THE EFFECTS OF AIR POLLUTION ON WATER QUALITY PEDCo ENVIRONMENTAL ------- 950R77O10 PEDCo ENVIRONMENTAL 11499 CHESTER ROAD CINCINNATI. OHIO 45346 (513) 783-4700 THE EFFECTS OF AIR POLLUTION ON WATER QUALITY Prepared by PEDCo-Environmental, Inc. 11499 Chester Road Cincinnati, Ohio 45246 Contract No. 68-02-2515 Task No. 2 Project Officer: Justice A. Manning Task Officer: Don Lokey Prepared for Strategies and Air Standards Division Office of Air Quality Planning and Standards U.S. Environmental Protection Agency Research Triangle Park, North Carolina 27711 March 15, 1977 BRANCH OFFICES CHESTER TOWERS Crown Center Kansas City. Mo. Professional Village Chapel Hill. N.C. ------- This report was written for the U.S. Environmental Protection Agency by PEDCo Environmental, Inc., Cincinnati, Ohio, under Contract No. 68-02-2515, Task No. 2. The con- tents of this report are reproduced herein as received from the contractor. The opinions, findings, and conclusions expressed are those of the author and not necessarily those of the U.S. Environmental Protection Agency. 11 ------- ACKNOWLEDGMENT This report was prepared for the U.S. Environmental Protection Agency by PEDCo Environmental, Inc., Cincinnati, Ohio. David W. Armentrout was the principal investigator and author. Mr. Don Lokey was the task officer for the U.S. Environ- mental Protection Agency. We wish to thank him and also Mr. John Bachmann of the Agency for their technical assistance and comments throughout this task. 111 ------- TABLE OF CONTENTS Page ACKNOWLEDGEMENT iii LIST OF TABLES V 1.0 EXECUTIVE SUMMARY 1-1 1.1 Conclusions and Recommendations 1-1 1.2 Survey Results 1-2 2.0 INTRODUCTION 2-1 2.1 Purpose and Scope 2-1 2.2 Evidence of Atmospheric Effects on Water 2-2 Quality 3.0 SURVEY RESULTS 3-1 3.1 Trace Metals 3-1 3.2 Nutrients 3-14 3.3 Toxic Organics 3-23 3.4 Airborne Acids 3-27 4.0 CONCLUSIONS AND RECOMMENDATIONS 4-1 4.1 Conclusions 4-1 4.2 Recommendations 4-8 4.3 Intra-Agency Considerations 4-11 REFERENCES ------- LIST OF TABLES No. Page 1 Estimates of Potential Trace Metal Input To 3-2 Lake Michigan 2 Estimates of Trace Metal Input To Lake Washington 3-3 3 Trace-Element Input to Lake Superior From Four 3-5 Major Source Categories as Estimated in the Upper Lakes Study 4 Trace-Element Input to Lake Huron From Four Major 3-5 Source Categories as Estimated in the Upper Lakes Study 5 Elemental Ratios in Rain, Soil, and Fly Ash in 3-8 the Walker Branch Watershed 6 Summary of Trace-Element Mass Balance Results 3-9 for January-June 1974 at Walker Branch Watershed 7 Estimated Wet Deposition vs. Dry Deposition of 3-9 Selected Elements in the Walker Branch Watershed 8 Nitrogen and Phosphorous Input to Lake Superior 3-18 From Four Major Source Categories as Estimated in the Upper Great Lakes Study 9 Nitrogen and Phosphorous Input to Lake Huron 3-18 From Four Major Source Categories as Estimated in the Upper Great Lakes Study 10 Drift vs. Particle Size Pesticides 3-24 11 Estimated Inputs to the Southern California Bight 3-26 VI ------- 1.0 EXECUTIVE SUMMARY This report summarizes the results of a literature re- view performed for the purpose of defining the known and potential impacts of air pollution on water quality. The literature survey was supplemented by a limited number of conversations with researchers who have been involved with intermedia pollution problems. Air pollution effects on water quality have been a significant part of environmental research in Europe, espe- cially in Scandinavia, for several decades. These effects have more recently become a topic of environmental research in the United States and Canada. For convenience, the potential effects are discussed for four categories of pollutants: trace metals, nutrients, toxic organic com- pounds, and airborne acids. 1.1 CONCLUSIONS AND RECOMMENDATIONS For every pollutant category reviewed, the research evidence indicates that the atmosphere can be a significant source of pollutant input to water. The development of continuous data bases to provide statistically significant data has been limited by problems with sampling methodology and by a total absence of interagency project coordination 1-1 ------- within the United States. Intergovernmental cooperation and information exchange among the United States and foreign governments has been good, but a serious effort to coordi- nate studies within the United States is needed. Too many small-scale projects are being funded through various agen- cies. It is believed that better data and better project quality control could be obtained through a planned research program dealing with intermedia effects. The survey concludes that expanded research is neces- sary to accomplish the following specific tasks: 0 Identify pollutant sources and relative source contributions to the total pollutant impact on water. 0 Improve technology for sampling the dry deposition of pollutants onto water surfaces. 0 Expand the data base through expanding geograph- ical coverage of precipitation chemistry networks and through more long range studies of specifi" watersheds. 0 Determine cross-media impacts, including direct impacts of air pollutants on water and the in- direct impact of air pollution control technology on water quality. 0 Expand research to define effects of pollutants on aquatic life forms and to determine the mechanisms of those effects. 1.2 SURVEY RESULTS 1.2.1 Trace Metals Insufficient data have been documented to allow reason- able predictions to be made of the impact of atmospheric 1-2 ------- trace metals on water in any specific geographic locations. The presumptive evidence shows, however, that the air is probably a significant source of trace metals detected in some water systems. Data from Lake Michigan studies show that the atmosphere could be a primary source of input for at least nine trace metals into that lake. Studies of the Lake Washington Watershed in western Washington and the Walker Branch Watershed in Tennessee present similar con- clusions for their associated water systems. The effects of trace-metal levels on the biota in water are not thoroughly understood. A major theory is that excess concentrations of trace metals are toxic at acid pH. More research is required to define excess concentrations and to determine the mechanisms of toxicity. The increased application of emissions controls on some fossil-fuel-burning installations may have little effect on decreasing the potential atmospheric loading of trace metals into water. Studies show that trace metals are concentrated during the combustion process, and that a significant amount of these metals are included on the particulate fractions that escape electrostatic precipitators. 1.2.2 Nutrients Lakes naturally evolve from an oligotrophic (nutrient- poor) to a eutrophic (nutrient-rich) state. Human activi- ties can accelerate the eutrophication process, causing an 1-3 ------- overabundant concentration of nutrients and subsequent stimulation of life cycles and depletion of oxygen within the eutrophic lake. Nitrogen and phosphorous are the most frequently studied nutrients. Several studies show that whereas pre- cipitation is a significant source of nitrogen in surface water, it appears to be a less significant source of phos- phorous. Other studies show that precipitation is not a significant source of either nitrogen or phosphorous to the watersheds or water systems studied. The importance of atmospheric input of nutrients to water appears to be a function of increased industrial or agricultural activity. Problems with sampling methodology limit the data available on the determination of dry deposition of nutri- ents into water. These problems also limit emission source identification. 1.2.3 Toxic Organic Compounds The toxic organic compounds most often cited in the literature reviewed are polychlorinated biphenyls (PCB's) pesticides and, more recently, nitrosamines. The toxicity mechanisms of the toxic organic compounds are not completely understood. PCB's and DDT have been implicated as causative agents of cancer, however, and since these compounds can bioconcentrate in fish, their presence in water presents a hazard to humans. 1-4 ------- It is estimated in a study of the sources of PCB input to the Southern California Bight that the atmosphere may be at least as significant as municipal wastewater systems as a source of PCB input to that area. Little significant in- formation appears in the literature to quantify the contri- bution of atmospheric sources of toxic organic compounds to concentrations of these compounds in water. 1.2.4 Airborne Acids Extensive literature is available concerning acid pre- cipitation. The results of acid precipitation studies in Europe, the United States, and Canada in recent decades present strong evidence that increased acidity of precipita- tion is the result of atmospheric pollutants from fossil- fuel combustion. The pH of precipitation in much of the northern United States is reported to average between 4.0 and 4.2. Values as low as 2.1 to 3.0 have been reported during isolated storms. In Europe, particularly in Norway, the consumption of fossil fuel, the increased acidity of precipitation, and the increased number of barren lakes can be correlated. Although more work is required to determine pH effects on aquatic life, previous studies indicate that low pH affects the reproduction and possibly the metabolism of fish. Low pH levels also may affect the toxicity of trace metals. 1-5 ------- 2.0 INTRODUCTION 2.1 PURPOSE AND SCOPE The Strategies and Air Standards Division of the Office of Air Quality Planning and Standards (OAQPS), U.S. Environ- mental Protection Agency (EPA) has become interested in the effects of air pollution on water quality. Preliminary evidence that air pollutants can have a significant impact on water resulted in the initiation of a contractual effort to further document that potential.3 This report summarizes research related to defining the effects of air pollutants on water quality. Runoff of terrestrially deposited atmospheric pollutants and leaching or runoff of residuals from air pollution control activities constitute two indirect effects of air pollution on water quality. Direct impacts occur as a result of gas phase interaction between air and water, dry fallout of aerosols on water surfaces, and rainout or washout of atmospheric compounds directly onto water surfaces. The direct impacts from trace elements, nutrients, toxic organic compounds, and airborne acids are reviewed in this report. a Internal memo by John D. Bachmann and Don Lokey, U.S. EPA, Research Triangle Park, NC. "Effects of Air Contaminants on Water Quality", September 2, 1976. 2-1 ------- The information in this report was derived from a literature review and from personal conversations with researchers. The sections discussing each of the four major pollutant categories summarize the extent to which both quantitative and qualitative data have been reported. The report also summarizes required further research and coordi- nation of interagency research activities. 2.2 EVIDENCE OF ATMOSPHERIC EFFECTS ON WATER QUALITY Several organizations other than the U.S. EPA have conducted research on one or more subjects relating to the effects of air pollution on water quality. The most sig- nificant among these organizations are: 0 U.S. Department of Agriculture 0 U.S. Geologic Survey 0 National Oceanographic and Atmospheric Admini- stration (NOAA) 0 National Space Administration (NASA) 0 National Science Foundation (NSF) 0 International Joint Commission (coordination of U.S. - Canadian studies) 0 Atomic Energy Commission [now Energy Research and Development Administration (ERDA)] Several of these organizations have carried on research in this field for some time. NOAA, for example, has cooperated with the World Meteorological Organization in precipitation chemistry studies for several years. 2-2 ------- The effects of air pollution on water quality have been studied more extensively in Europe than in the United States. The preliminary evidence derived from both U.S. and European studies indicates that a significant proportion of the atmospheric pollutants that impact on water surfaces are of anthropogenic origin. In Europe, the changes in water quality correlate with changes in the use of fossil fuels and with increased urbanization and industrialization in general. Emissions inventory techniques coordinated with water and air-quality-sampling projects in the United States lead to similar conclusions. Studies from both areas also indicate that atmospheric input of some pollutants into lakes represents a significant percent of the total loading of those pollutants. 2-3 ------- 3.0 SURVEY RESULTS 3.1 TRACE METALS The mechanisms characterizing atmospheric trace-metal input into lakes are functions of particle size distribution and particulate aerosol transport and deposition properties. The literature reviewed indicates that one focus of the research in this area has been a more precise description of the physical parameters involved in the atmospheric loading of trace metals onto lakes. Field measurements obtained from the studies that have been performed, however, are not sufficient to provide a reliable data base. Although most of the preliminary data found in this survey do not appear to be derived from statistically significant field measure- ments, they do provide evidence that the atmosphere is an important source of trace-metal input into lakes. More work is required to identify sources of each element, to measure deposition onto lake surfaces, and to build reliable histori- cal data bases. From the survey it appears that the most significant work in the United States has been performed on the Great Lakes and on the Walker Branch Watershed in Tennessee. 3-1 ------- 3.1.1 Loading Estimates Table 1 partially summarizes data reported in one study of Lake Michigan. The estimates shown were derived from emissions data and from National Air Surveillance Network (NASN) ambient air data for Chicago, Milwaukee, and north- east Indiana. The study concluded that the atmosphere may be a significant source of input of at least nine trace metals into Lake Michigan: cobalt, copper, manganese, nickel, lead, selenium, titanium, vanadium, and zinc. The conclusion that an element from the atmosphere may have significant impact on lake loading was based on the hypo- thesis that approximately 10 percent of the atmospheric emissions might be transferred into the water. Other more recent estimates of transfer efficiency have ranged as high 2 as 20-25 percent. Table 1. ESTIMATES OF POTENTIAL TRACE METAL INPUT TC LAKE MICHIGAN1 Element Arsenic Beryllium Cadmium Cobalt Copper Iron Manganese Nickel Lead Selenium Tin Titanium Vanadium Zinc Stream input, 103 kg/yr 59 6 209 19,958 208 9 91 6 91 26 590 Air emissions inventory, 10 3 kg/yr a 40 4.5 11 45 2903 78,019 4173 907 1996 18 12 2177 553 3538 Potential loading air to water" 103 kg/yr 4 0.4 1 4 290 7802 417 91 200 2 1 218 55 354 • Based on pre-1970 data. Does not reflect p^st-1970 con- trol strategy effects. Based on an assumed 10 percent transfer efficiency. 3-2 ------- A study of the Lake Washington Watershed in western Washington also included estimates of atmospheric loadings of six trace metals: iron, sodium, lead, zinc, copper, and mercury. Table 2 summarizes estimated annual atmospheric input of these metals compared to estimated annual stream input. Stream input was estimated from river samples; atmospheric dnput was estimated by subtracting net fluvial input from annual sediment input. Table 2. ESTIMATES OF TRACE METAL INPUT TO LAKE WASHINGTON3 (103 kg/yr)a Element Iron Sodium Lead Zinc Copper Mercury Annual Input Stream 2331 4917 1 30.8 5 0.06 Air 1996 163 27 13.6 2.5 0.05 Numbers are approximate and have been converted from English units originally reported. Neither the Lake Michigan nor the Lake Washington study is based on extensive field measurements of atmospheric fallout. Other studies found in the literature reviewed provide similar presumptive evidence of the significance of atmospheric input of metals into lakes or oceans. A joint U.S.-Canadian study and a study of the Walker Branch Water- 3-3 ------- shed in Tennessee appear to be most comprehensive in quan- tifying atmospheric trace-metal input to lakes by using field measurements. A 2-year study of Lake Superior and Lake Huron wag 4 5 recently completed. ' The study was coordinated by the International Joint Commission, and the research was funded jointly by the United States and Canada. This study of the upper Great Lakes was initiated to estimate the annual atmospheric loadings of 16 pollutants onto Lake Superior and Lake Huron and to identify the relative contributions from 30 major large-area sources. A mathematical model was applied to estimate pollutant loadings. Precipitation sam- ples were collected and analyzed from a network of 70 sam- pling stations on and around the two lakes. Tables 3 and 4 summarize the mean annual atmospheric loading estimates for the trace elements analyzed in this study. Inputs from four major source categories - direct municipal, direct indus- trial, tributary, and atmospheric - were estimated for Lake Superior and Lake Huron. The tributary loadings include all upstream sources, including municipal, industrial, and land drainage. The percent contribution of each element from atmospheric load- ing is expected to increase as cleanup of aqueous point sources continues. 3-4 ------- Table 3. TRACE-ELEMENT INPUT TO LAKE SUPERIOR FROM FOUR MAJOR SOURCE CATEGORIES AS ESTIMATED IN THE UPPER LAKES STUDY4 Element Cadmium Copper Iron Potassium Magnesium Sodium Nickel Lead Total Input, 10 kg/hectare (ha) Direct municipal 0.024 1.4 74 177 216 847 0.234 0.383 Direct industrial 3285 10 239 2825 3146 30,733 10 3285 Tributary 351,000 1015 65,335 86,140 361,350 196,000 613 1105 Atmospheric 55,000 369 9700 13,000 5584 15,000 120 650 Contribution from precipitation, % 13 26 13 13 2 6 16 13 OJ I Ul Does not include dry deposition, Table 4. TRACE-ELEMENT INPUT TO LAKE HURON FROM FOUR MAJOR SOURCE CATEGORIES AS ESTIMATED IN THE UPPER LAKES STUDY5 Total Input, 10 kg/ha Element Cadmium Copper Iron Potassium Magnesium Sodium Nickel Lead Direct municipal 0.8 7 136 628 668 6680 7 3 Direct industrial 0.8 6.8 2376 1565 650 10,950 3.7 5 Tributary 796 1358 32,960 76,285 456,250 405,150 1478 1200 a Atmospheric 79 760 4600 32,000 8200 45,000 210 780 Contribution from precipitation, % 9 36 11 29 2 10 12 39 Does not include dry deposition. ------- The atmospheric loading estimates in this study do not include the potential contribution from dry deposition, which is the fallout of particulate separate from precipi- tation events. If the atmospheric input from dry deposition is significant, then the total atmospheric contribution of trace metals to these lakes will be more important than the results of this study indicate. The sampling network used in the Upper Great Lakes Study is being expanded to include the other Great Lakes. U.S. EPA Region V has the responsibility for implementing and operating the surveillance network on the U.S. side. The expanded Great Lakes surveillance should provide an expanded data base, one that will allow more comprehensive estimates of atmospheric loading of trace metals and other pollutant classes and better validation of models. Accurate estimates of dry deposition of particulate onto the Great Lakes will require refined sampling method- ology. To date, most work to characterize dry deposition phenomena has been performed in laboratory environments. Measurements in the field are hindered by problems such as: 0 contamination by insects and other foreign matter 0 sample loss from wind currents 0 restrictions in sampling small particles 0 a lack of particle size distribution data asso- ciated with previous experiments 0 logistics of operating remote samplers. 3-6 ------- Current estimates of dry deposition are based on emission inventory and meteorological data applied to mathematical 126 models. ' ' The deposition functions in the models are based on laboratory experiments. Better estimates may be possible only by refining sampling techniques and emission inventory techniques. A second major study was performed on the Walker Branch Watershed in Tennessee. This study was supported by the National Science Foundation and the U.S. Energy Research and Development Administration, and the work was performed by the Oak Ridge National Laboratory, Oak Ridge, Tennessee. Field data were derived from precipitation, stream, and Hi- vol samples. Precipitation samples were collected during the period from June 1973 to July 1974. Dry samples col- lected in July 1974 were analyzed for 34 elements to esti- mate the elemental concentration in ambient air at ground level. Enrichment factors were calculated to relate the atmospheric concentration of each element to the expected concentration from natural sources. This technique provided estimates of the relative importance of anthropogenic source contribution to atmospheric concentrations of each element. Ratios of cadmium, copper, mercury, lead, and chromium to manganese were calculated for rain and soils in the Memphis area and for flyash collected at the Allen Steam Plant in Memphis. The Memphis data were the only data avail- able for this study, and they were determined to be suffi- 3-7 ------- ciently representative of conditions around the Walker Branch area. The calculations indicated that cadmium, copper, mercury, and lead in rain are enriched with respect to soil in this area; therefore, they must have another source. It was concluded that flyash ratios are comparable to rain ratios for copper and cadmium, and possibly for zinc. Chromium appears to be derived from the soil in this area, and lead is probably derived from automobile exhausts. Table 5 shows the elemental ratios in rain, soil, and fly ash. Table 5. ELEMENTAL RATIOS IN RAIN, SOIL, AND FLY.ASH IN THE WALKER BRANCH WATERSHED7 Rain Soil Fly Ash Cd/Mn 0.39 0.002 0.12 Cr/Mn 0.11 0.13 2.1 Cu/Mn 1.47 0.043 1.1 Hg/Mn 0.01 6x10-5 0.0001 Pb/Mn 1.8 0.04 0.27 Zn/Mn 3.8 0.7 13.7 Total input of selected elements from stream sources and from the atmosphere were compared to total estimated output. An estimate was made of total watershed retention as a function of percent total atmospheric impact of seven elements. Table 6 summarizes the trace element mass balance for the watershed as estimated for the 6-month period Jan- uary-June 1974. These estimates are significant, because they illustrate the possibility of predicting the total impact on the watershed resulting from an increase in atmos- pheric concentration of each element. Predictions of this type can be valuable in planning future land use in any geographical area. 3-8 ------- Table 6. SUMMARY OF TRACE-ELEMENT MASS BALANCE RESULTS FOR JANUARY-JUNE 1974 AT WALKER BRANCH WATERSHED7 Element Cadmium Chromium Copper Mercury Managanese Lead Zinc Atmospheric Input, g/ha O- 62.4-65.4 22.1 143.4 1.51-1.89 136.7 115.4-171.9 269.4 Total Stream3 Output, g/ha a 3.5 9.1 23.09-26.37 0.47 59 2.9 67.9-73 Watershed Retention, % of , Atmospheric Input lod *(*~3)/V 94-95 58.8 82-84 69-75 56.8 97.98 73-75 b, Total stream output is the combination of dissolved trace elements and suspended particulate species. That portion of each element entering the total watershed and not being exported by the streams as a result of runoff or direct deposition into the streams from the atmosphere. Table 7. ESTIMATED WET DEPOSITION VS. DRY DEPOSITION OF SELECTED ELEMENTS IN THE WALKER BRANCH WATERSHED7 (July 1974) Element Cadmium Chromium Copper Mercury Manganese Nickel Lead Zinc Wet Deposition, g/ha 4.5 0.5 6.5 0.04 4.4 0.9 7.9 16.8 Dry Deposition g/ha <0.21 0.13 0.15 <0.01 0.87 - <3.9 1.7 % of Total Deposition <4.5 30.6 2.3 <20.0 16.5 — <33.1 9.2 3-9 ------- Table 7 shows a comparison of total dry deposition input to wet deposition input for eight elements. Dry deposition estimates are extrapolated from only one month's data; expanded sampling would be necessary to draw firm conclusions. Although based on limited field sampling data, the preliminary estimates are nevertheless important because they imply that the contribution to total watershed loading from the atmosphere is significant for some elements. This preliminary evidence also illustrates the need for expanded studies to sample dryfall deposition over lakes and water- sheds. 3.1.2 Source-Receptor Relationships Determining source-receptor relationships for trace elements requires an understanding of the emissions from selected industrial processes and from the combustion of fossil fuels. A significant amount of work has been com- pleted or is under way to characterize the distribution of trace elements from various industrial processes and from fossil-fuel-burning sources and to study the atmospheric distribution of these elements. An improved means has been reported for determining elemental chemistry as a function of particle size using a commercial cascade impactor and the Q ion-excited X-ray analysis method. This has proved to be an important application for studying the enrichment of 3-10 ------- trace elements during fossil fuel combustion. Other studies concentrate directly on characterizing removal efficiencies and on the enrichment and particle size distribution rela- tionships of trace element emissions from fossil fuel com- bustion.9'10'11'12'13 All of these studies are important for understanding the potential loading of the atmosphere, and consequently of water surfaces, with trace elements from combustion sources. The expected impact of trace-element atmospheric emis- sions on surface-water trace-element concentrations appears to be a function of land-use configuration, of industry-type distribution, and of fuels burned within an area. A recent study of power plant boilers indicates that trace metals are volatilized and enriched during coal combustion. In addition, aerosol studies of six U.S. cities indicate that a significant percentage of the particle mass of aerosols in selected cities exists in the size fraction less than 2 micrometers (ym) diameter. These observations are signifi- cant because they provide preliminary evidence that long- range transport of particles enriched in trace elements, i.e., the smaller particles that may escape some conventional control devices like electrostatic precipitators, presents a potential threat to water quality. This potential threat might remain regardless of the degree to which major fossil- fuel-burning plants control their emissions. The Upper Great Lakes study estimated trace-element loadings by applying an atmospheric transport model. The 3-11 ------- particulate emissions from 30 major source areas were input into the model as trace-element constituent emissions. The trace-element constituent emissions were estimated by applying a constituent fraction matrix, derived from a literature survey, to the emissions categorized by major industrial groups for each major source area. The study concluded that major trace-metal input to Lake Superior came from Chicago, Cincinnati, and the State of Michigan, and the major trace-metal input to Lake Huron came from Detroit, Saginaw, and Cincinnati. This is the only study reviewed in this survey in which estimates were made of trace-element contribution from sources beyond the immediate area of the body of water in question. The Walker Branch Watershed study included a consid- eration of minimum atmospheric discharges from coal-fired steam plants in the Oak Ridge area. The purpose was to derive enrichment factors to distinguish between relative contributions from anthropogenic sources and from natural sources. A comparison of aerosol enrichment factors to fly ash enrichment factors led to the assumption that atmos- pheric levels of seven trace metals in this study were significantly influenced by the power plants in the area. Better estimates of source-receptor relationships, especially in the Walker Branch Watershed study, might have 3-12 ------- been possible with better sampling time resolution and particle size distribution sampling. Estimating specific source contributions to an area such as the Upper Great Lakes may not be possible using current techniques; or, if possible, it may be too expensive. An area like the Walker Branch Watershed appears to be better suited for source- receptor research because it is influenced by a relatively few large emission sources, thus allowing the collection of more precise source emissions data. 3.1.3 Effects of Trace Elements on Water Quality and Aquatic Ecosystems Little information is available on the effects of trace elements on aquatic life. One theory proffered on the effect of acidification on aquatic life is that extinction of fish populations is caused by heavy metal toxicity thab results from low pH. Zinc concentrations were considered to be potentially hazardous to fish in the LaCloche Mountain 14 lakes, near Sudbury, Ontario. Other metals appeared to be present in safe concentrations in this area. Mosses dependent on atmospheric sources for a large portion of their mineral supply have been shown to be strong- ly contaminated with cadmium, nickel, chromium, and mercury in southern Scandinavia. Sphagnum mosses appear to in- crease in density in lakes where acidification is a prob- lem. These mosses cover sediments and retard recycling of 3-13 ------- metal cations from the sediments, thereby depriving other species of the aquatic system of an important source of trace metals. 3.2 NUTRIENTS Lakes naturally evolve from an oligotrophic (nutrient- poor) to a eutrophic (nutrient-rich) state. Nitrogen and phosphorous are usually considered the most important nutrients in the eutrophication process. In some lakes phosphorous is the nutrient that contributes most to eutro- phication, i.e., is the limiting factor; in other lakes, nitrogen. Nitrates and phosphates are important plant nutrients, and their presence affects the rate of growth of many aquatic plant species. As algae die they sink and decay, consuming dissolved oxygen in the water and releasing nutrients for more plant growth. The decrease in oxygen and increase in temperature from decaying plant life signifi- cantly inhibit or alter other types of aquatic life, spec- ifically commercial and sports fish. Nutrient concentra- tions that accelerate the growth of the algae accelerate the total rate of oxygen depletion. The eutrophication process can be accelerated by such human activities as fertilization of agricultural land; discharge of pollutants from point sources into lakes, rivers, and streams; and pollution of the air. Several extensive studies provide evidence of increasing harmful 3-14 ------- effects on lakes resulting from human activities. Input of nitrogen and phosphorous from the air can come from wind- blown dust from fertilized soil, from direct fallout of pollutant emissions from fossil fuel combustion, and from precipitation scavenging. Dry deposition input of nitrogen and phosphorous to lakes has been modeled, but extensive dry sampling has not been done. Most of the literature reviewed in this survey indi- cates that the nitrogen content of precipitation is signifi- cantly greater than the phosphorous content; and precipita- tion seems to be a more important surface-water source of nitrogen for than it is of phosphorous. The National Eutro- phication Survey of the U.S. Environmental Protection Agency, Corvallis, Oregon, has studied lakes throughout the United States in an effort to determine their trophic status and to identify the factors contributing most to their eutro- phication. A sample of six working papers by the National Eutrophication Survey were reviewed in this project. These papers indicate that the percent precipitation input of both nitrogen and phosphorous into five of six lakes studied in New Jersey and Indiana is negligible.17'18'19'20'21'22 Conversely, a study of the Rhode River, a subestuary of Chesapeake Bay, indicates that precipitation input of nitrogen into the estuary, along with input from distant 3-15 ------- 23 point sources, maintains the estuary in a eutrophic state. The watershed has no significant point sources, and the percent nitrogen contribution from runoff is negligible. Studies of the Great Lakes also show that precipitation can 324 be a significant source of nutrients. ' The degree to which various point sources and nonpoint sources contribute to lake eutrophication appears to depend on regional land use, stream interactions, and land, soil, and vegetation characteristics. This review provides strong evidence that the atmosphere can be a significant nutrient source. 3.2.1 Loading Estimates Estimates from studies on the Great Lakes indicate that precipitation contributions of phosphorous and nitrogen represent a significant portion of the total lake loadings of those elements. One study of Lake Michigan indicates that approximately 13 percent of the phosphorous entering 25 the lake is from precipitation. The study estimates that the contribution from precipitation will be as high as 30 percent when the program to control 80 percent of wastewater phosphorous input is completed. Phosphorous appears to be the limiting nutrient in this lake. This study did not measure dry deposition; neither did it address the problem of source identification. 3-16 ------- The Upper Great Lakes Study includes estimates of the total precipitation input of nitrogen and phosphorous to Lake Superior and Lake Huron, as summarized in Tables 8 and 9. Dry deposition estimates are not included. Accord- ing to the study, a large portion of the total phosphorous loadings on Lake Superior come from the Chicago area; on Lake Huron a significant amount emanates from Chicago, Detroit, and Cincinnati. Estimates of pollutant-loading contribu- tions from 20 U.S. cities and from 10 Canadian sources were 4 derived from a long-range dispersion model. The model loading estimates were compared to precipitation chemistry results for several elements. The authors indicate that they obtained reasonable agreement between precipitation chemistry data and predicted values, considering the ac- curacy limits of the methods. Nutrient levels in precipitation have been measured in many studies. Most appear to be short-term sampling studies, usually 1 to 2 months or less, and care should be taken not to generalize from these results. The studies do not at- tempt source identification, nor do they include dry deposi- tion sampling. Gaseous flux of nitrogen as ammonia has been shown to be significant around cattle feedlots. Studies by the U.S. Department of Agriculture show that ammonia volatilized from 3-17 ------- Table 8. NITROGEN AND PHOSPHOROUS INPUT TO LAKE SUPERIOR FROM FOUR MAJOR SOURCE CATEGORIES AS ESTIMATED IN THE UPPER GREAT LAKES STUDY5 Element Nitrogen (total as N) Phosphorous (total as P) Total Input, 103 kg/yr Direct municipal 500 132.5 Direct industrial 565 99 Tributary 36,000 2832 Atmospheric 55,845 799 Contribution from precipitation, % 60 21 U) I M 00 Table 9. NITROGEN AND PHOSPHOROUS INPUT TO LAKE HURON FROM FOUR MAJOR SOURCE CATEGORIES AS ESTIMATED IN THE UPPER GREAT LAKES STUDY5 Total Input, 10 kg/yr Element Nitrogen (total as N) Phosphorous (total as P) Direct municipal 1044 190 Direct industrial 7081 81 Tributary 58,830 4416 Atmospheric 51,830 620 Contribution from precipitation, % 44 12 ------- cattle feedlots is absorbed by nearby surface waters. A study of northeastern Colorado feedlots attempted to deter- mine the rate at which ammonia is absorbed directly from the air at various distances from feedlots under different 2 6 atmospheric conditions. The study concluded that the amount of atmospheric washout of nitrogen was not large compared to the direct absorption of ammonia by the water surface. Effects were measured as far as 2 km from the feedlot. A study in Georgia produced similar results and indicated that the nitrogen contribution to streams and soils by ammonia absorption from the atmosphere appears to 27 be greater than from precipitation. The Georgia study also indicated that fertilizer nutrients in precipitation were negligible when compared to the content of the elements in the soils. Estimates of NO flux by washout and rainout, dry X deposition, and gaseous deposition are documented in the literature; however, our survey of this literature indicates the estimates are usually global in scope and based on NO ** absorption rates and reactions derived from laboratory experiments. The survey produced no information on NO H contribution of nitrogen to specific water surfaces. 3-19 ------- 3.2.2 Source-Receptor Relationships The nitrogen cycle is one of the most familiar ele- mental cycles in nature. Oxidized or reduced forms of nitrogen undergo a variety of chemical and biochemical transformations. Reduction of nitrate or nitrite occurs when these species are absorbed by plants or animals and are converted to plant or animal tissue. When the plant or animal tissue dies and is returned to the soilr microbial oxidation converts the nitrogen back to nitrate. Nitrate added to the soil in excess quantities required by plant uptake or losses by denitrification is subject to leaching and can be carried by percolating water into the ground water system and ultimately into streams and lake systems. Precipitation scavenging of nitrogen compounds can add to the total nitrate content directly and indirectly through soil leaching. Global estimates have been made of the 28 quantities of nitrogen present in the environment. These estimates indicate that natural sources of nitrogen con- tribute more nitrogen to the soil and water than do pollu- tion sources. Studies in Europe and in the United States indicate that increased industrial or agricultural activity can contribute to increased levels of nitrogen compounds in the atmosphere. A 10-year study of the Hubbard Brooke Experimental Forest in New Hampshire shows that nitrate 3-20 ------- concentration in precipitation and input of nitric acid are 29 increasing in the area. Precipitation nitrate concentra- tion changes correlate well with increased nitrogen oxide emissions from combustion of fossil fuels. This correlation provides presumptive evidence that increased industrial activity can increase nutrient loadings in the atmosphere, with a potential for subsequently increasing the nutrient levels in bodies of water. The literature indicates, how- ever, that the evidence is only presumptive, and atmospheric reactions may be contributing to the changes in nitrate concentrations. These atmospheric reactions, which result in acid precipitation, require further research. Very high correlations have been shown between the total input of sulfate and nitrate concentrations into the Hubbard Brooke Experimental Forest and the annual input of sulfate and nitrate ions from precipitation over the 10-year period from 1964 to 1974, indicating the importance of the atmosphere as a source of these ions. A study for the U.S. Department of the Interior pro- vided estimates of probable sources of atmospheric phos- phorous in the Chicago area. The estimates were derived by determining elements that could be used as tracers for specific industries emitting particulates and the atmos- pheric concentrations of those elements. An estimate of relative contributions of each source category was made from estimated source contributions to total particulate concen- 3-21 ------- trations, precipitation sample analyses, and estimates of percent phosphorous for particulate emissions from each industry. This study concluded that the atmosphere contrib- utes from one-fifth to one-third of the phosphorous going into Lake Michigan and that 60 percent of the atmospheric phosphorous input is from anthropogenic sources. The major source categories considered were: (1) phosphorous indus- try, including phosphate rock mining, phosphoric acid manu- facturing, and fertilizer manufacturing; (2) iron and steel manufacturing; (3) cement manufacturing; (4) refuse incinera- tion; and (5) fuel oil and coal combustion. Agricultural burning and forest fires were not considered. Although problems exist with the source identification technique, this study is the only one found in this survey, except the a feelot studies, that attempted to identify atmospheric A nutrient sources. Around croplands, runoff is considered to be a more significant source of nutrient loading than precipita- tion. ' ' No references to the windblown agricultural dust potential for nutrient loading of water surfaces were found in this survey. Loading factors have been determined for assessing water pollution potential from nonpoint 32 sources. The loading factor for phosphorous does not include a term for either precipitation or windblown dust. The function for nitrogen includes a term for precipitation 3-22 ------- but not for windblown dust. Although fugitive dust from agricultural tilling operations has been sampled, sufficient work has not been done to characterize either the particle size or elemental distribution of agricultural fugitive dust. Water-surface nutrient loadings from agricultural dust could be significant in high-agricultural areas, and the potential loading warrants investigation. 3.3 TOXIC ORGANICS Toxic organic compounds have been measured both in air and in water. The compounds most often cited in the liter- ature are polychlorinated biphenyls (PCB's) and pesticides, specifically DDT and related compounds. The U.S. EPA Office of Pesticide Programs, in cooperation with the U.S. Geologic Survey, compiles data on surface water samples four times annually and on sediment samples twice annually from 150 sampling points nationwide.3 The pesticides and herbicides of interest in this program are primarily water-insoluble compounds; consequently, significant amounts of pesticides have not been found. The Office of Pesticide Programs is currently conducting an air- monitoring program for pesti- cides. Although water-soluble pesticides in urban aerosols have been measured, it is not clear whether these are in the vapor phase or if they are adsorbed on particulates. Personal conversation with Mr. Fred Kutz, U.S. Environmental Protection Agency, Office of Pesticide Programs, Washington, D.C. 3-23 ------- Extensive surface organic films or microlayers have been observed in freshwater lake systems, but the rela- tionships between organic compounds in air and the nature of these organic microlayers have not been established. Many pesticides have a high affinity for soil parti- cles. Agricultural runoff, then, is an important source of pesticide pollution of surface water. The deposition of windblown particles on water surfaces has apparently not been studied in relation to pesticide loading levels. The potential of long-range transport of pesticide aerosols, however, is illustrated by Table 1.0. Solubility, vola- tilization, leaching, and surface runoff properties of 32 leading pesticides have been documented. Table 10. DRIFT VS. PARTICLE SIZE PESTICIDES 35 Particle type Aircraft spray Coarse Medium Fine Air carrier sprays Fine sprays/dusts Usual dusts/aerosols Aerosols Diameter, ym 400 150 100 50 20 10 2 Drift, a meters 2.6 6.7 15 54 338 1352 33,795 Distance particle would be carried by a 4.8 km/hr wind while falling 3 meters. 3-24 ------- PCB deposition into Lake Michigan is a current research 25 topic. PCB's have been detected in Lake Michigan fish, and they are a threat to selected fish species in the lake. Sampling is being conducted to determine the amounts of PCB's present in precipitation and in the air. Current research is hindered by sampling problems and by the many possible species of PCB's. Preliminary estimates indicate that 725 kg/yr of PBJC's are input to Lake Michigan from aqueous point sources and 645 kg/yr from precipitation. A study was conducted from 1972 to 1974 to determine sources and input rates of PCB's into the Southern California Bight. The sources of PCB's investigated were municipal wastewater, surface runoff, aerial fallout, industrial discharges, antifouling paint, harbor flushing, and ocean currents. The major sources proved to be municipal waste- water and aerial fallout. Table 11 summarizes the results of the study. The estimates of aerial fallout were derived from sampling data collected from a network of glass sam- pling plates sprayed with a mixture of 5:1 hexane and min- eral oil and exposed for approximately 1 week. Five sam- pling stations were used for three 1-week sampling intervals, with an estimated collection efficiency of 50 percent. The significance of this survey is that it points to the atmos- phere as a primary source of PCB input to water systems. 3-25 ------- Table 11. ESTIMATED INPUTS TO THE SOUTHERN CALIFORNIA BIGHT36 Route Municipal wastewater Surface runoff Aerial fallout Industrial discharges Antifouling paint Harbor flushing Ocean currents Year 1974 1972-73 1973-74 1973-74 1973 1974 1973 PCB ' s , kg/yr 1242* 4300 £550 - £180 <1 - - 1254a 1100 250 1800 60 <1 <150 4000 Total 5400 £800 - <250 <1 • - PCB mixtures with gas chromatographic profiles most closely resembling those of Aroclor 1242 and Aroclor 1254. The Office of Water Planning and Standards of the U.S. EPA is currently conducting research to determine safe levels of toxic substances in water. They are also trying to identify areas in which transfer of toxic substances between air and water appears to be significant. They will try to accomplish this by integrating historical data from air, precipitation, and surface water with sediment analyses. Part of this total effort includes the development of reli- able sampling methods for toxic organic substances. Personal conversation with Mr. Rober Greenspu|A> Water Qual- ity Analysis Branch, Monitoring and Support Division, Office of Water Planning and Standards, U.S. EPA, Washington, D.C. 3-26 ------- The two-year study of the upper Great Lakes included the study of organic compounds. Organics entering the Great Lakes can be classified as follows: 0 Compounds that readily degrade biologically or chemically reducing the available oxygen levels. 0 Compounds that bioconcentrate and may be directly toxic to aquatic life or metabolized to a more toxic form in higher organisms, e.g., PCB's and DDT. 0 Compounds that cause taste and odor problems in commercially important aquatic species. Studies of Lake Huron and Lake Superior fish revealed that fish from both lakes contained detectable concentrations of PCB's. Samples of rainbow trout from Lake Huron and a subspecies of lake trout from Lake Superior frequently show levels of PCB's in excess of Canadian or U.S. food guide- lines. DDT residues are still found in fish in Lake Supe- rior despite past efforts to control DDT contamination, and available data provide no evidence that DDT levels have declined in Lake Superior fish during the past five to six years. An 80 percent reduction of DDT residues in fish in Lake Michigan has been noted since the use of DDT was banned in 1969-70 in the states bordering the lake. DDT was banned in Canada in 1970 and its ban was extended nationwide in the United States in 1972. 3.4 AIRBORNE ACIDS Acid precipitation has been studied extensively in both the United States and Europe for several years, and its effects on ecosystems is documented in the European litera- 3-27 ------- ture. The atmospheric chemistry of acid formation is also 37 documented. A recent review by a leading researcher indi- cates that strong acids such as sulfuric, nitric, and hydrochloric dominate the acid precipitation chemistry in most of northeastern United States and northern Europe, and that anthropogenic sources are the major causes of acid precipitation in these regions. Results of acid precipita- tion studies in Norway show a relationship between atmos- pheric concentration of sulfate ions in precipitation and lake water acidity. Consumption of fossil fuels in Europe, increased acidity of precipitation in southern Norway, and an increase in the number of barren lakes there also seem to show a correlation. The pH of precipitation in much of northeastern United States is reported to average between 4.0 and 4.2 annually. Values between 2.1 and 3.0 have been reported during storms in some regions. Significantly more literature is available dealing with acid precipitation and its effects than is available for the other pollutant categories discussed in this report. Pro- ceedings of the First International Symposium on Acid Pre- 38 cipitation and the Forest Ecosystem documents a large volume of work on this subject. Most of the information reviewed and summarized herein is qualitative rather than quantitative. The following sections summarize the kinds of results that have been reported. 3-28 ------- 3.4.1 Acid Precipitation Effects on Lake Chemistry A large volume of acid precipitation work has been done around the Sudbury area in Canada, at the Hubbard Brooke Experimental Forest in New Hampshire, and in Norway. Water quality data for 150 lakes within a 100-km radius of Sudbury 39 were collected during 1974 and 1975. Surface water pH varied between 4.0 and 8.5. Lakes in the lowest pH zone in the study area have low buffering capacity and are oriented in line with the predominant wind direction that blows from the Sudbury smelting complex. High concentrations of . metals associated with the smelter emissions were found in these lakes, compared with low concentrations normally found in air, water, and soil, which would indicate that Sudbury smelting emissions significantly affect the water quality of these lakes. High S02 emissions from the smelting opera- tions are probably responsible for the pH depression in these lakes. Acidification of lakes in this area has re- sulted in the extinction of several fish species. The mechanism of toxicity has not been determined. The longest record of precipitation pH in the United States is for the Hubbard Brooke Experimental Forest, which is maintained by the U.S. Forest Service. Mean annual pH of precipitation in this area between 1964 and 1974 was re- 29 ported to be from 4.03 to 4.21. The annual input of 3-29 ------- hydrogen ion increased 36 percent in the same period. Pre- cipitation chemistry data indicate that precipitation in the eastern U.S. was already acid by the 1950's, and more recent data show a southwestward and westward extension of the re- gion of acid precipitation since the mid-1950's. Comprehen- sive precipitation chemistry data necessary to characterize the entire nation are not available. The contribution of sulfate to the ambient acidity of precipitation in this area is reported to have decreased from 83 percent in 1964-65 to .66 percent in 1973-74; the contribution of nitrate increased from 15 percent to 30 percent. Reasons for the increased nitrate concentration have not been firmly established, but the increased nitrate concentration parallels increases in amounts of fossil fuels burned in the area. Contributions from nitrogen fertilizers have not been determined. Precipitation chemistry has been extensively monitored 40 throughout Europe for the last two decades. At most stations the pH trend is downward. During the same period nitrate in precipitation shows a pronounced increase. Although the sulfur content of precipitation has increased steadily while the pH has decreased, the correlation coeffi- cient calculated between these data trends is not high. Chemical or meteorological interferences may influence short-term data irregularities. Extensive water quality testing shows a decrease in pH of surface waters in Scandanavian countries over several 3-30 ------- years. Low soil buffering capacity and increased acidity of precipitation are shown for the same regions. 3.4.2 Acid Precipitation Effects on the Aquatic Ecosystem In the United States, Canada, and Europe studies have been performed on entire lake systems to characterize the effects of acid precipitation on aquatic life. A 1974 survey of 155 lakes in Norway shows that unpolluted lakes in granitic basins in central Norway have pH values above 6.0, and bicarbonate is the major anion. On the other hand, lakes in large areas of western, southern, and southeastern Norway have pH values below 5.0, and sulfate has replaced bicarbonate as the major anion. This would seem to imply that the sulfate is derived from sulfuric acid in precipita- tion since the lakes in these regions have minimal local sources of pollution. The conclusions drawn from the study of these lakes are summarized as follows: 0 Bacterial decomposition is retarded at low pH levels. Decomposition of cellulose is reduced 50 percent when the pH is lowered from 7.0 to 5.2. Inhibition of decomposition results in an in- creased accumulation of organic debris and a de- creased rate of nutrient cycling. 0 Fewer species of phytoplankton and zooplankton are observed in water with a pH around 6.0 or less. This is true in both Norway and Sweden. 0 Some aquatic vertebrates that are important as fish food are inhibited or absent in low-pH water. Data on 79 rivers throughout Norway indicate that the salmon population in nine of the rivers declined to zero between 1885 and 1968. These nine rivers are highly acidic, 3-31 ------- with a pH range of 4.5 to 5.5. Of 2083 lakes studied in Sorlandet, 741 have no fish, and 516 in which the pH is around 5.5 or lower, either have no fish at all or their numbers have been greatly reduced. The disappearance of fish in most lakes occurs over a period of many years. In some instances, however, a shock effect from a sudden de- crease of pH can result in massive fish kills. This effect is related to the melting of snow and ice accumulating from winter acid precipitation. 41 In the Adirondack Mountains in the United States, fish populations have been adversely affected by acidifica- tion in approximately 75 percent of the high-elevation lakes. The effects range from extinction to alterations in density, size, structure, and growth rates of fish popula- tions. The lakes in the LaCloche Mountains near Sudbury, 14 Ontario, have also been studied. These lakes are believed to be significantly impacted by air pollution from the Sudbury smelting complexes. Sulfate in these lakes repre- sents approximately 90 percent of the anions compared with approximately 40 percent in the Experimental Lakes Area in northwestern Ontario. A survey of 150 lakes in the LaCloche Mountain area shows that 33 lakes had a pH of 4.5 or less, and one-half of the lakes had a pH of 5.5 or lower. Differ- ent species of fish apparently respond to different levels 3-32 ------- of acidification. Reproduction appears to be the most sensitive physiological process affected in fish popula- tions. According to this study, the extinction of fish did not appear to be an indirect effect of an earlier loss of food items. Major prey species remained abundant, while some predator species became extinct. It is possible that 'A acid stress may have caused reduced feeding or fooj£ utiliza- tion or both in some species. It was «e* noted that normal calcium metabolism may be affected by acid stress. 3-33 ------- 4.0 CONCLUSIONS AND RECOMMENDATIONS 4.1 CONCLUSIONS A significant amount of research on air/water inter- actions is documented in the literature reviewed. More work has been done in Europe, especially in the Scandinavian countries, than in the United States and Canada. A large proportion of the Scandinavian work deals with acid precipi- tation. The Scandinavian studies conclude that a signifi-* cant volume of pollutant input into lakes and watersheds is transported through the atmosphere and probably originates from anthropogenic sources. Studies in the United States and Canada support these conclusions, although few long-term, i.e. one year or longer, studies have been conducted in either country. Studies of the Great Lakes, the Walker Branch Watershed in Oak Ridge, Tennessee, and the Hubbard Brooke Experimental Forest in New Hampshire comprise the most significant work reviewed. In making this survey, PEDCo found more estimates of atmospheric input of trace metals and nutrients into lakes and watersheds than of airborne acids and toxic organic materials. The sampling and analytical techniques for particulates are better developed than are those for air- 4-1 ------- borne acids and vapor-phase organic materials. Vapor-phase organic pollutants present sampling and analysis problems, primarily because they are usually present in very low concentrations in large volumes of air, because they have a reactive nature, and because they are susceptible to adsorp- tion. Most estimates of source contributions to total atmospheric input of pollutants to lakes and watersheds result from the application of emissions estimates and meteorological data to atmospheric transport models. These source contributions, consequently, are first-order esti- mates. The best estimates of source contributions appear to be those made for the Walker Branch Watershed. Estimates of atmospheric loading onto water surfaces are usually based on precipitation studies. An estimate of dryfall input is usually not included. Extensive sampling for dryfall has been hampered by sampling methodology prob- lems. Precipitation scavenging effects vary by pollutant, and it is reasonable to believe that dryfall effects also vary considerably by pollutant. Effects of atmospheric pollutants on biota appear to have been studied more extensively in Europe than in the United States and Canada. Pollutant effects on water quality and on the biota seem to depend partly on the buffering capacity of the water. The most significant effects of acid 4-2 ------- precipitation on water quality and biota are observed in poorly buffered lakes. The mechanism of toxicity of biota in lower pH waters is not understood. It may involve metal toxicity that manifests itself below a specified pH. Our evaluation of all factors that have been reported and discussed indicates that expanded research is required to accomplish the following tasks:a' 'c 0 Identify sources and relative source contribu- tions. 0 Sample dry deposition of pollutants onto water surfaces. 0 Expand the data base. 0 Determine cross-media impacts. 0 Define the effects of altered water quality for various pollutants. The following sections expand upon these general research requirements and cite current related projects. 4.1.1 Source Identification Current estimates of source contributions of atmos- pheric pollutants to water quality are based on emission a Personal conversation with Dr. Martin Bratzell, Interna- tional Joint Commission, Windsor, Ontario. Personal conversation with Mr. Floyd Elder, Canada Centre for Inland Waters, Burlington, Ontario. c Personal communication with Dr. Michael Mullin, U.S. EPA Grosse lie Laboratory, Gross lie, Michigan. 4-3 ------- inventory techniques. Some work has been done to character- 42 43 44 ize urban aerosols. ' ' Most of these studies, however, concentrate on identifying trace elements as a function of total particulate loading. Source identification adequate for predictive modeling of trace element transport and depo- sition onto lake surfaces requires expanded knowledge of emissions from specific source types, the elemental compo- sition of those emissions, and the distribution of elements as a function of time and particle size. Industrial sources 45 46 47 of trace metals have been documented, ' ' and major efforts to characterize trace element emissions and their fate in the atmosphere are under way. The National Science Foundation Research Applied to National Needs (NSF-RANN) program has funded research related to trace contaminant 48 emissions and atmospheric transport. Current research on the emission and fate of trace elements includes studies of power plant emissions. ' Fossil-fuel-fired power plants are a major source of trace elements in the atmosphere, and the characterization and atmospheric distribution of emissions will become increas- ingly important as coal becomes a more important fuel. The techniques currently used to study urban aerosols include analyses of Hi-vol samples for multiple elements. Mass-balance approaches, using calculated enrichment factors for various source categories and paired correlation co- 4-4 ------- efficients in conjunction with reference elements, have provided the most data in the United States. Expanded sampling and analyses are needed for further characteriza- tion of these aerosols in terms of particle size distribu- tion. 4.1.2 Dry Deposition Estimates Most information on the process of dry deposition of pollutants onto water surfaces comes from laboratory studies. Consequently, deposition functions used in models also are based on laboratory studies. Since it is difficult to simulate actual lake conditions in the laboratory, these model functions may not be accurate. Increased field mea- surements of dry deposition over lakes is required for validation of the current models. Sampling experiments should consider fluxes in micrometeorological conditions and pollutant concentrations. Based on current knowledge of the mechanisms by which pollutants are transferred from air to water, large-scale efforts to develop models appear to be premature. Improved sampling and analysis methods are required. Limitations of current dryfall sampling techniques have been a major problem in estimating input to lake surfaces by dry deposition. Hi-vol sampling and particle sizing have not been used extensively in lake studies. Average aerosol 4-5 ------- composition with respect to time and particle size, however, is needed to estimate rates of dry deposition. Proton- induced X-ray emission (PIXE) analysis has been suggested as an analytical technique capable of providing the broad data base needed to develop and validate predictive models. Direct sampling from shipboard or platforms on the Great Lakes has been suggested. Parallel sampling programs at various locations throughout the Northern Hemisphere have also been suggested. 4.1.3 Expand the Data Base Expanded data bases are required for precipitation chemistry as well as for estimating the effects of dry fallout. The current expansion of the precipitation sam- pling network around Lake Superior and Lake Huron to cover the remaining Great Lakes should significantly improve the current data base. Dryfall samplers are also being used in this network, but, as stated before, the dry sampling tech- niques need improvement. Permanent sampling networks around other lakes in the United States and Canada would be useful. 4.1.4 Determine Cross-Media Impacts The direct impact of air pollutants on water quality appears to be significant. The enrichment of heavy metals on the smaller particulates that escape collection by con- ventional control devices on fossil-fuel-fired boilers is an 4-6 ------- example of the type of problem that should be addressed. The impact of pollutant fallout, both dry fallout and precipitation scavenging, on land-runoff loading of water bodies is another example of a problem that has not been assessed. Indirect impacts from air pollutants include the impact of combustion and air-pollution-control residuals. The soluble portions of ash and scrubber sludge represent a potential problem for both ground water and surface water quality. The U.S. EPA recognizes the potential for intermedia effects of control technology for air pollution. Recent studies have been conducted to define these effects and to 52 53 develop a methodology for evaluating cross-media impacts. ' A preliminary effort to develop a methodology for evaluating cross-media impacts resulted in an environmental degradation index for each possible combination of pollution-control measures by plant or industry. This was a joint project of the Council on Environmental Quality and the U.S. EPA. Example applications to kraft pulp mills and coal-fired power plants have been documented. 4.1.5 Define Effects Defining the problem of fallout of air pollutants on water surfaces requires more than measuring loading impacts; 4-7 ------- it also requires long-term assessment of the effects. For instance, the effects of air pollutants on aquatic life have not been studied extensively in the United States. One proposed approach is to study the effects on the entire ecosystem rather than just those on water. This approach may be more meaningful in defining cross-media impacts than fragmented research on separate segments of the ecosystem would be. Studies of watersheds presently tend to concen- trate on runoff and point-source impacts and to ignore the impact of atmospheric input. Inclusion of atmospheric considerations in these watershed projects would provide an significant extension of current knowledge. 4.2 RECOMMENDATIONS The status and nature of air/water research are sum- marized as follows: 0 Too little emphasis is placed on studying the effects of air pollutants on the ecosystem. 0 The lack of a large-scale monitoring network for precipitation chemistry studies hinders trends assessments and studies of cause/effect rela- tionships. 0 Too many small-scale projects are being funded by agencies that do not coordinate with other agen- cies that are funding the same type of research. The result is significant duplication of effort. A central clearing house to coordinate the re- search is needed. 0 Development of models is premature, and emphasis should be shifted to defining and solving basic pollutant transport and interaction problems. 4-8 ------- The International Joint Commission has categorized 54 current research needs as critical, essential, or necessary. These categories are used in the following list. Critical 0 Development and design criteria for an effective and efficient monitoring system for ecological needs 0 Optimization of design and operation of monitoring programs 0 Determination of sources of chlorinated organics 0 Determination of the behavior and fate of chlo- rinated hydrocarbons 0 Development of an analytical and sampling method- ology for toxic elements 0 Identification of source inputs of toxic elements 0 Determination of methods and rates of transport and dispersal of toxic elements 0 Identification of the fate of toxic elements in the water environment 0 Determination of the dynamics of PCB's within the Great Lakes ecosystems 0 Determination of contaminant budgets in the Great Lakes: sources, fluxes, transformations, and sinks 0 Determination of air/water interface actions. Essential 0 Determination of inshore-offshore energy and materials transport 0 Identification and monitoring of potentially dangerous organic compounds 4-9 ------- 0 Determination of the dynamics of potentially dangerous organics in the Great Lakes Necessary 0 Identification of sources of halogenated pesti- cides to the Great Lakes 0 Investigation of the environmental dynamics of halogenated pesticides. a ^r\ ft Conversations with leading researchers ' ' working on the subject of air pollutant effects on water quality con- firm our conclusions that the major problems that should receive research priority are: ° Expand monitoring programs to build a national trends data base. 0 Refine the sampling methodology to include reason- ably accurate estimates of dryfall input. 0 Develop a reliable method for identifying, pollu- tant sources. 0 Determine the relationship between terrestrial fallout and runoff contributions to stream and lake loadings. 0 Determine acceptable levels for hazardous pollu- tants and determine the effects of those pollu- tants. A recent analysis of the intermedia problem, recom- mended that in order to determine the degree of significance a Personal conversation with Mr. Floyd Elder, Canada Center for Inland Waters, Burlington, Ontario. Personal conversation with Dr. Michael Mullin, U.S. EPA Grosse lie Laboratory, Grosse lie, Michigan. c Personal conversation with Dr. Gene E. Likens, Cornell University. "Discussion of Nonpoint Source Water Pollution Interface With EPA's Air Programs, January 24, 1977." Prepared by Wallace E. Reed, Associate Professor, Department of En- vironmental Sciences, University of Virginia. 4-10 ------- of air pollution to nonpoint source (NFS) water pollution problems, EPA should: 0 Assess the location and magnitude of NFS problems by pollutant type through the "208" program. 0 Estimate the air contribution to each problem area. 0 Estimate the degree and cost of air pollution control needed locally and regionally. 0 Estimate the value of improvement in health and water use effects that will result from better control of air pollution sources. This approach would require continuation of existing EPA intermedia research as well as interagency coordination of research. It would also identify the most critical pollu- tant research needs. 4.3 INTRA-AGENCY CONSIDERATIONS At the present time, two interrelated programs are in various stages of implementation within the U.S. Environ- mental Protection Agency. These are the Air Quality Main- tenance Planning/State Implementation Plan Revision pro- cesses and the Areawide Water Quality Management Plan/"208" activities. There is a close analogy between EPA's Air Program/SIP Process and Water Program/"208" Planning. The entire nation is subdivided into "208" planning areas, but more emphasis is being placed in areas of major problems (i.e., large urban areas). State agencies are responsible for the development and submission of "208" plans on or 4-11 ------- before November 1, 1978, much as they were required to submit SIP's in 1972. One area of major difference in the two programs is that under the SIP process states were given specified regulatory requirements (i.e., Part 51 regulations) to follow for meeting NAAQS; whereas, under the "208" Plan- ning Process States are given more flexibility for develop- ing these plans through the use of nonregulatory guidelines to control sources to the extent feasible. Recognizing the close relationship and common object- ives of the two programs, the Assistant Administrator for Water and Hazardous Materials joined with the Assistant Administrator for Air and Water Management to produce "Pro- gram Guidance Memorandum: SAM-8," dated November 15, 1976, for distribution to all EPA Regional Administrators. The purpose of this memo was as follows: (1) to transmit revised guidance on Environmental Protection Agency policy on the relationship between air quality planning and the state and areawide water quality management (WQM) program, and (2) to identify air-quality-related activities eli- gible for funding under section 208. Relative to the second item, the memo specified various activities that are illustrative of eligible uses for section 208 funds. Keeping in mind that any ancillary data gathering or development project supported by "208" funding 4-12 ------- must be directly related to the objectives of the WQM pro- gram, the following activity categories appear to be eli- gible for "208" support and of real interest to OAQPS/SASD relative to the subject areas previously discussed: 0 Development of common data bases. 0 Development and implementation of common public participation program. 0 Development of common institutional mechanisms for plan adoption, implementation, and revision. 0 Development of control strategies or measures that will achieve the objectives of both air quality and WQM programs. 0 Development of air quality assessments of current or projected development to be served by waste- water treatment facilities. 0 Development of strategies for mitigating any adverse air quality effects from WQM plans. 4-13 ------- 4.0 REFERENCES 1. Winchester, J.W., and G.D. Nifong. Water Pollution in Lake Michigan by Trace Elements from Aerosol Fallout. Water and Soil Pollution. 1:50-64. 1971. 2. Gatz, D.F. Estimates of Wet and Dry Deposition of Chicago and Northwest Indiana Aerosols into Southern Lake Michigan. Proceedings of the Second Federal Conference on the Great Lakes. March 25-27, 1975. 3. Barnes, R.S., and W.R. Schell. Physical Transport of Trace Metals in the Lake Washington Watershed. In: Cycling and Control of Metals, Proceedings of an Environmental Resources Conference in Columbus, Ohio. October 31 - November 2, 1972. 4. Atmospheric Loading of the Upper Great Lakes: Volume I Summary Report. Acres Consulting Services Limited, Applied Earth Sciences Consultants, Inc. Canada Center for Inland Waters. December 1975. 5. The Waters of Lake Huron and Lake Superior: Volume T Summary and Recommendations. Report to the Interna- tional Joint Commission by the Upper Lakes Reference Group. Windsor, Ontario. 1976. 6. Hess, G.D., and B.C. Hicks. The Influence of Surface Effects on Pollutant Deposition Rates Over the Great Lakes. Proceedings of the Second Federal Conference on the Great Lakes. March 25-27, 1975. 7. Andrew, A.W., S.E. Lindberg, and L.C. Bates. Atmos- pheric Input and Geochemical Cycling of Selected Trace Elements in Walker Branch Watershed. Oak Ridge Nation- al Laboratory, Tennessee. June 1975. 8. Ensor, D.S., T.A. CahilL, and L.E. Sparks. Elemental Analysis of Fly Ash from Combustion of a Low Sulfur Coal. Paper 175-33.7 presented at the 68th Annual Meeting of the Air Pollution Control Association. Boston, Massachusetts. June 15-20, 1975. 4-14 ------- 9. Klein, D.H., et al. Trace Element Measurements at the Coal-Fired Allen Steam Plant — Mass Balance and Con- centrations in Fly Ash. Paper presented at a Meeting of the American Chemical Society. Philadelphia, Penn- sylvania. April 6-11, 1975. 10. Kaakinen, J.W., R.M. Jorden, M.H. Lawasani, and R.E. West. Trace Element Behavior in A Coal-Fired Power Plant. Paper presented at a Meeting of the American Chemical Society. Philadelphia, Pennsylvania. April 6-11, 1975. 11. Natusch, D.S.F. Physico-chemical Associations of Trace Contaminants in Coal Fly Ash. Paper presented at a meeting of the American Chemical Society. Philadel- phia, Pennsylvania. April 6-11, 1975. 12. Zoller W.H., E.S. Gladney, and G.E. Gordon. Elemental Fractionation in the Chalk Point Power Plant. Paper presented at a meeting of the American Chemical Society. Philadelphia, Pennsylvania. April 6-11, 1975. 13. Coal-Fired Power Plant Trace Element Study. Volume I. A Three Station Comparison. Radian, Inc. U.S. En- vironmental Protection Agency, Denver, Colorado, NTIS Number PB-257293. September 1975. 14. Beamish, Richard J. Acidification of Lakes in Canada by Acid Precipitation and the Resulting Effects on Fishes. Proceedings of the First International Sym- posium on Acid Precipitation and the Forest Ecosystem. Columbus, Ohio. May 12-15, 1975. 15. Wright, Richard F., T. Dale, E.T. Gjessing, G.R. Hendrey, A. Henriksen, M. Johannessen, and I.P. Muniz. Impact of Acid Precipitation on Freshwater Ecosystems in Norway. Proceedings of the First International Sym- posium on Acid Precipitation and the Forest Ecosystem. Columbus, Ohio. May 12-15, 1975. 16. Gorham, E. Acid Precipitation and Its Influence on Aquatic Ecosystems - An Overview. Proceedings of the First International Symposium on Acid Precipitation and the Forest Ecosystem. Columbus, Ohio. May 12-15, 1975. 17. Winona Lake, Kosciusko County, Indiana. National Eutrophication Survey, U.S. Environmental Protection Agency, Corvallis, Oregon. NTIS Number PB-256 577. Working Paper No. 348. April 1976. 4-15 ------- 18. Hopatcong and Musconetcong Lakes, Morris and Sussex Counties, New Jersey. Eutrophication Survey, U.S. Environmental Protection Agency, Corvallis, Oregon. NTIS Number PB-256 601. May 1976. 19. Long Lake, Steuben County, Indiana. National Eutro- phication Survey, U.S. Environmental Protection Agency, Corvallis, Oregon. NTIS Number PB-256 561. May 1976. 20. Pompston Lakes, New Jersey. National Eutrophication Survey, U.S. Environmental Protection Agency, Las Vegas, Nevada. NTIS Number PB-256 544. Working Paper No. 373. May 1976. 21. Sylvan Lake, Noble County, Indiana. National Eutro- phication Survey, U.S. Environmental Protection Agency, Corvallis, Oregon. NTIS Number PB-256 559. Working Paper No. 341. April 1976. 22. Hovey Lake, Posey County, Indiana. National Eutrophi- cation Survey, U.S. Environmental Protection Agency, Corvallis, Oregon. NTIS Number PB-256 552. Working Paper No. 329. March 1976. 23. Correll, D.L. The Rhode River Program. Symposium on Pollution Problems in Estuaries. Pensacola, Florida. February, 1975. 24. Murphy, T.J., and P.V. Doskey. Inputs of Phosphorous from Precipitation to Lake Michigan. U.S. Environ- mental Protection Agency, Washington, D.C. EPA-600/3- 75-005. December 1975. 25. Murphy, T.J. Precipitation: A Significant Source of Phosphorous and PCB's to' Lake Michigan. Paper pre- sented at the 10th ACS Great Lakes Regional Meeting, Evanston, Illinois. June 17, 1976. 26. Mosier, A.R., C.E. Andre, and F.G. Viets, Jr. Identi- fication of Aliphatic Amines Volatilized from Cattle Feedyards. Environmental Science and Technology. 7:642-644. July 1973. 27. Giddens, J. Contamination of Water by Air Pollutants, Especially Ammonia from Animal Manures. U.S. Depart- ment of the Interior, Office of Water Research and Technology. USDI/OWRT Project A-050-GA. 4-16 ------- 28. Robinson, E., and R.C. Robbins. Gaseous Nitrogen Com- pound Pollutants from Urban and Natural Sources. Journal of the Air Pollution Control Association, Volume 20, No. 5, 303-306. May 1970. 29. Likens, G.E., F.H. Bormann, R.S. Pierce, and N.M. Johnson. Hydrogen Ion Input to the Hubbard Brooke Experimental Forest, New Hampshire, During the Last Decade. Proceedings of the First International Sym- posium on Acid Precipitation and the Forest Ecosystem, Columbus, Ohio. May 12-15, 1975. USDA Forest Service General Technical Report NE-23. 1976. 30. Murphy, T.J. Sources of Phosphorous Inputs from the Atmosphere and Their Significance to Oligotrophic Lakes. U.S. Department of the Interior. WRC Research Report No. 92. September 1974. 31. Control of Water Pollution from Cropland, Volume 1, A Manual for Guideline Development. Agricultural Res- earch Service, U.S. Department of Agriculture, and Office of Research and Development, U.S. Environmental Protection Agency, Washington, D.C. Report No. ARS-H- 5-1. November 1975. 32. McElroy, A.D., S.Y. Chiu, J.W. Nebgen, A. Aleti, and F.W. Bennett. Loading Functions for Assessment of Water Pollution from Nonpoint Sources. Midwest Re- search Institute, Kansas City, Missouri. U.S. En- vironmental Protection Agency, Washington, D.C. EPA- 600/2-76-151. May 1976. 33. Schuman, G.E., and R.E. Burwell. Precipitation Nitro- gen Contribution Relative to Surface Runoff Discharges. Journal of Environmental Quality, Volume 3, No. 4, 366-368. 1974. 34. Andren, A.W., A.W. Elzerman, and D.E. Armstrong. Chemical and Physical Aspects of Surface Organic Micro- layers in Freshwater Lakes. Atmospheric Contribution to the Chemistry of Lake Waters. First Specialty Symposium of the International Association for Great Lakes Research. Ontario, Canada. September 28-October 1, 1975. 35. Methods and Practices for Controlling Water Pollution from Agricultural Nonpoint Sources. U.S. Environmental Protection Agency, Washington, D.C. October, 1973. EPA/430/9-73/015. 4-17 ------- 36. Young, D.R., D.J. McDermott, and T.C. Heesen. Poly- chlorinated Biphenyl Inputs to the Southern California Bight. Background paper prepared for the National Conference on Polychlorinated Biphenyls. Chicago, Illinois. November 19-21, 1975. 37. Likens, Gene E. Acid Precipitation. Chemical and Engineering News. pp. 29-44. November 22, 1976. 38. Proceedings of the First International Symposium on Acid Precipitation and the Forest Ecosystem. Columbus, Ohio. May 12-15, 1975. USDA Forest Service General Technical Report NE-23. 1976. 39. Conroy, N., K. Hawley, W. Keller, and C. Lafrance. Influences of the Atmosphere on Lakes in the Sudbury Area. Atmospheric Contribution to the Chemistry of Lake Waters. First Specialty Symposium of the Inter- national Association for Great Lakes Research. Ontario, Canada. September 28 - October 1, 1975. 40. Oden, S. The Acidity Problem — An Outline of Con- cepts. Proceedings of the First International Sym- posium on Acid Precipitation and the Forest Ecosystem. Columbus, Ohio. May 12-15, 1975. USDA Forest Service General Technical Report NE-23. 1976. 41. Schofield, Carl L. Lake Acidification in the Adiron- dack Mountains of New York: Causes and Consequences. Proceedings of the First International Symposium on Acid Precipitation and the Forest Ecosystem. Columbus, Ohio. May 12-15, 1975. 42. Neustadter, H.E., R.B. King, and J.S. Fordyce. El- emental Composition of Suspended Particulates as Func- tions of Space and Time in Cleveland, Ohio. Paper No. 75-28.1 presented at the 68th Annual Meeting of the Air Pollution Control Association. Boston, Massachusetts. June 15-20, 1975. 43. Rote, D.M., and L.E. Wangen. Analysis of Suspended Particulates in the Chicago Area by Particle Size. Paper No. 75-28.3 presented at the 68th Annual Meeting of the Air Pollution Control Association. Boston, Massachusetts. June 15-20, 1975. 44. Gladney, E.S., et al. Composition and Size Distribu- tion of Atmospheric Particulate Matter in Boston Area. Environmental Science and Technology 8(6), 551. 1974. 4-18 ------- 45. National Inventory of Sources and Emissions: Arsenic, Beryllium, Manganese, Mercury, Vanadium. W.E. Davis, Associates, Leawood, Kansas. U.S. Environmental Protection Agency. 1971. 46. National Inventory of Sources and Emissions: Barium, Boron, Copper, Selenium, Zinc. W.E. Davis, Associates, Leawood, Kansas. U.S. Environmental Protection Agency. 1971. 47. National Inventory of Sources and Emissions: Cadmium, Nickel, Asbestos. W.E. Davis, Associates, Leawood, Kansas. National Air Pollution Control Administra- tion. 1970. 48. Trace Contaminants in the Environment, Proceedings of the Second Annual NSF-RANN Trace Contaminants Con- ference. Asilomar. Pacific Grove, California. August 29-31, 1974. Lawrence Berkeley Laboratory, University of California. 49. Natusch, N.S.F. Characterization of Atmospheric Pollutants from Power Plants. In: Proceedings of the Second Federal Conference on the Great Lakes. March 25-27, 1975. 50. Winchester, J.W. Assessing Air Pollution Particulate Fallout Potential for Water Pollution in Lake Michigan. Proceedings of the Second Federal Conference on the Great Lakes. March 25-27, 1975. 51. Residual Waste Best Management Practices: A Water Planner's Guide to Land Disposal. PEDCo-Environmental Specialists, Inc., Cincinnati, Ohio. U.S. Environ- mental Protection Agency. June 1976. 52. Stone, R., and H. Smallwood. Intermedia Aspects of Air and Water Pollution Control. U.S. Environmental Pro- tection Agency, EPA-600/5-73-003. August 1973. 53. Reiquam, H. , N. Dee, and P. Choi. Development of Cross-Media Evaluation Methodology. Council on En- vironmental Quality and U.S. Environmental Protection Agency. PB-232 414. January 1974. 54. Great Lakes Water Quality Research Needs 1976. Great Lakes Research Advisory Board report to the Inter- national Joint Commission. July 1976. 4-19 ------- TECHNICAL REPORT DATA (Please read Instructions on llie reverse In-fore completing) \. REPORT 4. TITLE AND SUB TITLE The Effects of Air Pollution on Water Quality 7. AUTHOR(S) David W.Armentrout I i. PERFORMING ORGMMIZATION NAME AND ADDRESS PEDGo Environmental, Inc. ! 11499 Chester Road | Cincinnati, Ohio 45246 12. SPONSORING AGENCY NAME AND ADDRESS U.S. Environmental Protection Agency Office of Air Quality Planning & Standards Research Triangle Park, N.C. 27711 3. RECIPIENT'S ACCESS! OPr NO. 5. REPORT DATE Date of Issue: 3/15/77 6. PERFORMING ORGANIZATION CODE 8. PERFORMING ORGANIZATION REPORT NO. 10. PROGRAM ELEMENT NO. 11. CONTRACT/GRANT NO. 68-02-2515, T.O. 2 13. TYPE OF REPORT AND PERIOD COVERED Final 14. SPONSORING AGENCY CODE 15. SUPPLEMENTARY NOTES EPA Project Officer: Justice A. Manning 16. ABSTRACT This report summarizes a literature survey performed to determine the effects of air pollution on water quality. Emphasis is on direct fallout of atmospheric pollutants and on precipitation scavenging. Preliminary evidence indicates the atmosphere is a significant source of some pollutants in water. Expanded sampling, increased effects research, and improved interagency program coordination are emphasized as future requirements. . . KEY WORDS AND DOCUMENT ANALYSIS DESCRIPTORS Air Pollution Water Pollution Atmospheric Composition b.IDENTIFIERS/OPEN ENDED TERMS Precipitation Scavenging c. COSATI l-'icld/Group 13B AA IB. DISTRIBUTION STATEMENT Unlimited 19. SECURITY CLASS (This Report) Unclassified 21. NO. OF PAGES 67 p. 20. SECURITY CLASS (Thispage) Unclassified 22. PRICE EPA Form 2220-1 (9-73) 4-20 ------- |