorporated to the product sulfate, even though the stoichiometry of the reaction requires the net addition of one oxygen atom. H,0, + H,0-SO.— 2H* + SO2- + H,O (165) H!80!80H + H,O-SO,— HO,S'8080H + H2O (166) HO,S'80'80OH — 2H* + SO!803- (167) In a subsequent kinetic study of this reaction, Hoffmann and Edwards [31] found definitive evidence for both general acid and specific acid catalysis. They reported the following two-term rate law for the pH range of 4-8: Ee = k{H*][H.02][S(V)][H*]/({H*] + Kas)| (168) + k'THA][H202}[S(V)]{[H*]/(FH*] + Kaz)} where K,2= second acid dissociation constant {S(IV)] = sum of the concentration of sulfite and bisulfite within the pH range of 4-8 HA = weak acids used over this pH range as buffers The first term of the rate law reflects the contribution of specific acid catalysis and the second term reflects the contribution of general acid catalysis to the overall rate. Two alternative mechanisms that are consistent with the observed rate law and the isotopic labeling results of Halperin and Taube [42] are: _ Ka, H* + SO2> == HSO; (169) o- - KON HSO; + H.0, === Tu — OOH + H,0 (170) oO 85 CATALYTIC SOg OXIDATION IN NIGHTTIME FOG = 137 o- \ k _ S-OOH + HA ——~ H.S0,+ A (171) Oo” where HA =H,0* or a suitable weak acid such as phosphate, citrate, pivalate, acetate or phthalate According to this mechanism, the reaction occurs via a nucleophilic displacement by H202 on sulfur of HSOz to form a peroxymonosulfurous acid intermediate (Equation 170), which then undergoes a rate-determining rearrangement to HSOz assisted by H,;O+ or HA. Alternatively, a mechanism in which there would be an equilibrium protonation of the bisulfite followed by a proton transfer to A7~ _can be written as: Ka H* + SO2- == HSO; (172) HSO; + HA <= H,0-SO, + A- (173) bn ow A- +H.02 + H20-SO2 se )S00H +HA+H,0 (174) slow oO o- \s.00H — Hs0z (175) 0 / rapid In this sequence, Equation 174, a termolecular step, in which A~, H2O. and H,0-SO, react to form peroxymonosulfurous acid, is rate-determining. In the first alternative mechanism (Equations 169-171), HSQ3 is the reactive species of S(IV), and in the second mechanism (Equations 172-175) H2O-SO: is predicted to be the reactive species. The work of Martin and Damschen [33] appears to verify that HSO; is in fact the reactive S(IV) species and that the mechanism of Equations 169-171 is valid over a broad range of pH (i.e., 0-8). Penkett et al. [32] studied peroxide oxidation kinetics over the pH range 4.3-8.2 using acetate, citrate, phosphate and Tris buffers to maintain pH. They confirmed the role of these weak acids as general acid catalysts. However, Penkett et al. report a hydrogen ion dependence of [H*]°” over the pH range of 4-6 as noted in Table X. On the other hand, Martin and Damschen [33] report a first- _ order dependence on H* over the pH range of 1-3 and a reciprocal dependence below pH 1. This switch in pH dependence supports the notion that HSO; is ' the reactive form of S(IV) in solution. For the pH range of 1-5, Martin and Damschen [33] expressed the rate. 86 138 SOs, NO AND NOzg OXIDATION MECHANISMS law in a form that is convenient for application to open atmospheric systems as follows: —d[Sd SES = 8.3 X 10° [H,0.][H20- S02] (176) or —d[Sdv end = 8.3 X 10°Py.0.Ku.H202Ps02Kn.s02 (177) The concentrations of H2O2 and H2O0-SO, are expressed in terms of the products of their respective Henry’s law constants and partial pressures. They also report that the oxidation reaction is apparently insensitive to the effects of metal catalysis (Fe3+ and Mn?*) and organic inhibitors (hydroquinone, toluene, pinene and hexene). However, in the case of H2,O.2, Fe?* and Cu?*+ would be the most likely metal ions to exert a catalytic effect on the reaction rate since these metals are well known initiators of the Fenton’s reagent reaction [156-158]: , Fe?*+ + H,02 — FeOH?* + HO: (178) Cu?* + H,02,— CuOH?* + HO- (179) H* + HO- + HSO; — HSO;- + H2O (180) If oxygen were present in the reaction system a free radical pathway could be initiated by Equation 180 and proceed via Equations 3-8 or 18-23. Additional propagation steps can be envisioned to occur as follows: HO- + H,0.— HO,: + H:O (181) HO,- + HSO; — HSO; + HO- (182) Future work on the H202-S(IV) reaction system should address the possibility of catalysis by Fe(II) and Cu(II). There are few reports of catalytic influences on the oxidation of S(IV) by ozone. Barrie [159] has found that Mn?* promotes the oxidation of S(IV) by ozone in aqueous droplets. In the absence of Mn?*, gas-phase ozone alone failed to enhance SO, absorption rates in the aqueous phase. Chapter 2, in the section on ozone oxidation kinetics, reports no apparent catalysis due to Fe3+, Mn?+, VO?* or Cu?*; ’ however, Harrison et al. [43] reported that additions of Mn** and Fe>* at concentra- tions of 10 pM resulted in enhancements in the observed oxidation rates by factors of approximately 2 and 6, respectively. Addition of competitive complexing agents (phosphate or citrate) inhibits the apparent catalytic process. A maximum in cata- 87 CATALYTIC SO2 OXIDATION IN NIGHTTIME FoG 139 lytic influence occurred at pH 4.5. This behavior suggests that ferric-sulfito and manganous-sulfito complexes play a role in the oxidation process. Fe3+ O3 + HSO; ua HSO; + O2 (183) in + Furthermore, Harrison et al. [43] indicate that the rate of oxidation. increases linearly with an increase in Fe?+. These results are summarized in Table XI. Penkett [160] has proposed the following initiation of a radical mechanism to account for oxidation by ozone: 20; + OH” — HO- + O32 + 202 (184) HO- + HSO; — HSO;- + OH- (185) The bisulfite radical of Equation 185 would proceed to react according to Equations 3-8, thereby completing the free radical sequence of the Backstr6m mechanism. Based on isotopic tracer experiments, Espenson and Taube [162] concluded that in acidic solution the reaction between ozone and sulfur dioxide could occur via a simple bimolecular process in which two atoms of oxygen are transferred from ozone to sulfur dioxide or via a radical! reaction involving reactive intermedi- ates or via both pathways in parallel. In basic solution, they concluded that the free radical pathway was more likely even though more than one oxygen atom was transferred from ozone to sulfite. Table XI. Empirical Rate Laws for the Oxidation of S(ITV) by Ozone Rate (v = —d[SUV)]/dt) Rate Constant Conditions Reference v = k[Os][HSO3][H*]-°5 k= 1.45+0.7 X 104 M-05.571 pH 1-5 32 T=25C v= kPo,Ku,0,[HSO3][H*]-°" k = 4.4 X 104 M-09.571 pH 47 29 Kyo, = 1.23 M-atm™} T=25C v = k[Os][S(IV)][H*]-°5 k= 1.9 X 10¢ M-°5-s71 pH 0-3 Chapter 2 T=25C v = k,[HSO5][O.] + k[SO2-][O3] ky = 3.1 X 105 M-}-s71 pH 0-4 160 kp = 2.2 X 109 M-}-s7! T=25C E,.1 = 46 kJ-mol“! E,.2 = 43.9 kJ-mol™ v= k[Fe?*][S(IV)][Os][H*]" Kops = 6.8 X 10° M-!-min7!-ppm7!_ T= 295 K 42 pH 4-6 [Fe**] > 10 pM 88 140 SO2z, NO AND NOg OXIDATION MECHANISMS If a polar pathway were to occur as suggested by Espenson and Taube [162], a likely mechanism would be as follows: : [6] 0% No o” No oO 0% No \sZ% + H,O — HSO; + H* + O2 (187) 0” No Catalysis by Fe?* and Mn?+ may proceed through an initial metal sulfite complex followed by attachment of ozone to form a ternary complex. Fe? + H,O + FeOH™ + H* (188) FeOH** + HSO; 2+ Fe(OH)SO2 + H* (189) Fe(OH)SO2 + 0; 2+ Fe(OH)SO,02 (190) Fe(OH)SO,02 — FeOH?* + SO2- + 0, (191) Finally, S(IV) in aqueous systems can be oxidized readily by N(III) and N(IV) species in the form of nitrite, nitrous acid and aquated nitrogen dioxide. Nitrous acid will oxidize S(IV) in acidic solution according to the following stoichiometries: . 2HNO, + H,O-SO, — SO2” + 2NO + 2H* + H20 (192) AG° = —160 kJ-mol™! 2HNO; + 2H20-SO,2 — 2S02- + N,0 + H20 + 4H* (193) AG° = —431 kJ-mol“! Martin et al. [163] determined that the reaction was first-order in N(III), S(IV) and H*, and insensitive to catalysis by Fe?*, Mn?* and VO?* in the absence 89 CATALYTIC SQ2 OXIDATION IN NIGHTTIME FOG 141 —a[SdV)] a KEHNO.b{SV)I[H*] (193) | where [HNO2]; = [HNO] + [NOz] ; k= 142 M-2-s"! They found that NO was the principal product at low pH. Oblath et al. [164] found at higher pH (pH 5-7) that hydroxylamine disulfo- nate formed according to the following stoichiometry: NO; + 2HSO; — HON(SO;)3- + OH™ (194) Ross et al. [165] observed that a significant fraction of HNO, is converted to NO according to Equation 192 at higher acidities. They found that, in the presence of O2, an apparent catalytic cycle involving NO is established. The stoichiometric equations for this sequence are as follows: 4NO(aq) + O2 + 2H20 — 4HNO,z (195) 2HNO, + H20-SO.2 — 2NO(aq) + H* + HSO;z + H20 (196) Latimer [85] suggested the following mechanism for the reduction of nitrous acid by S(IV): HONO + HSO; —— HSO;NO + OH- (197) H,0 + HSO;NO — NOH + HSO; + H* (198) 2NOH — N20 + H20 (199) NOH + HONO — 2NO + H2O (200) HSO3;NO + HSO;s — HON(SOs;)2- + H* (201) Oblath et al. [164] suggest the following mechanism: HONO + HSO; = NOSO; + H2O0 (202) NOSO; + HSOz = HON(SO3;)2- (203) NOz + $:02- = ON(SO3)3- (204) H* + HONO = NO* + H,0 (205) 90 142 SO2z, NO AND NO OXIDATION MECHANISMS NO* + HSO; = NOSO; + H+ (206) which involves the formation of a nitrosonium ion, a fairly stable ionic species. In acid solution, aquated nitrogen dioxide (N(IV)O2(aq)) readily dimerizes to form nitrogen tetroxide, as reported in Chapter 4 and by Gritzel et al. [166]: 2NO,(aq) = N20,(aq) Ka=7X 10M" 207 Nitrogen tetroxide, in turn, is a moderately strong oxidizing agent [84] with E° values of +1.07 and +1.03 V, respectively, for the following reactions: N20, + 2H* + 2e7 = 2HNO2 (208) N2O, + 4H* + 4e~ = 2H,O + 2NO (209) Takeuchi et al. [167] reported that the reaction of S(IV) with NO2(aq) proceeds rapidly according to the following stoichiometries and rate laws: 20H- + 2NO,(aq) + S03" > HzO + 2NO3 + SO2- (210) i = k,[NO,(aq)][SO2-] (21) H,O0 + 2NO, + HSOs ~~ 2H* + HSOz + 2NOz (212) a = ke[NO.(aq)][HSO5] (213) where t=25°C k, = 6.6 X 105 M~}.s73 ko = 1.5 K 104 M-1-s7} Nash [168] described the role of Fe(II) catalysis in the above reactions in the nanomolar range. As pointed out in Chapter 4, translation of the above rate laws for closed systems to open atmospheric systems requires that the concentrations of the impor- tant gases be expressed in terms of the product of the appropriate Henry’s law constant times the observed partial pressures. In the case of HNO, and NOz, ‘their respective Henry’s law constants are 49 and 0.01 M-atm™!. These moderate -values of Ky indicate that the partial pressures of each gas-phase species will determine to a large extent the contribution of the above pathways to the overall rate of S(IV) oxidation. 91 CATALYTIC SO OXIDATION IN NIGHTTIME FOG 143 EQUILIBRIUM SPECIATION MODELS To use the preceding kinetic information reliably in dynamic models of SOz transfor- mation in the atmosphere, detailed chemical speciation in aqueous solution for reductant, oxidant and catalysts must be determined. Using data collected on the concentrations of chemical constituents in fog, cloud and rain water, equilibrium speciation models can be developed readily with the aid of the computer programs REDEQL2 and SURFEQL. Both REDEQL2 and SURFEQL are general-purpose programs for computation of multiple, simultaneous chemical equilibria involving acid-base, oxidation-reduction, precipitation-dissolution, complexation-dissociation and adsorption-desorption reactions [169-172]. As an example of the applicability of equilibrium models, we will consider the potential role of iron-sulfito complexes and sulfonic acid derivatives in increasing the apparent capacity of an aqueous droplet system for S(IV); the levels observed in fog and cloud water reported in Table I appear to be above those predicted by the observed partial pressures Total S(IV) in the aqueous phase is given by: S(IV)r = [SOz]eq + [HSO5] + [SO2-] (214) _ Ki | Kiks S(IV)r = KuPso, ( + (H] + Ay) (215) _,_(, , Ki, Kike as (1 tiny? i) (216) where Ky=Henry’s law constant K, = first acid dissociation constant for H20-SO. K,= second acid dissociation constant for H2O:SOz Using the appropriate values for Ky, K, and K2 [107], total [S(IV)] can be calculated to be approximately 1 weq-L~! at 16 ppbv SO, and pH 3. However, if the reactions of S(IV) with Fe?+ and CH.O are considered the following expressions are valid: K HSO; + CH,O = HOCH,SO; (218) K, Fe3+ + SO2- = FeSO? (219) S(IV)r = [SOz]aq + [HSO;] + [SO2-] + [HOCH,SO;] + [FeSO#] (220) 92 144 SOs, NO AND NOz OXIDATION MECHANISMS KK Tye tp (CHO) + SS [Fe**)) Ki, S(IV)r = KuPsog( i+ (H*] (H*? [H+] In this case, for Ks = 10° M-} and K, = 1012 M7}, the [S(IV)]- will be in the range of 100-600 py eq-L~ at 16 ppbv SO, and pH 3 if the observed concentrations of CH2O and Fe are used as initial input values and approximate values for the equilibrium concentrations of CH,O and Fe** are determined iteratively. These computations can be performed readily using SURFEQL. To illustrate the applicability of this modeling approach, a sensitivity analysis on the value for K, was performed. Carlyle [77] has reported a value of K, = 10° while Hansen et al. [173] report a value of K, = 10!8!. The actual value may lie somewhere between these two values. Metal-sulfite complexes are notably strong compared to the corresponding metal-sulfate species, e.g. Bo(Hg(SO3)2-) = 10%; B.(Hg(SO,)2-) = 104. Results are shown in Figure 2. These results suggest that, if K, > 10*°, 50% of the total iron in a system initially 0.1 mM in Fey will be complexed at FeSO} over the pH range of 2-4. Furthermore, if K, > 1045, the computer model predicts that all of the Fe will be found as the species FeSO}. Similar computations can be Fe + SO; = FeSO,* [Fe], = 0.tmM 100 Fe(OH)s + FeSO3 80 8 Fe>* x Fe(OH}o ™ 60 a Feci2* wv — HOCH, S03 > 40 +m fe} n 2 “, 20 0 Figure 2. Speciation of Fe(III) vs pH in aqueous solution as a function of the stability constant of FeSOj as determined with SURFEQL. Other species that form are listed on the right side of the figure. 93 CATALYTIC SO2 OXIDATION IN NIGHTTIME FOG 145 S(IV) SPECIATION HCHO + HSO3- === HOCH2SOz . 5 HOCH, S03 (K=10°) 100 “| Feso;* (8,= 10") MnSO3 (8, =10°) Bok 2 pe 5 HOCH, S05 [scmy) ],= 1.0 mm D [HCHO |, = 1.0 mM ” 60 4 2 [Mn |. = 0.1 mM 40 : oe SO2* H20 cor FeSO; Mnsoe | ee ee ac ae Ry 0 | 2 3 4 5 6 Figure 3. Speciation of S(IV) vs pH for the conditions on the far right of the figure as determined by SURFEQL. performed on Mn?* and HCHO. Figure 3 shows the total speciation for S(IV) over the pH range of 0 to 6 given the conditions listed on the right side of the figure. In aqueous aerosol! systems, important factors to consider are the nature and roles of dissolved organic molecules that can act as competitive complexing agents for metals. For example, liquid-phase autoxidation of benzaldehyde produces benzoic acid, which can act as a suitable complexing agent (e.g., pKa = 3.97, log; = 1.51 for Cu(C;H,_O2)*) and a similar oxidation of 2-hydroxybenzaldehyde to 2-hydroxybenzoic acid produces even a stronger potential ligand (e.g., pKa: = 2.78, logBi. = 10.13 for Cu(C;H,O,)*). The presence of complexing agents of this type and organic reductants will accelerate the dissolution of Fez03; and MnOz, which are the likely sources of soluble iron and manganese in aerosol systems. As shown by Cohen et al. [40], the catalytic activity of soot-derived aerosols corre- lates well with the total iron released to the liquid phase. To illustrate this effect, oxalic acid/oxalate was added to the computation illustrated in Figure 2 at a concentration of 1.0 mM. Results are shown in Figure 4 for the effect of 1:1, 1:2 and 1:3 ferric oxalate complexes with logB; = 7.5, logB2 = 13.64 and logBs = 18.49 on the equilibrium distribution of FeSO}. The net effect of the addition of oxalate was to lower the percentage contribution of FeSO} as a function of pH for a given stability constant. This effect is due to competitive complexation. 94 146 SO2, NO AND NOz OXIDATION MECHANISMS me Nd = | [Fe], 0.1mm @ [serv], = Lome 2 50 [ox], 0.1 mm | ood oO Ee ES 25 K=10" 1 a ee ae 1 ° 6 1 2 3 4 5 pH Figure 4. Speciation of Fe(III) perturbed by iron(IIDoxalato complexes as a function of pH and stability constant (8,) of the FeSO}. An additional effect that was not considered here is the possible formation of mixed ligand complexes of Fe3*, S(IV) and oxalate. As indicated above, certain metal-catalyzed autoxidations of sulfite proceed via the formation of discrete inner-sphere complexes between the reductant, SO2- and the catalyst as a prelude to electron transfer. Few metal-sulfite stability con- stants have been determined because of the thermodynamic instability of sulfite toward oxidation in oxic systems; however, those that have been reported [107] are significantly larger than the corresponding constants for metal-sulfate complexes as shown in Table XII. Sulfite can form complexes by forming a bond either Table XII. Comparison of Stability Constants®* for Metal-Suifito and -Sulfato Complexes at 25.0 C Metal B log B (SO2-) B log B (SO?-) Agt 0 AgSO3, 5.6 ) AgSO;, 1.3 Ca?+ 1.0 Cd(SO;)?, 4.2 1.0 Cd(SO,)3", 1.6 Hg?* 1.0 Hg(SO3)2-, 24.1 0.5 Hg(SO,)2-, 2.4 Ce3+ 1) CeSO, 8.0 4) CeSOZ, 3.6 Fes+ 0.1 FeSO3, 18.1° 0 FeSOj, 4.0 Fe2+ 0 FeSO, 2.2 uo? 1.0 UO,SOs, 5.3 1.0 UOSO,, 1.8 Cut 1.0 CuSO3;, 7.8 Cut 1.0 Cu(SO,)3-, 8.7 Cu2+ : 0 CuSO, 2.4 4 All constants reported in this table were taken from Smith and Martell [106], except the stability constant for FeSOf. > Hansen et al. [172]. 95 CATALYTIC SO2g OXIDATION IN NIGHTTIME FOG 147 through sulfur [174,175] or through oxygen [176-178] whereas sulfate is restricted to bond formation through oxygen. Part of the increased metal-sulfite bond stability may be due to the availability of low lying “d” orbitals on sulfur for pa-da and d-d7 metal to ligand backbonding. In addition to this added stability, sulfite can readily form bidentate [179] and bridging complexes with bond formation through two oxygen atoms or through an oxygen and sulfur. Based on simple linear free energy correlations, reasonable estimates for the stability constants of Fe?+, Mn?+, Cu?* and Ni?+ complexes with sulfite can be made using the data given in Table XII. With this information, models: for the speciation of fog water (Table I) can be constructed with the aid of SURFEQL. These computations are particularly important in the case of Fe°+ since it can exist as a solid (Fe(OH); or FeQOH) over a broad range of pH. Results of a speciation model for the chemical constituents and concentrations listed in Table | are illustrated in Figure 5 for Fe3+, STV), Mn?* and Cu?*. Iron(III) above pH 4 is predicted to occur exclusively as amorphous Fe(OH);; below pH 4, Fe%* is complexed by SO2- (logB, = 12) and by F-. Only a small fraction of total iron appears as free hexaquo iron(III). This fraction increases at pH <2. Copper and manganese speciation are relatively straightforward. At the micro- Fe(I/ S(T) U T T T T O8F fe) QQ T oO a x nN 3 wy \ | | | | | iN o6B- FeSO3 FRACTION OF TOTAL Fe (II) \ a FRACTION OF TOTAL S{I¥) oO Lo} oO nN b n T \ 4 i” / pH 1 Mni It) } T T T Teo _ “ =) 3 a S aN s OS Fe + COBR cu2* 4 = ¢ 2 a oe / MnSO3 a L M 4 = 4 © 06 n / 06 S x 5 z O4- / 4 2zoae f 2 / 2 CuNH 3* / f oa Z q / 2 , EFF / 7 EO2r- CuSO CuNHy)oX% M y uS0q u(NH3 )o i a O Lee ee 0 eed goo 1 2 3 4 5 6 1 2 3 4 5 6 pH pH Figure 5. Speciation of Fe(III), SAV), Cu(II) and Mn(IJ) in fog water as a function of PH. Input values were the high values for all measured components listed in Table I. (log A, {FeSOs} = 12 and log Bi {MnSOs} = 8.) 96 148 SOz, NO AND NO2 OX!DATION MECHANISMS molar concentration levels given in Table I, both Cu?* and Mn?* occur as soluble species over the pH range 1-6. Copper(II) forms amino complexes to some extent at pH >4. Manganese({I]) is found primarily as a sulfite complex (log8, = 8) at pH 4-6 and as a free hexaquo metal at pH <4. The speciation of S(IV) is of great interest. Bisulfite readily forms sulfonic acid derivatives, a-hydroxyalkanesulfonic acids (also known as bisulfite addition compounds) [180], as indicated in Equation 218, with aldehydes and ketones. Since relatively high levels of formaldehyde (Ky = 6.3 X 10° M-atm™ 1) (Chapter 4) and other aldehydes have been found in fog and cloud water in Los Angeles, the formation of these addition complexes must be considered in kinetic or thermo- dynamic model development. Dasgupta et al. [181] have established a conditional stability constant for the formation of a-hydroxymethanesulfonic acid of approxi- mately K = 1048 over the pH range 2-6. This constant can be used in the equilibrium computation. Results shown in Figure 5 for S(IV) indicate that the sulfonic acid derivative will be the predominant S(IV) species over a broad range of pH. Of secondary importance below pH 4 are FeSO8, HSO; and H2O°SQz, respectively. The occurrence of S(IV) as stable sulfonic acid derivatives may retard the net oxidation of S(IV) to S(VI). Dasgupta et al. [181] and Fortune and Dellinger [182] have found that addition of CH,O to sulfite solutions stabilizes S(TV) as hydroxymethanesulfonic acid (HMSO), and that HMSO was not significantly de- graded and oxidized to S(VJ) for at least one month. In conclusion, equilibrium speciation models will prove to be an invaluable tool in determining the detailed chemistry of atmospheric water droplets when sufficient time is available for reactions under consideration to be in a steady- state or approximate equilibrium. The kinetic and mass transport limitations for these approximations have been discussed [170,183]. In the next section, the models discussed above will be interfaced with a kinetic model for nighttime fog water chemistry. FOG WATER CHEMISTRY MODEL Background Fog and cloud droplets form by condensation of water vapor on the activated cloud condensation nuclei (ACCN) in the atmosphere and grow by accretion of water vapor. The soluble components of the ACCN on which the droplets form dissolve in the droplet and provide an initial chemical loading. As the droplets grow, they absorb gases from the atmosphere, and liquid-phase reactions occur within. During fall and winter 1981, water from nighttime fogs in the Los Angeles - basin were sampled and analyzed (Table I) [14]. The results of this study highlighted the efficiency of fog droplets as scavengers of air pollutants in the boundary layer; extremely high concentrations of sulfate ions, as well as nitrate and ammonium 97 CATALYTIC SO2 OXIDATION IN NIGHTTIME FOG 149 ions, were routinely found. The Log Angeles b basin is a highly polluted region, and the sulfate levels observed could be attributed to (1) scavenging by nucleation of the large primary sulfate ACCN; (2) scavenging by diffusion of the smaller secondary sulfate aerosol particles; and (3) absorption of atmospheric SO, and subsequent liquid-phase oxidation to S(VI). As discussed above, several pathways for the heterogeneous S(IV) — S(VI) oxidation scheme must be considered. The most important oxidants appear to be H202, O; and Oz catalyzed by the Fe and Mn?+ present at less-than-millimolar concentrations in the aqueous droplets. Hydrogen peroxide and ozone are both absorbed from the gas phase according to their respective Henry’s law equilibria; Fe?* and Mn?* originate from fiy ash and soil dust ACCN on which the droplets have condensed. The object of this section is to present a dynamic model for the water chemistry of nighttime urban fogs with which the contributions of the various oxidation mechanisms to the total S(VI) concentrations in fog water can be evalu- ated. A number of rain and cloud water chemistry models have been proposed in recent years. Adamowicz [184] modeled the chemistry of pure raindrops falling through a well mixed polluted layer with uniform and constant concentrations of SO2z, CO, and NH3. He assumed that the aqueous-phase oxidation of S(IV) to SCVI) occurred exclusively through iron-catalyzed oxidation by O2 according to the kinetic expression of Brimblecombe and Spedding [37]. Mass transfer at the surface of the drop was modeled with two-film theory, which Baboolal et al. [185] have since shown to be unsatisfactory in that it ignores the forced convection inside and outside the falling drop, which greatly enhances the rates of mass transfer. Durham et al. [186] added NO, to the gas phase and considered O, to be the only liquid-phase oxidant, which greatly underestimates S(VI) formation rates. Easter and Hobbs [187] modeled cloud water chemistry by using a wave cloud model, an open atmosphere with trace concentrations of CO2, SO2 and NHg, and a rudimentary S(IV) oxidation rate consisting of a simple first-order dependence on sulfite. In the model presented here, the chemistry is more detailed than in the above models; recent evaluations for reaction mechanisms, rate constants, activation energies and thermodynamic data are used. ACCN chemistry and gas-phase reac- tions are included to simulate the coupling between gas-phase, liquid-phase and aerosol chemistries. The system is considered closed because of the significant depletion of trace gases due to absorption by the fog droplets, which would make the assumption of an open system very inaccurate. Also, the physical processes relevant to droplet chemistry are examined more rigorously than in the previous models; as shall be shown, restriction of the model to nighttime fogs allows one to account for these physical processes in a simple but fairly realistic manner, whereas in rains and clouds, the large temporal and spatial scales involved make detailed analysis exceedingly difficult. The proposed model is a potentially useful tool for prediction of fog water chemistry, even though the present discussion will mostly be limited to S(IV) aqueous-phase oxidation processes. 98 150 SOz, NO AND NOzg OXIDATION MECHANISMS Model Development CGas-Phase Chemistry _ At nighttime the gas-phase chemistry is considerably simplified over that in the daytime because of the absence of sunlight and thus of the associated photochemical reactions. In the absence of OH, O3(P) and CH3Oz radicals, which disappear rapidly after sunset, SO2 gas-phase oxidation is very slow [188] and negligible on the time scales of a fog event. Although oxidation by ozone of organics (especially olefins) may be important, it will not be considered at the present time. Gas- phase chemistry is assumed to be limited to NO, oxidation, which proceeds through the following mechanism [189,190]: NO + 0; —~ NO2 + Op k, = 3.38 X 10° exp(—1450/T) (222) NO, + O; —> NO, + Oz ko = 1.76 X 10? exp(—2450/T) (223) NO + NO; — 2NO, ks = 2.94 X 104 (at 25 C) (224) NO, + NO; — N05 2HNO3 ky = 7.35 X 108 (at 25 C) (225) 2 No data are given for the activation energies of Equations 224 and 225; a doubling of the reaction rates for a 10° C increase in temperature is assumed. The kinetics of the hydrolysis of N2O;, which involves both water vapor and the fog droplets, is not well understood. The essential work in this problem area has been reviewed [191]. Hydrolysis of N2O; appears to occur much faster at the surface of aerosol particles than homogeneously in the gas phase. In the presence of fog droplets the reaction is probably very fast, because of the large droplet surface area and the possibility of heterogeneous hydrolysis. Therefore, this step is assumed to be instantaneous. Reactions 222-225 yield a system of differential equations that is readily solved by a fourth-order Runge-Kutta routine. The overall reaction rate is essentially limited by the slow rate of oxidation of NOsz. Liquid-Phase Chemistry Droplet chemistry is determined primarily by ACCN chemistry and by absorption and subsequent aqueous-phase reactions of atmospheric gases. Gases are absorbed and react in the droplet according to the equilibria listed in Table XIII. The temperature dependence for all the equilibrium constants in Table XIII is deter- mined by the Van’t Hoff equation: AH? /1 1 99 CATALYTIC SOzg OXIDATION IN NIGHTTIME FOG 151 Table XIII. Equilibria and Thermodynamic Data Considered in Fog Water Model [10, 32,196,197] . AG° 298.15 AA? 298.15 Kiagq) Reaction. (kcal-mol~!) (kcal-mol~') (M-atm™ or M) H,0(!) = H* + OH- 19.093 13.345 1.008 x 10-#4 H.CO,(1) = H* + HCO; 8.687 1.825 4.283 .X 10-7 HCO; = H* + CO2- 14.09 3.55 4.687 * 1071! NH,(g) = NH3(!) 2.41 —8.17 5.844 x 10? NH,(!) = NHi + OH- 6.503 8.65 1.709 x 1075 SO.(g) = SO.(1) —0.130 —6.247 1.245 SO.() = HSO; + Ht 2.578 —4.161 1.290 X 1072 HSO; = SO2" + H* 9.850 —2.23 6.014 x 10-8 HNO,(g) = H* + NOs —8.92 17.46 3.460 X 10° H20.(g) = H202(t) —14.5 7.1 X 104 Oa(g) = O3(1) —5.04 9.4 x 1073 It is interesting to note that at the pH commonly encountered in fogs (pH 2-6) all of the gaseous nitric acid and ammonia are absorbed by the droplets because of their acid-base chemistries in the liquid phase. For other gases absorption is limited. Oxidation of S(IV) in the droplet is allowed to proceed along the major pathways discussed previously. The important oxidants are HO, and Og; oxidation by dissolved O2 may also be important if catalyzed by Fe?+ and Mn?*, for which the catalytic synergism noted in Chapter 2 is considered. Oxidation by NO, and by Os catalyzed by trace metals other than Fe?+ and Mn?* is ignored for the present time. Speciation of Fe and Mn within the droplets may control their catalytic effectiveness. Fe and Mn are effective catalysts as long as they are dissolved (as free ions or complexes in solution); however, their catalytic properties are totally altered if they occur as solid phases. Surface catalysis may occur, but this possibility will not be considered here. The distribution of Fe and Mn species inside the fog droplets was obtained from typical values for the concentration of all major components of Los Angeles fog water with SURFEQL [172]. The results were shown in Figure 5. Mn?+ remains totally in solution at the pH values found in fog water, but Fe3+ starts to precipitate as Fe(OH)3 above pH 4. At pH values above 5 no Fe®* remains in solution; therefore, high pH values quench the iron- Catalyzed S(1V) oxidation mechanism. Total soluble Fe**, y, is given by the following _ felationship: y= *Kso[H* PaoFer 100 152 SO2z, NO AND NOz2 OXIDATION MECHANISMS where y is the dissolved fraction of the total iron in the system, *Kgo is the solubility product constant for Fe(OH)-; (S), and ao is defined as follows: ao = (1+ Bi[F-}? + BolF-] + Bs[SO2-] + *K./[H*] + *K.[H*P)" For all oxidation reactions the reaction rates and activation energies are chosen from the expressions suggested in Chapter 2 for atmospheric models, with speciation for Fe?+ added. Physical Description The model presented here uses a closed-box approximation. At t = 0, under condi- tions determined by the ambient saturation, size distribution and chemical composi- tion of the CCN, water droplets suddenly form and grow subsequently by accretion of water vapor. Scavenging of the smaller aerosol! particles by diffusion to the surface of the droplet is not included in the model. The possible limitation of S(IV) oxidation reactions by mass transfer has been discussed [185,188,!92,193]. Seinfeld [188] indicated that the liquid-phase chemical equilibria of concern are established very quickly and thus do not induce any rate limitations in the system. Schwartz and Freiberg [192,193] showed that, for stationary droplets smaller than 50 pm, the rate was limited by the chemical oxidation rates. Baboolal et al. [185] extended this analysis by showing that for the few fog water droplets larger than 50 xm that exhibit a significant sedimentation velocity (as opposed to the smaller quasistationary droplets), forced convection both inside and outside the droplet enhances significantly the rates of mass transfer as calculated for stationary droplets. Because of this enhancement the diffusion rate still is faster than the S(IV) oxidation rate. Therefore, the chemical changes in fog droplets are most likely limited by the specific reaction rates. Because fogs are generally localized events and occur on time scales of a few hours, usually under low wind conditions, no consideration of long-term trans- port of condensing/evaporating droplets was included in the model; also, evolution of the droplet spectrum through coagulation can be neglected since mass transfer does not limit the reaction kinetics. Droplet growth can simply be accounted for by the input of a time-dependent liquid water content (LWC), which is measured experimentally. Thus a reasonable physical description of the system in the model can be limited to the input of temperature and of a time-dependent LWC. Mathematica! Formulation At time t = 0, droplets are assumed to suddenly condense in an atmospheric “closed box” containing ACCN and trace atmospheric gases; the instantaneous formation of these droplets is associated with a certain LWC. This LWC may be made to vary with time following either experimental observations or theoretical predictions [194]. In the results presented here, LWC will be taken to be constant. When the droplets form they receive an initial chemical loading from the water- 101 CATALYTIC SO OXIDATION IN NIGHTTIME FOG 153 soluble fraction of the ACCN, immediately absorb atmospheric gases according to Henry’s law and the liquid-phase equilibria of Table XIII are established. The concentrations of the species at equilibrium are obtained by solving the electroneu- trality equation for a closed system (see appendix for details of the calculation). Formation of S(VI) in the liquid phase is calculated over small time increments At by a forward finite-difference technique: [SVDh+as = [SVDk + (¥ (ee) ) At (228) ox dt After each time step, changes in gas-phase concentrations from fresh emissions and NO, oxidation are calculated, and the electroneutrality equation is solved to obtain the new equilibrium concentrations for the species in the system. Time increments are taken as At = 2 X 10°-* n (minutes) where n is the time step number; this expression allows for a better resolution at the beginning of the fog event, where chemical changes are rapid. The use of smaller time steps did not change the calculation results. Results and Discussion The model was designed to interpret the results of the 1981-1982 fog sampling program [14] and to shed light on the relative importances of the various liquid- phase S(IV) oxidation mechanisms in ambient fog droplets. The present discussion will be limited to the latter point, in light of the fog water data collected in the Los Angeles basin [14] and the permanent atmospheric chemistry records of the Air Quality Management District. The following pre-fog nighttime gas-phase concentrations were assumed for all model runs: . COz = 3.3 X 105 ppb, NO = 100 ppb, ° NO, = 50 ppb, * NH3=5 ppb, HNO; = 3 ppb, and . SO, = 20 ppb. In cases H and I, different levels and emission rates for NO and NO, are considered. Gas emission sources per liter of basin air were taken to be -* §yo = 0.1 ppb-min7!-L7, * yo, = 0.01 ppb-min-}-L~}, dso, = 0.05 ppb-min-!-L~}, and * 6yn, = 0.005 ppb-min=!-L71. 102 “p-W-3 | = DMT pur ,_ulur-qdd co9'0 = “HNg ‘;_urw-qdd cog = 8g “gg add Oz “HN add ¢ “ONH add ¢ QD udd O¢¢ ‘suoNpuos jeuoMppy « ot Ol ol ol Sz ol ol ol ol (D) ainqesaduia |, sok ss SOA sox sox sax Sox ON sox sox UP /24 ‘WSkdsoUAS s-w-pou | ,_w-jowd | IDReN IDPBN HO#8N 'ONH IeN DRN D®N PRN [OPN uolsedwog NOOV 0 r-O1 100 100 10'0 100 100 100 100 70NQ 0 e-O] 10 ize) 10 ie) v0 ime) 10 Ong (;ului-qdd) *QN JO seosn0g S S S ¢ S S is if g +cUlWN ool oot oo! OO! 0.0) | ool ool 07 OO! +090 (;-7J-bev!) suonesmusouod [Ray aseyg-snoanby 0 ol os Os os os Os os Os 7ON 0 S oo! oor ool ool oo} ool OO! ON ol ol ol ol ol ol Ol Ol ol i @) it 1 I 1 I S I I i O*7H (qdd) sasen aaydsouyy jo suonesjuaaUu0D rf H D d a ad re) qd V asp spo] 243 Aq payenung syusaq 30,4 94) Joy sUORIPUOD SUaYydsow;yY “ATX F1GB1 103 CATALYTIC SO2 OXIDATION IN NIGHTTIME FOG 155 Concentrations of gas-phase ozone at night range from 0 to 30 ppb. There is some evidence of nighttime diffusion of ozone from the upper atmosphere down to the boundary layer as ozone is depleted near the ground by gas-phase oxidation reactions [195]; this source term will be ignored for now. Because NO is usually in excess of Os, the additional ozone would merely oxidize part of the excess ‘NO and thus have little effect on the droplet chemistry. Hydrogen peroxide levels are poorly documented, and the reliability of the few existing measurements has been questioned; a concentration range of 1-5 ppb will be assumed. Hydrogen peroxide is not considered to be an important oxidant for nighttime gas-phase NO, chemistry. A major feature of the Los Angeles fog water as shown in Table I is the high levels of trace metals. The model calculations were performed for the ranges observed, which are 20-200 eq-L~! for Fe?*+ and 1-10 preq-L~? for Mn?*- For all runs a typical value of 20 was taken for the Fe/Mn ratio in equivalents. Table XIV gives concentration conditions imposed on the model calculations. Calculations were first made under conditions typical of Los Angeles fog with a temperature of 10° C, a LWC of 1 g-m~3, prefog conditions of 1 ppb H2O02 and 10 ppb Os, and trace metal levels in the droplets of 100 jreq-L~! for Fe and 5 jreq-L~! for Mn. Synergism between Fe and Mn was considered for S(IV) oxida- tion, and the ACCN were assumed to be NaC! or some other neutral aerosol. The results of the calculations are shown in Figure 6. A basic feature of S(IV) nighttime oxidation is that the strong oxidants (H2O02 and Os) are not resupplied to the atmosphere in the absence of the daytime photo- chemical reactions. As seen in Figure 6, 10 min after fog formation, H2O2 and Os no longer contribute to S(IV) oxidation because they have been totally depleted from the atmosphere, H2O. by S(IV) liquid-phase oxidation and Os; by S(TV) liquid- phase oxidation and NO, gas-phase oxidation. Hydrogen peroxide is an efficient liquid-phase oxidant because it is very soluble in water, has rapid oxidation rates at low pH, and is not depleted at a significant level by gas-phase oxidation reactions. On the other hand, Os is poorly soluble; its rate for oxidizing S(IV) drops dramati- cally as pH decreases; and it reacts quickly with NO in the gas phase to form NOz2. Hydrogen peroxide is thus a very effective oxidant for S(IV) in ambient fog water, whereas oxidation by Os is relatively insignificant. In addition, it is apparent from the model calculations that rapid reaction of O; with NO not only prevents liquid-phase S(IV) oxidation by O; but also gas-phase NO; oxidation by Os, which is the rate-limiting reaction for the formation of nitric acid at night; thus, there is no nitrate formation during a nighttime fog event (although the fog water may have a high nitrate concentration because of the scavenging of nitrate particles and of the prefog gaseous nitric acid). Finally, the most striking result of the calculations is the dominating role of metal-catalyzed oxidation, which accounts for more than 90% of the S(VD formed in the liquid phase four hours after the beginning of the fog event. Unlike H,O, and O3, there is an infinite supply of Oz, and concentrations of Fe and Mn in Los Angeles fog water are high enough to make O, an effective oxidant. Although the corresponding oxidation rates exhibit an inverse dependence on H* 104 156 SOs, NO AND NO» OXIDATION MECHANISMS 4. ; CASE A é. . CASE A an ask 4 SULFATE FORMED (MolesL7'x 10 *) 4 = 4 bo 3.5 = 3 Hi L 250 50 100 150 200 250 TIME (Minutes) TIME (Minutes) c. CASE A ~ F fT T T wT ° ad ar 7 7 a 3 2 6- 4 Zz i] & 4 4 & E Fa ge TOTAL S( IZ) 7 oO ° x fe) 50 100 50 200 250 TIME (Minutes) Figure 6. Sulfate formed, pH and total S(IV) for the fog water as a function of time after fog formation, under conditions typical of the Los Angeles atmosphere (case A): 20 ppb SO, 10 ppb O3, | ppb H2Oz in the pre-fog gas phase, metal aqueous-phase concentrations of 100 eq-L7! Fe3+ and 5 peq-L7! Mn?*. concentration, dissolution of the ferric hydroxide precipitate below pH 4 maintains oxidation rates at a high level at the low pH values characteristic of Los Angeles fog water. Even at the lowest concentrations of metals found in the fog water (Case B, Figure 7) and in the absence of synergism between Fe and Mn catalytic effects (Case C, Figure 7), metal-catalyzed oxidation is still the major source of S(VI) formation. It should not, however, be concluded from the above results that HO. is only a minor contributor to S(IV) oxidation. Because no reliable method exists for measuring gas-phase H2Oz, its levels in the atmosphere are uncertain; at concen- trations higher than the 1 ppb considered in Case A it may be an important oxidant. This is illustrated by Case D (Figure 8), where a pre-fog HO, level of 5 ppb is assumed; the contribution of HzO2 to S(IV) oxidation is then important, and dominates oxidation processes until two hours after fog formation. High levels of H2O, also slow down the metal-catalyzed oxidation kinetics because of the abrupt pH drop associated with the rapid quantitative depletion of H2O2. _ Temperature dependence of the chemical composition of the droplet was studied by raising the temperature in Case A from 10 to 25 C (Case E, Figure 8). At this higher temperature the oxidation rates are higher, but the solubilities 105 CATALYTIC SO2 OXIDATION IN NIGHTTIME FOG 157 of the gases are lower. These opposite effects make the system rather stable to changes in temperature. Comparing Cases A and E 4 h after fog formation, some- what more S(VI) has been produced at the higher temperature (700 ymol-L~ at 25° C vs 520 pmol-L~ at 10 C) while the pH is slightly lower (2.9 at 10° C vs 3.05 at 25° C). Oxidation of S(IV) by ozone becomes totally unimportant at 25° C, as ozone is even less soluble than at 10 C. The effect of droplet “background” acidity was investigated by calculating the chemical content of the droplets in Case F of acid ACCN (1 pmol-m- of HNOs) and in Case G of basic ACCN (1 pmol-m7? of NaOH). These cases yielded ‘§nitial pH values” of 3.03 and 6.50, respectively; the calculated results are shown in Figure 9 (cases F and G). The effects of the pH changes are damped by the slow S(IV) metal catalyzed oxidation rates at low pH, and after 4 h, the pH values are reasonably close at 2.90 and 3.15, respectively. However, six times more sulfate is formed in the initially “basic” droplet. A main feature of this latter ’ case is that the high initial pH allows for a much higher rate of S(IV) oxidation by Os, which then becomes a major contributor to S([V) oxidation. In urban air, the levels of NO at dusk are generally higher than those of n~,A CASE B B CASE 8B «4 T 7 T T 55 T T T 1 2 7 5 4 43 3 45 4 $ [ wt Pi 35h 4 wu F3 ce] 3 i i 4 { ° 50 100 150 200 250 ° 50 100 180 200 250 TIME (Minutes) . Dd. CASE _C 34 5.5 T T T T 9° 7 7 5r 4 43 s FE + OP 4 g? . | 4 & wy 3 35h 4 of ee NO a 3 { 1 n 4 is) sO 100 150 200 250 o be 6) + 0,6) 150 200 250 TIME (Minutes) TIME (Minutes) Figure 7, Sulfate formed and pH for the fog water as a function of time after fog formation, when Fe3* = 20 peg-L-! and Mn?* = 1 preq-L~! (case B) which are the lows observed in the Los Angeles fog water, or when no synergism for iron- and manganese-catalyzed oxidation mechanisms is considered (case C). 106 158 SO2z, NO AND. NO» OXIDATION MECHANISMS a A. CASE D 8. CASE D ~ 6 T T + T 5 T T T T J x 5 4 45. : + Ps Za = Pa ry 3 5 4F 4 5 a 2 ww bE 35 4 z ain 5 w . oO 3 it 1 1 ° 50 100 150 200 250 TIME (Minutes) . CASE E o3 5.5 T T —T T S as Se 4 4 g = 455 4 = ° =z ¥ a 4 - § * 35 4 Ts) i= & 3 5 4 wo 25 i 4 i. i ° 50 100 150 200 250 TIME (Minutes) TIME (Minutes) Figure 8. Sulfate formed and pH for the fog water as a function of time after fog formation, when H2O; = 5 ppb in the pre-fog gas phase (case D), or when the temperature is 25 C (case E). O3, and the gas-phase oxidation of NO by O, depletes all the O, from the system. However, it may happen that ozone will be in excess of NO; in that case, after the oxidation of all the NO to form NOx, there will be ozone available to oxidize both NO, in the gas phase to form nitric acid and S(IV) in the liquid phase to form S(VI). Figure 10 illustrates the chemical behavior of the fog under conditions of Case H, where ozone is in excess of NO. Comparing the results to those of Case A, one observes a slight increase in the ozone contribution to the S(IV) liquid-phase oxidation, but mostly a large amount of nitrate formed. After NO is depleted, the slow gas-phase oxidation of NOz to HNO; thus becomes the domi- nant pathway for ozone removal from the system. Less sulfate is then formed because of the drop in pH due to the incorporation in the droplet of the nitric acid. It is interesting to note that under conditions of a NO,-free atmosphere (Case I, Figure 10) one does not observe an increase in the production of S(VI) by ozone as compared to Case H; this shows that production is not limited by competition from gas-phase NO, oxidation but rather by the inefficacy of ozone as a liquid-phase oxidant for S(JV) in low pH solutions. It is observed from the model calculations that there is no oxidation of S(IV) by ozone below pH 4 regard- less of the NO, concentrations. Overall the results of this dynamic model reveal some important features 107 CATALYTIC SO2 OXIDATION IN NIGHTTIME FOG 159 A CASE F 8. CASE F 25 T T T T 3,08 T T T T T 1 = 2.95 29+ . 4 SULFATE FORMED (Moles Lx 10 4) : 2.85 1 n 1 fo] 50 100 150 200 250 oO 50 0 50 200 250 TIME (Minutes) . TiME (Minutes) Cc. CASE G Dd. CASE G 10 T T T T : 7 T T T T _ SULFATE FORMED(Moles L7'x 10 *) i i 100 150 200 250 TIME (Minutes) TIME (Minutes) ° ts] Figure 9. Sulfate formed and pH for the fog water as a function of time after fog formation, when condensation takes place on HNOs ACCN (case F), or on NaOH ACCN (case G). of S(IV) liquid-phase oxidation in the fogwater of polluted atmospheres. They are: 1. The main contributor to S(VI) formation in the aqueous phase at night is metal-catalyzed oxidation of S(IV). This conclusion may be challenged be- cause of the lack of definitive data for H.O. concentrations in the atmosphere, which makes our estimated value of 1 ppb no more than an educated guess; it still appears that metal-catalyzed oxidation is more important in polluted atmospheres than has been hitherto recognized. 2. S(IV) oxidation by ozone is insignificant because of a combination of slow oxidation rates at low pH, poor ozone solubility and depletion of ozone in the gas phase by oxidation of NO. 3. S(IV) oxidation by H,Oz is very rapid, but is limited by the amount of H.0O, originally present in the system. _ 4. Precipitation of ferric hydroxide at pH 4, with the accompanying loss (or modification) of its catalytic properties, requires that Fe®*+ speciation be taken into account in the development of atmospheric water chemistry models. 5. If the level of NO is higher than that of O3, there is no nitrate formation in the fog. If it is lower than that of O3, O3 will first oxidize NO very” quickly to form NO, and then slowly oxidize NO2 to form NOs, which 108 160 SO, NO AND NOz OXIDATION MECHANISMS A. CASE 8. «CASE H T T ———F T T 55 T T T a pH SULFATE FORMED (Moles L-'x 10° ) Nn 1 1 Q 50 100 150 200 250 TIME (Minutes) __O. CASE I .* ~ T T T T 7 + 2° i) as 1 3 2 4 cad f. s & 8 E3 4 4 = E & 2 ow S 2 7 w a é 3 fo) 1 1 n n a ie) sO wo 150 200 250 TIME (Minutes } TIME (Minutes) Figure 10. Sulfate formed, pH and nitrate for the fog water as a function of time after fog formation when ozone is in excess of NO (case H), and sulfate formed under conditions of a NO,-free atmosphere (case I). will then react to produce nitric acid. The rate of S(IV) oxidation by ozone in such a system is not limited by the amount of NO, in the atmosphere but by the pH of the fog water. Below pH 4 it is insignificant. ~ Conclusion In heavily polluted atmospheric water droplets, such as those found in urban fogs, the model presented here shows that metal-catalyzed S(IV) oxidation is a significant contributor to formation of S(VI) in the liquid phase, and apparently is more important than oxidation by H,O2. Under conditions typical of the Los Angeles area, liquid-phase oxidation of S(IV) to S(VI) is found to account for sulfate concentrations in the range of 5 X 10-* mole-L~! 4 h after fog formation. Waldman et al. [14] found sulfate concentrations ranging from 5 X 10-‘ to 2 X 10-3 mole-L~! in the fog. Direct comparison of our calculated values with the field data is difficult, since Waldman et al. did not measure the LWC of the fogs they studied; nevertheless, it appears clearly that absorption of SO. followed by liquid-phase oxidation is an important mechanism for explaining the sulfate concentrations in polluted fog water. 109 CATALYTIC SO2g OXIDATION IN NIGHTTIME FOG 161 The present model needs to be refined further; the main limitations now are the failure to include organic complexing agents and reductants and the rather simplistic treatment of the formation of complexes. Absorption of formaldehyde followed by formation of its sulfite complex (HOCH2SO;) must be considered in an advanced model; including this complex and others requires the use of multidi- mensional calculation routines such as SURFEQL to obtain the equilibrium concen- trations of the species in the droplet. ACKNOWLEDGMENTS The authors acknowledge gratefully the financial support of the U.S. Environmental Protection Agency (Grant No. R808086-1) and the President’s Fund/Sloan Foun- dation Grant administered by the California Institute of Technology. They are also grateful to 8. Schwartz, J. Seinfeld and G. McRae for their helpful comments. APPENDIX: THE ELECTRONEUTRALITY EQUATION The ionic species in the fog droplets must satisfy the following electroneutrality equation: {H*] + [NHi] + [Na*] + 3[Fe**) + 2[Mn2*] = [OH-] + [HCO;] + 2[CO2-] + [HSOz] + 2[SO2-} + 2[SO2-] + [NO;] + [CI] where the iron and manganese complexes have been omitted for now to allow the equation to be solved by the one-dimensional Newton’s method. Fe3* and Mn?* are assumed to be introduced in the liquid phase through ACCN containing FeCl; and MnCl. In the closed two-phase system considered here the mass balance equation for component A can be written: P, Ar= 77 L[A(aq)] where P, = atmospheric partial pressure of A [A(aq)] = concentration of A in the liquid phase (where A may be dis- tributed among different species) L=LWC (liters of water per liters of air) Ar = total double-phase concentration of A in a liter of air (note that L << 1); a constant for the system For sulfur(IV), for example, 110 162 SOs, NO AND NO2g OXIDATION MECHANISMS _ Psoz y— 2~ Sr= RT + L({SO.(!)] + [HSO;] + [SO3}-]) From Table XIII: x, = SOx Pso, _ [HSO3][H*] Ks=" Ts0.0I _ {SO} H*] Ko =" THSO5] which yields K7KsKo 1 2-7 = Ss [S02] [H*}2 *( 1 +1(K EE EEE) RT "" TH] (HP In this manner the electroneutrality equation can be rewritten: [H*] + K;K.{[H*] [NHa]_ (z Kw 1 ee) + [Na*] + 3[Fe3*] + 2[Mn?*] Kw , K2K; 2K.K3K, K,Kg (H"] + 1H] Poo, + (H} Poo, + [H*] 1 K7Kg K7KsKg att Uk: Te) > TP ) [H*? fot ( K:Ks ER) RT +L K,+ TH) +e Ki, [NOs], 2] 4+ + ICI- RT [H*] which is of the form f([H*]) = 0 and is readily solved by Newton’s method. lil CATALYTIC SO2 OXIDATION IN NIGHTTIME FoG 163 REFERENCES 10. 11. 12. 13. 14. 15. . 16. 17. 18. 19. 20. Calvert, J.G., F. Su, J.W. Bottenheim and O.P. Strausz. “Mechanisms of Homoge- neous Oxidation of Sulfur Dioxide in the Troposphere,” Atmos. Environ. 12:197- 226 (1978). Middleton, P., C.S. Kiang and V.A. Mohnen. “Theoretical Estimates of the Relative Importance of Various Urban Sulfate Aerosol Production Mechanisms,” Atmos. Envi- ron. 14:463-472 (1980). Miller, D. “Kinetic Model of Atmospheric Oxidation Based on Published Data,” Atmos. Environ. 14:1067-1076 (1980). Cass, G.R., and F.H. Shair. “Transport of Sulfur Oxides Within the Los Angeles Sea Breeze/Land Breeze Circulation System,” in Proceedings of the Second Joint Conference on Applications of Air Pollution Meteorology (American Meteorological Society, 1980), pp. 320-327. Cass, G.R. “Methods for Sulfate Air Quality Management with Applications to Los Angeles,” PhD Thesis, California Institute of Technology, Pasadena, CA (1977). Cox, R.A. “Particle Formation from Homogeneous Reactions of Sulfur Dioxide and Nitrogen Dioxide,” Tellus 26:235-240 (1974). McMurry, P.H., D.J. Rader and J.L. Smith. “Studies of Aerosol Formation in Power Plant Plumes. I. Parameterization of Conversion Rate for Dry, Moderately Polluted Ambient Conditions,” Atmos. Environ. 15:2315-2329 (1981). Smith, F.B., and G.H. Jeffrey. “Airborne Transport of Sulphur Dioxide from the U.K.,” Atmos. Environ. 9:643-659 (1975). Wilson, J.C., and P.H. McMurry. “Studies of Aerosol Formation in Power Plant Plumes—I. Secondary Aerosol Formation in the Navajo Generating Station Plume,” Atmos. Environ. 15:2329~2339 (1981). Liljestrand, H.M., and J.J. Morgan. “Spatial Variation of Acid Precipitation in South- ern California,” Environ. Sci. Technol. 15:333-339 (1981). Gartrell, J.E., J.W. Thomas and S.B. Carpenter. “Atmospheric Oxidation of SO. in Coal-Burning. Power Plant Plumes,” dm. Ind. Hyg. Assoc. Quart. 24:113-120 (1963). Diffenhoeffer, A.C., and R.G. dePena. “A Study of Production and Growth of Sulfate Particles in Plumes from a Coal-Fired Power Plant,” Atmos. Environ. 12:297-306 (1978). Enger, L., and U. Hogstrom. “Dispersion and Wet Deposition of Sulfur from a Power Plant Plume,” Atmos. Environ. 13:797-810 (1979). Waldman, J.M., J.W. Munger, D.J. Jacob, R.C. Flagan, J.J. Morgan and M.R. Hoff- mann. “Chemical Composition of Acid Fog,” Science 218:677-680 (1982). Hegg, D.A., and P.V. Hobbs. “Cloud Water Chemistry and the Production of Sulfates in Clouds,” Atmos. Environ. 15:1597-1604 (1981). “Guide for Air Pollution Avoidance. Appendix B. History of Episodes,” PH-22-68- 32, U.S. EPA (1971), pp. 123-135. Wilkins, E.T. “Air Pollution and the London Fog of December, 1952,” J. Roy. San. Instit. 74:1-21 (1954). Wilkins, E.T. “Air Pollution Aspects of the London Fog of December, 1952,” J. Roy. Meteorol. Soc. 80:267-278 (1954). Holland, W.W. Air Pollution and Respiratory Disease (Westport, CT: Technomic Publishing Co., Inc., 1972), pp. 27-51. Friedlander, S.K., and A.L. Ravimohan. “A Theoretical Model for the Effect of an 164 21.. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 112 SOs, NO AND NOz OXIDATION MECHANISMS Acute Air Pollution Episode on a Human Population,” Environ. Sci. Technol. 2:1101- 1108 (1968). ; Larsen, .R.1. “Relating Air Pollutant Effects to Concentration and Control,” J. Air Poll. Control Assoc. 20:214-225 (1970). . Cheng, R.T., M. Corn and J.O. Frohliger. “Contributions to the Reaction Kinetics of Water-Solubles and SO, in Air at ppm Concentrations,” Atmos. Environ. 5:987— 1008 (1971). Beilke, S., and G. Gravenhorst. “Heterogeneous SO2-Oxidation in the Droplet Phase,” Atmos. Environ. 12:231-239 (1978). Dasgupta, P.K., P.A. Mitchell and P.W. West. “Study of Transition Metal-S(IV) Systems,” Atmos. Environ. 13:775-782 (1979). Freiberg, J. “Effects of Relative Humidity and Temperature on Iron-Catalyzed Oxida- tion of SO. in Atmospheric Aerosols,” Environ. Sci. Technol. 8:731-734 (1974). Fuzzi, 8. “Study of Iron(III) Catalyzed Sulphur Dioxide Oxidation in Aqueous Solu- tion over a Wide Range of pH,” Atmos. Environ. 12:1439-1442 (1978). Hegg, D.A., and P.V. Hobbs. “Oxidation of Sulfur Dioxide in Aqueous Systems with Particular Reference to the Atmosphere,” Atmos. Environ. 12:241-253 (1978). Kaplan, D.J., D.M. Himmelblau and C. Kanaoka. “Oxidation of Sulfur Dioxide in Aqueous Ammonium Sulfate Aerosols Containing Manganese as a Catalyst,” Atmos. Environ. 15:763-773 (1981). Larson, T.V., N.R. Horike and H. Halstead. “Oxidation of Sulfur Dioxide by Oxygen and Ozone in Aqueous Solution: A Kinetic Study with Significance to Atmospheric Processes,” Atmos. Environ. 12:1597-1611 (1978). Penkett, S.A., B.M.R. Jones and A.E.J. Eggleton. “A Study of SO2 Oxidation in Stored Rainwater Samples,” Atmos. Environ. 13:139-147 (1979). Hoffmann, M.R., and J.O. Edwards. “Kinetics and Mechanism of the Oxidation of Sulfur Dioxide by Hydrogen Peroxide in Acidic Solution,” J. Phys. Chem. 79:2096- 2098 (1975). Penkett, S.A., B.M.R. Jones, K.A. Brice and A.E.J. Eggleton. “The Importance of Atmospheric Ozone and Hydrogen Peroxide in Oxidizing Sulfur Dioxide in Cloud ‘and Rainwater,” Atmos. Environ. 13:123-137 (1979). Martin, L.R., and D.E. Damschen. “Aqueous Oxidation of Sulfur Dioxide by Hydro- gen Peroxide at Low pH,” Atmos. Environ. 15:1615-1622 (1981). Martin, L.R., D.E. Damschen and H.S. Judeikis. “Sulfur Dioxide Oxidation Reactions in Aqueous Solution,” EPA 600/7-81-085, U.S. EPA (1981). Thornton, J.D. “The Metal and Strong Acid Composition of Rain and Snow in Minnesota: Land Use Effects,” MS Thesis, University of Minnesota, Minneapolis, MN (1981). Overton, J.H., V.P. Aneja and J.L. Durham. “Production of Sulfate in Rain and Raindrops in Polluted Atmospheres,” Atmos. Environ. 13:355-367 (1979). Brimblecombe, P., and D.J. Spedding. “The Catalytic Oxidation of Micromolar Aque- ous Sulphur Dioxide,” Atmos. Environ. 8:937-945 (1974). Middleton, P., C.S. Kiang and V.A. Mohnen. “Theoretical Estimates of the Relative Importance of Various Urban Sulfate Aerosol Production Mechanisms,” Atmos. Envi- ron. 14:463-472 (1980). Barrie, L.A., and H.W. Georgii. “An Experimental Investigation of the Absorption of Sulfur Dioxide by Water Drops Containing Heavy Meta! Ions,” Atmos. Environ. 10:743-749 (1976). Hoffmann, M.R., and 8.D. Boyce. “Catalytic Autoxidation of Aqueous Sulfur Dioxide 41. 42. 43. 45. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 37. 58. 59. 113 CATALYTIC SOz OXIDATION IN NIGHTTIME FOG 165 in Relationship to Atmospheric Systems.” Adv. Environ. Sci. Tech. 12, 148-189 Wiley- Interscience, New York (1983). Mader, P.M. “Kinetics of the Hydrogen Peroxide-Sulfite Reaction in Solution,” J. Am. Chem. Soc. 80:2634 (1958). Halperin, J., and H. Taube. “The Transfer of Oxygen Atoms in Oxidation-Reduction Reactions. IV. The Reaction of Hydrogen Peroxide with Sulfite and Thiosulfate, and of Oxygen, Manganese Dioxide and Permanganate with Sulfite,” J. Am. Chem. Soc. 74:380-382 (1952). Harrison, H., T.V. Larson and C.S. Monkman. “Aqueous Phase Oxidation of Sulfites by Ozone in the Presence of Iron and Manganese,” Atmos. Environ. 16:1039-1041 (1982). Hoffmann, M.R. “Trace Metal Catalysis in Aquatic Environments,” Environ. Sci. Technol. 14:1061-1066 (1980). Frank, S.N., and A.J. Bard. “Heterogeneous Photocatalytic Oxidation of Cyanide and Sulfite in Aqueous Solution at Semiconductor Powders,” J. Phys. Chem. 81:1484— 1488 (1977). Mishra, G.C., and R.D. Srivastava. “Homogeneous Kinetics of Potassium Sulfite Oxidation,” Chem. Eng. Sci. 31:969-971 (1976). Boreskov, G.K., R.A. Buyanov and A.A. Ivanov. “Kinetics of Sulfur Dioxide on Vanadium Catalysts,” Kinetika Kataliz 8:153-159 (1967). Sung, W., and J.J. Morgan. “Kinetics and Products of Ferrous Iron Oxygenation in Aqueous Systems,” Environ. Sci. Technol. 14:561~-568 (1981). Beilke, S., and G. Gravenhorst. “Heterogeneous SO2-Oxidation in the Droplet Phase,” Atmos. Environ. 12:231-239 (1978). Backstrém, H. “Der Kettenmechanisms bei der Autoxydation von Natriumsulfitlésun- gen.” Z. Physik. Chem. 25B:122-138 (1934). Moore, J.W., and R.G. Pearson. Kinetics and Mechanism (New York: Wiley-Intersci- ence, 1981). Boudart, M. The Kinetics of Chemical Processes (Englewood Cliffs, NJ: Prentice- Hall, Inc., 1968), pp. 246. Coughanowr, D.R., and F.E. Krause. “The Reaction of SO2 and Oz in Aqueous Solutions of MnSO,,” Ind. Eng. Chem. Fund. 4:61-66 (1965). Ibusuki, T. “‘Manganese(I!) Catalyzed Sulfur Dioxide Oxidation in Aqueous Solution at Environmental Concentrations over a Wide Range of pH,” Atmos. Environ. (in press). Matteson, J.J., W. Stober and H. Luther. “Kinetics of the Oxidation of Sulfur Dioxide by Aerosols of Manganese Sulfate,” Ind. Eng. Chem. Fund. 8:677-684 (1969). Lunak, S., A. El-Wakil and J. Veprek-Siska. “Autoxidation of Sulfite Catalyzed by 3d Transition Metals and Inhibited by 2-Propanol: Mechanism of Inhibited and In- duced Reactions,” Coll. Czech. Chem. Comm. 43:3306-3316 (1978). Aubuchon, C. “The Rate of Iron Catalyzed Oxidation of Sulfur Dioxide by Oxygen in Water,” PhD Thesis, Johns Hopkins University, Baltimore, MD (1976). Chen, T.I., and C.H. Barron. “Some Aspects of the Homogeneous Kinetics of Sulfite Oxidation,” Ind. Eng. Chem. Fund. 11:466-470 (1972). Sawicki, J.E., and C.H. Barron. “On the Kinetics of Sulfite Oxidation in Heterogeneous Systems,” Chem. Eng. J. 5:153-159 (1973). Bengtsson, S., and I. Bjerle. “Catalytic Oxidation of Sulphite in Dilute Aqueous Solutions,” Chem. Eng. Sci. 30:1429-1435 (1975). 166 61. 62. 63. 65. 66. 67. 68. 69. | 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 114 SOQ2z, NO AND NO2 OXIDATION MECHANISMS Yagi, S., and H. Inoue. “The Absorption of Oxygen into Sodium Sulfite Solution,” Chem. Eng. Sci. 17:411-421 (1962). Davies, R., A.K.E. Hagopian and A.G. Sykes. “Kinetic and Oxygen-18 Tracer Studies on the Reaction of Sulphite with the Superoxo-Complex (NH3)sCo-O,-Co(NH3)§* in Aqueous Media,” J. Chem. Soc. (A):623-629 (1969). Linek, V., and J. Mayrhoferova. “The Kinetics of Oxidation of Aqueous Sodium Sulfite Solution,” Chem. Eng. Sci. 25:787-800 (1970). , Mishra, G.C., and R.D. Srivastava. “Kinetics of Oxidation of Ammonium Sulfite by Rapid-Mixing Method,” Chem. Eng. Sci. 30:1387-1390 (1975). Fuller, E.C., and R.H. Crist. “The Rate of Oxidation of Sulfite Ions by Oxygen,” J. Am. Chem. Soc. 63:1644-1650 (1941). Barron, C.H., and H.A. O’Hern. “Reaction Kinetics of Sodium Sulfite by the Rapid- Mixing Method,” Chem. Eng. Sci. 21:397-404 (1966). Hayon, E., A. Treinin and J. Wilf. “Electronic Spectra, Photochemistry, and Autoxida- tion Mechanism of the Sulfite-Bisulfite-Pyrosulfite Systems. The SO2, SO3, and SOs Radicals,” J. Am. Chem. Soc. 94:47-57 (1972). Schmittkunz, H. “Chemilumineszenz der Sulfitooxidation,” Dissertation, Naturwis- senschaftliche Fakultat, Frankfurt (1963). Bassett, H., and W.G. Parker. “The Oxidation of Sulphurous Acid,” J. Chem. Soc. (1951), pp. 1540-1560. King, E.L., and C. Altman. “A Schematic Method of Deriving the Rate Laws for Enzyme Catalyzed Reactions,” J. Phys. Chem. 60:1375-1378 (1956). Freiberg, J. “The Mechanism of Iron-Catalyzed Oxidation of SO2,” Atmos. Environ. 9:661-672 (1975). Bracewell, J.M., and D. Gall. “The Catalytic Oxidation of Sulfur Dioxide in Solution at Concentrations Occurring in Fog Droplets,” paper presented at the Symposium on Physio-chemical Transformation of Sulfur Compounds in the Atmosphere and the Formation of Acid Smogs, Mainz, Germany, 1967. Hoather, R.C., and C.F. Goodeve. “The Oxidation of Sulphurous Acid HI. Catalysis by Manganous Sulphate,” Trans. Faraday Soc. 30:1149-1156 (1934). Neytzell-deWilde, F.G., and L. Taverner. “Experiments Relating to Possible Produc- tion of Air Oxidizing Acid Leach Liquor by Autoxidation for the Extraction of Uranium,” 2nd U.N. Intl. Conf. Peaceful Uses for Atomic Energy Proc. 3:303-317 (1958). Karraker, D.G. “The Kinetics of the Reaction Between Sulfurous Acid and Ferric Ion,” J. Phys. Chem. 67:871-874 (1963). Pollard, F.H., P. Hanson and G. Nickless. “(Chromatographic Studies on the Oxidation of Sulphurous Acid by Ferric Iron in Aqueous Acid Solution,” J. Chromatog. 5:68- 73 (1961). Carlyle, D.W. “A Kinetic Study of the Aquation of Sulfitoiron(IID) Ion,” Inorg. Chem. 10:761-764. Lancaster, J.M., and R.S. Murray. “The Ferricyanide-Sulphite Reaction,” J. Chem. Soc. (1971), pp. 2755-2758. Veprek-Siska, J.. D.M. Wagnerova, and K. Eckschlager. “Eindequivalent-Oxyda- tionen. I]. Sulfitoxydation Durch Komplexionen,” Coll. Czech. Chem. Comm. 31:1248-1255 (1966). . Veprek-Siska, J., A. Solcova and D.M. Wagnerova. “Eindequivalent-Oxydationen. III. Kinetik der Sulfitoxydation Mittels Pentacyanoaquoferrats(III),” Coll. Czech. Chem. Comm. 31:3287-3298 (1966). 81. 82. 83. 84. 85. 86. 87. 88. 89. 91. 92. 93. 94, 95. 96. 97. 98. 115 CATALYTIC SOs OXIDATION IN NIGHTTIME FOG 167 Carlyle, D.W., and O.T. Zeck, Jr. “Electron Transfer Between Sulfur(IV) and Hexa- aquoiron(III) Ion in Aqueous Perchlorate Solution. Kinetics and Mechanisms of - Uncatalyzed and Cu(II)-Catalyzed Reactions,” Inorg. Chem. 12:2978-2983. Higginson, W.C.E., and J.W. Marshall. “Equivalence Changes in Oxidation-Reduction Reactions in Solution: Some Aspects of the Oxidation of Sulphurous Acid,” J. Chem. Soc. (1957), pp. 447-458. Reinders, W., and S.I. Vies. “The Catalytic Oxidation of Sulfites,” Rec. Trav. Chem. 44:249-268 (1925). Titoff, A. ““Beitrage zur Kenntnis der Negativen Katalyse in Homogeneous System,” Z. Phys. Chem. 45:641-683 (1903). Latimer, W. M. Oxidation Potentials 2nd ed. (New York: Prentice-Hall, Inc., 1952). Srivastava, R.D., A.F. McMillan and I.J. Harris. “The Kinetics of Oxidation of Sodium Sulphite,” Can. J. Chem. Eng. 46:181-184 (1968). Yatsimirskii, K., B. Bratushko, I. Yu and 1.L. Zatsny. “Kinetics and Mechanism of the Reduction of Molecular Oxygen Coordinated in the Complex Co2(L- histidine),O. by Sodium Sulphite in Aqueous Solution,” Zh. Neorgan. Khimii. 22:1611-1616 (1977). Abel, E.W., J.M. Pratt and R. Whelan. “Formation of a 1:1 Oxygen Adduct with the Cobalt(1I)-Tetrasulphophthalocyanine Complex,” Chem. Commun. (1971), pp. 449-450. Jones, R.D., D.A. Summerville and F. Basolo. “Synthetic Oxygen Carriers Related to Biological Systems,” Chem. Revs. 79:139-179 (1979). Hoffmann, M.R., and B.C.H. Lim. “Kinetics and Mechanism of the Oxidation of Sulfide by Oxygen: Catalysis by Homogeneous Metal-Phthalocyanine Complexes,” Environ. Sci. Technol. 13:1406-1413 (1979). Boyce, S.D., M.R. Hoffmann, P.A. Hong and L.M. Moberly. “Catalysis of the Autoxi- dation of Aquated Sulfur Dioxide by Homogeneous and Heterogeneous Transition Metal Complexes,” in Acid Rain, Vol. 1, Meteorological Aspects (Ann Arbor, MI: Ann Arbor Science Publishers, 1983). Laidler, K.J., and P.S. Bunting. The Chemical Kinetics of Enzyme Action (Oxford: Clarendon Press, 1973). Carter, J.J., D.P. Rillema and F. Basolo. “Oxygen Carrier and Redox Properties of Some Neutral Cobalt Chelates,” J. Am. Chem. Soc. 96:392-400 (1974). Rollmann, L.D., and S.I. Chan. “Electron Spin Resonance Studies of Low-Spin Cobalt(II) Complexes. Base Adducts of Cobalt Phthalocyanine,”’ Inorg. Chem. 10:1978-1982 (1971). Cookson, D.J., T.D. Smith, J.F. Boas, P-R. Hicks and J.R. Pilbrow. “Electron Spin Resonance Study of the Autoxidation of Hydrazine, Hydroxylamine and Cysteine Catalyzed by the Cobalt(II) Chelate Complex of 3,10,17,24-Tetrasulphophthalocya- nine,” J. Chem. Soc. (1977), pp. 109-114. Przywarska-Boniecka, H., and K. Fried. “The Influence of Additional Ligands on ' Autoxidation of Cobalt and Iron 4,4’ ,4",4”-Tetrasulfonated Phthalocyanines,” Roc- zniki Chemii. 50:43-52 (1976). Basolo, F., B. M. Hoffmann and J.A. Ibers. “Synthetic Oxygen Carriers of Biological Interest,” Accts. Chem. Res. 8:384-392 (1975). Schutten, J.H., and T.P.M. Beelen. “The Role of Hydrogen Peroxide During the Autoxidation of Thiols Promoted by Bifunctional Polymer-Bonded Cobalt Phthalocya- nine Catalysts,” J. Mol. Catal. 10:85-97 (1981). 168 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 116 SOz2, NO AND NOzg OXIDATION MECHANISMS Zika, R., and E. Saltzman. “Interaction of Ozone and Hydrogen Peroxide in Water: Implications for Analysis of HzO in Air,” Geophys. Res. Lett. 9:213-234 (1982). Egerton, A.C., A.J. Everett, G.J. Minkoff, S. Rudrakanchana and K.C. Salooja. “Some Improvements in the Methods of Analysis of Peroxides,” Anal. Chim. Acta 10:422- 429 (1954). Holt, B.D., R. Kumar and P.T. Cunningham. “‘Oxygen-18 Study of the Aqueous Phase Oxidation of Sulfur Dioxide,” Atmos. Environ. 15:557-566 (1981). Frost, A.A., and R.G. Pearson. Kinetics and Mechanism. (New York: John Wiley & Sons, Inc., 1961). . Anast, J.M., and D.W. Margerum. “Trivalent Copper Catalysis of the Autoxidation of Sulfite. Kinetics and Mechanism of the Copper (III/II) Tetraglycine Reactions with Sulfite,” Inorg. Chem. 20:2319-2326 (1981). Edwards, J.0. Inorganic Reaction Mechanisms (Reading, MA: W.A. Benjamin, Inc., 1965). Edwards, J.O., E.F. Greene and J. Ross. “From Stoichiometry and Rate Law to Mechanism,” J. Chem. Educ. 45:381-385 (1968). Bunnet, J.F. “From Kinetic Data to Reaction Mechanism,” in Investigation of Rates and Mechanisms of Reactions Part I, 3rd ed., E.S. Lewis, Ed. (New York: Wiley- Interscience, 1974). Smith, R.M., and A.M. Martell. Critical Stability Constants, Vol. 4, Inorganic Com- plexes (New York, Plenum Press, 1976). Langford, C.H., and H.B. Gray. Ligand Substitution Processes. (Reading, MA: W.A. Benjamin, Inc., 1966). Bernasconi, C.F. Relaxation Kinetics (New York: Academic Press, Inc., 1976). Taube, H. Electron Transfer Reactions of Complex Ions in Solution (New York: Academic Press, Inc., 1970). Fuoss, R.M. “Ion Association III. The Equilibrium Between Ion Pairs and Free Ions,” J. Am. Chem. Soc. 80:5059-5061 (1958). Davis, G.G., and W.F. Smith. “The Kinetics of the Formation of the Monosulphate Complex of Iron(IIT) in Aqueous Solution,” Can. J. Chem. 40:1836-1845 (1962). Basolo, F., and R.G. Pearson. Mechanisms of Inorganic Reactions (New York: John Wiley & Sons, Inc., 1967). Groves, J.T. In: Metal Ion Activation of Dioxygen, T.G. Spiro, Ed. (New York: John Wiley & Sons, Inc., 1979). Mathews, H.J., and L.H. Dewey. “A Quantitative Study of Some Photochemical Effects Produced by Ultra-Violet Light.” J. Phys. Chem. 17:211-218 (1912). Mathews, H.J., and M.E. Weeks. “The Effect of Various Substances on the Photochem- ical Oxidation of Solutions of Sodium Sulfite,” J. Am. Chem. Soc. 39:635-647 (1917). Mason, R.B., and J.H. Mathews. “The Effect of Ultra-Violet Light on the Oxidation of Sodium Sulfite by Atmospheric Oxygen,” J. Phys. Chem. 30:414—420 (1926). Alyea, H.N., and L.J. Backstrém. “The Inhibition Action of Alcohols on the Oxidation of Sodium Sulfite,” J. 4m. Chem. Soc. 51:90-109 (1929). Haber, F., and O.H. Wansbrough-Jones. “Uber die Einwirkung des Lichtes auf Sauer- stoffreie und Sauerstoffhaltige Sulfitlésung,” Z. Phys. Chem. B18:103-123 (1931). Norman, R.O.C., and P.M. Storey. “Electron Spin Resonance Studies. Part XXXI. The Generation, and Some Reactions of the Radicals SO3, 8,03, S° and HS- in Aqueous Solution,” J. Chem. Soc. (B):1009-1013 (1971). Dogliotti, L., and E. Hayon. ‘Flash Photolysis Study of Sulfite, Thiocyanate and Thiosulfate Ions in Solution,” J. Phys. Chem. 72:1800-1807 (1967). 122. 123. 124. 125. 426. 127. 128. 129. 130. 131. 132. 133. 134, 135. 136. 137. 138. 139. 140. 117 CATALYTIC SO2 OXIDATION IN NIGHTTIME FOG 169 Backstrém, H.L.J. “The Chain Reaction Theory of Negative Catalysis,” J. Am. Chem. Soc. 49:1460-1472 (1927). , Bickstrém, H.L.J. “Der Kettenmechanisms bei der Autoxydation von Aldehyden,” Z. Phys. Chem. (Leipzig) B25:99-121 (1934). Abel, E. “Zur Theorie der Oxydation von Sulfit zur Sulfat durch Sauerstoff,” Mon- atsch. Chem. 82:815-834 (1951). Matsuura, A., J. Harada, T. Akehata and T. Shirai. “Rate of Ammonium Sulfite Oxidation in Aqueous Solution,” J. Chem. Eng. Japan 2:199-203 (1969). Lunak, S., and J. Veprek-Siska. “Photochemical Autoxidation of Sulfite Catalyzed by Iron(III) Ions,” Coll. Czech. Chem. Commun. 41:3495-3503 (1976). Lunak, S., and J. Veprek-Siska. “Catalytic Effect of Metal Ions on Photochemical Reactions of Molecular Oxygen. III. Photochemical Autoxidation of Sulfite Catalyzed by Manganese(II) Ions,” Reac. Kin. Catal. Lett. 5:157-161 (1976). Veprek-Siska, J., S. Lunak and A. El-Wakil. “Catalytic Effect of Ferric Ions on the Photoinitiated Autoxidation of Sulfite,” Z. Naturforsch. 29B:812-813 (1974). Balzani, V., and V. Carassiti. Photochemistry of Coordination Compounds (New York: Academic Press, Inc., 1970). Langford, C.H., M. Wingham and V‘S. Sastri. “Ligand Photooxidation in Copper(II) Complexes of Nitrilotriacetic Acid,” Environ. Sci. Technol. 7:820-822 (1973). Langford, C.H., and J.H. Carey. “The Charge Transfer Photochemistry of the Hexa- aquoiron(IIT) Ion, the Chloropentaaquoiron(III) Ion, and the p-Dihydroxo Dimer Explored with tert-Butyl Alcohol Scavenging,” Can. J. Chem. 53:2430-2435 (1975). Lockhart, H.B., Jr., and R.V. Blakeley. “Aerobic Photodegradation of Fe(III)- (Ethylenedinitrilo) Tetraacetate (Ferric EDTA),” Environ. Sci. Technol. 12:1035-1038 (1975). Miles, C.J., and P.L. Brezonik. “Oxygen Consumption in Humic-Colored Waters by a Photochemical Ferrous-Ferric Catalytic Cycle,” Environ. Sci. Technol. 15:1089- 1095 (1981). Baker, A.D., A. Casadavell, H.D. Gafney and M. Gellender. “Photochemical Reac- tions of Tris(oxalato) Iron(III),” J. Chem. Educ. (1980), pp. 314-315. Taylor, D.D., and R.C. Flagan. ‘“‘Aerosols from a Laboratory Pulverized Coal Com- bustor,” in Atmospheric Aerosol: Source/Air Quality Relationships, E.S. Macias and P.K. Hopke, Eds., ACS Symp. Ser No. 167 (Washington, DC: American Chemical Society, pp. 157-172 (1981). Ouimette, J.R., and R.C. Flagan. “Chemical Species Contributions to Light Scattering by Aerosols at a Remote Arid Site,” in Atmospheric Aerosol: Source/Air Quality Relationships, E.S. Macias and P.K. Hopke, Eds., ACS Symp. Ser. No. 167 (Washing- ton, DC: American Chemical Society, 1981), pp. 125-156. Adamson, A.W., and P.D. Fleischauer. Concepts of Inorganic Photochemistry (New York: Wiley-Interscience, 1975). Brodzinsky, R., 5.G. Chang, $8.8. Markowitz and T. Novakov. “Kinetics and Mecha- nism for the Catalytic Oxidation of Sulfur Dioxide on Carbon in Aqueous Suspen- sions,” J. Phys. Chem. 84:3354-3358 (1980). Chang, S.G., R. Toosi and T. Novakov. “The Importance of Soot Particles and Nitrous Acid in Oxidizing SO. in Atmospheric Aqueous Droplets,” Atmos. Environ. 12:297-306 (1981). Cohen, S., S. Chang, S. Markowitz and T. Novakov. “Role of Fly Ash in Catalytic Oxidation of S(IV) Slurries,” Environ. Sci. Technol. 15:1498-1502 (1981). 170 141. 142. 143. 144. 145. 146. 147, 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 118 $OQ2, NO AND NOz OXIDATION MECHANISMS Kim, K.K., and J.S. Choi. “Kinetics and Mechanism of the Oxidation of Sulfur Dioxide on a-Fe2Os3,” J. Phys. Chem. 85:2447-2450 (1981). Faust, B.C., and M.R. Hoffmann. “Photocatalytic Oxidation of Dissolved Sulfur Dioxide on FeO, Surfaces” (in progress). Childs, A.L. “Studies in Photoassisted Heterogeneous Catalysis,” PhD Thesis, Prince- ton University, Princeton, NJ (1981). Childs, L.P., and D.F. Ollis. “Is Photocatalysis Catalytic?” J. Catal. 66:383-390 (1980). Benner, W.H., R. Brodzinsky and T. Novakov. “Oxidation of SOz in Droplets Which Contain Soot Particles,” Atmos. Environ. 16:1333-1340 (1982). Colton, F.A., and G. Wilkinson. Advanced Inorganic Chemistry (New York: John Wiley & Sons, Inc., 1980). Roberts, J.D., 8. Ross and M.C. Caserio. Organic Chemistry (Menlo Park, CA: W.A. Benjamin, Inc., 1971). Fridovich, I., and P. Handler. “Detection of Free Radicals in Illuminated Dye Solu- tions by the Initiation of Sulfite Oxidation,” J. Biol. Chem. 235:1835-1838 (1960). McCord, J.M., and I. Fridovich. “The Utility of Superoxide Dismutase in Studying Free Radical Reactions,” J. Biol. Chem. 214:6056-6063 (1969). Srinivasan, V.S., D. Podolski, N.J. Westrick and D.C. Neckers. “Photochemical Gen- eration of O7 by Rose Bengal and Ru(bpy)?*,” J. Am. Chem. Soc. 100:6513-6515 (1978). Zika, R.G. “Photochemical Generation and Decay of Hydrogen Peroxide in Seawa- ter,” Trans. Am. Geophys. Union 61:1010 (1980). Kok, G.L., T.P. Holler, M.B. Lopez, H.A. Nachtrieb and M. Yuan. ‘“‘Chemilumines- cent Method for Determination of Hydrogen Peroxide in the Ambient Atmosphere,” Environ. Sci. Technol. 12:1072-1076 (1978). Kok, G.L., K.R. Darnall, A.M. Winer, J.N. Pitts, Jr. and B.W. Gay. “Ambient Air Measurements of Hydrogen Peroxide in the California South Coast Basin,” Envi- ron. Sci. Technol. 12:1077-1080 (1978). Kok, G.L. “Measurements of Hydrogen Peroxide in Rainwater,” Atmos. Environ. 14:653-656 (1980). Kok, G.L. “‘Measurements of Hydrogen Peroxide in Rainwater,” EOS Trans. Am. Geophy. Union 45:884 (1981). Walling, C. “Fenton’s Reagent Revisited,” Accts. Chem. Res. 12:125-131 (1975). Baxendale, J. H. “Decomposition of Hydrogen Peroxide by Catalysts in Homogeneous Aqueous Solution,” Adv. Catal. 4:31 (1952). Uni, N. “Inorganic Free Radicals in Solution,” Chem. Rev. 50:375 (1952). Barrie, L.A. “An Experimental Investigation of the Absorption of Sulfur Dioxide by Cloud and Raindrops Containing Heavy Metals,” PhD Thesis, University of Frank- furt (1975). Penkett, S.A. “Oxidation of SO, and Other Atmospheric Gases by Ozone in Aqueous Solution,” Nature, Phys. Sci. 240:105-106 (1972). Erickson, R.E., L.M. Yates, R.L. Clark and D. McEwen. “'The Reaction of Sulfur Dioxide with Ozone in Water and Its Possible Atmospheric Significance,” Atmos. Environ, 11:813-817 (1977). Espenson, J.H., and H. Taube. “Tracer Experiments with Ozone as Oxidizing Agent in Aqueous Solution,” Inorg. Chem. 4:704-709 (1965). Martin, L.R., D.E. Damschen and HS. Judeikis. “The Reactions of Nitrogen Oxides with SO2 in Aqueous Solution,” Atmos. Environ. 15:191-195 (1981). Oblath, S.B., S.S. Markowitz, T. Novakov and S.G. Chang. “Kinetics of Formation 165. 166. 167. 168. 169. 170. 171, 172. 173. 174. 175. 176. 177. 178. 179. 180. 181, 182. 183. 184, 119 CATALYTIC SOzg OXIDATION IN NIGHTTIME FOG 171 of Hydroxylamine Disulfonate by Reaction of Nitrite with Sulfites,” J. Phys. Chem. 85:1017-1021 (1981). , Ross, D.S., C.L. Gu, G.P. Hum and D.G. Hendry. “Catalytic Oxidation of SO2 by NO, in Aqueous Systems” (submitted). Gritzel, M., S. Taniguchi and A. Henglein. “Pulsradiolytisch Untersuchung der NO- Oxydation und des Gleichgewichts NO; = NO + NO, in Wassriger Lésung,” Ber. Bunsenges. Phys. Chem. 74:488-492 (1970). Takeuchi, H., M. Ando and N. Kizawa. “Absorption of Nitrogen Oxides in Aqueous Sodium Sulfite and Bisulfite Solutions,” Ind. Eng. Chem. Des. Dev. 16:303-308 (1977). Nash, T. “The Effect of Nitrogen Dioxide and of Some Transition Metals on the Oxidation of Dilute Bisulfite Solutions,” Atmos. Environ. 13:1149-1154 (1979). Morel, F.M., and J.J. Morgan. “A Numerical Method for Computing Equilibria in Aqueous Chemical Systems,” Environ. Sci. Technol. 6:58-67 (1972). Hoffmann, M. R. “Thermodynamic, Kinetic and Extrathermodynamic Considerations in the Development of Equilibrium Models for Aquatic Systems,” Environ. Sci. Tech- nol. 15:345-353 (1981). Westall, J.C., J.L. Zachary and F.M. Morel. “MINEQL, A Computer Program for the Calculation of Chemical Equilibrium Composition of Aqueous Solutions,” Tech. Note 18, Department of Civil Engineering, Massachusetts Institute of Technology, Cambridge, MA (1976). Westall, J.C., and H. Hohl. ‘A Comparison of Electrostatic Models for the Oxide/ Solution Interface,” Adv. Coll. Interface Sci. 12:265-294 (1980). Hansen, L.D., L. Whiting, D.J. Eatough, T.E. Jensen and R.M. Izatt. “Determination of Sulfur(IV) and Sulfate in Aerosols by Thermometric Methods,” Anal. Chem. 48:634-638 (1976). Eider, R.C., M.J. Heeg, M.D. Payne, M. Trkula and E. Deutsch. “Trans Effect in Octahedral Complexes. 3. Comparison of Kinetic and Trans Effects Induced by Coor- dinated Sulfur in Sulfito- and Sulfinatopentaamminecobalt(III) Complexes,” Inorg. Chem. 17:431-440 (1978). Raston, C.L., A-H. White and J.K. Yandell. “Structural and Kinetic Effects in Cobalt(III) Complexes with Cobalt-Sulfur Bonds. The Crystal Structure of trans- Bis(ethylenediamine)-imidazolesulfitocobalt(IIT Perchlorate Dihydrate,” Aust. J. Chem. 31:993-998 (1978). Magnusson, A., L.G. Johansson and O. Lindqvist. “The Structure of Manganese(II) Sulfite,” Acta Cryst. B37:1108-1110 (1981). Johansson, L.G., and E. Ljungstrom. “Structure of Iron(II) Sulfite 2.5 Hydrate,” Acta Cryst. B36:1184—-1186 (1980). Johansson, L.G., and O. Lindqvist. “Manganese(II) Sulfite Trihydrate,” Acta Cryst. B36:2739-2741 (1980). Baldwin, M.E. “Sulphitobis(ethylenediamine)Cobalt(II]) Complexes,” J. Chem. Soc. (1961), pp. 3123-3128. Bordwell, F.G. Organic Chemistry (New York: Macmillan Co., 1963). Dasgupta, P.K., K. DeCesare and J.C. Ulirey. “Determination of Atmospheric Sulfur Dioxide without Tetrachloromercurate(II] and the Mechanism of the Schiff Reaction,” Anal. Chem. 52:1912-1922 (1980). Fortune, C.R., and B. Dellinger. “Stabilization and Analysis of Sulfur(IV) Aerosols in Environmental Samples,” Environ. Sci. Technol. 16:62-66 (1982). Pankow, J.F., and 1.J. Morgan. ‘Kinetics for the Aquatic Environment,” Environ. Sci. Technol. 15:1155-1164,1306-1313 (1981). Adamowicz, R.F. “A Model for the Reversible Washout of Sulfur Dioxide, Ammonia 172 185. 186. 187. 188. 189. 190. 191. 192. 193. 194. 195, 196. 197. 120 SO2z, NO AND NOQ2 OXIDATION MECHANISMS and Carbon Dioxide from a Polluted Atmosphere and the Production of Sulfates in Raindrops.” Atmos. Environ. 13:105-121 (1979). Baboolal, L.B., H.R. Pruppacher and J.H. Topalian. “A Sensitivity Study of a Theoret- ical Model of SO. Scavenging by Water Drops in Air,” J. Atmos. Sci. 38:856-870 (1981). Durham, J.L., J.H. Overton and V.P. Aneja. “Influence of Gaseous Nitric Acid on Sulfate Production and Acidity in Rain,” Atmos. Environ. 15:1059-1068 (1981). Easter, R.C., and P.V. Hobbs. “The Formation of Sulfates and the Enhancement of Cloud Condensation Nuclei in Clouds,” J. Atmos. Sci. 31:1586-1594 (1974). Seinfeld, JH. Lectures in Atmospheric Chemistry, American Institute of Chemical Engineers Monograph Series 76 (1980). Graham, R.A., and H.S. Johnston. “The Photochemistry of NOs; and the Kinetics of the N2O5-O3 System,” J. Phys. Chem. 82:254-268 (1978). Baulch, D.L., R.A. Cox, R.F. Hampson, Jr., J.A. Kerr, J. Troe and R.T. Watson. “Evaluated Kinetic and Photochemical Data for Atmospheric Chemistry,” J. Phys. Chem. Ref. Data 9:295-471 (1980). Richards, L.W. ““Comments on the Oxidation of NO to Nitrate—Day and Night,” Atmos. Environ. (in press). Schwartz, S.E., and J.E. Freiberg. “‘Mass-Transport Limitation to the Rate of Reaction of Gases in Liquid Droplets: Application to Oxidation of SO2 in Aqueous Solutions,” Atmos. Environ. 15:1129-1144 (1981). Freiberg, J.E., and S.E. Schwartz. “Oxidation of SO, in Aqueous Droplets: Mass- Transport Limitation in Laboratory Studies and the Ambient Atmosphere,” Atmos. Environ. 15:1145-1154 (1981). Rodhe, B. “The Effect of Turbulence on Fog Formation,” Tellus 14:49-86 (1962). McRae, G.J. Personal communication. Kirk-Othmer Encyclopedia of Chemical Technology, Vol. 16, 3rd ed., (1981), p. 701. Beutier, D., and H. Renon. “Representation of NH3-H2S,H2O-NH;-CO,-H20, and NH,-SO,-H20 Vapor-Liquid Equilibria,” Ind. Eng. Chem. Process. Des. Dev. 17:220- 228 (1978). 121 CHAPTER V A DYNAMIC MODEL FOR THE PRODUCTION OF HY, NO, » AND so,77 IN URBAN FOG by Daniel J. Jacob and Michael R. Hoffmann Journal of Geophysical Research 88, 6611-6621 (1983) 122 JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 88, NO. C11, PAGES 6611-6621, AUGUST 20, 1983 A Dynamic Model for the Production of H+, NO,~, and SO,?~ in Urban Fog Dane J. JacoB anD MICHAEL R. HOFFMANN "Environmental Engineering Science, W. M. Keck Laboratories, California Institute of Technology, Pasadena, California 91125 The chemica) composition of nighttime urban fog has been investigated using a hybrid kinetic and equilibrium model. Extremely high acidity may be imparted to the droplets by condensation and growth on acidic condensation nuctei or by in situ SIV) oxidation. Important oxidants of S(IV) were found to be ©, as catalyzed by Fe(III) and Mn{Il), H,0,, and O,. Formation of hydroxymethanesulfonate ion (HMSA) via the nucleophilic addition of HSO,~ to CH,O(I) significantly increased the droplet capacity for S(IV) but did not slow down the net S(IV) oxidation rate beading to fog acidification. Gas phase nitric acid, ammonia, and hydrogen peroxide were scavenged efficiently, although aqueous phase hydrogen peroxide was depleted rapidly by reduction with S{IV). Nitrate production in the aqueous phase was found to be dominated by HNO, gas phase scavenging. Major aqueous phase species concentrations were controlled primarily by condensation, evaporation, and pH. INTRODUCTION Concentrations of major ions in nonprecipitating clouds {Hegg and Hobbs, 1981} and fogs [Waldman et al. 1982; Munger et al., 1983] have been reported to be significantly higher than those commonly observed in acidic precipitation. In Los Angeles, fogwater [Waldman et al, 1982; Munger et al., 1983] was reported to have acidities 100 times higher than those observed previously in rainwater by Liljestrand and Morgan [1981]. Lower dilutions and higher scavenging ef- ficiencies due to reduced mass transfer limitations of gas ab- sorption and longer residence times may explain, in part, the higher concentrations found in fog (1-100 ym) than in rain (0.1-3.0 mm). A number of rainwater and cloudwater chemistry models have been proposed recently. Adamowicz [1979] simulated the chemistry of raindrops falling through a well-mixed polluted layer with uniform and constant concentrations of SO,, CO,, and NH,. Gas and aqueous phase equilibria were established at al] times, and aqueous phase transformation of S(IV} to S(VI) was allowed to proceed through the iron-catalyzed oxi- dation by O, according to the kinetic expression of Brimble- combe and Spedding [1974]. Mass transfer at the surface of the drop was modeled by two-film theory, but Baboolal et al. [1981] have since shown this simple model to be unsatis- factory since it ignores forced convection inside and outside of the falling drop. Durham et al. [1981] added NO, to the gas phase and allowed some NO, gas phase chemistry, considered kinetic expressions for all reactions instead of equilibrium re- lationships, and assumed O, to be the only liquid phase oxi- dant of S(IV) using the rate law of Erickson et al. [1977]. Easter and Hobbs [1974] modeled cloudwater chemistry by using a wave cloud model, an open atmosphere with trace concentrations of CO,, SO,, and NH;, and a rudimentary S(IV) oxidation rate consisting of a simple first-order depen- dence on sulfite. More recent models have been proposed by Middleton et al. (1980), Chameides and Davis [1982], and Car- michael et al. (1983). With this work in mind, a dynamic mode! for fogwater Copyright 1983 by the American Geophysical Union. Paper number 300722. 0148-0227/83/003C-0722$05.00 chemistry has been developed. The model has a hybrid kinetic and equilibrium structure: reactions which rapidly come to equilibrium are considered separately from reactions that are kinetically controlled. Gas phase chemistry, particle scav- enging by droplets, evolution of the droplet microphysics, and deposition were not included explicitly. STRUCTURE OF THE MODEL Aqueous Phase Reactions The chemical composition of a fog droplet is assumed to be determined by the following factors: (1) the composition of the activated cloud condensation nuclei (CCN) on which the droplet condenses, (2) the absorption of atmospheric gases at the droplet surface, and (3) the subsequent aqueous phase reactions of homogeneous and heterogeneous species. Proton transfer and most ligand substitution reactions proceed ex- tremely fast compared to the time scales of interest in this study [Haffmann, 1981]; therefore they are treated as dynamic equilibria. The same assumption is applied to gas absorption in accordance with Henry's law, although mass transfer may be retarded by the formation of an organic film at the droplet surface [Graedel et al., 1983]. Equilibrium constants K, have been adjusted for temperature T with the van't Hoff relation- ship: 1¢ o T [iain x - 23 [ aT a) 2 K200 298 T AH?® values at 298°K were obtained from literature sources (see Tabie 1). The equilibrium composition was determined using a MINEQL subroutine [Morel and Morgan, 1972; Westall et al., 1976}. In MINEQL the equilibrium constant approach is used to solve the chemical equilibrium problem, which is defined by a system of mass action equations. The computed concentrations of constituents are constrained to femain positive and to satisfy mole balance relationships pro- vided by the analytical information. Given a set of chemical constituents which have been defined operationally as metals and ligands, along with the corresponding stoichiometric and thermodynamic data, all the possible chemical species in a model system can be defined. The concentrations of these chemical species are written as functions of the free con- 611 123 6612 JacOs AND HOFFMANN: MopeL oF FoGwaTer CHEMISTRY TABLE 1. Henry's Law and Aqueous-Phase Equilibria Relevant to the Droplet Chemistry Reaction AM 15 . No. Reaction pK* kcal mole~* Referencet (Rt) H,0()=H* + OH” 14.00 13.35 SM (R2) SO,(g) + H,O=S0,-H,0 0.095 -625 SM (R3) SO,/H,O=H* + HSO,~ 1.89 ~4.16 SM Ra) HSO,~=H* + S0,?- 722 -223 SM «R5) HNO,(g)=H* + NO,” ~651 17.3 sw (R6) HNO,{g)= HNO,() ~-17 -9.5 sw (R3) HNO,()=H* + NO,~ 3.29 25 sw (R8) C08) + H,O=CO,-H,0 147 4.85 SM 9} CO,-H,O=H* + HCO,- 637 183 SM R10) HCO,” =H* + CO,?- 10.33 3.55 SM (Ri) CH, Og) + H,JO=CH,O-H,0 385 ~ 1285 LB (R12) HOCH,SO,H=H* + HOCH,SO,~ <0t Us R (R13) HOCH,SO,~ =H‘ + ~OCH,SO,~ 117 v SA (R14) NH,(g) + H,O=NH,-H,O 177 -817 SM (R15) NH,-H,O=NH,* + OH 47 09 SM R16) 042)=0,0) 290 358 P (R17) H,0(g)=H,0,0) —485 145 MD R18) O,e)=0,0) 2.03 5.04 LB (R19) CaHCO, * =Ca?* + HCO,” 11.6 —2.78 SM (R20) CaSO,(1)= Ca?* + S0,?~ 230 —165 SM (R21) NaSO,~ = Na* + S$0,?~ 0.70 —2.23 SM (R22) FeSO, * = Fe?* + S0,?- 420 5.4 SM (R23) Fe(SO,),” = Fe?* + 280,?— 5.60 u SM (R24) FeCi?* =Fe?* + Ci- 1.40 “—791 SM (R25) FeOH?* = Fe?* + OH™ 1230 0.04 SM (R26) Fe(OH), * = Fe?* + 20H 23.3 U SM (R27) Fe(OH), | = Fe?* + 30H™ 39.0 20.7 SM (R28) Fe(OH), *” =¢2Fe’* + 20H- 25.7 16.2 SM (R29) FeSO, * =Fe?* + S0,?- 10.0 U this laboratory RH) MnSO,()= Mn?* + SO,?— 230 —3.39 SM (R3)) MnCl’ = Mn?* + Ci- 1.10 —8.01 SM (R32) HSO,” =H* + SO,?~ 220 —491 SM *K is in M atm™' or M*. Temperature is 298°K. Reference code: SM = Sillén and Martell [1964]; SW = Schwartz and White [1981]; LB = Ledbury ond Blair [1925]; SA = Sorensen and Andersen [1970]; MD = Martin and Damschen [1981]; L-B = Landolt-Bérnstein [1976]; R « Roberts et al. [1971]; P = Perry [1963]. The pK for this reaction is very low. §Unknown, AH = 0 is assumed in the calculation. centrations of the constituents by mass action equations. These functions are substituted into the mole balance equa- tions with elimination of solids. The problem is reduced to a system of nonlinear equations in which the unknowns are the concentrations of the metals and ligands. This system of equa- tions is solved by the Newton-Raphson method. The initial solution is tested against the solubility products of the solids and a new set of solids, which includes those solids with the most exceeded solubility products, is selected for the next computation. A final solution is achieved when the difference between the imposed analytical concentrations for a constit- went and the sum of all individual species containing that _ constituent is less than or equal to a set value which is very small. Corrections for ionic strength were made using the Davies equation [Stumm and Morgan, 1981]. The thermody- namic data base consisted of 1300 equilibria; those found to influence the droplet composition are listed in Table 1. , For slower reactions, empirical rate laws were used. Specific fate laws for oxidation reactions involving N(III) and S(IV) have been incorporated in the model; they are listed in Tabie 2 Hydrogen peroxide, N(III), and O, sre potentially impor- tant oxidants, and O, may also be important if catalyzed by active sites on soot or by transition metals, such as Fe(III) and Mni(ll). Iron{11l) and manganese{Il) are effective catalysts when dissolved as free ions or complexes [Hoffmann and Jacob, 1983] but their catalytic properties are altered if they are present as solid phases. Surface catalysis may occur but in the absence of reliable data it has been neglected. In the kinet- ic formulation of (R35), [Fe(II] and [Mn(1I)] represent the summation of concentrations of all dissolved iron and manga- nese species. It should be noted that the kinetic data for this reaction is not satisfactory around pH 4 because of the influ- ence of Fe(OH),(s) which starts dissolving near pH 4. Rate expressions given by Martin [1983] at high pH and low pH fai] to extrapolate to the same value at intermediate pH. To minimize this problem, an average of the two expressions was used near pH 4. The reactions of absorption and aqueous phase dispro- portionation of NO, to form nitrous and nitric acid have been reviewed by Schwartz and White [1981]. These reactions could be major contributors to nitrate formation in fog drop- jets if allowed to reach equilibrium; however, because of second-order kinetics they are too slow to be important on the time scales of concern in fog. Alternative aqueous phase ni- trate formation pathways involve the oxidation of N(III) by H,0, and O,[Damschen and Martin, 1982). Overall, the rate laws shown in Table 2 indicate that oxida- tion reactions proceed on time scales of minutes to hours under urban atmospheric conditions. These rate laws are inte- grated with a simple finite difference scheme using adjustable 124 ERRATA p. 6613, reactions tions (R38) and (R39): “"d[S(VI)J" _ "A[N(v) J" 6613 d uu should read NV TABLE 2a. Kinetic Expressions sor the Aqueous Phase Oxidation Reactions of S(IV) to S(V1) and N(III) to. N(V) : _ _Activation > Reaction — : Energy, kca! No. «Reaction Conditions Rate, M37! at 25°C* mole! Referencet : a ea). 8 x 10*TH,OM)[S0,0)) . 33) SQV) + H,0, 0.1 +{H*) . . 13 M a +0, es son 19 W180 aaf0,0) ‘ M pH>3 aso = 419 x 10°(1 + a x 10" 2 FP NOsmitso,a0) 6 Ma a3) SIY4+0, [S0,{aq)] > 10-9, pH <4 aso), 471M0D)? (° Paso 218 M (with Fe (111) and fH") {H°] Moill)) (1 + Ae TMaty rota ) &3 x 10-* + [Fe(lit)) [(S0,faq)] < 1075, pH <4 ph =- 2{ soot Maan ]CHS0, “]+ oe eos 218 M [$0,faq)] > 10°, pH > 4 au - ce + 1 x 10°[Fe(HIN][50 (aq)}? 273 M £S0,t0q)} <10-% p> 4 ESET) ~ so00 pmtaangtHs0,-} 23 M 36) gavy+0, aso SSO) 2 54 x 107f0,0)" (with soot)t - , {SOxeq) __ . 7 B + 306 x 1050 ,faq)] + 15 » 10500) 37) saya NaI) pR<3 ae = 1420“ JDINO,(a9)}1S0,(29)) 125 M PH> 3 4509). 227NO,MITHSO,~} 125 o 38) Nal) + H,0, el = 46 x 10°TH*}{H,0,0))[HNO,()] 2 DM @2) NaM)+0, a) = 5 x 10°[0, INO, -} 138 DM *{SO,(aq)j = [SO,()] + [HSO,~] + [SO,?~); [HNO,(aq)} = [HNO,()] + [NO,~}. Mente M = Martin [1982]; Ma = Maahs [1982]; B = Brodzinsky et al. [1980]; O = Oblath et al. [1981]; DM = Damschen and artin 5 tC, is the concentration of active carbon in g I~ '. §Assumed. time steps raging from 10 s to 200s: constant for HMSA to be Kyo, = k,/k, = 7.5 x 10¢ M7', while S. D. Boyce and M. R. Hoffmann (unpublished manu- d[RJ\ - : script, 1983) have determined that ky = 3 M~' s~'. Pro- CRd.+a = [Rd + (x (= ye (2) duction of HMSA will contribute to the droplet acidity in- directly by producing a droplet that is supersaturated with where [R,] represents the concentration of the reduced com- Tespect to SO,(g), and because of the slow reversibility of (3), ponent R; (ie. N(IID) or S(IV)). HMSA formation may inhibit the production of S(VI). Reac- Of special interest in urban fog is the reaction of HSO,~ _‘0n (3) is integrated at each time step by using the integrated with CH,O and with other aldehydes, RCHO, to form hy- form of the rate law: droxymethanesulfonate ion (HMSA) and the corresponding sulfonates, HOCHRSO,~ [Munger et al., 1983]: in (HSO,7},.4.—P nate Enea zk — EO | Aq— p) Ar ay CH,O + HSO,” = HOCH,SO,~ )] . ‘in tHS0s" 1-2 “ Dasgupta et al. [1980] determined a conditional equilibrium [HSO,"],-4 ; TABLE 2b. Reaction Stoichiometries for Reactions in Tabie 2c Reaction . No. Reaction Stoichiometries (R33) HSO,” + H,0,)— HSO,~ + H,O0 R34) HSO,” + 0,0) HSO,~ +O, (R35) HSO,~ + 1/20,()— HSO,~ (R37) 2HNO,() + H,O-SO,— 2NO + H* +HSO,” +H,0 ; : 2HNO,(I) + 2H,0 SO, ~ 2S0,? ~ +N,0+H,0 +4 4H* (R38) : HNO,() + H,0,0)> ao, +H,0+H* (R39) NO,~ +0,0)— NO,” + 125 14 — (HSO, "J... — [HSO,"}) (5) where p and q are the solutions to the second-degree equation KX? + (K((CH,O()], ~ [HSO,"1) + 0X — ((HSO,~j,+ [HMSA},=0 (6) Gas Phase and Aerosol At night the oxidation of nitrogeneous compounds and al- kenes by ozone is important. The kinetics of these reactions {HMSA]., 4: = [HMSA], are dependent on the downward diffusion of ozone to the boundary layer [McRae, 1981]. In the boundary layer, nitric acid may be produced [Graham and Johnston, 1978; McRae and Russell, 1983] by a variety of pathways. No attempt has been made to model gas phase chemistry. Instead, con- centrations of gases and gas phase production rates (from emissions and homogencous reactions) have been estimated from local field data and predictions from the Caltech air quality model for the Los Angeles basin [McRae, 1981). The integrated mass and composition of the activated cloud condensation nuclei on which the fog droplets condense were estimated from field data. No subsequent scavenging of the interstitial aerosol by diffusion or impaction was considered. Physical Description A parcel of air is followed in which droplets form and grow by accretion of water vapor. The droplets are assumed to remain constantly within the air parcel (i.c., sedimentation and diffusion of the droplets are ignored). The limitation of S(1V) oxidation rates by mass transfer has been discussed by Schwartz and Freiberg [1981] and Baboolal et al. [1981]. Schwartz and Freiberg have shown that for stationary droplets smaller than 50 ym the net rate of oxida- tion was limited strictly by oxidation. Baboolal et al. extended this analysis to droplets larger than 50 jm. These large drop- lets have a significant sedimentation velocity, which drives convection both inside and outside of the droplet. This mixing effect enhances the rate of mass transfer as calculated for stationary droplets. Therefore the chemical changes in fog droplets are most likely limited by the specific reaction rates. Because fogs are localized events and occur on time scales of a few hours, advection of condensing and evaporating drop- lets can be ignored as a first approximation. Also, evolution of the droplet spectrum through coagulation can be neglected since mass transfer does not limit overall reaction rates. In this context the liquid water content of the air parcel and its time dependency are sufficient parameters for characterizing fog microphysics. At t= 0, droplets are assumed io condense, receive an in- itial chemical loading from the water-soluble fraction of the activated nuclei, and immediately react with the gaseous envi- ronment. The growth and evaporation of the droplets is simu- lated by the external input of an evolving liquid water content. - At each time step the equilibrium composition of the droplet is calculated along with the HMSA formation rate and the changes in component concentrations due to oxidation reac- tions. RESULTS AND Discussion _ Sulfate and Nitrate Oxidation Pathways In this section the relative contributions of the various + Sv) Jacos anD Horrmann: Mone: of FoGwater CHEMISTRY and N(III) oxidation mechanisms to the droplet chemistry are discussed. For this specific purpose a closed parcel of air with a constant liquid water content has been assumed (i.e. drop- kets condense on the activated condensation nuciei at 1 = 0 with no droplet growth or evaporation). A nighttime air mass typical of the industralized coastline of the Los Angeles basin has been chosen. Concentrations of gases and preexisting nuclei were estimated from local field data and transport models (Table 3). Gases are depleted by absorption except for CO, and Q,, which are held constant at a fixed partial pres- sure. The liquid water content is L = 0.1 g m~* and the tem- perature is T = 283 K. _Kinetic results shown in Figure ! clearly show that H,O, and O, (catalyzed by Fe and Mn) are the principal oxidants for S(VI) production in situ. Hydrogen peroxide is highly sol- uble in water and rapidly oxidizes S(IV) at low pH; however, if not replenished, it is quickly depleted from the parcel of air by reduction to water. In this case, in the absence of gas phase or aqueous phase H,O, formation, no H,O, remains in the system after 10 minutes. The catalytic effectiveness of both Fe(III) and Mni(I) is de- pendent on their speciation in the droplet (Figure 2). The principal manganese species are Mn(H,0),7* and MnSO,(aq). Both of these species are assumed to be equally effective as catalysts for S(IV) autoxidation. On the other hand, Fe(III) above pH 4 is found primarily as an iron(II) hydroxide solid, which has been assumed to be catalytically inactive. As the pH decreases below 4, the solid phase dis- solves and soluble Fe(III), the active catalyst, is released. The rate law indicates a decrease of the oxidation rate with pH but the dissolution of Fe(III) at low pH offsets this effect and the calculated rate actually increases between pH 4 and 3.5. Oxidation by ozone contributes ~4% of the total S(VI) TABLE 3. Composition of the Air Mass (Trace Gases and Con- densation Nuciei) Prior to Fog Formation at a Polluted Site Atmospheric Trace Gases Concentration, ppb . Reference so, 20 SCAQMD data* HNO, 1 Hans: et al. (1982} HNO, 3 id. NH, $ id. CH,0 » Grosjean [1982] oO, 10 SCAQMD data H,O, 1 Graedel et al. [1976] Condensation Nucleit Concentration Concentration, ng mm”? ee m=? $o,?- 10 NH,* 6.65 NO,” 10 Na* 61 a Li Fe(II) “48 co,?- 1.83 MniIl) 02 CAsoot) % Ca?* 1.22 The air mass is typical of the industrial coastline of the Los Angeles 30 nes for an inversion height of about 50 m. Pp, = 0.2] atm. Poo, = “Dats from the permanent records of the South Coast Air Quality Management District. +(NH,),S0,, NH,NO,, NaCl, CaCO,, metal oxides, and soot. Ref- erences: Gartrell et al. [1980]. Appel et al, (1980). 126 Jacos AND HOFFMANN: MODEL oF FoGWATER CHEMISTRY CONCENTRATION (mM) 3 Tv T T t T (a) 2 “ 1 “ ° 6 (b) 5 = = 4 > o 3 4 2 L l L 1 r ° 4000 g000 12000 TIME ( sec.) Fig. 1. (a) Profile versus time of total sulfate in the Sogwater and of the individual contributions to the total sulfate of sulfate aerosol and different S(IV} oxidants. (b) Profile of pH versus time. The fog formed under the conditions of Table 3, with liquid water content = Q.1 gm~?, temperature = 10°C. produced after 3 hours of fog. This is due to the Jow water solubility of ozone and the reciprocal hydrogen ion depen- dency of its reaction rate. Therefore little oxidation by ozone occurs below pW 4. Catalytic carbon sites in soot contribute <5% of the total SCVI) produced, even at the high carbon Jevels of urban environments. Oxidation by N(ITI) appears to be insignificant due to the low solubility of HNO,(g) at pH <4. Oxidation of N(III) does not contribute appreciably to ni- trate production. At 1 = 180 minutes, reaction with O, had generated less than 10°’ M NO,~ and reaction with H,O, less than 10-'° M NO,~. The aqueous phase oxidation of S(IV) has an important Consequence on the droplet acidity. Between pH 2 and 6 the Principal reactive species of S(IV) is HSO,~ [Hoffmann and Jacob, 1983). The oxidation product, HSO,~, is completely dissociated above pH 3. Therefore the oxidation directly pro- duces one free proton. Depletion of HSO,~ beads to further dissolution of SO,(g) which then dissociates to give HSO,~ and H* in the droplet. Because the solubility of SO, decreases with pH, each depleted HSO,~ molecule is not replaced in the Groplet. Depending on the overall droplet chemistry, oxida- tion of one molar unit of S(IV) thus leads to the production of One- to two-fold that molar unit of free acidity. The impact on the droplet pH is dramatic, as shown in Figure 1b. In the first half hour of the fog, S(VI) production is the fastest and causes & pH drop of over two units, Field observations are consistent with these predictions [Munger et al. 1983). Components in the gas phase are scavenged by the droplets With efficiencies dependent on their Henry's law constants and 6615 16 T T T T T ~~: (a) (=| Fe(OH), goer “ — E (a FalOn,® 06+ / mm 4 % y sone Ny . z ee o 04 : a _ - S i + a / Fe504 > oe i a re wre ton)? Fe(SO4)2- ' i . ° / 4s FeSOs", - b. ot . 1.0] T T T he {b) ~ O8F wnt? 4 a s =, 06h = J e e 04 & 5 o2b MnS0 4 3 '* ™ —- [-- o 4 ij 1 ai 1 re) 4000 @000 12000 TIME (sec.) Fig. 2. Speciation of (a) Fe(III) and (b) Mn(II) in the fogwater as a function of time under conditions of Table 3, with liquid water con- tent = 0.1 g m~? and temperature = 10°C. aqueous phase chemistry. Table 4 shows the fraction of each gas scavenged at { = 0, ¢ = 30 min, and t = 180 min. It must be kept in mind that these numbers vary with the liquid water content of the fog [Schwartz, 1983]. Nitric acid is totally dis- solved and dissociated, while NH, dissolves totally at pH < 5. These two components essentially titrate each other in the aqueous phase. Formation of HMSA over the course of a fog event was found not to significantly increase the fraction of CH,0 scavenged; therefore the CH,O partial pressure re- mains approximately constant throughout the simulation. Likewise, hydroxymethanesulfonate formation cannot be a significant sink for gas phase SO,. The solubility of SO, is limited, but because SO, -H,O is a weak acid, ~ 5% dissolves with the formation of the fog. S(IV) oxidation reactions further increase the fraction scavenged over the lifetime of the fog. Three hours after fog formation, over 80% of the initial SO, still remains in the gas phase, indicating that fog has little impact on the gaseous SO, concentration. Hydrogen peroxide TABLE 4. Scavenging of Gases by Fog Droplets Under Conditions of Table 3 Fraction Scavenged From Gas Phase. % Gas t=O t= min - t= 380 min HNO, 100 100 100 NH, 43.9 99.5 100 CH,0 5.0 $3 $3 > 27 118 175 H,0, 32.1 100 100 oO, 0.0 15 15 HNO, 3.8 0.0 0.0 Liquid water content = 0.1] gm~?, temperature = 283 K. 127 16 320 y ne 5 tT te) + ati 1 t § \ { : 0 a \~nocn,80,7 (weaSa) 7 = if ef J & 20 \ ae 5 e— HS05~ wt | Vien OTS) Peso, NLS Le 3 Ts + + t T ' ‘ L = (b) 10 H 1 HSO5” —~ st | | 2 \ a t t 4 By | = ae \ | eR § so2- \ : ~ FeS0;* 7 r Pi , a StS ° ! fy, 2 ri de and - 12000 ° @000 «0 TIME (sec.) Fig. 3. (e) Concentrations of S(IV) species in the fogwater as a function of time under conditions of Tabie 3, with liquid water con- tent = 0.1 g m~? and temperature = 10°C. (5) Same as (a) but with no formaldehyde in the atmosphere. is highly soluble in water. As was discussed above, reduction of H,O, in the droplet is an efficient sink which depletes H,0O, from the system. Ozone, on the other hand, is poorly scavenged, as is HNO,. Both react slowly in the droplet below pH4. The impact of HMSA formation on the droplet chemistry is worth discussing in some detail, as it has been the subject of recent interest [Munger et al, 1983; Boyce end Hoffmann, 1983; Richards et al. 1983}. Figure 3a shows the con- centration profiles versus time for HSO,~ and HMSA. The formation of HMSA is rapid but still slower than S(IV) oxida- tion. Consequently, the amount of HMSA produced is smaller than that of sulfate. Hydroxymethanesulfonate is the most im- portant S(IV) species in solution; if formaldehyde was not Present, the S(IV) concentration in the droplet would be much lower (Figure 35). It has been suggested that HMSA may inhibit S(IV) oxida- tion in atmospheric droplets by limiting the availability of Feactive S(IV) species. This does not appear to be the case, since oxidation proceeds faster than adduct formation; fur- thermore, since the partial pressure of SO, remains approxi- mately constant throughout the fog, the limiting con- Centrations of reactive S(IV) species are independent of the HMSA formation rate. However, the direct effect of HMSA On the fogwater acidity is of interest. Hydroxy- Methanesulfonate is the conjugate base of hydroxymethane- sulfonic acid, which is a strong acid [Roberts et al., 1971]. Hydroxymethanesulfonate is a weak acid (see Table 1) Be- Jacos anp HorrMann: MopeL oF Focwanen CxmaasTry tween pH 2 and 10 and in a system open to SO,, HMSA is stable and complexation of HSO,~ leads to further dissolu- tion of SO,, which then dissociates and releases free acidity in the droplet. Because of the smal] amounts of HMSA produced with respect to the droplet [H*] concentration, this contri- bution is small. In the first simulation, S(VI) formation buffered the droplet acidity at about pH 3, whereas in some extremely polluted atmospheres, higher acidity can be imparted to the droplet at t=0 by simple dissolution of the activated nuclei. Cass [1975] reports some cases for Los Angeles, mostly under high humidity conditions, where the aeroso! sulfate concentration was as high as 75 ug m~>. A second simulation was run (Figures 4 and 5) under the same conditions as the first but with 75 ug m-* SO,?~ in the air mass as NH,HSO, nuclei instead of 10 ug m~*> SO,?~ as (NH,),S0,. The pH of 2.4 at t = 0 (Figure 4) indicates the impact of the acid nuclei. At this low pH there is very little S(IV) in the _ droplet (Figure 5) and the only S(IV) oxidation reaction to proceed at a significant rate is that with H,O,. Therefore less sulfate and hydrogen ion are produced in situ. In the case of acidic condensation nuclei the pH of the fog may be con- trolied strictly by the nuclei composition instead of by S(VI) aqueous phase production. It should be mentioned that fog- water pH in the range of 2 to 2.5 has been observed in the Los Angeles basin following heavy ‘smog’ days. The lowest pH recorded was 1.69 in a dissipating fog along the coastline; in that case the dilution was probably less than the L = 0.1 g m~> considered in the simulation. (a) eat SO g qa 4+ 4 £ ; zr 7 tee of-——+ t + + + (b) 2.32 i. z 4 2.30- 4 2.28 4. 4. i ot ° a000 e000 12000 TIME (20c.) Fig. 4. (a) Profile versus time of total sulfate and of the individual contributions to the total sulfate of sulfate aerosol and different S(IV) oxidants. (b) Profile of pH versus time. The fog formed by con- densation on highly acidic nuclei (75 ug m~> NH,HSO,), other con- ditions as in Table 3, with liquid water content = 0.1 g m~° and temperature = 10°C. 128 Jacos AND HOFFMANN: Move. oF FoGWATER CHEMISTRY o6 T T T T T a4 y OCH SO (HesA) CONCENTRATION [nit 4 — — Cain til dante 4 oA ra d. a) — 10 ~ D : 000 TIME ( sec.) Fig. §. Concentration of S(IV) species in the fogwater as a func- tion of time. The fog formed by condensation on highly acidic nuciei (1S ng m™* NH,HSO,), other conditions as in Table 3, with liquid water content = 0.1 g m~° and temperature = §0°C. Fog frequently forms at temperatures lower than 10°C. Radiation fogs, in particular, often form near freezing. To check the chemical sensitivity of the system to temperature, a simulation was run at 1°C with L = 0.1 g m~? and conditions given in Table 3. Reaction rates are slower at 1°C, but gas solubility increases. Solution equilibria are shifted in either direction depending on the sign of AH’. The combination of these effects has an interesting impact on the sulfate pro- duction rates and the pH profile (Figure 6). The rate of S(IV) oxidation by H,O, is relatively unaffected because of its low activation energy E,, but metal-catalyzed oxidation by O,, trol $650) (a) nN T i Gheteas CONCENTRATION (mAf) t ' ' ' rn oes we) ’ Meee gO eae all 1 aOR ge NOT) fs] T 7 T "Tt (b) TIME (sec.) Fig. 6. (@) Profile versus time of total sulfate and of the individual contributions to the total sulfate of sulfate aerosol and different S(IV) Oxidants. (b) Profile of pH versus time. The fog formed under the conditions of Table 3, with liquid water content = 0.1 g m~> and temperature = I°C. 6617 TIME ( sec.) Fig. 7. Speciation of Fe(I]J) in the fogwater as a function of time under conditions of Table 3, with liquid water content = 0.1 g m~?> and temperature = 1°C. which has a high E,, is much slower at 1°C than at 10°C. Oxidation by ozone becomes appreciable at the beginning of the fog because of its low E, and the increased ozone solu- bility. However, between pH 4.5 and pH 4, H,O, is totally depleted and oxidation by ozone becomes insignificant; the SCV) autoxidation reaction is also very slow. The sulfate con- centration and pH profiles are almost at a plateau in this pH range. Then, as the pH drops below 4, Fe(IIIKOH), starts dissolving and the rate of oxidation by O, increases, such that sulfate and acidity are again produced at an appreciable rate. After 3 hours of fog the values for pH and sulfate con- centration are close to what they were at 10°C. The speciation - profile of Fe(III) (Figure 7) is similar to that at f0°C, but dissolution of Fe(OH), is retarded because of the slow drop in pH from 4.5 to 4. The speciation of Mn(II) is essentially the same as at 10°C. The scavenging efficiency of gases (Table 5) reflects their increased solubility; however, because the aqueous phase reac- tions are slower, little increase is seen between t= 0 and t = 30 min. After 3 hours the scavenging efficiency for SO, is similar to that at 10°C. An interesting feature of the chemistry at 1°C is the high concentration of HMSA (Figure 8), which is due to the in- creased solubility of SO, and CH,O. The reaction proceeds via a nucleophilic substitution of methylene glycol (S. D. Boyce and M. R. Hoffmann, unpublished manuscript, 1983) with an activation energy near 12 kcal mole~). Consequently, the reaction still proceeds rapidly at 1°C. TABLE §. Scavenging of Gases by Fog Droplets Under Conditions of Tabie 3 Fraction Scavenged From Gas Phase, % Gas t=0 t= WO min t= 180 min HNO, 100 100 100 NH, 76.0 97.0 100 CH,O 98 30.3 10.5 ; 10.7 13.0 18.2 H,0, 58.0 100 . 100 3 0.0 42 42 HNO, 10.2 3.4 0.0 Liquid water content = 0.1 g m~?;temperature = 274 K. 129 18 @0o wore - a H ‘ = 60 a — a i - 1 i é ri 7 = Py -~ T HSO,~ 20k ws, = ; Soe Fesoy iv — > ol =~ as = Pit ° @000 e000 000 TIME (sec.) Fig. 8. Concentrations of S(IV) species in the fogwater as a func- tion of time under conditions of Table 3, with quid water content = 0.1 gm~*> and temperature = 1°C. Simulation of a Fog Event The simulation with the present mode] of an actual fog event necessitates the knowledge of input variables such as the condensation nuclei composition, the evolution of liquid water content with time, and the gas phase concentrations. Unfortu- nately, our field investigations [Waldman et al, 1982; Munger et al, 1983] so far have been limited to the determination of fogwater composition. Field data for the above parameters is incomplete; therefore a direct comparison of calculated and observed values is not possible at this time. In addition the current mode! neglects gas phase chemistry, transport mecha- nisms, and aerosol scavenging by droplets. Fogwater composition and its variation with time for fog events at Lennox, a site in an industrial area near the Los Angeles coastline, have been documented thoroughly [Wald- man et al. 1982, Munger et al., 1983}. In this section, a fog occurring under conditions typical of this highly impacted site has been simulated. A plausible scenario for the evolution of liquid water con- tent in fogs can be obtained from existing data [Jiusto and Lala, 1983]. The liquid water content often rises linearly fol- lowing fog formation (neglecting small time scale oscillations), reaches a stable value after about an hour, and decreases lin- early when the fog evaporates. Liquid water contents range from 0.01 g m7? in very light fogs to 0.5 g m~°> in dense fogs. The profile chosen here is shown in Figure 9. Concentrations of condensation nuclei and atmospheric gases prior to fog formation are given in Table 3. From the discussion of the previous section, it appears that some gas- phase emissions should be included. Hydrogen peroxide is not Produced in the gas phase at night and thus actually disap- pears when the fog forms. Ammonia and HNO,, which are scavenged efficiently by the droplets, are, on the other hand, continuously emitted into the atmosphere from a variety of sources, Because fog droplets are a sink for these two gases, fresh inputs into the parcel of air must be considered. At Lennox, plausible values associated with an inversion height of 50 m are 0.01 ppb min™* for HNO, [McRae and Russell, 1983} and 0.01 ppb min™' for NH, [Russell et al, 1983]. Other gases included in the simulation are not depleted by fog, and a reasonable assumption is that their concentrations femain constant throughout the event. A constant temper- Sture of 10°C has been chosen. Temperature changes during the course of fog events [Jiuszo and Lala, 1983] are minimal. Jacos AND HOFFMANN: Mone: of FoGwaTer CHEMISTRY The concentrations predicted for the major ions (Figure 10) ate in the range of those reported by Waldman et al. [1982] and Munger et al. [1983], but no precise comparison should be made because of the reasons stated at the beginning of this section. The concave profiles observed (except for [H*}) con- firm the important role of dilution and evaporation that was suggested initially by Waldman et al. Aqueous phase oxida- tion of N(III) to N(V) is negligible and the only nitrate source is the slow gas phase production of HNO; followed by dissol- ution and dissociation. As a result, the nitrate concentration is controlled primarily by droplet growth. Similar behavior is predicted for ammonium ion, although in the initial stage of the fog the pH drop from a high value increases the NH, * concentration. From equilibria described by (14) and (15), it is seen that scavenging of gaseous ammonia is highly pH- dependent over the range of 5 to 8, which is typical of fog forming in rural environments (D. J. Jacob, unpublished data, 1983) or influenced by alkaline atmospheric components [Munger et al. 1983]. In such fogs, the NH,* levels are ex- pected to be controlled by acidity as well as dilution. Oxidation of SIV) contributes substantially to the sulfate level and the acidity in the early stages of the fog, and in the fully developed fog, 50% of the total sulfate present has been produced in the aqueous phase. During the first few minutes of the fog, S(IV} oxidation is in fact rapid enough to com- pensate for dilution; after H,O, is depleted and the pH has dropped, the oxidation rate slows down and the role of dilu- tion becomes predominant. Even with dilution the pH of the droplets does not rise because metal-catalyzed S(TV) oxidation by O, produces significant acidity in the droplet; instead, pH stabilizes at about 3.5. As the fog evaporates, concentration leads to a further pH decrease. The speciation of Fe(III} (Figure 11) correlates with pH in the manner discussed in the previous section. FeSO,* becomes an important species as the fog evaporates because the sulfate concentration is so high. Speciation of Mn({II) (not shown) is similar to that in Figure 2. An important question is, ‘Is the sulfate formation predicted theoretically actually seen in the field” The data of Waldman et al. [1982] and Munger et al. [1983] do not show obvious evidence for this. However, the bulk of the aqueous phase sulfate production is predicted to occur in the first hour of the fog, so that it could not be detected given the time resolution of the field study. A way to obtain experimental confirmation of this process would be to compare the amount of sulfate present in the atmosphere just before fog formation to that right after fog formation. : The HMSA concentration profile (Figure 12) shows the dominant effect of droplet growth and evaporation, except in e T T T T u E = 4 E z . 8 4 g ° i a 4 L l t+] wooo TIME (s0c.) Fig. 9. Liquid water content profile versus time chosen for the simu- lation of a fog event in the Los Angeles basin. 130 Jacos AND HofrMann: Mone. of FoGwaTer CHEMISTRY 619 5 CONCENTRATION (mM) (a) CONCENTRATION (mM) o... 4000 TIME (sec.) Fig. 10. Profiles versus time of the concentrations of the major ions in the fogwater under conditions of Table 3, with bquid water content as given in Figure 9 and temperature = 10°C. (a) Nitrate and ammonium ions. (5) Total sulfate and individual contributions of the different sulfate production mechanisms. (c) pH. the first few minutes of the fog. Hydroxymethanesulfonate is the major S{IV) species, and its formation may explain the high S(IV) levels found in fogwater [Munger et al., 1983]. It should be noted that in addition to formaldehyde, sulfite is known to readily form sulfonates with other aldehydes, some of which have been found in fog at concentrations comparable to formaldehyde [Grosjean and Wright, 1983]. These reactions would further explain the high S(IV) concentrations observed. 12000 CONCLUSION The chemistry of fogs forming in an urban environment has been investigated using a hybrid kinetic and equilibrium model. The most important conclusions are as follows. 1. Aqueous phase oxidation of S(IV) is an important source of sulfate in the droplet. The principal oxidants are H,0, and O, (catalyzed by Fe(III) and Mn(II)), although ozone can also be an important oxidant above pH 5. Oxida- 131 620 1 om T T T t Felon : fioer » fn 4 E / war Paton? Z / . . 5 oer f \ s OY a) \ / 4 BT / neti~w) / : / F ; Fetou)®* Hw Boy i Fy Ae ES -ne OO Cate pres ° 4000 8000 12000 TIME (sec.) Fig. 11. Speciation of Fe(III) in the fogwater as a function of time under conditions of Table 3, with liquid water content as given in Figure 9 and temperature = 10°C. tion by H,O, is very fast but limited by its availability in the atmosphere. Most of the sulfate production occurs within the first hour following fog formation. 2. When fog condenses on alkaline to slightly acid nuclei, important acidification occurs as a result of S(IV) oxidation. In the first case simulated, the pH dropped two units from its initial value of 5.5 during the first half hour of the fog; it then stabilized around pH 3. When the fog formed on highly acidic condensation nuclei, however, the pH drop due to S{IV) oxi- dation was very email because of the high acidity initially present in the droplet. 3. Oxidation of N(IH) in the droplet does not lead to sig- nificant production of nitrate. Production of nitrate proceeds through gas phase formation of HNO, followed by dissolu- tion and dissociation in the droplet. Nitric acid is scavenged efficiently by the droplets as it is formed in the gas phase. 4. NH,” concentration is dependent both on the liquid water content of the fog and the solubility of NH,. Below pH 5 the droplets are dssentially a total sink for NH, but above pH 5 the gas is partitioned between the two phases in a highly pH-dependent manner. 5. Over 90% of the S{IV) present in the droplet is com- plexed as HMSA, and this may explain the high S(IV) con- centrations observed by Munger et al. [1983]. Formation of HMSA releases free acidity, but its effect on the droplet pH is negligible. 225 : a oo ‘ : : ‘ L 40 H 4 1 = || = 30h 4 3 jon S05 ~ & H HOCH, 507 (HetSA} = 20+! 4 z i wy o f) an ‘1 \ 80," “ LS ° tee des eet c 9000 12600 TIME (e0c.) Fig. 12. Concentrations of S(IV) species in the fogwater as a func- tion of time under conditions of Table 3, with liquid water content as given in Figure 9 and temperature = 10°C. Jacos AND HOFFMANN: Monet oF FoGwaTer CHEMISTRY 6. Fog does not affect the SO, gas phase concentrations greatly; as a result the supply of reactive S(IV) species for S(VI) and HMSA formation is determined primarily by the droplet pH. S(VI) formation proceeds faster than HMSA for- mation at all times and limits HMSA formation by causing .the pH to drop and thus reducing the S(IV) supply. 7. At lower temperatures (1°C versus 10°C), metal- catalyzed oxidation by O, is slower and the importance of ozone as an S(IV) oxidant increases. Overall, sulfate and aci- dity are not produced as fast in the early stages of the fog but catch up later and become comparable in the later stage of the fog. Much more HMSA is produced at the lower temperature. The model has thus revealed some important features of the chemistry and production of acidity in urban fogs. Gas phase and aerosol chemistry, droplet microphysics, and wind fields will have to be incorporated in future versions to give the simulations a predictive capability. These theoretical advances will have to be accompanied by concomitant sets of field measurements. REFERENCES Adamowicz, R. F., A mode! for the reversible washout of sulfur diox- ide, ammonia and carbon dioxide from a polluted atmosphere and the production of sulfates in raindrops, Atmos. Environ., 13, 105- 121, 1979. Appel, B. R. E. L. Kothny, E. M. Hoffer, and J. J. Wesolowski, Sulfate and nitrate data from the California Aerosol Characterization Experiment (ACHEX), Adv. Environ. Sci. Technol., 9, 315-335, 1980. Baboolal, L. B, H. R. Pruppacher, and J. H. Topalian, A sensitivity study of a theoretical model of SO, scavenging by water drops in . air, J, Atmos. Sci. 38, 856-870, 1981. Brimblecombe, P.. and D. J. Spedding, The catalytic oxidation of micromolar aqueous sulphur dioxide, Atmos. Environ., 8, 937-945, 1974. Brodzinsky, R., S. G. Chang, S. S. Markowitz, and T. Novakov, Ki- petics and mechanism for the catalytic oxidation of sulfur dioxide on carbon in aqueous suspensions, J. Phys. Chem, 84, 3354-3358, 1980. Carmichael, G. R., T. Kitada, and L. K. Peters, The effects of in-cloud and below-cloud scavenging on the transport and gas phase _ reactions of SO,, NO,, HC,, H,O,, and O, compounds, in 4th international Conference on "Precipitation Scavenging, Dry Depo- sition, and Resuspension, Santa Monica, December 1982, Elsevier, New York, in press, 1983. Cass, G. R. Dimensions of the Los Angeles SO, /sulfate problem, Environ. Qual. Lab. Memo. 15, Calif. Inst. of Technol. Pasadena, 1975. Chameides, W. L., and D. D. Davis, The free radical chemistry of doud droplets and its impact upon the composition of rain, J. Geophys. Res., 87, 4863-4877, 1982. Damschen, D. E., and L. R. Martin, Aqueous seroso! oxidation of nitrous acid by O,, O,,and H,O,, Atmos. Environ, in press, 1983. Dasgupta, P. K. K. De Cesare, and J. C. Ullrey, Determination of atmospheric sulfur dioxide without tetrachloromerurate(!]) and the mechanism of the Schiff reaction, Anal. Chem., 52, 1912-1922, 1980. Durham, J. L. J. M. Overton, and V. P. Aneja, Influence of gaseous nitric acid on sulfate production and acidity in rain, Atmos. Envi- ron., 15, 1059-1068, 1981. Easter, R. C., and P. V. Hobbs, The formation of sulfate and the enhancement of cloud condensation nuciei in clouds, J. Atmos. Sci., 34, 1586-1594, 1974. Erickson, R. E., L. M. Yates, R. L. Clark, and D. McEwen, The feaction of sulfur dioxide with ozone in water and its possible atmospheric significance, Atmos. Environ., 1, 813-817, 1977. Gartrell, G., Jr. S. L. Heisler, and S. K. Friedlander, Relating particu- late properties to sources: The results of the California Acroso! re ieanal Experiment, Adv. Environ. Sci. Technol. 9, 665- 43, 1980. Grauedel, T. E, L. A. Farrow, and T. A. Weber, Kinetic studies of the photochemistry of the urban atmosphere, Atmos. Environ, 10, 1095-1116, 1976. 132 Jacos AND HorFMaNnn: MODEL OF FOGWATER CHEMISTRY Gracdel, T. E., P. S. Gill, and C. J. Wechsler, Effects of organic surface films on the scavenging of atmospheric gases by raindrops and perosol particies, in 4th International Conference om Precipitation Scavenging. Dry Deposition, and Resuspension, Santa Monica, Decem- ber 1982, Elsevier, New York, in press, 1983. : Graham, R. A, and H. S. Johnston, The photochemistry of NO, and the kinetics of the N,O,—O, system, J. Phys. Chem, 82, 254-268, 1978. Grosjean, D., Formaldehyde and other carbonyls in Los Angeles am- beent air, Environ. Sci. Technol. 16, 254~262, 1982. Grosjean, D., and B. Wright, Carbonyis in urban fog, ice fog, cloud- water and rain water, Atmos. Environ., in press, 1983. Hanst, P. L., N. W. Wong, and J. Bragin, A long-path infra-red study of Los Angeles smog, Atmos. Environ., 16, 969-981, 1982. Hege. D. A. and P. V. Hobbs, Cloudwater chemistry and the pro- duction of sulfates and clouds, Atmos. Environ, 15, 1597-1604, 1981. Hoffmann, M. R., Thermodynamic, kinetic and extrathermodynamic considerations in the development of equilibrium models for squa- tic systems, Environ. Sci. Technoi., 15, 345-353, 1981. Hoffmann, M. R., and D. J. Jacob, Kinetics and mechanisms of the catalytic oxidation of dissolved sulfur dioxide in aqueous solution: An application to nighttime fogwater chemistry, in Acid Precipi- tation, edited by J. G. Calvert, Ann Arbor Science, Ann Arbor, Mich., in press, 1983. Justo. J. E., and G. G. Lala, Radiation fog field programs—Recent studies, ASRC-SUNY Publ. 869, State Univ. of N. Y., Albany, 1983. Landolt-Bornstein, Zahi. te und Funkti Gleichgewicht der Ab- sorption von Gasen in Fhissigheiten von niedrigem Dampfdruck, 6th ed., vol. 4, part 4, sect. C, Springer Veriag, Heidelberg, Federal Republic of Germany, 1976. Ledbury, W., and E. W. Biair, The partial formaldebyde vapour pres- suse of aqueous solutions of formaldehyde, 2, J. Chem. Soc., 127, 2832-2839, 1925. Liljestrand, H. M., and J. J. Morgan, Spatial variations of acid pre- Cipitation in Southern California, Environ. Sci. Technol., 15, 333- 339, 1981. Mazhs, H. G., The importance of ozone in the oxidation of sulfur dioxide in gonurban tripospheric clouds, paper presented at the 2nd Symposium on the Composition of the Nonurban Tropo- sphere, Am. Meteorol. Soc., Williamsburg, Va, 1982. Martin, L. R., Kinetic studies of sulfite oxidation in aqueous solution, in Acid Precipitation, edited by J. G. Caivert, Ann Arbor Science, Ann Arbor, Mich., in press, 1983. McRae, G. J., Mathematical modeling of photochemical air pollution, Ph.D. thesis, Calif. Inst. of Technol, Pasadena, 1981. McRae, G. J., and A. G. Russell, Dry deposition of nitrogen contain- ing species, in Acid Deposition: Wet and Dry, vol. 6, edited by B. B. Hicks, Ann Arbor Science, Ann Arbor, in press, 1983. Middleton, P., C. §. Kiang, and V. A. Mohnen, Theoretical estimates 6621 of the relative importance of various urban sulfate aerosol pro- duction mechanismns, Atmos. Environ., 14, 463-472, 1980. Morel, F. and J. 3. Morgan, A numerical method for computing equilibria in aqueous chemical systems, Environ. Sci. Technol, 6, 58-67, 1972. Munger, J. W., D. 3. Jacob, 3. M. Waldman, and M. R. Hoffmann, Fogwater chemistry in an urban atmosphere, J. Geophys. Res. 88, $109- 5121, 1983. Oblath, S. B., S. S. Markowitz, T. Novakov and S.C. Chang, Kinetics of the formation of hydrozylamine disulfonate by reaction of nitrite with sulfites, J. Phys. Chem, 85, 1017-1021, 1981. Perry, J. M., Chemical Engineer's Handbook, 4th ed.. McGraw-Hill, New York, 1963. Richards, L. W., J. A. Anderson, D. L. Blumenthal, J. A. McDonald, G. L. Kok, and A. L. Lazrus, Hydrogen peroxide and sulfur (1V) in Los Angeles cloud water, Atmos. Environ., 17, 911-914, 1983. Roberts, J. D., R. Stewart, and M. C. Caserio, Organic Chemistry, W. A. Benjamin, Menlo Park, Calif. 1971. Russell, A. G., G. J. McRae, and G. R. Cass, Mathematical modeling of the formation and transport of ammonium nitrate aerosol, Atmos. Environ, in press, 1983. Schwartz, S. E., Gas-aqueous reactions of sulfur and nitrogen oxides in liquid-water clouds, in Acid Precipitation, edited by J. G. Calvert, Ann Arbor Science, Ann Arbor, Mich., in press, 1983. Schwartz, S. E., and J. £. Freiberg, Mass-transport limitation to the rate of reaction of gases in liquid droplets: Application to oxidation of SO, in aqueous solutions, Atmos. Environ., 15, 1129-1144, 1981. Schwartz, S. E, and W. H. White, Solubility equilibria of the nitrogen oxides and oxyacids in dilute aqueous solution, Adv. Environ. Sci. Eng., 4, 1-45, 1981. Sillén, G. L. and A. E. Martell, Stability constants of metal-ion com- plexes, Spec. Publ. 17, Chem. Soc., London, 1964. Serensen, P. E, and V. S. Andersen, The formaldehyde-hydrogen sulphite system in alkaline aqueous solution: Kinetics, mechanisms, and equilibria, Acta Chem. Scand., 24, 1301-1306, 1970. Stumm, W., and J. J. Morgan, Aquatic Chemistry, 2nd ed, Wiley- Interscience, New York, 1981. Waldman, J. M, J. W. Munger, D. J. Jacob, R. C. Flagan, J. J. Morgan, and M. R. Hoffmann, Chemica! composition of acid fog, Science, 218, 677-680, 1982. Westall, J. C.J. L. Zachary, and F. M. Morel, MINEQL, a computer program for the calculation of chemical equilibrium composition of aqueous solutions, Tech. Note 18, Dep. of Civ. Eng, Mass. Inst. of Technol, Cambridge, Mass., 1976. {Received October 4, 1982; revised April 14, 1983; accepted April 14, 1983.) 133 CHAPTER VI A FIELD INVESTIGATION OF PHYSICAL AND CHEMICAL MECHANISMS AFFECTING POLLUTANT CONCENTRATIONS IN FOG DROPLETS by Daniel J, Jacob, Jed M. Waldman, J. William Munger, and Michael R. Hoffmann Tellus 36B, 272-285 Tellus (1984), 36B, 272-285 134 A field investigation of physical and chemical mechanisms affecting pollutant concentrations in fog droplets By DANIEL J. JACOB, JED M. WALDMAN, J. WILLIAM MUNGER and MICHAEL R. HOFFMANN, Environmental Engineering Science, W. M. Keck Laboratories, California Institute of Technology, Pasadena, CA 91125, USA (Manuscript received November 8, 1983; in final form February 15, 1984) ABSTRACT High ionic loadings were found in fogwater collected at Bakersfield. California during an extended stagnation episode. The major ions were NH7. NO;, and SO2-, with concentrations usually in the millimolar range. Droplet growth played an important rdle in determining fogwater concentrations. The amount of solute decreased substantially over the course of each fog event: this was attributed, at least in part, to deposition of fog droplets on surfaces. The occurrence of dense fogs thus seemed to limit particle build-up during stagnation episodes. The sulfate fraction in the aeroso! increased appreciably over several days of stagnation. but no statistical evidence for in situ S(1V) aqueous-phase oxidation was found. The high ammonia concentrations present were sufficient to neutralize a large fraction of the ambient acidity. As a result. fogwater pH values rarely attained the extremely low values found in other polluted environments. 1. Introduction Concern about acidic precipitation has stimulated research efforts to understand the chemistry of non-precipitating clouds (Petrenchuk and Drozdova, 1966; Lazrus et al., 1970; Hegg and Hobbs, 1982; Hegg. 1983). Ionic concen- trations higher than those found in rain have been observed in clouds and fogs (Munger et al., 1983). In fog, the condensation of water vapor on pre-existent particles in the boundary layer shifts the aerosol size distribution towards larger sizes and produces a dilute aqueous aerosol. These physical and chemical changes in the aeroso! may significantly affect its characteristics with regard to air pollution potential and deposition. Even though fogs have been linked in the past to health-threatening pollution. episodes (Commins and Waller. 1967), few field studies of fogwater chemistry have been made. Houghton (1955) sampled fog at several sites in New England and observed high concentrations of components that he attributed to the dissolution of activated nuclei. He found relatively high acidities and suggested that they were due to the scavenging of free sulfuric acid nuclei. Mrose (1966) and Okita (1968) also found elevated ionic concentrations and acidities in East Germany and Japan, respectively. Mader et al. (1949) reported sulfuric acid aeroso] con- centrations as high as 150 wp m-? during fog events in Los Angeles. These values were comparable to those observed during some London fogs (Goodeve, 1936). Waldman et al. (1982) and Munger et al. (1983) found extremely high ionic concentrations in Los Angeles area fogwater: solute concentrations were in the millimolar range for the major components. and pH values were usually in the range of 2 to 4. Fog droplets appear to be efficient scavengers of boundary layer pollutants. and could provide a favorable environment for aqueous-phase reactions leading to the production of strong acids (Jacob and Hoffmann, 1983). The oxidation of absorbed SQV) to form S(VI). with the concomitant produc- tion of acidity, is suspected as being an important Teaction in the aqueous phase. Cass (1979) has observed that the worst sulfate pollution episodes in Los Angeles occur during periods of high Tellus 36B (1984). 4 POLLUTANT CONCENTRATIONS IN FOG DROPLETS humidities and coastal fog, and Hegg and Hobbs (1982) have reported SO, conversion rates of up to 1900% h-' in clouds. Laboratory studies (Martin, 1984) indicate that the oxidation of S(IV) can proceed rapidly at atmospheric concentrations in the presence of strong oxidants. This paper presents results of a field investigation of fogwater composition in the southern San Joaquin Valley of California. The San Joaquin Valley is a site of both agricultural activity and oil recovery operations, which lead to high at- mospheric loadings of particulate matter and trace gases. Fogs are common in the Valley during the winter months and can form every night under near-stagnant conditions caused by mesoscale subsidence. Because of the unusual stability in the local weather pattern and air masses, as well as the identification of emission sources (California Air Resources Board (CARB), 1982), it was possible to study the physical and chemical parameters influen- cing the fogwater composition, and to evaluate the cumulative effect of repeated fog occurrences on the chemical Joading of an air mass. 2. Site description and measurement techniques 2.1. Site description All measurements were made at Bakersfield (Kern county) from 30 December 1982 to 15 January 1983. Kern county is located at the southern tip of the Central Valley of California, which extends north about 700 km. The Valley is approximately 80 km wide over its entire length and is encompassed by mountain ranges (Fig. 1). Apart from a narrow gap at the Sacramento river delta, egress from the Valley is. possible only through mountain passes. The lowest passes to the east and south are above 1000 m elevation. The Valley is subdivided into the San Joaquin and Sacramento Valleys, south and north of the Sacramento gap respectively. In the southern part of the San Joaquin Valley the lowest pass to the West is at 600 m elevation. The wintertime meteorology of the Central Valley is dominated by the Basin High, which Creates a very strong subsidence inversion (Holets and Swanson, 1981). The base of the inversion is typically 200m to 500m above ground level (AGL), although it occasionally comes down to the Telus 36B (1984), 4 135. 273 Fig. J. The Central Valley of California. ground. This stable synoptic pattern is interrupted periodically by a frontal passage which often leads to precipitation on the Valley floor. As the subsidence resumes, the moist air is trapped inside the Valley. Loss of heat by radiation eventually leads to the formation of a widespread fog layer. Under these conditions, fog (locally called “Tule” fog) can form nightly and last into the late morning. with a persistent haze lasting through the afternoon. Because the mixing height is usually lower than the natural boundaries of the Valley, the transport of air masses in and out of the Valley during these episodes is very limited. A weak net flow north out of the San Joaquin Valley leads to a residence time of 8 to 12 days for an air mass in the San Joaquin Valley (Reible et al., 1984). The daytime up-slope flow and the night-time drainage flow associated with the heating and cooling of the mountain slopes dominate the wind patterns in the Valley itself. Flow divergence from the Valley floor in the day, and convergence at night, has been observed by Smith et al. (1981). In the winter, because of the reduced insolation, this flow is probably too weak to ventilate the Valley by transport over the mountain ridges (Reible et al., 1983). As a result. between frontal passages and the ensuing at- 374 mospheric cleansing, the air in the Valley is near-stagnant. Major oil fields are located in Kern county, and large amounts of SO, and NO, are released in the steam-injection oil recovery operations (CARB, 1982). In addition, most of the Valley floor is occupied by farming and ranching activities, which are important sources of primary particulate matter (CARB, 1982) and ammonia. At the high humidities prevalent in the winter months, primary and secondary H,SO, and HNO, react with NH, and H,O to form ammonium sulfate and nitrate droplets (Stelson, 1982). Because of the production of this secondary aerosol, Kern county experiences severe sulfate pollution episodes in the winter months (Duckworth and Crowe, 1979). 2.2. Measurement techniques Fog and aerosol samples were collected from the roof. of a three-story building in downtown Bakersfield, above a CARB air monitoring station. The surroundings are residential and commercial, with no major point pollutant sources. There are no nearby structures taller than the site. An east—west highway, with usually moderate traffic, is located 500 m south of the site. Fogwater was sampled with a rotating arm collector, which has been characterized in detail elsewhere (Jacob et al., 1984). This instrument samples air at a rate of 5 m? min~', performs well in preserving the chemical integrity of the sample at all stages of collection, and has a lower size cut of 20 wm diameter (determined by experimental calibration). Jacob et al. (1984) have estimated that on the average, 60% of the total liquid water is collected with this device. Fogwater samples were collected over periods ranging from 30 min to 3h, and were analyzed for major ions, trace metals, S(IV), formaldehyde, and total organic carbon. Sample handling and analytical procedures were identical to those described previously (Munger et al., 1983). Aerosol between fog events was collected on a set of two Gelman Zefiuor Tefion filters. One filter, which was surmounted by a cap to prevent preferential sampling of large particles by sedimen- tation, was open-faced and collected total particul- ate matter, the other was set downstream of a cyclone separator which removed particles larger than 2 wm. After 11 January, only the open-faced filter was used. Filter samples were taken over time 136 | D. 3. JACOB ET AL. intervals ranging from 2.5 to 8 h. The filters were extracted in 10 ml of water on a reciprocating shaker for 60 min, and analyzed for major ions using the same analytical methods as for the fogwater. Gaseous ammonia was collected by absorption on an oxalic acid impregnated Gelman glass fiber filter set downstream of the open-faced Tefion filter. Ammonium ion was determined by the phenol- hypochiorite method on a water extract using a modification described by Russel! (1983). Liquid water content was measured by drawing air at a rate of 1 m? min~' through open-faced paper filters. The filters were exposed to the fog prior to use to equilibrate them with the ambient humidity. The liquid water content was determined by weight. This method has been found (Calspan Corp., Buffalo, NY, private communication) to yield results comparable to those obtained with more sophisticated optical methods (Chylek, 1978). It must be stressed that no fully reliable method for measuring absolute liquid water content exists to date, and that variations in results of up to 50% commonly occur from one method to another; therefore measurements must be interpreted with caution. The CARB air monitoring station located at the site provided hourly averages of gas-phase NO, NO,, SO,, O;, and CO concentrations. Hourly mixing height measurements (from 30m up to 1000 m) were obtained at Bakersfield by acoustic sounding (Western Oil and Gas Association, Los Angeles, private communication). Hourly weather observations were recorded 8 km north of our site by the Bakersfield National Weather Service office (National Oceanic and Atmospheric Adminis- tration, 1983). 2.3. Weather pattern On 22-23 December, a frontal passage over the Valley deposited 7.6 mm of rain on Bakersfield. A strong subsidence associated with the Basin High followed, and a stratus deck started forming on 26 December. Temperatures dropped and were 3 to 7°C lower than normal during the period 30 December—15 January. Daily highs ranged from 4 to 8°C and daily lows from —2 to 3°C. Fog was Teported at the Bakersfield site on 29 and 30 December, but was patchy in the surrounding area. In the text, the date of the fog will be given as that of the morning on which it occurred, even though it Tellus 36B (1984), 4 POLLUTANT CONCENTRATIONS IN FOG DROPLETS may have already been formed the previous evening. On 31 December and 1 January, dense widespread fogs persisted through the night and well into the morning. The following three days, 2-4 January, thin radiation fogs formed at the site for a brief period in the early morning as the temperature dropped sharply. Dense widespread fog through the night again occurred on 5 January (not sampled) and on 6-8 January. On 9-10 January, a weak front passed through the Valley, resulting in some clear sunshine, high cloudiness, and drizzle in the morning of 9 January which deposited a trace amount of rain. On the morning of 10 January, a drop in temperature down to —2°C at the site caused the formation of a thin, shallow fog which lasted for 3h. After 1600 on 10 January, the subsidence inversion again limited the mixing heights to less than 500m AGL and dense, all-night fogs formed every night from 11 January to 15 January. Sampling was discontinued after 15 January. All fogs were sampled from beginning to end, except the 31 December and 11 January fogs, for which sampling did not start until a few hours after fog formation. 137 275 3. Results and discussion 3.1. Fogwater concentrations Table 1 gives the range and median value of fogwater concentrations for the entire data set (n = 108). Close ionic balances for most fogwater samples (+ 10% in over 80% of the cases) indicate that no important ions were missed in analysis. Ammonium, nitrate, and sulfate ions, by far the most important components present, had con- centrations usually in the millimolar range. These high concentrations indicate the dominant impact of agriculture (NH7) and fuel combustion (NO;, $O2-) on the fogwater composition. Occasionally, Ht, Ca?*, and Cl~ contributed significantly to the total equivalent loading of the droplets. Ratios of concentrations of Nat to other ions indicated that sea salt was not a major source. Iron and lead were the most prominent trace metals. Fig. 2 shows the evolution of fogwater con- centrations with time during the 7 January fog. Winds were light and variable throughout the night, and temperatures ranged from 1 to 3°C. As droplets grew at the beginning of the fog, they Table 1. Fogwater concentrations at Bakersfield, California* Component H* Nat K+ NH; Ca?* Mg?* ci- Range of concentrations (weq1-!}) 0.1-2750 14-325 16-368 490-13,300 7-3500 1.1-430 1-980 Median. concentration (ueqi-) 60 19.5 9.3 1440 47 6.3 47 Component NO; SO2- S(IV) CH,O Toct Range of concentrations (eq 1-') 200-6800 194-9400 45-3000t 53-710¢ 710-23,000+ Median ; concentration (ueqi-') 850 1160 §15¢ 165t 4000+ Component Fe Mn Pb Cu Ni Vv Range of concentrations (vg !~') 81-10,700 3-525 48-3340 6-717 22-1232 7-850 Median concentration (ug!-') 400 14 330 36 61 55 (a) Hydroxymethanesufonate ion coelutes with nitrate in the ion chromatographic column, so that nitrate concentrations could be overestimated. * 108 samples over the period 30 December 1982-15 January 1983. t Total organic carbon. ¢ For S(IV), CH,O, and total organic carbon, concentrations are in ¢ moles I~!. Tellus 36B (1984), 4 276 138 D. J. JACOB ET AL. (035) 3.62 = 20 - ~ if E i) (.055) _ (070) J z 365 (7 5.15 E 1.085) (.008) © 10- 4.34 (013) 38 + 5 " (036) 3.63 5} Heed (095) + et023) i ‘ (.062) is SB st mes 428 370 \"400 a sa 4 SS fr) 1 20 .22 00 O2 O04 O68 o8 10 6 JAN. 7 JAN. TIME (Hour PST) Fig. 2. Evolution of fogwater concentrations over the course of the 7 January 1983 fog event at Bakersfield. Fogwater pH and average liquid water contents (g m7’, in parentheses) are indicated on top of each data bar. became more dilute, and conversely as the fog dissipated, evaporation concentrated the droplets. The resulting concave profile for fogwater con- centrations was evident in all the events sampled over the course of this investigation, and has been found to be a characteristic of fogs in other locations as well (Waldman et al., 1982). Droplet gtowth is therefore a dominant process in control- ling fogwater concentrations. Accordingly, acidities were usually the highest at the beginning and end of fog events. The ratio of concentrations of one component to another was not subject to large variations from sample to sample within one event, which implies that the air masses were generally homogeneous. A trend of decreasing concen- trations, superimposed on the concave profile, was noted. We will see that this can be attributed, at feast in part, to deposition of fog droplets on surfaces. 3.2, Atmospheric particulate loadings in relation to fog occurrences The major components of the daytime aerosol were NO;, SO?-, and NHj. No significant dif- ferences were observed between the total serosol and fine aerosol filters for NH{, NO;, SO], and Ci- concentrations. On the other hand. concen- trations of Na*, Ca?+, and Mg?+ were higher in the total aerosol than in the fine aerosol: coarse soil dust is probably a major source of these com- ponents. Fig. 3 summarizes the evolution of atmospheric loadings over the 17-day sampling period. Mixing height data and concentrations of trace gases are given as hourly averages. Ozone concentrations were at or below the detection limit of 10 ppb in 96% of the hourly measurements. Time-averaged fogwater loadings/m? of air (cal- culated from liquid water content measurements) are given for each fog event, along with the aerosol loadings and gaseous ammonia concentrations between fog events. Prior to 6 January. the liquid water content was not directly measured: we estimated it by measuring the amount of water collected and assuming a 60% collection efficiency. In our calculations of the fogwater loadings. we assume that the collected samples are represen- tative of the ambient fogwater in spite of the high lower size-cut of the sampler. The individual contributions of droplets of different sizes to the overall fogwater loading have not yet been rigorously investigated: they are difficult to predict from droplet growth theory (Pruppacher and Klett. Tellus 36B (1984), 4 139 December Fig. 3. Fogwater (in bold) and aeroso! loadings of NH;. POLLUTANT CONCENTRATIONS IN FOG DROPLETS 277 600 ——__--_——+ _ -—r NO, pob i 200+ . ’ to 4 wo! 2 ; ppb "Ted ML bah | og ee Ue leah as re - : Mixing Heights m s00+ vt af . 1 Np f yer (rod nals oL_ te ; a a an oT sr TT TT. +. |. + tT T > | Filter ond fogwater loadings 2.0 Ww Ss £ g a —2 E ‘l 0 NO;. SO2- and H* (fogwater) measured at Bakersfield during the period 30 December 1982-15 January 1983. When measured. gaseous ammonia concentrations (p.p.b.: 1 p.p.b. = 0.044 wmoles m-? at 1°C) are indicated on top of the aerosol data bars. Mixing heights are given in (m) above ground level. 1978) because supersaturations in fogs fluctuate tapidly in a poorly understood manner (Roach. 1976; Gerber, 1981). The large droplets in the fog are likely to have condensed on large nuclei. and therefore are not necessarily more dilute than the smaller droplets. A recent field intercomparison study of fogwater sampling devices (Hering and Blumenthal. 1983) seems to indicate that variations in the composition of the fogwater collected by instruments with different droplet size-cuts are insignificant. The mixing height data show that two distinct air Tellus 36B (1984), 4 masses were involved during the 17-day period. The base of the inversion layer usually oscillated between 100m and 500m AGL. However. on 9 January, the inversion layer rose to above 1000 m and possibly vanished (no measurements were available above 1000 m). This allowed mixing with air from outside the Valley. and. with the. added contribution of the 9 January morning drizzle. effectively renewed the air mass in the Valley. After the afternoon of 10 January. stagnation conditions resumed and no mixing heights above 500 m were recorded unt] after 15 January. 278 Aerosol and trace gas loadings reflected the apparent change in air masses. Since emission sources are located close to the ground, the majority of pollutants were released below the base of the inversion layer. Due to the lack of ven- tilation, accumulation of pollutants proceeded. During the period of 30 December to 9 January, the stagnant air mass was well-aged; high particul- ate and SO, levels were observed. After the effective cleansing of the atmosphere on 9-10 January, aerosol and SO, loadings dropped sharply. On several occasions, aerosol samples were taken a few hours before a fog formation. The fraction of the afternoon aerosol found in the first fogwater sample collected the following night ranged from 0.1 to 0.8 and from 0.2 to 2.0 for sulfate and nitrate, respectively. These numbers varied considerably from fog to fog, and this can be attributed to variations in nucleation scavenging efficiencies and differences in the air masses advected over the site in the afternoon and at night. It is therefore not possible to make quantitative conclusions with regard to aerosol scavenging efficiencies, but it appears that a substantial fraction of the aerosol is incorporated into the fog droplets. Scavenging of trace gases is an additional source 140 D. J. JACOB ET AL. of chemical input to the fogwater. The solubility of several gases in fog droplets has been discussed by Jacob and Hoffmann (1983). Nitric acid is 100% scavenged, and so is ammonia below pH §. From the gas phase data, it appears that the occurrence of fog does not by itself affect partial pressures of SO,; thermodynamic considerations indicate that, in spite of the dissociation of SO,-H,0 in solution and the formation of S(IV) adducts, absorption of SO,(g) by fog droplets is limited to a few % because of the small amount of water present. No correlation appears either between NO, levels and fog occurrences. This agrees with the results of Schwartz and White (1981), who show that the dissolution of NO, at atmospheric concentrations is too slow to be of relevance over the lifetime of the fog. All the extended fog events sampled over the 17-day period showed a substantial decrease in the fogwater loadings over the course of the fog (Fig. 4). Loadings of major ions at the end of the fog ranged from 4% to 36 % (median 17%) of those at the beginning of the fog, depending on the event. However, the steady decreasing pattern was not established until a few hours after fog formation; in the initial stage of the fog, the evolution of fogwater loadings was irregular. Changes in the fogwater loadings over the course of the fog could be 10 T T T T — O08 3) wn £ = A 0.6 i ic : é O4r- - 2 vt) S oak z 0. oO fom N & 18 20 22 00 02 04 6 JAN. WS o6 O8 10 l2 7 TIME (Hour PST) Fig. 4. Evolution of fogwater ionic loadings over the course of the 7 January fog event. Tellus 36B (1984), 4 POLLUTANT CONCENTRATIONS IN FOG DROPLETS attributed to deposition of the coarse fog droplets, aghort-range transport, or displacement of the inversion base. Comparison of daytime aerosol or nighttime fog loadings to the mixing heights over the same time period did not show any strong correlation for mixing heights below 600m. In addition, oscillations of the mixing height over the course of the fogs did not show a consistent trend which could be associated with a decrease in fogwater loadings. Therefore, this factor is prob- ably not critical. Characterization of short-range « ransport at night is made difficult by the complex drainage flows converging to the center of the Valley. Because Bakersfield is located in the convergence zone, local winds oscillate considerably from hour to hour in both direction and speed (National Oceanic and Atmospheric Ad- ministration, 1983). Concentrations of CO, NO,, and SO, during fog events did not show the strong decreases that would be associated with the advection of much cleaner air masses over the site. Therefore, it is unlikely that short-range transport could fully account for the considerable decrease of fogwater loadings with time. Deposition of fog droplets on surfaces, on the other hand, could lead to the observed decreases in fogwater loading. Fog formation results in an important shift of the aeroso! distribution towards larger particle sizes; because the residence time in the atmosphere of particles above 10 ym is of the order of hours, and rapidly decreases with increas- ing size (Muller, 1982), the deposition rate of the gerosol must be enhanced. During the early stage of the fog, the droplets presumably do not grow to a size large enough to be effectively deposited; factors other than deposition influence the fog- water loading, and hence the irregular behaviour of the profile of fogwater loadings with time. As a population of ‘large droplets is later established, deposition becomes more rapid and overcomes other factors. Turbulent diffusion. to the ground, impaction and sedimentation are probably all important in contributing to the removal of fog droplets. The numerical study of Brown and Roach (1976) has shown that important droplet settling Must occur to explain the low liquid water content observed in fogs; Brown (1980) finds that the setiling process does not become important until a few hours after fog formation, which is consistent with our observations. If scavenging of gases and Particles by the droplets is followed by deposition, Tellus 36B (1984), 4 141 279 pollutants wil) effectively be removed from the atmosphere. Particle concentrations in the Valley are deter- mined by sources (fresh input of primary and secondary aerosol), sinks (transport aloft and out of the Valley, deposition), and intra-Valley trans- port. Under stagnant conditions, one would expect a pollutant build-up until a steady state is reached due to an eventual balance of sources and sinks. The characteristic time for deposition of haze particles (0.05-2 ym) is 3-6 days (Muller, 1982), but under fog conditions, particles scavenged by the fog droplets will deposit much more quickly. Dense fogs can therefore limit particle accumulation. The overall higher loadings during the 26 December-—8 January stagnation episode (where dense fog did not form every night) compared to those during the 10 January-15 January episode (where dense fog formed every night) could then be explained by a pollutant build-up due to the predominance of haze but limited fog. This accumulation is obvious for ammonium nitrate and sulfate salts, which are the main constituents of the haze aerosol; on the other hand, concentrations of Ca?* in the aerosol did not increase during the stagnation episodes. Calcium mostly originates from coarse dust particles with a residence time shorter than one day, and will settle out regardless of fog formation. 3.3. Heterogeneous S(1V) oxidation Because of the high SO,(g) concentrations in Bakersfield and the presence of an aquated aerosol. there is a potential for aqueous-phase oxidation of S(IV) to S(VI) by the strong oxidants H,O,. O,. and O, (Jacob and Hoffmann, 1983; Martin. 1984). Evolution of sulfate concentrations in the fogwater between times 7, and 7, can be expressed by: (S(V}), LO.) = US(VD),L@) f diS(v1)] + ee dt L(@jdrk(i—r), a) where L(d is the liquid water content of the fog. and r characterizes changes in [S(VI] (assumed first-order} due to physical mechanisms such as deposition. We assume as a first approximation that L(t) and r control the evolution of the chemically conservative quantity [7], defined as the sum of the equivalent concentrations of all 142 280 components A, in solution minus sulfate and its counter-ion: IT)= ¥ [4 —2SWDI i=l (2) With this assumption, we can write for a homogeneous air mass: and substitute (3) into (): f Seam Liar Lan(scvn, (3) x ~ - v0, a2 (4) A major difficulty in solving (4) for d{S(VI)]/dr and using statistica] analysis to verify or derive a rate law is that the variables on which d[S(VD]/dr depends are not satisfactorily quantified. Reactions of S(IV) with H,O, and O, are expected to be first-order in oxidant concentrations (Martin, 1984), but H,O, and O, concentrations were not measured at Bakersfield (ozone below the detection limit of 10 ppb, as observed, still cannot be neglected as an oxidant in droplets with high pH). Oxygen has been predicted to be an effective oxidant in solution when catalyzed by trace metals (Jacob and Hoffmann, 1983), but the form of the rate law has not been satisfactorily established at the pH and free S(IV) concentration ranges typical of the Bakersfield fogwater (Martin, 1984); further- more, it may be complicated because of aqueous- phase speciation of metals and S(IV} (Hoffmann and Jacob, 1984). Rates of sulfate production in the atmosphere have previously been derived from field data by Cass (1981) and Hegg and Hobbs (1981, 1982). These investigators expressed this rate in terms of a pseudo first-order rate constant k (% h-’) referenced to gas-phase SO,: dIS(VD]y, —k _ , 5 a 100 {SO,(g)] (5) where [S(VI)],,, and {SO,(g)] are concentrations per unit volume of air. Eq. (5) is obviously unjustified from a chemical standpoint, because it ignores the rate dependence on droplet pH, concentrations of oxidants and catalysts, tem- D. J. JACOB ET AL. perature, and photochemical activity. Still, it provides a convenient parameterization of sulfate production and will be adopted here in the absence of a more reliable expression. Substituting (5) into (4): f * * iso.@iar a 100 ; [7], = L(t, ((S(VD],, — — [S(vp},, (6) (71, Assuming that [SO,(g)] varies linearly with time over the interval considered: 200 L(t) IT],, IT], Table 2 shows that the & values calculated from our data set with eq. (7) are lower than those obtained by Hegg and Hobbs (1981, 1982) for clouds over western Washington State. and are of the order of those observed in the Los Angeles aerosol by Cass (1981). However, both ozone concentrations and pH were higher in the clouds sampled. by Hegg and Hobbs (1981, 1982) than in the Bakersfield fogs sampled in this study: in the same way, temperature, solar irradiation, and oxidant concentrations are all lower in Bakersfield (in the wintertime) than in Los Angeles. Therefore, comparison of the data sets in terms of k is difficult. Cass (1981) noted that gas phase oxidation could not be sufficiently rapid to explain the S(IV) conversion rates in Los Angeles. and suggested aqueous-phase oxidation as an alternate mechan- ism. Hegg and Hobbs (1982) pointed to S(VJ) in- cloud production from some basic statistical analysis of their data, a conclusion which has been disputed by Schwartz and Newman (Schwartz et al., 1983). Our data set does not allow us to conclude that & is significantly different from zero. Fig. 5 shows the frequency distribution we ob- served for k; the near symmetrical deviations about k = 0% h-' can be simply attributed to the non-Lagrangian nature of the sampling procedure. The presence of a few outliers on the positive side of the distribution could suggest that S(IV) in-fog oxidation proceeds under certain conditions, but these points are associated with SO, x (scm, - svn) ) Tellus 36B (1984), 4 POLLUTANT CONCENTRATIONS IN FOG DROPLETS Table 2. SO, oxidation rates in the almospnere 143 281 Location k (% h-'y* Reference Western Washington (wave clouds) ‘| 0-300 Hegg and Hobbs (1981) Western Washington (clouds) —600 + 1000-1900 + 1900 Hegg and Hobbs (1982) Los Angeles (aerosol, summer) 6.0t Cass (1981) Los Angeles (aerosol, winter) 2.0t Cass (1981) Bakersfield (fogs) 0.9 + 5.5¢ This study * kis & pseudo first-order rete constant expressing sulfate production as a % of [SO,(g)). + Hourly production rates averaged over 3 years. t Calculated from one fog sample to the next using equation (7). Data for nm = 80 samples. Samples were excluded from the data set when: (i) SO, partial pressure was below the detection limit of 10 ppb; (ii) ionic balances were off by over 30%; or (iii) only one sample was collected during the event (2 January) “16-120 BO 4 o 4 8 k(%n') oo 1) 2 t& 20 Fig. 5. Frequency distribution of the pseudo first-order fate constant for S(VI) in-fog production, k (% Ac"). Data from n = 80 fogwater samples collected at Bakersfield during the period 30 December 1982-15 January 1983. concentrations at, or close to, the detection limit of 10 ppb. The uncertainty on (SO,(g)] can then be as great as 50%, so that these outliers cannot be considered significantly different from the population as a whole. Sulfate production was also analyzed in terms of k/L, which characterizes sulfate production referenced to the aqueous phase, and in terms of a zero-order dependence on [SO,(g)] (oxidation limited by mass transfer). Neither of these analyses showed evidence of S(V1) in-fog production. In spite of this, the sulfate fraction of the aerosol increased over the course of the stagnation episodes (Fig. 6). After the two widespread fogs and Tellus 36B (1984), 4 o oO 1 w SULFATE EQUIVALENT FRACTION o w JAN. DEC. Fig. 6. Evolution of the sulfate equivalent fraction in the fogwater (—) and in the aerosol (@) at Bakersfield during the period 30 December 1982-15 January 1983. Lines have been added to show trends. persistent thick haze of 31 December and ! January, the sulfate equivalent fraction (SEF) of the aerosol (defined as SEF = [S(VI)}/[T]}) rose and stabilized at 0.40 + 0.05. In comparison, the two-month aeroso! averages reported by Heisler and Baskett (1981) for November—December 1978 at Bakersfield show an SEF of 0.22. The highest sulfate concentration we observed, 76 wg m~* on 6 January, is close to the highest 24-h value recorded in the area, 80 wg m-? (Duckworth and Crowe. 1979). On 9-10 January. along with the removal of the stagnant air mass from the Valley. the SEF dropped sharply, down to 0.10 on the afternoon of 10 January. After the inversion trapped the renewed air mass in the Valley, the SEF rose again to values similar to those before 10 January. The SEF values for the fogwater were lower than those 144 282 for the afternoon haze, but behaved qualitatively in the same way. Preferential deposition of components other than sulfate would result in an increase in the sulfate content at the beginning of the stagnation episode. Because nitrate does not show the increase ob- served for sulfate, even though both components are mostly present in the same type of particles, it seems unlikely that preferential deposition alone could result in such 2 rise of the SEF. It therefore appears that S(VI) production reactions proceed in the stagnant air mass, and that the stabilization of the sulfate content then indicates the attainment of a steady-state between sulfate production and the sinks discussed previously. In-fog oxidation at an average rate of 0.9% h-! (Table 2) with P,,, = 20 ppb would produce 16 neq m-? S(VI)/h of fog. This is not sufficient to explain the increases in the SEF of the daytime aerosol, especially since a large fraction of the sulfate that could be produced in fog is expected to be lost by deposition. Gas-phase oxidation of SO, by radicals is expected to be unimportant because of the limited photochemical activity (reflected by the low O, levels). Instead, the higher SEF values in the daytime aerosol than in the fog suggest that sulfate production mostly occurs in the haze droplets forming at high humidities. This could be expected because of the higher catalyst concentrations in haze droplets, and because both temperatures and SO,(g) concen- trations usually peak in the afternoon. Oxidation of S(IV) in haze droplets should proceed along similar pathways as for fog droplets, and has been observed to proceed rapidly under controlled conditions (Matteson et al., 1969; Crump et al., 1983). 3.4. Fogwater acidity Extreme acidity in fogwater, as has been documented in the Los Angeles area, may pose specific environmental problems because of the potential for injury to materials, vegetation (Scher- batskoy and Klein, 1983), and human health (Hoffmann, 1984). Sulfuric acid and nitric acid (which are the main strong acids present) can be emitted into the atmosphere as primary pollutants, or alternatively be produced in situ (Calvert, 1984). Alkaline components in the atmosphere will titrate free acidity. Gaseous ammonia, which is the most important alkaline component in the Valley, reacts D. J. JACOB ET AL. with H,SO, and HNO, at high humidities to form concentrated ammonium salt droplets (Stelson, 1982). Dissolution of SO,(g) in the aquated aerosol leads to S(IV), which can then be oxidized to S(VI) as discussed in the previous section. Alternatively, S(IV) can be stabilized in the lower oxidation state by complexation with carbonyl compounds (Mun- ger et al., 1984). The production of one molar unit of S(VI) releases one to two molar units of H+; formation of S(IV) adducts also produces H+ @acob and Hoffmann, 1983). The resulting input of acidity can be subsequently titrated by absorption of NH,(g) in the droplet. Therefore. the acidity of liquid water in an air mass under high sulfate and nitrate conditions is limited by the availability of NH,(g) (and other alkaline compo- nents) to neutralize the strong acids present. This has previously been observed for rainwater in the north-central United States (Munger, 1982). In Los Angeles, where NH, emissions are relatively low (Russell et al., 1983), extremely high acidities are frequently found in the fogwater. In the Bakersfield fogwater, nitrate and sulfate concen- trations are comparable to those in Los Angeles. but high acidities are much less common (Fig. 7). This is because ammonium concentrations are $2] T t oT + + T ¥ #4 Bobersfreic n- 20- E or cot sO # aor Los Angeies (fogs) Fe ee g > 8 zor E of a inl a; # «0b Los Angeles (strotus) » 30r gm F of Fe} a 2 oo 2 1 v2 3s 4 5 6 7 8 9 pH Fig. 7. pH frequency distribution for n = 108 fogwater samples collected at Bakersfield during the period 30 December 1982-15 January 1983. Results are compared with those obtained in the Los Angeles basin for fogs (nm = 45) and stratus clouds (nm = 156) (this laboratory. unpublished results, 1981-1983). Tellus 36B (1984). 4 145. POLLUTANT CONCENTRATIONS IN FOG DROPLETS 8 F Ly T T Lu M7 ms aor - # sot - > ¥ 3° 7 rd ° 1 i 4 rl n ° “a es oS 2 2 wo 35 wolfe] /{wog]}+[se2" J) om Fig. 8. Frequency distribution of the non-neutralized fraction of the acidity, as expressed by [H*+]/([NO;] + {SO2-]), for n = 108 fogwater samples collected at Bakersfield during the period 30 December 1982-15 January 1983. considerably higher than in Los Angeles. A few samples at Bakersfield also had high concen- trations of Ca?*, indicating an alternative neutral- ization pathway by scavenging of alkaline dust. Fig. 8 shows the fraction of non-neutralized acidic anions in our fog samples. In almost half of the samples, over 95% of the acidity was neutralized. At the same time, very low con- centrations of gaseous ammonia were observed throughout the 17 days (Fig. 3). Most of the fogwater samples had pH values below 5, which suggests an absence of ammonia in the atmosphere from chemical equilibrium considerations. Osten- sibly, the acidities present were sufficient to totally exhaust ammonia from the gas phase, and produc- tion of additional acidity resulted in a pH drop. 4. Conclusion High ionic concentrations in the fogwater were observed at Bakersfield, California over the course of two stagnation episodes. The main contributors to the ionic loading of the fogwater were NHi, NO;, and SO32-; this can be attributed to agricul- tural activity and oil recovery processes. Droplet growth was found to strongly influence the ob- served concentrations. Comparison of ionic load- 283 ings in the fogwater with those in the aerosol of the previous afternoon indicated that a sizable fraction of the aerosol was scavenged by the fog droplets. Fogwater ionic foadings decreased considerably over the course of the fog, and this was attributed to deposition of the large droplets on surfaces. In this manner, it was observed that a large fraction of the pollutants scavenged in the fog could be removed from the atmosphere. Deposition of fog droplets proceeds much more rapidly than deposition of the haze particles present at the high humidities between fog events, so that the occur- rence of fog could slow down or limit the accumulation of particulate matter in a stagnant air mass. Evolution of the sulfate content over the 17-day sampling period indicated an important increase of the sulfate fraction in the aerosol during stagnation episodes, but no statistical evidence for S(IV) oxidation in the fogwater was found. Most of the sulfate appeared to be produced in the daytime haze droplets. Preferential deposition of compo- nents other than sulfate could also partly explain the increase in the sulfate fraction. High concentrations of acidic anions were present in the fogwater, but most of the acidity was neutralized by ammonia. In most cases. the neutralization of acidic components totally depleted NH, from the atmosphere; although the excess of acidity was usually not enough to lead to highly acidic conditions, 8% of the samples had pH values below 3. 5. Acknowledgements This research was funded by the California Air Resources Board (CARB). We thank the staff of the CARB for their help and cooperation relating to our use of their facility at Bakersfield. We also acknowledge gratefully the CARB and the Western Oil and Gas Association for letting us use their data in this paper. Discussions with A. G. Russell were most helpful in the interpretation of some of the data. REFERENCES Brown, R. and Roach, W. T. 1976. The physics of radiation fog (II). A numerical study. Q. J. R. Meteorol. Soc. 102, 335-354. Teilus 36B (1984), 4 Brown, R. 1980. A numerical study of radiation fog with an explicit formulation of the microphysics. Q. J. R. Meteorol. Soc. 106, 781-802. 146 © 284 California Air Resources Board 1982. Emission Inven- tory 1979, Stationary Source Control Division, Emission Inventory Branch, Sacramento, California. Caivert, J. G. (ed.) 1984. Acid precipitation. Butterworth Publications, Boston. Cass, G. R. 1979. On the relationship between sulfate air quality and visibility with examples in Los Angeles. Atmos. Environ. 13, 1069-1084. Cass, G. R. 1981. Sulfate air quality contro! strategy design. Atmos. Environ. 15, 1227-1249. Chylek, K. P. 1978. Extinction and liquid water content of fogs and clouds. J. Atmos. Set. 35, 296-300. Commins, B. T. and Waller, R. E. 1967. Observations from a ten-year study of pollution at a site in the city of London. Atmos. Environ. 1, 49-68. Crump. J. G., Flagan, R. C. and Seinfeld, J. H. 1983. An experimental study of the oxidation of sulfur dioxide in aqueous manganese sulfate aerosols. Atmos. Environ. 27, 1277-1289. Duckworth, S. and Crowe, D. 1979. Sulfur dioxide and sulfate trends, Bakersfield, 1977-1978. Technical Services Division, California Air Resources Board, Sacramento, California. Gerber, H. E. 1981. Microstructure of a radiation fog. J. Atmos. Sci. 38, 454-458. Goodeve, C. F. 1936. Discussion in “Sulphuric acid as a disperse phase in town air”. Trans. Faraday Soc. 32, 1201-1202. Hegg. D. A. and Hobbs, P. V. 1981. Cloud water chemistry and the production of sulfates in clouds. Atmos. Environ. 15, 197-1604. Hegg. D. A. and Hobbs, P. V. 1982. Measurements of sulfate production in natural clouds. Atmos. Environ. 16, 2663-2668. Hegg. D. A. 1983. The sources of sulfate in precipitation (1). Parameterization scheme and physical sensitivities. J. Geophys. Res. 88, 1369-1374. Heisler, S. and Baskett, R. 1981. Particle sampling and analysis in the California San Joaquin Valley. Research report CARB-RR-81-14, December 1981. available from Environmental Research and Technology, Westlake Village. Catifornia. Hering, S. V.. and Biumenthal, D. L. 1983. Field comparison of fog/cloud water collectors: preliminary fesults. Proceedings, Air Pollution Control Association Specialty Conference on the Meteorology of Acid Deposition, Hartford, Connecticut, 17 Oc- tober 1983. Published by the Air Pollution Control Association, Pittsburgh. Hoffmann, M. R. and Jacob, D. J. 1984. Kinetics and mechanisms of the catalytic autoxidation of dissolved sulfur dioxide in aqueous solution: an application to night-time fogwater chemistry. In Acid precipitation et J. . Calvert). Boston: Butterworth Publications, 1-172. Hoffmann, M. R. 1984. Comment on acid fog. Environ. Sci. Technol. 18, 61-64. Holets, S. and Swanson, R. N. 1981. High-inversion fog episodes in centra) California. J. Appl. Meteorol. 20, 890-899. D. 3. JACOB ET AL. Houghton, H. G. 1955. On the chemical composition of fog and cloud water. J. Meteorol. 12, 355-357. Jacob, D. J. and Hoffmann, M. R. 1983. A dynamic model for the production of H*, NO;. and SO2- in urban fog. J. Geophys. Res. 88. 6611-6621. Jacob, D. J., Wang. R.-F. T. and Flagan. R. C. 1984. Fogwater collector design and characterization. En- viron. Sci. Technol. 18, (in press). Lazrus, A. L., Baynton, H. W. and Lodge. J. P. 1970. Trace constituents in oceanic cloud water and their origin. Tellus 22, 106—114. Mader, P. P., Hamming. W. J. and Bellin, A. 1949. Determination of smal! amounts of sulfuric acid in the atmosphere. Anal. Chem. 22, 1181-1183. Martin, L. R. 1984. Kinetic studies of sulfite oxidation in aqueous solution. In Acid precipitation (ed. J. G. Calvert), Boston: Butterworth Publications. 63-100. Matteson, M. J., Stober, W. and Luther, H. 1969. Kinetics of the oxidation of sulfur dioxide by aerosols of manganese sulfate. J. & E. C. Fund. 8. 677-686. Mrose, H. 1966. Measurements of pH. and chemical analyses of rain-, snow-, and fog-water. Tellus 18, 266-270. Muller, J. 1982. Rules in atmospheric behaviour of particulate substances. GAF-conference. Bologna. 14-17 September 1982, Fraunhofer Ges.. Grafschaft/ Schmallenberg, F. R. Germany. Munger, J. W. 1982. Chemistry of atmospheric pre- cipitation in the north central United States: influence of sulfate, nitrate, ammonia. and calcareous soil particulates. Atmos. Environ. 16, 1633-1645. Munger. J. W., Jacob. D. J.. Waldman. J. M. and Hoffmann. M. R. 1983. Fogwater chemistry .> an urban atmosphere. J. Geophys. Res. 88. §109-5123. Munger, J. W., Jacob, D. J. and Hoffmann. M. R. 1984. The occurrence of bisulfite aldehyde addition products in fog- and cloudwater. J. Atmos. Chem. (in press). National Oceanic and Atmospheric Administration 1983. Local climatological data (monthly summary) for Bakersfield. CA. Kern County air terminal. December 1982 and January 1983. Published by US Dept. of Commerce. National Climatic Data Center. Asheville, North Carolina. USA. Okita. T. 1968. Concentration of sulfate and other inorganic materials in fog and cloud water and in aerosol. J. Meteorol. Soc. Japan 46. 120-126. Petrenchuk, O. P. and Drozdova. V. M. 1966. On the chemical composition of cloudwater. Tellus 18. 260-286. Pruppacher, H. R. and Klett. J. D. 1978. Microphysics of clouds and precipitation. Amsterdam: Reidel. 418- 421, Reible. D. D.. Shair. F. H.. Smith. T. B. and Lehrman. D. E. 1984. The origin and fate of air pollutants in California's San Joaquin Valley (I). Winter. Aimos. Environ. (in press). Roach, W. T. 1976. On some quasi-periodic oscillations observed during a field investigation of radiation fog. Q.J.R. Meteorol. Soc. 102, 355-359. Russell, A. G. 1983. Analysis of oxalic acid impregnated Tellus 36B (1984), 4 POLLUTANT CONCENTRATIONS IN FOG DROPLETS . fitters for ammonia determination, Environmental Quality Lab open file report 83-1, California Institute of Technology, Pasadena, California. Russell, A. G., McRae, G. J. and Cass, G. R. 1983. Mathematical modeling of the formation and transport of ammonium nitrate aerosol. Atmos. Environ. 17, 949-964. Scherbatskoy, T. and Klein, R. M. 1983. Response of spruce and birch foliage to leaching by acidic mists. J. Environ. Qual. 12, 189-195. Schwariz, S. E. and White, W. H. 1981. Solubility equilibria of the nitrogen oxides and oxyacids in dilute aqueous solution. Adv. Environ. Sci. Eng. 4, \-45. Schwartz, S. E.. Newman, L., Hegg, D. A. and Hobbs. P. V. 1983. Measurements of sulfate production in Tellus 36B (1984), 4 147 285 natural clouds (discussion). Afmos. Environ. 17, 2629-2633. Smith, T. B., Lehrman, D. E., Reible, D. D. and Shair, F. H. 1981. The origin and fate of airborne pollutants within the San Joaquin Valley. Final report to the California Air Resources Board, Sacramento. California. Stelson, A. W. 1982. Thermodynamics of aqueous atmospheric aerosols. Ph.D. thesis, California Institute of Technology, Pasadena, California. Waldman, J. M., Munger, J. W., Jacob, D. J., Flagan. R. C., Morgan, J. J. and Hoffmann. M. R. 1982. Chemical composition of acid fog. Science 218, 677-680. 148 CHAPTER VII THE H,SO, - HNO, - NH, SYSTEM AT HIGH HUMIDITIES AND IN FOGS: A FIELD STUDY OF A WINTERTIME STAGNATION EPISODE IN THE SAN JOAQUIN VALLEY OF CALIFORNIA by Daniel J. Jacob, J. William Munger, Jed M. Waldman, and Michael R. Hoffmann Submitted to Journal of Geophysical Research (January 1985) 149 ABSTRACT Atmospheric acidity in the San Joaquin Valley of California was interpreted from a systematic study at a network of sites of the H,80, - HNO, - NH, system in fogwater, aerosol, and the gas phase. Spatial patterns of atmospheric concentrations were established that reflect the distribution of emissions within the valley. Fogwater was acidic - or not - depending on the relative proximities of local oil recovery operations (SO NO,); and agricultural and livestock feeding 2? activities (NH). The valley was efficiently ventilated when the temperature inversion broke up at some point in the day, but a severe stagnation episode was documented when a subsidence inversion based below the surrounding mountain ridges persisted for over five days. Progressive pollutant accumulation over the course of the stagnation episode was observed and interpreted in terms of production and removal mechanisms. An average SO, conversion rate of 2 % nl was determined in the overcast and stagnant boundary layer. Alkalinities decreased over the course of the episode due to the secondary production of strong acids, and inorganic acidities were observed at sites furthest from NH, emissions. The stratus cloud topping the boundary layer was found to slowly scavenge 50., but not NO,. The occurrence of fog during the stagnation episode led to general decreases in aerosol concentrations, probably as a result of enhanced removal by deposition. 150 INTRODUCTION Extremely high acidities have been reported in fogs and stratus clouds collected in Southern California (Munger et al., 1983; Brewer et al., 1983; Waldman et al., 1984; Jacob et al., 1984a,b). The acidity of these fogs was due to sulfuric acid and nitric acid, which had been partially neutralized by ammonia (Jacob et al., 1984a; Waldman et al., 1984). Therefore, one can attempt to interpret the “acid fog" phenomenon from the consideration of acid-base neutralization processes in the H,SO,-HNO,-NH, systen. Ammonia is directly released to the atmosphere from a variety of sources, but H,SO, and HNO, are mostly produced in the atmosphere itself by oxidation of reduced sulfur and nitrogen compounds. Oxidation of SO, to sulfate proceeds in the gas phase 2 (Calvert and Stockwell, 1984), in concentrated aerosol (Kaplan et al., 1981; Crump et al., 1982), and in dilute aqueous solutions (Martin, 1984). Oxidation of NO. to HNO, may proceed photochemically in the gas phase, or at the surface of particles, e.g., water droplets (Heikes and Thompson, 1983). Nitrite is slowly oxidized to NO, in dilute solutions (Damschen and Martin, 1983). Using data available for SCIV) and N(III) oxidation, Jacob and Hoffmann (1983) found that aqueous-phase oxidation of S(IV) in fog droplets could be an important source of sulfate under polluted conditions; they found that, on the other hand, aqueous-phase oxidation of N(III) provided only a very small source of nitrate. | The residence times of H,SO0,, 3 depend on the way they are partitioned between the gas phase and the HNO,, and NH, in the atmosphere 151 aerosol. Sulfuric acid is present as aerosol under usual atmospheric conditions, but HNO and NH, may have significant vapor pressures over 3 3 ammonium nitrate (Stelson and Seinfeld, 1982). The gaseous species are reactive and should be quiclky removed by deposition; Huebert (1983) has found deposition velocities of 1 - 5 cm s! for HNO,(g) over grass. On the other hand, the secondary particles formed from 4,50, - HNO, - NH, - H,0 mixtures are typically in the size range 0.1 - 1 um and have low deposition velocities, of the order of 0.01 -0.1 cm es"! (Sehmel, 1980; Slinn, 1982). Under supersaturated conditions, however, deposition velocities of aerosol particles increase considerably if the particles grow to fog droplet size. A general mechanism for the occurrence of fogwater acidity in polluted atmospheres must include five fundamental processes: (1) emissions of S0,, NO, and NE, (11) pollutant transport, (iii) atmospheric production of H,SO, and HNO, , (iv) detailed chemical speciation within the H,50, - HNO, - NH. - H,0 system, and (v) 3 2 scavenging within fogs. To attempt an interpretation of fogwater acidity in terms of the above processes, we have conducted a detailed field study of aerosol and fogwater chemical composition in the San Joaquin Valley of California. The San Joaquin Valley is the site of major oil recovery operations, which release large amounts of S50, and NO. In addition, agricultural and livestock feeding activities provide important Sources of NH, - Severe stagnation episodes, which are associated with fog and low-lying stratus clouds, occur frequently during the winter 152 aonths. They are caused by persistent subsidence inversions based a few hundred meters above the ground and below the surrounding mountain ridges (Holet's and Swanson, 1981). Tracer studies have documented the lack of ventilation in the valley during these prolonged episodes (Reible, 1982). Ina preliminary study at a single site in December 1982 - January 1983, Jacob et al. (1984a) documented the main features of aerosol and fogwater chemical composition during stagnation episodes in the San Joaquin Valley. The main features observed were as follows: (i) important accumulation of secondary aerosol proceeded under non-foggy but stagnant conditions, (ii) significant pollutant deposition occurred over the course of fog events, (iii) SO, conversion in fogs did not exceed a few percent per hour, and (iv) the pH of the fogs was determined by the availability of ammonia to neutralize the sulfuric and nitric acids present. In this paper, we report the results of a more comprehensive study conducted at multiple sites during January 1984. Spatial and temporal variations of aerosol and fogwater concentrations have been established that reflect both the geographical distribution of emission sources and the meteorological conditions. Pollutant accumulation and removal over the course of a stagnation episode have beem characterized under both foggy and non-foggy conditions. The H,SO, production rate has been estimated. A thermodynamic 24 interpretation of the partitioning of the H,S0,-HNO,-NH., system between the gas phase, the aerosol, and the fogwater is presented ina companion paper (Jacob et al., 1985). Deposition data are discussed by Waldman (1985). 153 EXPERIMENTAL The volumetric concentrations of aerosol, HNO,(g), and NH, (g) were monitored at eight sites over the period 31 December, 1983 - 14 January, 1984 (Figure 1). Samples were collected twice daily (0000 - 0400 PST and 1200 - 1600 PST) at the six valley sites and once daily (1000 - 1600 PST) at the two mountain sites (Tehachapi, Lake Isabella). The sampling stations were located on platforms 15 m above the ground (Wasco, Tehachapi), on the roof of a building or trailer (Bakersfield, Lost Hills, Buttonwillow, McKittrick, Visalia), or on the ground, 1.5 m above a grassy area (Lake Isabella). Two open-faced 47 mm Gelman Zefluor Teflon filters ( 1 um pore size) were operated side-by-side (10 lpm) to provide duplicate determinations of the inorganic content of the aerosol. The Stokes number at the filter inlet is 0.1 for a 50 wm diameter particle, so that even very large fog droplets should be sampled efficiently (Davies and Subari, 1982). A 47 mm Gelman Nylasorb nylon filter collected gaseous nitric acid immediately downstream of one of the Teflon filters, and an oxalic acid-impregnated glass fiber filter collected gaseous ammonia immediately downstream of the other Teflon filter. The filters were sealed in Petri dishes and kept at 4°¢ following collection. The Teflon filters were extracted in 10 ml of distilled defonized water (Corning Megapure) for 90 minutes using a reciprocating shaker; subsequent extractions gave satisfactory blanks. The extracts were analyzed for major fons using standard methods previously described by Munger et al. (1983). The nylon filters were extracted in a solution - 2 3 mM HCO, and 2.4 mM CO, ~ (a conventional ion chromatography 3 154 eluent), using the extraction procedure described above. Oxalic acid-impregnated filters were extracted and analyzed following the protocol of Russell (1983). Sulfur(IV) was found at low levels in the Teflon filter extracts, but since the S(IV) concentrations were small compared to S(VI), their contributions were ignored. Although most of the aerosol sulfur is expected to be present as sulfate, some of the measured sulfate may have resulted from the in situ oxidation of reduced sulfur species on the filter or in the extract. Fogs were sampled by event at four sites (Figure 1). Fogwater samples were collected with a rotating arm collector (Jacob et al., 1984c) for intervals ranging from 30 minutes to 3 hours. The rotating arm collector has a theoretical sampling rate of 5 n> min !, and has been shown to collect fogwater samples without evaporation or condensation. Laboratory calibration has indicated a lower size cut (50% collection efficiency) of 20 \m diameter. Because the instrument collects fog droplets by direct impaction and does not require drawing air through an inlet, large fog droplets are efficiently collected. Liquid water content in the fog was determined from the sampling rate of the instrument, assuming that 60% of the total liquid water sampled was actually collected (Jacob et al., 1984c). Fogwater samples were preserved and analyzed for major ions and trace metals following the protocol described by Munger et al. (1983), with the exception described below. In fogwater samples, significant S(IV) concentrations are found during conventional anion analysis by ion chromatography. A possible explanation is the formation of stable but reversible S(IV)-aldehyde adducts, such as hydroxymethanesulfonate (Munger et al., 1984). The standard ion chromatographic method 155 3 3 flow rate) proposed by Dionex (1981) and used by Munger et al. (1983) (Dionex AS-3 céluma, [ 3 mM HCO, + 2.4 mM CO,”-] eluent, 3 ml min ~ does not clearly separate S(IV) and NO, . Better separation is achieved with a weaker eluent or with the Dionex AS-4 column, but quantification of S(IV) remains unsatisfactory. To solve this problen, aliquote for anion determination were spiked to 0.09 M HAO, several minutes prior to injection; this addition resulted in the quantitative oxidation of bisulfite and hydroxymethanesulfonate to sulfate. Oxidation of hydroxymethanesulfonate by #,0, was probably facilitated by the presence in fogwater of metal catalysts (McArdle and Hoffmann, 1984). Sulfur(IV) concentrations were separately determined by a pararosaniline colorimetric method on aliquots preserved with buffered (pH 4) formaldehyde immediately upon sample collection (Dasgupta et al., 1980). Fogwater sulfate concentrations were calculated by subtracting the S(IV) concentrations thus obtained from the sulfate concentrations determined by ion chromatography. To date, there are no standardized sampling procedures for the collection of fogwater and aerosol samples for chemical analysis. Therefore, it is important to assess the errors associated with our methods. A detailed discussion of sampling biases, artifacts, and standard errors on our data is presented in the Appendix. Hourly average concentrations of 50, and NO. were measured at Bakersfield and Visalia by the California Air Resources Board, and at McKittrick, Kernridge, and Lost Hills by the West Side Operators (WSO). Hourly average CO concentrations and mixing heights were measured at Kernridge by WSO. Surface winds were measured by WSO (Lost Hills, Kernridge, McKittrick, Maricopa), Getty 011 Company 156 (Bakersfield), National Weather Service (Bakersfield), Lemoore Naval Air Force base, Visalia airport, and Kern County Fire Department (Tehachapi). Upper-level winds were measured at Edwards Air Force Base, located in. the Mojave desert on the other side of the Tehachapi mountains. Additional weather data were available from the National Weather Service station at Bakersfield. To account for the equilibria between gas and aerosol phases, we define N(-III), N(V), and S(VI) to represent the element at the given oxidation state, both in the gas and aerosol phases. Thus N(-III) includes NH, (g) and NH”, and N(V) includes HNO,(g) and NO, We further define [A] as the concentration of constituent A in fogwater Culeq 17! of water), and (A) as the concentration of A in the air (neq m > of air). The fogwater loading (A), is the concentration of A in fogwater per m of air, and is obtained by multiplying [A] by the liquid water content of the fog.. 157 WEATHER PATTERN AND POLLUTANT TRANSPORT Figure 2 shows the profile vs. time of boundary layer mixing heights and stratus cloud bases over the valley floor during the winter 1984 sampling program. Two types of mixing height diurnal patterns were observed. In the first pattern (the days of 31 December, 1 January, 10 - 12 January, 14 January), ground-based inversions formed by radiation at night and broke up briefly the following afternoon, leading to mixing heights in excess of 1000 m above ground level (AGL). This pattern was usually associated with clear skies or high cloudiness, but fogs in ground-based inversions were occasional occurrences (e.g., Bakersfield, 31 January). Figure 3a gives the average wind vectors on the days when this pattern was observed. A net slow NW flow was observed on the valley floor, consistent with NW winds at upper levels reflecting the circulation around the Pacific High. Terrain influences in the southern end of the valley led to convergence of the flow in the SE corner of the valley. Surface winds frequently shifted in direction, and erratic patterns of low winds were typically observed under nighttime stable conditions. Concentrations of trace gases at Kernridge were lowest on the days when this pattern was observed (Figure 2), and so were aerosol concentrations (Jacob, 1985). Because of the weakness of the NW flow on the valley floor, it is unlikely that surface winds ventilated the valley by transport over the mountain ridges in the southeast corner of the valley. Aerosol concentrations at Tehachapi and Lake Isabella remained much lower than in the valley, which is evidence against such 158 transport. Instead, pollutant removal was due to rapid vertical mfxing as the inversion broke up in the afternoon; this vertical mixing diluted the polluted air parcels and allowed their rapid transport by strong upper-level NW winds to the surrounding air basins. | A different mixing height pattern was observed on 2 ~ 7 January; during that period, a very strong subsidence inversion based a few hundred meters above the ground persisted over the valley. Subsidence was due to the presence of a stationary high-pressure center (Great Basin High) over the valley. Upper-level winds at Edwards AFB switched from the usual NW (circulation around the Pacific High) to E (circulation around the Great Basin High). A stratus cloud persisted in the valley below the inversion (Figure 2), and frequently intercepted the McKittrick site 250 m above the valley floor. On the night of 4 - 5 January, the stratus base lowered sufficiently to cause fog at elevated sites on the valley floor (Kernridge, Bakersfield NWS). On the nights of 5 - 6 and 6 - 7 January, the stratus base lowered sufficiently to fill the entire boundary layer, and dense fog was observed everywhere in the valley. The cloud layer deepened considerably on 7 - 8 January, and mixing heights rose to above 1000 m AGL; drizzle fell on the valley floor. On 9 January the inversion base dropped again, but upper-level winds had switched back to NW. After 10 January, the first mixing height pattern (surface inversions at night breaking up in the afternoon) was observed. The capping of the valley by a persistent inversion based ata lower altitude than the surrounding mountain ridges obviously restricted ventilation. Concentrations of trace gases at Kernridge 159 (Figure 2) rose to high levels during the period 2 - 7 January, but dropped to low levels on 8 January when the mixing height rose above the mountain ridges. Concentrations began increasing again on 9 - 10 January, and dropped on 10 January. The resultant winds in the valley during the 2 - 7 January period are shown in Figure 3b. Cross-valley winds were dominant in the southern end; further north, however, a slow net southerly flow out of the valley was apparent. Such a flow pattern has been previously documented during stagnation episodes by Reible (1982), who found from tracer studies that pollutant transport in the southern San Joaquin Valley (SSJV; see Figure 1 for definition) was extremely complex because of the upslope flow/downslope flow system associated with the mountain/valley breeze. Reible suggested that the SSJV could be crudely modeled as a stirred-tank reactor that was ventilated by the weak southerly air flow. He reported residence times of 2 - 8 days for air parcels in the SSJV. A net flow of air into the SSJV was observed on the west side. The dominant westerly surface winds in the southern end of the SSJV suggest that upslope transport over the ridges of the Tehachapi mountains and southern Sierra Nevada could have been a significant source of outflow for the SSJV air. However, the inversion did not break down or lift near the slopes: the stratus cloud intercepted the Bakersfield - Tehachapi highway as a well-defined fog layer, and the community of Tehachapi always remained sunny and very clear. Winds in the SSJV were entirely decoupled from the circulation aloft, as determined by the E upper-level winds at Edwards AFB and the SE afternoon winds at Tehachapi. Aerosol concentrations remained very 160 low at Tehachapi and Lake Isabella throughout the episode, confirming that upslope transport was not an efficient ventilation pathway. It is likely that shifts in wind direction occurred along the slope. Considerable fluctuations of both wind direction and speed with altitude have been previously documented during stagnation episodes in the SSJV (Aerovironment, Inc., 1982). Therefore, slow transport north out of the valley was the main outflow for SSJV air. Average winds at Lemoore, Fresno, and Lost Hills were consistent (Figure 3b). Projections of these winds on the 1 0.26 ms 2, valley axis (150°) yielded average speeds of 0.24ms_ and 0.22 m s! at each site, respectively. Since the SSJV is about 100 km long, this flow led to an average residence time of 5 days for an air parcel within the SSJV. The accumulation pattern for CO at Kernridge (Figure 2) suggests approach of a steady-state on 6 - 7 January, which is consistent with a residence time of 5 days. 161 AVERAGE AEROSOL AND FOGWATER CONCENTRATIONS AT EACH SITE 1. Aerosol and fogwater concentrations The SSJV is the site of important SO NO,» and NH, emissions 2’ 3 (Table 1). The geographical distribution of emissions is shown in 3 is emitted from confined feeding operations concentrated on the east side of the valley, especially around Figure 1. Most of the NH Bakersfield and Visalia. Another important source of NH, is cropland, which occupies most of the land in the valley floor not used for oil recovery operations, and the associated fertilizer use. Emissions of SO, and NO, are concentrated in the east side and west side oil fields 2 of the SSJV; the o11 field emissions originate mostly from small boilers, which release their exhausts 10 - 20 m above the ground and therefore affect primarily the surrounding air. Mobile sources (two major highways, off-road farm equipment, city traffic) also contribute to NO. emissions. Spatial patterns of aerosol concentrations and fogwater loadings (Tables 2 and 3) directly reflected the distribution of emission sources. The ionic content of aerosol in the valley was dominated by s0,7", NO, , and NH,” which contributed over 90% of the total measured ionic loading. Concentrations of N(-III) were highest at Bakersfield and Visalia, and lowest on the west side (Lost Hills, McKittrick). Mountain sites (Lake Isabella, Tehachapi), which have no important local NH, sources, had very low N(-III) concentrations. Concentrations of S(VI) were highest at Bakersfield, which is in the immediate vicinity of the largest oil field in the valley; they were 162 also very high ‘at McKittrick, which is located within the west side oil fields, and at Buttonwillow, which is directly downwind. Concentrations of S(VI) at Wasco, Lost Hills, and Visalia were lower, which reflects their respective distances from oil recovery operations. ‘A chemical balance on primary aerosol (based on Cooper and Watson, 1980) showed that all but a negligible fraction of the total S(VI) in the valley was of secondary origin. However, in spite of the time required for SO, conversion to sulfate, the areas of SO, emissions matched the areas of high S(VI) concentrations. Concentrations of S(VI) at the two mountain sites were very low and indicated little impact of the valley air. Because most of the NO, was emitted from the same sources as 50,5 one would expect the spatial distribution of N(V) concentrations to be similar to that of S(VI). Indeed, N(V) concentrations were highest at Bakersfield; however, concentrations at McKittrick were low. The lack of NH, at McKittrick led to frequently acidic 3 conditions, in which N(V) would be mostly present as HNO,(g) and therefore quickly removed by deposition. This point will be addressed in more detail below. Concentrations of N(V) were higher at Visalia than would be expected from the spatial distribution of S(VI); Visalia is an important population center, and it is therefore likely that emissions from mobile sources were the dominant source of N(V) at that site. Fogwater loadings of Ni and V, which are almost exclusively associated with residual oil burning (Cooper and Watson, 1980), were high at Bakersfield, McKittrick, and Buttonwillow, and low at Visalia. Loadings of Pb, a tracer for automobile exhaust, were higher in the 163 population centers (Bakersfield, Visalia) than at rural sites (McKittrick, Buttonwillow). Overall, the large differences in S(VI), N(V), and N(-IIT) concentrations, that were observed from one site to another over distances of only a few tens of kilometers, show that the composition of the H - HNO, - NH, system at each site was strongly determined 25°, 3 ~ NHS by the nature of emission sources in the immediate vicinity. This is especially interesting in regard to the secondary constituents S(VI) and N(V), and supports our observation that there was little net horizontal pollutant transport on the valley floor. The spatial distributions of SO NO, and NH, emissions are such that the 2° 3 acid-neutralizing capacity of the atmosphere should vary considerably within the valley. Indeed, Table 3 shows that fogwater acidity differed greatly from site to site, with low pH values at McKittrick and very high pH values at Visalia. Fog samples collected simultaneously at four sites during a widespread fog event clearly showed this spatial pattern (Figure 4). Acid fog was consistently observed at McKittrick, but never at the other sites. 2) Alkalinity and inorganic acidity We explore further the spatial variations in the acid-neutralizing capacity of the atmosphere by calculating the alkalinities at each site. Although alkalinities were not directly measured, they can be estimated from our data. We define the alkalinity of fogwater [ALK] as the deficiency of protons with reference to the system of “neutralized” fogwater species (caT™* cl, 164 NO, , SO, > s(Iv) , HA). The symbol HA refers to undissociated weak ‘acids in solution. In the pH 2 - 7 range, the main S(IV) species in the fog are expected to be HSO, and stable monovalent S(IV)-aldehyde adducts (Jacob and Hoffmann, 1983; Munger et al., 1984; Boyce and ‘Hoffmann, 1984). Therefore, the monovalent form of S(IV) is most appropriate for use as reference. For the subset of samples analyzed for carboxylic acids, we have calculated the fogwater alkalinity [ALK] as follows: 2 + [HCO,-] + [NH,(aq)] + [08] - [H"] (1) [ALK] = [HCOO ] + [CH,COO ] + [CH,CH coo }] + [CH,CH(OH)COO } where [H’] was directly measured and the concentrations of alkalinity~contributing species were determined from the aqueous concentrations and the equilibrium constants at 5°c (see Appendix). fo) _ —5 For NH,(aq), the equilibrium constants at 5 C are Ky .NH3 = 1,5xl0 ~ M, 15 vw? (Smith and Martell, 1976). K_ = 2.0x10 Ww Estimating the alkalinity of aerosol (ALK) , is less straightforward. We assume that the ions on the right-hand side of (1), plus possibly additional weak acid anions, constituted most of the aerosol ionic content unaccounted for in our chemical analysis. By a charge balance on the aerosol (concentrations in neq m >), we obtain: (ALK), = (Nat) + (KT) + (wH,*) + (ca?*) + CMe”) 2- - (C1”) - (NO) - (80, °) - (2) 165 We define the atmospheric alkalinity (ALK) by considering, in addition to aerosol alkalinity, the gaseous species HNO 4(g) and NH,(g): (ALK) = (ALK), + (NH,(g)) - CHNO,(g)) . (3) Negative values of ALK correspond to a situation where strong acids (mostly #,S0, and HNO, ) are not neutralized. When ALK is negative, we will refer to the absolute value of ALK as the inorganic acidity. This quantity is also sometimes referred to in the literature as the "mineral acidity” (Stumm and Morgan, 1981). Average alkalinities at each site are given in Table 4a. Due to the possibility of HNO, or NH, volatilization from filters collected in fog (see appendix), (ALK) and (ALK). were calculated only for non-foggy conditions. (ALK) and (ALK) , were usually small numbers determined by the difference of two large numbers, and so the errors are fairly large. The calculation of the fogwater alkalinity [ALK] involved subtracting a small number from a large number; the resulting standard errors are small and not indicated explicitly. On the average, the air contained substantial alkalinity at all sites except McKittrick and Lost Hills. This clearly reflected the spatial distribution of NH 50,, and NO, emissions, and the 3? deficiency in ammonia on the west side of the SSJV. Alkalinities were considerably higher at Visalia than at Bakersfield, even though N(-IIL) concentrations were similar at both sites. Further, alkalinities at Buttonwillow and Bakersfield were comparable even though N(-II1) concentrations were much higher at Bakersfield. Most of the alkalinity at Bakersfield was depleted by high H,SO, and HNO, 166 inputs. The partitioning of alkalinity between the gas phase and the aerosol is of interest. Scavenging of NH, (g) to form ammonium salts of weak acids could be a source of important alkalinity in the aerosol. Significant fogwater alkalinities were found at Bakersfield, Buttonwillow, and Visalia; however, under non-foggy conditions, the aerosol was never significantly alkaline. Although the error bars on the determinations of (ALK) are large, the absence of positive (ALK) , values even in the presence of large excesses of NH, (g) (as at Visalia) strongly suggests that the alkaline ammonium salts are volatile under non-foggy conditions. This hypothesis is supported by concurrent sampling of aerosol and fogwater at Visalia, where wu,” was in excess of NO, and s0,7- in the fog but not in the aerosol (Jacob et al., 1985). Under non-foggy acidic conditions, the aerosol contained significant mineral acidity when S(VI) was present in excess of N(-III). This occurred in 6 of the samples, all at McKittrick. In the remainder of the samples collected under acidic conditions, the aerosol was neutralized and the inorganic acidity was present exclusively in the gas phase as HNO,(g)- This observation is in agreement with aerosol equilibrium models (Bassett and Seinfeld, 1983); NH, (3g) is scavenged by acid sulfate aerosol until this aerosol is neutralized as (NH, ),S0, « Excess NH, may then combine with HNO, to add NH NO, to the aerosol phase, but HNO, in excess of NH, remains in the gas phase. The contributions of different species to fogwater alkalinity are shown in Table 4b. The main contributors to [ALK] at Bakersfield 167 and Buttonwillow were formate and acetate. Formic and acetic acids are efficiently scavenged in fogwater at pH > 5; they are highly soluble as indicated by their large Henry's law constants (Hor 298 7 . > 3 -I 3.7x10° M atm ; Hacer, 298 are mostly dissociated (pK.op 597, = 3+773 PKacer 078 and Smith, 1977). At Visalia, the fogwater pH was much higher than at = 8.8x10° M atm); JANAF, 1971) and they = 4.77; Martell Bakersfield or Buttonwillow, and the contribution from HCO, to [ALK] was correspondingly larger. Carboxylate anions did not provide higher contributions at Visalia than at Bakersfield or Buttonwillow, since carboxylic acids are already efficiently scavenged at pH 5 ~ 6, raising the pH higher leads to little additional scavenging. The speciation of N(-III) and N(V) in the atmosphere was directly related to the ambient alkalinity. Under alkaline conditions, only a small fraction of N(V) was present in the gas phase. At Bakersfield, Wasco, Buttonwillow, and Visalia, the average fraction of N(V) as HNO, (g) was less than 10%. Even under neutral to acidic conditions (Lost Hills, McKittrick) there was sufficient N(-III) available to draw most of the N(V) into the aerosol phase. At -Lost Hills and McKittrick, less than 15% of the total N(-III) was in the gas phase, but this fraction was larger at other sites. Gaseous ammonia constituted 64% of the total N(-III) at Visalia, which reflects the considerable excess of N(-III) over N(V) and S(VI) at that site. 168 EVOLUTION OF THE H,S0,-HNO,-NH, SYSTEM OVER THE COURSE OF A SEVERE STAGNATION EPISODE Figure 5 shows the evolution of concentrations at the six valley sites over the period 1 - 8 January. With a few exceptions (discussed below), concentrations increased over the 2 - 7 January stagnation episode and dropped on 7 - 8 January. The decrease of CO concentrations at Kernridge indicates that the increase in mixing heights on 7 - 8 January ventilated the valley; S(VI), N(V), and N(-III) were removed more efficiently than CO, which indicates that wet deposition was also an important contributor to pollutant removal. Dense widespread nighttime fogs on the valley floor (6, 7 January) were also associated with decreases in S(VI), N(V), and N(-III) concentrations. No drizzle or increase in mixing heights was observed on those days which could account for these decreases; therefore, they were likely due to rapid deposition of fog droplets. Jacob et al. (1984a) have previously suggested that fogs efficiently limit pollutant accumulation during stagnation episodes. In the winter 1982 — 1983 study, Jacob and co-workers compared fogwater ‘loadings in successive samples collected during the same event and found rapid decreases with time of the amount of material suspended in the fogwater. Unfortunately, such an analysis could not be adequately conducted for this year because of the paucity of extended fogs on the valley floor. Figure 4 suggests that fogwater loadings decreased with time at Bakersfield, Buttonwillow, and Visalia during the 7 January fog event; however, this evidence is not conclusive because liquid water contents were also decreasing. The apparent decrease in 169 fogwater loadings could possibly have simply been due to evaporation of fog droplets to below the collector size cut. Scavenging of NO. and 50. by fog can be evaluated from their concentration profiles. Concentrations of NO. at Kernridge increased steadily throughout the episode, even during the foggy 5 - 7 January period (Figure 2); NO. at Kernridge behaved similarly to CO, which is not water-soluble. At McKittrick, NO, concentrations increased steadily (Figure 6), even though the stratus cloud frequently intercepted the site during the 2 ~ 7 January period. Therefore, it appears that NO. was not scavenged by fog droplets. This is consistent with laboratory data documenting the poor solubility of NO in water at atmospheric concentrations (Schwartz and White, 1981). The profile of SO, concentrations at Kernridge was very different from that of NO. even though both gases originated from the same combustion sources. Concentrations of SO, rapidly reached a steady-state after an initial increase, and this is consistent with fairly high deposition velocities of 50, over grass-type surfaces ( 1 cm sl, Sehmel, 1980). Concentrations of 50, then decreased progressively over the 5 - 7 January foggy period, in contrast to CO and NO,- This is strong evidence that SO, was being slowly scavenged 2 by fog droplets. We infer from the SO, concentration profile at Kernridge that the rate of 50, scavenging was of the order of a few percent per hour. At McKittrick (Figure 6), concentrations of 50, fluctuated considerably while NO. concentrations increased relatively ‘steadily; this is consistent with SO, scavenging by the frequent fogs at that site. Steady increases in S(VI) concentrations were observed at 170 Bakersfield, Wasco, and Lost Hills over the period 2 - 5 January, when ‘no fog or drizzle occurred to limit pollutant accumulation. Sulfate concentrations did not appear to reach a steady state over that period. We have argued previously that a stirred-tank reactor ‘approximation could be used to interpret pollutant accumulation in the SSJV. For the period 3 —- 5 January, the average mixing height h was about 400 m AGL. The characteristic time T for S(V1) removal is given by: l Yso4 1 ns Ce a (4) r a where Ts is the characteristic time for advection out of the SSJV (5 days), and Vso4 is the deposition velocity for sulfate aerosol. Considering that there was still no evidence of a S(VI) steady state 1 on 5 January, should be smaller than 0.05 cms. This is in ‘s04 ‘agreement with deposition velocities for particles in the 0.05 - 1 Lm size range at low wind velocities (Sehmel, 1980). Deposition velocities for S0,(g) are of the order of 1 cm s (Sehmel, 1980); therefore, we expect roughly steady-state so, (g) concentrations to be reached on a time scale of about one day after the onset of the stagnation episode, as was observed at Kernridge. We can then estimate the average rate of sulfate production from the increase in sulfate concentrations over the period 3 - 5 January, assuming that sulfate deposition can be ignored over that period. The average rates 171 -~1 -1 of (S(VIL)) increase were 9 neq n>? h ~ at Bakersfield, 3 neq n> h at Wasco, and 3 neg n> nl at Lost Hills. Modeling of S0, concentrations (Aerovironment, Inc., 198 ) indicate that the SO, concentration field in the SSJV during a stagnation episode should be bounded on the lower end by the concentrations at Wasco and Lost Hills, and on the upper end by concentrations at Bakersfield and McKittrick. Over the period 3 - 5 January, 50, concentrations averaged 3 ppb at Lost Hills and 25 ppb at Bakersfield. Therefore, the average pseudo first-order 50, conversion rate k should lie between 0.8 and 2.5 % nl, Alternatively, we can calculate the S05 conversion rate from an estimate of the average SO, concentration in the SSJV based on steady-state attainment in a stirred-tank model: E $02 (S0,)., = : ; (5) $02 k + at a where E599 = 0.95 leq n> day | is taken from Table 3. We obtain by this method (S0,)_ = 9 ppb, and 1 % ni¢k<3%h 1. This is consistent with the above result. During the period 3 - 5 January, the stratus cloud filled a large fraction of the boundary layer. Although we have presented evidence that SO, was slowly scavenged in fog, this does not necessarily imply enhanced sulfate production in fog; S(IV) may be stabilized in the aqueous phase by adduct formation (Munger et al., (1984), or removed by deposition before being oxidized. In-cloud sulfate production cannot be reliably interpreted 172 from our fogwater sulfate data, because we cannot ascertain the history and fate of droplets collected in successive samples at a given site. We tried to account for transport by using Ni and v as tracers for sulfur, but were unsuccessful at obtaining a significant rate of sulfate production in fogwater. Details of the calculation are given by Jacob (1985). In addition to H,SO,, HNO, was produced in the boundary layer. Under the alkaline to neutral conditions found at Bakersfield, all S(VI) and N(V) combined with N(-III) to produce a neutralized (ne,”, NO, , SO 2~) aerosol. However, S(VI) and N(V) did not accumulate in 4 the same way (Figure 5). Concentrations of S(VI) did not begin increasing until 3 January, showing a time lag consistent with the slow rate of 50, conversion: SO, must first build up before important S(VI) production proceeds. After 3 January, S(VI) concentrations rose steadily, consistent with the slow deposition velocity for sulfate aerosol. On the other hand, N(V) concentrations rose rapidly on the first day of the episode but did not increase after 3 January. This could be explained if nitrate was present in coarser particles than sulfate and therefore deposited more rapidly; attainment of a Steady state on a time scale of 1 day would indicate a deposition velocity for nitrate over one order of magnitude higher than that for sulfate. The presence of nitrate in coarse aerosol has been documented in the Los Angeles basin (Appel et al., 1978) and can be explained by absorption of HNO, (g) onto primary aerosol. In the San Joaquin Valley, however, the relative paucity of cations other than NH,” indicates that nitrate ehould be present mostly as secondary ammonium salt particles. The Kelvin effect leads to accumulation of 173 nitrate in particles larger than sulfate, but the resulting difference in sizes is very small at high relative humidities (Bassett and Seinfeld, 1984) and could not explain an order of magnitude difference in deposition velocities. Previous wintertime aerosol measurements in the San Joaquin Valley (Heisler and Baskett, 1982) did not reveal differences in the distributions of nitrate and sulfate between the coarse and fine aerosol. A better explanation for the constancy of N(V) concentrations after 3 January is that conversion of NO. to N(V) was limited by the lack of appropriate oxidants such as OH or 0 Little photochemical 3° activity is expected under overcast conditions; ozone concentrations at Bakersfield were below the detection limit of 10 ppb. The lack of nitrate production would explain the persistent accumulation of NO. observed over the course of the stagnation episode. Low photochemical activity indicates that conversion of SO, to sulfate proceeded predominantly via heterogeneous processes. As H,SO0, was produced over the course of the stagnation episode, alkalinities decreased at all SSJV sites. At Wasco, Lost Hills, and McKittrick, significant inorganic acidities were eventually produced; at Bakersfield, alkalinities were neutralized but no significant excess acidity was found. Strong acids titrated N(-III) into the aerosol, and consequently the air parcel became acidic when the supply of NH, (g) was exhausted. At Lost Hills, acidification coincided with a brief increase in HNO,(g) concentrations, followed by a drop in total N(V) concentration. This can be explained by displacement of NO,” by so, in the acidic aerosol, followed by rapid deposition of HNO,(g). Because of its high deposition velocity, HNO, (g) should be 174 removed from the boundary layer on a time scale of less than a day. McKittrick frequently intercepted stratus clouds after 3 January, and received occasional drizzle. Concentrations of N(-III) peaked on 2 - 3 January, and decreased afterward; the decrease may have been due to removal by fog and drizzle, but also to advection by dominant westerly winds of ammonia~poor air parcels from the coastal mountain ranges. Concentrations of S(VI) at McKittrick rose at the beginning of the episode, but accumulation was apparently limited by fog deposition. Concentrations of N(V) did not stabilize in the same way as concentrations of S(VI), and fell to low levels after 3 January. Transport is not by itself a satisfactory explanation for this behavior since N(V) and S(VI) at McKittrick are expected to originate from the same sources, and NO. concentrations did not show the drop observed for N(V) (Figure 6). As in the case of Lost Hills, the drop in N(V) concentration coincided with acidification of the air parcel and was preceded by a brief increase in HNO,(g) concentration. Displacement of NO,” by so,77 in the aerosol followed by rapid removal of HNO,(g) is a likely explanation. An additional explanation is that conversion of NO. was limited by the low availability of oxidants; that is, N(V) was removed in the fog by deposition and was not replenished in the atmosphere. As opposed to N(V), S(VI) concentrations did not decrease because of continual production by heterogeneous processes. Concentrations of N(-III), N(V), and S(VI) at Visalia (Figure 5) did not show the same patterns as in the SSJV. Visalia was overcast until 5 January except for a 1+h fog on 3 January, but dense, precipitating fog occurred on the morning of 5 January from 0200 PST 175 on. Concentrations of S(VI) remained low throughout the stagnation episode, reflecting the distance of Visalia from the SSJV oil fields. Nitric acid was not produced in sufficient amounts to deplete the very high alkalinity at Visalia, and large amounts of N(-III) remained in the gas phase. On 5 January, precipitating fog efficiently removed aerosol nH,” and NO, from the atmosphere, but did not affect the concentration of NH,(g)- At the very high pH values found in Visalia fogwater, NH, (g) is not scavenged by fog droplets (Jacob et al., 1985). The above discussion has shown that the profiles of concentrations vs. time during a stagnation episode can be succesfully interpreted based on stirred-tank considerations of pollutant accumulation and removal; however, the differences in concentrations from site to site clearly indicate that a stirred-tank model for the SSJV as a whole is not an adequate modeling tool. The major reason for that is the lack of internal mixing, and therefore the persistence of concentration patterns directly reflecting the local emissions. Modeling could possibly be approached by subdividing the SSJV into a number of cells where the stirred-tank approximation could be invoked. Such an exercise is beyond the scope of this paper; however, for the sake of illustrating the above discussion. we will describe two different patterns to be expected from such modeling. The accumulation of constituent A in a stirred-tank is described by equation (6): mE, + k(B) - RNA) EC Avy + (Ay + Ap yy) - d(A) dt g a £ a 176 are the gas-phase, aerosol~phase, and where Ay» A> and A, fogwater-phase species, respectively, k is the pseudo first-order rate of conversion from precursor B to A, k' is the pseudo first~order rate of destruction of A, and E, is the emission rate. We simultaneously A solved the coupled stirred-tank equations for SO. > NO,» S(VI), NCV), and N(-III), under the conditions of Table 5. An exponential decay for the N(V) production rate was arbitrarily chosen to account for the limitation of the reaction by the availability of oxidants. We assumed that the aerosol was a neutralized mixture under non-foggy conditions if N(-III) was in excess of S(VI), and that the formation of NH NO, aerosol was sufficiently favored to prevent HNO, and NH, from coexisting in the gas phase. Further, based on the results presented in Jacob et al. (1985), we assumed that (a) 30% of the aerosol is present in the fogwater at any given time, (b) HNO, (g) is 100% scavenged under all foggy conditions, (c) NH, (g) is not scavenged in fog under alkaline conditions (N(-LII) > N(V) + S(VI)). Figure 7a was obtained from Case I of Table 5, which considers average SSJV emissions as calculated from Table 3. Aerosol accumulates rapidly under non-foggy conditions, and is partially removed by fog. Concentrations of N(V) and S(VI) show a short time lag at the beginning of the episode: SO. and NO. must first build up before significant production can occur. Excess alkalinity as NH,(g) remains present throughout the episode. Acidic conditions as observed at McKittrick and Lost Hills can 177 be interpreted by a simulation which assumes higher SO, and NO, ‘emissions, and lower NH, emissions, than the average SSJV values (Case II of Table 5, Figure 7b). Under such conditions, inorganic acidity appears early into the episode, and a large fraction of N(V) remains in the gas phase. Because HNO, (g) is rapidly removed by deposition, N(V) concentrations drop after a brief initial increase. The presence of N(V) as HNO.,.(g) also increases the scavenging efficiency of N(V) by fog, and the subsequent rapid deposition. 178 CONCLUSION Fogwater acidity in the wintertime atmosphere of the San Joaquin Valley of California was interpreted from the composition of the H,S0, - HNO, - NH, system at a network of sites. Fogwater acidity was found to be directly determined by the proximity of oil recovery operations (S0,, NO.) and livestock feeding activities (NH). Thus, a region of prevailing acidic conditions was identified on the western edge of the valley; alkaline conditions were found elsewhere. Considerable alkalinity and very high fogwater pH were found at a site distant from the oil fields but near major confined feeding operations. In view of the close balance of acidic and alkaline species presently found in the southern San Joaquin Valley, changes in the activities of either the oil industry or the cattle and dairy industry could dramatically affect fogwater acidity. This can explain why fog collected at Bakersfield was often acidic in December 1982 - January 1983 (Jacob et al.,1984a), but not in January 1984. Pollutant concentrations were strongly affected by atmospheric stability, and by the occurrence of drizzle and fog. Instability in the boundary layer extending to above 1000 m AGL efficiently ventilated the valley. Accumulation in the boundary layer was evident under non-foggy conditions when strong and persistent inversions limited the ventilation, but decreases in concentrations were observed following fog. Aerosol removal in fogs was attributed to rapid deposition of material scavenged by the fog droplets. . Concentration profiles at each site followed a stirred-tank type of behavior over the course of a severe stagnation episode. The 179 accumulation patterns were consistent with very low deposition velocities for secondary ammonium sulfate and nitrate aerosol ( < 0.05 cm s'), and a high deposition velocity for HNO, (g) (> 1 em a). In the absence of fog, slow sulfate production (of the order of 2% a) was observed on the valley floor, and enhanced scavenging of 50, was apparent in the stratus cloud above the valley floor. Sulfate appeared to be produced by heterogeneous processes, but we were not able to obtain statistical evidence of in~-cloud sulfate production. Nitrogen oxides were not scavenged in fog, and the conversion of NO. to HNO, was limited by the low degree of photochemical activity. 3 Steady decreases in alkalinity associated with H,SO, accumulation were observed over the course of a stagnation episode, and inorganic acidities were observed at sites distant from ammonia sources. Under acidic conditions, nitrate was displaced from the aerosol and HNO,(g) was apparently quickly removed by deposition. As a result, decreases in total nitrate were observed as the air parcel became acidic. ACKNOWLEDGEMENTS We thank the organizations that provided us with sampling sites: California Air Resources Board, Western Oil and Gas Association, Buttonwillow Park and Recreation, Kern County Fire Department, and U.S. Army Corps of Engineers. We further thank the organizations who provided us with atmospheric data: West Side Operators, California Air Resources Board, Getty O11 Company, Lemoore Naval Air Station, ‘Tehachapi Fire Staion, Edwards Air Force Base, and Kern county Air Pollution Control District. We express our gratitude to D. Buchholz 180 and K. Mayer for their help in the field, and to M. Lemons of Getty Oil Company and K. Carter of the Boy Scouts of Lake Isabella for their contributions to the success of the sampling program. G. R. Cass (Caltech) provided many helpful discussions. This work was funded by the California Air Resources Board (contract No. A2-048-32). REFERENCES Aerovironment, Inc. 1982. Development, validation, and application of the AVKERN model. Report AV-FR-80/603R, Aerovironment, Inc., Pasadena, CA. Aerovironment, Inc. 1983. AVKERN application report. Report AV-FR-83/501R2, Aerovironment, Inc., Pasadena, CA. Appel, B.R., Kothny, E.L., Hoffer, E.M., Hidy, G.M., and Wesolowski, J.J. 1978. Sulfate and nitrate data from the California Aerosol Characterization Experiment (ACHEX). Environ.Sci.Technol. 12,418-425. Appel, B.R., Wall, S.M., Tokiwa, Y., and Haik, M. 1980. Simultaneous nitric acid, particulate nitrate, and acidity measurements in ambient air. Atmos.Environ. 14, 549-554, Bassett, M.E., and Seinfeld, J.H. 1983. Atmospheric equilibrium model of sulfate and nitrate aerosols. Atmos.Environ. 17,2237-2252. Bassett ,M.E. and Seinfeld,J.H. 1984. Atmospheric equilibrium model for sulfate and nitrate aerosols-II. Particle size analysis. 181 Atmos.Environ. 18,1163-1170. Boyce, S.D. and Hoffmann, M.R. 1984. Kinetics and mechanism of the formation of hydroxymethanesulfonic acid at low pH. J.Phys.Chem. 88, 4740-4746. Brewer,R.L., Ellis,E.C., Gordon,R.J. and Shepard,L.S. 1983. Chemistry of mist and fog from the Los Angeles urban area. Atmos.Environ. 17, 2267-2271. California Air Resources Board 1982. Emission Inventory 1979, Stationary Source Control Division, Emission Inventory Branch, Sacramento, CA. Calvert ,J.G. and Stockwell,W.R. 1984. Mechanism and rates of the gas-phase oxidations of sulfur dioxide and nitrogen oxides in the atmosphere. in Acid Precipitation: S0,, NO, AND NO, Oxidation Mechanisms: Atmospheric Considerations. Calvert,J.G., ed. Butterworth publ., Boston, 1-62. Cooper,J.A. and Watson,J.G.Jr. 1980. Receptor-oriented methods of air particulate source apportionment. J.Air Pollution Control Assoc. 30,1116-1125, Crump,J.G., Flagan,R.C. and Seinfeld,J.H. 1983. An experimental study of the oxidation of sulfur dioxide in aqueous manganese sulfate 182 aerosols. Atmos.Environ. 17,1277-1289. Damschen,D.E. and Martin.L.R. 1983. Aqueous aerosol oxidation of nitrous acid by 0,, 0,, and H,0,. Atmos.Environ. 17,2005-2011. 2’ 22 Dasgupta,P.K., DeCesare,K. and Ulirey,J.C. 1980. Determination of atmospheric sulfur dioxide without tetrachlormercurate(II) and the mechanism of the Schiff reaction. Anal.Chem. 52,1912-1922. Daum,P.H., Schwartz,S.E. and Newman,L. 1984. Acidic and related constituents in liquid water stratiform clouds. J.Geophys.Res. 89, 1447-1458. Davies, C.N., and Subari, M. 1982. Aspiration above wind velocity of aerosols with thin-walled nozzles facing and at right angles to the wind direction. J.Aerosol Sci. 13,59-71. Dionex,1981. Determination of anions in acid rain. Application note 31, July 1981. Dionex Corp., Sunnyvale, CA. Heikes, B.G., and Thompson, A.M. 1983, Effects of heterogeneous processes on NO,, HONO, and HNO, chemistry in the troposphere. J.Geophys.Res. 88, 10883-10896. Heisler, S. and Baskett, R. 1982. Particle sampling and analysis in the California San Joaquin Valley. ERT Document P-5381-700, Environmental Science and Technology, Inc., Westlake Village, 183 California. Hering,S.V. and Blumenthal,D.L. 1984. Fog sampler intercomparison study: final report. Coordinating Research Council, 219 Perimeter Parkway, Atlanta,GA. Holets,S. and Swanson,R.N. 1981. High-inversion fog episodes in central California. J.Appl.Met. 20,890-899. Huebert, B.J. 1983. Measurements of the dry-deposition flux of nitric acid vapor to grasslands and forest. in Precipitation Scavenging, Dry Deposition, and Resuspension H.R.Pruppacher, R.G. Semonin, and W.G.N. Slinn, eds.2,785-794. Jacob, D.J. and Hoffmann,M.R. 1983. A dynamic model for the production 2- in urban fog. J.Geophys.Res. 88, 6611-6621. + - of H', NO, , and SO, Jacob,D.J., Waldman,J.M., Munger,J.W. and Hoffmann,M.R. 1984a. A field investigation of physical and chemical mechanisms affecting pollutant concentrations in fog droplets. Tellus.36B,272-285. Jacob, D.J., Waldman,J.M., Munger,J.W. and Hoffmann,M.R. 1984b. Chemical composition of fogwater collected along the California coast. Environ.Sci.Technol. (submitted). Jacob,D.J., Wang,R.-F.T. and Flagan,R.C. 1984c. Fogwater collector design and characterization. Environ.Sci.Technol. 18,827-833. 184 Jacob, D.J. 1985. The origins of inorganic acidity in fogs. Ph.D. thesis, California Institute of Technology, Pasadena, CA. Jacob, D.J., Waldman, J.M., Munger, J,W., and Hoffmann, M.R. 1985. Comparison of field data with thermodynamic calculations for the H,S0, - HNO, - NH, J.Geophys.Res. system at high humidities and in fogs. Submitted to Kaplan,D.J., Himmelblau,D.M. and Kanaoka,C. 1981. Oxidation of sulfur dioxide in aqueous ammonium sulfate aerosols containing manganese as a catalyst. Atmos.Environ. 15,763-773. Katz,U. 1980. A droplet impactor to collect liquid water from laboratory clouds for chemical analysis. in Communications a la geme Conference Internationale sur la Physique des Nuages. Publ. by Laboratoire Associe’ de Météorologie Physique, B.P. 45, 63170 Aubfere, France, pp. 697-700. Martell, A.E., and Smith, R.M. 1977. Critical stability constants, vol.3. Plenum, New York. Martin, L.R. 1984. Kinetic studies of sulfite oxidation in aqueous solution. in Acid Precipitation: SO,, NO, and NO Oxidation Mechanisms: Atmospheric Considerations. Calvert,J.G., ed., Butterworth Publ., Boston, pp.63-100. 185 McArdle, J.V., and Hoffmann, M.R. 1984. Kinetics and mechanism of the oxidation of aquated sulfur dioxide by hydrogen peroxide at low pH. J.Phys.Chem. 87, 5425-5429. Munger,J.W., Jacob,D.J., Waldman,J.M. and Hoffmann,M.R. 1983. Fogwater chemistry in an urban atmosphere. J.Geophys.Res. 88,5109-5121. Munger,J.W., Jacob,D.J. and Hoffmann,M.R. 1984. The occurrence of bisulfite-aldehyde addition products in fog and cloudwater. J.Atmos.Chem. 1, 335-350. Reible,D.D. 1982. Investigations of transport in complex atmospheric flow systems. Ph.D. Thesis, California Institute of Technology, Pasadena, California. Russell, A.G. 1983. Analysis of oxalic acid impregnated filters for ammonia determination. Environmental Quality Laboratory open file report 83-1, California Institute of Technology, Pasasdena, California. Russell,A.G. and Cass,G.R. 1984. Acquisition of regional air quality model validation data for nitrate, sulfate, ammonium ion and their precursors. Atmos.Environ. 18, 1815-1827. Schwartz,S.E. and White,W.H. 1981. Solubility equilibria of the nitrogen oxides and oxyacids in dilute aqueous solution. Adv.Env.Eng.Sci. 4,1-45. 186 Sehmel,G.A. 1980. Particle and gas dry deposition: a review. Atmos.Environ. 14,983-1011. Slinn,W.G.N. 1982, Predictions for particle deposition on vegetative canopies. Atmos.Environ. 16,1785-1794. Smith, R.M., and Martell, A.E. 1976. Critical stability constants, vol.4. Plenum, New York. Spicer, C.W. and Schumacher, P.M. 1979. Particulate nitrate: laboratory and field studies of major sampling interferences. Atmos.Environ. 13, 543-552. Spicer,C.W., Howes,J.E., Bishop,T.A. and Arnold,L.H. 1982. Nitric acid measurement methods: an intercomparison. Atmos.Environ. 16,1487-1500. Stelson,A.W. and Seinfeld,J.H. 1982. Relative humidity and temperature dependence of the ammonium nitrate dissociation constant. Atmos.Environ. 16,983-992. Stumm,W. and Morgan,J.J. 1981. Aquatic Chemistry, gnd ed., Wiley Interscience, New York, Waldman,J.M., Munger,J.W., Jacob,D.J. and Hoffmann,M.R. 1984. Chemical characterization of stratus cloudwater and its role as a vector for pollutant deposition in a Los Angeles pine forest. Tellus 37B (in 187 press). Waldman, J.M. 1985. Contribution of fog to the depositional flux of acidity. Ph.D.Thesis, California Institute of Technology, Pasadena, California. 188 °BIeP ygol ArenuBL — (Ce61) qeoEr (Pp) *uoF{onpold ees pur [jo ATIwow (9) *wIEP 6461 — (7861) pABog seOANOBAY APY BTUIOFTTED (q) “BI6P 796] JeqMEIOR — (796) “SUL ‘qUeMOITAQIay (2) ap/aw0d 61 "WiOl z LoFIENqUOD Tang £ 73 sam0g ol aon Jaz} F704 81 TI°s 94 9poysaF] (Asp/su03) suojecpme a0anos (P) Zep/euor 061 ‘WIOL rao S20INOE BTIIH Stl (2) nee AXBUOFITIG (Aep/euo3) eucpesyme so1n0e x q) & uoFonpaad 110 (4ep/au0j) suoyssTae aamoe , z (wy “*P/09) “os 2 cbs Hay *(wOFIBITFEP 303 | eanByg see tarss) LOTTRA UTNbEOL uss UEAYINOS ayy 303 sePIOWRAUT UOPETAM “EN pur ON “og °T @TSRL 189 SY ST * £ ra WI ra 8 4 1-8 8 bf-LI> § 8I-"> «=Le> «OS TG-0D> E> = -9ZI-OT af Averumep { 03 JeqmEoay TE woIaz (2) Areruee y] 03 Azeruep ¢ Worx (4) *J08 BJEp ayeTdIOO ayI 10F (CBI) Gooer eee (F) ra ra CL-@> OFI-y> C64> Cl wus1 a z 6t 7 6 6 oo » TW & O 1D A> LD H*-S E> E> 0D 9 EO My BED? 9 wl om ¢ 68 Sa om » @ 6 » &£ 690-489 TELL 7OH-VBT SI-b> OZI-SE BIE-EL TOU SZ 16-9 GE-BOl Ly 9I-> ne > 2 TOOI-SET SOLE SBT LE-Y> ELSIE OO-HE —TP-O2> II-Y> 8-9 9-9 OID ISH | <> A ISy-LE -LYE-IZ_SUR-LID. YOT-Y> ZOB-9E 9NZ-OT LOD OID EE9 HB ET-Y> 01-5 we ott 1] & of wt oo > af w@ » U LI9-SS YOY «= BE-LID HLT TEST Ole” © EPOZ> TI sy ESS LD BD ac 8080Os(atitk(<té‘iSC(<;eSCTC CST CO OTCESC( BYO-S6 «TEDS««ELZ-LID 9G-H> ZIE-BT GTZ 6-00 O1-D> EY AS-HL O~H> 98D oc wz =o Of 9 Of OR Fl lll WOCT-60Z TL-E6 CRP-ET «HD SOREL L9H-SB (9-02 ZI-H> SL-9 GHIT-HVT 6I-Y> 95> an beu 7 afr (3% 1 § 4 qa-y NN fon Ys fon TIN HN “(e) weet Areruer 9] = ga6T JeqE0aG Te “exp (3)SN PR «(3)% om qdepeyay, usa Il ue1 STTeqesy axe] usc bt aBuer (a) FA Upgal Qt aurz (qy OTT HMOF I ueae 62 aBurl 2~HRVPPR usc GZ aBue1 STTTH 380] Upe 6% ue COsEY, upall Q¢ Bue2 PLeTssroxeg Josoxey °% 8TqaL 190 any] Suy[dmes [e}07 ayQ pur peWeTTOO Je7eM JO JUNG [eI] By] UO peseq ‘JUeqUCD JajqEm pyNbyT aBerane (q) Jas eJEp oJaTdMoD BYI JOJ (GR6T) Gover ses (B) 0 ot S9 €s 9 8 Il L 19 rh 1] BTTESTA 0 € 6 vat @ ce vu ce va ra ran 9 MOTT HII ING, 0 é € ez 92 16 ey ral Le 9 92 A 3PHIIPPH 6 ST GST Ss? é TY 6€ 61 eT T€ 8tY S PTOF SIMPY wii yt Ba ayeuotdoid = 8BOeT aqeIe0R «BHILIOJ u A Ww 114) qd Uy aq “a _ @ys 670°0 = aT G9 Tv SII ¢ 6 O80T S°S TO = €2°L-1S°S €1 PTTPSTA o0so°O. #4. 09 ces LY Ot 2 L£90T_~—s ET 9°S = 6L°9-T0°S £ MOTT PMUIOI ING 1°O)—& ove 0SZ ST Gc «6 o8y =o €6 = €2°S-89°C 8S PHAN £80°0 78t 040 618 ral €€ 691 OLcE ah O°S = =26°9-01°S ) PTeTssrsyeg as 3 aa bert (a) 1 (AL)S _/08 ton _D 47 47 "HN (NH aBuez Hd “a 2778 * (ey 7861 MteMmEL yI-E861 TE TeqmoAT “suoyIeAIUEOUOD Joqendoj ederane peqysfam—toqem pynby] *¢ PTqeL Table 4a. Average alkalinities at each site. Site Bakersfield Wasco lost Hills McKittrick Buttonwillow) Visalia’ n AX ~3 —— mga 111431 46419 -1H16 -12H8 131431 441430 AIK a -13H18 2416 20417 14428 7418 191 a AIK var 2 210 38 98 2 183 6 220 p® gm 0.11 0.11 0.05 0.03 ALK is the alkalinity of the atr parcel, ALK, 1s the aerosol alkalinity, AUK; is the foguiter alkalinity. ALK and ALK were calculated only for non-foggy periods. ALK, was calculated for the Error bomds are the standard errors on the determinations of the means. (a) average liquid water content for the subset of fogwater samples. (>) from 5 January to 14 January (c) from 31 December to 7 Jamary Table 4b. Contributions of different species to fogwater alkalinity. Site Qn Bakersfield 2 McKittrick 3 Buttorillow 2 Visalia 6 free (4) acidity’* formate “1.1 45 -123 22 -7.7 142 0.0 53 (a) free acidity = [H*} - [af] acetate lactate 1% 1.1 t of samples analyzed for organic acids 2 propionate HOD, NAL,(aq) contribution to alkalinity (yeq 17) 15 9.5 6.3 0.2 1.6 0.0 0.2 0.0 3.7 0.0 Ll 0.0 77 0.0 92 1.5 192 fee «w wiigeg = FAG «ep a wtgez = Na +) dep wm wriget = 8a sry ae kep mrtgtt = Ny «sep | w WtET “Na «Aap. Hl G6°0 = OS; 21 388 Seyel UOTSSTIY “(shop UE 2) (32-)h@ OT “ou <— “on OL (203) "0s <—~ “os zZ (03 ou) _,"0s <— “os - rs BI}E1 UOSIGAU) I (oon € (3) one ¥ sjatdorp Boy *skep £ = 3 03 Shep Z = 3 wory Boy soo ("0s ‘on *,"iN ‘,H) Tosoze “0 = 2:98. 0 = (Sy) = Conn) = (“os’H) :s00737pu00 TeyATUL To “ON W OOF :34sTay BUH I eos *skep ¢ :sTeored aye jo aly. ausppsey 1? wo SefIPooTsA WoT y}sodaq * UOPIETMUTS Tepoul yuE]-paliyys Joy sieyamereg * ¢ STIL 193 APPENDIX: ERROR ANALYSIS 1. Fogwater concentrations A recent intercomparison study of fogwater collectors (Hering and Blumenthal, 1984) has demonstrated that the Caltech rotating arm collector provides reproducible and representative samples under both light and heavy fog conditions. The uncertainty on the concentrations of major ions (determined from samples collected with two collocated rotating arm collectors) was found to be about 15%. Errors due to chemical analysis in the laboratory were about 5% for all analyzed ions. Hering and Blumenthal (1984) found that our rotating arm collector (lower size cut 20 ym) and a jet impactor (lower size cut of 2-5 ym; Katz, 1980) yielded samples with statistically undistiguishable concentrations of major ions, in spite of the difference between the lower size cuts of the two instruments. This suggests that small fog droplets are not sufficiently different in composition to affect the total solute loading. Ionic balances are an indicator of whether all ionic components in the sample have been accounted for in analysis. Ionic balances in the fogwater samples were 0.98 + 0.18 at Bakersfield (n=15), 1.02 + 0.09 at McKittrick (n=53), 1.11 + 0.18 at Buttonwillow (n=7), and 0.60 + 0.17 at Visalia (n=12). The fonic balances were calculated from the following equivalent ratio: [c1™]+{NO,”1+[80,7"}+[S(IV)] + Ionic balance = [w']+(Na*}+ (NH, *}+1ca7*}+IMg7*) (Al) 194 where each mole of S(IV) was assumed to provide one equivalent (Jacob and Hoffmann, 1983). The most reliable tonic balances were found at McKittrick, while at Visalia there was a considerable and consistent anion deficiency. Fogwater at Visalia had a consistently high pH, and as a consequence could contain important alkalinity; therefore, additional weak acid anions must be considered in an ionic balance. In fogwater with low pH such as at McKittrick, weak acids are mostly present in undissociated form and thus equation (1) represents a reasonably good balance of cations to anions. The principal ionic contributors to alkalinity in the pH range 4 - 8 are expected to be HCO, (H x Kal = 7.07x10 2 v2 atm! at 5°C) and carboxylate ions. Fogwater concentrations of some carboxylate ions in the fogwater were determined by J. W. Munger (unpublished data; average values are given in Table 3). The acidity constants of these ions at 5°C are 1.3x1074 for lactic acid, 1.7x107" for formic acid, 1.7x10> for acetic acid, and 1.3x10> for propionic acid (Martell and Smith, 1977). For the data set where carboxylate ion concentrations were determined, we calculated the dissociated fraction by assuming equilibrium at the fogwater pH; we then added the resulting anion equivalents (plus bicarbonate with Poo2™ 340 ppm) to equation (1). Before inclusion of weak acids, the ionic balances for that data set were 1.03 + 0.09 at McKittrick (n=25) and 0.49 + 0.15 at Visalia (n=6); after inclusion of weak acids, the ionic balances were 1.06 + 0.11 at McKittrick and 0.75 + 0.15 at Visalia. The McKittrick samples have little alkalinity, and the tonic balance remained close to unity. At Visalia, on the other hand, ionic balances were greatly 195 improved. There was still an anion deficiency at Visalia, likely due to undetermined weak acids or to an underestimate of HCO, (equilibrium may not be reached). 2. Liquid water content We determined the liquid water content in fogs from the collection rate of the Caltech rotating arm collector, with an empirical correction factor of 60%. This method has been found to provide reasonable agreement with Hi-Vol filter and laser transmissometer measurements (Jacob et al., 1984c; Hering and Blumenthal, 1984; Waldman, 1985). Further, this method provides measurements which coincide in time and space with our chemical characterization of fogwater. The correction factor of 602 is consistent with the lower size cut of 20 im for the rotating arm collector, in view of typical fog droplet size distributions (Waldman, 1985). At the present time, there is no standard method for ground-based measurements of liquid water content; Hering and Blumenthal (1984) have found that biases between different methods are usually systematic; i.e., there is a good correlation between liquid water contents obtained by different methods. Although the absolute error on our liquid water content estimates cannot be ascertained, we are satisfied that the collection rate of the rotating arm collector adequately follows variations in liquid water content over the course of the fog. 3. Aerosol, HNO.(g), and NH.,(g) measurements 196 There are two major errors incurred in the filter method used in this study: (1) N(V) and N(-III) artifacts, (ii) random errors from the sampling process. We will address each in order. Stelson and Seinfeld (1982) have shown that an increase in temperature at constant dew point during sampling may volatilize ammonium nitrate collected on Teflon filters and result in artifact HNO,(g) and NH,(g). Further, absorption of gaseous nitric acid on the Teflon filter may result in artifact aerosol nitrate (Spicer and Schumacher, 1979; Appel, 1980). In an intercomparison study of gaseous nitric acid measurement methods, Spicer et al. (1982) found good agreement between the dual filter method (used here) and other methods. Further, Spicer et al. (1982) found that the dual filter method was accurate in measuring total N(V). The study of Spicer et al. (1982) was conducted under hot, dry conditions. Potential biases are different under the cool, humid conditions found in the San Joaquin Valley. Nighttime samples (0000-0400 PST) were probably unaffected by volatilization because temperatures during the sampling period either remained constant or decreased, During the day, temperature changes were usually small because of the overcast conditions; temperatures recorded hourly at Bakersfield between 1200 and 1600 PST increased on only 5 of the 15 sampling days, and never increased by more than 1°C except on 14 January (when a 3°c increase was observed). Still, an increase of 1°C in temperature at constant dew point may increase the dissociation constant K = Puno3* nH3 by a factor of 2 under high humidity conditions (Stelson and Seinfeld, 1982). The formation of aerosol 197 ammonium nitrate is strongly favored thermodynamically, and aerosol concentrations could not have been significantly affected by volatilization; on the other hand, high relative errors may have occurred on the determination of the gas present at the lowest concentration. That concentration was often near or below the detection limit, in which case the error was inconsequential. Teflon filters run in dense fogs accumulated drops of liquid water at the surface. In those particular cases, samples were dried in the open before being sealed. Nitric acid scavenged in acidic fog volatilizes during drying, leading to underestimate of total aerosol nitrate. No significant nitrate loss should occur in non-acidic fog, because in that case NO, remains in the aerosol phase as the fog dissipates. Similarly, ammonium aerosol should not volatilize in acidic fog. Volatilization of aerosol ammonium from filters collected in alkaline fog depends on the stability of the ammonium salts of weak acids. Some volatilization was found to occur (Jacob et al., 1985). There are three types of "random" errors: (1) uncertainty in the flow rate through the filter, (11) uncertainty in the efficiency of recovery by extraction, and (iii) analytical error. Because the first two sources of error affect the aerosol sample as a whole, we expect a correlation to exist between the errors on the different species. To 2- test for these errors, concentrations of 50, > NO, , and NH,” determined in duplicate for 45 pairs of filter samples "1" and "2" were collected side by side. Concentrations of Cl were also determined, but the errors on those were generally controlled by the filter blank. The relative differences on the determinations of X for the 45 pairs of duplicate samples were statistically analyzed as follows: 198 ' 2 |x,-x 27*1] DX = (A2) x, +X, . Standard deviations Spx on the determinations of 80,7", NO, and NH,” were 18.6%, 18.8%, and 19.1%, respectively. These results are comparable to those reported by Russell and Cass (1984) for similar measurements. Student "t" tests for paired data at the 5% level of significance did not show significant differences between sampling locations 1 and 2 for so,” NO, or NH,” at any site; therefore,no significant effects from the differences in backup filters were apparent. The standard deviation Spx was resolved into its component Sp associated with pump operation and filter extraction, and its component 5 AX associated with chemical analysis. The error tJ characterized byS, was assumed to be the same for so,” NO, ; and + NH, » while the errors due to analysis were assumed not to be correlated, Because Spx is small, we can write as an approximation: S =S§ "+5 . (A3) By statistical analysis of the differences between mH, NO, and sO, concentrations, we obtained Sp= 18.1%, $4 s04™ 4,42, 84 NO3™ 4.0%, and Sy NHA" 6.3% Most of the error in aerosol determination was > due to uncertainty in the amount of sample collected and recovered, and the errors on different species were thus strongly correlated. This correlation is an important point to take into account when 199 assigning error bounds to alkalinity calculations, or when predicting gas-phase concentrations from aerosol equilibrium nodels (Jacob et al., 1985). No duplicate analyses were made for Na’, KY, ca2t and Mg? Replicate analyses of standards indicate Sax of about 5% for these four ions. Duplicates for NH, (g) and HNO3(g) determinations were not collected, but the S values should be the same as for NH,” and A,X NO, , respectively. In addition to the analytical error, concentrations of all constituents were assumed to be subject to the same error Sp: This assumption implies that the variability of the flow rate through the filter is the major contributor to Sp» which seems justified since filter extraction efficiencies are better than 95% (Russell and Cass, 1984). Filter blanks for all constituents except Cl and NH, (g) were below the detection limit of the analytical methods. For a 4-hour sample, the detection limits from the analytical methods corresponded to 4 neq n> for NO3 , s0,"", HNO,(g), and cations other than NH,” and 8 neq mn? for NH,”. Because of substantial filter blanks, effective detection limits for NH, (g) and Cl were 17 neq n> and 20 neq m~, respectively. 200 FIGURE CAPTIONS Figure 1. Sampling sites in the San Joaquin Valley of California. Important emission sources (oil fields, major highways, confined feeding operations) are highlighted. The dashed line is the Kern county line, which is the northern boundary of the area referred to in the text as the “southern San Joaquin Valley” (SSJV). Altitudes in meters are given for each site; reference elevation is Lost Hills, the lowest site. Figure 2. Concentrations of S0,, NO, and CO at Kernridge (the bold lines were obtained by smoothing the data with a digital filter). Mixing heights at Kernridge and stratus cloud bases over the Bakersfield National Weather Service station. Kernridge hourly mixing height data were missing for 8 - 9 January (dotted line in the mixing height profile), and mixing heights for that period were interpolated from scattered measurements at Kernridge and cloud top data from pilot reports. Discontinuities in the mixing height profile at the altitude of Kernridge indicate a surface inversion at Kernridge. Discontinuities in the mixing height profile at 1000 m AGL indicate a mixing height in excess of 1000 m AGL. Figure 3. Flow patterns during the sampling program. The wind roses indicate the frequencies of wind directions. The wind vector is the resultant wind (calculated by averaging the hourly wind vectors). (a) Non-stagnant conditions (31-1 Jan., 11-12 Jan., 14 Jan.). (b) Stagnant conditions (2-7 Jan.). 201 Figure 4. Fogwater composition determined simultaneously at four sites, pH values are indicated on top of each data bar. Concentrations of S(IV) were not determined in the last two samples at Bakersfield. Figure 5. Aerosol and gas-phase concentrations, and atmospheric alkalinities, at the six valley sites over the period | - 8 January. (*) on top of the data bar indicates that the gas phase species was not measured. (tft) on top of the data bar indicates that volatilization from the filter in fog could have affected the aerosol Measurement (see Appendix). Figure 6. Concentrations of 50, and NO, at McKittrick. The bold lines were obtained by smoothing the data with a digital filter. Figure 7. Stirred-tank simulations of aerosol accumulation over the course of a stagnation episode. Conditions adopted are given in Table 5. Two patterns of emissions are considered: (a) alkaline atmosphere (Case I of Table 5), and (b) acidic atmosphere (Case II of Table 5). 202 T oansty 8u0}}019d0 Bujpee4 paujjuoy s* skomyb}y s0foyy mw (,-Aop 2_wy suo} GOO<) sucjssjwg 2 Og 3: (O00'OS <) suejues uoyoindog C) SOS 4410 7 sais Buyjdwos 6040 8811S Buyjdwos josquay + (pSIt) - Idoyodoya, + suajaw Ul JN0jU0D ‘S, (Wy) 3TVOS OS Ge f T O 7 203 10 re) ad SO} 2 ° T T T T t T i T 100 ———— lrwersien bow (Kernriige} “7 o—~ Stratus bose (Bokersfieid NWS) ’ 750) f 4 7 r — Loke sabeilo wm H | 300 Pia i = ! f A — MeKitinck | = Kernridge it = Bonerstied NWS ry + + + fut + Lost Hills 3i ! 2 "W 2 3 ” DECEMBER Figure 2 * (9) 204 € ean3sTy 205 THE (PST) Figure 4 206 G ean3tq | (ut | | TT “aT rt et r ‘rt roy TMT wyTty T = erent err maT | puny Seer a LTTE] TT] TY ep TT, — Hl Fanaa! j | any Hi] bey ala 1 a Hoot oe gu fh i0 ye hl ee Pre [ T My | hi ll ly I { i | amo sve] | es 207 os ee ee ee | | ee eee ee es ee ee | en ee eee ee 200r 4 150 PPD 100 40 ppb oli 3 DECEMBER JANUARY Figure 6 208 1250 | UM ils, FOG ‘ 1000 750 neq m79 S00 250 | YLILLLL EL | 1 FOG OMY CO, C 3 TIME (DAYS) Figure 7 (a) 209 CHAPTER VIII COMPARISON OF FIELD DATA WITH THERMODYNAMIC CALCULATIONS FOR THE H,SO, HNO, - NH, SYSTEM AT HIGH HUMIDITIES AND IN FOGS by Daniel J. Jacob, Jed M. Waldman, J. William Munger, and Michael R. Hoffmann Submitted to Journal of Geophysical Research (January 1985) 210 ABSTRACT - Aerosol and fogwater concentrations of 80,7" ’ NO, and NH,” and gas-phase concentrations of HNO, and NH,» were determined simultaneously in the field under cool and humid conditions. The data were compared to thermodynamic predictions for the chemical speciation of the H,S0, cad HNO, - NH, system between atmospheric phases. The observed products of HNO,(g) and NH, (g) concentrations under non-foggy conditions were usually of the same magnitude as the ammonium nitrate dissociation constants predicted by a recentlydeveloped model, but the model predictions were consistently too low. Concentrations of HNO, (g) in fog were below the detection limit under non-acidic conditions, but detectable concentrations were found under acidic conditions. Concentrations of NH,(g) observed in fog were of the same magnitude as those predicted at equilibrium with the fogwater. 211 INTRODUCTION The partitioning of the H,SO,-HNO,-NH,-H,0 system between different phases in the atmosphere has been the subject of much recent interest. Concentrated sulfuric acid droplets form by homogeneous heteromolecular nucleation under usual atmospheric conditions (Kiang ‘et al., 1973), but nitric acid remains in the gas phase at relative humidities up to about 98% (Nair et al., 1983). Ammonia is scavenged by acidic sulfate aerosols until eventual neutralization is achieved, and combines with nitric acid to form an aerosol of ammonium nitrate or mixed ammonium-nitrate-sulfate salts. Tang (1980) and Stelson and Seinfeld (1982a) studied the effect of relative humidity and droplet pH on the vapor pressures of HNO, and NH, over aqueous aerosol. Stelson and Seinfeld (1982b) determined the dependence on temperature and relative humidity of the ammonium nitrate aerosol dissociation constant Ry = 3? and later reported (Stelson and Seinfeld, Puno3*? NH 1982c) that addition of 4,50, to the mixture does not lower Ry greatly unless the H,S0,/HNO, ratio is very large. However, addition of H,S0, upsets the acid-base balance and may therefore considerably affect the individual concentrations of HNO,(g) and NH,(g)- Saxena et al. (1983) and Bassett and Seinfeld (1983) incorporated thermodynamic equilibrium calculations into comprehensive multiphase models. Bassett and Seinfeld (1984) reported that droplet curvature did not increase significantly the vapor pressures of HNO, and NH,. A number of investigators have tested the applicability of the above thermodynamic calculations to field data. Stelson and Seinfeld (1982b) and Hildemann et al. (1984) calculated the products of HNO, (g) 212 and NH,(g) concentrations from data collected under warm, usually dry conditions, and found numbers of the same magnitude as the predicted dissociation constants. Hildemann et al. (1984) reported that agreement could generally be improved by taking into account the association of NO,” with cations other than NH," Tanner (1983) found good agreement for one series of samples collected at a temperature of 14°C and relative humidities 55 - 70%, but large discrepancies in another series of samples collected at temperatures 5.5 - 8°C and relative humidities 75 - 85%. Harrison and Pio (1983) reported measurements consistent with thermodynamic predictions under cool and humid conditions. The hygroscopic aerosol is an aqueous solution at high humidities, and Table 1 gives the reactions determining the speciation of HNO, and NH, under those conditions. Ry decreases rapidly with increasing humidity above the deliquescence point (Stelson and Seinfeld, 1982a). At relative humidities in excess of 100%, fog droplets grow on activated nuclei and the considerable increase in the atmospheric liquid water content leads to enhanced heterogeneous condensation of HNO, and NH, - Even though fog droplets are generally in disequilibrium with the surrounding water vapor, the speciation of HNO, and NH, can be conveniently determined as a function of liquid water content from the reactions of Table 1. Fog droplets are sufficiently dilute solutions that simple activity correction expressions (Stumm and Morgan, 1981) can be used. One readily finds that HNO, is exclusively present as NO, at the ranges of liquid water contents and pH values found in fog; i.e., HNO,(g) is 100% scavenged under all foggy conditions. On the other hand, the speciation of NH, 213 is strongly dependent on droplet pH, liquid water content, and temperature, as shown in Figure 1. “Measurements of HNO, (g) and NH,(g) in fogs or clouds are few. Daum et al. (1984) found concentrations of both gases to be below the detection limit of 0.4 ppb in acidic stratus clouds (pH 3.2 - 4.2). Measurements in acidic fogs (pH 2 - 4) showed no detectable NH,(g), and very low concentrations of HNO, (g) (Hering and Blumenthal, 1984; Waldman, 1985). No measurements are available in non~acidic fogs, where a substantial vapor pressure of NH, is expected to persist. This paper reports on simultaneous measurements of gas-phase, aerosol, and fogwater concentrations in the H,S0,-HNO,-NH., systen, obtained over the course of a field experiment in the San Joaquin Valley of California (Jacob et al., 1985). The observed speciation of the H,S0,—-HNO,-NH., system will be compared to that predicted by thermodynamic equilibrium calculations. Concentrations of HNO, (g) and NH, (g) were determined by dual filter methods as described in Russell and Cass (1984). The reader is referred to Jacob et al. (1985) for a description of sampling sites, sampling methods, and a detailed analysis of measurement errors. 214 NON-FOGGY CONDITIONS We applied the Bassett and Seinfeld (1983) model to data collected at Bakersfield (Table 2). The Bassett and Seinfeld(1983) model determines the equilibrium composition of the atmospheric H 250,- HNO .- NH, system as a function of temperature ‘and relative humidity. It does so by minimizing the Gibbs free energy of a system composed of the gas-phase species H,S0,(g), HNO,(g), NH,(g), the 2~- aqueous-phase species H’, HSO,, S0,"", NO, , NH,”, and the solid phases NH, HSO, (s), (NH, ) 3H(S0,),(s), (NH, ),50,(s), NH,NO,(s), (NH, ) 50, ¢3NH,NO,(s), (NH,),S0,¢2NH,NO,(s). In the special case of the HNO, - NH 3 and Seinfeld (1982b), which uses the same thermodynamic data and ionic 37 H,0 system, the model reduces to the model of Stelson strength correction procedures. The advantage of the Bassett and Seinfeld model is that it takes into account the presence of a number of sulfate-containing species. The Bakersfield data set is well suited for comparison with thermodynamic models, because the ions SO , NO, , NH,", and 2- 4. 3 , typically contributed over 90% of the total ionic content of the aerosol (Jacob et al., 1985). Therefore, the effect of other ions on the thermodynamics of the system are minimized. Further, temperatures and relative humidities. remained stable over the 4-hour collection periods because of the prevalent low overcast conditions. We applied the model to describe two different partitioning modes for S07 and NO 3: (i) S07” and NO, present exclusively in different aerosol phases ("external mixture"), (ii) so, and NO. present in the same aerosol phase ("internal mixture"). Because ammonium ion was always in excess of 215 sulfate, the “external mixture" case was treated by assuming all sulfate to be present as (NH, ) 50, and conducting the equilibrium calculation for the remaining HNO, and NH, with no sulfate present. The presence in the same phase of s0,7~ and NO, increases the 3 solubility of HNO, and NH, (Stelson and Seinfeld, 1982c). 3 Temperature and dew point were measured hourly at the National Weather Service (NWS) station 8 km north of our Bakersfield sampling site, and standard deviations for temperature over the 4-hour period were calculated from the hourly temperature record; 0.5°C was added to the standard deviation to account for the sensitivity of the readings. The dew point did not change significantly during the sampling periods. The model requires as input the total concentrations of H,S0, 5 HNO,, and NH., the temperature, and the relative humidity. The 3? standard errors on the determinations of concentrations were about 20% for all species, and strongly inter-correlated (Jacob et al., 1985). To calculate the errors on the predictions from the errors on the measurements, 100 sets of input variables were generated for each data point from the observed values and standard errors for each variable; the correlations between measurement errors were accounted for with the scheme described by Jacob et al. (1985). The expected values and standard errors for the output variables were determined from the outputs for the 100 sets of input variables. Results are shown in Figure 2 (external mixture) and Figure 3 (internal mixture). The aerosol was predicted to be entirely aqueous except in a few cases (indicated on the figures). The standard errors on the predicted values of Ky were mostly 216 due to errors on the determinations of temperature and relative -humidity. They were not affected by errors on H,S0,, HNO, , and NH, concentrations in the external mixture assumption, and only very weakly so in. the internal mixture assumption. On the other hand, errors on the concentrations of the individual gases were often strongly affected by errors on the determinations of concentrations; as a result, error bars were sometimes very large. The products of partial pressures observed in the field were usually of the same magnitude as the values of Ry predicted by the model, but the model was consistently too low. Predictions from the “external mixture” assumption were closer than those from the “internal mixture” assumption. The model appeared to give better predictions for individual gases than for Ky but this is deceiving. First, the errors on the predicted gas concentrations were large; second, Ry was often so small that one of the gases was almost entirely depleted, and the remaining gas was then simply present at its concentration in excess of the neutralized aerosol. Stelson and Seinfeld (1982b) also found that the products of vapor pressures over aqueous aerosol were systematically underpredicted by the model, by factors similar to the ones we observed. Harrison and Pio (1983) did not find such a systematic trend, although their determinations may have been subject to large errors because of the considerable fluctuations in temperatures and relative humidities over the course of the sampling periods. Bacause of the many complicated processes occurring in the atmosphere, it is difficult to assess the significance of the relatively small model _ underpredictions. Laboratory data of vapor pressures over 217 concentrated ammonium nitrate solutions would provide a check on the accuracy of the thermodynamic calculations, but are not available at this time. - Systematic experimental biases could not satisfactorily explain the larger discrepancies between our measurements and the predicted values of Ky A systematically higher temperature at our site (located in downtown Bakersfield, pop.122,000) than at the NWS station (located on the outskirts) could occur because of the “heat island” effect; however, a few temperature measurements taken at our sampling site were within 1°C of those taken by NWS. Measurements by the NWS confirm that temperatures and relative humidities at the NWS station are the same as in downtown Bakersfield (D. Gudgel, Bakersfield NWS, private communication). In comparison, an increase of Ky by a factor of 5 would require a difference in temperature of about 5°C at constant dew point. Such a discrepancy is not realistic. The dual filter method is known to be subject to positive interferences i.e., concentrations of HNO,(g) and NH,(g) may be overestimated. Volatilization of NH NO, from the Teflon pre-filter leads to artifact HNO,(g) and NH,(g), but in our case this problem is minimized because temperatures and relative humidities remained stable over the course of the sampling periods (Table 2). Another source of positive interference is the displacement of NO, or NH,” by non-volatile material collected on the filter, e.g., acid sulfates or alkaline carbonates. However, if such displacement reactions occurred on the filter it would mean that the atmosphere itself was not at equilibrium; in that case, HNO.(g) and NH, (g) concentrations could have differed substantially from their equilibrium values. 218 Indeed, one cannot exclude the possibility that the atmosphere ' was out of equilibrium with respect to HNO, (8) and NH,(8) + To illustrate this possibility, we assume a monodisperse aerosol (particle radius a, number concentration N), and calculate an upper limit of the equilibration time T by considering the rate at which a gas of very low vapor pressure is scavenged by the aerosol. From the equations of Fuchs and Sutugin (1971), we find: - be eeBHOeTE! | 4-2) -1 30 1 + Kn T= 172 (1) 4 2 BRT where Kn is the Knudsen number, © is the accanmodation coefficient, and M is the molecular weight. Depending on the value of the various parameters, T may be rather large. For example, with a = 0.1 um (Rn 1), N= 10' em, « = 10 , and M = 63 g (nitric acid), we find t = 17 minutes. 219 FOGGY CONDITIONS Samples of fogwater, aerosol, HNO,(g) and NH,(g) were collected concurrently on 10 occasions. Aerosol, HNO,(g), and NH, (3) were simultaneously sampled, over intervals of about 4 hours; fogwater samples were collected over shorter intervals, covering in general all of the corresponding aerosol sampling period. The fogwater data are given in Table 3a. Fogwater was acidic at only one of the four sites (McKittrick). High alkalinities were found at the Visalia site (Jacob et al., 1985). Concentrations of NH, (8) at equilibrium with the fogwater were calculated from the overall reaction (Table 1): NH,(g) + HY(aq) =S5 wH,*(aq) (2) where KA » _ &3Ky Kk’ = 7 (3) 1 (NH, 7] CMt ~My yy (NH, (g)) = RIK’ = exp( R CF - 308 »? . (4) a7] and (a’] are in ueq it, and (NH,(g)) is in ueq m. The standard errors on (NH,(8)) were determined from standard errors of + 1°C on temperature and 15% on the determinations of [NH,*] and [H’] (Jacob et al., 1985). Time-weighted averages of predicted NH, (8) 220 concentrations and fogwater loadings are compared to the aerosol, HNO, (8), and NH, (8) data in Table 3b. Total aerosol concentrations of NO, and NH,” in fog may be underestimated under acidic and alkaline conditions, respectively, because the aerosol filters that were run in fog were allowed to dry 2- 4 > and fogwater was found to contain between 10 % and 60 % of total aerosol so,” . At Visalia, NH,” was in excess of NO, and 80,77 in the fog but not on the filter; this suggests evaporation of NH, (g) in the open before being sealed. This was not a problem for SO from alkaline fog droplets drying on the filter. Alkalinity present in the Visalia fogwater (in our case, mostly as ammonium carboxylate and bicarbonate salts) would thus return to the gas phase upon fog dissipation, explaining the absence of aerosol alkalinity observed under non-foggy conditions (Jacob et al., 1985). The ratio of aerosol/fogwater NO, concentrations was much higher at McKittrick than at other sites, and this similarly suggests volatilization of HNO, as acid fog dries on the filter. Concentrations of HNO,.(g) in fogs were at or below the detection limit of 4 neq n> under non-acidic conditions (Bakersfield, Buttonwillow, Visalia). Fog was present for less than half of the time that the filter sample was collected at Bakersfield on 31 December, and the non-zero HNO,(8) concentrations on that night can be satisfactorily explained from the average relative humidity of 96% observed during sample collection. Concentrations of HNO..(g) in the range 5 -— 11 neq n> were observed under foggy acidic conditions (McKittrick). These concentrations were much lower than those typically measured at that site under non-foggy conditions (Jacob et 221 al., 1985), but still were significantly different from zero. Although HNO, (g) is totally scavenged at equilibrium with fog under acidic conditions, equilibrium may not be fully reached because of the slow rate of gas-phase diffusion to droplets (Chameides, 1984). Further, air parcels undersaturated with respect to water vapor are likely to exist within the fog (Gerber, 1981); in such acidic air parcels, HNO,(g) would be present at equilibrium. Further, if fogwater collected on the aerosol filter was evaporated by undersaturated air during sampling, HNO, would have volatilized and caused artifact HNO,(g)- Concentrations of NH, (g) at McKittrick were below the detection limit at all times, whereas substantial concentrations were observed at other sites. This is consistent with the low fogwater pH at McKittrick, and the high pH at other sites. Although there were discrepancies between observations and equilibrium calculations, the magnitude of NH, (g) concentrations was correctly predicted. In comparison, several investigators of rainwater chemistry (e.g., Ayers et al., 1984) have reported [NH 7] rainwater concentrations that differed by orders of magnitude from those expected from equilibrium with ground-level NH, (8). As pointed out by Ayers et al.(1984), a number of factors could prevent attainment of equilibrium between rainwater WH,” and ground-level NH,(g)« Fogs are more likely than ‘rain to be near equilibrium with the surrounding gases because of longer contact times and smaller drop sizes. Due to the low solubility of NH, at pH > 5, equilibration of NH, with the fog 3 3 droplets should proceed within seconds (Chameides, 1984). The discrepancies between observed and equilibrium NH, (8) concentrations 222 can be explained by non-uniformities in droplet pH, fluctuations in liquid water content, or formation of aqueous-phase ammonia-complexes which complicate the thermodynamics of the system (Hales and Drewes, 1979; Nielsen et al., 1979). 223 CONCLUSION Aerosol and fogwater concentrations of SO 2~ NO “, and NH,”, 4 ” 3 and gas-phase concentrations of HNO, and NH s were determined simultaneously in the field under cool and humid conditions. Data collected under both non-foggy and foggy conditions were compared to predictions from thermodynamic calculations. Measurements under non-foggy, cool and humid conditions (temperatures 3 - 13°c, relative humidities 60% ~ 100%) were compared to predictions from the Bassett and Seinfeld (1983) model. The predicted ammonium nitrate dissociation constants were usually of the same magnitude as the observed products of HNO,.(g) and NH, (g) concentrations; however, the predictions were consistently too low. In fog, concentrations of HNO,(g) were at or below the detection limit of 4 neq nn” under non-acidic conditions, and were very low (although generally non-zero) under acidic conditions. The observation of detectable HNO, (g) in acidic fog was attributed either to the presence of undersaturated air parcels within the fog or to the slow rate of HNO, (g) diffusion to the fog droplets. No detectable NH, (8) was found under foggy acidic conditions, but substantial amounts were found under foggy non-acidic conditions. The observed NH,(g) concentrations were of the same magnitude as those predicted at equilibrium with fogwater, but significant discrepancies were noted. These discrepancies can be explained by non-uniformities in droplet pH, fluctuations of liquid water content, and formation of ammonia-complexes. 224 ACKNOWLEDGEMENTS | A.G. Russell and G.R. Cass (Caltech) are thanked for many helpful discussions. This work was funded by the California Air Resources Board (A2-048-32). REFERENCES Ayers,G.P., Gras,J.L., Adriaansen,A. and Gillett,R.W. 1984. Solubility of ammonia in rainwater. Tellus- 36B,85-91. Bassett,M. and Seinfeld,J.H. 1983. Atmospheric equilibrium model of sulfate and nitrate aerosols. Atmos.Environ. 17,2237-2252. Bassett ,M.E. and Seinfeld,J.H. 1984. Atmospheric equilibrium model for sulfate and nitrate aerosols-II. Particle size analysis. Atmos.Environ. 18,1163-1170. Chameides, W.L. 1984. The photochemistry of a remote marine stratiform cloud. J.Geophys.Res. 89,4739-4755. See also Discussion by D.J. Jacob, same journal (in press). Daum,P.H., Schwartz,S.E. and Newman,L. 1984. Acidic and related constituents in liquid water stratiform clouds. J.Geophys.Res. 89, 1447-1458. Fuchs,N.A., and Sutugin, A.G. 1971. High-dispersed aerosols, in 225 International Reviews of Aerosol Physics and Chemistry, vol.2, G.M. -Hidy and J.R. Brock ed., p.1-60, Pergamon, New York. Gerber ,H.E. 1981. Microstructure of a radiation fog. J.Atmos.Scl. Hales,J.M. and Drewes,D.R. 1979. Solubility of ammonia in water at low concentrations. Atmos.Environ. 13, 1133-1147. Harrison, R.M. and Pio, C.A. 1983. An investigation of the atmospheric HNO,-NH.-NH NO, equilibrium relationship in a cool, humid climate. Tellus, 35B,155-159. Hering, S.V. and Blumenthal, D.L. 1984. Fog sampler intercomparison study: final report to the Coordinating Research Council. Available from Coordinating Research Council, 219 Perimeter Center Parkway, Atlanta, GA. Hildemann,L.M., Russell,A.G. and Cass,G.R. 1984. Ammonia and nitric acid concentrations in equilibrium with atmospheric aerosol: experiment vs. theory. Atmos.Environ. 18, 1737-1750. Jacob, D.J., Wang, R.-F. T., and Flagan, R.C. 1984. Fogwater collector design and characterization. Environ.Sci.Technol.. 18,827-833. Jacob, D.J., Munger, J.W., Waldman, J.M., and Hoffmann, M.R. 1985. 226 The H,50, - HNO, - NH, system at high humidities and in fogs: a field study of wintertime stagnation in the San Joaquin Valley of. California. Submitted to J.Geophys.Res. Kiang, C.S., Mohnen, V.A., Bricard, J., and Vigla, D. 1973. Heteromolecular nucleation theory applied to gas-to-particle conversion. Atmos.Environ. 7,1279-1283. Nair, P.V.N., Joshi, P.V., Mishra, U.C., and Vohra, K.G. 1983. Growth of aqueous solution droplets of HNO, and HCl in the atmosphere. J.Atmos.Sei. 40,107-115. Nielsen,A.T., Moore,D.W., Ogan,M.D., and Atkins,R.L. 1979. Structure and chemistry of the aldehyde ammonias. 3. Formaldehyde-ammonia reaction. 1,3,5 - hexahydrotriazine. J.Org.Chem. 44, 1678-1684. Russell, A.G. and Cass, G.R. 1984, Acquisition of regional air quality model validation data for nitrate, sulfate, ammonium ion and their precursors. Atmos.Environ. 18, 1815-1827. Saxena,P., Seigneur,S. and Peterson,T.W. 1983. Modeling of multiphase atmospheric aerosols. Atmos.Environ. 17,1315-1329. Schwartz,S.E. and White,W.H. 1981. Solubility equilibria of the nitrogen oxides and oxyacids in dilute aqueous solution. Adv.Env.Eng.Sci. 4,1-45. 227 Smith,R.M. and Martell1,A.E. 1976. Critical Stability Constants, vol. 4, Plenum, New York. Stelson, A.W., and Seinfeld, J.H. 1982a. Relative humidity and pH dependence of the vapor pressure of ammonium nitrate~nitric acid solutions at 25°C, Atmos.Environ. 16, 993-1000. Stelson,A.W. and Seinfeld,J.H. 1982b. Relative humidity and temperature dependence of the ammonium nitrate dissociation constant. Atmos.Environ. 16,983-992. Stelson,A.W. and Seinfeld,J.H. 1982c. Thermodynamic prediction of the water activity, NH,NO, dissociation constant, density and refractive index for the system NH,NO,-(NH SO,-H,O at 25°C. Atmos .Environ. 4) 280,-Hy 16,2507-2514. ‘Stumm,W. and Morgan,J.J. 1981. Aquatic Chemistry, gnd ed., Wiley Interscience, New York. Tang, I.N. 1980. On the equilibrium partial pressures of nitric acid and ammonia in the atmosphere. Atmos.Environ. 14,819-828. Tanner, R.L. 1983. An ambient experimental study of phase equilibrium in the atmospheric system: aerosol HY, NH,” so," NO, , NH, (8), HNO, (g). Atmos.Environ. 17,2935-2942,. Waldman, J.M. 1985. Contribution of fog to the deposition flux of acidity. Ph.D. thesis, California Institute of Technology,Pasadena,CA. 228 Table 1. Dissociation and vapor pressure equilibria of HNO, and NH. Reaction Koos ABbge reference Mor M atm 2 keal mle! 1. HND,(g) = BNO, (aq) 2.1x10° - 4 White (1981) 2, HNO,(aq) = NO, +H 15 3. NL(g) = NH,(aq) 74 ~6.80 Hales and “3 "3 Drewes (1979) 4. NE,(aq) + H,0(1) = mat +O L.7x10> 0.9 Smith and Martell (1976) 5. H,0(1) = H+ af 1.0x10 24 13.35 Smith and Martell (1976) 229 Table 2. Bakersfield data for aerosol and gaseous H,S0, ~ HNO, - NH, species. Date Time) ma” No, 0,7" Nig) BO) T pe) neq n> °c % 31 Dec. 0000-0345 573.351 351 19 17 7.80.9 96 1200-1610 669 467 297 158 4 10.810.6 93 lJan. ~0215-0815 242 170 289 212 <4 10.440.6 96 1200-1615 275 161 = «161 68 35 13.0.7 76 2Ja. 0000-0415 273 «(168-109 67 4 7.241.189 1200-1620 M4 = =201 = 124 107 25 11.640.7 74 3 Jan. 0000-0415 545 370 226 483 <4 9,140.8 87 1200-1615 769445 AS 113 ll 9.240.6 81 4 Jan. 1200-1700 1057 427'—S «660 55 24 9.541.3 80 5 Jan. 0100-0510 11441 442 731 63 6 6.90.8 95 1200-1615 923 %2 596 60 19 8.60.6 85 6 Jan. 0000-0415 11449 388855 47 7 7.40.6 95 8 Jan. 0000-0405 272 «142—Ss«d118 52 6 6.810.6 95 9 Jan. 1420-1825 260 85 = 219 139 8 7.50.6 86 10 Jan. 0000-0415 621 117 570 60 46 6.20.6 89 1200-1615 533 «152.417 48 14 6.940.8 85 11 Jan. 1200-1615 44 10 78 65 46 9.040.771 12 Jan. 0000-0415 ¥%5 204 149 157 <4 3.00.9 91 1200-1615 331 244 = 169 3% 26 11.41.0 66 13 Jan. 0000-0400 486 283 407 208 <4 4,640.7 96 0700-1115 729 34975 122 6 5.90.8 91 1200-1615 855 437 507 240 14 9.91.1 73 1800-2210 692 402 306 161 10 6.440.8 85 14 Jan. 0050-0410 567 423 229 180 5 4,941.2 89 0735-1130 641 423214 470 6 6.441.0 85 1200-1615 318 263 129 118 43 10.741.5 61 (a) local time (PST) (b) calculated from the expected temperature value amd the dew point. 230 Table 3a. Chemical composition of fogwater samples used in comparison. Ao Site fakersfield 31 Dec, 1983 T=8%C 13 Jan, 1984 T=4C McKittrick 5 Jan. 1984 T=5 C McKittrick 6 Jan. 1984 T=5% McKittrick 7 Jan. 1984 T=5% McKittrick 7 Jan. 1984 T=5% MKittrick 7-8 Jan. 1984 T=5% McKittrick 10 Jan, 1984 T=5% Buttonwillow 7 Jan. 1984 T=6 °C Visalia 7 Jan. 1984 T=6C Tine 0215-0230 0805-0415 0130-0200 0200-0300 0300-0330 0005-0100 0100-0200 0200-0300 0300-0435 0035-0100 0100-0200 0000-0100 0115-0150 0200-0300 1030-1230 12530-1440 1440-1540 1540-1655 2035-0035 0035-0205 0205-0405 0010-0120 0120-0525 0040-0140 0140-0410 0410-0620 0120-0250 0250-0430 yp) gu 0.13 0.10 0.11 Ou 0.03 0.07 0.32 0.16 0.24 0.21 0.25 0.22 0.17 0.15 0.12 0.19 0.30 0.21 0.07 0.20 0.10 0.07 0.17 0.22 0.07 0.07 0.02 0.10 0.07 0.08 0.07 0.05 0.06 0.07 5.99 5.63 6.38 5.92 6.74 5.18 5.40 5.33 6.97 7.23 us Sess 2346 SS8S8GG atansw wae — ann ~B 4 + Ne 4 2510 3190 2800 5920 3300 Ka] 347 BA 357 241 294 499 333 345 13% 18 162 204 166 683 870 740 522 599 555 658 845 351 1270 969 1070 860 678 Ca 2+ 73 75 143 75 492 294 13 21 10 8 10 5 3 10 9 47 10 14 % 24 116 145 7) 27 yog Me peq 1 13 4 BY Ke ee ey WAN eH a -) 3 40 314 hm he s~wBeR wowta Puwe BEB Soo ex yvkG a8 175 lai 196 158 142 270 131 1006 331 318 198 162 4920 267 272 255 317 371 310 210 255 278 827 467 842 418 492 916 900 176 122 NAKRAN ARAN Ctl oll eel ed tt ee ee RAAAA bat pee tet pee NAAN | ell ell coe onl wv AA + pene a 88 oS ~ ele 141414 Ba WW hh (a) calculated fron smpling rate of rotating am fogmter collector, aeeusing 60% collection efficiency (Jacob et al., 1964). Cb) calaulated from equation (1). 231 °eE STQRL WIZ peyepPOTSEO ‘adeiave wy UO S119 PlepuRye ale s10ITY “PL BTqRL, UT peqoypead suo; waquscuco jo aBeraae peyysyas—auyy Aq poqeTMoTeo ‘JayenZoy yIpA ENPAqTT Abe Je suozJBAqUEOUOD oseyd-se8 a7e sasayqueued uy s1equew (q) -_ "aE eTqRy, Wory Seflonons, Pasa SET) (®) *sepoods [oearae Jao [Te Joy _ w bau y puw * TD oy. Ww 0% *“HN 403 _ w bau g ‘(3)"ONH 303 | w bou y “(3)4N 203 w bau /] agam ex pNP] WoT DeIaq *(Se61 ‘*T8 72 q ) sopads tre x03 fiz qoge asan s10170 ques *quaquod’ Jaye PRMbyT aya 4q suopABzquaOUKD AsyEMBoy BupATdTITMM Aq peupeqqo alam I;e JO GW Jed BRUPpeOT asyantog € > 265 oY €L o> 81 el 68 Josor78 cURL { (D>) = (06#55€) 90°C aI I TO 9° OS 470 90° By SEEO-S#TO BTTESTA 9 ) VAL td 0c > 9 2 «2 osor38 “usr (D>) (6) L0°O «xs Te z eo Zz 69 «6 Z70s«dS BOY 00S0-00TO «OT TH#H033Ng ot Ww ez LE 6z > 61 we 6 yosos3e ouBr OT (I>) (ID) g0°0 «8 ral 1 z0 OZ fe 80 7 Boy SV0-SH00 PIR S dl 0z1 8 ze y> 9 v1 8 Tosol37 ouRr g (D) (TH) 90°O $2 9 Oe A A i OT?0-0000 [PFI NW iat LD cozy 74 > Of il 8 Josor3e cuey f (ID) (D) ZO BL 0 T 1 ol c8 OO WNC*—s« BOT G191-0021 «PHAN 6 Ww 6ce UT & o> ZI el ST yosorse cuer f (I>) (ID) slo | ay 6 r4 70 =«S 0O¢ WwW Ith = Roy ST70-0000 0 PR PON Z LD 92 = o> > OF gz Ol osaTaa *uEr 9 (I>) (1D) 6T'0 66 9% r4 €0 2 oo 62 $I'y = Boy OT70-0000 (P39 RPN > iD we 89029 > » @ isto yosorse suRr ¢ (D) (PD) 0z"0 «€S hy z €70 2 uw ot 0°, Boy STt0-S00 OP HIIRON 9 902 op = eBet—~<i«iR Ss 6 %F 4 Joso1se *uRt €1 (D>) (9840/6) 80°O OSZ = SY aI z a 082 9 Wy By 0070-0000 «=: PTe1J8197eg at ot Ise 9 9 a 8 64S WN Joso738 "oa IE (>) (61469) 2T°O)—s EST ey S z 6 et 7e°S Boy 000-000 PLeyseaeyEYy bou 3 So ba - - — €, ¥, £ ¥, au q@@in wt os fa Dp ym yo JN amy, aaTs -eaep (8)%n pur *(3)%ona ‘Tosoras uaz equp zo7uMHos Jo meTIBdWD * oF BTTEL 232 FIGURE CAPTIONS. Figure 1. Fraction of gaseous ammonia scavenged by fogwater. All fog droplets are assumed to be at the same pH. Liquid water contents of 0.05 g n> (——) and 0.5 g n> (--~) are considered. Figure 2. Comparison of Bakersfield field data to the Bassett and Seinfeld (1983) equilibrium model: products of concentrations (HNO,(g))x(NH,(g)), and individual concentrations of HNO,(g) and NH,(g). The aerosol is predicted by the model to be an aqueous solution in all cases, except when indicated by (*) (mixed aqueous/solid) or (tf) (solid). Two limiting cases are considered: (a) sulfate and nitrate present exclusively in different aerosol phases (external mixture assumption), and (b) sulfate and nitrate present .exclusively in the same aerosol phases (internal mixture assumption). 233 10 | | { | 8 & 8 ¢ = —_— (%) 903 AS GAONSAVOS NOILOVYSs oO Figure 1 234 “a & -~ ‘) 2 T T T T 1.8 ie i re -——=—« ————— eee oy i 4 = ——. 2 r c»— cs 4. x 4 Ll s oc a [ Fy 7 co oe! a ex | oa EEE fren Ft ———— mI ee 4 “" 7] oEH oe cc cay ————s do = am 4 § cHe« J | i co we a i we J COM Lf cos a E § i 4 , d £3 fF uC ee 4 }- on aaa ca es ft cs i eo it 24 wt =} oe L 4 4 4 aoe t —J =< io. ————— oa 4 on == | coo [a =] co -_— , Q Ls T T T Li Tr —S55= —r = a a ——— | | cot =—F} coy co laa a = [ a en —-—« g =a sc—~C<‘ OD = fr a) 4 a — SF z oa” -— ca ee ony S44 — ay a co —* 4 cor J coe ca —t 4 2 a ry | E 8 7 ” “ sit 2c we Oi ———— — eS on ei nen morn oe a N eH wee co we, = ——f 4 a> . . co ey oe Fenn el ene 4 iz. LL SSS — } ce 2. 5 i. ce om + cr —_ 4) | ce i = | Gg _— = w—t=$ —— F4 L r£—} . ie ore I 8 g & ° 2 2 2 * ? i z € 2b Fi 236 CHAPTER IX FUTURE RESEARCH 237 The results of the research conducted so far have made the “acid ‘fog” problem a serious environmental issue in California. Several studies on the environmental effects of acid fog have already been published, and more are under way. Eventually, regulatory action by air quality control agencies may be considered. Still, more research is needed to assess the extent of the “acid fog” problem, and to understand the means by which the acidity is produced. The following are a few questions which could be addressed in the next few years of research at Caltech. 1. Provide air quality control agencies with a fogwater collector suitable for routine monitoring. A screen collector has been developed for this purpose (Chapter 2) but has not yet been automated or experimentally calibrated. Another promising scheme for sampling fog droplets is by virtual impaction of fog droplets in a cavity. Such a method would have three major advantages over the screen collector: 1. Collection surfaces are easy to clean. 2. Droplets collect in a quiescent environment. 3. Fogs can be collected under freezing conditions. On the other hand, two drawbacks of the jet impactor are: 1. Anisokinetic sampling biases. 2. Impaction on the inlet walls. 2. Fog and aerosol sampling in the Santa Barbara Channel. The Santa Barbara Channel has been found to contain considerable oil and gas 238 reserves, and the federal government is encouraging the development of these resources by proposing lease sales in the Outer Continental ‘Shelf. This would considerably increase offshore SO, and NO. 2 emissions in an air basin where dense fogs are frequent and the acid-neutralizing capacity is low. Tracer studies have already been conducted, and have shown that offshore emissions affect the coastline. Further meteorological studies and tracer releases are planned by the California Air Resources Board. The fogwater acidity in the Santa Barbara Channel should be documented in detail, and predictions made of the increase in acidity to be expected from increased offshore activity. 3. Modeling of the smog-fog-smog cycle. One-dimensional models of fog microphysics have been proposed which predict droplet activation, growth, and evaporation over the course of a fog event. These models also predict the amount of liquid water deposited to the ground. It would be of great interest to interface such a model with chemistry packages for fogwater chemistry and the chemistry of non-activated aqueous aerosol. Considerable information would be gained regarding the effect of fog formation on the aerosol composition, and the importance of acid deposition during fogs. 4. Determine sulfate production rates in fogwater and aerosol. Such a project could be conducted in the southern San Joaquin Valley under stagnant conditions, with the help of tracer releases. Emissions of SO, in the Valley are concentrated in two small areas, and daily emission rates are available from the oil companies responsible for 239 these emissions. Tracer releases from both of these areas, followed by tracking of this tracer within the valley, would allow an estimate of sulfur transport. A the same time, a network of sites would be operated to monitor surface winds, mixing heights, 50, concentrations, aerosol and fogwater sulfate, and deposited sulfur. By a mass balance on sulfur it would be possible to determine SO. conversion rates. 5. Develop an engineering model for predicting fogwater mineral acidity in the San Joaquin Valley. At the present time, the San Joaquin Valley does not have a systematic “acid fog” problem. However, the decline of the cattle industry means that a decrease in NH, emissions is to be expected in the next few years. Since there is no foreseeable decrease of oil production activity, Kern county may soon be a “hot spot” for acid fog. The California Air Resources Board is currently working on a NO /0,/hydrocarbons air quality model for the Valley. In parallel, a detailed inventory of ammonia emissions should be prepared. From this inventory, determinations of sulfate production rates, and results from Chapter 7, an “acid fog" simulation package can be written to interface with the Air Resources Board model. Chapter 7 shows that such a simple model can be expected to reliably predict inorganic acidities in fog. 240 APPENDIX A INVENTORY OF AMMONIA EMISSIONS IN THE SOUTHERN SAN JOAQUIN VALLEY OF CALIFORNIA 241 L°6 TVLOL o°Ee iA L°eel OSY pFAbyT $°9 or 6°€S 70€ Aap € ke suo 4 sso 9 ke oatTtdde suo] ke ayttdde saz qaey suo I- P “HN (Z) IN Z@ (S)T- P PPT N (S)I- P PPT TTF J asn JeZETFazaeg Z°Lt IVLOL T°O> T (4S5°0) OF puelT ysa10j y°€e T (%E°€) O€7‘E pueyTesuer 9°Z 9°E (Z€°8) 099 spaeyoi0 8° IT 9°€ (42°) OLDSE pueTdors T°O c¥°0 (42°47) O€E pue,T dn-jzT Enq ke t suo. w aad. ke t 3 eoae , APP “HN (z)7m P/“HN 34 (y)&g™ STFOS €°94 ‘IVIOL T°Q> 200°0 000 ‘04> shayin} pue sueyoTyD Z°0 9€0°0 0009 s30yq Z°T 220°O 000‘ ZZ deasys Z°E (¢)80°0 000 ‘Ov S9ATBO Z° Iv $7°O 000‘47ZT SMO00 pure s1990q8 9T378D fe t suo eay aod se t 3 qJequnu I- P “HN q (z) peoy P/"HN 24 (1) q YIOYSIATT] Wy O€6L Peay “(IIA zaqdeyp ‘] aanByzq 3es) auz{T AqunoOd ulsy 3yz JO yRnOs AvTTBA Vy JO Javed yey. se pSutjep St ,APTI@A U;NbeoG UBS UTaYINOS,, au] 242 t 1-4ep “HN 8u0Q 6°// AQTIVA NIQOVOL NVS NAXHLNOS AHL NI SNOISSIWH €HN ‘IVLOL c°T TVLOL €°0 °T 1 OTX6E*Z 006‘6€ se3 [einqeu €°0 B°Z Ot*Z0°T 699‘LT TFo T So Fisut yoy c°0 7°T yor evv"€ oo0E ‘ZS ses [eanjzeu 7°O 8°z ppor%se'T 000 ‘ZZ [Fo ¢ € $10}e10Ues Weeds ke suoq 3 se sep slesase ;-4eP “HN (z) 948 gol/ HN34 (@)1- P ALd -APP ST q seoinos Areuoytjeqjs — uo; ysnquos Teng L°% ‘IVLOL 9°0 c-OTX2°% (1000 ‘€0€ $389 6° c -0TX6°9 (10005 €8¢ $30q z°O 01*8°9 (9)£69 Sze (uof _eizdsiad) suewny Aep CaN su0j Aep peroy/suotssyue t zequnu I- | (Z)I- 4 H q oF 3seuN0g 243 °908S6 VO ‘OjUeMBADeg *9CzI XOg “O'd Saanq[NoTAZy pue poog jo JUuaeMJABd|aq BTUAOZTTeQ ‘aozArag Buy _aoday yoo ysaaF] pue doap SFUIOFTTED “CRG YJsOdor-Tenuue :yOJssATT] SFUAOFTIEQO “HEEl °N ‘UND puB °-Ie¢ “ND SAayVON, ‘*y°y ‘OBeag SW “SZII6 WO ‘Buepeseg ‘ABo0TOUYyIa]L Jo a4nQFIAsuUl ByuloszTeg ‘ArojReIoqey] ARTLTeNH Te\uamuoryAug ‘9-7g JYaodel aT]Ty usdQg ‘*evsIe UBqIN UB UT sisydsowje ay} 02 sUucTSsywe eTUCMME JO UTSZIO aul “ZRSI CM’ SUSPTTL pue *H ‘uosiaqeg **s ‘qyaBeyy ‘°y°y ‘e8BD “CZI1I6 VO ‘Buepeseg ‘ABoTOUYDSL JO aQnay sul ByUAOFT[eQ ‘szpsoauy ‘a°yd ‘“seTeszuy so] 02 saTdwexa y{yM JuoWsaUeseueM ARZTTeNb Ife azez[Ns JOJ spoyyW “*g/6l “UD ‘88RD "71866 VO ‘OqUsMeIDeS ‘s2naq YOOJSeAT] pue ‘tezzT} wag ‘paeq “FREE YoIeW — AIeHAgey - Arenuer *jiodai aseuu0j - sleyiejem BuyZzz[}3eg *ygG] ‘Vang [NoOyAZy pue poo Jo Quowzavdeg epTUAOZT TED *ga0ualazyoy °(8261)88e9 Wory S108] UOTSAaAUOD (¢g) *(O86I‘snsuap *s°n *YI8) ASS 243 UF uOTQeTNdod uBWNY AQUNOD UIey ayy Jo UOTIBAZ |aYyQ Aq ps qdaazIOD sazaasquNny *[oOrQUOD [eMyUY JO QUsujaBdap AQuNOD urey Vy} Woz UOT ABOTUNWMMOD TRUOsIad Aq pauyzEyqo sTe 07 AquNOD (/) (0861) snsueo uozze{ndod *s*pn (9) °(7861) eanq[noyasy puw poog Jo Quawjziedeg eyuI0Z;TeD (¢) *°(0861) dew 2zas0d puet/esn pue{, AvAing [eo;Z0TOe) saqejg paqyyUuQ Woy peinseseu (4) *suoTssyWe 190]38s pue AOD WoIy WYSzam Apog Aq paqoazi0d (¢) *(Z861)°T® 39 BsBD Worz (7) *“avFAI9g Buy jaodsay yoo ysaaTy] pue doig eyus0sTT BQ “ung TFen 4q uoF}e9TUNMMOD [euosaed WOAj payeUy {ss saaATeo Jo JaqUNN “eANATNoOTAsy Jo quawjaedag Aqunod ursy ‘suBsg UuBAY wo1lj Uo} {BO FUNMWOD [Buosied sq pauzeyqo ASS UF UOTFIDBAG °(H¥8GI)* 1S 38 AOZVIVI_ Wo1z sTeqoQ AQuNOD (7) 244 APPENDIX B CALCULATION OF SULFATE PRODUCTION IN FOGWATER FROM SUCCESSIVE SAMPLES COLLECTED AT ONE SITE » 245 During the period 3 - 5 January 1984, a stratus cloud filled 1/4 to 1/2: of the boundary mixing layer in the southern San Joaquin: Valley. Over that period, a sulfate production rate of the order of 2 ys n7! was observed in the overcast boundary layer (Chapter VII). This sulfate production could possibly have been due entirely to in-cloud SC(IV) oxidation followed by downward transport of the produced sulfate. Alternatively, some of the sulfate could have been produced in the non-activated aerosol below the cloud. The enhanced 50, scavenging in fogs (Chapter Vilsuggest that fog (or cloud) droplets could be a favorable environment for sulfate production. However, SO, scavenged in the fog may be stabilized as S(IV) by adduct formation (Chapter V ) or removed from the atmosphere before undergoing oxidation. Ideally, in-~fog sulfate production rates should be determined by following the evolution of SO, and fogwater sulfate concentrations in 2 a given air parcel. Dynamics of droplet growth and evaporation make this very difficult to achieve. The unknown errors in liquid water content measurement methods further complicate the estimate of in-fog sulfate production. In the San Joaquin Valley 1983-1984 study, we investigated sulfate production in fogwater by comparison of successive fog samples collected over the course of an event at a given site. The successive samples were identified by the subscripts 1 and 2, t, and ty being the corresponding midpoints of the sampling 1 intervals. Samples were typically collected successively over intervals of 1 - 2 hours. The concentrations in samples 1 and 2 were assumed to be those of the fogwater at times ty and to» respectively. 246 Three processes could affect the sulfate concentration measured in fogwater: (i)change in the liquid water content, (11) advection or deposition, and (iii) SC(IV) oxidation reactions. We use the same notation as in Chapter 8 to describe concentrations in fogwater and concentrations in air. The time dependence of the sulfate fogwater loading (S(VI)), can be written: d(S(VI)), ( d(S(VI)), d(S(VI)), dt * dt ox + ¢ dt dep ,adv . (B1) The deposition/advection term can be estimated by following the change in fogwater concentration for a component T assumed unreactive and with the same origin and behavior in the atmosphere as S(VI). Over a short time interval: (S(VI)), Assuming that (T), varies linearly with time between t, and tos we 1 obtain an expression for the oxidation term: t t &(S(VI)), (Me (De, (SID) g(t) t dt ox (ty-t)) t, (T),.(t) (B3) The integral on the right-hand side is estimated from an average of the (S(VI))-/(T), ratio at times t, and ty? 247 t fees (SQVI))¢ -(SCVID)¢ 1 (SCV) g (Te » (SVD 5 (Me eet = 3 3 + 2 > - > > dt ox 2 2(T)¢ 2 2(T)¢ 4 . 1 t 1 (BA) This oxidation rate was used to derive the pseudo first-order rate constant, k (2% holy, for the conversion of S0,(g) to S(VI). By taking an average of P over the interval [t,, to]: $02 t 2 d(S(VI)) x = ——100 RT (_—_——*) at. (BS) (to-t Peo» dt Ox 1 We also obtained directly from the sulfate production rate a pseudo zero-order rate constant k' aw ny to describe an $0, conversion process limited by mass transfer of S0,. | We calculated k and k* from our complete data set, using as tracers nickel and vanadium. Nickel and vanadium are almost exclusively contributed by fuel combustion, and therefore should trace sulfur concentrations. Results are shown in Table Bl. No statistically significant conversion of S05 (g) to sulfate was observed from one sample to the next during fogs at any site. To test whether the lack of significance was due to short-term fluctuations, we examined separately a restricted data set with (t,-t,)> 6 hours. For these samples (all at McKittrick), the fluctuations for k were less, but the rates were still not significant. 248 It must be pointed out that the use of a pseudo first-order rate constant to calculate the in-fog sulfate production rate has the major fault that it ignores the dependence of the rate on the reactive species: oxidants, S0,-H,0, HSO, , and so,” S(IV) speciation can be expressed by a function of the form: The dependence of k on (B6) However, we were not able to find a significant fit of k to equation(B6). In Chapter VI, we calculated sulfate production rates in winter 1982-1983 Bakersfield fogs by a method similar to the one used here (with nitrate as tracer, and approximating the right-hand side integral in (B3) using the (S(VI)) -/(T)¢ ratio at time ty). We found rates of similar magnitude to those found here, and an average rate not significantly different from zero. It must be stressed that our failure to find significant sulfate production rates in fogwater does not constitute evidence that this production does not occur, but it does suggest that the production rate does not exceed a few percent per hour. 249 o ore ~ es = domed +i+] +14] +141 MOTEqG suOy eAUaDUOD “os) BTTeSTA JOJ pue (eIep é *uofFIonpoid ayey[Ns Jo ayer dy. SF ,y ANN ww ce ce on + 1+ cq)" *UOFSAVAUCD (8) 3 £€ u *STPAIeqUT INOY-g je peaT[OO seTdwes wo1z ATUO payeTNoTeD (q) *(qdd QT Jo IWF UoPI~eIep un; peuea Toyo Fu un} peurA Tayo Fu mn} peuea Tayo Fu un} peuea Teyoyu ADVIPI S$ OU) MOTT FMU0}INgG JOJ pazeTMOTed Jou sem y OS JO e381 Aepio—JZsizy opnasd ayy st y (#) CTTeESTA MOTT Fao yng TIAN PTeTysasyeg aqqs *ia7eMZ0J UT uoTIoNpord ayey[ns Jo saqey “Ig eTqQeL 250 APPENDIX C FOGWATER CHEMICAL COMPOSITION: CALIFORNIA COAST, 1981 - 1983 251 reese BOTS LZBETOT ‘ot OTe oe GTB es SSOW Octtt GBTT zo0o't £¥6°0 veB’o S4N/0+H £S6'°0O 096°O 9TOrT v2.40 vert t/~ Gtere ov7'o Cl ee Teves bar's 69ETE eas CUNN0d Midd NY) NIW/EW S = SLY SNTTdWUS 9 SNIWNSSY ANY DSLIS TION YSLVN904 40 SWNIOA BHL CAWMOTIVAG JT) WSaYNNSYSW MAHLIS LNILNOD YALYM CINOIT NO Waseda SI Iv 84T'°O vee?o £90'O 009'O 689'°0O StN/H 0°00 orore O' See O'06£ O°sdé bas Tte' ot Cote Tag:or OT6* Tt vB9°zE STOLE £0N, OSZ'O ¥26°0 0468°0 6TZL°O £va'o £0N/H Q*sed o°oult O°'G4et O°O96T O°S22e £0N LIE eSt°o 2680 B08'T 6EL°E G00°6 W Ove tO 070'°0 8Se'O £L8t°o L4ET°O 0zs'o A 9ZE°0 £00°O atoto BETO eSe'o 6¢9'°0O OW 6£T*O LE0°0 zvo'o 6£0°O 8cz°O OcEe"o vo YSLAW STAND 7 WYHOONOTH SS6°0 C8£°E Lee'a ST9"e 6tore £8/é/£ DaSIAay 206°O c9b'O SfS'o LeLe'o S9T*O LOTT BBS°0 9L49°O 9Te'O tté"o Ost Tats £ésg°e 060°E SSé2'o bOS/H S+N/bHN £ON/~HN bOS/VHN SOITLYY LNA TVAINGDS o's? o'ovt 0°05 O°S9S o°oort Ww Ode T'é 0°0e 0°06 oror O*bt o*ss O'té O°Se 0°0S o°ast o°r9 o°Oovt O°S9S 0°c8 4 OW a) MaLT1/0a ONIN boots g9ort eca't PELE ere 850°C vHN $ £8 NYT 80 oct'o SEare oos*ét EL49°E OOr'zTt 608°E wert 9E0°S T99°O Tass TEL*O vOS/£0N YN/ FOS z280*O oTo'o toto 900'°O cOT*O <OT'o MN oHo't Sé0'O £80°O Sé6°o res’e Tea'e YN £v0°O ose*t O95°O 640'°O £0T'o 940°O YN/99 SAIdWHS °S toe*o £10°O O'T- L90'O OOF 6ST°O O'T- IST*O Ofee LITO Ort ETE'O O'bE £9E'GO O'T- LBT'O O'9G 2G¢°O O'T- Sét'o O've *syan “od H (2W/99N7F 0A gec'o0 6950 o0r'o 9ge°O £9E°O cvc'o YN/OW ~ 9045 C49 T3) 093910 NYS ZT°t Z90'b 00? B9E't 856'0 60E'T ON 19 $ £8 NVC BO - 9043 (YYW T30) O93STT NYS 0°948 O°0L9 O*Sert o'sts 0°Sas VvHN O'ot O°sEe o'sse 0°06S o'szor YN O'tét e4*f O'OSOT 86°S O°o0oTt 00°F O'otet Sere O'oves £9°S H Hed $ £8 NYUF 80 - 5O4 CuVW Tal) O93T NYS 404 00°O£ 00'St 00°09? 00° OF 08’ ve CNIW) 4g *e a2MVL 1 SOILYy £S6°O 00°0£ 096'O 00°Sb 9TO*T 00°09 eZ44°O 00°09 ver tT OB ee +/- (NIM) LO ‘é Ay. G2°ce 00° 0E £T'ZZ O0O°Sb OO*TEe 00°09 90°02 00°09 96'B8T OBE 1 0IW (NIW) Ld “tT 3791 a9¥MaN¥ 00° ee OG'Ze OS'te 0S'0E se‘ét eaWib WONd TSLYINIWI YO £W Yad 4LM10S Ssuw iqvsyvas oo'ke 0S'ce oS? te 0S '°0€ Sé*ét cawis oo'fe OG*te og'*té OS'0e Sc°ét CAaWIL OS‘'ze BL°TE 0G°02 OS*6T £9°8T TAMIL OS'ze GLITE 06° 0Z os 6t Zorat TAWIL OG de BL°TS OS°0E oS'ét £9°@T TAWIL AAMT sam eu * an mt e 252 6460°CT 9Tr’s 4£18°OT 409° BE SSYW 8rB'o BoOEtT 6£2°O S+N/V0+H etter 9ro't 989°T +/~ Tée'o veb's G6S°f Cb6L°S vos CONNOG Y3ddN NY) NIW/EWN S = FLY ONE TANS Y SNIWASSY FINY 31997109 43LYM904 40 AWNTOA SHL TSYNSYIW YSHLTS LNALNOD Y3LYM WINOIT NO Waseda SI yI¥ £00°O 420°0 é£T0°O S4N/H o'ozotT Ortee O° SHE vos TIO°O ££0°0 £20°O £0ON/H O°Lke o°sgetl OTE? £ON OLE TE Z6T°O eS0'o 88T‘o 468°S Sré'o 000'0 990°O 62T'O £oG°T 70b'°T £ET'O 8c0°0 680'0 eEGte TEb'b Bio S90'0 L0£'o O9e'b aM) 4 OW La) vHN MALSW JIN 7 WHNDONDIW CAM TIVAY JED £8 NUT £E/L/L£ WASINAY coe*t £ob°O ¥00'0 £v0'°O 9v0°O bee*o BcT'T cEz'o 6288'S 8ttee Loree v4a°t T96°T £09°O STZ'T vOS/H S+N/tHN £ON/VHN bOS/tHN bO0S/£0N SOTLYYN LNA TWATINOA £B/£/L£ M3SINSY SOTO OIT*o ££0'°O E9t'o M vre*o 9£E°0O OST*O £LE'O YN 9TO°O SAMHVS ¢*f 404 000°O O'T- 4£970°O O°0T 00°O£ vO0'O O'f- LEZ'O OSE 00'0E£ 900°O O°T- EEZE'O O'0S 00'0£ *SvaW ‘a9 (NIW) H C£W/99N7 TON iu - 904 CAMYW NSSND) HO¥V3E 9NOT oct'o vee's bor tt écLr'se YN/¢0S £bo'o bec’o 6£49'O £G0°T UN/¥O 8Cz°O 9Sb'°0 é8£°0 B9L°O YN/OW ater cece TL0°9 Té9'"s YN/19 £8 NUT 90 - 904 CAMNYW NASNO) HOVSH ONOT £B/£/L£ GaASIAaY 0°00 0'o 0°e8 O*vet o°98Eet O°OLT O'os o°ot oat 0'809 O°sde 0°99 O'9T O°9o¥ O°ot?e aM) 4 OW 92 vHN “MALIV/03 ONDIW £8 NYT 9O Tce T'b ver M ~ 903 o°ost O°BE Stee YN CAMYW NESND) HOWS 5 a'y T°st 9°9T H cL'S cate aL Hd 9NO1 *e 31avL cttt 9rO'T 9B9't +/- SOTL¥M 00°O0£ 00°0L£ 00°0£ CNIWD LO *é 37a¥1 Sé*b St*b SL°E i Wiw 00°08 00*0£ 00°0£ CNIW) LE ‘tT aval a9VMaNY 00°S% OS’ 00°b caWIh WO4A WALYINIWS YO £W Y3d JLN10S SSYW L1vsv3s 00's OS*t 00°*b cawih 00°S OS’ 00° eawil 0S’tb 00° OS‘ T3WIL OS't 00° oss TAWIL OS*t 00° os’s TAWIL mt * aaAM an * 253 £L9°O bca'o SEE'O St9°T 692*T T46e°T v9e'T 9LE'T S4N/0+H 94E°T Teo'tT 646°T 162° L12*0 S880 zea‘o aeZ'o +/~ £OL*O 659° T~ O59 tz~ canagad 000'0 €00°O To0"0 &10'O 260°O FOT*O E8910 808'°0 S+N/H O°9ods 0'0E4 oroost Ororlt ore 0°csB 0°00e 0° Ove ¢0Ss L0E 806 '°G~ 08s 'S~ BTZ'O Bee't Ortet 268°T >) ae youth £0ON Hdddf NY) NIW/EWN G = SLUM ONT IWYS Y ONTWNSSY 000'O £00°0 €00°O BTO'O £40°O TET'O Té6°0 azo't £0N/H 0'00orT o*oost o'aect O°SG&S O°Sst 0°04e O°OLG -0°O2L £0ON Sber’o bectt- Ste o- 000 '0- tst°o B90°O 459T*O0 90c'O £090 WwW Ov0'O 640 '°O- TZt°0- 000°0- ZTeE*o Ort 'o vét'O 990°O 9TT*O 4 €00°O vl? o- 00 °0O- £00°0 600°O oT0'od £T0'o 610°0 £50°0 OW OoTO*O- P6L*O~ Ove o- or0'o £20°O GLo*o cbo'o £vo'o B4T'O A) MALSW JIGND / REMSOMOIW t00'o S00°O £00°O 2££0°O L£9E'O g£ev'o Seve grere £8/£/£ T3IsSiAay £2940 ecB*o vke£'o £09°T TtL*t O6e'T c8s'o 895'°0 0S6°O 2eLn°t TEP*O 282°C 427%E Tao't 9A 'O £542'°0O 60£'C £9TtS O&£*0 €8B'b 000°8 €£S'°S Ove tse Tete Tee'o 9EbT~ 6vTte~ O£T*O TEO'T z0z'0 92"0 BES'0 1a6'0 vHN Tiv'é OL? *T 6ET'T Svo'e eL9'£ E6GE OS8'é 000°E vOS/H StN/VHN £ON/¢HN bOS/PHN bOS/Z0N SOILYY LNATWAINGSD 0°0aS o°oat t'o- 0°Or o'dt o°cey o°48 orodt Yh £G/E/2 WASTAAY O'0ce O*oOst T'0- o°cot o°s9? 0°06 O°asS ort? af O°odt O°SLe 0°68 Ore Lan B°6 O'L£s o°dy OW O° OLE 0°00 O°9EE Ove 9°St S°et Orde 0°66 v2 MaLIWV/0a ONOIN orore}r 0°056T 0° O80T O°des O'ves O'yse O° Bee O°SesS VHN 900°O 90E*Q- 990°0- 000'°Q- £E0°O atoro S£O0°O TEZ0°O 0£40°O N OET"O TLe't TLb6t2 SE£'9 000°S £as'e SO6'f 605°E aeEre YN/ +O éato Qo OB8G'O- 0 OFTtO- Oo 9£0°0O o ¥S0'0 oO TEO*O Tt SS0°0 b £80°0 S avero 8 UN $ £8 NYT 9O £¢0°0 98L°0O ved'*T ¥S6'O Sect OO£'T Ta8*oO T9S°O £16°O0 8 YN/VO # £8 NUT 90 O*0gb O'ett O'ére Orce oret Or'te O'ds o*BOor ON voto SAWMWYS °8 OO'O- O'T~ O90'~ O'6- OO°O- O'T~- OF70%= O°S- 00°O O't- 200'O OFF 00°O O°T- ZoTt?o O°oTt 00'O O't- L£TT*O O'FE 00°O O'T- ETT*O O° bE £LO'O O'T~ £90°0 O°0E 40°O O'Tf- OOT'O O°0£ *s¥aW ‘awd H (£W/9) INT TOA QNY 31937100 Y31¥M904 JO 3AWN1OA BHL CWE TAY 4D) GSHNSYAN YAHLIS ANSLNOD YAaL¥mM UINHIT NO dasevd SI yI¥ - 902 XONN3) BEz'"0 Ltr g00°O tect Tée°e £6E°T 9L4E°O 000°0- ate'o atart £19°O Evert L9v'O 000°2 vov'o 9E5'T SEP'O beG't YN/OW ¥N/ 1) - 904 XONNZT ST"? BLS BE°s 6£°S batbo Sb’ BEL Thre - ° > OVA Ts Ci Met uy eOMMan TN * ~~ = N NS Hd $ £8 NUT 90 - 904 XONN31 MOd 00°O£ 90°0£ 00°0£ 00°0£ 00°09 00°09 00°09 00°09 (NTH) id ‘£ a 1avL SOTLYY 962"T 90°08 Teo't 00°O! 626°T O0°0£ T42°O > 00*0£ £T2Z°0. 00°09 S89°O 00°09 c£EB°O 00°09 8b2°O 0009 +/~- (NIN) 1 ‘e a1avl SL°E ek S2°e Se’e os°f os’é OS't os°o *y aay oo'or 00°OE 00° 0f oo°o" 00°09 00°09 00°09 00°09 i dIW (NIW) La agvuaag 00's ose 00°S os’ 00ty 00's 00'e oo't @awis WOM T3LV1NIIVD YO £W Nad 3LNIGS SSYN i 1s¥3S 00*8 osté 00°S OS*¢ 00'¢ o0°s 00°s oort cawits 00's Ones 00°s os'¢+ 00° oo'k oo'e oo't cawih ostd 00° OS‘ 00"F oo'e oore oo't 00°0 TAWIL Osté oo'd os’ 00'y 00°£ eo°e oo°t 00°90 TAWIL OS's 00'S os’ 00'+ * 00°! 00°2 oo't 00°O TaWIL ANAM eHORD ANM eNO * SUM TN OND * 254 zv°o.sT*0- «TO s««T°O- ««T°0- «T0- «T°O- «T°O- «60 SS isi (ii etsiéE € 6 Zz 8 zy'o. = 1°O- «T°O- =T°O- T°O- T°0- 1°O- 8 10-67 )=— OL sate Z ezI Zz SI 61°0 =«1°O- 9 3 9 962 «16 ol zy ve OS z Sy 61tOo@ 61°O I0- St vy 8 8 €8% ZZ 1I 66 862 9 9 z 9 1 62 61°0 0- SZ 69 €% 8 162 £6 ot ws Ok % 9 1 y 1 Oo 670 190- €2 ZL 6 zSt 26 oT os OW 9 z an an | 60 T0- BT wz ££ 9 961 £6 6 0 6 i or it 6 2 @y 81°O «= T°0- «TTO- T°O- T°O- TO- TO- ZyI oT Tr A | 1% 9% 8 OF + £6 1S0°O 1°0- T'O- T°O- T°0- T'O- T°O- T°O- T°O- GES €% egie o8f ¥Z— T0T 9S Ove 1L0°O —-W°0= «-T°0- ~T*O- T°O- TO- T°O- 89% = OT Ely 69 10€Z S8E 92 Ol 96 O29 650°O I°O- 12 dy Zt» zZ4OU 1°O0- «=6T0- seek ery] O62 $¢ 611 1& £€66 SIO T'- €§§ OL OF Z og 0- T0- Io Zé zu ty =O €T°O TO TO T0- 1°0- T°O- T°O- C01 ZI 981 Ze siz it 8 “e 2 49 O6°O 10- ZZ 86% 69 ST ghe 29 L tol ¢T goz 96 OT ez ysO8T ST°O T'0- SS SS 2 6 9% «£6 oT 4€Z 09 46€ Zt €& es Sl tly Te0°O «6° «88H TZsTZSiCaT CTO 79 ZHOL OOZT Sez Lit ZOT OOLr 990°0 «T°0- oh 6h COT OS 86 6 ose ze 6vse 229 It TOT 29 £92 g60°0 TO €9 Of 2 8 yl 6% 8 £96 €2 oeze OVS IIT vl oS O12% 9s0°O T°O- 602 48 49 9 88h «68S O6T 1BZI 97S «= ZOE. «OHH HZ SCLZEsC*dGSC(OUSE 960°O T°0- 82 OS 7 OT O98 T°0- TO- 6IE 2B 1s@ OLT «£6 est 7S 8y9 GL0°O = T°0~ T*0- T'0- «T*0- T°O- T'0- 1° T°O- GS8Z 4E vI8z O1S 661 é21 O6T O9SI oz°0 «TO 12) «dS CUT COTSOOTCOZSts«C*#K™T esl vol 266 Il 16 oor dt It gvo°O 21 406«1Z—iZsi(‘<iéi‘i:SC‘“‘wKBS:C*dTERSC*CGGY SZ 09SZ OZSZ OOSE O60T Bee 95% SIT S9Z¥ 990°O zt SE €9 62 006 €€9 92 OVLT OV61 009% S28 Of2 €97 S6 SEE 670°0 8 9» 3 86Gh) OOECLL HE CTO =O) OZEZ«-BZOZ «OHS s«OLHT «~OFY «= EZS_—s«OHT. «(ONES TS0°O «= T°O- SPO «T°0- «TO- «T°O- «T°O- Ley GT OB0Z SEST O08S O6ET OOF S8Z€ ET 0195 ovo°o « @tiST CG (iiLSCi«i‘éEE:C*dT'Q™—Ci‘édTQ~—Ss«OGEZ «OBL «= -0029-s«OOZBT «59S SSH SLT O(OLZL y70°O «== T*0- «T°O- «T°O0- «T°O- «=T°O- «BOE «=OT*O- ST 096Z OZIT OOFOT OESZ SoZ LOS 04% OOZOI z90°O «6 Ty) SZECSStiCiELCC(GsCETCsZCRGT = sC«CT°O- = GZOT.s«BZE Ss CHS «COTE «S8h)=— ee) OLT $69 TI°O) «EO T°0- ST°0- «TO «TO T°O- TO 1°O- 0627 0062 OSOT Sel 46T 098% 2 Sz rt ct? 1-03 wi 1-3 beri Z 4 € 9 om OA mW ™H 64d UW Od OOH (ADDS -z7 98 _ON _19 47m 42 4 AN LN =N rN * 6 ois es eo © © © @ o 2 © 8 @ SQISNQSeTSsSrovrRxgnssgs RINK DOR DOSKR AMIN 6 oO > Are FOsTrrerrnsewsaewswsn ° 9T°% hd 8500-2000 0000-2EEZ STB0-TELO 0€20~-S790 99790-6SS0 8SS0-€1S0 eTS0-62 90 L£240-£4€0 SSTI-€0T1 TOTI-O00T 8560-0680 4040-2290 0790-9€50 9e£0-1920 0420-0020 SUIt-Zy01 1290-4990 8ZE0-0£20 0000-2072 S@Ce-G2T2 0060-0280 0090-0050 6S20-8SS0 9990-02470 8170-0160 g0E0-0ST0 £9710-2000 S000-SE2Z O€SZ-STIZ OSTZ-SOTZ (Te00T) atl s0018W €8/02/8 uBS 78/21/8 79/11/8 28/01/8 28/6/8 sekay" 3a Z8/ET/8 923Ng° IW Aeg ze/¥I/L = OW pueys] ee OYDIN 78/92/8 ueg Ie TeP zB/L/Z1 —- BUOIOD o79g a3TS 255 60 TO. 6 Zt & $$ o¢ 4 € cSI IZl Sh ET OTC Cty sk’ 002T-OOTT zo y”!~C«C(<‘ézETS!COCUCC‘éSC(C‘dTTO:SCSV sll 78 wy Zt 8 ysl 9 66 «ZS*Y 8S0T~000T yO 10- O1 wt 6 € 8% @ 1 96 #+€& 6 YF 861 £ 82 89°Y 8560-0060 0 «10 Of 467) OE CC (‘ESOC z 6s90lC«adS( (<i ke lol Z Ot TY" 0060-0080 0 T- T § “46 € ww z£ z Pr, ne | en a 3 6 6 << €S*% 0080-0020 yO. T°0- «T0- T0- T°0- TO- T0- T0- T0- & @% 7 F @ 2 zZ LS 00L0-2E90 60 1O- 2 ST €f Z 6f 9 I 62 6h CG (iC WS Z Z 09% 0€90-0090 so TO z € gL t 8 £ z wm @#% € %«. 7 co Z@)COdTtCSC*é« «CTE HO-OOO 0 10-2 «Of OSTCECOTD z Ih 69 6 @ 2 1g zZ € 4£9° TEEO-00E0 6S°0 T'0- T'0- T'O- T°O- T'0- T°0- 10- TO v 89 EF 4% 2 60ol € 6 99°% €0Z0-0STO 09°0 «T°0- T0- T'0- T°0- T%0- TtO- T°0- TO 8 989 GI € 9% € %£ SS*% OSTO-SzTO ; iso TO 8 € a € St £ I w% 6 @&% % £01 zZ Ol zv*y SZT0-00TO saeg Zy0 T°0- TO- T0- T'0- T'O- T%O- T°O- TO- 7% 921 OF € E€ £6 Z 8 tty 8500-2000 80035 2°0 «T'0- T° T°0- I°0- T°O- I°0- 8 T0- 6 I zw 9 £ szl 2 ST. oTH*y «0000-Zeez £8/02/8 ueg 1-0" ni 1. bat cqroot) oT m ad mm ea om (ats "os "ON To ,_ 3H Oo "HN M (NBO OTL 26g 231s 256 APPENDIX D FOGWATER CHEMICAL COMPOSITION: BAKERSFIELD, CALIFORNIA, DECEMBER 1982 - JANUARY 1983 257 COCO POOTO = LTOrO 1o99ta L008 000°0= 00070 rgora A IN Th) re] Nw Law QIN 7 WeMOOMO TW TLO°OT Ethie PS LOL OGarO 000° O- BEO*S ELT*O 040*0 cbet £4b'O TZ¥*o 990°O voto Tee*2ze OG£°S vea'OT BETO I SSW +0S ‘S0N 1 MALAW DLA CONNOR NAMOT NY) NIW/EW CATV TTVAY AT) KKKEK oo°9 096°O eort BB°k Otort sAONSNS AATI-Ad-Hde Hd ot/~- 03198 arto A QOMIULN't Tha OOLED FF OONAT Ot eae OroerT Tt0- b°O~ Tae Tto- "yeetl “OOCET T*O~ ial IN 3 | NY ad Maueg 1) Wd HOM OTH $ CB Vat 6c T4SygNed ¥aTeo Lat 4£0°0 TeT to 000°0 £90°O SP0°O £0o0+o veoro Brod 900°O OOtT~ TOOO O€L40 vAYtE cBT°o TLENO TOO OO'T- 20040 *SYUSN 49 Mo eHN N UN H CENZ9) 9 f WEMHOMIEH G = ALUM ONT THUS 9 ONTHNGSH MING CLOaTION WaLemood JO UNIOA yyO0. WayASvaW MAHL EA \ MEOW OLAND 7 WYMOOMIIN 7 $ 28 Oa 46 - 904 WIT ASYAMUd £8/Oe/e WAST NaN 00010 £26°0 Tora £O0%O O0tT~ AOO'T OOOTT- O0tT- TOO"O Ort~ OOOtT £4Anrd TOE TO OO*T £00°0 Ort “Suan Wo 30) OfHO CADIS CPN DOM 201 NU/NW GINN0d YIMO1 ¥) NIW/EW GS = ALYY ONE VANES Y ONTWIISSY TINY UELIETIOD M4alemedd dO aK loA cate ira OooOtT 000'0- 2£B2°O Getto _ £80°0 vEne OEHI/CATIS « bOS/ CATS S+N/U+H SOTLUM MYON O'T~ OfT- ort- Z£446'O *OORST O'T- 8998E OTL T9P*T “OezZyT d01 OCH CATS t/- POS NU/WW/ ON OCTV/W OMOTW 7 VNNSN £O/GE/E 000890 zO0*0 Sth/H £B/GC/T “OGbTT "O0R8OY SCN NON WAUWASCIN AAHITA ANAILNOD Aye tae TNO Tasvd SI $ €B8 Dal 6€ ~ OUsa OVaALASMANUA WASTAAN OELO eroro ‘oO 饰t Ovg'o ORG gest ece*o vara £66°O S4N/7HM YOS/¢0N YNZOOS UND VN IN ga tin SUTLOM LNA brite $ 0 Jad 4€ - 904 WaT ASMaNeA Was iAany O'OeTtT OrGng OORT OOTTE OF6LS Otoget OF O0b *QOLOE O8'00"E a wo YH ™ Wn “OA OMATW “ff ¢ 29 an 42 - O08 W1atscy °S Ww m1 IHL INSUNOD Mabe HIPer) NO Waseda ST wry 9 4 0¢ Nl oro OOTP SBS RE? e orn Rate LBeg gee? t H Het GaWiL TaWll 4 aay , SIWUWOS 88 MOA SJOUMAAY NALHO LSM awk 690 OO'0R LHS £B49 é O°t OO'O£ Lary Loe t CNTH) WN Li CSWTL VIWil + NOas Wao WwO YO £W Yad ALMWS Ss aaa S31dNYS *S NOS FOvNaNY WALHOT AM SW °T- Ott £a°L £89 é 9OBE O'f¢b LBD EE 1 HD (ATS eSWIL Tawi #£ /W OMOIN / FHL NOMA MELVIN WS YO MIyv ‘NOMad SLN10S SSuW *S Fav. SOTLYY LS ¥3S £22 O0'O9 Late £9°9 € acre OuTor £BtP LEP 1 LQYN CHIN 10 GSHEL TAMIL o€ ve 0 oO 44 oy ATL ‘T OOF FAL Q "eet Bare “£a°9 t H Ha SaWNTL: Tawil # Sf Dae 258 «TIWNANAAAD I 9IT*T GLTtT T£4°O vTZ'o cBB°0 L4E°T elect 646'0 eToOrt B40°T eLttt arort OZHO/ CATNS SOTLY O'T= O° 8GEe Oe O'Tét O'T- O'eTEe . O'Oort ‘ O's MM OG NON OCC TT UN Oa 201 OzHD \UHWS \CI/M FOE Pats Gere Fe pe cere 9btE. Ore ObTS OTtb 40% te CATE. 99° BOtb e0't Gate O9tt Ad-Hada He LoO°T OPT TO TOT goer ONTT= CLO O't- GATT ON9"D NNT E60°O ‘Cc ObTet gee a ONT BLOOD _ Roepe ONT COTO erp £00 °O One T~ OROrO 400T OO T~ Of0°O Tro'T OO°T= SEEO THOIO OO°T- TET*O NAOT CO*T- £TE'O OcetT Ott Trea gtett OO'T~ &4e*o O48°O O0°T- £EEO “SUA SID e4/- WALIAMYOD, +/- DOL OfHD CATS CEM 79) ANT OL \ MALAW OTHND 7 NYMOOMITIW 7 NVR NOW OMDIW 7 CONNOD YAMOT YY NIW/EW & = ALUM ONTTAWeS Y ONTWAGSY NY 192991109 YalyMeod JO aWITIOA AHL WOMS TaLeINIWI MO TOL*O COLT FOS/(AD)S 4 Me IOW O° 88E O'TOE O°S9T O° 00T ores CAT)S t/~ OMOTH 7 CATTVIT RAY AT) SEDINSVAW MALTS LNSLNO9 MaLYm i # €8 DAT Te - Os WVALASNSNYA £8/GE/E AASIAAM Oat geod PEO 2£T*T . BOLtO 90T*O £99°O GLZ9°O AnOTTR b6eyeG 489'°O TOO*O- FOLTT 66840 LEO) E280 QOOWD ELETE Ov 0) = O9LSE gB2Ttt ECHO vtc*o 802°O EbH'4S OLE ETG*O w4ory ObT ST TaB*O vat oO £49 °O BIZ'°b BOZtT OGESO Ge.ty LOStT 650°O T04.*0 NOOtTE wee OLR°O O0L'E are’? 696° tard BBO *OG 9TF*0 4T2°e O90%T £94'O £t6SO {c4*E9 Ber O beat GIT*t LE0°T 256°0 Oe! LBLtY TL0°T £68°O 6£3°0 f4S°TE 4E0°£ réoTet c08'o HYe*O SEBO eG4tEs eTEtT SEBtO £OB*O OLB RE bEO's Geet oeret STtet vIOTZt Oftert 290°S ¥G6*0 SHN/U4H S+N/H S+N/YHN POS/OON YN7eOS YN 7D YN 19 +/~ SOTLYM INAIBVATANSA $ 7 ORT TE - 90d 1 TaLASMaNed L£a/Ge7e VASTIAAY Orocet T*Q- E°4e oO°ate T*6t O' 008 OHS Tes 6'4e O°Oeet 2ty O09 Orns G9 Trag O'sstT £99 O'or? 9°08 ate e O°O0OT £9 O°Ock Os £°6 + eae & O°ed'e e°e O° Ov! O°468 O°6 beng 9°ON8T 99s O° O9S O° 0OT 6*e ftdt O°NPET bes O°OLS NOuT ate £tbb O'OOeT Tet O'OES OtOgT &°o O°ANeT ETT Or OAR Orort £t9 Gite O'oertT Ts arord One bro £° Rb O'OTeT aeet bese O° 089 O°0L4 O°odT Bed E°og O°'OrsT £°GE O° OG NS fon WW a ON $HN M YN NNN NNN ce WE OA IAT ASM ANE tyr NO Mase? SI YI¥ EW M3d ALNIOS SSUN SAWS "eT MO FOWMaAY ALHOTAM-aWwId Omens Of AA Cts eT O'r¢gT O'Tos 00° iT Outs O8'feT OO” OT O°OrT Qrnnt Bat & O'fé orcs OOtF 9g Oadt O'4tc 29°C é O'rOT O° 60E Oot 9 eee Z1T'O ) ete ceeo- ob O'STe O'9£LE OO°T- AGtT- £ Oats O'9GE Batt ATe- = O'Gre O'G6E “4T*E- EB'E- TF 02H) (AIDS caWil TAWII #€ ‘ek aay. SOTLYM J WSs yas c6'6 00O°'O? chrOot THs et 05°8 00°09 00° 00°8 Tt id oe 4 Of*GS c6°eZ 00°! or BE? OF'tS £899 c6°S 4 opty Of'GS Zé6'v 00" 8 ave’ 00°OF LOE LP é bors O8 yo BGs ost 9 9e'O0 O8'rE GLtO £T°O 2) Z£T°O- OO°O£ B0t0 ce O- ob 4é*T- O8tbE OO'T- BEtT~ £ 8B°T- ObfGE OG'T~ “ZTtE- @ OGtE~ O9'SE LT E- EB'E- TF LIM (NIW)L awh + TaWTl ‘e 37ayL O' ere E9°L Ebeot evs ét o°sit P4°£ 00° 00's UT Seen cées OOT!’ oT “eRe SS'f £8'9 c4'S OredétT P4°E Gate oo'V O°ssé TOtb LOE LOE 0°49 OTe BGs Og'T o*ott 94°£ GLO Zt*0 o°rs BN* be B0°O avtO- MTO CN OH o°198 40° OO'T- BET T- 0°9S Sete gett 4tte~ = Orage OFte 4T°S- £BtE- TT M Hed Cawtl TaWrl f ff lh ft Tare), 259 ooo" E00° c00' Z90* TOO? c0o* roo' oot 900° £00" OTO* Oto? A Les TOO*O 8 000° O 4 TOON @ 900°0 Z 00040 TANTO TONG TOO*O roasa TOQ’O £O0%O ca mOO*O S NW resid 9£b°O TOE*O TeAtT OOT*O- OOTTO vOA'k ogT*a PLT EO £rO'b PETYO TET*O 0 Foo'O § TO0'D § ZTOrO © £00'O 100'9 § Tz9"0 0 Z00'O TOO'O = BTOtD 0 ZOO'O ZOHO OZOrO 0 Z00'O THO'O FTO 0 £00'0 E0O'O <£FOrO 0 900'0 400° = asaro 0 800°O F00°0 — HROTO 0 800°O0 §£090°0 0 800°0 §=FO0r0 0 vTO'O TEOrO 0 020°O fz0°0 IN ad MILIW OLAND / WeNoMoTH O99°OT 9GE%b EGE 2bo°oT 845°8 ETL*OR EATS eoS°9 «beTTT | OLB TE rete = OTONS 1 BLh'e. Stel’ 4eB°AT b60'p BLES Ste PE LEG “bOSTTT LOE*LE BPL ET 865° 4F O29 AT Bib bb 6fe oT ssew bos £ON ai) CONNOR MAMOT NY) NIW/EW GC Ca J MALWW O19 OT/9 17 TO°O O0°O 1O*O 1O%O 80°O £T*o FB ETO"O - Taro 900" ea0'd 900'°O GO0tO 40090 40090 STO*O BTO'O ATO"O LEO TEQ*O Ow no 4 WOM ALM roAe Zeta beard ROO TEOtO wIOro f9O'O erred Beha ; uo Sou NOR bY Soo'e BNO'b O18? ak do QtbE ort of Ores are One Okt Ott Oery O'aT Od O'od no MALIT / * 28 9£0°0 GEOO reO'O ATOO coro Panto 9L90°O Panto bTT*oO £9T*O Toro M JNA LNOD * 28 ad ores Oc ore Ort neg O°dT or? 9°8 Org nro QttT Ortrt NY WYMOOMITH a7 TE - GOA WVATASHAMHA = ¢G L200 SIT*o oGT"O 64T°O MALY Tne oT NO Te 904 WUSTISYA Or ees O'ToT OBESE OrTeT O°94T Ot LHe OrTEt or eoT OtZOT OrZit 0109 Ose ad Oto OOtT~ EFOrO O0°T= £690 00'T= ZL9°0 OO'T= €OT*O OO'T= 980°0 09'T= 980+0 00*T= SATO OO'T- TET'O 09'T- £TE*O ONT TeE‘O OO'T= PEO OY T- EEETO ‘suau OTH (FN/9)901 qagva SI y1e *y eocecoeoooosooyo fg cee x 9000 C9O5S9090599090 14a) SA 39) O°or O*Be O°0€ O° BE OE Orbe 0°99 O°B8e Oree O°sb o°eSs O°ro TOA WO. ew 3VIHL LOE bare See SG°£ PLE zo'ev OTe Tae a BO't 40°¢ O9'b Hed TAWYS WMaAY 00°09 00°09 O€'SS O9¢bS O2°6S 20°09 OB te? OB EE 00° O£ O8 tre Ob Ge O98 6E CNIW) Aid zbroT THs 00°6 E6tL £9 tote Lore ase G20 80°O OO'T~ aStt~ BST £1ts- 2t¢E- gBtE- ZaWIL TAWEL ‘ZT MO A3LHOLIN-3WTL Ev OT Tbté 90'6 00°8 Z6°L = 00° £89 = 6G c4th 00TH LPR L0E agra oget BL9Q LTO B0°O | ZbO~ Oot BEtT~ BET LTE 2tee- gBre~ TIWIL = TaWTL WwW) NO ad SLrios Soe BOM FTNOVR OH » SUM TLIC N OD - 260 Ze WAST ABM A9b Of got 90Gs0 SrO"O SATdWYS *S MOd ANNA NALHOT 3M- AW WKEKK PETG L7T8H°O Aré'o OOO8T= OONTT~ OOOtT= OO°TH- O't~ Ort OtT- 0S°8 Bo°s 4 KEKE GIAO vTO*T OOOTT= ADOT ANOTT~ OOT- ETONO O't- ort- O°T- ago ORT a AO Cay 459 °O NON T- ZAC P4ETO NOT STOO O°T- O'STE O°es GI's GE°E Z {OOO 91's LOB TO £946'°O PLutt £OTO zee O ONT Gyr as O°SEé O'f4a GLTE CO's ? T's 608'°O T48°O ONO00T= HOOT = OOOTTH= ONT ZAOTD O°T~ O't- O't- oor? oorT & 00'S 4TO'T LAO°T OBE 4BT°O £94'O 998TH OOT*O bis or'e9 OTOL oorT 00°0 v toe O56°0 TEotT O97°b OBTso LOTT NO'T- LTTO Ste Ores O'Poe O0°O Oo't- £ o0r9 Bra'o O68°O Lotte 49T°O £cv'o NgeT~ £rOOd 9°98 ‘ELT O'40£ FB'O- “4T*E- = cles 44£°0 £9b4O ON0tT ZETO £POO Ont FOO Otte O'SOT O'Sp EL'L- GEte~ 7 “SYUIW fad sFINAMSAAATI-AA-H Hd o«t/- QHLIAMNOD. = 4+7~- J0L NZHO CATS C£N249) 9M on OEHI (AIDS caWIL TAWIL & N MALS ODT 4 WYMOOMOTU 7 NSV/MHS ON CW OMDIM 7 CONNOA MAMOT ¥) NIW/EW SG = FLU ONT INES Y ONTWNSSY NING T9LIITION YaLYMaNs JO AWNIOA FHL HOMA TaLYINIWI YO CATV TEVA AT) WEMNSYSH MAHL INGULNOD ALUM OIA T NO Wasevd SI yIv EN 4 3LN10S SS¥H $ £8 MOF TO - OA WIATASMSNH”T ‘€ DPaYL £8/SE/E MASINAM RETO 4vet SOILYMY Lvs vas f00"O- eso'T 000°O £H0°T aTg*a ONE TO OTa*O Brs'o e°stt O88 B5°e & C00 O- 940°T ?00°O O40°T 6£R°0 O9Z*O BOLE vtort 08 hr? AE? 0G5°S 8 609°T STO"O Sbart Gord eget T 659 'O O08 BINS Boe é baltt BO0*O AeT eT aT et 8408 £E4'0 O°GOT B2°E oor 9 090°T Bret 500'°0 £retT L£50°T C58'Re TARO QO'OF OO°E oo°T S HLL ob O84°0 ROO'9O BLS*9O bared Bh £40°T O0°O?9 OO'T 00°90 v 9ELIS bEeT*T 940°0 AOTTT LEE EELSL TEOT 00°OF 00°0 OOrt- ¢£ Tee's GOTT HOO*O Agort ALOT eos T 949°0 vort £8°O- ATE EF GEv Oo Teors TOO*O TeOE O£e*o cob ty 29640 OO'GE EEE Glte- 7 OCHI/CAIIS bvOS/(ATOS StN/¥+H S4+N/H SHtN/hHN YOS/SON YNZbOS YNZ HD UNO YN/ 19 +/- AQIW (NIWA GAWIL TAWTL #¢ SOTLYM NYOIOW SOTLOM INABA TIO $ CO NOC TO - AOA WISLASYSNYY 42 BTL RB/GE/SE MASTAIIN Ott- BrAtO OTOROT OF 846 aon N°oneT O° 46S 94'S OSB 4 OrT- eTOtT O'OrOT OORT OHVs OObrT OLE 6B°b 8aee? 8 Ott- SO9%O Or Ooob Orek gran par feb GIG 4 ate £96°O 6°S GL ° OrT- T6é8°O : £°9 47'S 00°C & ¥°e £407 Oeste CO'G = OOrT t oe TLOTT O° Oba O'OEsT £°9 bE th O0'O E oe O48'°0 Oey O'OB4E O'OE OOT? ERO ATE F O't- £9b°O 9'oGET O*OTTH O'o£ E2°5 LEIE- Gete~ TT - OCHO +/- bos ruth 4 d HN N Hd SSMIL TAWEL & \ VW NNN NNSA * £8 NWT TO - DIBAIASMIMT of YT 261 000°0 000° TOO'O €00°O S00°O S00°O £00°0 £00°O £0O'O ia) o v - 000°O0- 000°0- ~- 000°0- 000'0- 100°O TOO°O £00°0O %00°0 900°0 £00°0O S00°0 SO00°O 800°0 £00°0O TIO*O c00'O 800°0 c00°O IN no 000° O- £6/C/8 W3YaLNa 000°0- =000°O- 000%0- 000°0- ONDtO- £00°O 000°O 200'O 0TO'O ~— 000° 920°O = TOO"G £20°O = TAO"O TEO'O = ZOONO £TO'O —800°0 S00°O = Z00"0 ad MGLAW JTAND 7 WYMOONIIW S s 4 8 9 b2tot SSUR O2E'°O LEGG POET LOE % iS 7 ea 8 £L°O £b'O 6£°T ts a 3 60''d vos é 00E'0 ETO"0 £50'°O vco'o S50°O Toe oO TeET*o G4Z°O £ET*O vEr'o Ww NW 34 SZT*O TTO"O PE0°O Z00°0 T9O'O E00'0 e£0°O TO0°O £000 £eE°O 500°O SBE" G00*O FELO B00°O eST*o 9£0°0 Z£5°O TEO'O d MaLaW OTAND 7 QUNNO YAMO1 NY) NEW/ZEW GS = FLOM CATO TIVAY OW fee T*9- O's? Ores o*ag O°eg o'r? Oruet 0°66 £TT*O ZTO0°O £EO°O 9TnO tTEnso send ££0°0 £E0°O FORO 608 °O vo WOMOOMIT MW ONT TANS dQ) TPT O choo goo geet 808° eaegre BLATE CPOE OLTS bun T°O- T*o- ore OZ Or4 O°oT O°4at Orest Ores Nu TOO*O 00040 oonrod g00°O 909°0 TOO*O TOO"O HOO" 000°0 900'O T°O- T*O- T*o- T°o- Oty O° EO’ 0°46 Ora’ ObE O°aé6e Ores O'8ve Ove O°e£e O°er O'eok orty OvETT ng ad MALI 7 WVHOOMITH $ £8 NUT TO gco'o 9e0°O 800°O OTO°O £rO%0 £TIO 400°O ETO"O TIO"O PnOrO £cOtO Z00*O BH19°O 4TO°O 9£9'0 £00°O ono'od ¥9Ord TaNed 26910 N uN #¢ £8 NU TO H JQINML) NO W3ased nt T*o- T'O- O° ERE Oreo O'aTe O°ng? 0°060T Tto- Orocrt a4 ~ 004 OTSLISYaNed 9vO*O °S oor t- £00'°0O OO'T- ZTO*O 9O°T= S6TO'O O°8 OO'T~ 9bO'O Of bE OO'T= 28040 O89 OO*T= OOT*O an*t- ETTO 90°T= £rO"O OOtT- £90 *syay 89189 CPY/D) OMT AQ AMADA FHI tes St MIYy 'b 0'o 94°S o°0 4are oro Sa°e 0°90 9U°S 0°90 41°sS o°o 0O°sS o'o bEtb oro a0%9 oro gL°S H Hd 31401 Sa dws JOYVMAAY Oro E°sit O°b OB be 00' tr o'aor 00°09 O'OE 00°09 Ofee 00'0? O'os brOort OST OES (NTH) WA 47 os’ ase Stes SLE 00°e oot 00°O /£8°O- ££°e- zawiL *6 MOA MHLHOTSM-SWTdL og*8 Bh°9 GIG BLE 00°8 o0°T o0*0 £a"o- LET EM CaWel TaLvinwe yn 3LN105 SSUW acre ofS GLE 00re oo'T Q0°0 00't- 2Ttg- Sz*e- TOWTk ear? OSS GLtE 00°s 0o'T 00°06 00°T- A ee No GE'e- TaW Ek rmUM TOR Oa * “aM TOWN Oe -™ 262 £B/00/8 WAMAING 000°O- = 000°O- O00°O- 000°O- 900°0- = O0040- Tey- T*o- T*9o- TtO- Ten- Tte- A IN no ad NW a4 ra) IN ny bs MALAW OLAND 7 RYMOONIIN MaIEV 7 Weg a4 $ EB NOT cO - NJ WILASYaNYd 4 ae» sva’o £468°O 9vO'0 TOT*O Tr9*0 Lis ae svea‘o £6B8°O 9¥0°0 TOT‘O TIO*O £ereo Hozrt TvoOro r90°0 £eTo Q0gtT TeO*O ¥eoro 000°0 £TO°O o°o 40°R L9°9 L£9%L t H Ha cAWIt TAWIL & °S a1v1 Sa twvS *T MOd BSUM3IAY DALHOTIM-3WIL 900'O OO'T- ETO°O Of O0°OF 249°B 49°L tT *S¥aW fdwO (RTH) : . MALaW JIN / WYNOOMDIN CANNOT YARM01 NY) NIW/EWN GS = FLOM ONT IINOS © ONTRASSY FING 119909 NaILuMgO4 40 ANN ION CAMP TIVAG 3D) aNNSVAW MAMTA ENGUNOD MALOM HINDI NO w4suva ST $ £8 Nur oO - 904d NIATASYSNYA Sesez/e WASINAY 000°0 900°0 900°0 £TO"O KEKE L0°S 6cv'o B4r°O OOO'T- OOO*T- ONO*T- OO*T- £TO'O O°t- “SYUAN SITY) « SONSMSAITO-AR-Hd Hd et/- TaLIgNN0o. = t/- TOL OHI CATYS (£/9)9M1 JoL : NO YSL1SN OFAN 7 NYMOOMIIW / NUANH/ GINNOA YAMOT ¥) NIW/EW GS = ALVY ONIVWYS 9 ONINNSSY ANY TILITIIOI MALYMOOA 40 ANNONA CATO ITUAG 4D) CEYNSOAW MGHLIS LNCHNOD NaLYM INIT NO W4sba SI * £8 NUE CO - 904 TIAL ASNRNGA £e/SE7E WASTAAM 9z1O0 FeOO RECO Zt*t 600'°T €00°O- 280°E vo0°0 £90°S 818°0 vac? O4TtE Oreo 2£90'°O 84¥°0 OAM CATIS vOS/CATIS SUTLYS 4Y TOW StN/0+H S4N/H SEN/VHN bOS/EON YN7ENG UN7HD YNZ OW H/T t/- SOT LOM JNA WALANG $ eh NUE ZO ~ UOA WIaTASNa td Le/ne7E VASTAWM O'T~ O°T~ ort- B4b'O OT OCET O° 9801 Oths Oran argo 0°09 CO eT 4 O°OTEe 901 OCH CADIS +/~ oon £0N 1D F| OW (sy WHiN M UN NUARH/ \ CIW OMIT 7 NNN NNN MALE V/O OMIT 4 f Ss FA ner FO OO4 ay asyanud C£W/9) M7 TOA Lil cAWIL YSWIL ¢ SHL WOad T3LY INI IWI YO MIV CW 494d GMOS SSN *y BNL SAIdWYS *T NOS FOVMAAY TALHOLIM-aWTL ort- Ort- 29°68 L£9°L Tt OZHI (AIDS caWTL taWIL + \ VW OMDIW / FHL WOMA MALIN WI) YO MIV EW ONGd JLNIOS SSUN *f 21841 SOTIYM LTWS vas ame) oC*'0? £9°8 L£9°E T AWIN CNIHD LH @aWIL TaWil it ‘Ee DVN $°a O° 29°B LOE tT H Hed ZaWwil Tawil + Leste “T DVL 263 WA'eT BSE 2ees £LTO oro 900°0 vIo°o 22) kd cand GE9"O 900°0 GTO°O God*t e62°9 £02°0 ¥8z'o £90°O OOF*O~ OOT*O- BSF'O OOTSO- B800°O 000'O 00°T~ co Teget £88°0O aL££°0 070°O zvO°O OOT*O- NOTSO- 944°0 NOT*O~ = FO0'O 000°O O0°T~ oO “G4Gtce «=GSttOt £BC'9 vot o OREO 900°O Bro BES z460°0 O2O°O TOO*O = OO'T~ te *S¥aW o°OD SSUH vos £0N mi) 4 OW De) tHN N YN H CEW/9) 9 MALAW IJLAND / HYVNOONIIW GINNOd Y3MOT NV) NIKZEN GS = FLUM ONT TNS Y INTWNSSY ONY IKILISIIOS NALYMADS 40 AMITION CATIVTIVAG 4D) MaMNSYEW MUHA TA INSLNOD MAL9M WINGTT NO Wasv"q ST $ £8 NYE £0 ~ 904 W131 ASNAMYR ge/se7e Wasinay FFL°£ 68nd 8vg*o STO*O0 40°S 84'v vao't £sart N004T- O0O'T~ O0O'T= O0°T- 200°0 Ort- C6°S ces TeOT*T vecrt 900°T- Bz0°O HOO'T~ OO°T= ¥O0°0 Ort- as’f vé'y Z££0°T corr FELSE £5T°O 8va"o OO°T- vrOrd Ord ‘SYAN THD « SONSNSAATO-Ad-Hde Hd at/- GRLISMNOD. = +/- 20L Q@HD ODS CLW/OI INT 204 N MALO OTHND 4 WHMOONOIN 7 NVA (QNNOG HSMOT Y> NIW/EW S = 31V4Y ONTIdWYS VY ONTNNSSY ANY 11937909 YALYM9Od JO ANNONA C3IWVVAG 4D) VSUNSVSW NIHLTa LNSLNOD MALYM TID NO WaSya ST $ SHONUD FO - NA AIZTASHINYA £O/GE/Z A3SIAgay otTrO Ford gzc*o Zv*tT 000'T 000'0- 282°O TOO'"O £62°O £z74°0 Slt fe TOOTO~ FOOD GOES FST 700°0- 000'0- 958°0 000°O 95840 e79°9 SESS E008O~ EOOO- 446° yeett tSO°S ose*o GG6'0O c00°O £56'0 64°90 QNO0O'0L basta e9ta 60g t cotet OEHI/(ATIS bOUS/(AT)S StN/U4H StN/H S4N/YHN YOS/S0N YN/H1S YN7YD BN/ OM YNZTS +/- SOLL¥A YY OW SOTLOY LNTWAL ADA $ £8 MYC FO - 904 WET ASMINGE £e/GT/E WISIN O°T- OfT- O't- ESAT 0°00%6 0°0089 0°008¥ 1*0- Tho ‘ogee? Tan O'Zel O'T- O'CEZ O'T- bOET*T O°009r O'OOTE OTF YT TeO- To~ O°04659 Tth- Or ee Or£ O°OTT O°968 COTTT O'O92¥ OTOKEE O'KAT Oorrt oreg OTL O'S o'a9 30L OTHI (AIDS 4/- ens YON VW F OW v2 HN N oN VN NN NNN Ma) 1/02) ONT . “of NU/WW/ \ CI/H OMOTH 7 $ £3 Mer ro ~ 904 MW 1aT ASYaNod SAMINVS °£ 4Od 39UN3IAY WALHOL3M-3WIL H°0 S90 00°09 E4b°2 zoey £ o'O O°T O86 £B°9 00'9 é 0°0 Of0T 00°She 00°9 BESS 1 Wd (NTW) - mt WA 4 TaWIs TaWIL ¢ 4HL WONS TALYINIWO YO MIV £W N34 31N10S SSH ‘oe 31vL SAWUNUS *£ MOS JNUNaAY WALHOTAM~SWIL Orf- Q°t- f6°L O°eke O'T- £a°9 O°aTT O°946S 00°9 O@HD (AIDS cawil NVM OMIJIW 7 3HL WOYd TSLVINITWD YO MIV £W 43d ALMOS ssuu *s AVL SOTLYY LIVS ¥3S cord OO'O9 § cb's Teg O8*ée £9°9 cos H0'Se 00°F AQT (NINA COWEL ‘es Bes Sor Bathe core bee “vrG fargo os*e 00°? H itd Cant 44 4e °r adel cae £ 00°? 4 gees t TAWIL @ cae? oo" Bers Se al TSWIL # ea'g9 oore BE aoe * Tawi 264 000°0- G00°0- 100°O 000°0 * 000'0- 000'O- A- IN MAL AW 000*0- 000°0- TO0*O 100'O 000°O- §=BTorod na at DTand / WYOMSOMOTW 000° 0~ 600°0 TO00t0 NW £8/0T/8 090 °O- 000°O yoO"O =F] WaNSiNG T*o- O°Oee TtO- A T*OH~ Orcay Th Ge IN NH EO gn TtO- Oroet O° 4bG dd Tagywaned oro Oro Oro H “Co DVL B4'¢ Lee vary Had Zate £3a°9 @BWTh 265 £A/0T/0 @ANFiNa 000°0- 00040- 000°O- 000°0- HH0°0- 000°0- Tea TtO- Tt9- TO- TO- ThA~ oro ST'? OS'B8 wre? cd v00°O 600°0 c00°O 800'0O 000'oO 200°O OOOE = OFSFF o'cH Or Ace oat O° 96E oro 00'S FEO aF'S 1 fn In ng ad NW ad aA IN ng ad NW aw u Hd zawtl awIl ¢ 4aLGW J1and / MUMOCNITW LTT 4 BYNOON TT $s fA Nye bo - 904 AIBTASNaMYR =°S FTAVL £tS°S asd‘t 840°C 590°O veo'oO 1TO*O WPO°O Gort otorn nono £90°9O SaTdW¥S “co NON 669°T £E9°0 Sisto £s0%o 070°O $00'O O60°0 A9E*O BLOt eT 8SCtr SF's 960°O 4icto £TO'O 500 THLE A0UMANY ABLHOTAN-3WOL g00'0 900°0 OO'T- 200°O OTT F°4cT 05°86 preg zt BInd OOO OO°T- 1290 O'F OF'BS HED 4E°S T "SsuaW ofr) CNIH) SSOW vos f£ON vy 4 ow w yun N YN 4 (PS/79) INT TON iq ZaWwIl Tawri @ MALIN JIAND / PHYSOMITW GINNOd N3RO7 NY) NIN/EN S = 3Led ONE TANYS Y ONTUNSSY TINY TAaLISTINI YALYMONS JO 3uNTIOA CAIAVTIVAY AL) GINSVAN MAHLTA LNUINDD M314 DINHIT NO dasva SI * £8 NUF 6O - 904 @VaLASNaMud £B/GE7E TASINAY OUT Pato £6£°0 £00°O Aoi Sto IT6'O £846°0 AOO*T~ NOO'E- ANDTT~ OO'T- 2000 ort- WREKE 00's 9sa'o e16°0 OL4tE ogo FéP°O OO*T— TEO'O err “sya 'owd eDINSMIAALI-Ad~Hd Hd o4/- GALDANHOD., = 4/- OL Ore CFU/OP IT 301 N M9LIW 7a 7 WHOONDIN / NVNH/ CONNOA H3MOT ¥) NEIWZEW S = SLUM ONTIGNUS & ONTHNSSY ANY O1LISTION wILHRIOS 40 SHMIOA CAVAVTIVAG JE) CINSYAW MAHA INALMOD YSLYM QIU NO Tasea St # £8 NOP #9 ~ ON4 TIBTISYINYA LA/GzZ/Z AASTAAM orto ghao ereey 2ter 000*t 000°0- 0S4°O 000°O = 0G440 Bcare PAGIIE Q9N'h 6etey £56°0 geet 442°O TSt't tToO°O ent groryr CFO PET VALE bees: cTé*o OZHIZ(ATIS = vOS/ CATS SHN/U+H SEN/H SENZEHN YOS/OON YNZENS EN 7HD YM A UNsID tn SOTLYM NY TOW SOTLYM LNCCIMATONT + FA NEE YO - end OTATISMANHA FA/SE75 MASTAT Ofte OtT~ o't- FS4°O OTAHRH AsARER Hors OONETE HPTET OfReE OtaET ATT O'F9R O84 ZIS*O OFMOFE AtNOEe Otec? Ze ater *HOOOT O° OG o 701 OFHT (ATI +7- bOG £0N vy d ou wo “PN ‘ uN NURS SOT/U TH 7 Vv v wns MEET typ ae FEO MEE fo end matzo SHL WOMA MALO INIWI yO NIV FH 43d 3110S SSYH *e aqui S3IWKYS *2 YOs JONI. M3LHOTAM-AWIL ors ort- ose Fee O'Feh O'96S WEF ZETS Tt OCHI CAINS TaWIL TaHIL ¢ NOCVW OMDIN 7 BHI HOME T3aIVIAD IW YO ¥I¥ EH Yad 31N105 SSsuH ‘£ 2 7a¥L SOILYY L1¥S ¥3aS zere P'6ct Osta vet? a 98°s OraG FED C295 v 201M (NTWOLO THWTL TaN k OF ‘coat 29 ste? octal bree c oroT oors eETO ra eat is i) Ha TAWTL TWh FF ff he 4y yaey 266 «dONANA A ITI- 000'Tt Z-) ae Tké*e S9e'¢ 089'*¢ £90°S 6cL's 6Se°e TLGté 206%? 6L9°E wee ore 2t't 40° eK [Pee 8 | Lot bore KKK HEE SOK Ad~Hde CUNNOH H5NO1 Y) NIN/EW GS = T00'O ECBO 928°O £06°O G460'T ECO'T 982°O 2£E9°O S99°O Bert 2S0'T bitty “ity Bere Ae ROM Hoey T98e TO*te O£°S RES 4e°S Hd Q2eHIsCADIG vOS/( ATS SOlLvy NY OW Ort- OfT- Get” oO°oae aes o'9ar. raed oretTt Te o°eor s't O°S6 ere o'ne Z‘y o'odt (Ets oretEe £tbv OtPZt. OT O'Zée 304. OcHD NVWHS N/K OrTt- o* sce 0°806 . O° é4b O° cay O° Tee O° ERE O° bas O°Ses o°sos ‘OTOT CALT)S OMDIW 7 gS6°O z90°T 6v0'T gLOtY £BL°O £66'0 Toc*T ceort SG4°0 9FB8°O L£96°O at/- rrotr 400 gbort Zen FLO OoTet £v0°T e04'°0 L££EL50 206'0 QaLoqy4s0o. fag A9aACT Abate C4001 Le. G46°0 yout cat eh? Q oFEro £98°0 t/- oeeth enn: Waa EG wee OOOtTH DO £e/er7s dasitAay 978°0 #G2°0 Tva'o ST4*o GGT Bo0'T 62°90 £26°9 vyort £GG'T ate't €Ts'o 00c*o £E5°0 StN/YHN VOS/5.0N £8/b1/G WASIAAY 6z8°O £00°O £42°O ecoro eL6'°0 TRO'O Sé6°O 600'°0 £9E"T 4£00°0 £20°T 9TO*O TeZ*o ZeOtO £ES‘O 900"O 9vo't c00'O 2£ss°t vO0*o oeret €90°0 S+N/¥+H = St+N/H 0°006e O° O00rT O'OvOE O° 090T O'O9TT O°ObS O'OOTT Ooty O°0HB8 OTCRE O°ors o'0a8e Orne. 0*00r O° Oust se O°OreT O° Ore o°088 O° Orr O'Oz4t O° 00TT vos £CN NAN NNNSN O°oT O'8y T*O- TtoO- T*O~ bre Ores O'es T tH O° Ove 0°O%6 13 O° O£T O°eel Oreo O° 6 O'ée 0°48 orot o'78 O'aut Orete O° not 4 weber AONE garg OcH O27 0 OTt ey ZE9* OY £99 3E “yT vetact 660°T6 G4etoe LID*OD ore te YN7eOS 0°98 O'K9 T°O- ere orev bb OrTs bego S°OT bes B°Rs OW ewe rer? T O00 Tm CATS # $8 NUE 90 - Od WET ASMaNGd grord Givi et £58°OT YNs99 SOYLOYN LNA TEAINGS $ FE NYT 9O - HOA T1alsASYaNed 0'088 O°Ses TtO- St erEE Otay Or'o9 Oras o's? O'es O°Tst ve) MaLTV08 OMITH # £8 NUP 90 ~- 904 WAT ASYSNVd as OvTto ObT*O ay ANGth C£y79) 307 \ MBLSW OLA) 7 NYNGOMIIN 7 REEMO éteet Bdert £007 O~ 8é6L°O Z25°0 L99°O £E9°4 HELO B9z°O 40b 90 4£e9°0 ON OW 0 Oase O° oLrec O°O09T O°00rt OOENT OVOret O° Oss O° 00ers 9°0460E O° 0808 O° OnsE vVHN ALva OND VINES 8 ONTHNSSY ONY DaLITTION yaLymaGd JO aHNI0A CATV 1A 4D) TanSvgW MHHLTA LNSLNOD MaLY¥M TInt dT NO dasvg ST Zuet ‘ESTO 9S4'O £00°0- eT0°O- vTo*o~ Sisg°oO TST oT é5c* €00°0- 269° OT nTé°og YN/T9 ass Ty Tt9- 6rtt 06 a2 ge zor eae beat Zt6h N « é Te: St eve o°ce rand OtoeT eg fer O°vel Ort~ O°22e£ a0L OcHID NVR NOUN Oo ‘ea o54°0 OZ2°oT e70't £Q°G 6vo't £9°8 gzort LOL £aZ°o 98°9 £44*0 £Ss'sS Toe't os’ z90't eS £ 954°O 99S 7280 £e¢t £96°0 c6*0 +/~- 4aTh ‘2 214 g°agy Stet e*os OL Ong Gres £°3 Stet ord £°9 9'°9 6'0e £°s Svc Treat Gat £°Ov ors eer Zte 6*Se e's YN H L444 AS ‘To av GA TANES CADIS MOTM 7 Wl SHL WONA TALYTNIIVI 40 yIv £W wad aLN1OS ssun SOTLYY LWwSs Yas 00*SY 90°09? 00°09 00°09 00*Sy 00°09 08° v9 oé*Gs oc "Ov 00°0£ oc*or (NIM) Ld OL vate Zi’ Bey vary 80'S Bob T9'b Taty ofS gF°S “es Hd VL TT oT SOM TONS OH * TT AAMTH CR DAO - ad Tr SK aM TON DBAS _ ‘ft AOA TSLHOT SM AW BOTT Lat £LOF ELS ets 47°86 Zt°s fb‘ Zed Eees. EES £4" 80° Oo*e oOte Bore OOr' ELE B0'c BG°T 00°T £8°O CSWIL TAWIL BOTT LESOT EL°OT ELS ra Sr) £9 <T'8 4782 rane 4 avrg £“0rge EEtS 80'S 00% 00° 80°£ 0o*e FRE Bore ecrt oot £8°O @AaWIL TAWIL eO°TT f£EOT £E°OT LES £T°6 eg év°s £006 ra rs cee EEC? £ERS 80'S 00'v 00° Bore oor LEE 80°E BS°T oo't £B°O TaWTL TSWIL - 267 000°0 £OO"O cagro £00°O £00°O £00°O 800°O oc0'o ETO'O OT0*0O c£O'O fa) Ss 4 8 Z 8 ~ O000°0- O0OTO- O00'O- O0010- FOOTO beat) c00°o £00'°0 oTo'o 800°O 600°0 6to'o ¥toro OTo"o $60°0 IN EO0'O 9To"o $oo"o econ G00°O 9ToOro TEO'O 610°O 2000 ZTO'O GOO°O erorn 900'°O- %e0°O OCOO'O- Stor 600°O 94T°O no ad 00040 - POOrod TOO"D OOO ZOO%O TOO"O 000°0 900°O £00°O £00°O 90° NH MALaW OTA 7 WYYMOOMITIW TREE 99L'S O4b°£ £ayr 0 6b ber oO 20 4 “tT KOETT 00 664°E oo oo 8T nO. 4£ EL°SE LOO TT éT Ho'et 659°@ f 00 SO'LT vta's 61a°¢e 06 eo°Be Ostet Beas dod SOW FOS £0ON CUNTOH MAMOTT NW) oro eT o°o 4 T'0- oe T°9- TT T'O- GA o'o av vO Te T° Tt T'O0- 40 ay ro Ltt ge Nebleew © £8/TI/9 ITAMALNT Onto TIL°O 650°O #40°0 a4 ro zcO*O oro iD TARE¢) o°O ROO'O o°O ONT O- To 900°O 9°0 8O0°O a*o 800"0 9°0O zaT+o eto [Taso £°O GTO'O o O00 vo bro 4 au WAND 7 heey yu 6TT*o RHO'O GOT HO 9OT*O- 590'O vated veto ATSO enT'd gered £9G°O THe 1 = zag beh AtTS gers OG oat OrSET Ste 0'089 IN GOVE 6T 4Te'O ZEOrT o94'e S9TtS OLOCOT rHN T*0- O° bob Neve Otel ort? Qtarl ore? OrHe T'O- T0- oreo qa MALIT / $ £9 S900 TTO“O 9T0°O 90T*O- 9200 $50°0 Srora bento 480°O ZETIO Potro 0 T°O- T*Q- Qrgases O'OCe O°oGr Orne O°ebt 9°79 Own orbT O°ert o°8 Orent or'e O° Ges Orne O°eze oe o°ont Orit O°CZGET Of eeT Hd MH WYMOOMOT HW ner 90 - 904 WIATASMSNYS GLOro 2000 ETO'0 fE00 SEOrO 6z0°O £7040 azoro TEND FOTIO Evo SHINO uN z90*0 009°0 THONO ENO £00°O TO9*O £o0'd0 Latte nee) ZO0°O TOO'O TOHO*O 1O9°O H Tt*O- T*O~ 0*99£9 O°OTS o'oT? O° ode o°Ta O'OZOT O'oTor O'ltr O°nLg ad 99 500° 9TO°O 9£0°O OGTtO satro OST*O GPT°O Orta orto OrTo orreo °S Ells ‘. . ry Vasu £E*OT LENG 21'S ATs oreo LEG 00% 80°F £L°s eat £9°0 TaWTL NQALHOI3M-~-3wIl £E°OT ££°6 20° ated cy'g LEG 0°0 vate BOTT o°9 ét*b FE OT 0°90 Bote ZT°S 0'o bate 21°88 o°o go's 4T°Z 0°90 B9°b EET? o°o T9*e BOS o*o Tate oO°F o°0 O£°S 00° 0°99 £e°S Bore o'o 4£e*S oot H Hd eawil *S 378¥L To Sa 1awys *TT ¥O4 BOVMAAY 60090 O°2% 00'Sh BOTTT £L0°O OFOT OO'OF E£E*OT £50°O O'9T HO°O9 ZT £240°O O°FE 00°09 4T°8 ZOT'O O'bE CO'GH LT°Z £90°0 Of08 O0°OF EFTF TTIT*O Of9£ O8tb? B'S 289°O Ofbe OES 00% 940°O OFT OC °Ob COPE EET*O OTOZ OO'OE Hore £6790 OfB OCtOT OOrT ‘awd (NTH) 247 WA Lil CAWTL OA WL MOM PPLE Iny Teo ST ote Od VAT ASMA th AVIS Tawrl TT or A NM TOWN OO > ia! AAMT NWR Deo ~ 268 Bele UA LA SaitHus *OT MOA HOeNAAY : W3LHOLAM- AW #20°O Ort O*TOL §94OT GEtTtT B80°S6 or $659 CLEO 458°O Bok £e°e BLGtO OO0'T~ £890 ROTO hae £9'E TOOT Bao ONO'T= Seto EGE YO £10°0 Ott O'NEL O'SYB BOrSs Bors 4 wre 00't THANO a14°o OO0tT= bTTNO LAE*O £LEOrO OrT- O'OrRS aotd Bor? a eee CPE, vento »AB*o ON08T- FLTIO £7G*O avord OttH O'9T O'48bh GOO'? L9°¢ é LE°E. Ofek 66o'T N9O04T- PTE GITey zro°O Ort O'4?T O°F9S 00° Bore 9 WRK Beth SE6°O OOOtT= &EGIO BeZrt Z£99°O Ort O'rét o'c¢S cert geo S eee boty ELONT SObeE Gro ezs't £50°O g°S OrFe O°08G OS'0- GétI~- bv HORE ST's 0446°O ¥9O">” 900° T= 000° T~ 6hO°O 5°S O't- Ort B2°T~ Ebtt- £ 70°8 Pre LOTEL oF ED 499° O84°0 Zyarg 6°S O'67E O'fGG ER'E- Gere- TF Bete Ce a Scett OOO'T- ¥hZ£'oO ece't gt0°O G*t= O°60Z “O8TT EF*b~ ZT'G- TT *SYAW SIV “sSONSMAASTA-Ad Had Hd at/- WALIANHOO., 4/- oO QeH9 (AIDS CERO) IMT col OZHD (AIDS ZaWI) tawil +¢ : \ MEW ITAAD 7 WOMEN TEM / NVNW/ NN T/W OMDIN 7 : CONNOT MAMO7 ¥) NINZEW & = FLOM SNTDAWeS 8 OONTWASSO ONY NISLIATION YaLym90d 4O ANNION BHL WONG TALYINIWI yO CSTV TIeAe AL) WANNSVAW MILI J NEUNOD Malem HINT NO W4Se4 ST MIVY FW Mad 3LNINS SSH A * £8 NUP (H)40 ~ 80d WIL ssesed "2 SVL £B/GE/C AASINAY ocT*O £bo'o Hze'o 2° SOTLYY LI¥S 93S 868° 629'°0 e2B'o orto mG4*o 465°O0 NOVTE OObTET O9reT 994°0 296°O 9T'Or ceont Gettt 80's OT Tepe org'o THBYO 650°O Prado Ste O GEE Sb SOETD gTgto aTg*oO acott g0°S O°OTT B06 BOrL 4 £L4E°F. TTZ'O SET TrO'O BlOrT c45°O 08 L2OT TESTE eres o 000°0 AT4é'o eo"? 00°O? = B0°s B0°9 8 000°f TeZ°o Z£TTet POtto TTO'T etZ*0 GvGtra saret 4eT*O, 000'0 vaso vers o8°62 00°? 29° Z BEETLE oreo ELOCT c40'°0 Te6*O Bryto eTTsT 900°E 6ce*o0 WG Bb f4e°T 6e°£ oc'se 00°F Bac 9 OTO°E ZEL°0 £G0°T €20°O TLOCT cee'o0 56°OrT SO4'T FFI O ACbtT 9P6'°O oort 00°04 G4tt Garo S GOets 9cZ'o SC4*°O 600°9 914+ s9f'o Reergg wart yanrd f00°'O- = BEO'T ET*T- 00°G2 OG°O- GLtT- + OOO'T T00°0- 680 €00°O Zao T95°O T8E°B 4TH4O TRO'O £4E°O0 c00't 6Z°E~ O8'bF9 Gett- EA*E- £ 60G°T B9a°O cHL°O £FO'O ebceo TT9°O yeaect feet 45T°O E45°O cert vStE- OC°SB EBtC~ Gere- Z voort 91LE°O 6E9°0 6T0'0 O19'°O 6fb'O LO6 OF BAT e Orb Oo £8L°O O9E°T Bl'b~ O05 FE'b- “4T*Ge TT QCHIZ(ALIS wOS/CATIS S4N/¥4+H SEN/7H SEN/SHN bOS/FON YM/tOS YN79 ON/OW aN) t/- AQIN CNISOLT GAaWIL TSWIL ¢ , SOIL¥Y YYIOW : SOTLYM LNCPIVATMOA $ £8 NUE (840 - Od WISTSSMINYA = ¢S AVL ‘ OfT- OTHE 8 ObOT 49440 OF OBOE O'ObAT Orne O°es O'on~ee BI7E O°0S 0°06S fete Gettt sos of O°T-. OFORE O'SPB BEOt*T O'OP9T AOhrT Oren bier OTOSVE BOE T8243 O°ees £e°k Bord Bore & O°t~ O'R9T. OO BT4'O O°OEET 0°004 0°96 4ty O'O9GE §=B°ES Tee o*ooT 00th Bord gor? 8 O'T- O°LFT O68 PABTO OT OCET O04 age Pre O°'serc EteT otbe O'ore ev°£ 00rF L9%b é O'T- O'469T O'FPG ELST OT BEET 0° 0bS OT OLE ory O'OG0T 8tcT Ord 0°00 OL°£ OOF Bae ¢ O'T- OTST O'GLE> 99440 O'ORbT Of0OTS OE ae O'adre OFet Sot Ges Bete G“tt SiO S £°S O'EGE OF OBG BGO*T O'OonE Of OORT Oot an) greg Oty? é°oer whe OG°90- Sette + 9°64 O°t~ orT~ EOO*T O'Ob9T O° ORET O°OLT 6°Ge ONE OT TAE O°SsTse iors GI°G Gétt- £B8a- £ 4° OT49E O'4EG THT 'T OFO9TR OO NbEE Ote“t Tere OTATOG O'LeT Oreee O'eee Sok ELBE Bette Fs O'T~ 0°602 OBTT O9F°T O'09GB O'oeze Ot eee O°erT OOTEL O'F46 O°LLe O° Ord cof EET LIGHT J0L OfHD CALS +> vas £ON d OW : oHN M YN H Ha SaWLL TawTL + NU/WHS NCU OMDIW / VN NNNSN YVaLPVea OY fff she ee ZO = OA VAT ASMINUE 8 T 2 IdeL 269 FB/TT/8 NAYALNA SAM TOON DAD * MAM TOON DRO a - 400°0 TOO°O ra0°O OToO'O Tone 9TN°O o°net ore O'66 0°090E 0°0 Erk seert €00'0 <00°O To0'’o GOOD TaOra 9TO°a Ot45 O° 6h O° ave 0°o0 £9*L BO°S TON*O c00'o C00°0O bang 900° En9°n org O° 0°49 o*0 oo*e 80rd “£900°O ¥o0'o Too'’o a00°O eoatn TION" grat Or4 DOA o°0 coe O0°9 F00°0 900°O T00'O £ETO*O GOTO L£O0'°O O°eT Or O° aor o'o O4'£ 00% £90°O TIO'O €O0'O STO'd TOO'O TEOSO Ota os o'cr 0'9 8c'b SLE eto°o zTO°O TO0'°O LTO°O TOO"O HOO ore OroT O°oLor 0*0 be*b 06°0- 920°O 410°O TOO'O LEO%O £000 ERTSO Orbe O°Fb 0°006T 0°0 aT*G Géet- Trarg 020'O TIO"O brood GOO*O HNO*O~ : OTE o°oot ro- oro Goth £B°s- OF0°O £rOro Z00°O TEOO 810°0 £90°O O°ece OeeeT Oar O°GEG Q°006T 0°0 THE LESH LTS a IN ng q4 NW 34 fa) IN ag ad NH a4 H Hd cTaWIh TaWIL MALAW ITAND 7 WOMIOMITW MALIA 7 WeMOOMIEH * £O NUP (820 - 904 WISIASYSNYA °S DL BSE'EE FEATS £rT*o TET*O OTN'oO aur OFBIS TPT&O LOTTO 900°O Sroro Sais Ot MOF SOVNAAY NSLHOLAM-IWIL OTA'S FATT ¥TO°O zTO'O £00°O e“e'd ETO"O 600°9 G00°O HOO'O TTO'O O'bT CtOLT Gerttr 80° SS2°2 Loot AON 'O T=0°O §00°O eRL°O TOO “Tyg ENN'O ETO*O ZTO*°O O°OT OOcT BOTS Bord Ett ALott 000'°0 etOro TO0°0 O90tT TZO'O Zoned ZOO'O ETO*O ELO*O OFOT OO'OF BOPL B0°9 T96°S THE'S 000'0 é zN0'O Trust E£SO*O oco'o 400°O) FFO°O OFO'O O'9T 08°42 00°9 29° TABTE LEL°O 4Ge ho TAO"O REE TC TEO"O OT0°o ETO'O TPO'O TOO OBI OE'GB OO'v este abs[g TS50°O 68T'°O rooed ZEcte £90°0 ECo'O0 600*O G60°O £90°O OF0£ 00°04 GzL't GE°o Ses ot OOT*O- LATO 200°0O A0R'o £TA°0 Etro f00'O «68800 £O0°O O'0E O0'GL OG'O- Grit- 0488 £IE*O TAT*O aearo Lacte eraro 205°0 9900 O2N°O 4h0°O OF9T 08? GérTt- £B'T- Tet'9e CBbtOT £ETHO zar'o GENO 994° TEE*O EHO GOO CvO°O O'OE OC*GR LA*E- Gory Sc4*he THET?T 457° OLz°O ceT'o TSO"O TE2*b 2eT'O 800'O SEO'O 9T0°O O°b OF OS LEtb ATG “Sun 89TH (NIW) SSUW vos £0N aie] 4 AW “9 eHM M UN H C£H79 M7 ION ta CaWTL TAWIL | MaLaW Drang ¢ weManare Woyd Waly INIT WO SWoMad GLNIOs ssn QNSTANYS BOONTWASSH HAY TI ITTIOD Maeno d JO 3PNINN BML AD) WAMOSPAH MAMI TS ENAENOD LVM WENT) NO Washed GIoMIY GINNO MAMOT NY) NTIW/EH GS - FLY CVpv Ti vAY $ SA wer Cy) sO - ONd MIATASMAMHT 8 AML 270 ERLZOT I TE OOOTO OBO OOOO OOO40— OOO ODOTO= TO Th T10- T8~ THe COOTO- H90°¢ O00'O- OO0'O- OOOH Onto T8O8 180m Tro. The. Tha. 4 IN no 1 NW ad fa) IN my Td MILSW ODIN 7 Wet Dun HET 4 WeMOOMOT €¢ FA NOE CHILO - OA WVALISMaNed PELICO TOO'GE BOTTTT O64°0 L4E°0 RETO Spare TerttTT Lage TOTO T9O'D 9a0°O GETTI OOO'bE GTE'Qe 610%) Bebe Oo a 27° TEO°O =OG0°O £0 vo9tG9 OTLE vet tec 1860 TLENO a ESTtO EHO'O 920°O £1 . es) SSUW vos SON a on wo M HN H qPus9g MALAW OTA 7 MEMOOMUT CONNO8 MANO) NY) NTH/ER Go oe FPO ONT Tee GF ONTUNSSY INE LOI MaLeMoOd 40 SHA CATV TIVAG AT) QUYNSUU MOH TS Leo Yale Ie NO a3ased SI € £8 NYT CT)70 - SOA TALISMAN Sg/ce/a WaSINIy 9gn*od asTARa) HONO egno Seek 64°C LP4ND eTorr O0OtT~ BAT COOtTH ARYA FADO get Tae £4%e TEONT THOTT OOO OOGtT = ONdt T= Agana ETAd Ot'T= «JON ~AT—Hels cdooad/- TSLISMYOD. $76 oH) HALES cayaaey \ EN 4 UEMOOMI TW 7 NT CONNIOS MARMOT OF) NIWZEW Go FLED ONT Ties EF OANTHNE SG Me ISIN MaLymooy JO aMNADA CAV TTUAU JT) TEMNGPSM ACTHLTS INALNOD MALU ID) NO maseva SI € 88 NVC C970 - WA TVWTASMANYA £97607" OCASTAWM LVnrg Z£t'T O0Oo'T OOO O- caseo T9940 ggorn Ek eeyT LEE eTorTt 000°T 000° 0- erro 49040 £Le%0 OOGts BOTT GETYL Tanet NCHIZCATYS bOS/ CATS StN/V+H SEN/ZH SEN/PHN YONG ZCOM YM7EO5 YN74) wn zp Uns 1) t/- . SOTLVY yeoIOwW SOT LEM LM ATTRA TAIT # £2 NYP C940 - A OVAL ASSMAN Ee/eese VISTA OFT Otte Ort- eTO'T *ADNOT NEHER FANATT Hart Oraee O'T OT Ott~ THOTT Oroeng QrAeel OF OOLh OFOn erost mln! OTH (ATIS +/~- OS SON i 4 te} M HN Whe NIZE yw NNN NN wrerpsaeny sf PA MME CFL = OO VT ASSMAN oro 44°60 OO8RN 9°°O SE's Carol H Hed TANIL TAWIt *S S191 SAWS to YO JOVMANY TALHOTIM-aWIl O°O O'S BtveT OO'CE TZ4t4T 0°0 O°b 00°09 Té'4T T4°RT Ww (NIW) m1 “OA 1 TINIL TAWIL AHL WOMA WALvWIWI YO MIV £W M3d ALNIOS SSUW “e 214a¥1 a4 a can | SIwWMYS *c YOA JOVMANY TALHOTSM-3WTL Ort Ot OO'TE Ebt4T Ort- O*t- a6°éT cé'et OZHI (AIDS TAWIL TSWIL N VW OMOTHW 7 SHL WOMA WaLy INIT) yO MI¥ CH Yad JLNIDS Sse fe Syl SOTLYM LWWS ¥3S 94°OE BthTT OO'ET Z4°ST Ee ot OO°09 Chol C4*BT LTIW CNIW)I0. C3WIL = TAWIL ‘e 291 O'OEOT 66°F OO*EE ChAT OOLTT £6'E EH°6T COtGE H Hd Taw TJ. TAaWI 444s : “Toa, e T + i 271 Santo &TO°o agor o Gaus 9S MOA BOWMAIAG SISLHOTSM-aWT dL OPO 'TT vrate TE9% TTOta OTT? a? SOG SEOtO £E¢'°O Trea oO0'0 $OOSO caer Aga) NOTHO— OOATHO- ONOTO GOO*O £O0"O OFT AB*YD Gere L720 S 459°T beso TONO T2O*n TANTO ATATO yQero TONG TOO*O =EEO'O TUONO O'BT OC'OL 49°9 O5°S v Gere HO0t0 TL£0°0 Poor £TO°O Pn O°O TTO'O £00'O OFT*O G9O'O OE O'SOT O5°S GLE £ 68° £00'O ert 500"0 Lead S490 TEN"O SEO°O OOTTO £50°0 O'9E O'GEST aLee ontt é vor’? vEO'o vearo tez+o edt 699°O cayeo EvO*O GEO FSO°O Obs 90°06 OST oor T "Sue 8D (NIWD vos SON VW a Ay ia) PHN N WN H CFM/9) 941 TON Ail CaWIL TtaWwil ¢ MaLSy JTAND / BYMOOMIIN CONNO NAQVOT NY) NINZEN S = SLUM ONTTVHYS Y OMTHNSSE ONY T3LIST IN MALYMNOst JO SWHATIOA FHL WOMs aL INITWO YO CAWUTIVAU AT) TSMNISUAW SEIHLIG LNSLNOD MALUM DINNIT MO WaS¥a SIOMTY EW Yad ALNINS SS¥W # 79 NYE 90 - 994 NIZIASYENOR %b D1a¥1, £9/fe/7e WASIAa’yY gnn°9 Goreo L4P°O ORO'O S31dHWYS *G NOs F08NaAY W3LHOTSM-3wWit HHH £yto B99°O 929°O O90'T= OOO'T= ANDOST- GOO'O £00°0 o°t- Ort Ort Gerd £99 & HORE 446° €86°O g00°T azb°0 90T*O ere*o Z£0'O 150°O Tt O°Y¥OT O'42E £9°9 OS'S ¥ KAKKK oS*b £EL5O 9PL°O vcore OT£°oO 626°O OZT*O 590°O Te 0°98 O'cve OS'S BLtE £ REERK bees £52°0 BAL°O ovearg cAg*o '49°O OOT*O FEO'O 2°s O°Sst O8'4TE GLE os't ba HKKKE vere 6T8'0 geno OTB YT YEr oO $82°0O £40°9 £50°9O Oret O°Odt O°age Ost 00°0 t *SYAW 4397149 sSONSMSIITO-AA-Hede Hd oat/- WALISYNOI. = +/- 901 NEA CATYS C£R/9 IMT ond OZHI (ATIS eaWIL TSWIL # \ MBLSW DTAND 7 WHNCOMIIN 7 NVWN/ \OTZH ONDIN 7 GINNOA M39M07. 9) NIW/EN S = 3LYM ONE TINYS ¥ ONTWNSSY ANY T31IFTION MALYM904 AO ANNONA AHL WOMS NaLYINIWI YO CAJIWIVTLVAY AT) TENNASYAN SGHLIS LNALNOD MBLYM DINOIT NO WASYA ST MIY £W Mad aLNIOS SsvHu # £8 NOT BO - HA WIATASYSANGD fh aTa¥L EA/GE7E WASTAAM : OeT"O gear RETO éT°T SOTLYY LVS ¥3aS 000'T TOO°O~ TZett STO"O 9geet ered WORE EY EEA ER 00'0T- 949°0 TEe*Z O8teo GLts L£9°9 S Fats OZ 'O G64°0 z10°O £84" 9Gerd 00°00 AGveTT 9oret ££B"o0 soot 69°9 90°04 L9°9 06° Vd vieté Oc8°O Bent 6E0°O ge0rt cTL°0 Borst GHZrt Te9*O £TS'oO 9¥Z'O an O°sot O8°S G2L°E £ £TT*T rEEro 98 T T9T*O PoTST teg*o ago os goat L£9E°O0 évtto a84*o0 £9°E O°SET G2te ost é BIS'T vee?o errTet zoTta $B6°0O g£RAt0 GOteT Te6teT FABIC ons°O 898°0 S2°0 00°06 OST 00°0 1 OCHI/CATIS vOS/CNI)S StN/¥+H = S+M/H S4+N/UHN YOS/EON YN/NS Ns) YNZ OW YN/19 +/- LOVIN (MIWA GBaWIL TAWEL # SOTLYY UYIOn SOLLOM LNATHATOOS $ 88 NUT 90 - OOF TVET ASHaNYA 8S STAVL £8/GE/e WASTAW O't- o't- O't- 929°O OFOCRT OF One ors O°4E Orns Heo’ O°04ek T*0- T*O- O°de “Sete Shee £9°9 S Tt O'vOT O'42e SOOT O'OROT Ofser Oe res, gts 4a O'OrhT bee bee O°4t Z4'e £9°9 OS'S ¥ Te 0°98 = OFSbs FeL9O 0°06S O°orr Ore OTE git Ore 9° 4c O°6e bGte 05° BLE £ “£°S O'G4T O'4TZ AALSO OOORT Of O54 Ore Ora 6°*s Lier £°¢e oe O°Sok beth GLtEe Ogtt z OTST O'OLT O° 8G HVBO OFNORT oroT O°OrT L°4E orce4nr 2°ut z*or O° 2S werk OsttT 00°O T “01 O¢HI (ATIS +/- £0S LON Ww d OM uy ve N YN H Hd caWIL TAWIL F&F NUZWHS NN CT/H OMDTH 7 VN NNN NS 4/07 OMDTW ra A : & FA NYP BO = O04 QISTASMANU fT FL 272 000'0- 100° £00°0 BOO*O ETO*0 A 000°0- TOO'O vo0'o bro°o Z£TO°O IN MALSW 000°0- 9004 0- 000°O FO0'O £00°O £00°O SO0'O TIO" ret O 9RO°O n3 ad OIAND 7 WYMSOMOIW 000°0~- TOO*O TOO'O TOO"O ¥oO"od NH £B/1T/8 WANALN]A 900°0~ 800°0 ATO" 39040 L£LT*0 dd T*O- Orbs OE 0°ne O'SAT OORT (a) IN Tt0O- OrTT O's ort Orere no MALT / T*O- O°os o*t9 Ornate O° EG ad T'o- Orog Org Org Orty NW WER TOOM EW Mer Re O04 WTS ASIN °S DVIOL Lee Lib vS't bys bere Hd B2°L L£9°9 OS's BL°E os¢t eaWit £9°9 08°S Bate Og't Q0'o TaWrd ame * 273 4££EE'eT FTO PBA ES ryaeo c9E*O oro'o orerg 6TESE F90°O £490 On0°O 9r0'd Sass ¢e YO BOYVM3AY WALHOTAM-3WITL LOA e BTZ*T TEaatt £TE*O PL£N°O L20°O wse°o 920°0 A50°O ONO NTO*O 20090 OTE 00°09 Ehr°OT Tes v S49'ST O92°S Ore’ 9O8'O GROTTO o0T*O gta LOTTO 28T°O 990'O GEO'O T19°O Ofo OB +E OG'B aa°d £ T42°TE O8STT O89°8 9TEts ETL°O £TOO Gite L£OT+oO 949°0 000°O SET'O B24O°O Of¥T OO'FE GLre STL é Sets O94 bibs 66£°0 vOL'o £20°O £9LS 450 9900 900°O ONO ONTO OFF OO've STs GL°9 T *SuaW OTH) CNIW) SSYN v0S £0N W F | OW ee) vEIN N eM H ceM79)9M7 TWA ad TaNIL TAWIL ¢ MALAW QIANI / WeMOOMITH ‘ GINNOT NaMOT NY) NIWZEW GS = VLOM ONT IdWeS Y ONTHNSSY FINY ITGL737909 MaLemgod JO TMMION HHL 4OMA DSLYINIIW! YO CTO TLVAY AT) WAMASYIN YBHLTS LNALNOD MALYM DWINOTT NO dasvd ST YI¥ ©&W Yad 3LNIOS SSeW * £68 NOT OT - 9NA TVAIASYINYE %b FTAYL LR/PT/S WASIASY O56°4L 909°O 900°O 9v0°0 SSIS *e MO4 FOVNANY WALHOTSM-AWIL XFHKE 00°? L£96°O ez2orT NNO T= OO0'T= OOO'T- OTO°O £200°0O OrT- Ott- O*T~ ce Oot chs v 7. oor b¢B°0 2£98'°O NOO'T= ANO*T- OOOTT~ SEO'O TTOO O*t- Ott ort- 0s°8 cars £ WORE OL°9 v4B°0 T26'0 OGatZ NOO*T= NOO*T= GIT'O 829°O ES Ort- O't~ GétZ BTL é HKKKK Or'9 TT9°O TOL*O O0O'T~ O0O*T- O0O'T- O50°0 O80°0 Ort- O*t- O't- STté SL°9 T *SYaW Saw «SONSNSdATO-A~Hae Hd o«t/- HALOISNNOI, = +/~ DOL O&@HD (ATIS CEWs9)9N4 20L OfHD (AIDS caWIL TAWIL #€ NO MELOW STAND / WEOMOOMIIN 7 NUNW/ NCI“ OMDIW 7 CANNOT YAMO1 ¥) NIW/EW S = ALUM ONTTAWYS 9 ONINNSSY NY 131997109 YALYM90d JO ANNOA FHL WOMNd TALYINIWS) yO CAMMY TTVA 4D) C3YNNSVAN MASHLIA LNSLNOD M3LYM DINHIT NO 14Sva SI MI¥ FW Y3d 3LNIOS SSUN $ £9 NUF OT ~ O04 WIATASMAMYA “£ PWVL £e/vT/G TAsInay : OETO £raro erc'g 2T°T SOILYM LI¥VS ¥3S O00°T 180° 0- ZOL°O 000°O TOL*O atrro OffteT Oras PERO NObNE eoort £6°S 00°09 cabot hts v O00'T 000° 0- bcB8'o 900°O yen o Fagard 494°rT Teed GTOeT N08°S £98°O To*a O8'v£ 08'S EGS £ 900°T TOO*O- Leer?r 000°0 avert ° ° '6t*GG Orere beTro 902 rT T6'O ot Ae 4 OCO'SE Gerd Sid é O0OtT €O0°O- LBG'T 900° Z3G°T £970 R4TOE L68°S e42°0 Lba°e TOL°O g6°9 OO'bE GTed BL°9 T OCHISCATIS FOS/ (AIDS S+N/¥+H S+N7H S+N/tHN FOS/¢0N YN7~NS YN/HD UNO YN/T2 +/- ATIW (CNIWDLT GAaWIL TAWEL #¢€ SOLLYM MY TOW SOLLOM LNIWAINNA # OO NYC OT = HOA WIATASNINMYA = °E DTWVL LB/bT/S WAST AIM O'T- O'T- 9 O'T- bzZOrT 0° 0RGE O'OESE 0° 009 O'one Ores OONTE Ot ooge 0° one ort 0O'? ezhtot zeié b O°t- OfT- OT LODO OTOH O° OORT Oract Ogee NONE O'oers O°ere T°o0 oo°£ OS's E4°s £ £'S Ort- Ott- T46'O Of OBRT O° OctT grange £°9 o°oTT OOvEe Ore c'0 OL'9 Gd stté c O'T- O°T- att TOL*O O° OSGTT A’ Orors Ore O° GSE OOL0E Org wo Ove? GTts GL°9 T O01 OTH CATIS +/- bos yON + ad OW wo tHN M YN H Hd caWIL THWIL # NVRWS S14 OMIT / NN NNNN MALIAV/02 OMDIW vs hse * £8 MUP OT - AA TAT SeMaNeT fT aDTeL 274 £8/TT78 QayALN] Too'o 000°0- €00"'O) OTO'O = FO0tO L£00°O O°RET TtO- o°'set Oetot Or'Fle O° 189 0°0 O0°9 ebhot Tes 000°0- 000'0- 000°0- 000'0- 000'0- cootO- T'Q- T°O- The T*o- T'O- T*O-~ o'o O0°f O8°8 cb°L 500°O cTo'o 800°O £20°O ETO"O TOS* Ory 9°24 o°L9 O'Pag Ot O° OTOb oro OLt9 Glee STtZ 600°0 0c0'o £c0°O bIT°o Z00°O | 240° O'¢aT Ofeee O'ocr O'ORTE O'OT O° OGéT o°0 Ob? BTS BL°9 f IN. no qd NW ad A IN ny qd NH dd H Ho faWI) TAWIL MALSW OTANI / WOMOOMIIN : WALID / WOyOOMOTH € £8 NOE OT ~ 904 TTALASMAN *S 3TIYL, sams 275 000°T OK ¥00'0- 000'T 9aa't LE9'E LOLS BvOTE GEETE E09°E TOL 6£2°S STé°S 920°S OTB ‘Ee TO0°O 9gG°O c00'O z00'0 899° OcHO/(ATIS § ¥vOS/CATIS SOcLyyY YY TOW O't~ O°T- O°T- Ot O'T~ O'ste O°t- OfT- O'T~ 0°66 O'T~ 0°6S O'T- O99 “OTT OT BOT O'Tt=- O°90T 't O° HoT O'T- -O°dtt oT O°EetT att Ordtt art Otett as o°sor 904 O¢cH NU/WH/ N\ T/W ort- O° zed Ort O'T- 4 O° Za O°GST orest orete Or 2EE 0°68 ores? or ege ozae OrEdS 0'00r (AIDS OMDIN / 476°O vetit LvO't FETT TETT S90°T 9Tort 4£Tttt 660°T £G0°T 6ETT 8rs'o Or6'd 92B°0 6289 t/- EB/Ge/Ee WASIASY Peso TIZ°O S99°O GO9*O eLrZ'o £TZ'0 sTa*o 8E6'O 86L°O 408°O #9RtO0 EnO°T vorTt E9etT zz StN/HN 9 OS/E0N gst+t eater coat ogort trast Lact 0602 gaqet vsg't TH9*T £eert boat Trott 6z4't BROS oeT?O T9G°L 656 °0E LLLI9T 289° TS Tie * Be Pte Ly O45'°9¢ GEL SPE Tovey 995°Ob UOT 9oe Q00*GE GEG*O£ TaS* se 942°6 YN POs SOILYM LNDWWALNOT £a/G2/E WASTASY 926°0 TBESO GB6'0 £4E°O Sv4°O Onze °O 998°O T9E*0 £TOtT TBCtO 6S46'O £ea'o 9IOtE 00c'O 42T*T cOE"0 £96°O SOTO BO6"0 TOR‘O v24'o E90°0 eter 620°O véttt 600°O elery 800'0 20c°F G1O'O StN/¥+H = S4tH/H O'OSsT 0°046¢T O'STOT O° O4TT O°oTs O° 098 0'8se O°Sr6 o°'ode orord OroLe O° E99 O° ese O° Ov. O° Ook o's OTOL O'seg O° Ory Onde 0° o0£ O° ouE O°06E Oore O*vét eS £0N NNNNANN o'aorT O° OES Oty o°ov 0°06T orTe O° o£ 0'09T Ores o*0 o°art Orat O'le Orb Ores i) a*o4t O°cs O°OS o°oT Oro£ O° O°ob orang O98 oro o°ay Ole argo ong 0°99 4 £°1Z Tee e*et £°8 ate vib ae o's an Gt ors ere bt art re OW MALIW/08 OMOTH geod ecco 9a4ES Breto £GEE Tebr'o 4oTre £0r'O Gree 00c'O tanee 29£'0 obors #95°0 Heed 905°0 botec oTa'o Gores 9I£*O vane ferro B0r'o P4TO Leto v6Tto ToS'T £ST'°O UNE YN/OW # £8 NOC TT 0°909 O° O0SbE O'EoT O*eosT 6°29 O° OTTT T£e o'seat OPE O' Yes o°ee O°sbed e'at O°c4a StS 0°2G4 O°9T Oreos ccd oranges gttt O'vAsg 2° OroOns grat o°cRe O°oT Oecd $'OE OEE Lia) vHN $ £8 Mer TT eet £z5'0 62G°0T zovet OTR Zverat eles Leave TES*9Z GOT. 900°O BOLE BIG6T T1Z°e 9IS*y TTT'T YN/ 19 T*O- T*o- T'O- Ore OG 9 9°S £°9 ote £°ot a’t T‘9- ate Tty. vit N - 904 W1a1ASHaNva os‘ot £T6°O 00°09 betet GLY6 00° 0£ Zvort GE* 00*°0£ 9ETeT G2°8 00° OE TEET ge*e 00*0£ G90rt Gdte 00*0£ 9ToO°Tt Gerd 00°OEk Z0tet BL°¢9 00°O£ 460°T geo 00° 0£ £Sorr GL°S 00'0£ 6ett S'S 00°0L ars'o osty 00°09 Of4*O oles 3 00°09 9L8°O osté 00°09 éc98*o Zort Oc Ov t/- AIW ONT) Li ‘se 2WVL o*soc O'fo9 ec°e£ T*0S O° go? cere £'OE O'89r £e°e 9°9T O° Eer LLO°=£ £°7t O° Ice Bre BZ o*tce 09° an oréte 99° a°é 0602 aore an o'ssl oBre 9° OT OLtT £4°ES at9 O°0S oL'¢y eet o°es bo't até Or2 fY°G £°6 ors O£°S Bt éT ors 80'S YN H He S44 44 «4 "tT aay ~ 804 WaT sSyaned SOILVY LIWS AS oo*tt ooroT Os'sé 00°6 os's 00'S ogreé oord On'? 00°? os's 00° oleh 00k oo°e cSaWid OO°Tt ootot OS*é 00°S ose 00*8 OSs 00% Ost? 00'9 oss 00°S 00% oore 00°s caWil 00°OT 0S°6 00°64 os*s 00°8 ost 90°% os? 00°? O5°S oo°s 00° O0°£ 00'd £“0°T TAaRIL 00°oT 05°s 00°6 on*8 00's osté 002 on? 00'9 oss 00°S 00*t core 00re EET TaWLL mf ot £T ra YT oT MANN TL ORG OH * gt ames wt te «4 ~ AMM TOWN OAO “ 276 BOL eT "9 Tegra 970°0O 90E° OOT*O- = $4090 ETO*O OO°%T= O2O°0 HOT*O~ ETOO P00"O ATO'D LE0°O ONTO TON ETON OLO'O EGOTO TENSO Z££0°O 980'0 29040 KEOO GeOed £60°O O9T*O £40°O orotd rend BrO'O O4THO EETLO TAA OTeO — ZANO Pant oT O1Eetd ARH +O AeOTO BEOrO RPO O&O LvT'O broreT etere GLO 4onro aad OfOO Ovrtd LETIO LAGS ALE NO LOOTO gyotd beord OrO'O aGTeO ZOT*O £OTTS 000°O STO*O 407 990° HEO°O O47°0 O9T*O 29L'°8 é2T°O LO0'O rao" Zbord eTO"O OREO OFT*O o80's HLT°O B00'0 20T'G= GZLOrd £0090 0600 OFTO ee2’s LEE*O 5900°0 9aQrO e90°O c00'S OnE'O LeT*O bGsot eased Cov 'O 290*0 gga" 999*0 TON'O OTE'D LITO baat aT vast. cvE‘O ETO"O CATO egcte £S0°O TeT*o £00'O OTE'O 45T"O *SUaN STH) SSuW bos £0N 49 4 au 43 YHN N UN H C£W/9)0 9M MALaW JTAND 7 CINAOd MAMO1 NY) NIW/EW S = FLYM SHL CATE IEBAG MIV $# C8 Mer TT - O04 WUALASMINOd tb EH/GE/E TASIAAY EOeO 908 'O 9CBrE 905°O $3 aq . soon o rosa TS Ae AN at eocoeoco395 * O°ee Oecd O*eb 0O'OsS O°ee ON WHYS §Gt MOd OMAAY WALHOLAM~3WIL 00'n? oo°oT GT 000k 05°64 eT 00°OE 00°6 £1 90° OL Oo8°8 et 00° Or 00'S tt 00° OL og* OT OOrO£ OO°L 6 oor os OG? 9g 00°0£ Qo"? Z 00° O£ Ones 9 00°08 00's s 00°09 00° Ud 00°09 00°k £ 00°99 00°c é Of 'OF £ert t (NIK) La TAWIL TAWIL - Woyd daLvimno wa] yo £W Yo avy aan’o £T4*°0 ONO*T~ OOO'T= OOO'T~ OO*T- 9Z0'0 O't- o*T- bet ecget NNO*T O00'T- 0460°0 OTO"O Z20°0 Ort ort 400°T 4yort O00'T= OTz*O Q00'T- OfO'd ESO*O ort- o°ere orret onrTt 900*T ONO'TH NOOTT= OBOTO £9080 Ort- Ort- LO0°F TEpeT NOOTT ELbtO OB88O O9T*O £4040 ort~ 0°64 FEOrT 9fLo Evo tO O6T'O LETTO ort- O°4S BL4°O 9Lb tO OL0°T Ofe*O LeT*O Ort- 0°99 490°T Srz+0 PL9tT Orso LET*O ort O'vot 450°T o4Z*o 9N9ts Onto ZOT*O Ort- O'90T GLE TROtT 228° Tees O42°O OFT'O ott o*soT bb's O40°T £B40 LO? OfT*O OFEO O°t- Orgtt ORK Té68°O £946'O0 O99t9 O4E°O OFTTO ort O°erT OOK 6ra'o £G0°T gers ONf°O LeT*O att O'dTT MHHEK 808° O OP0°T PAerG OTE*O LOTTO grt Otett OEE 4bs°o 84650 B94°L oTe’O 40T°0 ere o*cot “Suan 'I1) OM CATS (LH 79 IMT SVAN) 7 WOMOOMIIN 7 at? NLOIYMOD, 4 7— sHONAMASATI- Ad ~Had GUINNOA MIMOT Y) NTNZEW ¢ IeTeAY AL) *# £8 NOC TT -~ 80a VET sSMaNNT Q2HD (fn \NOVW ODI / 4 2LiN 0S SssuH SANUS “ST YO4 39bNanY W3LHOLAM-AWTL Ort OO*TT OO'OT &T O'cBe OO'OT 05°6 bt OtT~ OG'6 00°S gt Ort 00'S os's cr O°'zsT OS's oo0°s Tt O'sst 00°8 OS's oT O'est O82 o0°% 6. Ofte 00° ont? a O'SLE OS'9 00°? é O'468E 00°9 OS'S 9 O'es9 OGG 00°S G O'ss9 O0°S 00% v O°T459 00% OOrk £ O°'E4S OOF oor é O°00r OO'c Lert T DS cOWTL TSHIL + = UVM ONTIAWUS WY OONTHNGSY ONY TSLISTIOD MILYMOOd AO TMATIOA BH WON WaLVINITWO YO : i NOD MaLumM Tia NO Wasva ST MI’ CW N34 3LN 0S ssuu “f FIL 277 EQ/TI/Z9 IMAM NOO%O- OOOO ANOTO- AOOtN OOOO nnAO- Tt9- T*a~ O°0 oo°oT OOOO. (O4040~ CANTO OOOtO- OAOtO- OnarO- TtaQ- T*O- o*0 08'S CODD ANDO QnA E.- OOOH. BNnOtNnH Ten Teo 0°O 00°S fonre TON Tong TOQ te APorn O°OT aga? ora oss OOOO ONTO annen. DADO OBO Tae Tto- 0°90 00's OOD Am OOAEO MONI ONE HANTS COO Om Tem T8an Oro (O8tS DONO. OO0tD- Qant on OAT Mm OOO CON Ae Tto- The oro oor ponte rOOtO TOHtO bron TAnA reaey ra | NTA orn egre OOOO OOOO! CONN GOOF BANA OCB em THN Tth~ Qr9 Core RAeO POOrO FOO TOGO Taaeg OMAaAT MOLaT aro O8°6 Onero= aon DOOD NOOO Onto Ona ge The Ttae 9°0 00's £O0'O LOO" LONE PEON TOON TéQto ore OreTy 9°90 Oot boot TONED OOO boon TANTO o4ary oer Orang o°9 OOrf OOO roeta KONO THEO TAOTO yEeqro Ar’ S*ib Oro 00%e COOtO LOO" - BPQaNtH biraen FOOtO hema) Orr OLE ord Leer 4 IN np a4 N ad fa) TN na 4 NY 34 H Hd EaWIL TSWIL MEW OLSND 7 HeMgoMo ti MATT 4 Weorearnyiey Tot FER NOP TT © Od TRIAS INeY 8 aa TOON DH m mite = 278 £O/GC/e WASIAAM 990°O Nbe'o vooet 9£9°9O SANUS *4 MOA 294RA9 NWALHOTAM-AWTt L£5°E Toro ToET GG0°O £470°0 Ort~ orase O'TTZ O8°6 OO'L £2 Bo" LLENO Seatt canard T9O°O OrT- O°'BeT O°495 00° £e'¢ 9 Boer ONOTT~ FARES OTT*O £49°O Ort- Ort- O°TTZ £8%v 00's S Aor Ee SELt0 O29°T OTT*O 990°0 Ort O°4ec O'Z4TS 00° oo'0 Ld ae & Go SITs ASOr'O OFOtO ort- O'BTE *OvTT 00° OO°'e- £ IRs NOOtT= PTESO 4N9°O O20°O Ort- Ort "O60T 00°@- 49°E- SEES L420 OLOT O0O'T~ Z2T0°0O O°T- O°TZS *O86E GL°E- Eétv- TT “SvawW S91) «FINAMASATO-Ad-Hae Hd oet/- NALOANMOND, elaen NZHD (AIDS C£EH/9)3M7 90L OCH (ADDS caWIL TSWiL & N OMALAW UTAND 7 WYMOOMITIN 7 NUH \ CVN OMDIN 7 CONNOT MAMO Y) NIW/EW S = FLY ONT IWS & ANTWNGSY NY N3LITIIOD MALYM9Od JO AWNTIOA AHL HONS ALY INIT yO CPWMO TION AT) TOYNSVGW MGHLIN LNALNOD NaLUM HINOIT NO d3svd GI YI¥ EW Yad 3LNIOS SSYN $ £8 NYC Tt - 904 WIALISMANYA *£ a14y1 FO/GE/E WASINAM OzT*O £ro'o CLT*O 2Tet SOTLYY LIS ¥3aS BOLE e9e°T 469°O ELT*O beso THEO CPL Gh STTe oeGg"0 000°O TTO*T s2°s Orost 08°6 00° 4 TEA TE vaTeT 92£6°O ROTO son 0 gonrd N4E*TO enert £TT*0 Braet 4c0'T T6°S COLT OOr4 £B'e 9 EEK 906° O T446°O FET*O ALBTO ZTe*O bBYATIO tees LEO éT8tg S£0°T Tee OaeT Late 00°s S ALEtE LBLSO vROST 4ET*O L£046°O Bo4'o T4G°S8L Trees Tce°o Obe'Y 986°0 O0°T Ofoct O0°s 00°0 4 SBGtE B48'0 ecneT S40°O TE6*O WOL TD EGETO Otay Bee oO zeart 9490 O0°'T- O'0ET 00°0 Oo'd~ £ OEE Zeo'o 122° GET EN 445 '°OQ ELS*O bGStoe acert Tee*o 4oG°s O4E°T vBtc- E°NOT OO'E- LZP'E- FZ 41E°S BAO SAGO O£T°O eGbeo Lot aan EL4Gtb L£4646°9 Tez°0 stats Serer bete- Oc°0L GLIE- E6*b- TT OCHIZCATIS eOS/CNT)S GEN/¥4H SEM/H SEN/YHN YOS/CON YN7EOG YNZ YNZ YN/ 19 +/- LATIN (MINDED COWELL TAWIL £ SOTLYY ME TOW SOLLYM INATATANA & LA NYC CT - OOS DISTISMIMYA fc DY L SO/GR/E asin OT OfShE OffTe TAs O°GéER oro oro 6°ET O° OOTT Beye OSE O8°6 008! £ O'T~ O° MbT OF49G BeOrF O° Oger O°be arag ste OtaneT S°St 99°e 00°" £B¢e 9 Ott Ort~ OTTTL bFO8T %°*SGo O°ag orna ag NO1ET Tet £G'L LHe oor G Ol OF42F O'ZTS PRATO O'HOOT NF EA OenoT be O°orzTe FEE B°9T £S°£ 00% 0070 + O'T= OFBTE FOFTT 940 NOrat OF apr arn £°4T N°OdTe E'O~ vet C 4 Z4E°C 00°O OO'S— Ff O'T. O'T~ “O40T O68 °F orate OVret One aradae T°O- Te OrETs 4O°£ COE 49° OTe OTL 8ORAT GOEST orange o°eT? O°LZTC O'greg TO- O'ESE O'OT9T = 424*T GLE- Thtv- TF DO. OSI CATIG tie bas SON Ww d ow ao vHN N ON H Hd cAWIL (TAWIL € NVU/HWZ NIZE On) NN NONONN MULT V/Oa ONT S444 * et ~ 9O4 WISTASMSNYd 8 T SMYL 279 500°O 900°0 TOO'O O£0°O T00*0 8£0'O 600°O vto°o f00°O BT0°O TOQO'O 260°O £00°O 600°0 TOO"O £v0'O T00"O ££0°O 800°O 800'0 too0'Oo 098°O c00°O ¥90°O v00'0 200'O TOO"O £40°0 £00°0 c80°O TOO"O 100°O ©00°O 9T0°0 TO0*O 820°0 100°O v00'O zt0°o 450°O £90°O STT*o a IN no add NW a4 YILaW BTN / WeHOOYNIIW L£2e eT OSES BL4°Es £510 TOT*O oTo*0 LENS O46°S 2vert 000'°O 000'0 40070 L65%L 9LO°£ veett 2£20°O £90°O £00°0 940°LT O60'°8 Lovie SéT'o L9T'O £00°O -BOTT6E OL4°9 OcBt>9 veE'o oFETO £00°O TZ9°*SF TZ0°L 9Lb°9 bOt'o 000'°O E100 9oe tt BaZL'T £90°O 620 '°O c00°0O 804°S v9G's bee'o TOT’O Sro'o SSOW vos £0N £B/TT/B WAMALNS OW o'ss8 O'9oT Ort? O's8e Orne ort? 6£T°O etTo Evoro 020°0 980'°0 £4rso 650°0 vOL90 92 WHUMGOMITH ANT TdwS Y aiid 5 Ot 0°90T O's6st £°22 O°9d oratt O'Gs& OreTe IN 4gort 909°T eBL°£ 3 aed forty Q£e°0 4e2°T YHN O°*st O°ers Qr'9E O'ete O'd Oks Orct O° Sted O've O°eeeT Oredt O'eraal OrLTe O° Oree na ad O°at O°9e? Ofer O'0L9 O's O°coEe ovat O*bas O°é6e O'Oert o°eTt O° 080k OFS 0°0489 NW ad MALI / WOMPOMOTH $ £8 NYP CT c90°O evoro O0T*'O~ TEO"O orord Ofo0*0d £gG0°O0 £EOrO 960°O £eOrO OOTtO- = GEOr0 COT O~ TOO OOTTO~ GTT*O N UN 904 W131 4SYaN0d £z20°O 020°0 bTO'O GhOrO £c0°0 6100 2800 T90°O eLO'O = OTEHO LrO°0 ELO*O OTTO 090°O GE0*O gG9"O OvOtO £00°O =4600'°O O20°O BEOtO OO'T~ LTOtO 'S “SYUAW Sov H (2W/9) 987 ONIWASSY ONY WAIL VIO MaLymgo4d 40 SWNI0A AHL Woy # es ner ey SHLIA LNSNOD YOLeM AMT NO IPAS OO) OW ATASMANBH Si uIv “t acre 99° E £G°E £G°E£ Z£°E£ 60°£ 62°S * + eoooecoso > * xz eoocooos Hed aay S3A1dWes S9¥MaANY O'0e O°OST O*ob Cort O°or B°49T O'9e O°0cT Orbe O'ocT O°OT €°o0T O°F 08°04 «NIW) wa 4d HOM TaLYINIWI YO- £W M34 FLNIOs SAW OS*6 00°2 £B*b 00°e 00*0 00*e~ Sé*e~ caWIL *£ NOH 00% £8'y oore 00°0 00tc- 2£9*E~ E6Se- TAWIh WALHOTaM-3WIL 086 00° £a'y 00% 00°0 00°E- cawis SSYN oor £Ate¢ oor' 00°0 00°E~ LOtE- C6 fe TaWTL AM TION, AOI T LION = 280 €8/81/5 WeaSIAas beso EOE STTSO SAWNYS §6 MO AIAN TALHOT3M-3WIL BETES ERGO NEOTT NOOTT= OHO'T= OO*T- STOO Ort~ O't- O't- ££e'TT cerot & ; es O£46'°0 é9z*o £990 OYO'O BPO°O o'r O'Tre O°aTS chor 4°86 8 Ate L940 rZb*O gNTeO £90°O £°e O'sst O°eE £9°S LIL é chee FSBO B4T'O ETO OOT*O tT oss O°OT? L£9%% 00’! 9 ogte LPBea 6L9°0 SoTtO £TTO Te Oot Of109 ZT°9 hed g Tote TL45°O OZT°O 980'0 ete O°4Bt 0°008 Oftb £T°f y Ost OAT*O 4OT'O ere Q*7ce O'sTB BO°E 8O°s £ Zor G2T°O 9800 ert O't- O'érS Bors £B°O é OL ie OTT*O 41rO°O eT Ort- O*ab? OS*O as'o- TT ‘awa . Hd atf~ L O2H9 CAINS CLN/99M1 JOL OcCHI (AIDS cBWIL TAWIL #¢ MAJAW OFAN 7 WYMYOQOMITW 7 NVWH/ NVM ONDIN 7 GINNOA NAMOT Y> NINE 7 : IDLIWS 8 ONTHNGSY CNY ALIS TIOD MALYMo0d JO AWNTIOA AHL HONS TSL91NIWS YO NSVAW MHL TA LNCHNOD MALYM EIT NO Jase ST YI¥ EW YSd 3LNIOS SSUW $ £8 nur ft ~ 904 WaT ISYaNed “f AWYL £E/8T/S WASIANAY geod Reco év'tT SOTLYY LIS 94S 000°T TOO" O- TET*O bond gFageo ZEL°*EL BTS Ob 486°C EST*S avort SB°OT OF°FS EESTT Ebtot 6 6b ts oor T £PE MO TEAto pagro Sbers? GELIP 400°T 906° OL46°O 43°64 E°ooT eetot 84°98 8 TL9°G ZOr tT ZET°O veh*o 4940 NOO'OR Leatey £22°0 90% £96'°0 ame.) 00°09 £9°8 LOE Z T460°Tt Ore se 4rO°O TSt*t 995 '0 bt'ése, T4909? 9aZ*o 4G9°ET E£5ato bed Z°O0e L9°S oor 9 OTT vOG'S S£0'0 40E°T GT9*O ooo'o? =ETBTT EPEO Gets 2vB8'O BEtS Or°Gé LT°9? BS°b Ss ELE Te FObTE 820°O arort Brord GLB? ERGIE Z4e'o 468°T T24°O vOre O8'42 O8'+ 2eE b 6£48°2 400°E ELOtO GL6°0 saeco resis’ TSe°O0 EOL te a7o't asc 00°09 BOE B0°S £ KKK BELG rf oro'o £18"oO N0n4O £ eaatb Tearo Savrk TST*T See OO'Ss£ B0te £8°0 z ROK 00L%E OT6"0 bE0°O gas" 90490 086° voo'h 2£65°0 000°O eLo't vO'0- OB? O5°0 es'o- OEHIS(AIDNS bvOS/CATOS S4N/U+H S4+N/H S4N/¥HN FOS7EON YN7POS YN YD YN7OW YN/19 +/- A0IW CNIWDA0 GAWIL T3WIL # SOTLYM YS IOW SOTLYM LNAIWAENOSG . # £8 NUT ET - 904 WISTASNIMYA te BMVL £B/8T/S WASIAAY Ort Of'T- Brot OfOObZ O'OBTT O'OZT Orore 0°26 O'SrGE O'B80ZT O°Sb O°L£e O'lby GE'S ELTE eEvtot 6 Ory O'Tes OF6'O O°ObE O°oTe O*re O°bs 8° tt Bad O'9C4 bee é°tt O° coe eGt£ cetot S2ta 8 gts O° ast L£96°O O°009 O°Oone o°aT O°Bbe Bs oer Tt? Pars O°etT G6*£ 29°83 LOE Z Bt O°as O°oT? FEB 'O O'00S5 o° O°aT Ott T*T 0'9428 oT bet o'se ebte 29s 00rd 9 Te O'OrT OFTO9 ZOH'O O°O8b ct ordt Ortg re O°L¥4 4°S o°a B°Le 9S'b LT? eS't S sce O'48T O°00B8 TE49O O06 Or one O°Rt oreo Tee O'ves 9d 6°Z S'be Tete OS'b Z2T°E > ere O'42e OTETH BFO°T OOS O'oZe Oo£ 9°02 \ ane O° én8 aes £°9 0°0E OL tr BO" 80°C £ Aer Ort~ O°ZeS TSE'T O° 00% o°00d O°4t 9°08 ee O° RAP £°6 S's Ove coe BOTS £8°O é 8°T O'T~ | O'MbF THOTT OOH O°Obs 9'O O°Be B°g OBES ae ars edt 92° OG°0 eS°O- TF 204 OfHD (ATIS thn bas £ON WW Fe] OW PHN N YN H Hd TaWIL Tawi # NVWRHS N CIZH ONDTW 7 NNN NUNN MALTW/OF OMOTH C4s4s¢le $ £8 Mer ET - Od WIAIASNANYH °T S1d¥L 281 O00 O~ EON OD 00°O 600°O S00'o ¥OOr0 900°O A OO0°0- 000° O- €00°O 100° boo'o ZO0°O 900°O €00°O POO" 600°0 BOOTO vOOrO 900°O 700°O bO0°O ¥00°O £00°O c00°O IN no 000°0- 000° O- OEY Tone FOCO TOO°O 6200 400%O F8/TT/9 MWANALNG GTO"'O TOOTO £E0°O ONE T49°O TQ0°O RPO'D 600°O 000°O Z£TO4O £TO*O 900°O broro0 Hel MALAW JIAND 7 WYMOONIIW Teee'’ vEOre CLEC yBO*O ro EE bos CINNOA MAMOT NY) NTH/EN G = (PVMVITAY 6'E Beet LOTT 8800 Tév't ZTO'T BYO'O o8ats PELIT €90°O OBBTE £B0%S Z££0°O T25°£ seere £400 90btte Ch ae 040°0 POO & batey T6T'O OPES. OLT°E BTT*O _vEGTS L£9°T 900°O Ww ad TIT°O £00°O esoro Zt0°o veo'o POO O T40°O 490010 GEO'o fO08O OST*O vOO'd £Oa°O 7000 S£0°0 hOO°O £9040 £00°0 650°O 8009 4 MaLaW OTaND / LYM 000*0- Tene Orn Orne Ore 0° OF O°ys Orde of O° (a) TOT"O EGY'O £9OrO 8999 Qc0'O Brad 890°O 989°O PAD tO TOT*O 49 HESOOMD TW Gt ab Tae O° By rere IN es2't oG¢'°O AA" £9e'T Obert bIor? anes ecBte 4part L£9E'T HN rEOrO 920°9 890°O &rorg B00°O ecard toO"o0 egard r90°O FLeoro N Se OMe WIT] INBLNOD MILOM WIT NO aeed SI yIe * £98 TtO- Tho TtO~ OrTtg o°ot O° ZO5 O°ere o°s ons Orebe ora O° £0, bss O'S Otte O'ToT ors O°Oer Sty Ore O°'SZE ores oT O94 Otett ore O'OTs ad NW ad WOMOMI TY NeC LT - 904 MUHTASM ING ¢°S 9co°o 990°0 6TT*O TT0*O £00°O O0°T- STOO TIO°0 ETO*O = OFOO BFOTO £19°O TIO'O §=HOTTO £9049 ¥o00°0 590°0 OET*O NOT'O 420°O 6900 =GGTtO £IT"O TEOTO 7900 O4T*O 080'O PEO"0 £00'O O8T*O ZOT*O z20°O vO0'O GZT*O 080°0 6eo*o0 290*O OTT*O 4bO°0 "SuaW f° 9149 UN .H C£W/9) 997 auma9gs JO 3vNq0N FHL Nur £T ~ 90d WTITASMEMOd tb gooooe seo °0 Sere *0 Pe 3 °0 84° ‘0 Zeib ‘Oo dd ‘0 T9*%b ‘0 O4*b *0 co'b 0 GLIe H Hd a1avL sald S9¥ManY Otb O9' eS O'be ZtOOT O'0E 90°09 0°00 Oc *0Ob O'vS Ov'GS O°ee 08° 462 O°ze 00°09 O°O£ 00'GS O°9T O8'b? CNIW) wh 4 YONA TALvINIWI MO ch Yad SiN10S Sssey SAVE £“2°TT shot cr or 82°68 29°8 29°% L942 OOr!e at°? ast Og't+ £0°e BOE B0te 80'S £8°o 05°90 as*o- GaWIL TAWIL *6 401 . THLHST3M-3HT Lh enttt ceor cr*ot S2°8 2998 £98 LOL OO0°e LT9 eStv OS*b 2T°s 80'k 80° B0te £6°0O onto B5°0- caWIi = TAWIL RAM TOWN OH *- HOM TOON OD * 282 £e9/GE/S WASIASM Z£Z9°OT = 909'°0 ceeer £90°O0 Sa1dW¥S *B NOs 329bNaAY WALHOTAM-3WIL ve eTh°O 684°0 O00*T- 900'T- 4600'°0 TO9*O GO0'O O't- ort O'bse GL'OT ch¥té 8 Lee SEBt0 TO6t0 Tater eas°o zTe°oO SSO*O GbO'O $e O'Gee O'8rE Eb'sd LL! é oa*e 8046'°0 026°O B946'Y ¥9L°O gag't 060'0 ££O'O ob O'feee O°ObS LE°s 80°S 9 Tes 6£4'0 ooo'tT c46°ce O1G'oO 9e9'T S400 Ov0'O etc O'62T O'SkS 80'S 80°? S cere Te6*0 6946°O vous T4Z°O OS9'T 060°9O BG0°0 arZ O'sse O'L45 00°C Z£T°O~- + B6°S £L6°O Z4E0'T 9bE'S volo overt O£0°0 6b0'O 6*tt O'voe O'SEL OOTT- Eé°~E~- £ ee°e Tere ¥66°0 g420°T 8b2°8 Tacro gac'T Sbo'o B£oro ergot O'ofy 'O¥ct OOTE~ Eb tb Z £L°S 69S £46°0 650°T vR6IE eve*o vont O20" TELONO 9°9T O°94G *O0TS Eete- OG'G- TT "SUAN * 3109 »FINAMAAATON~-Ad—Heae Hd at/~- WALISNNOI, = +/- OZHID (AIDS CEN 79) 97 901 O¢ZHI (AIDS ecaWIL TAWIL +F acangd / WHYOOMIIW 7 NVWH/ NOV OMDIN / CINNOA M3NO1 YD NIWZEN GS = ALYM ONT TINOS & ONTHASSY TINY AALITZVIOD YaLem904 JO SWNIOA FHL WOYA TaLYINIWI yO CAWVTTVAY 4D) USNNSVAW NSHLIA LNALNOD NSM OIMr 1 NO Waseda SI MI¥ €W Nad BANOS ssuw $ £8 NUE br - 904 WaTASNaNHd ‘eg JL £O/S2/e WASTARY orTro g£rord e¢c'o rane SOTLYM LIvS 3S FORK IK O8E'O GL8°0 £LL*O zrs'o EV9*°O 60T°?S Lette EGET LTT? bB6'O GO°OT O8°6£ GSZ*OT ch°sS 8 The*t BEL oO 6ETtT S8E*O ¥seBro g0z2°0 Cebc4S 4nr'e S09°0 BéT*t z06'°0 8f°8 y'set ced cee é 806'T ves 'oO | est'tT o00'Tt GE9*O Té6ort 28E°0 G2Z*Tt 0246°0 oc’? o*acel reté g0°s 9 68H6°S% £82°0 vLort rand 99°feT Ooze 18c°0 oG2°8 o0o't SS°e o°08st 80°S 80° g 9GSTT TSE9*O PBO'T ¥9B°0 EL°OLT LeEOre 9EL*O 2Ore'd 686°0 c4'0 c*oLrt oote éT'O- 4T0°S 88S °O ELOtT 9GLIO 6Fk'BB CTAty oT9*0 208°C ZTort 94*T- EtGIT OO°T- E6te- E£ vuB Ss 60L°O ££6°O GL9°9 90b'Eo won? ELL°O roe" G2ort T2Z°k- Of °SGB OOTE- hte OE UO’ £14*0 e860 e690 neeree “4tas eoort Evert 4gort 94°*v- O8' V9 Tete OS'G~ TF OZHIZACATIS vAS/CAI)S SUN/¥+H GHN/H SHN/HHN 6OS/6ON UN/bOS = YN7H YN/OW YNSTD +/- ATH (NIWA ZAWTL TSWIL ¢ SOTLYY NWIOW . SOTLON LACTIVAT NOS $ £8 NUC bt - Od DISTASMONYd = ¢e FTL £8/Se/e WASIAAY 6H6°O O'OOTE OOSET O'Gb orecet 9°O¢8T O'OSTT 6° GLO Ebrd 8 E06'O O'0946 araay O08 O'oeT O°OOrT z £L0°L certs REL Z : O46°O OFOHET 07008 O'Aot O° ORT 9° OnE OrFlEe OGTE “LLL 80'S 9 Ot elt HOOT OF OOET O°008 Ores Oreet OOLST O° 4BE Terk BOrG Bore g O'Lae 48690 O'N7ZT O° 096 ornet OOF oroge? O°END Berk OOS Z£T°O~ O'ePEe Z10°T O'O0SE OfOGST O° ogT ON9OE O°OcTT Gé6tE OO'T~ hte EF O'ore BLOT O'OOGE O00¥E 06 O° OnsE o°ocs Tate OO'L- Tete E OOS 9009 OFOOCT One O° OL Ae O°OvOE § 49'S Cote OGG TT SOL OEH) | CATIS t/~ bos £ON d ¥vHN M UN H Hed CaWIL THWIL + NU/AWW/S NN CI/W OMDTM Z V NNN NN thf ses $ C8 MOP bT - aod A Va1lseMeD “ET ay. 283 000° poo® 600° 400° OTo* vTo* TTO® 900° n £8/tt/@ DayaLNy O- G00°0- 000°O- O00'O- 000°0- 000°O- 0 80040 TOO"O 620°O T00"O 9E0°O ) 9TOro TOO*O bEOrO TOO0*O e£L0*O ) té0°O TOO'O £Z0°O T00°O GEaro oO £T0°O TO0*O BrO"O TOO"O ££0°O (¢) 610'o v00°O 8L0*0 £00'O 040°O 0 9EO"O £00'O TZ0°O ¥00°0 ceyeo 0 B00°0O c00'o ¥LO'O £O0°O vIc°o IN no NW a4 MALAWN JLand / WeNgOHITIW TGb’LT 8bSS OOk e ETLE*O Ttz°0 800°0 6L8e'O TOT'O v80°O z00°O TOO"O Goole oy Ss Sree 6£0'0 400°9 vOrtoT Ofs's vovie PESO 8000 Téa tet coc’ e1s'v £8c°O Of" O £00'0 EGELY KOPTL EG2°5 6TE°O bec* 80070 FAC OE OOb'B LeL[ >? eZe'o Sef °O GTO*0 Zeertar o9eeeé 969'"9 LPO recto £700 280 OT 9Thte ‘OOT*E BIE'O vIT'o KTO*O SOV vos £ON aM) | OW : . AGLAW ODI 7 CONNGH YAMOT NY) NIW/EH G = ALUM ONTVWUS Y ONTNNGSY FING WGL9T 09 YaLvmaod JO ANNIOA AHL HONS WALYINITWI YO fH Mad GLNIOS SSuH T*O- ores o-vor Ord O'rTt o°noe 9° 96% O°Pee A £0T°O etoro o8o°o svo°d0 Troro Noord nated ger eo vtz°o yg WOOO ITH T*O- O° oet orcet O'ece O° 9RT O'LLe O°ecRS O*oRe IN O4LTE veord sant OLnrk ogg’ £08k GRE cake “ert PHN $ £0 Ta0°0 TOO'O Tad £90'O E5000 4g0+0 z9n*d AIO'O gEord M Tto- 0°6tS O'ede O'oRE O°eES O°séZot orzaat O*vodT ad WYMOOMD Nur ot 8T0°0 gard T°0- o'er 0°4 oro o°8 O°tb O°sé 0°42 NH mW 020°0 Tro 40 T H T°O- O'cie O°zcLe O°S9e O°9GE O°0aeT 0°004E *O0Z0T a4 ~ 904 WTAI4SHaNYA ¢°S £90°O TAN"O GO0°O SoO'O GhOto O40°O LEO0"°O u60'O OF0°O O4N'O BCO*O OL0°O Bbord B£0°O 0 TLO°O *SugW SWI CFW79)9N7 (ATV 1eny 40) TaNNSV3W YGHLTS LNALNOD YMaLym dint NO Waseda ST yy $ £4 NOP oT ~ 904 1431: *b 7 ee ++ ooscooscoo xz eococoocco eS aVvt $4 a9 Os O°Be Orce Ofer O°ee o°be Orot 0°01 OA 2VvvL vbte Ente£ OStf Tere ezre g6°e tec 69°E Hd T4WYS MING 0864 v'set ‘set o*oet e*ost erate oe tne Outre? CNTW) ud scot ceed £0°“4 80°S oor 00tT- 0ort~ cern eantl ‘8 MOA cveé £EeS 80°S g0°s £T*O~ cé*e~ chee OS*s- TAWIL ALHOTAM-AWE L S4°OT evtsé feed Gor O0rs oor tT 0O'E~ ériye cAWIL zed fete go's Bor? £080- B6te~ O5'S- Tanti AAMT HVWVAG * Sam THON - 284 250°OT = 4c8°0 £LS TE TAT SO Satdivs ‘ot 404d JOVNaAY G3LH9T3M-aWI Oot SALT NNNET- NNATT- O0OOTT- GOO'O ZN0°D Ort- Ort- ort- oSs*tt os'ot of ont STOTT LAO ONO'T- OTE*O rcs°o £CO'O beOtO Oort~ OtrOE FO9TT OF*OT GL°S 6 9OrE chara 096°O BIL? £92°0 ects PANO ON90°D 9 O'42ZE O'0GL GLtS6 S2°a 8 COE C6047 4O0'T Ede°at § £TOvE OrE'O 1T60°O Bb O'eTT Offch GLa OS's ra OF’ 9A0°T 470°R eoT'a grat GTS OLT°O LOTTO St O°'TSt O'eae Gerd 00°? 9 Soe 6bT¢t OATT Nebo et cacet TA6'S OSLE°O otto ce O'NZT O'cOB 00°? Scr S Thy 650°T Zat*t vALS? Eveo (on: he Ovo OZT°O B'f O°S6T O'TEP Serr S2°c ¥ en't PALO een'o NPTtT RESO voce 092°O TIT*O a°e O°f£6 Orcas SLs set £ Th by BNE *O ZGe*O O2LG*TT Thr oO veere o0z'O NaC’oO By ore o*ars Sert as'o- Zz 40'S 558°0O TE6°O O4E9 abe’o OLT*S $29°O 980°O Te 0°66 O°vOS BS°O- Szit- Tt . “SYUAW *I1H9 «BININS ATT N-AG-Ha Ha at/- QALDIMNOD. = + /- J0L o¢cH) CATS C£H79) 207 anh OcHI (AIDS @aWIL TaWIL ¢ N W3LSW STAND 7 NYMOONIIN / NVSHN/ \ V/M OMDIN 7 CONNOA MAMOT ¥) NIW/EW S = ALUM ONT TANYS Y ONTWNSSY CMY A3L97II07 YaLyMa0d JO AWNIOA 3HL WOYS DALY INIWS YO CATA TT¥AY AD) G3YNSYSW MAHLID INSLNOD 4YALuM MWINDIT NO dasvd SI ¥I¥ £W Yad 3LNI0S ss¥H # £8 NUF GT - 904 MISTASNANYE °£ 319vL £A/vI/S GASTAIY : ozT*o £bo°o0 eze°o Zuer SOTLYM Lvs ¥3s ooo'T 000°0- vi9°o 6tc'o gobo bred GES*OW 900°0- NOorO- OB4'T BEST OO'TT 00°09 OS*TT OG'OT Of LAS: be 3 tSE°O o9a'n 82250 “AGO ySEso e98*hL 84£°8 045°O ecast £6o't £T*OT OOS OS*OT S2°6 6 ARPT B94'°0 840°T O9£'oO BE2L*O0 0950 FAG6'L9 = 60G°E vibeo Stay 096°O Gz'é 00°09 SZ2°s s2°a 8 eS2°E ST8toO v0Z2'O Abt to PAGO £ eG°CeT wearc BAE'O POLS covet zea 00°Sss GZ°8 osé é TLE CE vcort 126'0 BT0‘O £456°0 E9¥°0 BT'see 496°OT FLort PEFET LL0°T £9°9 00°GE Sere 00°9 ? RIZ*Y 6S£°T 99B°O £TO"O 758°O aTe40 oorege 009°0 00E°9 ott ats O°sot 00°9 Sc*e S vette PERS c54°0 Gbro*o 70640 ££4°0 0£0°82L GRE*O BPEL Zav*t OS'S 00°06 SEF GLE ¥ 042°9 ABT’ é G9etT gYoro one't 94740 OG42*45 oes°o eeres a8a'o 00s 00°06 GLte se't £ veo? COS't . Gert £LO'O geet oz9"o GLE°b!S 598°O ates 2Gd°o ¥e°o e°60t sSzit eS'0O- TZ £bS*¢ That vrort £90°O TeOsT £59°0 vOLt ey efeet orev 126°O f6°0- O2°ORW BG*O- Scrt~- 1 O¢HI/(ATDS vOS/(NI)S StN/¥+H S+N/H GEtN/GHN %OS/S0ON YN/bOS UN709 YN/OW YN/ 19 tl- AQIN (NIWDI0) CAaNIL TAWIL # SOILYYM YY TOW : SOLTLYY LNA TIWAINOS #8 NUE ST - 904 GISIASNINYA 8S ATAVL £6/¢61/S T3ISTASY O'T- Off- O'T- BES*T O70066 OOKTF O°00b O° one T*O- T'O- 9°002ZS vfs 0°20 O°OGLZ 9S*E OG*TE O8TOT OF O'T- O'FOE “O9TE F£40°T O°00Tr O'SAPT OThOT a'act Ore Orore gorge <Ztet T°9S o°oent o8'f OG°OT SL*6 6 o'9 O'64Z O'0GZ 094°0 OOEGE O° HOR O°OTt Brot 0°98 OTSRLT 8°S arcs o°tca 9O°L GL°6 62° 8 Btt OfvET O°8TH LOW'T O'OSOT O°04S5 Orr ££ Orbe 0°006 6°S 5°83 O° ore cok S28 Ont4 é gre O°TST O° HB ZZ0°E OFF O° Ore 0°o£ ge O's OTe Gee pany tse Oo°e Gere 00°9 g ere O'odT “a0B O6TTT OTOBRTT O°cbb OrTe OTe 9ST O°Scet or2 o's veces S9*e 00°F See Ss Bre O°FST OTRO ZETTT O° Sbe orate O'ce pany 9°B8T O° ORL 6s “es 6°8f£ Te'b Score G2tc. + [and O°f6 o'ces BAATO O'04F oO ORE Oree Gib O'ce Q°0L0T £°9 OAE:) c*O£ “eG*b Sete gest £ Ate o*°cA O'SES TAL*O OFOTL O° Obey 0°OS orot £04 orty O1Oe9T wes 6°9 o°se weer Sct BS'0- € ara O°'64T O'b0G TEE'O O°004T Of0rs O'neT oral Gite O° ase O°STrs O's OTE 8 60°S @5°0- Séri- Tf 304 OCHS CATS +/- FOS FON me) A OW ¥2 YHN M YN H Hed c3WIL TaWIL @ NV/HHS NOU OTM 7 SN NNNAN MULT UZ0 OM TTH “444d * ER ur ST - 90d WWLASYANYE ¢°T A1aYL 285 000°0- 000°0- ¥00'°O F00°O £90°O OTO°O ¥00°O G00°0 £00°O ¥o0'OQ fa) BaS* 000°0- 000°0- 000'0- 460°0 000°O- = 000'O- 900°O £0040 £90" g00'°O 900°O 950°O OTO*O £00°0 O£0°O 9TO'O O1O'O OrO'O 200°0 £00°0 8c0'°O TTO'O 700°O 6£0'O Bboro £10°O T50°0 £80°O T£EO*O TeOra IN MALSW OTANG 7 RYYSONOIN 6t 4c£°S8 Fear’ PGES v45°B LEP'b B66'GT bb4'9 900° Tec0° ge BHorttT 4é nd ad 000° 0O- 080°O £09°O €00°O £00'°O voO'o €00°O $00'O 800°O 9T0°O NH weet £2e°0 Sa £46'°O TZ0°O £b BITS BaO'O BZ 650°S L9L4O £b GLLbL 62T°0 Lae) 99R LL LBE “9 6S 927°OT SBETO 64 TOPE 80 "0 erétb 84e°O &46 9Se°s "SE °O aT BLE TY TLEO TZ FON VW aLaW CONNOT YAMOT NY) NIW/EW SG ca £B/1t78 TaMALNa 9004 O- 900°0- F400 Or0'o0 Ss0*O SOTO 950°O 20190 GOTTO bcL'o ad z°o £TO*O 0°0 o0T*O~- oro 690°O e*o ETO"O T*O 600°0 £°0 Eto'o £°O £10to P9E°O boo'od £0 TTO°O ¥'oO &T0°O0 z'o BEOrO A OW T*o- TtO- atte T*Ge bye Or0s gece dtbe Orde erro oOoT'O- 9tz*0 OGT*O e0T'O OLTHO O£TtO en0°O rOT'o veto ZE5°0 a) OTIN 7 WYMIOMITW = FLU ONTTINYS & ONTWNSSY CINY GALIATION 4SLUM90d AO ANNONA FHL AVILYAY JT) CAMASUSN MAHLIS LNALNOD MALYM TINT? NO dasva ST TY T*0- O'estt Ord? 2°ce S°ae 6°Gb O°Le 6°96 O'4e'e O'vod IN 6£2'° £TS*O odert OcOrEe vogte bb? 446°8 9OA4°T ECerE FO6'S ood: tHN T°O- T *O- ‘ce Orne Orit O°noe O's O°6E ores o'oty ng wand / $ £7 e60*0 OT9°0 ETOAtO zcO*O0 SBOO Bboro 960°O TEN*O Srorod g50'°0 Se0°0 N $ £8 T*9- T*O- T*O- O° O8FEe OG99 6°9E orsee £°OT O°oTT 8°46 Ott S¢it oreo G°ST oraest &°2T Of ECE T'4e O'9le ad NW WOMOOMITH cc eo ee ee ee m oosooosooos oeooocoooesoso Nur ST - 904 WTal4SMSNya = ¢°S a1dVL gLoro £e0°oO 9£0°O 6b0°O £v9°0 vTo"o ovnro B10°0 BL0°0 Teo*o SSO*O uN 6T0°9 vIgro 9£0°O =80°O £50°0 £99°O 800° 500°0 9090°O 900190 TOO'O H Tato sa 5000 £20°O ¥60'90 Orzo 948°0 9SE£°0 OrT*O 9025*0 002°0 S20°o *SUaW 39 £0990 O°? vroro O'OT 090°0 O°s8T 160°O O° bE ZOT*O O° 0b vIT*O 0°09 OFT*O O'bS TIT*O O°0S OB0'O O'by 080°O O°9F “aW) CEN/9) 997 TOA NUP GT - 804 WWLASYSNYA fb ATL 98° oa°é FO°£ Eerek O9°b So°b Te’ cate trey 60°S Had VdNYS UMAAY 00°09? 00'S 00°09? 00°&2 00°GL o'sot 00°06 00°06 B°40T Oc OV (NIW) dd OS'TT OS*OT OS'OT G2Lt6 G26 g2°s8 B2°s8 OS'Z Sete 00"? 00°? Ste sat G2ts BLE sect sett 85°0- B85°O- GBErt- ZaWIL TAWIL “ot NOA NQALHOTIM-3WTL oS'Tt os°or OS'OTt S2Lté S2°6 S2°9a Gate ontd Bote oor? 00°? Sa'v gat GLE BLE scr set AS*o- BS'O- Seet- caWIL TAaWIi WO¥d MALYINIWI YO €£W Mad ALNTOS SSH ~ SUM THWN DAS - NM NMTLIONR ORS : ~“ *» 286 APPENDIX E FOGWATER AND AEROSOL CHEMICAL COMPOSITION, HNO, (g) AND NH,(g) CONCENTRATIONS: SAN JOAQUIN VALLEY, CALIFORNIA, DECEMBER 1983 — JANUARY 1984 (287 feed IOSOM IY SIL AO OND WV DINGOD AD ALO UI Wo 81 tHe *QSONENMILST LON SYM NOTLBM LA UINOD GH dA TdG1O Gt leat HaHa a2aites WICT = 8 *d/O07L ALWIOLNSONN = 6 CONTE OND des & “SNE WINYH = b 10) HOdd 1 MMOH ONL NUALS Fae SaWOL ‘eH/SAIONUMOTW NE S2 HOLHM CADIS LdduXd £709 ON0TW NI 3YY SNOTLVYLNGINGD 119 ~ bOv8179 Wa: O°O FEG'O HIE'O 000'O F20°O O4S*e OOTAT HIST $00°O Tév"O OO6'S OBS°L IFT OoorO stor OSE*E OLBtO UT/T Z00°O £C0'O OST*h OO°BL KT/I BO0'O LEON Ode OOTET ETT Goerd €T0°O Ghéte H0Gts elt BTE*O £eOtO 10% Bebo O4t9O ¥OGtO TO0%~ bLETO DELO HOON 160% 6ee%o TPEO OTOTO TOO% FOTO ObetC rletoO EhOt- 4oGgto CERO FOOT 100%- ELETO OnUTO HEO'O Huara Goro Gullo & EEOO Pret 9NO'O OvOtO aren O40°0 4£0'O D) FONTO TO0% LOR oO GOO'O 4200 Ov'0O wtG'O GTO OOOO ETT 0 Goarg 920°O TOO'- 49T'O G10°O VOo') £20°O OFE*e OO'ET CIVT oO ror OO'O 4C8'O &bl'O S00°O GOTO Gbeto ¢60°O BE0'O OGE'e O00°O ET/T o Soot 9rO'O = 100%= B820°0O 7O0"'O 4bO°O bhEtO G00°O 600'°O Ofer OO'cT TIT 0 £1O°- TOO*- 420°O AELSO bOO'D OFO'O S9E'O 00050 900'D O90%b O00°O: «=TI/T 9 9Stt~ TOO'- 41T'O bhOrO b1O'O O61'O Z00°O FOO'O 90070 OOT2T OO'ET OT/T oO croro 9bO°0 §=T00'- O45'°0 £00°O OF0'O Te9?G 000°0 400°0 042% 0000 OT/T ) B1o'o BO0'O F00'- 478°O SHO'O GOO°O Trv'O O9E'O ¥O0°O B00°O OTT*® £f¢vt 6 /T t) 9TOr- TOO'- O60°O “49T*O SBO°O £00'O 800'O #9E'O 000'O SOOO O42'¢ O0°2T 8 ZT °o 200° ~ 900°0 160'O HIT*O CeTto Z00*O 800'O Z4z'O0 000° SO0°O OTT*e 0000 8 /T Qo 900° - TOO*~ GIT*O Br 'O 9AT’O S00O'O 2c0°0 B£7'0 SOO*O 8000 OTE*w Ostet 2 /T ) £90°O 620°O GPE°O F99°O HTE*O 2400°O TS0°O £84°O 600°O O10'O O8e*y OO'ZT F /T Qo ¥S0'O 200°O ¥60°O SG&B*O Berto GOO'O 6£0°O 4FF*E 800°O TFO'O O6e*e 00070 9 /T o 420°O 610'0 bveOTO 965'0 ZOETO 400°O £bO'O £E6'O FOO'O OTO'O O8%*H COOrET SG /T 0 040°- 700°O £6T'O TEL'O Erb‘O £00°O GO¥O'TO TeE'T 4F0°O 950'O OGT*h OOO'T GS /T 4 0go’- ¥z0*O §=BFTtO «OFF*O Lehto O10'O 480°O Z50°T £00°O 910°O O00°S OO%cTt Fv /T 0 g90*0 TTO*O «=GPT*O Str'O Sbhh'o B00'O Lb0'O 492°O GOO'O ETO'O O£e'y OOtcT £ /T 0 600'°O ‘100'O T00°- 92z°0 OZf'0 §00'O 120°O SeG'O 900°O ¢I0'O OBc*e 000°O £ /T 8 O£0°- Gz0'O T00°- weT'oO TOE'O voO'O £Z0'O FHETO 000°O 400'O OT£'b OOrcT & /T 0 eto*o wTO*O FOO*- 407°O BPT'o G00°O Be0'O £22'O 000°O 900'O O42'v O00'O ¢ /T oO 8to'o SE0°O E£T0tO THT*O TPT*O ¥00°O BIO°O GSZE'O 000°O 400°O OFe*'h OOtCT T/T Oo ~ 900'°0 Z00°O «90040 480°O O4T°O T0O*O §=9900°O «eeeto «100°O £00°O cStot Sére- T/T t 250°0 vOO'O 2F0'O «46Z2'0 = L9b'O G00'O 420'O 499°O F00'O £00'O Ofe*e OO*cT TE/ET °o BET*O Z4T0*°O «=9FO'O «TSE'O TSEtO 900°O 8Z0°O £450 FTO'O B00°O OG42°F O00'O TE/zT 9 G40°- S£0o'O T00'- vIO'oO £10°O £00°0 060°O £20°O vOO'O 4600°O OTE*S OOG*B <e/at 0 200°o foo*- £70°O B8f0°O AbE*O 900'O ZEO'O BeT*O F00'O CIO*O OSO'e OO%Et Oc/cT 0 860° TOO'- TOO'- E£LO'O £80°0 £00°O0 OG0°O 4¢T°O 000°O 800°O O£0tY OO*HT 4T/cT °o zt0'~ TOO*- S9O'O ¥ST*O GOETO £00°O F£0°O ATH'O 000'O ZTO'O OO6'E cetyT BI/ET o TeT*- TOO*- 920'O 1460°O FOTO G00°0 TE0'O ZTZ°O 000°O SIT°O OFGte OO'S6T “TVET o zto"o TOO'= 070°0 ZOT'O TOTO E10O°O 900°O GEO'O 9EZ*O 400'O STO'O OOS6*E OOS*E 4T/ET o T2T*O TOO'- GOFtO £4E'O £6E'O TOO TOO'- TO0'- FAO TOO*- TOO'- O86*E OOCET BT/ET 0 Z200'O TOO'- £60'O fFOO'- TOO'~ FOO'~ TOO'- TOO*- FOO*- TO0°- TOO*- OHOtw OQ00°O FT/ET o voo°’- T00'- 920°'O E£TO'O ZzO'O 000°O T00°- FOO*- £vO'O TOO'- T00'- OSE*E 49'CE St/et 0 662°O TOO*- pSO'O bET*O 9ZT'O 6£0°O TOO'- $FOO'- FOO’- TOO*- FOO'- O46°F O00°O St/2t yOuya tH? QEONH (ADT)S --vOS ~£0N beat I) +49W +499 t¥HN +N +N 4d TL saivd 4x DTALASMaNYH 4% WUN9ONd ONT TANYS HBT NYF-Le4t Jad AATIVA NINOVOF NYS SONTOVOT OLHN ONY “SEONH 410S0N3Y 288 red - reretv9 Tas00Nd “WOSUMAY GHA NO TUNE IVA GINOD Ad atv naw ST Ane TANTWHAL I LON SYA MUOLLYMINHOMO AHL TRAY STH SL TOO’ NUH AD4wWS WIOT = 8 6470072 ALNIGIMIONIL = b COMIL ONT FdW8S = C CONT WINYH = Fo aloo sou] ‘)S$a SMNUH NI NAAIY AMY SHWIL ‘EN/SATIONOMOIW NI SI HOLHM (ALS 4d39X4 £N/OFONITIM NI ayY SNOTLYYINAINOD 1 o £OO*- B90°O f0'O TOO'= GEE°O 48E'O ElGte GBIOTO FeOtG SEED OOOTO 4YEO"O OdetH vt/st o t10'D 260'O0 900° TOO BLT'O STbtO OFO'O HUOTO LEOTO 4EETO GOO PTO'O O£O'Y vist o 940'O SEO'O OF0'O TOOT EGTTO G6E1O FEOTD 9O0°D BLOT AVEO OOOO GOO OTF git € O7O'D FSTO FOOD TOOT- ZETTO GHE'O 20G°O £UOTO 210°O Heb TO OUOTO ETOTO O42'4 £1/t ° £1TO0°O) ¥40°O) ZE0'O TOO'= ZET*O FBTIO FFO'O GOGO THO'O LOE'O GOV L20°O O4e'b Zist o ThO’- PST*O FLOdO'TO TOO'- OOTTO GBTTO OCOD FUO'O BEOTO S9E'O 490°O VIO'O O£E'Y erst 9 Z410°O 860°0 6€0'0 TOO'= TEO'O ZbO'O £70°O TOO'O HOOTO FLO°O 060'Q GO0'O Oadc'l Test 0 OvO'O 000°C ZbO'O FOO'= SET*O THT*O LEOTO ENDO OOO SBE'O £00" FOO"O OcE'e TT/t o EvO'O 920°O EZO'O TOO'- FEZ'O 6OT'O EZO'O LOO'O BlO'O ¥6c'O 000'DO F000 O82'¥ ost () TEO*- TOO'- ETO'O TOO'= TESCO FOTO T£0°O ZOO'O ZTO'O FLE'O FOOTO GO0TO O9h'F ost 0 020'O 000°0 T00'- 960°0 OeT'O ZOO F£EO'O T00'O ZTO'O G60°G O00'O 900°O O£O'b 6/1 0 ZOO'O =ETO'O Z2Z0°O TOO'- 8460°O 4£70°O TIO'O 100° 9O0'O KETO 000'O £00'O O¢e*r 6/t ¥ GEO'G ZO0'O T00t- BZTtO £60'O EZOO LH0'O TOO" TOO'- FET*O FOO'- TO0°~ O8C°F g/t 0 £Z0°O ETO'O T0O'- FEO'O E£PO'O Z7O'O TEO'O 100°O ¥00'O SOT*O 000°0 £00°O Ooch est 0 ET0'O BEO'O OT0'O FrO'O ETT'O £60°O 6£0°O Z00°O E100) BTZ'O 090°) FOO"'O OUctY gest 0 TT0°O TOO'= FOO°O TIF'O TAT'O PET'O S1O'O F00'H 9O0'G 0462'O 00070 800°O 094'F é st oO £00°O 610° 910°0 890'0 26Z'0 80Z°O FTO'O ZO0'O ZTIL'O 2éb'O 000'O 400°D O4ct¥ 9 /t 0 SIE‘O ZIO°O ZzZ0'O 650°O ZTE'O 460F'O <ZEO'O £00°O YTO'O vAS'O SOO*O 900°0 O8c'y 9 7t o 190'O vRO'O 900°O FrO'O S6T°O FOE'O ZOO FOOTO 4T0'O BIS*O OOO 9O0'O OBEY gS /1 ° ST0'O veOtO ETO'O E£ST*O TeT'O EPL TO FEO'O £000 TZ0'O ES¥°O O00'O FOO'O O2z"b vst ° BL0'O OTO'O TeO'O £SO'O S£2'0 S7E°O ZIO'O £00'0 ¥TO'O 616°0 000°0 900°0 050°F v s/t o 91O°O 9ST°O vOO'O T00'= GZT*O LEE'O 420'°O0 FOO'O 410°O v4eO O00"O = GO0'O Odv'£ Oo'ct £ ZT @ 900'- 920°O 000°0 T00'- 960°O £OZO H4O'O FKOOTO FIO'O GLE'O OEN'O PROTO OB'Y Q900'o £ /T S LT0°O «¥G0°O 920'O TOOt- GZLOtO ESTO TEO'O COOH ETO'O OZE"O ¥OO'O %00"0 OSe°e oOrTcT <¢ st 0 TOO'= SSO TOO'OD TOO°- ZET’O BOL'G E000 viOeO Rov oO GOOD GOO'O OBEtR OOOO G/T 9 G00’ vhO'O T1S0°O FOO'O HELO PELtO #OONG 910'O THETO UOo'O 600°O anette Gorter Tt o ETO £47°0 90090 FTO*N 41O°O BOETH O00TO B00'O e&T'O OOOO £OOTO OuTrte OOOO FT /t 0 LTOt~ 2£0°O Z4TO*O 000'0 Of6'0 8460'O O00°O $0070 TET'O OGO'O £00°O O4E'w OOTET TEST 9 9O'= FGO'O eTO'G OOOO 460°0 FOTTO OGO'd TTO'O «FEE TO O00°O HTOTO OBd*E OOO%D TEZET o Z201°O OFOO LPOTO BOOTO 4TOTO H10°O £10°O 9800 O00'O 900'O Ofb ty O0O'B eset °o ToOt~ TOO%- 4¥0°'O HOOTO EFO'O 800°O TZO0'O =£E0°O O00'O ETO'O O0O'H OOTET O&/ET o YOO" FOO FOTO = L90°O LHETO ErOTO BTILO YEETO TOOt EEO O46'E OOtHT ST/ET v TOO*~ 100%- 4¥0'O O80°0 O42'U LT0°O visto GeEtO 060°D 600°D OF4'E OOTET BIZCI 0 TOO*- = 100t- TOO'- 290'O FYIT'O Ob0°O REO'O FIT'O COO'O TTOTO OFL*E OOTHE 4TZet 9 TOO’! TOOT bAOTO FHOTO ESO'O ZTOTO TOOt~ 100% EVTTO FOQt*~ POO OVETE O00°O FILET 0 TOO*- fQO*- 0000 MEOTO «—Z90TO «=1EO°O «TOO'- 1OO'- FEOTO FuGr- TOO? O76'E 000°O ST/ct 4Ouad 4H? OSHN OLUNN (AIDS --FOS EN “HW + bE 4400 t+VHN +N +¥N ia Th aivd xe GCISVM FX HYNOONd OND IINYS PESTINYE -CUST Ua ATTINA HLABYOP NYS SONICVO) OEHN UNY fOCORN S1OS0uae 289 red - '8/8t7? 13400Nd *1OSONSY SHL NO JONV WH JINOT Ad WAL INIW ST 4H? “MGENIWNSLIT LON SYM NOTLYYANSINGD JH THAN TdSI SIT T00’- NAHM BOMNDS 19907 = 8 44/9071 ALNIVINGOINOA = & FANTL ONIDANYS = ¢ SONI WINYH = T 3ad00 YOUN *I1Sd SMNOH NI NSAID 3M SaWIL ‘EW/SHIOWOMIIW NI SI HOTHM (AT)S Lda0Xa FYN/OZONIIN NI 3YY SNOITLYMLN3ONOD TW £00'O TO0'- £90°O TOOt- £80°O STT'O O70'O OT0'O %z20°0 BET'O ¥OO'O G6£0°O OF2*e OO'ZT IVT 500'O 900°O TEO'TO TOO'- BzT*G OvT'O F10'O FOO'O TIO'O 4¥E'O 000°O 9IO°O O8tte OOOO F¥t/t £G0'- TOO'- EFEO°O TOO'- 620°O BTE'O 800'O 400°O F£0'O Z44T°0 000°O BTO°O OcE*e OOZT ETT 420°O 000°O 420°O TOO'- 99E'O 4ET'O FEOTO FOOTO BTIC'O HLE'O F700'O BTO'O O240°R O0G'O ET/T STO'- £20°O £FO'O TOO'= PST'O C6T*O ZTO'O 700°O 6£0°0 bOLTO 000°O 620'O Och OOTET ZI/T TLO*O =TEOTO GEO'O FOO'- 420°O FES0'O O20°O ZOO'O O70'O 120°0 000°O 920° OOC*b O00°O ET/T TSO*- BEO'O OFO'O TOO'- 450°C 9T0'O 800'O bFOO'O Zz0'O G40'O 00D 600°O OcZ'r OO'ZT TI/T OvO'~ 000°O 400°O TOO'- ZEO'O 120°O 000'0 OOO FIO'O 4TT*O vOO'O TE0'O OOC*e OOO'T TI/T SBO'- TOOt- ZTO'O [TOO'- bIT*O 460°O 120°O0 700'O £b0°O Zét‘O £TO°O 040°O Ofc*he OOtET OT/T ZOT*O =FOO'~ OTO'O TOO'- F46z°0 ZEO'O 470'O FOO'O 4T0'O Zzz*O 000°O 4600°O Ofo'e OG2*O OT/T O£0°O 600°O T00*- %S50°O PEO'O YTO'O BZ0'O T00°O FT0°O 8Z0°O 000°O SO0'O beet OQOrET 4 ZE0*- 000°O TOO'- ZET'O B4T°O OTO'O 000'0 200°O ZI10°O £AT'O H00'O B£0'O O0Z*H 000'°O 6 SS0'O0 000°O T00*= GbT*O 40Z'0 TEO'O BZ0'O ZO0'O FO0'O L4T*O 000° 400°O Oce'e OOTcT B TZ0'O =£00°O GO0O'O BPO'O OT*O BzZO'O TZEO'O T00O'O 900°O HhE'O 000'O 800°O OBT*w 000'0 & 650'O ¥00'O TTO'O HET'O SOTTO ZeO°O FZ0'O Z00°O OFO'O Z6T'O 000'D BO0°O Ofe*w OOET £4 FOO T00'- 600°O ZIC'O S220 ETO*O BZO'N EOO'O ETO*O vht*O O00'O GIO'O Oze*y 000°O 2 BS0'O TOO*- 6£0'O 090°O GRE'O £40°O 0700 900'O bEO'O EVETO O00°O LTO'O Oct*he OOvET F #Z0°O 2£00'0 2000 080°0 922'0 BZ0'O 000°O Z00'O OT0'O B0F°O 000'O OTO'O O4T*hr 000°O F Z50°O 000°O bE0'O G9T'O H2z'O BPO'O OTO'O £O0'O ZTO'O 29Z°O 000°O 800° OF4te OOtET S/T TZ0°O 000°O T00'O Z4z'O ZOZ'O E90°O STOO £OO°O ETHO*O ¢BT*O 000°O O10'0 O4b'f O54'°0 & 080'O 000°O TOT'O bEZ'O FOL'O TEO'O 020° GO0O'O ZEO'O 9Z2Z'O 00O'O FIO'O Ofete GOt'ET & Tbe 'O 000°0 O20°0 940°O 2O8tO PEO'O FEO'O GOO'O vIOtO GBEO 000°O bE0O'O OOcte 000°O ¥ BF0'O 02040 ¥9T'O TOOt- O8Z°O EBT*°O TEO'O ¥OO'O 610°O vero 000°O 400'D OBt*e OO'aT £ £90°O 94T°O POO TO0'- 99T°0 F4O'O 000°O FOOTO FIOTO T9Z°O O00°O ETO'O O80'w O00°O £ 9T0°O 060°O 690°O f00'- OFT'O PbT'O OZO'O ZOO'D FTO'O EBE'O O00O'O 400°O O4e*bh OO*ET E BEO'O GOZ'O TZ0°O TOO'- ZOT*'O THEO SEO'O ZOO'O T1IO'O FRE'O 000'O 400°O OFZ*w OOG'O EZ £00'- £90°O B20'O 900°O B4£0'O OFT*O 000'O Z00'O FTTO'O B4T*O 000°0 4600'O OOF*H OOrET FT 900°O TIT*O O0O'O PTO'O ZIT*O bET'O ZEO'O SOOO zZvO'O F£4T'O 000°O EzO'O OST*h BOrET TE/ET TOO'- 8T0°O @bOtO BTO°O 960°O SZT*O Ez0'O £OO'O Zz0'O 45Z'O 0000 OTO'O COB TE 049°0 T£k/ET BO0'- 2460'O £Z0'O FEO'O SE0'O 9200 600°O £00'O STO'O £40°O O00'O 400'O O£O's Gétet <evct ZbO'O =T00'- TO0'- £50°O O80°O EBOO ZI0°O 600°O 8F0'0 §20°O 000°O 9E0°O 000°F OO'ET O/T ZEO'- TOO'- 100% 960°O GZ0'O 140'O 800'O FOOTO SZ0'O bET'O T0O'~ ¥OO'O Oeore EFT 6T7ct vZ0'O TOO'- TOO'- BET'O 920°O BTE'O TZ0'O LOO'O FIOTO FSET°O #¥OO'O ETO'O O44°H OO'ET tet 9T0°O TOOt= TOO'- 100'~ 470° O8O'O SEO'O ZOO'O LTO'O . wET*O 000°0 Z400'O OB6*E OO'HT <4T/ET ELO'- TOO'- TOO'- BIO'O BIO'O ETO'O 000'0 £00'O STOO 6£0°O O00'O 400°O O84°E OL4e°t 9t7ET amerTeoeooeooononsoocoonoonseootnadT@soooanSoosg youNg tH} §£HN SEONH (CAINS -~b0S ~£ON “ID ++0W +409 +¥HN +N +UN 4a TL aLVvd wk MOTMLLIN JW kk WONDONd ONT IINYS FEST *NOF-LEST IA AATIVA NINOvOr NYS SONITYOT OFHN ONY SOLONH 4 10SONIY 290 cra ~ ¢87et79 1a40084 *WSON3Y FHL NO GONY IVE GINOT Ad WALVINIW) SI CHD ‘WANIWNS190 LON SYM NOLLYHINGONGI GHL daAVTdSI SI 100*- N3HM agynus WaI01 = B 44d/O¥/L ALNIVINGONT = & SONIA ONT TdWYS = € ONT WINGH = To Sa09 YOMNS *1S4 SUNOH NI N3OI9 3YY SAWLL *fW/SAIOWONDIW NI SL HOTHM (AIDS Ld350X3 EW/DZ0NIIW NI ANY SNOTLYMINGONOD T10 9Z0'O O20'O B20°O TOO'- TeT*O GIZ*O EvO'O TOTO GE0O°O0 692°O 000°O BrO'O O4e'H OO'CT bI/T 6TO'- 920° TCO'O TOO°- BOT*O 982°0 FTEO'O POO'O GTIO°O B6£°O FOO'O EFO'O OCO*H 000°O HTT £20'O 000°O £90°O T00'- FIT'O E£ZT*O ZTO'O FOOD FZ0°O 94T°O 000°O E20°O OGEre OO'ET ET/T Seo'o Z2¥O'O S60°O TOO'- ELT'O BBl'O O20'O 400°O 250°O Z6T*O 000°O OFO'O OGE'Y OO'ET ET/T ZIO*- BROTO SOOO TOO'- ¥S0°O LOT°O 4IO'O F£OO'O FZ0°O ELT*O L00'O "FZO'O O£FO'H OOOO ET/T G00*~ Z80°O, STOO $fTOO'- 6£0°O 660°O PTO'O TO0°O ZTO°O FE0'O 000°O 210°O O4e'H OOrCTE IT/T TTO*- €00°O OT0'O TOO*- SS50'O -440°O 000'°O 100°O 900°O bST*O 000°0 ¥00'O OFe'h 000°O TIvt 0Z0°O ~FhO'O B£E0'O TO0°- B9T°0 020°O PTO'O TOO'O F1O°O GTZ'O 000'O FOO'O O4e*w OO'ET OT/T T£0%- TOO"- E10°O T100'~ 990°0 BZ0'O0 000°0 L£OOTO TroOto LET'O O0O°O BO0°O 000°O OT/T GEO 000°O 6¥0°O ZHT'O ZEO’O FOO'O BHOTO EOO*O Y10°O SS0'O OO0O F00°O ooOr'et 6 S/T ¥00'= 000°O ZbO'O TOO ETY*O BFO'O O0O'O TO0'O Elv'do Belo O0O'D L00°0 000'O 6 /T BTO'O 000°O TOO'- GEO'O ZEOTO BtrO'dO FZO'O 000°O FOO'O OBOTO 000°O F00'°0 pe*’et 8B /T £ETO*O FOO'O TOO*~ O40'O EFGTO OF0'O ZTO'O T00'O 800°O S4O'O OO0'O £00'O O00tH 000°O 8 /T TSO*- TOO%~ TOO'- TOT'O TZOO bZO°O OOOO ZOO'O FTO'O 1420°O 000°O Z400°O Ocety eect 4 /T weo"'O TOO'- OOO E€S0°O é £0'O BrO'G 1T00'G SOOO 666°0 00°C 9U0TO OE9'e OOO'O £4 /T voor- FTO'O £EO'O HOO ELT*O = 91LO'O ZOUTO TEO'O ERETO GO0'O G00°O O4e*e OOTETE F /T veEO'O #90°O FEOTO FZLT'O 298°0 EOD EOOTO FIOTO ESS'O F00"'0 400°O OScte OO'ET G/T voo'd 600°O G40'O0 9E0°O pel tO #lo'O POO'O GEO'O FOTO 000° 900°O OSc*e OOTET ¥ /T 6c0'- ¥EO'O HLt*O TOO'- Ore'O O00O'O BGO'O 620°0 Z6E'O GO00'O OFO'O O£O'R OOrET £ /T TTO°O «=TeOTO. TO0O"'O) T00'~ OTE'O 6LOO COOTO HTOTO SevtO 0000 B00°O OO£*w O0O'O £ /T STOt- £v0'O GSPO°O TOO*~ BAl?O BOO'O TOGO FTIGO %4E'O 000°O £OO'O Oe'e OOtTT Ec /T vOO'- 000°O Z00*°O TOO'- SHO'O ZET*O OOO TOO'O B00'O ZOZ"O 000°O FOO'O OBé*h O00°O c /T STO 8Z£0°O G00°O 900°0 910°O FEO'O 000° COTO EIO'O F9OTO 000'O 400°O OOF tH 000°0 T/T ZLO'~ HEOTO SEOTO 900°O SFOTO 60OT'O 000°0 Z00'O BOO'O FSEO OGO"O OTO'O ete OO%EE FEET E1OtO EGO°O LTO’O 000° 190°O OZLTIO 4TU'O ZOO GOO'O BTE'O 090°O FO0'O OLBE O00°O TEZAE FEO'O T9O'O E£FO°O FTEOTO 9ZO'O SLO'O TEOtO £0O'O 400'O ATOTO OOOO EIO'O O49"E KBL- LasEt ELO'- TOO’ TOO'- TOO'- E£TO'O BE0'0 SZO'O 400'O BZ0'O 490°O 490090 420'O OUO'H OOTLT OE7ET LOO'- TOO'- TO0%- OF0O'O ¥AO'O OFTO HHO'O GOOTO TZOTO SBz°9 H00'O BOO OB6°E OOtFT GT/ET TEO'- TOO°- OTO'O ZOO 4TO'O FZ0'0 21O'O £OO"O OFOTO SOTO 060°'0 600°O 000% Oor'fTt Bt/ET BIO'- FOO'- TOOt- Z80'O0 600°O 600° 000'O %O0'O STOO 000°0 OCOO'O ZTO'O OvdtE OO'BT 4T/ET amMooconocoocewesconoceoceococec@tdoo yOuNg CH? QZHN OEONH (AIDS --?0S ~£ON “19 4+t9W +100 +bHN ™ +¥N du TL aLv0 aX STITH LSOT ** HYMOUNA ONE HAWYS VEST NOP -£A6T*IaT ADTIVA NENOVGE NYS : , SONINVQT OSHN ONY “OLONH 4 TOSONAY 291 Crd ~- v8/8T/9 14as00Md *JOSOMSY FHL NO BIND WA DINO AP ASL INIWI SI CH? “TININMALAT LON SYM NOLLYMLNAINGD AHL WGAV TSIM SI TOO*- N3HM SOMNOS WIOT = 8B 44/90/L AINIVINAONN = & FSAMTL ONT WWYS = & fONITMINYVH = T $4009 4OuNa °LSd SYNOH NI NAATD AMY SAWIL *YW/SSIOWOMIIW NI SI HOTHM (ALS Ld39XR LW/NSONTIN NI 3MY SNOTLYMLNAINOD TW cco’- E£OT*O TOO*- TOO*~ Obf'O GBc'O ELO'O TTO'O ZvO'O BBE‘O 000°0 FEOTO OO OOtET T/T HO0O'O L92'0 900'O T0O'- 49T'O “4EE'O FEO'O BOO'O GEOTO ZTS*O 00°00 F¥zZ0'O 000'R O00°O FeT/T 9£0°O £970°O FEO'O TO0'- ZéttO SEF'O PEO'O TTO'O 870'O 4é4f*O O00'O B80'O OS4S*E COTTE ETT 6£0°O O4T°O 9700°O TOO'~ 4ST*O OOF'O OFOTO G00"'O EFO'O ZBF'O 900'O FE0°'O Oc4*E OOO'T EET BEO'O Sb?'O ¢4£0'O TOO'- FZT*O OTL'O ZFO'O OFO'O E9O'O PSE'O OFO'O GOTO OFO'R OOTET ETZT 7TO*O £6E'O OOO'O TOO'- 1420°O &460°O [TEO'O HNOO'O EThO'O 120°O 000°O TE0'O O46°E 000°O ET/T OLO'O B4t°O STO'O TOO'- TEO'O THO'O ThO'O ¥00'O FEO'O FLO'O 000°0 600°O Ofek BStct TIT ve0*O G90"'O BO0'O TOO*- B8T*O ZéT°O FOO'O T00°O STO'O LTE*O OOOO FTO'O OS" 000°O TT/T BE0"O O090°O OCO'O TOO'- TLT*O SBO°O OFO'O 900°O GEOTO 40%°O0 O00'O B00'O 006°F OO*ET OTZT £CO*O =100'~ 97O00'O TO0O'~ SHT'O 460°O eceO'O Z200'O FTO*O S2E'O 900'O ¥00'O OFT te 000°O OT/T eSoococomeosooonseowroocuws Z£00%*~ €£0°O FTOO'- £9O'O 99T'O 850°O0 TE0'0 F00'O TOTO G66T'O 000°0 900'O O€Otb OO¢ET 6 /T v£0'O 940°0 100%- TOO*- FOT*O GrO'O TEOO LOO*O FTO'O O€T'O 000'O F0OO'O OF2*E 000°O 6 /T vvO'- BTO'O TOO*- ¥£0°O OfOTO FHEOTO FOTO TCOOTO TOO ZIT'O OCO'O TOTO OHS TE OOtZT A/T B40°~ ZLeO'O TOO'- TSO'O Z4£EO'O FE0'O 000°D0 L£OO'O CTO*O Seto 000°O 920°O O046tE 000°S B st TEO'O BEOTO B00'O EST°O SEE'O $F20'O BEO'O FOO'O GEeor'o SZT*O 000'O OT070 066° OO'ET 4 s/t BEO*- 690°O bOO'O ZET'O TLE'O £L0°O OF OO EOO'O 9000 #42°O G00°0 GE0*O OOStE OOO'T 2 /T 6C0'- 620°O bIO'O BOT'O OFF°O ZET*'O O00'0 6O0'O Z£E0°D0 S6£690 000°0 900'O O40'H OO%ET 9 /T ZEO°- OFO°O ETO'O 690°O bHE'O BSETO POO'O FOO'O ETO'O FH9'O OOG'O 900'O OSO'e O000°O 9 /T 920°O 4E0°O ZE0'0 440°O Ter tO <HE'O TEO°O SO0*O 4€0°0 £99°O Of0°O 900°O 000'% OO'sTt G/T 446°O ETO 100°O T00*- 625°O0 BYL'O GTO'O TOO*- TOO%- TOO*- TOO'- TOO*- O£0°F O00°O S/T MOMMA tH} SEHN OLONH (AT)S ~-bOS -£0N “VW +45" +409 +bPHN +N +UN Lil TL ALU ** MOTIIMNOLINA ** WONOOM ONT TINGS PBST NOT-LBST IAT ANTIVA NINOVOr NYS SONTNVOT OSHN ONY 4ORONH SOSONSTY 292 *1Sa SYNOH NI NAAT GCO*O 200°0 etot- cco 600°0 200% ~ 6£0°0O TEO*- ¥£OroO T0O0'O 600°0 900'0 900°- 9T0°O Tco*o 6S0°- TLO't~- ZLO*- ¥00°O socoocscoceocoeoocooc]|cse“ 4Oudga tH} WONDON ONITINYS PRET NYT-£B6T°94N ABTIVA NINOVvOr NYS eed - ceG'oO Lév'o 685'°0 €65'0 Grs'o c99'*O L£9¢'°O Sov'o vat'o Gke'o Sé6L'°O B8s'o FILO O8E°O 665°0 T00*- TOO*- TOO*- to0*- OLN b8/8T/9 N3a400Nd *JOSOMAY BHL NO SONVIVA DINOI Ad SLY INIWI SI {H+ 8t0°oO £00°O 500'O 000°O £00'O TT0°O £00°0 500°0 ¥00°0 £00°0O 200°O g00°0 STO*O éT0°O Ot0°O TOO*~ TOO*- TOO'- T0O*= 9L£0NH *TAININNALI LON SYM NOTLYMINJONOD AHL MAAYTIdSTO SI T00*- NAHM 39NNOS WIOT = 8 44/OV/L ALNIVINGIND = & 49HTL ONTTWYS = & “ONT WINVH = T 83109 YOUR 3My SHWIL “EWN/STIOWOMNDIN NI ST HOLHM (AIDS Ld39Xd LW/OSONDTW NI ayY SNOTLVYALNSINOI “119 050°0 FETTO S49£'°O ZEOtO GOO'O ZEOTO SFYTO 000°O ETO°O OCO*y OO'ET T/T SE0'O FOOt- TOO FOOt- TOO*- TOO'- TOO'- TOO%- TOO*- OGZ*E OOS*T T/T £IT'O S6ZE°O F£80'O ¥OO'O £00'O STO*O £OZ'O 000°O 200°O O2Otw OO'cT ¢ 990°O 9rO'O £20°O 810°C GSOO'O BtO'o Bet*O 000°O B00°O OfB't O62°T € 870'O 92£0'O BOT'O ET0'O c00'O FTO°O ¥9T'O 000°0 vO0'O O£E"S 000°0 F £¥0°O O80°0 TET*O Sc0'O G00°O HEOTO OBT'O O00°O 400°O O86tE OCO'cT G Ov0O°O EIT'O GLZ'O TZ0'O Z00'O OFO'O #¥SE°O 000°O £00°O 000*% O00°O & 200°O TOR‘O £82'O COO'O FN0'O ZE0°O 6GE'O 000°0 SOO*O OS46*E OO'ZT bv 000°O 960°0 SIf£'O BOO'D SOOO FZ0"O BETO 000°O 400°O OB86tf 000°O vb /T 200°0 620° T4t'O TeO'O SOO'TO FEO'O Src'O 900°O YTO°O OFO*F OOvET E£ 000°'0 650°O B9T'O GzZ0'O cOO'O OTO'O BTzZ'O Z£00'O 700'O O446"£ .000°O F£ 000°O 260°O ZOz'O ZIO‘O £OO'O $FEz0'O TBZ*O 000°O G00'O O446°£ OO'ET CE 000°O £40°O vBl'O ZI0'O 200'O ETO'O F5Z'O 900'O Z40OO OO6'E OOS'O ¢ 0£0°O €60°O Z4T'O Fz0'0 £00°0 STOO 492°O E00°O £00°O O49°9 GZtTT fF 000°O SEO°O 840°O EFO'O £00'O 700°O BOT’O ¥O0*O F00'O 040°4 OGc*- Tt TOO*- 960°O EIZ*O TZO'O 400°O TEO'O BOL*O ETO*O TEO'O 08699 COTTT STt/ctT 000°O 12Z0'0 bGZ°O 000°O vOoO'O 6f0°O SEO ¥O0'O FOO'O OOE'H OO'CT BI/cT 0v0'O STT*O GbF'O g00°O £O0'O STOO E2b'O ¥00'O 900°O OOc*he O00’Ef BT/ct 000°O TZ0'O O£E'O OZOtO 000°O 000°O ZTE*O 000°O 000°0 OOd*e OS*¥t Ztvct (AIDS -~-bOS ~£0N “VW ++9ON +400 +¢HN +N +0N ig TL aly ax YVITWSIA &x SONTQVOT 9EHN TINY ‘OLONH *1OSONSY | 293 ced ~ °1S4 SHNOH NI NAAT STO*- 600°O B00*- 600°0 BTO*- 200°0 SEO'- £00°0 £00°- Ted*o 800°~ STO*0 bco'o § 410°9O TEO*- Gearod G0O*- GEroro 600°- ‘¢¥O*O eccssesoocs 4yOuNa CH} SSHN HYMOONA ONT TANYS PEO NYUP-LBST Oa AD TIVA NINDYOr NYS b8/s8tv9 Mas00Nd *1OSOMAY BHL NO JONYTHH DINOI AD ASLYINIWS SI {Hy PEO'O £20°O £00°O TOO*- TOO*- £00'O bTO*O S00°0 £TO°O £T0°O SL£ONH “NINIWYSLS0 LON SYM NOTLYYINSONOD AHL SAV TdSIO GI T00*~ N3HM 3ONNOS WIOT = 8 44/9W/L ALNIVINAINN = & ENT ONT TWdNYS = ¢ SONIWINGH = TF 23d090 YOuNa AMY SGWEL *SW/STWWOMDIN NI SI HOTHM (AT)S LdaoXa £W/ORS0NDIW NI 3aNY SNOTLYVYLNAGNOD TV TOOt= FTO'O EZO'O F£OO'O 7200'O 4600°O Z£0'O 000°O ETO*O O008°S OO'OT ET/T 000°O Zt0°O0 £20'O £00°O £00°O FTO'O 020°0 900°O SlO'O O08°& OO*OT ET/T G00'O 900°C £00°O 9000 FO0'O FTO'O O00°O Of0'O ZT0'O O08°S OO°OT TI/T A10'O £20°O 600'°O TTO'O £00'O cEO'O FIO'O ¥OO'O Bc0'O O08'S OO°OT OT/T 600°O £00°O £90°0 000'O T00°O 00°90 000°C 000°O c00'O OTB*S OOTOT 4 /T 850°0 Z00'O 000°O 000°O 100°O 900°O 000°0 000°O £00'O OB4'& OO'OT 8B /T TZ0°O «=FTOTO «6ZTO'TO: «bTO'O: «61000 «6400°0) «9900'0 «=FO0°O =F¥00'O 0442'S OOTOT 4 /T GG0'O 000°O ¥90'O0 000'O T00°0 900°O 600'O 000°O 8B00°O O82'S O0°OT F ZT GE0'O £00°O $£00°O wTO'O TO0°O 600°O 400°0 000° 900'0 O82'& OO'OT & /T ZZ0°O) «900°O = 900°O £00'O ZOO'O TTO'O 000'0 v00'O 2£00'0 O£8°S OO'OT & /T C(AIIG m= bos ~£ON -W ++9W +409 +PHN +N +¥N Li Th aLvd xx YTTSIAYSI ANT Xx SONLIVO1 SEHN TINY “9fOQNH 4 10SONaY 294 req ~ *1LSd SHNOH NI NAAT9O 0 GZ0'O =100*~ t 060°O ¥50°O 0 040°~ BT0rO ) OOT'- T00'- t EbO'= BOO 8 6b0*- vZoro 0 600'0 £50°0 1) ET = SEOO 0 920" Gz0'O ty) S00'~ Soto if Oz0'- T00'- 0 zbO'- 000'0 t ETO*- 000°O MOMMA CH} O£HN B10°O ¥O0'O GO0*%O TOO’ - T00'- 600'0 800°O 900°O £00'O TOO! ~ TOO'O 200°O SEONH bB/8T/? NWAsOONd 320Nnes 14901 ANY SaWTL TOO’- O£0°O TOO*- ceONO TOO'- ZT0*O EO'O crO'o TOO*- = £00'O 000° § 4eO"O 810'O TTO'O €£0°QO 600'0O TEO°O =£00°O ec0°O ETO'O TOO*- OOO FO0'O §=460°O TOOt- TTO'O (ATIS --bOS WOY9OM ONT TIWYS HAST ONYT-£B4T Ja AATIWA *TOSONSAY S3HL NO JINY IVE JINOI AG WALVINI WI SI CHF *HONINMSLAD LON SYM NOLLYMLN3S9ONOO AHL W3AANTdSIM SI TOO’ N3HM 49WIL ONT WWYS = ¢ = 8B 44/90/41 ALNIVINSINN = PrO'O o50°9O FO0'"O 000°0 000'O 6O0'O 900'°O £00'O 000*0 600°0 000°O £00°O 800°O ~£ON 600°C cOO'O 000°O S1O°O NInHvoOr NYS 600°O TOO' 800°0 200°C 700°O £rO'O Too'o £00°O £00'°O rO0'O cQo'a c00'O coa'a ++9W b50'O TOO" 4650°0 c90*O 050°O 650°0 OoTO*O FETTO 9 0'°O beo'd &T0°O Teoto L£1O°O +4+¥9 £40°O 800°O 000°0 000°O 000°0 OTO*O bTo'o ro0'O 800°O OOOO 000°9O vo0rd Teoro +PHN Ort to TO0o0*- £00°O §00°O 000°O 900°O 000°0 000°0 000°0 COO* 0 OO0'°O 000'o 000°0 +N £60°0O TOO*~ EEO*O £40*0 £00*0 500°O £00°0 £00°O Tro"o Z00°O £:00°O 200'0 700°0 +N ‘ONT WINGH 86°S OG6'S OS6'S O&6°S OS6'S 046°S 026°S Och hs OOTTRS ogg Oe es cer’ OGL A xX TdAYHOVHAL kx SONTQVOT SLHN TINY = 1 00°Or 00% OT oo'oT oo'oT 00°oT oo'oT oo°oT 000°8 oo°Ttt os tT O0*OT oo'ot oo°ct Th 23009 YOUN *SW/SAIOWOMIIW NI ST HOLHM (CAI)DS LddoXa £W/NROMDIN NI ayY SNOTLYYLNAINOI 110 vist £T/T et/t tts OT/T 6 /t 8 /T 9 /T gs st v /T SOSONH #050449 295 eFONIMAIATO LOS 50°S Té*v SOKO KK KKKK Hd ~AU-Hde 000°T O£L°T olor OCHIs CATS NN SOTL¥Y OO*vE oOreT OO'gT z*0- z*0- 0°0065 10N Ws OGL NVUW/S vESs'O ¢ee'O v9ESO SEBO £S2°9O | £84°0 tho eMNOde te Zane G&6° eZtt €460'°O 982°O vEe'O vOS/(ATIS StN/Y YOUN 4 7 NNN Orde Orbe O'eb orze + ONCHI CATS \N VW OMDTH 7 O'BET 4&9 O°Slt 5 PB/SOV/G HAST ARM TOh69SELHONO —-vLT OO OOTO- ¥S0'0 Bg0*0 OOT'O- 608'0 4880 Sy 'S9SE G90'O —-9BT*0 30L © -OZHD cS \ MaLaW SLUM ONT TANYS NTW/EW S Z°O~ Z*0- 0°006S O°fb DOL *NO/WENSOMIIW 7 NIZWRZ ON VoLY ADNSTITAIS Z0S SNTNNSSY Ores O'ed OFHI ¢ WH ON 850'0O Orde —-GbOTO OTHE O'BET Y60'O O'FT O'STE OG0'O O41 ADS a aM1 “A OIW 7 (EH/9) CW SHL WOYS SLE WIIV LNSLNOD YSLYM TINOIT NO WaSva ST Miy¥ 4O EW Mad JLN10S SSuW va/sot/S dasingay t 900°0 CETet ost’t o 800'°O 2b6°9 vTe*o t 600°0 £OUtT vero +H SEN/H = SEN/@HN ¥OS/EON \ SOTLYM va/9T/S Wasingy £8'O O0°208 O'9tS 0°S~ 0°968 O°eer O°oT 5L°O O'fB4 Le O°Ze t/~ vOS FON i) VNNNANSN $ £8 O30 BT - oeT'O £v0°O 9SS'be B42'é £49B8'°6E LOT TU9*eS §=688'e YN/vOS = YN/YD INS WWAINOS 904 Wal ASNaNed BECO vero ooT'O ££0°O YN/OW ZT°y OOT*O- £LE*O oos't YN/19 4 ff $ £8 937 BT - 904 WIS ASMaNed o°8 ore o*st SW oro 0°0 -0°O ¥ S 0°0 S oto “ o°o a vH MaLt1V/03 OMIM $ ¢8 947 Bt - 261 ¥OT vt IN o°at O°o0£ 0°at VN 4 S44AL LA YO Ua AS ING *T a1avL GINS £ MOA SUN AAY UCDO TAM JWOD E°Grt bers “1° nats £ ¥ots OO'P BOR Le S4°t 80'°'E f8°oO 61 At LWIW CAWTL TAWIL #€ CNIW) TaLIaTIOI MaLyMyOd sO aWNTOn *e A Td0L SSOTLYM LWS ¥3S vsa'o t2°8 2£T°6 8é*L £ S9B°O osre 00% Ho'f € £52'°O Sé6°T 8o°£ rea‘o of t/- LIIW CAWIL T3WIL ¢# °e WW¥L T*o 40°L LTS GEL £ 6°B S0°G 00° gore ¢€ Etat Té6'&’ Bore £e°o of H He THWrlL 296 OOT*O- O0TtO~- Z00°0 )=—- F000 Too*o S00°o A IN VANS ALON O60 £28°T ooTto- t00°0 900°O no MaLaW OTUAD / ONT 1dkY¥sS Bv9a' WONNOM vOsS ESE ¥o0'o 800°O €00'°0 qed NIW/EW ¢ VASO se veoto 000°G> FOr Bvo'o at) iW O01 'O- 100'0 £00°O NW / ¥oLY ADNITOL 700°O G00°0O ootd 600°O OW 11S LAaM 900°O “Lord Ono Oo ad 4s GIAaY £90°O 4b0°O ¥so'd £40'°O vo QW OTANI 7 WHNMOMITIN 1°O- OPGk Ore A NON 4eort BELT OGL? T Ooret vHN ‘ALUM ONT VWYS NIW/EW GS HL WOMS GELYVINIWD LNILNOD MaLom TINO ro- b°O~ 0°94 GEE vor ores Oth Oritt O'ue IN no Hel NN MILET 7 WOOOMSIN 1 408 ONIWNSSY WALI 1I09 INGO MALYM ENOL) NO WaSVa ST MIV GO MULOW ONO Mad SW Sova dO ONTOVO $ £8 930 OT - Set WaT AGMANGHE “oy 4e0'°O Q900°O | 0¢0°O Q00tO G90°O T00*0 £20°oO 1OO'O YN WW ¢4teé Bw OLY AONHLO FAT 205 SNEWS HALO vo GUMIY 40 2W Yad ILI SSUW * £8 93 BT ~ 04 8G0'°O 4¥0'o ¥40°O OL0°O ont «rH7 NO) WAT ivngod T°o- O'eTe LO°L LEG ocol Or6at g0'S Ooty O'ds O'so9 ta’ gore NW Ad bed “s¢thes 40 AWAIOA THE OL GATOS F Orel Gait eeu 2°6 Oret esce 00°F O°rdt Gé°t gorse “WA Ltr 4uIW GIWLL cw CNTWD LYm90d AQ AWN ION MVE ‘ey TIGL x) $é gore £u°O £ é Lf VAWEL = #€ Gere TAWTL + 297 6E°e 69'°9 £vo°t £ RXKEK OSSD 828°0O 8 aKKKK OTOL 2EL'°O - sSONDMaAATO Hd t+/- ~AT~-Hde oHdNOT OoT*O~ ooT*o- 9a7°o 9S0°O 9EE°O 020°O OZHI/CAIDS vOS/CAIDS S NN SOTL¥H NY TON / / N OL£'Z Zt0O- =TtO- iF t0- OS'*ET O°00S O'ZOE 0°02 OS'Zy 0047S O'4E = OST TOA I01 OZHO (ATS NIW/ NVWHS ON OI/W OMDIW v8/9T/G NWASIAIY B46 9LE4250°0 BL0°o vo't OOT*°O- OOT'O- OOT*O~ 28°90 4E°9BbSE ZTT°O 980°O g2°o HARRAH OLESO 9ZET°O +/- 901 OcHI (AIDS 319M ONITVANYS NIW/EW SY LY AONSTOIAAD 40S ONIWNSSY SHL WONS T3L91097193 LNZLNOI MALYM UTNE NO WaSvd SI YI¥Y 4O EW Yad 3LNI0S SSW ve/9T/S WASINaY 6880 000*0 S88°O 飣°O £80°T 000'O 9B0'T 2£ET°O vLe°T 000°0 wle't £59°O +N/YtH S+N/H StN/@HN 60S/E0N NNA vB/9T/S VASINIAY £vo’t O'orss O'O4bz O°SB B28°O O°'046eZ O° T6E Oot ZEL*O ORE 0*08E O°ee t/- vOS £0N aM / VNNNNANAN E°O- — T0- 0'00bS O'ZOT 0°04zS 04% a04 O¢cHd \ MALAW SNMO/WYNSOMIIN 7 NU/WW/ NOV/W OMDIW / $ £8 330 6T OzT°O £ro° SOTLyy IN3TWAINGSR ovo'o T*o- ptoro te O'0g efoto S’tt ost £ec'O =S'cy CATS oa1 TON (£W/9) CTW) - 804 WaLsSyaNed 0 ecc‘o aver Tee os CéT°S 2£95'°T Z18°0 00S'Z40E LEETS L£97°O ££a°O €£58°O? 98°72 6cv'O eTé*y YN7-OS YN/SYD VN/OW YNS TO 4f4let $ £8 Jad 4T ~ 90d WW ASNINGA O°Lot O°0rs 0'0498 «(O° VOT o*s O*v9 O°OLTe oct ore o°9t 0°026 ore OW Ze) HIN YN 4Y3117/039 OYDTW 4? $ £8 9390 At ~ 904 Mv Sa WdWys £ yOu S9VNIAN WALHOTAM~-SWIL OrGé 88°s 0&°6 O°0cF £F°S £E°9 0°09 ater 2£9°T i ATW 2aWIt (NIW) W4ALIATIO) YALYVANOA 4G AWNTON tSOILyy LS yas £vo't 68‘ 0S°6 aZ8°o ££°S £“°9 2eZ°0 2ttt L£9°T +/- 40TH eawii *e aay zo 69°9 OG°6 £°o0 OS LE°9 10 OF 'L LOT H Hd caWis 4444 "tT 31d¥l ses £ EE'e 27°0 «6 TaWIL # ‘ef av ce°a £ £o°e ~ 2o°0 «6 TaWIL # geese £ £e'v CZ £90 + TaWIiL # 298 000°0 ooT’o- » €00°0 VNNN 6c£'oT 682°6 GGL. eator wns N NNN 000°0 OoTto~ €00°0° IN YALSW JTAND / "SL0Y4 ONT TWdWVS £TO°S 020°S vas’ 24Z°S vos 000°O oot *o- ¥00°O na Lve°z gocte 6€6°0 616° £0N vG/ST/E WASINAY T00*O 000°O vo0"Oo o°ct <'st O°O£ OOF*O- OOT'O- OOT*O- T*0- t*o- T0- 800°9 z00°O SOt’o v's o°9 o°st a NW 34 A IN na WYYNDONIIN 4/4 VN NNN NIW/EW SY LY ADNAIDIGIS ZOS SNIWNSSY W3LIaTIO ANFLNOD N3L9M UINHIT NO WaSvd SI XIV 40 YS13W $ £8 94 6T - vs/9T/G WASIASN £60'O £to'o 0460°0 Bva’c 920°O 000°O £v0'O 620°O 9ST"O CGEre v£o"o 000°0o vto’o voo'o 600 vote TT0O'o 000°0 TEE'O oto*o T40°O aS6'b 9v0°0 000°0 13 OW vo vHN YN H MALAW DLTAND 7 WYMOONITW 444 ‘ALYM ONT WWYS NIW/EW SY LY ADNSTOTAAS 3HL WOMA WALGINDWI ANALNOD YaLoM arn $ £8 930 6T - i) 904 / 40S om 904 0°02 O'eT T*O- t°o- O°é6c Org ad NW MALIT 7 WYYNODONOTW O° tee r°o- O° s9e ad S44 69'°9 OS? OTe Hd OS's ££°9 £9°T ZaWIdL sc°s £ £0°v ¢ £9°0 «6T TAWIL MaLom9G4 40 AWNT0NA SHL WOMA TILE INI IVD "ND Mad GIWLAW 30¥eL AO ONTMVOT SSOW SHL WaT dSyanvad 680°0 ¢To°O £°é Br0'O Sttt £BE'O G'ty on 104 C£W79) CTW) *S AVIV O°aZ o°oeT o'09 ig CNIW) SAWS £ 390NaAY TILHOTAM-3WIL 98°8 £L‘S avy LOIW 05°6 £L°9 £9°T cAWIL 404 Sé°s £ ree f £7°O 6 TaWIL ONIWNSSY MALIATION NaLvmM90s 40 aWNI0N NO O4aS04 SI MI¥ JO £W Nad GLNTOS SSVW WaT ASM ANYS *y 31ayl 299 xy BLD Sea WOOK 6D £94 KAKKK CII 088 eK £9'°S 4688 HXHKK 66°S 2b eFONANA SATO Hd +/ ~AMa-Hde o MM S0Te 6462 °O O9F't cLE‘O &te't 895'°O TLUT v4r'o 26e°T BSb°O QEHI/(ATIS = wOS7EN NN SOTLYM MOIOW 00'¢ z*O- o'set OStET Z°O- O*brt OO'ZT Of 00H O° CBE OO°zE O° OSES 0° B6Y OS'Be O'O0BE O'TET 100 901 OZHIO NW/ NIJWHS ON OVW v8/9T/S WASIARY GLT*LO9SZY9°O B16'0 9£0°O Saws S&S Od S9VN3NY WALHOTIN~AaWIL ‘oO S£B8°O OOT*O- 88070 B6T°O e° on O'GoT o'aee BTO"O OF 0°06 os'8 GE's SLta & ‘Oo £824°0 OOT'O- S&T*O Sre'o ato- O°vbt O'6hc SbO'O Bret O°OZT SL°9 BLL G2°S bv. ‘Oo ofB°o 66'°2466E L98°0 £re'T O°00ry O'FBE O'2bS F40°O O'4T 0°04 00°S B4L°S Sf°v £ ‘oO 688°0O 08'SOc8 SZB'T Lents O°0SbS O'B6y O°FBS GET*O Ofc @°0L IFES Sé'y 80° Z ‘oO é¥S'oO BSttTtts 846s‘o €4B°O O°'008e OTST O'LBT ~ST'O S'*He O°sé 4B°t os°z Szrt fT - +/~ 901 OCHO (ADDS JOL Q¢HI (ATS ant TOA ig 1GIW @AWIL TAWEL #¢ O03. . \ NALAW NO/NYHOOMDIN 7 NI/WW/ \OI/W OMDIN 7 CEW/O) CTW) CNIW) SLYY ONTTdNYS NIW/EW SY LY AINSTIIAAA XOS ONIWNSSY 31937109 YALyM|Os 40 aWNTON 3H1 WOMS MALY INIWI LNSLNOD NAL¥M AINGIT NO Tasvd SI Iv 40 EW Nad BLNIOS SS¥H $ £8 O30 TE ~ 904 NIST ASYSNG *e Fal vB/9T/S MASIAAM ozt*o £v0'o 8zz'0 ZU°T tsalivy LWs vas z40°T 000'0 c60°t 026°0 OOT*O~ OOTSO- OOTTO~ OOT'O- SEBO os's GE's GL'2 & SEl't 000°O Getet 1S6*°O G68°4e Tavis 6£25°0 BOLT £BL°0 GLP SL°% S4°S + 0246°0 T00°O 0€4'0 TéL°oO ESESTT OOT*O— OOT'O- TéLb°E 088 °O 00°S S2°S Sé*v £ £28°O T00°O z248°O0 Oce*O0 Z0b°CB B44°E 6TS'*oO Tav't 688°0 ID'S ge'v sore = Babet T00'°O L90'T 2£S£°O 826°6t Geert Berto 056°0 405°0 2ert os’d gert +t 19S S+N/U4+H SHN/H S+N/PHN FOS/EON YN/~OS YN/YD YN/OW vNs 19 +/- LdIn TAaWIL YTAWIL # SFNNNN SOTLYMY INA WAINDA a a ‘ $ £8 930 T£ - 904 NVALASHSNYS ‘e a1avl ea/9tsS Gastnay O'abs SEBO OOLES O'O9TE O°96 0*0¢- o°7ét orotos T*0- Z*0 6L2°9 GE'S 8=SL94 O'6EE £8L'O O'OhPT O°O4ET O25 Orce O°OkT O'O9Te O°BE £°O 6y'9 BL°4L SL2'G O'ebS O8F'O O°OTST OETP O'cy O°or- O° 0L£- oroere Oct are S9°S &ZL°S Sete © O'sBsS 688°O O'09EE O61 0°04 Oret o'Gce O'OTse Off £°S £9°*S Got¢ gore & O'far LbS'O 0°4662 O°S8e o*S8e oret ored O'O4SF O0b ot 66°S O6°% cerr (ATIS +/- vos £0N mie) SW va vHN YN # Hd ZaWIL TaWIL ¢ OMDIW / VN N NNN MALTV/OS OMDTW a a $ £8 930 te ~ Sod q731dSNaNed "T 214a¥L 300 va/Tt/£ Wasiagay OOT*O- OOT'O- OOTTO= OOF'O- OOT*O~ OOF*O- T°0- T'0- T*O- T*O- T*o- T°0- GL£°9 GES SZ°*Z S ¥00°O To0'O to0'o £00°O T00°O 600°0 0°O£ £°be 0°0e orgd O'LEe O°04T oeto SLed Sé°s £00°O £00°O T00'O 800°O To00°o 0z0°0 O°se £°6E o°st orottT O°st O'09e B9I'S GLYS Gé°ev £ voo"o ¥O0°O z00'O £t0°O £00°O 9£0°O O*ee coe O'9t o*ooT Ore O' ree £9°G Geb 80's Zz £00°O voo'o £00°O 0c0'O vo0°O £v0°o o*ét G*Sé Q°etT o'er Oo O'E8e 66'S OS°S se°t fF A IN no 44 NW 34 A IN na dd NW ad Hd caWIi TAWIL # NNNN MSLAW OTAND 7 WeYNsOMOTW 44 VN NNN MALI / WOY9OMOTH 4/444 *31¥M ONT TGH¥S NIW/EW SU LY ADNSTOIA43 40S ONTWASSY UALISTIOD NALYM90d JO AWNTON SHL WOYS DALY INIWI ANSLNOD NSLYM HINDI NO GAS¥H GI MIY 40 YALEW *ND M34 GIVLAW FJOvNL 4O SNIDVOT SSYW AHL $ £8 9a Te - 904 TIATsASNANYA ‘Gg aay : v8/9T/G WASTAAYN : Br9'°cT 99BNS LOT'S 4TT'o 810°O Toto SBE'E 990°O 900°0 940°O GAWNYS S 4Od SOVNAAY COLHOLAM-aAWTL Tre? 066°T T4e*2 ¥S0°O0 000'0Z~ 020°O 2oort a0Tt'o- 000°O 8t0°o ote 0°06 o5°8 BE°sé S2°a GS TGL'6 e1ree ezere’ £80°O Zoro Z£TE*O S9o°e 6F£0'O 000'0 Svo°o S°et oot S2°9 G2te G's + ZOTTER Ta? e48°E Etro O00'OT- 000°O£- CPT'E oro'o 000°O 920°O orét 0°06 00°S GZ°S Géte o£ She ee 602*oT bLOte 84T°O FZ0°O gsTt'o GLONG aZoro Q00'°0O sét*o Ore E°Od 9OE Sa°y 680°L s OLv ct £Le'S 989S Goz"o0 bc0rd zL2'o O9f*tb ovt o 000'O est*o S°8e O°sd Z40°T osté geet wns vos £0N Ww OW wo vHN YN H ont WA 1g QU ZaWIL TAWIL # NNNAN MGLaW DLGND / WVHSONITIW 444 C8W4A9) CTW) CNTW) ‘ALUM ONTIdNYS NIW/ESW SY LY ADNSIDI Aaa 40S ONIWNSSY HLI99T109 MaLem—od JO aWNIOA 3HL WON TALYINDIWD ANALNOD NALOM INOLT NO Gasvd ST YIY 4O £W Mad ALNIGS SSUW $ £8 930 Te ~- 904 (aT ASMaMvd “te 31a¥l 301 Sv8' ot ves° vrE st T9B* oT PSS 0E Fab OL wns vos NNNN Bere vet's Tete Ors 9B°E£ St*s vere TEs eFJONSMAAATO Hd ~Ad-Hde O0T*o- OOTtO- Bke's TGG*t _ OCHI/ CATS NN. SOILYY ie aa Tto- oL°T T*0~ 9L9 c'0- BL OT c*o- 0A DOL NIW/ NU/HHS bAa/9T/E MASI AaM Ot ZéS*b BT£LtO 090'°O ¥GE*o 258° £50°0 900°0 é£0'o S31dnYS b Od SOVYINY NALHOT3IM~3WIL 456 Toc'o OOT*O~ OOT*O- vRETE OOT*O- 000°0O Tz0°o Ve vas Bort LTE Btrt v GTOrEe céc'o OOT*O~ OOT'O= L6E°E£ O0T*O- 000°0 E1O"O att v's 89°0 ét°t gT’o £ £68'b Boro evoro vse*o 605° TEOO 000'0 9v0'O 8°9 €°es TE'O- LTO oB'o- 2 v6srd 697'°O 440°0 O0T*O- FTL? $20°0 Too'o 490°0 vor 0°09 £eE'T- E£B8°O- L£Bt'tT- FT £0N a) ow ¥2 bHN YN H 207 0A aq AQIW CAWIL TAWIL ¢ MALSW JIGnd / WeHOONSIW C44 ¢ (EWS OTR) (NIH) ‘ALUM ONT TIWVS NIW/EN GY LY AINATIT4AIa ZOS SNIWNSSY MALIZTION YALYM9DA AO AWNIOA SHL WOMS W310 INI WI LNALNOD 43L¥M TINOIT NO dasva SI YT¥ 40 £W Yad ALNTOS SSYW $ 68 NWT 2 - 904 ATFTASNANYA ‘eo Del v8/9t/2 AASIAAY 900°0 voT‘o TEE*O £v0"°O S31dWUS & NOd BO¥NIAY JALHOTSM~-aWIL boc°r ¢éerTt OOT*O- OOT*O- OOT*O- TrO- T°0- T*O- Teo'o Te v° 6S 89°T cv ett ¥ tae°t Sek OOT*O= OOTTO- OOT*O- TtO- t*o- T*o- etoro att v°4S apo ater arto £ £co*t Lcort OOT'O- £6zrad S2O't c*o- O°oTe O'ceL P¥O'O BF 2°BS TE*O- =ZT°O OBt0~- cot*t cot OOT*O- 9£2'0 atett e*o- O'vSE O°'6¢S 490'O b'OT o*09 EL*T- £8°O- E£B'T- +/- t/- J0L ocHd (ATS 301 OCHO (ATS 3a 10" 1d LQIW @SWIL TAMIL ¢€ eyo. \ SALON “NO/NYMOONIIN / NIZWWY NOCIVW ONDIW 7 CEW/9) CTW) CNIW) ALUM ONIWNYS NIW/EN GY LY AQNBIDTSSD 40S ONINNSSY NALIZIIOT YALYM9NA JO aWNTOA 3HL WONS TaLYINIIWI LNSLNOD Y3Lvm HINDIT NO WaSsvd SI YI¥ JQ fW Yad 3LNIOS SSYN $ v8 nor 2 - 904 DIGLASNSNYGT ‘g aay va/s9t/c DASIASY OET*O £v0'O Beco cT°T tsOl.yy Livs 93S OOT*O- vod'0 190'0 wacto 098° OOT*O- OOTPO- QOTTO~ OCOT'O- HactTt go't 2Ve ett ¥ OOT*O- 428°O 900°0 6cg°o0 £B2°O OOT*O- OOT'O- OOT'O- OOT*O~ TSE'T 89°0 att at'o £ SOL°O 8L£8°O T90°O éee8'o 6580 KOTTEST bles cssté Bre? Leo°t TE*O- €T°O og°o- é gketo Eea'e TOOtO vearo BL4E"0 25°65 OOT*O- “LE6'T v90'v cott E£'T- £8°O- LB't- + bOS/(ATIS StN/Y4H StN/H SEN/HHN #OS/£0N YN/POS = YN/I VN/OW UN/19 tf- ATIW e3aWIL 'SWIL & MVTOW / /NNNN SOLLYH ANJIVAINOA “fle $ v8 Nor 2 - 9G4 WIALASMANGA ‘Ss avi vB/9T/Z DASINAM T'o- T°0- vdE'T OTOTSL O'OB4T O° O4E T°0~-- Tto- O°OTeL = TtO~ par 4 eT's “tee att ¥ T*O~- T*O- THE*T OFO29eT O'OSTe Of Fos T'O~ T'o- -O'009ET Ten~ 6°!£ Ore €Tt Btto 8 £ OroTS O'tel LEO'T O° OEE O'004T O° ZT Otbe'd OORLES O'O8kS O°4e Ted SES éT°O Oo8°O- = O'VSE O'6S SOT'T O'0RSr OFO4cZT OTST 0°c4 Poe O°ORES Ode ate TI'S «£8°O- f£eBtt- tf OCH (AIDS t/~- ros ON WwW OW 8. PHN YN Hu Hd CAHILL TAMIL # N VW OMOTLW 7 NNNNAN MALT V/O2 OTR ar a an a $ yO ner 2£-~ 904 WVaTaSyaNed "Tr aL 302 OOT*0- oot*o~ OoT'o- g00°O ia) VN NNN OOT*O- OOT'O- OOT*O- OOT*O— OOT*O- O0T*0- 800'0 £00'°0 IN no MAaLaW JIAND / *Ji¥M ONT TdWYS v8/ST/E MASIAAY OOT*O- OOT*O- O0T'O- T*O- OOT*O- = HOT'O- OOT*O- T*O- OOT*O- OOT*O~ = O0T°0- T*o- £60°O 9900 cet'o orcet 4d NW 3d A WHH9OMIIW 444 N NNN NIW/ER SY LY AINGTOIA Tt0- T*O-~ T'o- Or9tt IN t0- T*0O~- T*0- O'S ng MALI / WYNSOMOIW T*O~ T*0~ T°o- O° OLE qd INALNOD MALYM GINOLT NO dagud ST yI¥ 40 YALaW ‘NO Mad T*O- T*O- r'o~ o°cs NW $ ba NOE 2 - HOA WTSLASMANYA TtO- vttS LTE T*0~ ots éttt T°O- SI'S é1°O0 o°sost it’s £B°0- 34 Hd eawit “ft 406 ONIWNSSY DALIITWI YaALeM9O4d JO AWAION 3H WONA MAL INI WI SLAIN JIWAL 30 ONTVOT SSYW AHL avyL ert BT°O 08° 0- £B°T- TAWIL net * 303 Loe oore KXKER TORK WKEKK ober eS INIYBS ST ~AY~-Hd oot ToZ cz0 oot est 0c8 OZHI/ NN oL°£ o9°st 00°'9 O2te o8'ét OS'S mina) C6"? SS‘? vL'g cL'g 24'S BL°? Hed ‘o- *T 'é +o- ‘Ee °é CATS SOILYN T*o- 2*0- Z*0- Fto- c°0~ OTELEE d01 780 LO £e6 as-) 206 vé0 +/ oud ooT"o 9ET'O g0z°o ooT'o ée°O ete'o vOS/(AIDS StN/¥+H yv1OW 7 7 NNNN t*9o- O*vet O°svt t°o~ oreTt O°cEt O€HQ NIW/ NU/WHS ON CIR va/s9t/s W3asl ¥08° 0205) OOT’O~ O OOT'O- 0 Oor'o- 8) OOT'O- 0 ooT'oO- ¥ 08°040S + 301 0 Ag £r°o ot'o- spro St'oO Ot*O- veo 9¢'°O eHo SaT't o0T*o- £18°O GLL°O0 ooTto~ T4s*t Z62°T (ADDS T°o- t &°0- O° €*o- O° t°'o~ tT &*O- O° O°ELEe O° 304 oz o- bet obt o- att zet HO ¢ T*o- O°sere O'zoe T*o- O'9Ze O'vbe AS N MALSW ‘NO/WENDONIIW 7 NI/WW/S NO T/W ONDIW 7 SLUM ONITANYS NIW/EW GY LY AINSTIIAAa XOS SNINNSSY SHL WON T3LYINGTWWD LNSLNOD Y4L0M TIADIA NG G3S¥d SI YIY JO £W Y3d SJ1N10S SSUW ¥8/91/G WASIAIN *T 980°T *t Zorrt *O £86°O ‘O #96°0 *O £046°O ‘tT é0'T - +/- 03. - BS8'0 O1Z'o 9T6'0 - £86°0 BL6"°0 0&S8°0 000°0 000°0 000'°0 000'°0 900°0 000°0 9s8°O OTZ°o 9T4°O £846°0O BE6tO 0S8°O O£Ee'o ecto Scc°o vec*o PFE TO ¥STt'o S+N/H S+N/*PHN bOS/E£0N SOILYY LNA WAINGS va/9t/S GasIn3y T*o- 9B80°T O'090S O'sezZ LoEtT O°04688 O°COE £2440 0°0b462 T*o- v96'0 00246 O°9Ze 206'0 O'ORGE OtvrE vo0'T O°0SCE CATIS t/- vas OMDIW / o'00et 0°90E O'Gée O° E99 O°dbt Ordvt O'OOTT O'Tte O°Gey o°cos £0ON VNNNNSN 9°08T O'ete ai) $ 68 NEF OcT*O ooT'o- 828°bs ooT*0- ooT'o- LB OL S80'SS YN/¢0S $ va NUT O°sSs 0°6 Oreet Ores orot O°é OW ¥so°o zco"o vor 'o 080'O 9£0°O zet’o LET°O am C£H/9 £°e 9°st org Le B°4tT S°é ta) > CW) et - 904 WaT ASYyaNVA £b0°O ooTt'o vse'e ooT'o oot'o BOTS vere YN/¥D Be - 00 0c - 00 - 00 £6 £3 YN Z*o T*o- 0 z'o + T’o- 0 T*0- 0 Lane) 6 To tid OW vi Zt*t Ot *o- £9°E Ot*o- OT*o~ Aare cL°S aa) Sst £t - 904 T1a1ISYaNvd ord ore Oty ore Oru ore ¥ 8S WT 6€ 6 at vt a NOLO OMIM $ ¢8 NUT £t- 904 Or OoLe? O*°ooTe oO 00LE 0°0c6S 0*008E O'O4TE vHN IVT AGM t*O- o'ty T*0- Fto- o's o°4S UN / Satdwys 9 MO4 NWALHOL3M- 3WIL aovyany O°o? OS'S 0°09 OS’t O°0£ SZL°E O°Or Sere 0°09 oS'Z o°or SZ°t A ATW (NIW) ‘gf DWVL tSOTLyy 980°T OS°S 2o£’t OS*b 2£26'°O SZ°L£ ¥96°O Sc°e 206'°O OB°s vsort s2rt +/~- AQIW *e JWVL tte f6°9 £°0 sc°9 é'o v9 ato fL°9 et C4°G vo BL°9 H Hed S444 44 aNva ‘T al 0o'9 00°s 00'v os’: oor°e 00°% eawil W4aL937109 43L9M904 40 3WNTOA L7eS 93s 00'?9 00°S 00’ oss oore 00°z caWth 00'9 00°S 00'¢+ os’: oor’ 00*2 ZAWIL 00'S oot OoS°e 00's. 00° os't aANM TOS TaWIl #¢ 00°S 00*¢ On°e oo°e 00°? ost TaWIL #@ mam TNs 00°S o0'¢ og't 00°£ 00°% os’t wom Tine TAWTL # 304 6te°ze Tetett LTE *6E TSE *O0E worst 656°LE £I6*CE WNS NNNN voorert O92°S Tee’ at L£6E°1T £08°8 OLE gt G42 6T vos GLO°E Bob's £PO'E vac'£ Sobre B4b°£ era's £0ON vaAs9t/S GASIAAY 825°0O T9T°O 6bs'o Ztbo 695'0 zes'o Otr't 19 MaLaW JTAND 7 WY 620°O LvE"o gTO‘O TIO'O oET'O £EO'O 9E0"O yTo"o SW ‘JLUM ONT TINYS NTW/EW FHL WONd TAL INDI LNSLNGD YALYM OIAe vsT°o o0T*o- 840'°O 00T*o~- oot 'o- e&E'O eét°o YN 000°0 000°0 000°O 900°0 000'O 000°O 000'O H Sst vs80°O czO"O vort*o 80°90 9£0°O zeTed 4et'o amt (£W79) Fe 9°St o°9 L°S Btét 6°6 wa CW) $ 68 NUT £T - 904 TIaTASHANVA 0°09? 0°09 o°o£ O'0£ 0109 O'OEr a CNTW) Sa1dwyS 9 x08 JOvNAAG DALHOLAM-3HIL OS'S Ost SL°] Sere OG°S Gait LUI ‘ye WWUL 00°? 00's 00° OS°f£ O0°£ 00°%s caWit S ¥ LY AQNSIOTASS %OS ONIWNSSY 9791997909 HALYMgOs 40 SWNT1OA 11 NO aSva SI MI¥ 40 EW Yad JANIS SS¥W 00°S 00'+ OSte 0or£ a0°s ost wANM ££ O TaWIL # 305 OOT*O~ OOF°O- OOTTO~ OOTTO- OOT'O- OOT*O- T*o- T*o- T*o- Tto- T°O- T'0~ OT*O- OO°OT 00%) & £T0°O S10°0 00°0O SE0°O oTo'o0 £60'°O O°tve O'ore o's O'OL£s O'OST “O'eGet B9'e 00°6 oo'n oF A IN no ad NH a4 A IN no ad NW a4 Hd GAWTL TAWIL # VN NNN MaLaW OTHND / WYYNDOMOIW 4447 VNNNS MALI 7 WHsOOMIIW S44 “SLUM ONT VAWYS NIW/EW S 8 LY AINSIITA4S 40S SNIWASSY T43LI9 TION MALYMD04 40 3WNIOA BHI WOMS aLyInoqw3 AN3LNOD Y310M DINGIT NO WaASvd SI YIV¥ 40 H3L4aW °ND Nad SW13W 39D¥M1 40 ONTUYO SSYUW JHL # 6B NUC £ ~ 904 NOTNLLINOW °S a1a¥L ¥B/9T/S d3ISIN3y OOF Ze L£90'S OZ4E°4T EEO £S0°0 S22°0 vias 000°0 TeT*o 0S0°O SANUS € YONA SOVYINY WALHOTSM-3WIL B6c°SS 659° 6£Z°UT = STe*O B£0°0 OOT*O- Lea's OOT*O~ OOT'O- £FO'O o°s 0°09 05°6 O0°OF O00°6 € cor*ze 922°9 TOS*OZ = O092'0 £90°0 GL2°0 TaZ'v OOT*O- Tet'o £90°O o°oT 0°09 os°B 00°6 oo*s Wns vos £0N ny OW vo VHN YN H on 0A in 40IW ZAWTL TAWIL ¢ VN NAN Y3aL9W JTANS / WYNOOMDIW 444 ¢ (8/9) CW) (NIW) “SLYM ONITGWYS NIW/EN SY LY AONSTOLAdS 20S ANIWNSSY 3L93TIGD YALYMANA 4O AWNTION HL WO4Nd DALYINIWI LNBLNOD Y319M HIADIT NO GaSed SI MI¥ 40 £W N34 31N70S SSeW $6 vB NYUr £ - 904 NOIMLLINOW ‘vy ale ¥8/91/S J3aSINay 000°0 690°O 9ET‘O 0S0'°O SATMWUS 2 NA S9UNSIAY UALHOI3A-3WIL OFS OT*o~ ¥tO°S wtore OOT’O- 9£0'0 zot*o Z*0- O'9£ O°sé ££0°O O°S 0°09? 0s'sé 00°OT 00°6 Zz 6S°S B9"s B8Z0°T TOTT OOT'O- ZOT*O TLT°O z*0- o°ts 0*08 £90°O §=0°0r 0°09 os*s 00°6 oo’s ff aFONSYNS 440 Hd +/- t/- 904 Q2Ho CADIS I01 QZHO (ADDS ant wa Ait 40IW ZAWIL UTOWIL ¢ ~Ad~Hde «4400. N MALSN *NI/WOHOOMIIW / NI/HH/ NN IVW ONDIW 4 CEN/9) CIM) CNIWD 31049 SNITANYS NIW/EW S Y LY ADNATDI SSS 40S SNIWNSSY G31937109 431¥M904 40 aWNION SHL WOMS T3LVTNIWI LNZLNOD YALYM GINOIT NO Gasvd SI MI¥ 40 €W 43d 3LN 10S SSUN $ 08 NUF £ ~ 904 NOTHLLINOW *e AML *8/9T/S U3SIAIY ozt*o £vo'o0 8cz°O aver $SOTLvy L105 vas . 629°C §90'0 00S°0 OOT'O~. O0S*O BLS OOT*O- OOT*O- OOT*O- OOT*O- H#TIO*Z 0S°6 OO'Ot 00°6 Z 69S'T Z80'0 9t8°0O cOLtO vts’o TES*S OOT°O- OOT*O- OOT*O- OOT*O- TOT*T os's 00°6 oo's 6f OZHI/CAIDS —vOS/( AIDS SHN/YtH S#N/H StN/PHN bOS/E0N YN/bOS YN/YD YN/OW ¥N/19 t/- AdIW @aWIL TAWIL @ NN SOTLYY (HU 10H ZFINNNSN SOTLYY LN TVAINOA 74s $ 68 NOT £ - 904 NOTHLLINOW *2 PML v8/9T/S dasInay 00's z°0~- O°oe 0'S6 vIO°'S O°'OTS62 0°0489 O° IGF O°bs T*o- 0°006b) T*0- t*o- OT°O- OO°OT 00°S Z oo°or Zo- ots 0°08 TORT 0°096T 0°096% O°0TT o’re 0°O8S O'O9SE = T*O- £'6BOr BF*'Z 00° oo's fF WwA 201 OCHO (AIDS t/~- vos £ON as) OW vo vHN YN H He ZaWIL TAaWIL ¢ NW/ NIVWWS ON OV/W ONDIN / NNNNNAN 43L119/03 OMDIN 44444 «4 $ 68 NUF £ - 904 NOINLLINOW *F Jd. 306 OOT*O- OOT*O- OOT*O~ OOTTO- COT*O- O0T*O~ £T0°O 800°O 200°O 800°0O Too'oO 8T0"°0 A IN no 4d NW ad N NNN MALaW DIAND / WYNDOMIIW 44st “aly 9TE"S 8462°T 9vE'S OOT*o TTO°O 040°O 628°O 000°0 8t0°o0 8r0'O SA dWUS Z@ 4O4 SSUMAAY N3LHOTIM-AWIL cEb mS £08'O act*t STO"O 000'0é- vv0°O 6Z£°O OOT*O- 600°O oTo'o gt 9°09 OS*6 OoroTt 00°6& Z 000°8 682°E vISts vat'o TTO"O 9460°O 6Z2°T OOTtoO- 20°90 £80°0 oret 0°09? os*s 00°64 00's Tt wns vas £ON a OW ¥) vHN YN H ant WA 4g AMIW Z3WIL TAWIL #¢ NNNN MALaW DIGND / WYNOOHIIW 44 4 ¢ CeW79) OW) (NTH) ‘ALUM ONITdWYS NIW/EN SY 10 ADN3IDIASS %0OS ONIWNSSY W4L93T1I09 H3L¥M904 40 AWNTON FHL WOMNS (3L97NIWVI LNILNOD YaL¥M GINOIT NO dasva SI YI¥ 4O £W Yad ALNIOS SSUW $ wB8 NUT & - 904 NOIMLIINOW *e ATV vB8/9T/G M3SIASY £09 L6VE9TO*O 620°O zZ0°O Sa dWHS c MOA S9VNANY DALHOLAM-AWIL 06°S vores 60T°T ZeTet OOT*O- OOT'O- OOT*O- T*O- $°O- T°0~ O10'oO §=S°Tt 0°09 OG*°6 OO*OT 00°64 Z ove 1S’°f Wao°t £20°T OF 'Z6¥E Lv0'O 980'O OTO9ER O°BT ote 480°O «O°ET 0°09? os’s 00°4 oo*'s 6f sFONSMaAATO Hd +/- t/- 901 OZHI CAD)S 904 OZHI CAINS oN7 OA ig AIHW ZAWEL TAWIL ¢ ~Ad—-Helo e4HOI. \ MELSW 'NO/WYNDOMIIW / N\I/WW/S \OT4W OMNDTH / CRW/9) CTH) CNIW) ALYM ONTIdNYS NIW/EW G ¥ LY ADNRIDIA49 %OS SNIWNSSY W31L93TI09 HaLvmM904 40 SHN TON BHL WON4 T3L9INITGD LNSLNOD YALYM WINOIT NO W3SVE SI YI¥ JO fW Yad SLNTOS SSW $ v8 NOT & - SOA NOTMLLINOW ‘fg aay vesot/S WaSINIM : Oct"o £r0°o 8ec°O aver SSOLLYY LIS vas oot *o- ooT'o~ TrB°o SSc*o 2as'o TET'E OOT*O- OOT*O- OOT*O- OOTTO- att OS's 00°oT 00°64 & cect £60°O 006°0 £ECIO 999*O 9426°0 OOT'O- OOFTO- OOTTO- OOT'O- 440°T os's 00°6 oo'8 fT OZHI/(AIIS bOS/(AT)S StN/¥+H S+N/H S+N/VHN bOS/£ON YN/bOS YN/¥O VN/OW YN/ 19 +/~ STW COWIL “TaWIL # NN SOTLYN HYTOW ZZ NNNN , SOLLYM LNAWAINGA a A $ vA NUT & - 904 NOTMILINOW ‘2 a1av0l vB/9T/G WASINAY os*t tT 10- T"0- T*0- ZET°T O'089T Of004T O'c¥ O'oc- Oecd o°00TE = T'O- orctlé vo's OO°OT 00°6 € OO'eT Of0O9LE O°8T O'le Z40'T 0°0L9 O°bS? 0°09? oot O°ss o°cs8s8 t°0- O°S60£ TS*E 00°6 oo’s Ff wA 301 OZHOI (AIDS +/- ¥OS £0N a) OW v2 vHN ¥N H Hd caWIL TAWIL # “NWS NUZWHS ON OOT/W OMSIW 7 NNNNNSN MBLITV/0O3 OMDIW L444 . $3 6H NUP & - 904 MOTHLLINOW *T 31avL T*o~ T*o- T*o- T*0- T*O- T°0- ro'g O0°OT 00°64 & 0°ost T'64 O°be 0°96 O°dt O'v0d TS*S 00°6 oo's. tT fa) IN no qd NW a4 Hd eaWIi tAWIL # N NNN MALIA 7 WOMN9OONIIW S444 M ONIVNYS NIW/EW GY LY ADNSTIIAST 40S ONIWNSSY M3LISTION MaLYM9NA JO 3WNIOA AHL WONS WALYVINIWS LN3LNO9 N3LYM DINWIT NO G3aS¥A SI YI¥ JO N3L3W OND N3Ad STWLGW J0VML 40 SNTTVOT SSVW SHL v8/9T/S MASINIY $ v8 NUT & ~ SO NOINLLITIOW ‘Sg 31aVL 307 16° Bb'e SEt¢ ti a 2 ary co’ “2° TO’ 68°L 80° eSJININASSI ~-AB-H: Bt bre gEetb Leo eo grees Leib 4a'b To°e cote 00°¢ co'e 1 Hd de tte L09E Té0'’ 00 T*0- BEL°E BL vc LOGE gag’ t /9ve'T £2z°T OFHI/ CAINS NN 00'bz 00° 0£ 00'bT octt 00ST 90°09 00°09. 00' Re ONb6E oO'RS 4A oN? SOTLUM O'segt oroset z*o- Tt0~ O°O4or O'sne OSes OGbs OrEle Onan Ww V/s 200't 8960 vzort oos*t ToOO't 656°9 GSE6'0 200'T £90°T £26'O +/- SOTO 62T°O T6c°O ooT*o- eSE'O é£te*o vI£E*oO £te°o 94ENO SOT*O e8/9T/S MASINAY £BZ*ES774680°O oTo’t Te *yadét 9¥0°O 996'°O 48°8ETS O50'°O sco’t OOr*o- Bf0°O Tos’t OOT*O- OOT*O- e0o't TE*T94T 090°O 096'°O 00°E48L 9ZT°O 286°O SS*boce vit*o 600'Tt LO°FELE F¥T'O 990°T 94° V99T SL0°O 646'°0 cLtvtte 2ct*o +/- 30L O@HI \ AL¥Y ONITdNYS NIW/EW GY LY AQNALOTI49 XOS ONINNSSY M3LSW SNI/SWHHOOMIIN 7 NT/WH stc'o vorT"o zert'o 99T*O ooT*o~ Stc°o 6S¢7°0 Scr'o 8rz°oO 60T'O cLtto (ADDS orsest or4t O*OabT O*bT €°o- O°bt r°o- T*O- OfOZ0T O°£t o'rss = Of9t O°scé O°St O°zZ6L O°¢T or4tZ Oret 0°268 O'TT 904 O®HD ¢ O°°"k O°Le or4s T*o- O'by O°6e 0°0¥ O°Ze o°ee O°br Als \ VW OMDIN 7 S6T°O 060°0 ozt*o T60°0 0z0'0 £St*o £9£°0 egc°o £82°O £6t*O vas’o a7 ceH79 O'be O°or O°vtT stt O'ét 0°09 0°09 O°Eb 0°4e o°Ees 10n >» CTW) B°90T e*oot at? 0°Or 8°6r ¥*s? B's 0°O9 o°09? 2°ss ia CNIW) S3IWMWYS OT MOA SQVNSAY DALHOTaM-3WIL Tae tr 99°6 crs gLg 60°9 ees 62°f 0s°z os't vs‘o0 10TH ozat oS*ot £8°8 00°e os'9 L49°S esty OOrE 00°¢ oort caWIL cG*Ot OF £8°B osd os’? L£9°S 8s’t 00°f 00's oo*t 80°0O mw NM TOON OD TBWIL ¢ 931937703 43aLvm904 4O SWNTON 3H. HONS D3SLYINITVD LNFLNOD Y3LYM INDI NO 13S¥4d SI MIv JO EW Mad JiNTOS SSsuW # v8 NEF S - 904 NOIMLLINOW vAs9T/S MASINSY 592°0 GAL°0 bao 1S2Z°0 958° 916°O 6260 Z16*0 Zba'o 994°O vOS/CATDS S4N/ 94H YYTON ZZ NNNN SOT*o 580°O ¥90°0 eto"o 9Tt*o cET"O £eT*O B6T°O 48To 28T'o S+N/H 099°O 669°O oGZ*O 6£2°0 6£4°O vac'o O£B'O vtZ:o £89°O 644°0 S+N/CHN vB/9T/S TASIASY Ordt Oren oto’t O'et O'LE 996°O O'et O°<S Scort T*o- T0- Tos't Ost O'te cOOtT O°et O°éEr 096'9 O'ST | O° Ob £56°O oret ode 400°T Orel Orbe P70°T ort Orel 64£46°O CATIS tf o'bag Orth O°e6E Orare 0°0GE O'29E vO NN Orth Ose O'ort o'esl O° VNNS 809'°O £6S°O 26£'°0 Brr'o Et?*oO 689'°O 989°O 606°0O Te6é'O H16*o v05/f£0N SOLLGY 8cc°0 2tt ET 290°t ooc'*z 6€S°O vcB°O 00L£'o 092°0 ~ GOT'O- OOTtO~ £99°0 ££S°O 00e'O O9r"O 90s*oO O08 *OT £EE‘O ££6°O EEESO OSS*0 osc'o SL6°O ¥N/OW vN/ TD 4st nor S ~ S04 NOIMLLINOW Ozt'o £v0°oO 009°Sh O002't bere G9LI9 O0E*S6E OOS OOT*O~ OOT*O £LL*LG LCLL'L O0Z'Se OOF'T 005'Z2T 000° EEL PB LLEE ELE'Sh OOS*E OSLZ°99 = OGE'E YN/bOS YN/YD Ana twAINOA $ +8 o°9T ore 0'6 Ors Or£ O'S r°0- T*0 ore 0'0 ort Ore ort 9'8 ort 0'o0 ore Ort ort Ore OW Bt $ 6S NUE OG 1 tid 13 3 é T T é 1 Naat O°SEL 0°09 0'66E O'EPE 0°86 O°Ods O°45E O'eee O° APE O° 66F vHN MOM LL ost arét o°or Foe Or£ 0°s Ors Ore o"9 O°b BN / ‘Ss a1avl tSOTLy9y LVS ¥3Ss OTO'T 9946°O Szort TOS’T coo°Tt 096°O £56°O 600°T 990°T 626°0 +r- ‘2 a1avL BteIT c'9S “O°RE 6°S a9 2°ss 261g? G*G6 o*oor GB°G6 H an a ae TNOW ‘Tr ani Ta°tt 99'S ze°B SL°9 60°? £US 62'S ose ost ws°o idiw vhtf Ge'bv Leite LE°S LEY Late Té'¢ E0te 0Ote 2o°v Hd o2*et os*ot £8°s 00°24 0O8°9 £9°S Ba" oo°e 00°% oo't CaWIk 26‘OT OF £8°B8 08's os*? £9°S BS‘ 00°£ oo°e oo't 80°O SAAN TNVA GH TaWIL & Z6°Ot OF fae : ost. os"? 29°S este oo'e 00°% 00°T 80°O nem Tonka TAWTL # 308 TIO'O 900°0O eto'Oo 700°0 900°0 ¥O0'O QOT*O- OOT'O- 800'°O g00'0 610'°O TTIO"O 200°0 900°O zTO*o 200'0 oto’o 900'°O g10'O €10°Oo A IN VN NNN ‘AL¥M SNITIWYS NIW/EW G ¥ LY ADNSTOIAIS %0S ONIWNSSY MAL Lote?’ oes'é v86'9 £56°E T46S°S 9LE°E TE TE 90L'T 489°O 9EL"O TOT crest BBE’ OF OLE’ ZEL02 TOFTE 6906 £8¢ tL 656°9 9ES°E vsé'et Beaty “Wns vos NAVAN 200'o T90°O TO0°O oot 'o~ TOoO"0 ¥O0'O 100°O ToO"O T00°O 900'°O nD coos 20Lf°s 6tert 28Z°0O BAL YO Beet PAPE Lec es OL0°t 670°E vEaSS £0N vA/ST/£ d3SINay 500°O v00'o T00°O O0T*O~- 500°O o10’o 900°O 600°9 800'O £10°O ad M3LaW JTHND / WONSOMITIW 200°O £e0°O t00°0 T19°O 000°O £00°O Q0T*O- =OOTTO- 000°0 400°O TOO'O S00°O T00°0 €00°O 100°O £00°O T00°o eto'o T00°0 Sto°o NW ad S4tt vB/9T/S NASIAAY 890°0 SOLt*o 450°O ve0*O $rord 600°O O£0°O gato B20°9O £e0°O £50°0 mie) MAL3W DTHAD / WHNDONOIW ‘FLOM ONTTWYS NIW/EW GS ¥ LY ADNSTOLAST 40% SNIWNSSY MaALIATIOI MALYM90 40 AWNTOA 3HL WOME CALYINDWI ANSLNOD NALYM HINOET NO Wasa ST MIV JO £W Nad FLNTOS ssun £0040 Z£190°0 £TO"O £00"0 OoT*O~ ¥00'°0 600'0 £00'o £00'O §00'0 500°0 OW eet’o veLr'o OLE" gv0'd oot 'o- TEO*O TS0'O Tro°o 280°O TB0'°O ooT’o vo Orétt O*O0r O°S? T‘o- O°ts o°Tts O°8E O°ee 0°OS O'°8E A VN NNN CPE NT GLUT véd°O e59'°O TET'O 0c8°O Btls O£9'°T 6442°T oTe't e9L°S bHN O'2? 0°0sS O° 6s T*0O- O°sEe 0°6E O°fe Ove O°0E o'te IN se ee co Cr coe . - * eoococoroos 5 Oo WQMNTTONROSD eal n Orts Ores oroT T*0- O'te O'9G Oke o°0Ee 0*°0b Ore qd MALI 7 WYMNOMOIW z = $ v8 Nor S - 903 NOTMLLINOW 8c0'0 T£O'°O Z£v0°0 Tz0°O o0TtO- TI0°O evoro eTo"o oc0t0 £60°0 S£0°O YN STo°o oto°o £00°O £00°O 000'°O 200°9O 0E0'O 9to'o 820°O oco'o L£O°O H 7f4rt vac’ m7 (£W/9) O'bz O°0£ o'er st O°éat 0°09 0°09 O'ee 0°62 ores WA CW) $ ¢B8 NYC GS ~- SOA NOTMLLINOW Ores O°vé o'B8e T*0- 0°9S O°vE o'a ore 0°09 0°6£ a4 a *G a1dVL e'90t e*oot arg 0°0£ B°6Y v's? B°vs 0°o? 0°09? e°ss Ld (NTH) ‘vy aye va Bete aocu £e°sS £e*e Zaye To*e cO°¢ 0o°¥ ZOte Hd oZLtet os'or £8'°S 00°24 os’? L9G 8S*v 0o°e 00°S oo't canis Z46°OT OF £8°8 o8'2 0S°9 49°C as'y 00'Ee 00°Z 00'T go'o AAMT UN TO TAWIL # 1109) NaLYM904 JO AWM ION BHL WONS TALYINITWI IN3INOD NALYM AINOIT NO @aS¥a SI YTY 4O WALSH 'NI Yad GIWLSW FIvML 40 ONIMYOT SSYW BHL SINUS OT 40d BOVNAAY MALHOTAM~SWIL TA°tt 99°S 飰s S2°9 60°? Lis 62° oS°é os't eS'O 40TH o4*at OS‘°OT £8°8 00%! O5°9 L£9°S gS'¢ oo°e 00°? oortT e3wis Z6°OT £8'S oBrd os’9 £9°S 8S’ 00° o0°s 00'Tt 80°oO “~ ANMSTBON DRO TAWIL # 309 cere. 40° BEte Sbie BO'b bite £0°¢ eB «eJINAMASATO ~Ad~Hds OOT*O 00c'S 29v'E £06°S Tete woe BTece SL's. L0° TERY? ett Oc'¥ cate £0'¢ £o°b Hd vA/T OE DASTAAM \ MALIN SNO/WYNGOMIIN 7 NT/WH/ NOUN ONDTW Z ALYM ONIVANYS NIW/EN SY 1Y AINATIOEASR XOS ONTIWNSSY TTT PULTIGO"O OTTtt ZT t OOTIO- O0T*O- T46°0 = 66°O OOT*O- vod 6440 T86'0 COT*O- = 6£0°O ¥TS°O 9TS*0 OOT*O- &E0"0 Br0'T OS0'T 40°OLGe 450°0 £E4°O 8c6td OOT*O- FTTtO 446'0 coOorTt bt cree cot'o o78°O 298" OOT'O~ 840°0 +/- +/~ 20L O@HI e409. ba/tc/£ WASIAZY ooT*o- 066°O LET IO £89°O 9r8'o TZT°O S29°O ££6°0 TEteo £08°0 996°O B9T*O 662°O 9bB°O SETt'o Eré*o 984'0 90T°O 988°0 ¥ae'o £1cro T29°O 9426°0 Tar°o 982°0 S+N/H S+N/bHN GEHI/(AINS vOS/CATIS S+N/¥tH MYIOW SJ NNNN NN SOTLUM 00°F bo- 00's &°O- OO'kT BtO- OO'TE = ftO- 00°24e. O° 0G0T O0'OR = Z*O~- 00°6E (0° 082 00'0e §=F°0- wa 904 T°O- oot orat o'er ocr oréet oret O'tt T*o- O°csS O'°<e O°ae O'¢e O°cr O'te O°cb QOH (ATS NIW/ NV/WHS ON CTH OMDTW / *8/Teé/£ TASINAY 4TTt ¥64°0 1as°o0 916"0 os0'T Bc6°O z00't e980 t/- OrELe? 0° 06k Ord 0°06 o'oTe Ortde Orzte 0'oue vos NN 0°08 o°sot o'set O°ceT O° est O°96T o'ter 0°6S LON N NNSA OTz*o 90TtO- 820°O £OT*O T42T*O etco 696'O éve'o 642°O (AIDS 9ET0 44E°0 BIg'0 06r°O o1rs?o gc5°0 eeeo g£e'o ¥OS/¢0N T*o- c*0- c'0n- cto o'osot c'o- o°oceé zto- 301 T*O- oot o'st oretT Ott o'6t oret ott QfHI TO- Ores Ores O°sE Orde O*cr O'tt O°ce CATDS PSt'o £50°0O £b0°O 280'0 Ovt'o O8t*o 00c*0 ose'o 962'°0 Ia7 (£W/9) O'v ord oret o°te O°de O° o£ O'os 0°0E aA cw) O'o£ 0°09 0°09 0°09 0'09 0'O9 0°09? Orde 4a CNIWND SIS @ MOJ ZOUNIAY WALHOLAM-AWLL Ger!’ O5'? OS'S Ost OS*f£ os'é os't 82°O LadIw caWil oor 00°? 00°S lhe ae oore 00rd oo't SS'O RAM TION TaWIL ¢ 931947109 N3L¥mM904 40 AWNTON aHL woud MaLYINIWD LNALNOD NaL¥M TINOTT NO WaSva SI YI¥ 40) £H Yad 3LNI0S SSYW $ b8 NUE 9 - 904 NOIMLLINGW SOILYY LNATYATING Ost o°st O°st Ore o*8 0*6 oral o°s mie) oto £v0ro0 ecto ett OOT*O~ COT*O- OOT*O- OOT*O- LTTE o00'9e §=498'T 90T*0- O00'T ¥46°0 OS2°*t? oOOg'T 000°0 OGLE 186°O gfe'ebe oostt 2970 00S'O 916°O E8t°sc 60640 T60°O 2é2*0 oso’t £eate oog't 2910 oos't 86°90 9BE°Sk LeTtS 98c°0 ceres’ coo't OSe*te£ oSe't acto Q00'T e98°O YN/bOS .YN/V9 ¥N/OW ¥N/19 t/- 4/4 ft $ vA NOC 9 - 904 NOIMLITYON ‘tf a1avl T°o~ T*0- 0°04b T'0- oroot TO- o°8e O°9Ek o°st T*s8 cto O'9 O° TOE or’ O'ét aed O° O'StE oO’? vice ort .OroT O° LEE O'rt Tee? ort 0°S O° sé o°9 £°09 ore o°at O° bse Ord £°£6 ort o°oT Ot the o'e 6°es OW v2 vHN UN H NaLIWV03 O8DIW S444 * 6& wor 9 - 904 NOINLIINOW ‘tT 314a¥L ‘eg 21401 SSOILYM L1vS ¥3S Gore os'9 os's OS*+ os’s ose ost 82°00 Larw 84° 40° tate vite Oc'v cc’ £0° £ctv He OSté 00°e 00°? 00's 00° 00° 00°S oo't zaWIl OS*Z 00°! 00°9 00°S 00°F 00°£ 00rd oo'T eanrl oor 00°9 00°S 00° eore 00%e oorT Ss*o ANAM THWN SG TaWIL #¢ o0°e 00'9 00's 00*v oo'e 00% oo'T gs°o aM THON TAaWIL #€ 310 OOTSO~ oot o- 200°O £TO"O FtO°O ato'o eFo'o 0€0°O “A VNNNA AGL ib cove’ Test GG Ee Heke TL9 8 hs EEG EL LLG" AvT'? WAS VN NNN OOTtO- OOT*O-~ voo'n 600°0 TIO*O vtIoto fEOtO TI0°O TN MALSW DDIND 7 WeD 201 0~ OOT*O~ 200°0 TAO*O Z90'O c00'0 £90°O €90°0 no vB/SI/E UASTASY OOT*O~ voT*on TOO'O £O0°O voo'o bOOr0 qd ‘ALUM ONT WWY¥S NIW/EW SPE TS TEagtt bLBtO B20'T bOS nacet B9EO etntO RBP"O 3 KPLT Tasted raort FON ANSINOD MALYM INDI NO vas £50'°O 9£0'0 GEO OOT*O~ OOT*O- OOTNO= OOT*O~ 000'0 eO0"O 000*O ts 900°O 900 T00'O £00'°O TO0'O 490'°O T00'O 690°O NH a4 TW fret S ¥ te ADNIDAAT 4OE ONINASSY WALIETIOI N319M904 Syd GI MIV AQ M3L3M ND MAd SIWLEW JVML JO ONTTVOT SSVH AHL Té7e WASTAIN £00°0 2£0°0 A010 OOT'O~ NATHO= PEO 90910 O10’? z90'0 560 °O eLoro ZO0+0 9900 —-B40°0 v0" = -4GOt0 OW “9 FTAA 7 WENA Ew T*O- T*o- O's Ores 0°38 0'é8 Orr? 0°49 fn NN ¥94t0 Téb°0 £ecro Téb'o 1Z28°O TaOTT ooBtT 44627 Bact tvHN "ALUM ONT IWS NIZE SHL WOM Le dob - T0- T*O- 0° ae Ores ore 0°69 TO o9fnn orgs IN NN T*0- T°0- O°" Ors O°tt Ortt ore}T o'8 na T'O- TtO- Ort O°d6Tt Orbe O'te ordt ore ad MALI / NYNEOMIIW JO 3WN70N SHL WOMS WALYTIAITYD $ v8 Nor F ~ 90d MOTMLLINOW 9c0°O OOT*O- 9torod 800'°0 610° ord 820+0 evoro ¥S0°O YN UoLY AONATITAda ZOG-ONTHNESY NWALIATION MaLYMand JO BWI TON 40 £W Mad GLMIOS SSY¥N NOHLNOD TIO°O 400°O bOO"O booted otarod 1TO°0 zTO'O GEOrO e100 H Ae aA vSTt'o £50°O Zr0°O £80°O OvT*O OgBtTto 0020 O90 9bE*O om CEM) wa CW $ pS NY Fe ~ 90d NOTMLILTMOW T*0O- r0- Q°4tT O*er O°°£ oree o°sEe O°46E x FI 444d ‘Ss aL O°OE O'O9 0°09 0°09? 0'09 O°o9 0°09 O°4Ee Lit (NTH) *e FVMYL BLE 4o'e T£*v vith Ora 4 céty £0°¢v roo HA OS°s oord 00°9 00'S 00°v oor’ 00°s oo't caWtL oore 00"? 00's 0o'¢ oork oor oort g5°O aANM THON DB TAWIL # Sa 1dwes 8B Od SOVMAAY NALHOLSN~AWTL Bere OS? OG°S OG't OSE oss OS'*T aZ'o LOH OGt£ 90%% 00"? 00's O0*b 00'°£ 00°? oo'T cawik oot? 00°? 00°S 00%b 00°! 00% oo°T SS'0O AAMT MOND TAMIL # 311 ooT’O- 00r*Z 00z*z 692°T ooT'o- B04'T weet 9gc't r4c'T east oteet 09746°0 160°S WIte C49 L£ee’g ZS9't voL'v OL2°s 9BL ITZ 6ZEe't 626°T OZHI/ (AIDS VN 00°8 00°Bt 00'zs oo’st 00°02 o0*zy 00°os 00°0£ 00°82 00°SE 00°zt 00°SE 00°CT 00°9T 00° 9E oo*zt 00°8L O0°LEe 00°SL 00'0%" 00°SE 00°é6 an NW SOTivy z°0~- Z°o~- o°ovoe Z°0- T°0- z'o- 0°O02at Z°O- O° oree 0*0s2e Z°o- o'ozéet 0°O00ST z*0- Z*o~ O°00kZ Z*o- o°ocel c'o~ O°OSTT z°0- O°OwsT a01 NVA OOT*o- 2£22'°O cec'o eecto o0T*O- zst'o Z9T"o £Etso £er'o zIt°o cLEO 9EZ*O TS2°O £0r°oO SbE'O zEL°O TRE*O 25¥°O BELO B0r'0 9St°O vor'o v8/9T/S NASINgy o2Tt"oO £ro'o 888'o £ToO*o s¢2a’o £S2'0 oos'4e <2£6°O B22°9O cro'o FELIO vec'o 000°TOTF O00'T 998°0 980'O 084°O 20v'o 00S'Sh O00'T 696°0 atc’o 142°0 ter'o SES Acate OOF*O~ OOT*O- OOT*O- GOFTO- OOT*O- OOF'O- 6580 esto S0Z‘o0 ocr'o OOT*O- OOT’O- veae'o ztt*o tL4°0 Brb'O ooT*o- O0T'O~ 068°0 6S0°0 Tr8*0 £1b'oO OOT*O- OOF TO- vtB‘o 050°0O £92°O B2£°O ooTt*oO- = O0T*O- #S2'o 090°0 ¥49°0O tzz°o oOT*o- OOT*o- 6¥S'O £90'0 var'o T4Z°0 ooTto- O0T*O- TE9*O 9£0°O GSS*o 960°O ooT*Oo- OOF o- £E9°O 6£0°O vac'o az1'o OOfo- OOT*O- a99'°O Z££T'O Orso £40°O Qot'o- cor to- 659°O atta Teo*o vot’o GoTto- OOF'O- B18°o0 ¥ET'o S89'o 69T°O ooT*O- = O0T*O- G4Z2°O toa*o £440 ecc°o Oot‘o- O0F*o- 298°O 6vT to £T2°O £ec'oO OOT*O- OOF*O: o£e’o O飰oO 6S¥'°0 vBt‘O OOF*O- OOT*O- o1e"o 9LE°0 v£s‘o agtto OOT‘O- OOT*O~ tva’o Z4Z°O 645°0 cat'o OOT’O- OOT*O- 896°O 6Z0°0 ¥é8°0 GoTo Oot*o- OO0T*o- #OS/(AT)S StN/¥4H S4#N/H StN/WHN #OS/E0N YN/¥OS YN/YD yOION SJ NNNN Tto~ o°se O°oe O°9% Tto- Orve orZe o°2z o°ce 0°dz orte 0°0S Orlt o'er oret o'9 oO°re O'rr O'Bt Ore 0° se O°BE O@HI \ VW OMDIN 7 T*o- o°va O79 a'%" T°O- o'lv O°é6f O°ty o'rs o°se o'er o*ar O°985 O°cs o'8e O°lr 0° 8t O° 8b o’ty 0'°6£ 0°04" O°bs (AIDS soliva Na IwAINOSA ¢$vo ner eat é - va79t/S gastnay S40°T 0°009 o°zst o*or oot ost ZETE 0°909 o*Brt o'é ort o°9 6ZQO*T = O°9bS 0'22z 0°s o°% orct T26°O O°OTr o°*Tsar o°zt ore 0'0e T0O"O- T*O- T*0- T'O- T'0- T*O- 9Z£0°T O°8Es O°9ce ae 4 Ore O°st TZO°T Of49% 0°60e o'r Ore or2z 1846°O O0°OR9 O'09E 9°9 0's ort wzo't O°¢ca orets O've Oret o°set Gao't 04908 o'ect O°et oer orstt Srz't O*cber o'oet O°ve o°ctT O°Ooegt ZSz°t 0°90b o'er Orle o'9 o°d¢ 6B2°T O° Sry ores O°st o'9 O°és 924U°T O°ase o*se ran’ ort o's T9E*E Odd O°6d ar o°e Ove G26°0 O°GGE o°ey lar A o°9 aoroyr 190°E O°fcr 0°0S arzZt Oe Ooret EHO OOT? Orts £°e ort o'ol ¥S6°O O'04K2 o'ev O°9E ore orry fOO*y O'tst O°ee O*ve Ont O°9E B9640 OGRE o'ty ors ort o'er ¥CB'O O29E O'vy £°r ore o'et +/- ¢os EON Ww ow vo NNNNNSN Mah rsd OMIT S$ oO NUF Te ucc'o 4Sc9°O L£9T'O 29t'0O £F“£'"O ooTt'o- oot*o- OoT*O~. oot *o-~ oot *o- oot*o~ OOT*o~ oot *a- oot o~ 00T*O~ ooT'o~ OOF’ o- oor *o- o0T*o- OOT*o- ooT'o~ oot’o- o0t'o- VN/OW get Sco"o oos'T océ'o EFL °T ooT'o~ 00T*Oo~ 0oT*o- OoT*o- oot*o- oot *o~ 0oT*o- OOf*O- ooTt*o- ooT*o- OOoTtO- ooTt*o- o0T*o- ooT*o- ooT*o- ooT'o- ooT*o- ooT*o- YN/ 1) 4/4 904 NWOIMLLINOW o°8s? 0°S&SS 0°4646S O'vrr T°0- O°4ES o*zes 0°0b~. 0°0248 o'rag 0°24e O° Lez O'bée oO°OatT O99 O° 602 Ore9t oro Or9eTt Orett O°erT O'uLEe vHN Ww oot 0*9 oret 0°86 to- to- t0~ tro- Tro- to- T0- to- T0- Tto- To~ T*o- T'0- t'0- Tto- Tto- T*0- Tto- UN / SZ SSOTLON LIS 84S o°t ery 6C Té or 6°0 T00*0- 96 té ort ort TR4°O ve G6 Sb 2s 68 94 9 Ge ort ort zt ent zt tt mer 6*0 T9O't ae Lats zo a? ee <4 4°90 4°O O°F 6°0 B°0 t/- avi 8°46 Ots T99 s'ert s*est S°Ztt 6°SZ S°aS S°4s 4°8s £°9L 6°ee S°st 6° Or EOF B°4Y Aru 9° 6or 2°19 Geld. ane H Stlhs¢re 904 MEM LA ENOW “f VG L Rot LEe EENGE BS°CEe RO'0E vost TL°ht OL gt “vest oorot 00°s os’? G2's To's oS‘ at's 68°£ oa8rEe £4°L£ ct*y Bc*¢ vice £e*v vey iver Té'e Tete ee'y v Ib ob 96° v BO*He 80°99? BE’ bE ag°0¢ aG'ét o5°at 24° 9T e£go°at L9°oT as*ct O° Ot 00°6 oo°2 Osts 00°s 00'e oo°e Za't o0°t 00°90 00* T- HO*z~ CaWtL 80°8c B0°9e BS*ve 88°00 85° 6T OS* BT zé°ot £9°RT L9°ot OS St os'oT 00°6 00° 08S 00°s 00°’ oor? Alea t oo°t 00°0 00° T+ HO e~ COWL Hogg DPE fd B5°OS B5°6T OS" Ot ca°ot £9°St L9°¢T OS*at OS* oT Os'é 00° 00°9 00°S 00°¢ 00°e 90°Z S2°r 00*o 00° f- OS*Tt- 60°f- TaWTS 80°97 BS*¥c astoc BS’ 4T os*et ZH°9T £9°ST LP vt os‘et os‘ot oS*°6é 00°? 00°9 00'S o0'¢ oor' 00%! Et 00'0 00* {- out tH BO't- PAWEL € Té o- ra aN FIN OREO ce CGaANMENVWVADA OH tet et ot tot mt ot ot OY Ty IM THON Oo ~ 312 Of*e ae £0° Otte OL°9 a“ee 90°%b Ort 60°¢ 98° pare £6*L T6°E £T'e bith vite 9b O4*b cO*b Ta'v Oty KARKK «JONSMS AAT ~Ad~-Hds Hd 8L40°T TEE*T LE0tT £16°O to0*0- £L0°F 4T0°E 4£6°0 £EO'T E4o't geert T6ée*t oBE'T e2ZT°T 6St°T bZ6*O 6S0'T 026'0 086°0O 000'T G96°O £E8°O tf e403. vO/S9T/G HASTA SOG *Ve9ELaO "0 S40°T OOT*O- OOT*O~ cLttt OOT*O- = +80°0 4E0'T eSkeEts B20°O TEA O OOT*’O- vA0°O TOO*O~ OOF TO" OOF TO~ 92L0°T OOT'O~ FETtO T20't 8b '°4686G FIE'O T846°0O OOT'O~ E9T*O veort OF LOGE L£BOTO S60°t BB°EGTE ZOT*O Seer 00T*O- = O50°0 ese't 92°TE4T GET*O 68ett EL Teel 920° 94T°T OOTSO- = 040°O TOE T OOT*O- #460°O Géé*o CO*OTZE vTO*O T9O°T OOT*O- SLTtO Z26°O OS°OLes ZITO ¥S6°0 QOT*O- ETO c00*Tt LETLBIPE TITO A96°O OOT'O-. O462°O bcs'o £L°SvO? SLe°O +/- 904 O&HO \ MSLEW SNO/WNYNDONDIW / NIJAW/ NN OIZW OOMDIW / 3104 ONTTdWYS NIW/EW SY LY ADNATIIAd %0G ONIWNSSY ELT SO OoT*O- STéE*O £aT*o 2£4T°O OOT*o- ELE'O ELLEO 69E°O Tero B9T*o £60°O B8ET'O vbot'o 9EE'O 9L'°0 SOTO 60£'O 9eS'O 408°O £LEO 2eero 99S*O CADDS zt0- zt0- 0'0b0E zt0~ Tho zt0-~ 0' 0481 z'O- orotaz or oszz zt0- 01 0z6T o' 00st z*0~ z*0- 0° 008% z°0- O° 0zLT z*0- orostt z*0- o'orst 301 T*0~ orcs O°0F 0°98 T'0~ Orbe oree ode orze O°6Z ote 0°06 O°rt Orb Ore oF ore Ort orat O'bT 0°6e 0'8z QZHI orte O°BE O°ab O*tb O°é6s O°0b O'vs CALIDS SoT°o 2E0°O ORn0°O Z80°O OET*O Leto 44440 L£9E°O O0r'O 980'O ZTT°O 080'0 £80°O 080°0 ETZ*°O OvE'O 080°O gee SSE"9O £LEO L£92'°O £EEO 2en°O ant c£W79) o°8 O°'Bt O°ES o°sT Or0e Ore 0°0S o°o£r O°he o°se orat O°GEe O°et o°9T 0°98: oretT O°RE O°fe orae O°0r O'°SEe O'é6b mitts) cmH) O°oET 0°06 O'obs 0°09 Hees Btes O°S4 0°09 z°ORT O°0eT 0°09 o'oEet 0°09? o°0£ 0°09 9°09 0°09 ele 0°09 0°09 O°0£ 0°09 4g CNTH) SI WWYG Ce Od OStTt 00°OT 90°8 Og'9 So°s OS*y 0S°f oG'e 9ST os'o 0S*o- Se*t- as'c~ JOIW WALHOTAM-AWTL BO'BE BOTVE FE BO'9E Bree Tz Hu'be aStod 0 85°02 at 83°6T at os‘at “t Z6°9T 9t 29°51 st L9° bt a og¢zt £T OS'oT ras 00°6 TT o0'e ot os's 6 00°S 8 00't é 00°r 9 Za Ss 00'T * 00°0 £ 00'T~ z 80°Z~ t ZaWIL T3WIL # 3193717103 MALYMOO4 AO aWNIOA SHL WOM AALYINIWS LNALNOD YALYM WINGIT NO 13994 SI YI¥ 40 f£W Yad ALNTIOS SSYW $ ba NUT 8 8 4 - 904 NOINLLINOW ‘gf aay 313 *B/9T/S WASTAAN : : L£E0°S 416°@ 6TT*T = &GO'O £00°0 POT*O TLO*t 810'O 400°O Svt'o SIWWYS Ze MOA S99MNAAY WALHOTLAN-: TSE°O 600'°O £00°0 B00°O ZTE °O O10'O 000°0 £20°O o's o°oat 8o°4e §=80°BE v“eZ°O 9E0°O TO00"O oh Tenge) TORSO T10°O £00'°O ‘ O° aT 0°06 ELSE BOSSE £4TtT BE0"O c00°O TEO"O C490 vEord 900°0 Ores O°Ove B8G°Ze Balke Zeger TG0*O vo0'Oo Hro°o T946°O 800°O 9TO'O O°aT 0°09 80°02 §=8S°0E OOT*O= OOTTO- OOT*O= OOTTO- OOT*O- OOT*O~ O2O"DO o*0n arr? wO°Sst BG°6T GL5'° Lae s <Gvoro 700t0 L90°O KEL SE 9OT°O- = TZ0°O O*%cb B'vs T4°4t OSS OT T86°S SSe°e BFOrO OTOtoO veto TIste O0T*O- O0£0°0 O°0S O°SL OF T9T EGt9T esor9 vee Le 2£0°O 610°0 S8e'o0 OL9°E OOT*O- TTOCO O° OF 0°a9 21T°*ST .49°ST LETS b99°T £20°O STO°O BE'O OSE*T OOETO- G00'O 980°O oO°e8s f° ORT 64°RT LOter 9S? 88E°T ¥S0°O 4TO°0 T£6°O OOT*O- 400°O ZtT'o O°se O°odT OGTTT OoGtel A69'T S45°O 8900 éTo*o E4T°o OOTTO- L£O0"O 00 °O Oat Oroy9 Qo'OoT OS*Or Gee't EP0'O 900°0 B40°O OOETO~ LOODO FROTO O'GE Oot 00°8R 00° bret £bea'o 900°O £90°O OOTTNO- OOO oral a°O? 00%! bers FTO'O £000 PEOro ONL TO" BOOTO Ort oror Ons 00°R 4 SOE te o70to 600°O STt'o OO1O- 600°0 oroe O'o9 O08 g 086°0 Téo%o 700°O 850 '0 FOOTO oret oto? OnE O0'°b é Hotto 90070 TZ0°*O OOLTO- #TOrO ORE O°O9 One OOek 9 vEOrO vooto TZ0°O OOT*O- TOPO Orer Ene 9ST LAE : S 2?00°0 O¢ OOTTO~ PEO'O O°se Oro? : oort 00°O =F £O0'O 980°0 OOF TO- ZTO0°O 0°00 O'o9? 00°O OO*T- £ aS ¥oo'o ¥60°O OOL*O- 920°O O°GEe O° OF 008TH OCT = L58°9 g00°O £80°0 ae T OOT*O~ =800°0 O° 4b O° 09 BO'E- GO'Ee T RNS vas £ON my - SW BO FN YN H onl 110A A AWTW TAWIL TAWEL # VN NNN MALaW JTAND 7 WYNOOMOIW 44 4 f CEWS9) OW) (NTW) “ALUM ONTTIWYS NIW/EW GY 10 AONSTOTSNa 20S SNTWNSSY MWALIATION Y Lynsos 40 awnton SHA WOMA TALE TNO LNBUNOD YALYM INO? NO HaASbA ST MTY 40 EW OY SSYW $ boa NUF 8 8 2 - HOd MOTALATNOW “te SVE 314 Oot o- 400°O oT0*o ALO'O 00T*O- atoro bt0'O ATO'O £000 G10'0 ¥00'9 400°9 900° 600°O 920°O 700° zz0°0 Tz0°O Bv0'° T£O"O ££0°0 £10°0 fa) NNNA OOTtTO- OOT*O- ro0ro tooo 800°O TOO"O 900°0O TO0'O QOT*O- OOT*O- OTO"O TOO"O TIO°O eo0*o e1O°O €00°O §00°O EOO°O 500'0 000°O 500'O TH0°0 v00°O £00°O £00'O Ta0td vo0"O 7=00°O 600*0 ¥o00°0 vo0'd =00°O 600°O g00°0 200°O T00*O 600°O T00'O 400°O £00'0 £T0°O €00°O O10'°o €00°0 IN Ag wAsET sy OOT*O- T00'o To0'o ¥00°O 001 *0- v00'O §00°0 “T0°O Fon" Oo 200°O 200'O TOO'O TON*O T90°0 ‘E00°O TOO'O ENO'O 200°O $oo'o ¥00°9 §00°O £00°O Ad 1AIM OOT*O-— O0T*O~ 000°O ROOTO o90°0 n0"oO o00°O 500°O OOE'O~ OOTPO- O0T*O- Sa0"d T00'°O #10°O ro0'o Z£T0°O =00°O ¥LO°O 290°O oTo*d To00°O gTO°O TOO" 4000 000°0 TIO"O 900°0 elora 100°O TEO*O 000°O £To"o T00*O B10°O 000°O eto"o 000°O ETO TQ0°O gto'o roo0*o TEO"O T00'°O 210°9O NW a4 MALaW JTHND / WeMDONIIW 444s *SL9M ONTTWYS NIW/EW GY LY AINFTIIASI 405 ONTWNSSY MALIZTIO) Maem T*0O- O° 40T O°4TtT Orect T*oO- O'Z0T Ores $ OS O'eR O'GEe Othe O' rot O49? One O° RYT o'ta o°8e oOrog 0°90 O°Ott O°rtt OFkG fa) N NNN JNALNOD SLM UTNE NO HASYA ST MIY Tt9- 9 eo ato T'o- ote or? 0°6 o°se or Or9t Ove o°st 0°64 Orat Ort orat ore ore o'orT org Or4 Ag Oro’ o°4aT O'vEe O°9T ore O° o°8 or7t O°6 Or4 O°st O'eT O°9T o°oT ad NW MALI / WMOONOTH AO HAL AW T + o- orsot ots Orr TO orte O'S 0°93 O° 4st O° Ast o°as O°¢4gT 0°02 O'GE O°es O°8S OL? Ores Ad Sst TO'S Ost BT'b 68° ON’ re Etty Bab very re'o be eb Lee T° Tob ve tb Ort 96° bo ieee GEE. lee “TI b9*b Hd Z4°9T 29°GT L49°¢! Ostet OG' OT 00°4 O0°4 08°S 00°S 00'F 00k LEE 00°t 0010 00° T+ 802 ZAWIL Of 40 SWNIA SHL WOMA TALFINITWI *ND M34 SILAW JOVMNL dO SNTOYO SSYW FHL $ ¢a nee Be 4 - NO MOTMILINOW *S SVL 80°9E aoe 9 ont at T4°9T L9°RT L9°RT ogeat One Or O5°46 00°! a0r9 00°S 00%b 00°£ 00° sett 00°90 oo*t- out t- #0 °k- TAWEL ae Té 8t gt vas st al et _ - StAMTH ON BRO ~ > 315 OS°f Ore Hore £6°e B2°L vo’ eFINAN IAAT Hd ~AU-Hde ooT*o- verter 001,*0- OZHI/(ATIOS vOS/CATNS OOvE = =tO- 00°8) = Z*0- 00°4 e*o- OA 301 BESO T ZG MA EAUIM €00'O 900°O fO0%O xO0t0 SVdWYS FM TOM ANY THOT IM IW 46T't KOET 9OFO~ OOTMO~ OOTTO” E'0~ teo- Tto- 610°O O'F utegy Alta GSEETT ovltt QOTO~ vara THOPO c*o~ Or HY o°ce 9¥0"O 0°b 0°06 £L°OT 460°C OOTTT OOT*O- OOTTO. OO1O- ctO~ Tto- T*o- S£0°O Of¢4 H°4'd t4°8 85°64 uc’e 6 +/- t/- 04d DEH (ADS J0L BEHD CATING on “WA Aa ATW COWIE TAWEL ¢ eHdOD N MAN *AO/WUNSONUTW 7 NWA NOCIZWOOMOTW 7 CeW79) CTD ONTHD HA¥4d ONILINYVS NIW/EW GY LY ADNAT EAS 20S ONTWOSSY MALITTWO MUEYMOOs 40 AWANTOA SH. WONS HALO INT INALNOD MELUM OTT NO Wise GIOMTY gO) SW Maa aantos ssuw $ be NOrF 8 ~ YOA NOTMIL INOW ‘es ay vA/set/S astaay : O@1°O ¥reoro aur SSO7VL9N LWwS 938 OOT*O~ o7g'o z2t°o 8469 '"0 TEE'O SEC°OR OOTTO- OOTTO- Testo LOE T G62°TT ETAT GEetT £ esto ¥462°0 Getto 9v?'"O sero L90°6E Feet L9EO ELT Ovt't £E°OR BO°tt acté Z ooT*o- ocu'o Szo°o0 G42°O 9E2°O Té4B°BT =240'T TIT"O 920°0 our’t t6° BES ge'8 6 StN/¥+H S#N/H StN/PHN GOS/EON YN/F0S = YN/OD VN/OW VAs TO t/~ 40IW ZAWIL THWIL # NN SOTLVY YOIOW YS NNAN ‘ SOTL¥Yy LINZ WAINOR S4she $6 08 NUP GB - 904 NOINLLINOW *é ay. v8/9T/S UASIAIY T*0- tto- Lozg*t O° vag 0° Ove O°or T*0~ 0°O0£- O°9£9 o°ét S°ast O8fE LEST GS°iwt £ o°Be O°sb Ovt*t 0°26s5 O°vte orat O'v O'9z orzes Ost S*ctr £6°E GO°TT a9°6 2 t*o- Tt*o- OOT'T O°OZTE O°vee Oot Otor 0's? O°OWTET arco £°9E vu'y asté sc‘a % OZHD (AIDS t/- vos £0N aM) ow va VHN UN H Hd C3WIL TSWIL 6 NNNNAAN 43117/039 OJIN SSIS E NW/ NVHW/ \ 14H OUDIW 7 $ -8 NUT B - 904 NOTMLLINOW ‘tT a2 1ayL 316 be/St/y Wagtaan OOT*O" OOTTO~ OOTTO~ OOTTO~ OOFTO” OOF ro. rto- 1°O~ T*o~ OUT EF TEE Th OOF TO” -OOT'O- OOFTO~ O01 90. OOTTG- QOETO- fOr 1*0O~ t'o~ ro THe HOVE UGH S 800 °O 900°O foo'o 000°O 000° 900°O O°eGE OreZt O°éT Ore? vey *a Tt “A IN na Hed NM ft IN na NW ad Het WWok | TAWIL N NNN MALSW OTHOS 7 WENOOMO TW 4st NN NN HALE) 7 WeMoy S444 *ar¥y ONT IWWeS NIW/EW GY LY AONE AIS 209 YNTWASSY COMI TIO MELOM90d 40 AWNION FHL WOM UAL Wd ANSINOD METLYM NOT} Nu Be ST MT’ JO MaJaW SOO ed SUV LAW MAU dO ONLVOT SSOW AHL $ PO NYC 8 ~ SUG MOO ° Bays vA/9OE/G MASINI cat?! 4ee't vSe'o Z10°0 £00'°O eEO"O LEbTO veoro £00°O TEO°O SIMWIS £ NOd S9UMAAY TELHOLAM~3WIL Zttet 8090 9480 T1O"oO CO1'O~ O00" OL £00°O £00°O 4T0%O ore Bree 64°TE eeret Gert & O88'T TLort 940'O TE0°O €00°O 6to'o ETO%O voo'o PF OO One 0°06 E£"OT BOTT BE°6 & B8r "Ef ELéct véS'oO £T0°0 vou'od 8bo*O 050°O T00°O GFatd Or 8°62 14° Bors Bea oF wns ¥OS EON Ww ow wl UN HN Ont WA ae AUIW GAWIL FSWT # NNNAN MAaLaW JTAN / WYNOOMITW 44 4 ff CMB CHRD CNIW) ‘SLUM ONT TWOS NTW/EW G 8 LE ADNETIV AT? Zou SNIWNSSY DALIT BVMOOd AO AWA 3H) WOMA HSL INI TNO ACLLVM FLO NO EIS¥s Sr oe a0 gw 4 a10I0S soy $ b8 NYP B ~ OOd NOTMLLTMOW ‘oh DWIEL 317 e8/ct/e Wasting port e¢B'o vév'o ’zo°o £00°O £20°O tez*o £10°O 500°O acoro SINS £ NOS BSYNAAY SSLHOTAM-SWEL 520°0 OOT*O- OOTO- OOT*O- OOTTO- OOT'O~ £240°O OOT'O~ ZO00°O voo'0 ort 0°06 SBc*rt Ostet oorrt & tes't 2£22°O oLero £E0'O £00°O 660°O BrE'O F10°O 700°O 0co'"0 o'9 a*0zT OO'OT GOTTE 00°6 @ 868'T 284°O £eS°O SZo*o0 100*0 vto°o £EL'O oro'o 900°O 9£0°0 o°s 0°08 OO'G- Bete GL'G- FT wns vos £0N mie) OW vo PHN UN H a7 10" | AUT TAWTL TAWIL @ NNNA MaLaW JIGND 7 WYNOONOTW 444 ¢ CERO CD (NIW) *3LUMN ONITINYS NIW/EW S ¥ LY ADNATII4AR ZOS ONIWNSSY 041937109 Y3LeNHo4s 40 SWN TOA BSHL WONd MALVINITVI LN3LNOD YaLem WINWIT NO Wasvd GI MI¥ 40 £W Y3d JLNI0S SSUW $ va NOC 4 - Od NOTHILINGW °o JL eB/2Zt/e MaSINay 000°0 tt0°o TI0O"O FIOtO S31dWeS £ YO4 ‘ 39904300 DSLHOISM-3WIL HNREK COTE 000°0- 000°O- OOTTO- OOT*O- OOTTO- T°O~ rt0- ¥°O~- ¥00'O OF 0°06 S2tit OS'ZT OOTTT E vee SS'f vZO*t 2ao't OOTF*O~ OOT*O- OOT*O- Z°0O- Fto- t0- O20'O O°9 OTOTT OOTOT OOCTE 00°46 € TPs 4é°f BOTttt wit't ooTt'o- croto z£0°o z*o- O°0£ o°'ae 920'O 0°8 0°06 00°S- Sa*r- SéL's- 7 eSINANIIITIO Hd +/- t/~ 301. OZHI Cans I01 OZHQ (ADS aa aA id 40IH ZBWIL TSHIL ¢ -Ad-Hde e400, \ MBLOW ‘NO/NUMOONDIN / NU/WHS \ OT/W OMDTW 4 (CEW/9) CTH) CNTWD LVN ONITGWYS NIW/EW S Y LY AINATIIAda ZXOS ONINNSSY G3LI91IOI WaLvMg904 40 JWNTOA 3HL WOM4 G3LY1NI1VD 1N31NOD Y3LYM UINVIT NO DaSvd SI YI 40 EW Yad JINIOS SsUn 5 $ ba NYr 6 - 904 NOTHLLINOW ‘eg a1a¢ w8/2t/2 G3asin3y ocT*o £r0°O aze°o éT°T sSO1L9d LWS vas oot *o- ooTt’o- OOT*O- OOT*O— OOT'O- OOT'O- O0T'0- OOTO- O0T*O- O0T'O- 000°0~ G2°T? oStZt OOtFT £ 00T*O- oot ’o- Sv8'o 9bz'O 665'°0 vér'o 6955'S Loree ztr’o Tré‘o cHo'T OO°OT OO°FE 006 Z ££6°0O £60°0 9vB'O 602'°O 8fe'o gteso L£T6°Le L99°T ose*o L£99°T wttst 00O°S- SZ*v- SzZ*s- F OZHO/(ATOS WOS/CAI)S StN/¥+H S+N/H StN/¥HN bOS/EON YN/F0S YN/09 ¥N/OW ons 19 +/- iaIn caWIL TAWIL @ NN SOTLYVY YOTON 7 ZN NNN SOTLYYM LNAZTWATIND! 4444 $ bB NUF 46 - O04 NOTALLINOW *e 31avl v8/2T/2 AASINaY oo'T T*o- T*o- T*°o- 000°0- T*O- T*O- TtO~ F°o- t*o- O°ZT6 T*o- Bert 飥°£ OS*2T OOTTT E 00°? z*0- T*o- T*o- fBO'T O42 0°62 o°ce O°ot Orne 0°989 O're at tez SE'E OO'TT 00°S Z 00's f*0- oro o°az ortet ofces O°6ES o'0g ore O°0d O°6TS O°zT a'6oT 4é2°£ S@*e- G2°s- T ‘oA 301 OZHO (AIDS +/- vas £0N mie) OW wo PHN UN H Hd TaWii TSWIL ¢ NTW/ NU/WW/ NOV ONDIN / NNNNANAN NaLIWV/03 OMIIW 4444s $ v8 NUT & - 904 NOTHLLENOW ‘t ayes 318 ecb 60°¢ Ete. wey St’t+ 96S £B°L See £4°E£ T2°s eFONSNG AST Hd ~AB-Hds 4b8°O 6Se°T Trs*o L£EF*O £52°9O OZHI/ (AIDS NON SOTL¥Y OO'trT Z*0- go'or = c°O~ 00°hE) = E*0- OO0°ssS E°0- 00°00 O° 0E9E 10h 301 NIW/ NVA B8246'O TB4* BEO*T 4£0° 6v6"O CS6° 9To't ftco* vE6'O 6£6° +/- tse +MHOD« £0z°O £06 eve‘o 958 Z£4T'O 096 £2410 268 ecto Sté vOS/(ATI)S StN/ MMTIOW 4 / NN ores 0°62 o'se o'2Zot ore? Ores 0'°6S o°ze Orde o°as OCHO (ATS \ OV OMDTW / vOs/TE/E WASIASY o ¢ 0 oO T 0 ° oO I 0o cLG'GLE Ott o~ OT O~ OT O- OT *O- 690°O ¥ST*O vse°*o vero 4ST'O 0 OT*LIPE 990°0 N SLUM ONITNYS NIW/EW GY LY AONSTIIAAG “0S ONINNSSY SHL WOMS US3L9INIWI LNALNOD HAL¥M INNIT NO Wasvd ST YIV JO EW Yad 3LNTOS SSUW 9021 O¢HO vO/tzZ/£ WASINAY *0 "0 “0 ‘Oo ‘9 UtH NN o8o0'o £L0°O 9eT°o £Se°O GLT°O S+N/H veste/£ MASINaYy T86°O 6ro'Tt e86°O teo't 6£6'0 tf- O°8ee o'va8 0°8ES Oster 0°crs vos \N $90'0 orto Treo T0E"O 90T°O vtceo (ADDS 2en'o 9OE°O £eB"0 €5E°0 £1B‘O véE'o bE9°O £ET£'O OF7Z£°0 Tee*o S+N/tHN ¢0S/£0N SOILVM o'er 0°6 O'Tes o's orete orgt o*tet OroT O'0z2e O°or £0N w NNNS &°O- a°o- e*o- e°o- O° 029E 901 $ 68 OzT*O 008° 42 00b'88 008 °£S vIZ'é6S 140° OL YN/bOS LNA TWAT $ +B OW Se0°o0 Ores 0°64 SSo*o o*SB O'ZzotT oO0T*O Or“? 0°46 Ores BIT*O Orde. 060°O 0°42 o°ss STT°O OGHO (ATS on MALEW *NO/WHMOONIIW / NI“HW/ \OT/W ONDIN Z CERO ortt 0°0e O'SE O°Ss 010€ 0A > CW) Nur OT - 904 NOIMLLIMNOW £¥0°O 006°S ooL'é 009'°T 985° 982° OND nos etrto “ter 00r'o 0064°0 00s*o 008°S QOL TO 009°T 98 °O 9822 98 "0 I 20°T YN/OW YN/ 19 44st wer OF - S04 NOTHLLINOW 0°6e OZ Orot oot Ores a) 0°0r8 O°1ts O°ot9 O° rTse o'sre bHN MALT 1/09 OMDIN $ bA nvr OT - 904 MOT ororT o°or Orot ord O°BE YN +f INOW B°6L v°08 9*bR &'Sez PA? id (NIM) tsorivy Lws vas T86°O 6£0°T €56°0 Téo°t 6£4°O +/- ‘e avy £°Ts £O9E 9°60T £°tet O°céer H Ssh ‘Tt ATa¥L Satdwvs G 40d 39VMIAY G3LHOTAM-AWIL vas OS'Z £19 BL kr SZL°O IW va’ os’Z £19 BEE S2°0 LitIw 60°¢ vey 96'S. Sere TL°E He OS°6 24t°a £B°9 ze°s £“ort caWIs gaL937TI09 MaL¥M905 40 aWNTOA OS°6 cts £8°9 ce's £E'T @awit 0S°4 Z21°B £8°9 eoes geet CaWTt 4v°3a £a°9 eves cet éT°O TaWIL # SS at ie we? *£ aay év°a £8"? eb's £Ee°T 2t'o TaWTL #€ FAM sw £1°8 Far? Eve's get £140 aOoM TiO TAWIL ¢ 319 ooT*o- TT0°0 TTO*O OT0°O L10°O * a) VN NNN TTIO°S 624°E. ri BEA 850'9 GOE*ES vEL°B WAS VANS OOT*O+ 400°O 900°O 600°O TTO°O IN ooTta- T00°O GaGQ*O TOOt0 g00°0 no PA/ST WasTAayN OOT*O- OOT'O- OOT*O- £00'°O 100°O Tto0'o 00*°O 900°O 800°O 000°O 000'O ¥00'"0 £00°O Z00°O 9TO°O ad NW aa MALSW JTAND 7 WeygOWIIW C4 Ls ‘ALYY ONT WWYS N 662°% 090°Z Btetb oSO0'£ oos't Be9'd vas 2oTer vis*o LET wss't geZ*O bcs't LON vA/sTE/E WASTANIM ¢90°O v00°O £S0'°O BT0°O £00°O zkoro 660°O 900°O vSa'o £90°O ¥00'°O 8£0'0 1S0°O €00°0 660°0 zezT*O TIO'O «— -OBT*O YO on vo M3LaW STAND / WYYSOMOTIW ‘ALUM ONTTdWOS NIW/EW GY LY AINSTITAAA XOS SNIWNSSY WILIAZTION NSI¥MAod JO AWNTIOA 3H. WOME TILYINIWS INDINGD MALYM INDI NO WaSvd ST YI¥ 4D EW Nad JANOS SSeW TW/EW SY LY AINATOTSAAS ANSLNOD MALYM HEIANIT NO T*0- O°ett O°eé O°40T O'OET A. NNNSN “OG ONTHNSSY WALIATIONN MALYMNGA 4O AWNTOA FHL WOMA DALY INIT N3808 ST MI¥ JO NALAW SND MAd STVLAW AWM! 40 SNTdYO) SSYW SHI aTo't 9F8°O Brot tort R95'°0 es2°t PHN TO T*o~ O°0e O°eT COS oO Tb O°9G ahs O°46 Et9v IN n3 MALIW / WEMOOMITH T*0- Orve oOeF ore 0°8S Hed T°o- oo O'bv ore O°er NW $ vba NYC OT - DOA NILTALLINIW Ge0"o zTO"O £TO"0 $00°O £20'0 #00'O 420'°0 £10°0 ¥tO"O £10°O v20'oO £20°O ON H S4 se ¥40°O s50'0 ootto BIT*O 060°O GIT*oO amy (£W/9) Ortt 0'0g O'Se O°sS 0°08 10n cw $ VB NOT OF ~ SOA NOLTHLLIMOW T*0~ 0°60T Orbe 0°0S a4 44st 60° 2 Ae 96° SA°k T4°2 Ha *S 3J1a¥). B64 6°08 9°v8 & See 9°69 Ait CNIW) OG°6 208 FRI? eyes EFT Caw 47°68 £O'9 ees 5 aa | 2T°O aNM TD TaWIL # SA1dWHS GS OJ A99NIAY WALHOTIM-SWEL ve°s O5°Z £1°9 BE°£ G2L°O0 AqIW *y aL OS's 2t's £8°9 tv's £“e°r @aWtl a7*a £8°9 zees KENT ét°o TAWIL # am. i 320 £66°CT BS9°OT £L9°ET ae9' et RAS NNN ILE 60° 6 Z a S \ 96 £c eeesk Te oDJONSHIIATI ~Ad~-Hd H TSS*t Z8zZ 62S °é 4 eA/Te/£ MasSTAay os’e 9EL°9 veT'o vco'"o 6£2°O 9L°E 92¥°S 160°O oTo'e 9ET'O vec GEE"? tet’o Z££0'O c£“e'oO Sf B9E'L wét*o S<0'0 vée'o vos £ON aie) ow 93 Y3L3aW OTAND / WYeNSOMIIW £9StS 650°S 25G°E ra 2 vHN Set‘o zto° att*o zZt0° Bvt to vto* cv) 9LT°O 0 Z0T*o oO reo cét'o zt0°O Zatto UN # “ft amy 7 (¢W79) oret o°ce 0°ee WA ctw SPINS © NOd S90MAAY ULHOD aM AW 0°09 Gob 5L°6 S2°e € oro? gee 62°0 G2°e 2 0°09? Gate GL°e s2°9..t ig LUIW ZSWTL TAWIL ¢ (NTH) “3LUM ONITIWYS NIW/EW S 6 LY AONAIOT Ada XOG ONIWNSSY 194931109 HaLvm90d 40 FWNION 3HL WON4 G3197NIID INBLNOD Y3L¥M QINVID NO 235d SITY 40 EW Nad 3iNI0S SseW $ va NUr £T - 904 NOTMLL va/ste/e astaay SVE 966£8C0'O s70*d zvoro0 ‘£ tso*t e£S0°T vVB*¥99C 090°O 660'°0 o°og0€ 2*8T 0°42 Zoro ‘vy oro’t rro't 08°6168 vOT’O yse*o O°OvOE 6°bT O° bE £LE"o ‘vw * 006°0 106°0 9S‘Z08E 660°O cero O'OSZT £°8 O°ce cato r | t/- +/- 304 QéHd CATIDS 904 OocHo (AtDS 9M edHODe \ MELON ‘NO/WUNOOMIIW Z NINH \OVVW GMDTH 4 (END JLVM ONT TdW¥S NIN/EW SY LY AQNSTOTI43 X0S SNTWASSY INon oral orse 0°82 WA > Cw) ‘ob aay 0°09 0°09 0*0? 4a CNIW) Sa1WYS £ 4O4 SOVHSNY DSLHOISM-AWIL Se°s Sc*e scrd 4aIW GL°6 G4°a SL°eé c3MIL gé°ea o£ GSLtk z gLto fF TAWIL # d31037I03 YaLymsod 4G aWNTON 3HL WON T3ILVINIWI LNSLNOD N3LYM DINOIT NO Cased ST YI¥ dO EW Yad SINS SSOW ve/IZ/E Gastnay Z0T°o z98*o 080°0 cBz*o 6éT°o tte'o £20'O 8£2°O eitto 996'0O 090°O 906°O OZHIZ(AI)DS vOS/( AIDS StN/94H = StN/H StN/PHN NN SOLLVH WW TIOW SS NNN NAN 0o0'st 00°SE 00°82 TA NW NUSWHS NOTH OMOIW / vastz/£ GasIaay 0°0802 2°BT 0°62 £So°t O° OFS 0°87z8 oO'ovor 4*bT O'vEe TEO'T O°62£ Oley O°oSet 24°a o°ce T06°0 §=O0°26E£ o°ac9 _ 904 OZHD (ATS +/- vos £ON NNANNNSN £“s°t ceter zov't vas/EON Ssorivy O°ve oret Or9e Ww SSOTLON LWS vas £So°t T£O°T T04°0 t/- ‘fe a1av. 9*40T 6°8S Lye H A444 he $ 68 NUF £T ~ 904 NOTHLLINOW -02T°O £r0°o0 82z°O aur OSZ°TT 6cc°T 29t°o 00s’o LEO HT 299°C T8v'°o OLF"O 008°6 00's S2e°O 0S9°0 YN/vOS = YN/¥I UN/OW YN/ 19 ANS WAINDS 44h $ v8 NUP ET ~ 904 NOTHLLINOW o's 0*é6S 0°040T 0°80 o°’et O°eZ 0°846S O'4é artt o°o00t O'ves O'°Oe OW vo PHN UN 4SLT1/049 ONDTW $ bod NOT £T - SOA NOINILITNOW *T ae BE°6 Gz°3a G2°d 40TIW 96°E got tee Hea GL°6 Sé°s8 Ger! Gawls ‘BLt6 G2tea BL°6 ZUWIs *g aay s2°e £ GL°L = SZ°9o 1 TAaHIL &. G2°8 £ Glie & ad°o ov Tawi. ¢@ 321 HR KK TO°sS Te°s aoe vere 84° etre .FONSNIAATO ~AH-Hde 4Z2°9 wees £ees ££°S O's ares TO°S Hd ooT*o- 689°0 cao't 8cé°oO Q00°T cce"o 829°0 OZHI/ (A ‘N80 Ios IL¥y 00°*f T°Oo- OO'vT Z°0- oo'st c*0- o0°ét c* OO'OE 0° 00bz: OO°ST a o0°sc O° o- o- O8be 00 304 NW/ \VWHS ON CUR ONDIW 7 ¥S6°0 Soort TOO*T Svorr SZT°t OETt 2éS°t +/~ 440d. 00T*0- Z£81T°O cotto TéT°O £8T°O FEEO ebvt*o vOS/(NT)S StN/¥4+H YYTOW ZS NNNAN ba/FT/S WASINIAY bS6°O g00°T c00't Sro't S2T°T TEt*t ecs't t/- OTT 69bET OOT*O- 0 QOT*O- 6 ooT*o- Oo OOT'O- 9 Tt'90Lre £ OOT*O- & 2U*eSés ¢ 30L 0 \ MBLSW *MI/WYMNDONIIW 7 NI/WH/S NIH OMDIW / SLUM ONITdWNYS NIW/EN GY LY ADNSTIIAAD 40OG ONIWNSSY SHL WOMS DALY INITHD LNBLNOD YALVM GinvIy NO Wasvd SI yI¥ JO £W Y3d JLNTIOS SSYW 8t°o OT‘O- 40°O Tr°O vt'o ZT°O 82°O 6v'O eH vB/9T/S WASINAY 0zv°O 688°O £98'°O 288'O 082°0O zS8°oO 809°0 000'0 +00°O ¥o0'"O ¥00°0 £00'O ¥00'O s00°0 Ocbr'o s88°O £9B°O 82480 24£°0 are'o £09'"0 v9T*a O0T*Oo- £20°O ZeT°o Sbt'o SBT'O £9c°O 9SE°O CATS Tee°o T9E'O GLE'O £SL°O T9L'O Seore 988°O S+N/H S+N/VHN ¥OS/L£0N SOTLYM LNA TWWAINGA v879T/S WASINGY T*0- t*o- ¥S6°O O° Tse O'90T O°fe SOO'T O°2£6 Oorlg 0°99 coo't o'BAg o'ce O°2é Svo't 0*0046 O°ed O°ZL GZT°T O°9T6 O°bs o'cs Tet't O° eéy O'StT oO'ad 825°T 0° 46460T O¢HO (ADDS +/- vOS NNN O°vOG O'ere o'é6art O'sTEe O'Tes 0'900 O'tsa £ON NNN O°éke O've O°? O'4E orat T 0°95 Ore? Ww tto- a*o- €°0~ 2*o- O° 00bE e°o- O° OBbEs o0L T*o- o°90T O'19g o°'re orcZ O'es o°stt OcHI ¢ T*0O- Orne O'99 O42 oreZ 0°28 0°84 ADDS 090°9O oTo’o TEO*O 090°0 650°O 080°0 ooTto ert o any (f£W/9 ore Orot O°st O'ét O°0£ o°st O'°sa 10n , CW) $ vA Nor 2 - 904 MOTIIMNOLING ocTt*o 90T*O- KETSP C9L EE ata’ts 00S* rT 008 *°eL CPE TEL UN/bOS £v0'°O ootto 000% o's SSb’°s T Goose o0g*T oov't YN B82e°O ét*t - O0T*O- O0T*O- £LL°O L9ETE vIZ'o O4T*z GSv‘o Bob's GE?°O OGG °S L9E°O ELLE ££S*oO ELT? YN/OW YN/ 19 4/447 $ bo NUr 2 ~ 903 MOTIINNOLING O°et o°rt ost ors O°sS o'r o's SW T 0°0 oro Or46 org 0°6 o*s ort ¥ 00 9 S @ & I < 9 MALIV/03 OMOIW oroztt O'orort O°462 o*osot 0°696 O'oéet o°oscet “VHN O'ot- o°at O'tse Ort o’s Oat O°ST YN /¢f $ 68 NYT Z - SOF MOTTIMNOLING o°ort o*ostT z*oot 9° S6ET o*ost 0'o9 z*02 iq CNIW) *f aL SSOTL¥N LS ¥3aS ¥S6°O Goort zoo't svo't G2T°t Teter 84E'T +/~ *e 3avl H 44rs *T 316i S31WdNnvS 2 Od S9VMSINY DALHOLAM-AWTL 00°at 08'S OTe GE°S 26°% curt 60°O AdIW 00°ST 05°64 OTe 62°sS c6%e aut 60*O LdIW 62°9 bEtS £b°G £L°S O's BT's “TO"°S He OO°ET oo°tt 00°8 ££°9 ctty Lert 27°90 f3aWIL G319397109 waL¥e9Ns 40 3WNIOA 00°et OO°Tt 00°8 £L°9? atte 2o°t £9°0 Zawts oo°nt oortt 00's LES? éttb £9°T £9°O cAanIb oo°rt 00's ££°9 Ltt aot Lo°o os*0- TAWIL OO*TT 00's £E'9 iad a L9°E £9°O. 05°0- FaWIk aN Mm TON ANAM € DON AAMT ON 322 ve/st/e£ TaSIARY OOT*O-— OOT*O- OOT*O- OOT*O- OOT*O~ OOT*O- T°0- T*O- T°O- r*o- Tto- T*o- 6£°9 OOTLT OOTTT 2 000°0 T00*0 900°0 c00°oO 000°O $00°O O°9oT £°Le O*v O°bS o°ert OrnEet be'S OOTTE 00°8 c00'O €00*O 000°0 vo0'o T00°o 4E0°O O°oLr cay o'r Oro ors O'bere £b°s 00°R gE G c00°O €00°O 000°0 200°O 000°0 §00°O o*Sse tebe o°9 O'4E Or O°RS £2°S EEO atte ob £00°O £00'O cz0'o £00°O 000°O %o0'o Orbe 4°04 O°04E O'eL or? O'be Ors di*t¢ 49°T £ voo*o 500°O £00°O voo"o T00°O zto*o O°se o°eb O°6E Ofc o's o°stt BT'S 49°T £9°O € $00°O £00°0 G00'O 700°O £00°O 0c0'0 O've O° 8b o'er O° 6 o*Bb O°zetT 8 T0°R £9°0 os°o- T A IN a3 ad NW aj nf IN ng 4d NW 3d Hd ewWiL TAWIL # VN NNN MALSW JTHND / WENSOMIIW 444 VN NNN MBLIW 7 WOMOOGMOIN ya A A *9LUM ONDINYS NIW/EW SG 9 LY ADNATILAAT Z%OS ONIWNSSY MALIATIOD YALYMIOs JO ANNA SHL WONA TALYINIWI IN3LNOD N3LYM HINOIT NO GaS¥a ST YIV 4O MBLAW “NI Nad STWLAW BIVML 4O ONTMYOT SSYW FHL $ ba NYP 2 - 904 MOTIIMNOLLING *S a1avl vB8/9T/S ASIAN 406°S TO9°S beat 860°0O £00°O 8460°0 Ost*t TE0°0 000°0 090'0 . SaidWes 4 404 S9UMAAY TALHOTAM-SHIL est F60°T z1e*o OeT*O vTtoro Tor’o 000°OT- 000°0 OTO*O ore oroct OOrST OOTET OOTTT Z Sess £6r'°T 697° Z££0°O voo'o | ZkOTO TYO°O 000°O TLEO°O O'eT o*O8T Os'é oo'Ttt oo'8 9g LOLtE 4Z6°T 202°0 8460°0 TT0°O 140°O 620°0O Q00'0 090°O orst Zoot OTe 00°84 £“o°9o GS O£S'b SES*E 9GT*T 980'°O v00'O zL£0o'o STO*O 000'0 650°O Oat 9° het GE°s ££°9 atte ob LL9°9? 0gS* evel TSO0°O g00°0O 9r0°O S10°O o00'd 080°0 O°0£ o°ost a6°S LT't 27°t £ POTS TT LPL LEEN9 66T°O S00'o P£0°O vEOtO T00"0 ooT'o o'st 0°o? ét‘t Z9°T Z£9°O & T22°6t 61S°Z z09°8 £TE*O ¥to°o 090°O 4¢9°O TO0"O evTo o°ce e°0d 60°0 £9°O os‘o- T WS vas £0N Va OW v2 vHN YN H om 0A AIF AMIN ZAWIL TANTL # NNNAN MAL9W OTdNI 7 WEMNSOMIIW SAL fF RWI CTW CNIW) *ALYM ONITVAWYS NIW/EW GY LY ADNATIISAa %OS ONIWNSSY WALISTIOD NaLyMaos JO AWnrIOn SHL WOMS CSLYINI1V9 ANFLNOD YSLYM CINOIT NO dasva ST NT¥ JO £W Yad S1N10S Ssvw $ pO NYOF 2 ~ 9OA MOTIIMNOLINA *e DMYL 323 OLt'R «= STO‘T «= 9¥Z*H HETO «CTO, FSO) = wbATT «= PLOT = 00D zz SITWYG vo NOd BOVNYAY HE HOTAM- WIE Tap't = Z46Z°O LSb4O | BETO. — 6000-000 _— ONTO 0900 HOT *O~ TOTO OT ESE HTH OOGTA NOTE oo 9B9'S « £62°Os*TL9*E—WSF*O ©=— $000 6G OLL*T E00 )=— 0000 ROTO OST EtG BETH UTA RTA OF SLB GOTT 949'L 9SV'O fZ0°O)-BFOTO ESE «= 4400 000° G40'0 OEE HS ETS BGs L¥TD Z WEO'TE 9EZ'O 6Z6°S «= dT’ HOO ELE «= GGO | HOZTO OT O- -FTOTO OT ese 9 wGteGvteogytz oF wns vos £0N 13 on va vHN ON H ant ina Li LW @aWt) TAMIL & NNNN M3L9W OTdND / weYsOHIIW S00 ¢ 8079) CW ONTHD ‘4104 ONITUWYS NIW/EH S Yo 10 AUNSIDIAd9 20S ONTHNSSY 1997103 Wiivmand 40 aWITOA BHL WO¥d MALO TNIWD LNALNOD YALUM HINGIT NO TASHA ST YIY dO SW Mad 3ENINS SHWH $ £8 D3T Bt - 904 YIWSIN ‘by FTL v8/9I/S WaSIAIY vaz'2as 900'O §~=—- 1000 900°0 SATUNYS # MOd ANVMANY GAIHOLAN- IRTL KERR OT'O- = £E9'O E90 OOTTO- OOTTO- OOTTQ- FtO~ F°O- TtO- FTOTO OTF FEZ ZB ETH BOTH oY HRER 909 = -9L99D -9LF'O--BZ*HNLZ 65040 §=— 900° =O 0HSZ O'ZZ «OTE HHO OOHT «GB GAA 00TH: ATE ween 18'S T4810) 128°) | AG ZA FBO'O «= TEOTO «=O OBE OTOL «= OFT, G40°O «OEE «6 thS RTTL «HA LODZ BEER OT'O- «= 9960) -99H'0) OOF 0- = OOT'O- =OOT*O- ~TtO- = T'0- FO- 9toro Off etsz wStZ Getz EETS fF eaONaua4aIT = Hd t/- te 301 02H) (ADDS JO OTH CATS = NTO tit 40IW ZaWIL TaWIl ¢ ~Ad-Hds s8YOI» \ MALEN NO/WYNOOMIIW 7 NIZWHS NOH OMDIM 4 (CEW/O) CTW) (NIN) BL¥Y ONTTVANYS NIW/EH SY LY ADNATITI4a XOS ONIWNSSY T3L99T1I09 WLvAD904 AO aWNTOA 3HL WONd UALVINIWS LNFLNOD YALYA DINOIT NO W¥Svd GI MIY 40 EH Mad FAN IOS HSU $ £8 930 Bt - 904 YIWSIN ‘fe aay ¥8/91/5 Casing OZT°O «= EKOTO | BAZ*O LET 1SOl1va Aqws v38 00T*0- OOT*O- «=—«-PRL*T = QOT*O- «FELT «O4T*T «=v 9ETZ «0000 64770 BNET «=EEVTO |= BZ*4. O8'6 HOH t40°o ZZ0°0 SvP'T = 100'O «who TE céS*Z PBT TET CSEtc | «ZSE'0) «6 000°BPZ9°O) 4H OTH BSL EF £fL'0 S¥0'O OBT'T 100°O 62T°T 804° «FESS oFbbtO «CTTS'O §«=— oor se8OET*L «STL LODZ 00T*0- OOTTO- «ZED OOT*O~ «zETO «= ZOHSE «ZSS*O = ELOTTZ «TTe'O |= BBO 99HO STZ GLTZ LENT OT OZHI/(AT)S FOS/CAI)S SEN/Y+H SHN/H S4N/PHN WOS/EON UN/6OS YN/YD =YN/ON UN/ 19 +/- AMIN) Z@aWIL «TAWTL NN SOLLVY MUTOH ZS NNN AN SOlLva LNA WAIN ree $ £8 330 Bf - O04 VIWSIN °% FWvL vO/9T/S GasIngy OO'E T40- T'O- =T40- = FEPTO -OTOKE «=O FH OHZZ_—C HOO O°00SF O'S9T F40- © OFO- 05°6 80's F& O0'6T O'0BSZ O'ZZ O'Z = =-FLZP'O- OO FBT «OT EBR «OO TT One Ove = OTOOTT OWT = FO 909 00'S HGTL £ OO'FT O°OWBE OTOL O'OT 12810 O'dry O'OOET O'SET O'RZ =—OTOZ =—-FO9GE «Oke GIT HS ane 499 z 00°F «T'O~ = =TO- = T90- = 99H -O60E «= OZOH «—O'F6Y «= OTOFE §=—-OODBTT O'040z O'09G) = F*0- OF O- Getz ENTE OF wa 304 OHO (AI)S +/- ~ vO8 £0N 19 ow vo vHN oN H Hd «= ZaWIL «TANI ¢ NW/ NWR N VW OMDTH / NNNANN Ya1I7/03 ONIIW Vr . $ £8 930 BT ~ O04 VEWSIN oT aul vos ver "eB atte 324 oOoT*o- T0000 T00'0 OOT*O- a] VN NAAN OOT'O~ OOT*O- ¥o0'o Z00°O 810'O £00°O OOF*O- O0T*O~ IN na HAL3W OTang / “Hive ONITdNYS va/stve aasinay OOT*O- OOTIO- GOT'O- ¥o00*o Z00°O vto'o BT0°O 800°O T#HO°O OOT*O- OOFTO- OOTtO~ ad NW a4 WeHNOMIIN 4 4 4s NIW/EW GS ¥ LY ADNATOTIS9 T*o- t*o- T*o- T*O- O°tt o'er o°82 o'éy o°ot o°cat 0°49 O*vat Fto- t*o- T*o- Tto- fa) IN no a4 VN NANA 43117 / WYYOONSIN 20S ONIWNSSY UWSLI97II09 YS1¢n904 30 AWMIOA 3HL WONS G3L97NITVO 4ANJZLNOD YALYM GINOIT NO G3Svd SI YIY JO NALaW ND Mad S1v19N 30¢NL 40 ONTIVOT SSYW BHL T*o- 0°2e 0°88 t°0- NW % £8 33d Bt - 904 YITSIN T*9- or°est o'9va T*o- 34 444 ‘Ss a1avl otto- 909 Te*s OT*o- Hd os'é 00°6 ac'é S2te eanit 80°46 OS*Z £9°9 ££°z anNnM + Taw © 325 L328 JORIOIOK FOR IOK FORO OO «3JONSNIAIIA ~AP-Hdo 60£°O etett Seto 000°0 000'°O OQCHIS OA \N\. 80 00's et corset OF oo°T£ O° oo*0c, Of oorat &* THA N\TWe L£4°9 Sv’9 vL'9 oa’9 Tore Hd Ds TL¥M o- OeEk oLet OF8T O- 201 NTVAWW/ \ MALSW SNO/WENIONDIW / NT/WH/ NN OCI/W OMOTH 7 ALVM ONT WWYS NIW/EW GY LY AONATOTAAN 205 ONTWNSSY SHA WOME ALY INIT LNBLNOD M3L9M HINDI NO 73asvd ST MI¥ AO EW YSd4 BLNINS SSeW ve/stese WAS LAAN BEb°£9S TIO°O Zva*o zeBto OOT*O-- 920°O Z£8'O 4£BtO £0'%460F 620°O 10Z°O §=T04°0 TB*94TE 6L0°O TtIg°O =TTE*O bE*S9bT FFO'°O TZE°O «TE '°O OOT*O- Z90'O +/- t/- 30L QEHD «4409. vasTe/f WASIAIN £50°O Soret 000°0 BOTtt Tet’o Obit 000°0 Serer 020'O L£5¢'T 000°O 960'T 000°0 beets TOO*O £LOOL 000'O Zeer 000°O 2bere VOS/(CATIS StN/V+H S+N/H = StN/tHN MYIOW / SN NNN O°ss Ores Orbe orat O'8E OFM ¢ O°et 0°Ob ore o°0 Oto ADDS ve/stese UCaSTAaIN crsa'o 2EB°O TOL*0 Tir'o TLE*O +7- N CVUW OMDTH 7 O° ae? 0°898 0°99 Or°ety 0°56 0° ORF O'éb O°Est O°stt 0°96 vO SON VNNNNN ¥00°O 600°O £OT*O O10°O 000°0 000°0O CAL)S THEPT 8c2°0 9020 cote bL5°C ¢OS/f0N SOTIYY O'ott 0°8S OF O'0E O° 6t Y e°O- o°ss O'OEeE OLE O°O42ET Of bE O°OLBT O° BE c'o- O°9e I0L OcHI Orat 0O'Ov Ore o°o o'o0 CATDS Tt0°o 9tO'O OTF 080'oO OfeT £OT*O O'TEe 290°O = 0°0E 080°O O°éT a7 10n Cen79) CW) O°Sé of09 o°oat O°oet 0°o9 Ai (NTH) tSOILYM LIS ¥3aS cea*o ££8°O TOZ°O TIF*O TZ£°O t/- $ £8 D407 6t - S04 VIWSIA OET°O 8cc'o 2tet Ore Le Z99°O 9bEte 0099S O00 9008'S LEL*BS L99'°B 000%. L£98°S 4EENG £L2°T YN7vOS = YN/YD ¥N/19 ANIWATINOA 44 tt $ £8 OAT 4T o°st Oty Ore OE Ore OW $ £8 DHT 4b 904 ¥IWSIA ‘e AWyL e*o ¥*0 <'o ato TO. H L44EL Ts O°ooRT O°4e O°'oETT OfOT O° ORL os. 0°0469 or¢. O°OLTT Ofce HN uN SOA VEIWSTIA “T ARTE Sa1dWwuS G MO4 SOVNSAY DALHOTAM-SWIL £9°8 osté 00°9 00% OS°% LOT £9°8 O8°2 00'9? 00%b osté tatW ae FS bs[g? On'? ror" Hel BC°6 00'S 00r2 00°Ss 00°: caWit WALISTION NALeM904 AO BWATNA Gas 0o'8 00°4 00°Ss oO°s cAWIL CAWTE 00°8 00% 00°S 00°f 00°s samen TaWIL ¢ ‘fg a1ayv 00°8 oOOr!£ 00'S oore o0%% TAWTL 00's 00°4 00°S Q0°£ 00r' Paws aN) Ty eam + 326 OOT*O~ 100°0 TO0O"O 000°0 100°0 A VNANNN SBE TE 694'T Olt’? TOv'y TO9'T vad Ee WNS. NV NNN ooT’o- Too*o 990°O T00*0 200°0 IN M3L3W OTAND / WYYOOMIIW O0T*O- T00°O $90°O 000°O Foo0’o no v8/ST/£ d3asIn3ay OOT*O- OOT*O- OOTtO- 700°0 T00*O etoto S00'°0 t00°O TTO*O T00°O 000°0 t00°0 Z00°O £000 8To"o dd NW ad 4444 NNNAN t*0~ T*0- 4°6 £teT O° se? O°EL? S°et O*+ Orne ort IN no T*0~ O°se O*be oO°ot 0°6e a di MALIT / WYMOOUDIW T*0- 0°6 o'9 OrZ O°BE NW T*o- orest o°dot o°at O°Bcz ad 444 *“AL¥VM ONT TANYS NIW/EW SY LY AINSTITAAT ZOG ONIWNSSY WALIATION YaLYM9NA 4D BWNIOA AHL j ANSINOD MALoM WINNIT NO WAS¥A SI MIY 40 MAL3W {ND N34 SIWLIN JIVML 4O SNIMVO1 SSUW 3HL £48°O O4b°O GAT? vor't Z8tto crr'o vos S6T*t T98°O eeors 60b°T 829°O B9b°T £0N ¥8/TE/2 TASINAY 680'°0 £00°O TLO'O 990°0 voo"o 610° e9T*O ¥o0"o 880'°O §460°0 £00'°O Sz0'o £¢o"o €00°O 9T0'°0 TIE'O Z00°o é£0'o WY OW i) MALAW JIAND / WYNOOMIIW O9T't ocS'0 9T9ttT B6E°T O£B8°O 499'°T vHN $ £8 931 61 Stoto 900°0 oT0°0 000°0 B8T0*O 000°0 200°O 000'0O TIO"O 000°O oro'o 000°0 YN H 4fStt 904 YIT¥SIN 220°O 9T0°O o80'0 LOT*O £90°O 080°o a7 CEW/9) Orr o°eT OrTe 0°00 ore mite) cw *S a1avl O°sd 0°O9 O°ogt o°0et 0°09 CL°9 Seo veo? oa’? FO" Hd G26 00's 00° 00°s o0°e cawtl 00°98 00rd 00's 00°e 00°% TAWIL WOM GIL INIWS S3IWYS S Od S9VMANY TALHOTIN-3WIL £9°8 Garé 00°8 OSS 00°8 00°! 00°9 00°? 00°S 00*b 00°s oor oS*e oor 00°S AWIW @AWTL = TAWIL Aut CNTW) "SLVM ONT UWS NIN/EW SY 10 AINATOTAAA 20S ONTWNSSY NSLIFTION MALYMONA AO aNNION SHL WONA WSEL9 INOW LNAINOD NALYM HINDI T NO WAS¥VA ST NI¥ 40 £W Mad ALNTOS SSYW $ £8 Dal 61 4d ¥IWSIA ‘ob ant ann? “OM TH + 327 GELS 862°O 6£6'°O GEESE 8462°O 6£6°0O wns vos £0N VN NNN KEXEK TSS 80S°0 eSONSMSASTIO Hd +/- , ~Ad-Hd ae «MMOD e oot *o~- ooT°o- OZHI/(ATIDS wOS/(AIDS NN SOILYY HYIOW 7 7 ost T°O- T*O~ T*o~ 805'O O0°OT+ O'00OT O° LOE ore O*OvT orokee T°O- re vB/2Z1/5 MASIAAM £9T"O 000°0 £eO"O £88'O 000°O 000'0 ST0*o - SA1dWuS TT MOA JOVNSAY WALHOTAM-AWIL £9T'O 000°0 £v0"0 £88'°0 OOT*O- 900°O sto*o s*t 9°6L 00°0 ££°O £r°o- 1 W SW v3 tHN UN H om wn 41d AQIW Z3WIL TAWIL #¢ M3aLaW DTand 7 WeNDONIIW S44 CALI) CTD CNIW) *3LYM ONTT4NYS NIW/EW SY LY ADNATIIASA 20S ONINNSSY TALIATIOD NaLymSO4 40 aWn On 3H1 WOME TaLvINDWI LNALNOD YALeM INDIA NO GaS¥d SI 4I¥ 40 £W Yad FINS SSeW $ ve NUF £ - HOA YIISIA ‘e J1H¥L vyB/L£T/S MASINI ° 000°O 000°0 000°0 T00"O0 S31WdNvS T YOd SOVMAAY ASLHOTSM-AWIL 80S5°9 OOT'O- OOT*O- OOT*O- T°O- T*O~ T°o- Sto°O StT 9° 6 00'o £r'o e£°o- T +/- 901 OZHI (ATS 204 O@HD = (AIOS an TOA iu AMIW Z3WIL TSWIL # \ MBLSW SNO/HUMOONIIW 7 \I/WWS \ CIZH OMDIW 7 CEW/9)> CTH) ONTWD ALYM ONITAWYS NIW/EW GY LY ADNAIITSSS %OG ONTWNSSY TALIATION YALYM|NA 40 ANN IOA AHL WOMd TSLYINIWI LNBLNOD NAaLYM WINOIT NO WAaSVa ST NIV JO fH Nad ALNIOS SSYW $ vA NUT £ - GOA YIIWSIN *S ay va/ZT/S JaSInay : ; OET"O £v0°O BzEe"O 2tet {SOTLYM LS 83s £62 '°S z00°0 T6E%E Shes OOT*O- OOT'O- OOT'O- OOT'O- 805°0O 00°0 £E'O £E°O- 1 S4N/¥4+H S+N/H StN/PHN FOS/E£O0N YN/bOS YN/YD ¥N/OW vNs 19 +/~- iIW caWIL TAWIL # VN NNN SOTL¥M LNAWAINGS 44f¢4 4 $ 6A NUT £ - OA YIWSIN ‘e aVIVL bA/ZI/S WASINAN TS*S €£°O ££°o~ 1 0A J0L OfHI (ATS +/~ e0S £ON wm OW va bHN YN H He cIWIL THWIL # NWS NUJWHS ON T/W OOMOTW / VNNNNSN MALTIV/03 OMOTW S444 tht $ vo Ker f ~ 904 YITVSTA *T AJ VYL 328 90° £ree £t°"E KKHKK JO IOK KKK TORK K OOK HR KKK 68°9 06°? oT'o- eSONSNIAATIT H ~Ad —Hde OO0T*O- O0T*Oo- 28£'°O oot *o- Tee*oO é1S'°O F | OEHI/ CATS NN. SOTLOY 00°S 00’ oo°s 00° o0'v 00's 104 z*0~ z*0- z*O- z*0- z*0- z'O- I0L NW/ NOU/MH/ PE/9T/S WASTARM 000°O 600°O ¥00°O zTo°o SAIWYS 9 HOA S9OMAAY CLHOTAM-SWOL B89Z2'O B9Z'O OOT*O—- OOT*O- OOT*O- EtO~- 1°0- 0- 090'O O64 Oto9 OSs 00'8 00's F ve.°O = vbsto OOT*O- = OOTTO- OOT*O~ E°0-~ T°0- Tto~ £50'O O'¢ o'or GL°9 00°! og°9 & 069°0 = 069°9O OOT’O- e£0'od £10°0 z°0- Orle oer vEO'O O'S é°as GO°9 os'g v 0920 O092'°O OOT'O- OOT*O- OOTTO- 2°O- T°o~- T*o~ LE0'O Ob &° as ¥O°s EGkG £ gat8°oO S8T8'o OOT*O~ SbO'Oo ST0°O &°O= a°Be O°6 £E0'O Ofb oroe Se*e OS*tv é 949'°O 9749°O OOT'O- 840°O £vo°o &*o- O°6e O°st 060'O 0°S eee cue 00% 1 t+/- +/- J01 O€H3O (ATIS 901 OCHO (ATS 07 WA iu LIW GAWIL VAWIL ¢ «44096 \ MALSW 'NO/WYNDONDIN 7 NUVWW/ N OCT/W OMDIW 7 (END) CTH) (NTH) ALYY ONT IWYS NEIW/EW GY LY ADNATOTASS 40S ONIWNNSSY NALOSTIOD YBLeMgod AO aN OA SHL WOM WSLYINIWD LNSLNOD Y4L9M GINOIT NO Qasva ST YI¥ 4O) EW Y3d 3LNI0S SsoW $ vO NOT S - OA VIWSIA ‘£ ay e8/9T/S WASTARY ogt*o £v0'o0 8ez°0 2tet sSOILYY LWs VAS oot *o- Gov't 000°0 Gov't S0E°T 000°89 = OGE'E OSZ°O G4e°ee B9d'o Outs 00°8 ooT’o- bert 000°0 bcL’t ecort oo7'sé OOT’t 00c°0 008 t bez oO GZL°9 00%! 190'°0 best 000°O bes’t oto%2 L£99*EL BONE OSe*o 000'°6 049°O ZO°o OG oot *O- Bée°T OOT*'O- B2ébwT SSe°e Sev'esat £82°0 £Ta°O £86'°8 O9L°0 vO°S eG'S zv0"0 9GE'T 000°O 9GE°T BEE ts OOS eTe O0O'rT OO0'E o00'ée BTBSO Gee On'¢ 250'O Tées't 000°O Tés* £98°T O00"ERE OOG*ZT O00'T OOG'HF = P49°O care Oo'b POS/(ATIS StN/¥4+H StN/H StN/PHN VOS/£0ON YN/b0S YN/OD YN/OW YN 1) +f- SUIW CAWIL VAaWIL # MVTOW ST NNAN SOILVY LNDWAINOA 4/44 $ 68 NUT G - HOA YT WSIA °c SRIVL vE/9T/S WASTAaM T*0- T*o- 892'°O Obes O°etd O°Zsat or9 O'Ove#t tO 00°8 9 T°0- T*o- beZ*O OFF Ores O'eT org or0Z46T TO Ss O'TEe orerT 069'O é O°so0r ore OToTat [eat] ¥ T°O- T*0o~ 09440 O° 908 Org O° ovo? Teo~ F o°8e Or6 BTea'o Ole? ore ood ro é O's Q°st Pb9°O 0° 086 O° eel Ore To Ooty t O¢cHI (AT)S t/- v0S £ON WwW ow vo H Hel WEL t4Wtt 4 \ VW ONOIW / NNNNNN MALI OTH irs $ vo NYY S 329 oot *o- ooT*o- 000°0 oot’ o- oot’o- Too'o . A NNNN 66S°S cS9°9 6SE'°9 209°E Bcb°£ BOT? Lett WAS N NNN ooTt*o- o0Tt*o- t00°oO oot'o~ oot’ o- £00°O In. 4a.13aW a20et B9S'°T See't 2v9°0 075°O 660°T 94E°E vos oot*o- ooTtto~- T00°O GOT *O~ OOT*o- Z00°O no 31ana / WYYDONIIN *3LYM ONITMNYS NIW/EWN S ¥ LY ADNATOT 44a XOS SNINASSY DWILDATIOS MaLYNDO4S JO SIWNIGA SHL NOMS ULV INIW) ANBLNOD NALYM GINVIA NO GaS¥d SI YIU 4O HALAW *19 Nad SW1aN BOvNL 40 ONTOVOT SSYW JHL ver'z 609s 280% 6d9'°T 6E9°T TOTES vle's LON v8/ST/£ daSIAay OOT*O~ ooT*o- zt0°O oot *o- OOTtO- 200°0 ad ooT*o- OOT’O- ooTt*oO- a0T*o- 900°0 Sto0'o OOT*O- OOT*O- OOT*O~ OOT*O- £00°0 4v0°0O NW 34 44st vA8/9T/S GasIn3ay £2Z'0- 86£°0 vEeo°"o cEetto T0E*90 orto ete to ar] MBLEW OLN / WYNSOMIIN *3luM ONITdW¥S NIW/EW SY LY ADNSTOTAA 40S ONTWNSSY ELIT TIOD YaLYM90d AO SWNT0A €00'0 vo0'o tToo"o T00"0 Z00°O T00'o Z00°O OW 2£20°O TEO*O vro'o éT0°oO oto'o O£0°O £90°O vo t*o- T°o- o'9 yto- t*o- o's (a) VN NNN GL9ET cé6'T G48'Tt cert ztO'T vLort ee2°£ vHN t*o- t*0- vot T*o- T°o~ Oée IN T*0- T°o- O°eZe T*o- tT’o- O'°9% no MALIT / WYMOOHIIW T*o- Tto- o*oct T*o~ T*0- O°9Z dd T*o- T°o- o°et T°Oo- T*O- Ore NW $ v8 NUE S - 503 YIIVSIN OTO*O tro"o0 cto°o 600°O ST0°O <00'O ¥00°O ON 000°0 000*90 900°9 000°0 Oot *0- 000*°o 000°0 H frre aroro 070°0 £S50°O veo'o £50°O 060°0 ont (£W/9) Cw) T*o- T*o- O'tey T*0O~ T*o- O'tes x: Fe ffl *S Adv. 0°09 O° Or 2°8s é*8S Oror e°ce Lil CNIW) GOCE £1%Z £t°2 OT’o- 68°9 06°9 Hd A 00°8 oo'd on'9 éS°S OS’ Q0*v CaWIt 00°% os’? £S°G SS*¢ 00*+ £9°L =aNM TS TaWIL # Sa WdNvsS 9 Od A9gvMaIAG GALHOTSM-3WLL 08'S SL°9 Cor? vors Sé°v ere ALdIW 00° oor ost? eS'S OS 00'y ZaWwt 00° os’? £S°*S SS*v 0o'v £9'E BAM TS TaWIL ¢ 3HL WON4 TALYINITWVD LNILNOD NALYM GINVID NO WaSvd SI MI¥ JO fH Mad ALIS SSUW $ ve NOP G - 904 VITSIN ‘ep F1avL 330 Zee ated weexxk Tote xaEK 269 akaeee ZTOS weenk FS aueek 26°9 sFON3MAAATO Hd ~Ad~-Hde ooT*o~ 69£°O ET9°O SSr’°o certo ooTF*o- OZHI/ (AIDS NN 00° oo°st o0°2 00°12 00° tz 00°9r A NW? soliva Z'o- o°odte c*o- o*oret 0°O009T o°ozét 901 eevee S9S°o TSe°o Btr'o 60¥°0 esvr‘o t/- eddOIe OOF *O~- £zr°o ztz*o ELt'o 790°O oot'o~ vOS/(AIDS Mv TON / / vaset/s Ua81nay ALVY ONITNYS NIW/EW GS Y LY AINSIDT 449 %OS ONINNSSY FHL WONS WALVINITWI LNGINOD Y310M UINVID NO dasvd SI YIv JO f£W Yad 31N10S SSuW ZLE*SEOTSZO°O Zeus Q0T*O- O0T*O- S9S‘O T6*SOST ETT*O TS£°O OOT*O- £F0°O alv’o Seeecczt 1S50°0 600°O So°ttet gso°o zsv'o LO°S9¥T 070°O t/- J0L OZHI N vas9t/S d3sinay S6T°e 000°0 S6Tté 419°T 000°0 L£19°T TO9's 000°0 TO9"S OLe°% 000°0 a2e°s BBETS 000°0 L482°S OoL'z 000'0 662°SZ StN/¥4+H 84N/H = S+N/UHN NNNAN ve/9T/S GaSIAay t0- t*o- ZETS o°s¢ O've S95°0 O°tEe o°ét TS£°O o'zz o°ot cad Akt) o*z2 Oo’ 40¢'O 0°82 t*o- esv’o OZH2 = (ATDS /- NVWH/ NOV OdOIW / o* tog orctt 0° 06£ 0°86 O*6<ét 0°66 0'OST orctt O°ceT orest O°vet o*eé! vos £0N NNNNNSN oTo’o OOT*O- ct0- vvo'o o°02TzZ Bc0"O c*0- Sz0°o O°oret Tro°o0 o°009T OOT°O- O*02Zt DS 301 €sS*0 1SE*O £os°o oad*o ace*t SZEST ¥OS/20N tzo° r*o- t°0- Ovo: o*s? O°be aso’ O°'te oO°sét Zvo’ orze o°or ££0* o°ze o°v vao* 0°a8e t°0- té0° O2@HI (AIDS ont M319 *NI/WUNOONDIW / NI/RW/ \ O1/H OMDIW / CEWS soriew .iNaWwAINGS o°ce4lt T*0- t*o- T*o- Ore 0°SZ ” $6 va Nor Z ozt°o £v0'o Q0S°O0T O0T*O~ 000'°S6T 000'bT 00S°68 000°S 000°OS §£FLt o08'OL OOS'T 9SS°6T cez*t YN/bOS YN/YD 6 v8 NOF 2 oz 0° oT- ors O*8z% ore Oot ort o'r o'r o'9 org ort OW vo yaitWvog oyIW - 904 VI 8zz°o o00°t ooo°T 000’T £Er*o Ogc'O zee°oO YN/OH ~ 004 ¥I 0°489 0° 482 Ofec~s 0°909 O°BZ? 0°098 vHN $ va Nor 2 - 904 YI oO O°0£ tT 0°06 9 £ £ Z 0°09 I is ° Oo Oo o Oo Oo ote e°6oT Oo ote z*oor Oo o°oF 0°06 WA iit 9) CW) CNIWD WSIh Ztrt sSOTLYN 1L1¥S vas 000°99S ZETS 80° fed ooT'O- S97S‘°O 80°a £8°B OOTtO~ TSE*O £8°9 LEE OOT'O- STr*O Tw's £“'°9 00S‘0 60¥'°O 99'°o On'¢ ALLE 2Sv'o 80°% Lea's YNs 19 +/- 4aTW ZaWIL 4444 Wsin ‘2 awvi ore T'o 9T'£ ff°6 or°e T*o0 Te*é £6 ore to 469 ERS o's T*o at*L ££? o'v T'o £o°L£ OS* 0'6 T'o L459 EBS ¥N H Hd Zawtl S4lhité WwsiIn "Ty 21a S31dNUS 9 NOS SOVNSAY DILHOL aM~ IWEL 80°6 g0*s £a'9 Tes 99°E g0°% 4dIW EL'A £8°s EES £e°9 ost £e'z cawih GaL9371109 44aive904 40 aNnnon £8°@ ££°% L“°9 os*v £a°z ge°r TaWIL @ anh ese Hwo- *s aay £a'a EL°Z £E°? OS' €a°S £ECT aN MeN ~o TaWIL ¢@ £8°s KEL £“°°9? os’ £8°Z egret mNM TIS TanIi # 331 b8/ST/£ WASIARY OOT*O- OOT'O~ OOTTO- OOT*O- OOT*O- O0T*O~ T*0- T*o- T*o- T*o- T*0- T'o- 9T*L £E'°S6 ra°s 9 T00*O T00*O 000°0 £00°O 000°O £00°O Oror ortt o't O'9v O° O'és to's £88 ge°2 SS OOT*O= OOT*O- OOT*O- OOT*O~ OOT'O- OOT*O- T*o- T*O~ T'o- Tto- To- T'o- L£6*F ELOL £L°9 000*°0 000°O 000°O £00'O 000°O Z00°O oo an ort O°eEe Ore o°Ze ets LE°9 OStb £ 000°0 000°O 000°0 voo"o 000°O £00°0 ors ve Ore orse o'r o°438 £ets OS'b rere ¢ to0°0 too*o T00°O £00°0 Too’o gTo"o oroT 9'B8t octt O°9E ors O°vee L6°9 EATS fer “A IN na ad NW 34 A IN na ad NW 34 Hd eaWwilL {TAaHIL # N NNN MEL3W DIAN / WYHYODOMIIW 44 VN NNN MALI / WVNOONSIN a an ae *FLYM ONIVAWYS NIW/EW GY LY AINSIDTAAS 20S ONIWNSSY WaLIaTIOO M3LYM904 JO AWNTOA SHL WOYS GALYINIWS LNJINOD YALYM GINO NO dasva SI MI¥ 430 Y3LaW ‘ND Yad SIWLaW 30041 40 SNTC¥OT SSUW JHL $ v8 Nor 2 - 904 YI WSIA ‘Ss aL vaset/S dasiIn3y ztTS'S 619°O sas'o ébz'o T00°O STo*o0 £28°0O 900°0 000°0 £90°O Sa1dw¥S 9 YOJ BOVNaAY CILHOTAA~-AWIL B9L°T 98£'0 822°0 so9't 100'°O 000'OT- 96¢°O z00°o 000°0 ovo’o ore O°o£ 80'°6 £E°S6 £8°8 9? 262°S c8o't TSE°O OOT'O~ T00°O €£0'0 zce’o £00°O 000°O aco°o oret 0°06 80's £8°S8 £e°e S 60f°T Tov'o 98c°oO OOT*O- T00°O 600°O 609°O zo0'%o 000°0 ébo'o org 0°09? £B°9 Lez ££0°9 SSé°T TSS*O SSS'O OOT'O- 100°O 97000 9£8°O S00°O 000° £20°O orte 6'40T Te’s ££°9 ostr £ £BE°S téer'o cva’o 900*0 T00'°O oto'o sczo°t 800°O 000°0 ¥80°0o orle z*oor 99° OS't £a°c fz ZL9°S TOv"O £28'O =£90°0 €00°0 9T0°0 £ot'T St0°O 000°o TZ0°O oror 0°06 B0°S £82 coer of wns vos £0ON me] OW v3 vHN UN H amt ‘10 iu A0IW ZaWIl TAWIL ¢ NNNN : YALSW BIND / WOHXOONITIW 4447 (tW79) CTW) (NTH) ‘310M ONIIMNYUS NIN/EW G Y 10 ADNSIDI449 40S ONIWNSSY W4L9471090 NaLvM904d 40 SWIN10A 3HL WON CT3LYINIWI LNBLNOD N3Livn WINOIT NO Wasva SI YI¥ 40 EW yad JLN10S SS¥W $ va nur 2 - S04 YIIVSIA ‘ey av