National Academies Press: OpenBook

Acid Deposition: Atmospheric Processes in Eastern North America (1983)

Chapter: 4. Empirical Observations and Source-Receptor Relationships

« Previous: 3. Theoretical Models of Regional Air Quality
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 87
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 88
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 89
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 90
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 91
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 92
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 93
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 94
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 95
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 96
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 97
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 98
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 99
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 100
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 101
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 102
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 103
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 104
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 105
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 106
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 107
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 108
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 109
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 110
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 111
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 112
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 113
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 114
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 115
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 116
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 117
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 118
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 119
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 120
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 121
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 122
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 123
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 124
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 125
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 126
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 127
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 128
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 129
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 130
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 131
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 132
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 133
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 134
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 135
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 136
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 137
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 138
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 139
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 140
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 141
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 142
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 143
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 144
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 145
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 146
Suggested Citation:"4. Empirical Observations and Source-Receptor Relationships." National Research Council. 1983. Acid Deposition: Atmospheric Processes in Eastern North America. Washington, DC: The National Academies Press. doi: 10.17226/182.
×
Page 147

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 Empirical Observations and Sou rce- Receptor Re~ati onsh i ps The analysis of data taken in the field complements the development of theoretical models as a means for understanding both the phenomenon of acid deposition and the responses of the atmospheric system to alternative emission-control strategies. Field measurements not only reveal insights into the nature of the atmospheric processes involved in the deposition of acid-forming materials but also may hold the greater promise of providing direct, unequivocal evidence from which responses to mitigating strategies might be assessed. Data from which the spatial and temporal distributions of SO2 and NOx over large regions of North America might be derived have not been available until recently. Similarly, reliable sampling of the chemistry of precipi- tation in North America is a recent development. Even so, sampling of ambient pollutants in the atmosphere and sampling of those in precipitation generally have not been simultaneous. There are no direct measurements of regional dry deposition for gases or particles. [For a description of the monitoring systems in North America and data interpretation, see U.S./Canada work Group #2 (1982).] Although data are relatively sparse in North America, those that are available tell us much about the phenome- non. A more extensive data base exists for northern Europe, where atmospheric SO2 and sulfate in precipitation have been monitored systematically for several decades. Much of our understanding of acid deposition has come from analysis of the European experience and the transfer to and replication of those data on this continent. That being so, differences in patterns of emissions, climatic factors, ground cover, and influences of marine air between Europe and North 87

88 America require that our understanding be tested against North American data. In this chapter we review the existing data for North America and use the data to assess the extent to which the relationships between emissions and deposition can be inferred from observations. The limitations of available data, based on conventional sampling and measurement methods, are discussed first. Results from some pertinent field programs are surveyed to illustrate the importance of meteorological processes for the variability of ambient air pollutants and acid-forming components in precipita- tion. Results of statistical analyses of data on ambient concentrations of pollutants are reviewed, and data on both emissions and deposition are analyzed to infer the influence of sources on deposition. Our survey is not a comprehensive one; it focuses on the results that bear most directly on atmospheric transport and transformation. In this context, the issues addressed are whether data are sufficient to infer (1) the extent to which a given region of sources affects receptor sites in remote locations and (2) the importance of nonlinear processes in determining the relationships between the magnitudes of emissions in source regions and the quantities of acid-forming substances deposited in · . ~ receiving regions. One of the greatest difficulties in establishing relationships between sources of pollution and conditions at receptors is accounting for the influences of atmo- spheric processes on the behavior of pollutants. The atmospheric processes involved include airflow, mixing, and chemical transformations. These processes are responsible, directly or indirectly, for the distribution and rate of deposition. Attempts to discern the influences of atmospheric processes on source-receptor relationships have taken different routes, including (1) descriptive accounts of observations, (2) analysis of data segregated by airflow from source areas (trajec- tories), (3) statistical analyses of data, and (4) inference from source tracers. In this chapter we describe the approaches used in these types of analyses and the results obtained. We also analyze existing data on emissions and deposition to discern trends and the relative behavior of sulfur and nitrogen emissions in the atmosphere.

89 AEROMETRIC DATA AND THEIR LIMITATIONS Most of the historical data on ambient air concentrations describe urban conditions. A large body of monitoring data exists in the National Aerometric Data Base (NADB), but few of these observations have been analyzed or interpreted. One of the more reliable and complete data sets available that describes regional air quality in the eastern United States was taken during a single year, 1977-1978: the Sulfate Regional Experiment (SURE) (Mueller and Hidy 1983). Unfortunately, few precipita- tion chemistry data were collected during the period of the study. In contrast to other data sets, those obtained from the SURE experiment have been analyzed in detail. Complementary data are or will be available from the western United States for 1980-1982 as a result of several studies (Allard et al. 1981, Pitchford et al. 1981). Observations of precipitation chemistry have bee n made periodically in the mid-1950s, from 1959 to 1966, and from 1972 to date in the programs described, for example, by Wisniewski and Kinsman (1982). Air-quality and precipitation data have been collected in Canada since 1974 in the Canadian (CANSAP) monitoring program. Precipitation data provide a direct measure of we t deposition. Since there are no direct measurements of dry deposition, regional patterns are estimated from ambient concentrations and deposition velocities (Appendix C). Since deposition velocities depend on the airflow near the surface, surface properties, and cover, these calculations are believed to be uncertain for quantitative evaluation. The quality of the data describing regional-scale processes is variable. The precision and accuracy of the measurements are generally not well defined in work reported prior to the late 1970s. Recent programs, however, have made considerable progress in reporting data with supplemental information on the calibration o f instruments, as well A: l~h" "rr~rc: ^^n. :~;r`=A ;r' she observations . A definitive discussion of aerometric data is included in Mueller and Hidy (1983). The quality of recent data on wet deposition is less formally documented, but extensive information on the comparability of data is available from the Illinois Water Survey Laboratories, the Department of Energy's Environmental Measurement Laboratories in New York, and the Environmental Protec- tion Agency's Environmental Monitoring Systems Laboratory. A statistical analysis of data from two independent

go precipitation sampling networks (SURE and MAP3S) covering a similar region in the eastern United States indicated that there is good agreement between data on the concentrations of the major ions, H+, sulfate, nitrate, and ammonium, between the two networks (Pack 1980). Documentation of the quality of data is essential to ensure that comparisons will be possible with data from future studies. Experiments and monitoring programs should incorporate such efforts to avoid or at least minimize controversies about data interpretation. The integrity of the sampling and analytical methods employed is another important consideration in measurement quality. Despite years of effort in the development of methods, the instrumentation used in past and current programs is subject to serious interferences. Uncertainties in results can be large. Interferences or ambiguities in the methods are particularly serious under rural or remote conditions, in which concentrations of pollutants or deposition levels are low. Listed in Table 4.1 are examples of sampling or mea- surement uncertainties that exist in currently available methods of sampling and analysis. From the table it is evident that few of the important chemically related measurements can be made without ambiguity because of uncertainties in the methods. RELATIONSHIPS AMONG AEROMETRIC PARAMETERS Atmospheric measurements provide a basic description of spatial and temporal distributions of pollutants. On a regional scale, the distribution of SO2 shows strong gradients near sources, but airborne sulfate shows relatively weak gradients. Concentrations of both SO2 and SOi are elevated over the industrialized or urbanized parts of the eastern United States (Figures 4.1 and 4.2). High concentrations of sulfur oxides are found through the Ohio River Valley into Pennsylvania and New York to the Atlantic Coast. From this region of high concentration, there is a strong gradient in ambient SO2 concentration northeastward toward the Adirondacks, Vermont, and New Hampshire. The distribution of nitrogen oxide concentrations is generally similar to that of sulfur oxides, but it may show more localized maxima near urban areas (e.g., Mueller and Hidy 1983). Paralleling the distributions of ambient concentrations are the distributions of sulfate and nitrate in precipi

91 c~ o - - ;^ c c 04 ~: - en c~ - ct o . - ~: :> c ~: .- c~ - c ·ct c) c :~ - m ~: E~ _ E U I, ~e y ~ ' ~ i ~ I 3 1 ' ~o ~C ,= ~ o C ,_ f~, Y `,: ~ y e ~ 0 's ~C 0 O ~O ~- ~ ~e ~0 3 ~C C o .o ~3 ~ ~ · ~ ~ ~ c 3 ~ 3 ~ ~ P~ ~ ~ ·= ~ E c o`> to o ' ·° E ~ C E o c 3, ~Z~= ._ c o E o ~ o _ Ut o eD c C o~ C Co _._ ~_ o~ ·_~C ~4.) X3 C)~, ~o _._ e~_ 3~ _ o _~ - C)~ e~ ~3 c D~ ;>, ~ O ·- C~ C ~.- o os ~ - C _ ~ ~ o ~ ~ en ·- E ~ D D ~ r ~ ~ Z _ 0,~ O C . _ Ct ._ _ ._ ~ .- ~ ~_ ·- 4,) <: 3 o ~ . _ _ ~c, ~ 3 ·° E ~°° ~ .- · ~e ~e E v D ~E es 4' ~ * U, -c ~o ', ~E E ~ ° y U ~ e ~ y ~ ~ ~ y c~ - C`3 ._ ca et - ._ ,~ ~o ~ ~0 O O - D ° ~-=v E E E 3 _ ~ c ~ E! _ _ ~ ct 3 3 c c ._ ._ ~ CJ u' Z 3 c ; - 6 4D en c~ c, - c.> ~ 04 c~ 0 0 0 0 ._ ._ _ _ u~ C~3 c, u, c o - -

92 ~ ~_ J2,~6,\) I\ . ^? 2(6) ' 7! /~( 20 (601~' 1 ~ ~ , ) ~ ~~~4 ) Off \ ~ _ _ fin 1 _J ~10 ('30) , - _ lo/ 10 (30) ~ ~\ N ~ _! j \ 'N 1 \ 1 ~l 4_ ~ _ _-~ 1-Hour SO2 (ppb) FIGURE 4.1 Composite spatial distribution of 1-hour average concentrations (ppb) of SO2 from one representative month in each season between 1977 and 1978. This average is approximately equivalent to an annual average. Dashed isopleth is based on limited quantities of SURE data for 1977-1978 and on data taken at Whiteface Mountain, New York, after 1978. Numbers in parentheses are calculated values of dry deposition rates in kilograms of sulfur per hectare per year assuming a uniform deposi- tion velocity of 0.8 cm/s. SOURCE: Data on concentrations from Mueller et al. (1980) and, for Whiteface Mountain, from V. Mohnen, State University of New York, Albany, personal communication (1983~. tation, as indicated in Figures 4.3 and 4.4. These distributions may be compared with that of the hydrogen ion concentration in precipitation in Figure 4.5. From these figures, the pattern of deposition of hydrogen ion in precipitation appears to correspond to regions of elevated sulfate and nitrate concentrations. This finding does not necessarily follow without accounting for all the cations (such as NH4 and Ca++) and anions that

93 7 'it ~_~l \~z 8 ( '\~ J ~ 24-Hour c~-~- _;- ~SO4 ("g/m3) ran ~ a./ .~ FIGURE 4.2 Composite spatial distribution of 24-hour average concentrations of sulfate (,ug/m3 ~ from one representative month in each season between 1977 and 1978. This average is approximately equivalent to an annual average. Numbers in parentheses are calculated values of dry-deposition rates in kilograms of sulfur per hectare per year assuming a uniform deposition velocity of 0.2 cm/s. SOURCE: Data on concentrations from Mueller et al. (1980~. may be important factors affecting the acidity of precipitation. A qualitative comparison between dry- and wet-deposition rates can be made from observation. Wet-deposition rates for sulfate are shown in Figure 4.3 for data taken in 1980. Estimated dry-deposition rates for SO: and particulates are shown in parentheses in Figure 4.1 and 4.2 (estimated values in kilograms of sulfur per hectare per year). The rates have been calculated using uniform deposition velocities considered typical of values reported in the literature for SO2

94 ~4 3 4, ~ l l ·5.1 i " ·10 " ·16 I - t, v ' \' - r t - ~ /~6 3 ·5 2-' t . , 1 t__ j .~ L-_ 49 ~48 \ \` · 53" an, I ? ~37 t~'-~-~w 1 m mole/m2 ~ 0 961 kglha l , ._ _ ~ l _ ~1 4 I ·45 ~ t- _ 1 ~ - or'_ ~ ~ 11 · ~ ~] I I 2 :~,~!, -~I' In' ..~-.;..:c..',fyIt. ~ - - ~ \ , ~ , '' 20 FIGURE 4.3 Spatial distribution of mean annual wet deposition of sulfate weighted by the amount of precipitation in North America in 1980 (mmoles/m2~. SOURCE: U.S./Canada Work Group #2 (1982~. gas and for submicrometer particles (sulfate). Note that in much of the region of high ambient concentrations, dry deposition apparently dominates total deposition. This result suggests that dry deposition of SO2 exceeds wet deposition in parts of the Ohio River Valley and the Ohio-Pennyslvania-western New York area but becomes progressively less important farther from the region of major emissions, to the northeast in New England and Canada. At large distances from sources, ambient concentrations of sulfur oxides are low; wet deposition will dominate dry deposition far from sources if precipitation is significant.

95 ~=O.5 _ ~ 0. 9, ~ I ~ I 8.2 e/ ~ 3.5 / ~- 5.5 / 4 ~'\ _q ~1 ~ A_ . t_ _ 11~ Of) ~0 Cl i_ _. ~< ;~3~\ '"' 8~.9 ''it ~ _ _ l ' - I =~ ~ <-'9 ;W -' to _ ~ ~6.0 ~ $43~~\ 1 m mole/m2 ~ 0.62 kg/ha / FIGURE 4.4 Spatial distribution of mean annual wet deposition of nitrate weighted by the amount of precipitation in North America in 1980 (mmoles/m2~. SOURCE: U.S./Canada Work Group #2 (1982~. THE INFLUENCE OF METEOROLOGICAL CONDITIONS The spatial distributions of deposition for acid-forming materials are similar, and elevated concentrations appear to be associated with areas in which emissions are high. The temporal behavior of sulfur oxides and nitrogen oxides is dominated by meteorological variability. For example, analysis of the SURE data indicates that the variability in ambient concentrations of SO2 was as much as an order of magnitude greater than the vari- ability in SO2 emissions over the eastern United States (Mueller and Hidy 1983). This finding suggests that advanced and sophisticated analyses are needed to separate the influences of emissions from those of aerometric parameters in such data. These analyses have

96 '` .1.0 ~ \ ~ aim_ ;ir;; TV j i ret I\. 1~. 5.7 ~ ~ i,> . ~, ~ 7-- ~ use\ tj ·0. ~\ ~ . · - ~-0 6-' I ·3. _ _ _ + _ _ ~ 0 .4 I ~,) ,' t~o.5 ; $~ i\ 1 m mole/m2 ~ 0.01 kg/ha = 1- FIGURE 4.5 Spatial distribution of mean annual wet deposition of hydrogen ion weighted by the amount of precipitation in North America in 1980 (mmoles/m2 ). SOURCE: U.S./Canada Work Group #2 (1982~. followed two directions in the recent literature. The first is descriptive, taking into account the behavior of sulfur oxides in the atmosphere as a function of climato- logical or meteorological features. The second stems from the first by adapting statistical techniques to separate influences of different processes. In this section we review results of analyses of data according to meteorological conditions and studies of air-mass trajectories. In the next section we present the results of statistical methods of analysis. Classification of Meteorological Conditions One of the primary conclusions of many studies of field observations is the absence of a definitive relationship

97 between SO2 emissions and sulfate concentrations in dry air or in precipitation patterns. Sulfur emission rates are relatively constant, whereas the concentrations of sulfate aerosol and SC: in the air are highly variable and dominated by meteorological conditions (Electrical Power Research Institute 1981). The concentrations of sulfate particles tend to be high in summer, presumably because of more rapid photochemical oxidation of SO2, high in the central and western region of high-pressure systems that move slowly toward the Atlantic, and high in maritime tropical air emanating from the coastal region of the Gulf of Mexico. Concentrations of SC2 also tend to be high under stagnant air conditions of slow-moving, high-pressure, anticyclonic systems but tend to be low in the maritime tropical air. Trajectory analyses applied to specific transient anticyclonic systems have further demonstrated the tendency of such systems to favor air- stagnation situations in which high levels of pollutants accumulate (King and Vukovich 1982). Evidence based on chemical and meteorological analyses as well as satellite photos suggests that, on occasion, pollutants originating in the Midwest and Northeast can be entrained in the clockwise flow around a transient high-pressure system and transported to the Gulf Coast region. The pollutants are believed then to be adverted back through the Midwest and Northeast in the south- westerly flow of maritime tropical air (Wolff et al. 1981, 1982). The slow-moving flow of southwesterly maritime air from the southern states entrains pollutants from sources in its path and eventually moves over the northeastern United States and eastern Canada. Summer- time convective storms that occur in this air tend to be quite acidic compared with precipitation during the cooler months (MAP3S/RAINE Research Community 1982, Raynor and Hayes 1982a). Precipitation during autumn, winter, and spring in eastern North America also tends to occur in moist southwesterly air that frequently emanates from the southern states. During these seasons, most of the precipitation occurs in the vicinity of warm fronts associated with cyclonic low-pressure storm systems. Precipitation develops because the southwesterly air travels faster than the warm front, which represents the boundary of colder and heavier air lying to the north of it. The southerly air ascends over the cold air and is cooled, producing large areas of precipitation. The situation is illustrated in Figure 4.6. The center of -

98 ~__ warm air ____ - - Cold air In_ Cold air ,,,,,,,,,,,,,,,, ~ - Cold air ~ Cold air (Dc)lluted) \ Cold air `~{ / I' ~< C~ Preci p. ~ l h , '\ i I, ~ , Warm air ,, a, ,' ~ Warm air ~7W Cold air _~: ~_ _ FIGURE 4.6 Idealized depiction of air masses and precipitation typically associated with warm fronts and low-pressure systems during autumn, winter, and spring in eastern North America. SOURCE: Lazrus et al. (1982).

99 the low-pressure system tends to travel toward the Northeast, and naturally the associated warm front moves with it. A map of preferred tracks of cyclonic weather systems (Figure 4.7) indicates the tendency for movement toward the Northeast. The bands of precipitation asso- ciated with the warm front move northeastward, depositing pollutants incorporated by the southwesterly air during its passage from southern or midwestern states. Although warm frontal precipitation is less acidic than summertime convective rain, warm fronts may deposit more acidic material in the Northeast because they deposit more precipitation (Raynor and Hayes 1982a). Figure 4.8 shows precipitation and wet deposition of chemical species as a function of the type of precipitation and type of synoptic weather system at Upton, New York. The data of the SURE network have made it possible to quantify pollution episodes in terms of duration, sulfate concentration, extent of area, frequency of occurrence, and association with meteorological conditions. This information is summarized in Tables 4.2 and 4.3. Table 4.2 defines events (episodes) according to the number of stations with sulfate readings in specific ranges. The notations in Table 4.3 referring to meteorological type are illustrated in Figure 4.9. Warm southwesterly maritime tropical air (indicated as mT) referred to above occurs about 12 percent of the days in the region of the SURE network. The air involved in warm frontal precipi- tation is included in a transitional category indicated as Tr in Figure 4.9 and Table 4.~. Air manors Of continental polar origin (cP) are identified with large areas of high barometric pressure and anticyclonic circulation. Continental polar cold (cPk) air refers to flow from central Canada southwest. Air-mass stagnation under the high pressure area is referred to as cP2 conditions, while cPw refers to warm airflow from south to north on the west side of the anticyclone. Recent studies of fluctuations in SO2 and sulfate concentrations along with other evidence indicate that sulfate aerosol advec ted from distant origins makes a large contribution to the particulates in the New York City area only during the summer (Lioy and Morandi 1982, Tanner and Leaderer 1982). A correlation analysis of sulfate particle concentra- tions in dry air between various SURE sites indicates that significant regions of correlation generally extend 200 to 300 km from major SO2 sources during pollution episodes and rarely beyond 500 km (Electric Power Research

-of ~ - -o-o 100 Y: _- ~ ~ ~ --I if_17 1: - ~ 1 1 , ~ hi, , ~'10~ ~, , , i 1 1 of . , , , - ox - Ct A> C) . Lo . . c) - o CQ . Cot U) a · ~ s CO U. o Ct CO Cot , Cot C) 3 o G ._4 - C~ :^ o Ct Cot . ~ A_

101 C c o o ~ ~, ~ C ~ ~ 2 - ° ~ " ~ ~ IL IL a x E ~ o ° C C, ,' ~ o C ~ ~ C + a: u~ o o 8 - ° CD ~ ~ (ww) uo!1el!d!oa~d z 1 ~ o z _ o , C ' ~ a CO 0 [L C) J ~ a U) LL O IL J ~ ~ a ~ LL 0 c) ~ IL J ~C~ O CD O4 (zu~lbaw) uo!l!sodap 1aM · · · I ~0 g 8 8 8 X (ww) uo!IeI!d!oa~d ~z a + cn * z o~ cC 3 Z C: C~ q o U' + ~ Z z C _ . o CD C ~' 00 ,CO ~ ~ O CO ( zw/baw) uo!l!sodep 1aM ~oo ~o o a' Q ._ Q o ~n - - ~ + U) ~n LU + a: tl: + a: ~ U' ~_ - C~ o Ct ._ C) C) _` _ O O ~ 4_ - C) CO Ct ~: C~ ~0 Ct Ct Ct .o Q~ c~ ~> · c) o o · · - o o ~ s~ 3 Q -, Z 4 -4) ~ ~3 ~ oo to QCt =) L) C~_ _ Ct _ ·_ ._ ~ C~ ~ .= Ct 3 oo r~ O V ~

102 TABLE 4.2 Definitions of Pollution Events by Intensity and Coverage of Sulfate Concentration Percentage of Stations with Sulfate Concentrations Days per Event- Event Group >10 mg/m3 >15 mg/m3 >20 mg/m3 Range Mean Enlarged regional Regional Subregional Nonregional No event 40to93 25to72 10to61 ~ 40to70 15to25 Sto 10 J 25 to 50 <35 o 15 to 25 5 to 15 SOURCE: Electric Power Research Institute (1981). atoll 5 <5 1 to7 O _ o Institute 1981). The exception to this observation occurs during dusting of pollutants from southern regions to the Northeast by southwesterly air. Another indication of the relative importance of meteorological processes in determining wet deposition can be found in European data. In eastern North America, the warmer months tend to be the period of greatest precipitation as well as of most rapid photochemistry. At certain coastal sites in Europe, however, the season of greatest precipitation does not correspond to the season of most rapid photochemistry. Figure 4.10 shows the seasonal variations in sulfate (in excess of that from sea salt) and amount of precipitation for three groups of sites in Europe. The triangles indicate coastal sites where periods of greatest precipitation do not correspond to the period of greatest sunlight intensity and hence to accelerated photochemistry. These cases demonstrate that gas-phase photochemistry may not be the dominating factor controlling the seasonal trend of wet deposition of sulfate. The influence of meteoro- logical and precipitation processes (and possibly of aqueous chemistry) appears to be of major importance. Air-Mass Trajectories The influences of various source regions on given receptor sites have been calculated by following the trajectories of the air parcels bearing the pollutants. The most widely used method calculates the trajectories along isobaric surfaces. The inaccuracies of the method arise mainly from the assumption that vertical air motions can be ignored, because the necessary meteorological data on which the calculation depends are available only twice

103 TABLE 4.3 Annual Percentage of Event Days by Air-Mass Category Air-Mass Categorya Annual Percentage of Days Event Group cPk cP2 cPw Tr mT in Each Event Group Regional 0 12 7 2 9 30 Subregional 0 4 4 4 3 15 Nonregional 2 10 2 18 0.5 32.5 No event 6 4 0.5 12 0 22.5 Annual percentage of days 8 30 13.5 36 12.5 100 in each air mass SOURCE: Electric Power Research Institute (1981). aSee Figure 4.9 and the text for descriptions of the air-mass categories. daily at widely spaced locations, and from the diffi- culties associated with calculating trajectories in the vicinity of precipitating systems. The periods over which samples of precipitation are collected are usually so long (one week or one month) that the source regions of the air masses contributing to the sample may change during a single collection period. This problem may even arise in event sampling in the case of cyclonic storm systems in which the precipitation falls from air masses that frequently change markedly over a period of hours. Recent studies of air-mass trajectories at Whiteface Mountain in the Adirondacks (Wilson et al. 1982) indicate that 62 percent of the sulfate ion and 65 percent of the nitrate ion are deposited by precipitation associated with air parcels emanating from the Ohio River Valley and midwestern states (Figure 4.11). Wilson et al. also point out that 56 percent of the total annual precipita- tion at Whiteface Mountain is carried by those air parcels and that the wet deposition of acid may be related more to this fact than to the higher concentrations in the precipitation originating upwind to the south and southwest. They also point out that the geographical gradient of acid deposition (normalized to the amount of rainfall) is not very steep. The normalized depositions vary from 35 mg/m of sulfate at the source region to 27 mg/m2 at Ithaca, New York, and University Park, Pennsylvania, to 18 mg/m2 at Whiteface Mountain. They postulate that atmospheric mixing of pollutants together with varying patterns of precipitation tend to make the spatial distributions of pollutant material deposited per unit amount of _ precipitation relatively uniform when considered over time scales of a year or more.

104 ~ ~ I CP >'art a,\/ I ~ ' ~4~ Tr ~A, >~< or ~ Or 1 / FIGURE 4.9 Categones of air masses used to classify regional air quality in Table 4.3. SOURCE: Electric Power Research Institute (1981). The tendency of atmospheric mixing processes and varying precipitation patterns to attenuate the gradients expected between source and receptor regions is illus- trated by isopleth maps of annual deposition of hydrogen ion (Figure 4.5), sulfate (Figure 4.3), and nitrate (Figure 4.4), which place most of the Adirondacks, a receptor region, in the same isopleth area as the Ohio River Valley, a source region. A similar trajectory study at Ithaca, New York (Henderson and Weingartner 1982), yielded the percentage deposition for trace constituents in precipitation as a function of the trajectory of the air parcel and time of year. From October through March the percentage deposi- tions from the southwest quadrant were H+, 72 percent; SO4, 77 percent; NO3, 70 percent; and NH4, 84 percent. Of the total precipitation during the period, 71 percent emanated from the southwest quadrant. Simi- larly, the values from April to September were H+, 63

105 percent; sod, 68 percent; NO], 63 percent; NHt, 76 percent; and precipitation, 52 percent. It is interesting to note the following: (1) during the cool period the percent deposition of acidic sub- stances corresponds closely to the percent deposition of precipitation; (2) during the warm period southwesterly precipitation deposits a higher percentage of acid per unit percentage of precipitation; (3) the Midwest is an important source of the ammonium ion (derived from gaseous ammonia), which tends to neutralize sulfuric acid; and (4) nitric acid, derived from NOX emissions that are partially due to vehicular traffic (about 44 percent of NOk emissions nationwide), is associated with southwesterly precipitation systems. The authors found no direct correlation between emissions and depositions but were able to develop empirical relation- ships with rainfall rate and air-mass velocity that 0.30 0.20 0.10 0.00 0.10 0.20 0.30 0.20 0.10 0.00 -0.10 0.20 0.20 0.10 0.00 -0.10 -0.20 - /7~"" - // ~\ / / I i, ~, ~ _ _ , _ _, ~ ,'? "it, /_" ~ ,, ' /' \\ /~) _ \, ~'%, ,, 1 1 1 1 1 1 1 1 1 1 1 1 I F M A M J J A S O N D J ~- ~4~ J ~ , ~ grit ·?' ' , . . j ~ - - ~ . . FIGURE 4.10 Seasonal variation in deposition of excess sulfate (solid curve) and amount of precipitation (dashed curve) for three groups of stations in Europe. SOURCE: Granat(19784.

106 9 8 7 6 - o 5 . _ 4_ . _ .= 4 C) 3 2 o (a) 15 N - C o . _ . _ o Q _ ~ I :, 10 o (b) _ Whiteface 1978 Ohio Valley/Midwest 17.3 liter Canadian/Great Lakes 7.9 liter Total precipitation 30.6 1 iter l l I /~/t:~a\i\a;kl\\ by/// Midwest ////,, Great Lakes ~ 1 1 1 1 ~/~/////~/~\\~\\\\~\\~1 30 60 90 120 1 50 180 210 240 270 300 330 360 Trajectory sector Whiteface 1978 her Ohio Valley/Midwest 23.3 mg/m2 Canadian/Great Lakes 11.7 mg/m2 Total deposition 37.4 mg/m2 I<i\\\\d 1 ~ ////, Midwest ~//~\ Great Lakes ~\\\d I ~ 1 ~/~/////~////~///~\\\\~\\N 30 90 120 150 180 210 240 270 300 330 360 Trajectory sector

107 600 500 - a, 400 - o :~ 300 o ~ 200 o En 100 (c) 400 - ~ 300 - o . _ ._ o Q c, 200 100 o (d) Whiteface 1978 Ohio Valley/Midwest 1087.9 mg/m2 Canadian/Great Lakes 531.6 mg/m2 Total deposition 1700.5 mg/m2 oL _ _ ~ 31% of total 30 60 90 120 150 180 210 240 270 300 330 360 Trajectory sector Whiteface 1978 ; Ohio Valley/Midwest 646.7 mg/m2 Canadian/Great Lakes 283.4 mg/m2 Total deposition 994.6 mg/m2 65% of total ~ 28% of total Is /~ 1 ~////~///~////~\\\~ 30 90 120 150 180 210 240 270 300 330 360 Trajectory sector FIGURE 4.11 (a) Precipitation, (b) hydrogen ion concentration, (c) sulfate ion con- centration, and (d) nitrate ion concentration in precipitation as a function of the directional sector through which the air parcel passed to reach Whiteface Mountain, New York, in 1978. SOURCE: Wilson et al. (1982~.

108 accounted for 92 percent of the variability of sulfate deposition with SC2 emissions along the air-parcel trajectory. Similarly, 93 percent of the variability of nitrate deposition with NO2 emissions was accounted for in these empirical relationships. A third recent air-parcel trajectory study conducted in south central Ontario showed similar directional expectancies. During the warm period the depositions associated with the southerly and southwesterly air-mass trajectories were H+, 75 percent; SO4, 70 percent; and NCi, 57 percent; and during the cold period, H+, 85 per- cent; SO4, 87 percent; and NO3, 86 percent (Kurtz and Schneider 1981). The percentage of total precipitation emanating from the southerly and southwestern sectors was not indicated. The evidence of the air-parcel trajectory studies at specific receptor sites in remote areas of the north- eastern United States suggests that the industrialized region to the southwest is a major source of both nitric and sulfuric acids deposited at the sites. However, the high percentage of acid deposition in the Northeast associated with air parcels coming from the southwestern sector may be a result of the high percentage of total precipitation that is delivered by these parcels. More- over, observed gradients of deposition demonstrate that atmospheric mixing processes and precipitation variability obscure the source-receptor relationship to the extent that simple transport models may not realistically relate specific source regions with specific receptor sites STATISTICAL METHODS OF ANALYSIS . Statistical techniques can help to develop additional perspective on the significance of meteorology and chemical processes on the behavior of ambient sulfate. The resulting empirical models are used to relate emissions and ambient conditions. Conventional statistical techniques have been used to establish correspondences between explanatory variables such as ambient SCAR or NOk concentrations, other aerometric variables, and the end products of atmospheric reactions (sulfate or nitrate). In interpreting the data, it is assumed that ambient concentrations of SO2 or NOX are proportional to emissions. Linear regres- sion analyses have been performed to relate covariations in the end products to covariations in the explanatory

109 variables. The results (estimates of regression coefficients) were sometimes inconsistent from location to location and often indicated negative correlations among observables that Ghoul ~ ; n Ph=~rv h" ace; - ; .,^1', ~ Or ~ ~ ~ ~ ~ ~ ~ ~ correlated. Although misleading estimates may result for a number of reasons, one of the more important problems in aerometric data is collinearity among the explanatory variables. Belsley et al. (1980) discussed this problem, which results when one or more columns of the matrix of the explanatory (or independent) variables is an approxi- mate linear combination of the other columns. If there are strong correlations among the explanatory variables, there will usually be approximate linear relations among the columns. The result is that the standard errors for the estimated regression coefficients may be drastically increased and thereby may even cause estimates of true coefficients with small magnitudes to have the wrong sign. To avoid these problems, the common features of large bodies of aerometric data have been examined using alter- native multivariate methods. Two such methods are regression on principal components, which eliminates problems with collinearity in explanatory variables, and empirical orthogonal function analysis (principal- component analysis) in which intersite covariations between end-product variables are examined. These methods also offer an efficient means of screening large amounts of data for internal consistency in the relationships between the explanatory variables. Regression on Principal Components Principal-component analysis (Marcia et al. 1979) applies an orthogonal transformation to the explanatory variables, which results in a new set of uncorrelated variables that are linear combinations of the original variables. The transformation matrix is the matrix of eigenvectors of either the correlation or the covariance matrix of the explanatory variables. The first principal component is the linear combination of variables that explains or accounts for the maximum variabiliEv in the ~ . . original variables. The second principal component is that linear combination, uncorrelated with the first principal component, that explains the maximum amount of variability not already accounted for by the first component. The third, fourth, and other principal components are defined similarly. There are as many

110 principal components as there are linearly independent explanatory variables; however, the first few principal components usually account for most of the variability in the explanatory variables. The new variables or prin- cipal components have been interpreted in terms of physically significant factors such as chemical or transport processes (e.g., Henry and Hidy 1979). When the principal components are found using the correlation matrix of the explanatory variables rather than the covariance matrix of the explanatory variables, the coefficients in the regression of end products on the new variables are "standard partial regression coeffici- ents" (Snedecor and Cochran 1967). They provide a measure of the comparative "strength" of association between end-product variability and the new principal- component explanatory variables. The reason for this interpretation is that principal components are uncor- related, so that a coefficient pertaining to one come ponent is linearly unrelated to coefficients pertaining to the other components. In addition each component has unit variance so that the component is scale-free. Considerable progress has been made in interpreting the temporal behavior of ambient sulfate by this type of approach. For example, Henry and Hidy (1979, 1982) analyzed large data sets from New York City, St. Louis, Los Angeles, and Salt Lake City using regression on principal components. In New York, St. Louis, and Los Angeles, the regression of sulfate on Sol at the same measurement location was not significant. - - ~ Instead, the explanatory variables, ozone, temperature, and absolute humidity, accounted for the variability in sulfate data. In Salt Lake City, sulfate was found to be related to SO2 concentrations and to conditions promoting atmospheric mixing. Evidently, variations in ambient SO2 levels in themselves have only a weak influence on variations in airborne sulfate concentrations in these cities, compared with the influences of variations in atmospheric processes or meteorological conditions. Rural data from the northeastern United States have Iso bean examined Gina these techniques (Henry et al. Regression on principal components of rural data indicated very complicated relationships between aerometric variables. However, significant evidence for a direct relationship between sulfate levels and ambient SO2 concentrations of 3-hour and 24-hour averaged SURE data emerged from the analysis. With regression on principal components, a linear 1980, Mueller and Hidy 1983). . . .

111 proportionality between ambient SO2 and ambient sulfate variability at the same site can be derived objectively. If regional SC2 concentrations in air are proportional to Sol emissions and if dry deposition is proportional to ambient sulfur oxide concentrations at ground level, then one can infer that reduction in SO2 emissions should logically result in reductions in dry deposition, even with the dominance of meteorological factors. Empirical Orthogonal-Function Analysis The influence of the spatial distributions of SO2 emissions on ambient sulfate distributions also can be obtained using empirical orthogonal-function analysis to analyze sulfate measurements at a number of stations over a period of time (e.g., Henry et al., 1980, Mueller and Hidy 1983, Peterson 1970). This technique derives eigen- vectors from the estimated covariances of sulfate concen- trations between pairs of stations. Use of covariances rather than correlations gives greater weight to sites with larger sulfate variability. Spatial patterns are then derived for each empirical orthogonal function (or eigenvector) separately by plotting the value of the eigenvector for each station, interpolating the values onto a grid, and then drawing contours from this grid. Examples of the first two empirical orthogonal functions derived from the SURE data for July 1978 are shown in Figure 4.12. The patterns shown are contours of constant values of the empirical orthogonal functions, where stations with the same sign covary together (when one is high, the other is high) and stations with opposite signs vary in opposition. The shading indicates areas of high SO2 emission density. fine empirical orthogonal function that accounts for the largest amount of sulfate variability in the data, identified as the ~first," is shown in Figure 4.12(a). The function accounting for the second largest amount of the variability, identified as the "second" function, accounts for the next largest fraction of sulfate variability [Figure 4.12(b)]. For the July 1978 data, the first two empirical orthogonal functions account for 74 percent of the deviation of the sulfate concentration from its mean value. The spatial patterns of August 1977 were similar to the July 1978 patterns shown in Figure 4.12; however, the patterns obtained from the SURE data differ from season

112 ~ Jo 14 i , :-~] . Cat ~ .. ILL ¢) pL14 0 3 C) c~ =: "D ~ C) o ~ · = - ~ 0 5 ~ - O Ct ¢) ~ 1 ~ <A Q) Cat ~ _ o O ~ Cat 4 O ~ Let 4 ~ E a to _ o _ ~ J; _~t J JO ~ O 3 $ Id V: oo _ ~ ¢., ~ ~ ~ 3 _ Cat i_ C<S U. . - ° o ~ 'Cat C) Cat Us ·_ et ~ Q. O ~ Can .= V: <~ Cat ° X .= ~ ~ O O ~ ~ I.= C~ Vet _ Cat o -= ~ ~ . st - ~ a' ~ ~ c) o ~ ~ =,._ 3 ~ I_, o _ .o .- Cs ~ ~ _ c3 £ ~ ~ o o ~V - _4 ~ ~ 3 c~ o ~ ~ .~ ;^ . _ .,. ~ l, ~ C o ~

113 to season (e.g., summer to fall or winter). In all of these cases, the patterns can be rationalized to be consistent with the dominant weather observed during the periods examined. In summer, for example, the pattern of the first empirical orthogonal function was identified with large-scale meteorological conditions leading to persistent mass transport of air around zones of high barometric pressure northeastward across the Ohio River Valley toward New England. These conditions were identified as being most likely for long-range pollution transport (Mueller and Hidy 1983). The second important empirical orthogonal function in summer was identified with large-scale air stagnation under conditions of poor ventilation on the west side of summer anticyclones (e.g., Vukovich et al. 1977). (It should be noted that there were 10 days with conditions of air stagnation during July 1978, whereas the 40-year average for July is 2 days.) The association between SCAN emission distributions and the important empirical orthogonal functions found in the SURE data is circumstantial evidence that changes in the source patterns should induce changes in regionwide sulfate variability over the eastern United States. However, empirical orthogonal-function analysis and other statistical techniques have not progressed to the stage at which they can be used to predict consequences of selective changes in emissions. Elemental-Tracer Analysis Sources often emit particulate matter that may serve as chemical or elemental tracers of the emissions. Sampling and analysis for the tracer substances at receptor locations can provide evidence of the influence of various types of sources on those locations. Because monitoring stations are in the receiving field, these statistical approaches to analyzing empirical data are often called receptor models. The technique, developed originally for studies of urban air quality, has recently been proposed for application to long-range transport of acidic sub- stances. The method relies on the fact that particles associated with different sources have different elemental compositions (Gordon 1980, National Research Council 1980). For example, sodium in particles collected in aerosol monitors is usually associated with

114 a marine source, such as salt from windblown sea spray. Other elements in particles that are strong indications of particular sources are calcium (limestone, hence construction and demolition activities), lead (motor vehicles), vanadium (fuel oil), and zinc (municipal refuse). Particles from coal combustion are more dif- ficult to distinguish because their elemental composition is much like that of soil. However, coal combustion produces fine particles that are significantly enriched in selenium and arsenic with respect to soil, so these elements potentially can be used to distinguish the contributions of coal combustion from those of windblown dust. The tracer method has been tested and improved for application to urban air sheds in several U.S. cities, including Washington, D.C. (Kowalazyk et al. 1982), Los Angeles (Friedlander 1973, Gartrell and Friedlander 1975), Portland, Oregon (Core et al. 1981), and St. Lout s (Alpert and Hopke 1981, Dzubay 1980). The method has also been applied on the scale of global circulation to determine the sources of pollution particles collected at sites remote from human activities, such as in the Arctic (Cunningham and Zoller 1981; He idam 1981, 1982; Rahn 1981a,b). It has been suggested (for example by K. Rahn, University of Rhode Island, in an unpublished paper, 1982) that the elemental composition of particulate matter in aerosols might also be indicative of the origins of the aerosols after regional-scale transport and might help to resolve the question of the relative contributions of distant and local sources to acid depo- sition in eastern North America. Among major stationary sources that contribute substantially to the total burden of emission of SO2, those in the eastern United States are fueled by coal and oil, whereas those in the Midwest burn mainly coal. Thus an aerosol collected in the East that is enriched in vanadium would be presumed to be of relatively local origin, whereas an aerosol characteristic of coal combustion would be taken to be of midwestern orlgln. As vet there has been no systematic evaluation of the _ , ~ method, nor has it been applied to a number of available sets of data. Care should therefore be exercised in interpreting the results of preliminary analyses using the method, which may be more difficult to apply on a regional scale than on either an urban or a global scale Urban air quality is usually determined by local emissions, so the elemental composition of particulate

115 matter collected at urban monitoring sites can be expected to reflect the types of sources in the urban air shed, with the possible exception of aerosol particles, such as sulfates, formed in the atmosphere from precursor gases and transported long distances. Polluted air reaching remote sites, by contrast, has usually undergone so much mixing that its elemental composition can be considered to be a composite representative of a large source area. When polluted air travels from the source region to a remote site mostly over the ocean, however, there are few additional sources contributing particles to the air mass during transit, and the method can give unambiguous results. At rural continental sites, the dominant local source of airborne particles is usually soil. Most other types of particles come from distant sources such as large power plants or major metropolitan areas. There appear to be two critical problems in the application of elemental tracer analysis to regional transport that must be solved before the method can be useful. The first problem arises because of the poten- tial that the elements of interest as tracers may not be transported at the same rates and over the same distances as sulfates and nitrates, which are of primary concern in acid deposition. For example, manganese and vanadium have been proposed as characteristic tracers of emissions from Midwestern and eastern source regions, respectively. Manganese, most of which arises from entrained soil, is typically found on relatively large particles that settle to the ground close to the source; only about one third of air- borne manganese is found in fine particles that would be expected to undergo long-range transport. Elemental tracer analysis attempts to correct for the presence of crustal material. Because most of the crustal manganese occurs on large particles, the technique focuses on manganese in fine particles. The source of airborne fine-particulate manganese is uncertain; coal is deficient in manganese relative to soil, for example. Almost all of the vanadium in particulate matter in the East is associated with fine particles. Whether depo- sition processes for sulfates and nitrates are similar to those of manganese and vanadium is unknown. The elements used as tracers are normally emitted in the form of solid particles, whereas the precursor gases SO2 and NOx are transformed into particles by chemical processes after emission. To the extent that formation of sulfate and nitrate is dominated by in-cloud processes, it seems

116 likely that different processes will govern transport and deposition of the acids and the tracers. Coal combustion is predicted to be a major source of airborne selenium, appreciable fractions of which leave stacks in the vapor phase and condense on particles as the exhaust gases cool in the atmosphere. The behavior of selenium may therefore be more like that of sulfur in the atmosphere than is that of manganese. The second, and perhaps more difficult, problem is that of differentiating contributions of midwestern sources from those of eastern cities over which an air mass originating in the Midwest may pass before it reaches sensitive receptor areas. While experience with the technique indicates that it is practical to differ- entiate the contributions to urban aerosols of types of sources in a local area, it may not be possible to assess the contributions to regional aerosol burdens of sources in different urban areas. It may indeed be possible to distinguish eastern air masses from midwestern air masses that are uncontaminated by emissions from eastern cities but not possible to identify a Midwestern air mass that has passed over and received inputs from eastern sites. It seems likely that arrays of elements, rather than a single one, will have to be used as tracers of pollution of Midwestern origin. ANALYSIS OF HISTORICAL TRENDS In this and the next section we analyze the relationships between emissions and wet deposition using data obtained from monitoring networks. The central question is whether the data can be used to judge whether deposition is lin- early or nonlinearly related to emissions. (See Chapters 2 and 3 for discussions of the processes that might lead to a nonlinear relationship.) If the relationship is linear, a change in emissions would be reflected in a proportionate change in deposition; if it is nonlinear, the change in deposition, if any, would be dispropor- tionate. Data on precipitation for analysis are available from both Europe and North America. Considerably more data are available from Europe, where monitoring programs have been in existence for many years, than in North America, where systematic monitoring of precipitation chemistry is a comparatively recent development. In this section we use the body of historical data as direct evidence of the

117 relationships, and then, owing to the paucity of these data, we assess the indirect evidence for linearity or nonlinearity in North America in the next section. More precisely, we consider whether currently available observations are consistent with conditions that should theoretically prevail if the relationship between emis- sions of SO2 and NOX and wet deposition of sulfate and nitrate in North America were strongly nonlinear. The possibility that changes in deposition rates are not linearly related to changes in source strength initially arose in the examination of historical trends in emissions and deposition in Europe. The body of European data provides the only direct evidence for this phenomenon. For example, from about 1960 to 1975, most of the areas in the European network for precipitation sampling experienced either constant or declining deposi- tion of sulfate (Figure 4.13) despite significant increases in SC2 emissions during that period. This apparent nonlinearity of response has been attributed both to chemical factors and to changing climatic conditions. The data analyzed by Granat (1978) were collected using the bulk sampling technique, by which collection containers continuously exposed to the atmosphere collect combined wet and dry deposition. The observed trends are made somewhat uncertain by changes in analytical techniques, sampling techniques, and analytical laboratories throughout the sampling period. Further- more, bulk deposition samples are subject to contami- nation from windblown dust, leaves, and insects, for example, and consequently are considered to be of poorer quality than current methods, which collect only wet deposition. Extensive direct evidence, comparable with that for Europe (Kallend et al. 1983), that can demonstrate a disproportionate relationship between emissions and deposition does not exist in North America. Efforts to reconstruct historical trends from the sporadic and disparate data that have been collected in the United States since the late 1950s (Likens and Butler 1981) are beset by large uncertainties (Hansen and Hidy 1982, Stensland and Semonin 1982). The monitoring of precipitation chemistry at the Hubbard Brook Experimental Forest appears to provide the longest continuous record of deposition data at a receptor site in the northeastern United States. These data, as in the European network, were obtained from samples of bulk deposition. The Hubbard Brook samples were collected on a weekly basis,

118 on o 1 ~ o In-'.. 4~,, o o -N o o so or in. - : ~" ~ in 'if Ct Cot ._ V, o ._ Cot ._ ._ Cot - o ., .. ;^ U) . C) As: o As Cat o V) o V, % a' o Cot Cot ¢) x o o :- cot Do o ·_ Cot _ o ~ · - ~ ._ . . ,_ > o CQ o tD ~ a, ~o, _

119 usually with three samples collected simultaneously. We regard the Hubbard Brook bulk-deposition data as reasonably reliable because samples were collected frequently and the collection of several samples at one site permits the detection and elimination of contaminated data. The Hubbard Brook data reveal several trends (Likens et al. 1980), which are borne out at least qualitatively in bulk deposition monitoring with relatively unreliable quality control from nine stations in the New York State area (Miles and Yost 1982, Peters et al. 1982): (1) There has been a decrease in sulfate concentration sine e 1964 but an increase in nitrate concentration over the same time (Figure 4.14 and Table 4.4). (2) The annual pH of precipitation showed no long-term significant change from 1964 to 1977, though several short-term changes did occur. (3) A linear regression equation of data points from 1964 to 1977 indicates no statistically significant trends in H+ deposition from 1964 to 1977. (4) Recent changes in H+ deposition correspond more with changes in nitrate deposition than with sulfate deposition, even though sulfuric acid is the dominant acid at Hubbard Brook. The contribution of NO3 to total acidity has been increasing, whereas that of SO4 has bee n decreasing (Galloway and Likens 1981). Year-to-year changes superimposed on the long-term trend may be related to climatological influences. A linear-regression analysis of the Hubbard Brook data against time indicates a decline in sulfate concentra- tions between 1965-1966 and 1979-1980 of about 33 + 18 percent (95 percent confidence limits). The regression of sulfate concentration against time gives a Slope of -~.074 (standard error 0.016) and an intercept of 3.186 (standard error 0.138). Using the t test, the slope is significantly different from zero at the 0.0003 level. The assumptions of the regression--that errors on the fi t are independent, have zero mean and a constant variance, and follow a normal distribution--were tested and do not appear to have been violated. The Hubbard Brook record, unlike the European data, appears in toto to demonstrate a reduction in sulfate concentration similar to the general reduction in SO2 emissions (Figure 4.15 and Table 4.5). Between 1965 and 1978 SO2 emissions in EPA Region I (comprising Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, and Vermont) declined by about 38 percent (excluding data for Vermont, which are not reported in

120 3.0 ._ c - 2.0 o ._ a, 1.0 o C: \soi . ~ .' NO3 r--~--~- _, ~.~' 'it, ,,^~ A 90 85 80 ~/ 75 \/\ 65 60 _ 0 WA~> . / - ~ l 4.25 _ 4.20 _ I 4.15 _~ Q 4.10 _ 4.05 _ 4 00 _ ,,' 1 1 1 1 1 1 1 1 1 .1 1 1 1 _1 1969 1971 1973 Year 1975 1 977 1979 1981 FIGURE 4.14 Annual weighted concentrations of sulfate, nitrate, ammonium, and hydrogen ions and weighted pH of precipitation at Hubbard Brook Experimental Forest from 1964 to 1977. SOURCE: Likens et al. (1980) and G.E. Likens, Cornell University, personal communication (1983~. the reference cited for Table 4.5). In Region II (New Jersey and New York) the reduction was about 40 percent, while that in Region III (Delaware, the District of Columbia, Maryland, Pennsylvania, Virginia, and West Virginia) was about 11 percent during the same period. Aggregate sulfur emissions in Region IV (Alabama,

121 TABLE 4.4 Annual Average Concentrations of Sulfate and Nitrate from Weekly Bulk Samples at Hubbard Brook Weighted by the Annual Amount of Precipitation (mg/liter~a Year SO: NO3 1964/65 3.16 0.70 1965/66 3.33 1.39 1966/67 3.13 1.49 1967/68 3.27 1.56 1968/69 2.42 1.18 1969/70 2.24 1.14 1970/71 2.75 1.71 1971/72 2.67 1.74 1972/73 2.87 1.74 1973/74 2.84 1.67 1974/75 2.54 1.52 1975/76 2.14 1.22 1976/77 2.20 1.66 1977/78 2.04 1.32 1978/79 2.55 1.69 1979/80 1.91 1.44 1980/81 2.36 1.66 SOURCE: G.E. Likens, Cornell University, personal communication, January 9, 1983. aData are recorded in the water year, from June 1 through May 31. Florida, Georgia, Mississippi, Kentucky, North Carolina, South Carolina, and Tennessee) increased by about 33 percent between 1965 and 1978. Region V emissions (from Illinois, Indiana, Michigan, Minnesota, Ohio, and Wisconsin) decreased by approximately 18 percent over the period. Total SO2 emissions from EPA Regions I through V, which comprise all states east of the MiSSiSSiPPi River plus Minnesota, decreased by about 8 percent between 1965 and 1978. The nitrate data at Hubbard Brook suggest an erratic trend toward a maximum around 1970, followed by a leveling off or a slight decrease. The emissions of NOX in the Northeast increased 26 percent between 1960 and 1970, and then decreased 4 percent by 1978. The nitrate deposition data also appear to reflect emission trends in the Northeast. The data of Peters et al. (1982) were obtained from a network operated by the U.S. Geological Survey (USGS). The USGS network employed bulk samplers; single samples were collected on a monthly basis. The scatter in the USGS data is quite large. According to Peters et al. (1982), the data do not suggest statis

122 21 20 19 18 17 16 10 9 7 _ 6 5 4 3 o 1960 1965 1970 so2 Regions - IV la Year 1975 1978 aExcludes emissions from Vermont, which in 1980 amounted to about 7,500 tonnes of SO2 and 12,000 tonnes of NOX.

123 12 11 10 9 8 7 6 5 4 1 No x - _ _ - _.. . ~~ Regions la v __ IV _ __ I l I J 1960 1965 1970 1975 1978 Year FIGURE 4.15 Annual emission of SO2 and NOX in the eastern United States by EPA Region, 1960-1978 (million metric tonnes). SOURCE: U.S./Canada Work Group ~3B (1 982~. tically significant long-term trends except for sulfate at one site (Hinckley). Although the direction of change is consistent, these results suggest weaker trends since the mid-1960s than the Hubbard Brook results. Uncertain quality control, especially during the early years of operation obviates the usefulness of these data however (Miles and Yost 1982). Another source of data is the New York State Department of Environmental Conservation (1976, 1978,

124 ._ ._ Cal Cq ._ U. Cal .~ s - o - cn o o - - :: Cal Cal Cal CQ - Car o O '7 ca ~ a ~ O Cal _ 0 ~ Up C) c40 o It_ ~ o O ~ ._ .0 ca ~ ._ Ct cat At ~ O _ S _ ~ 00 in 0 · \~ <: ._ 0\ 0 ~ Z_ 0\ 0 0\ C~ U' \0 ~ ~O O \0 ~ O _ `0 _ _ ~ ~ 0\ ~ X ~ X O ~ X ~ ~ Q ~ ~ ~ x _ ~ ', y;, t_ .D \0 - ~ ~ - ~ \0 0 O - ~4 U~ ~ oo X ~ ~ t~ oo ~ ~ ~1 ~ x ~o~ cr v~ 0 - r~ ~ ~o ~ ~ ~o ( - , _~ ~ O ~ ~ q" - r~ ~ ~ 0\ ~ t- O ~ O ~ ~) ~ ~1 0 ~n ~ ~- ~o ~ 0 ~4 ~ ~ ~ r- ~ _~ ~ ~ ~ ~ ~ ~ C~ so X ~ ~ °\ q SQ ~ ~} ~ ~ ~ (~ ON X 0 0 ~ ~ ~ ~ 0 _~ ~ X ~ - C~ ~ _ 0\~o ~ ~o ~ ~ ~ ~ a~ ~ c~4 X 0 ~_ _ ~ t_ ~ ~t =~ ~) ~ \O ~ ~D - - ~ \0 ~ ~ ~ - - ] ~ , _ _ _ _ x ~ d. oo X _ _ _ ~ _ 0 X oo 0 X ~ ~ V) ~ ~o c4 ~ 0 - 0 ~ ~P ~ _ ~r, x oo ~ \0 ~ ~ O _ _ _ _ o~ ~o ~ C~ ~ ~ - ~o crs r. ~ v~ ~ ~ ~ ~ ~ ~ ~o - ^ ~ ~ - ~ ~ ~ ~ _ _ _ ~ ~ ~o ~ - ~ C4 0 - ~ ~ ~ - ~ ~ X ~ C~ oo ~ - d. ~ ~ ~ ~ X ~ - r~ ~ ~ X X ~ X X ~ ~o _ t_ ~ 0 ~ 0 _ - ) _ ~ ~ ~ ~ ~ o~ - ca , _. ~ ~ ~ 5 ~ ~ ~ ° ~ 3 ~ ·= ~ ~ ~

- Q ~ x ~1 ~ ~ oo O ~t ~ `0 ~ O ~ ~ ~ _ ~ ~ O ~ 1- ~ ~ - ~1 u~ ~ oo tr) - O ~ ~ ~ ~ et _ _ -} ON ~t - \0 ~ - 0\ ~ O X ~ ~ ~ O _ - , ~ _ oo _ o~ oo ~ ~ ~ ~- ~ - ~ ~ ~ ~ ~ ~ O ~ ~ ~ ~ ~ e' oo ~ ~ ~- ~ Q - 1 ~ O O d. O Q ~ O c~ ~ x ~ x 0 ~ ~ x ~ - ~ cr~ - - ~ ~ ~ ~ ~ ~ O ~ ~ ~ ~ ~ ~ - - oo t- ~ kD ~ ~- ~ ~ CJ, ~ O ~ \0 0 - c~ ~ ~ ~ ~ ~ ~ ~ cr. _ - , ~ X ~ ~ O - - X ~ O - ~ 00 - ~ ~1 _ -) ~ ~ ~ ~ \D ~ - ~ ~ O ~ 0 \0 - CN ~ ~ ~ X ~- ~ ~ ~1 - O- X ~ - - ~ \0 ~ O X C ~ ~ ~ ~ ~ ~ ~ ~ O - - ~ 'i <'] d 0\ c-, ~ _ ~ C~ ~ X ~- ~1 0 ~ C~ ~ _ _ ~ _ - oo ~ - - ~ X ~ ~ oo ~ X O ~D O ~ ~ ~ - ~ X ~ d. cr - O ~ o~ ~ ~ 0\ - O ~ O - ~ _ _ _ SQ ~ r- ~ - O O ~ - ~ ~ ~ X ~ O ~ ~ oo ~ - X - ~4 v, \0 00 ~I X ~ X O - Co ~ X C~ _ oO O ~ ~ - - ~ ~ ~ ~ O ~ 0\ 0\ Q ~ ~ X ~ ~0 0~ -> - Os ~ O~ t~ X c4 - ~ - \a ~D ~ _ ~ _ ~ ~ ~o ~ ~ o' 0 ~o ~ ~ x ~ ~ ~r ~ ~ ~ ~ r~ oo 0 ~ ~ ~ O - ) ~ ~ ~ ~ ~ _ - \0 ~ _ ~ ~ o 5.c E a 5 5 5 5 ~.= C ' O ~ _ ~ ~J ~ ~ U ~ 'C ~ 0 U, o C) .o D c~ O c~ c; e e ._ 0 c, 0 o~ ~ ~ c; m * .- c`, e 2 '-o. ~8 ~ e ~ ~ 0 e~ . _ .. c, ~ 0 O ~ c~ ~ "o c' D e o u, O' 125

126 TABLE 4.6 Three-Year Running Average Values for Sulfate Concentrations in New York (,ug/m3) New York Statewide Rochester Station Sulfate Levels 2701-01 964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 10.78 11.01 10.68 9.88 9.46 9.04 8.89 8.84 9.01 9.00 9.15 9.32 9.45 9.02 8.72 8.3 8.42 8.42 9.6 9.37 9.40 8.83 8.23 8.13 8.70 8.77 8.57 8.53 9.40 9.8 10.0 9.53 9.5 9.65 SOURCE: Department of Environmental Conservation (1976, 1978, 1981). 1981). According to these data, statewide concentrations of sulfate particles in New York have also decreased about 22 percent between 1964 and 1978 (Table 4.6). This decrease corresponds fairly closely to the decrease in Northeast SO2 emissions (Figure 4.15). However, several stations, e.g., at Rochester, reveal more complex trends throughout the period (Table 4.6). These data are diffi- cult to interpret. Many of the stations are located in urban areas so that both the statewide average and values from specific stations should be strongly influenced by local urban sources. In addition, controls for both SO: and primary sulfate emissions were implemented during this period. Both the USGS and New York State data are therefore of limited usefulness for assessing relationships between emissions and deposition influenced by long-range trans- port to rural areas. The most reliable data available in a continuous record are those from Hubbard Brook. These data comprise the direct historical evidence for an emissionsdeposition relationship and show no indication of a significant nonlinearity.

127 ANALYSIS OF RELATIVE BEHAVIOR OF SULFUR AND NITROGEN EMISSIONS Because of the small amount of direct evidence available, we resort to examining less direct empirical evidence that may reveal for eastern North America a nonlinear relationship between emissions and deposition. An explanation of the inconsistent emission and precipitation trends observed in Europe comes from work of Rodhe et al. (1981), who hypothesized that (1) gas- phase photochemistry is significantly responsible for producing the acid incorporated in precipitation and (2) conversion of SO2 to H2SO4 is indirectly influenced by emissions of Nat in the atmosphere (see the detailed discussion in Chapter 3 on the Rodhe et al. model). In the Rodhe hypothesis, HNO3 production is thought to be favored near the source region, while H2SO4 is favored in more distant regions. Nitric acid vapor formed by the reaction of NOk and the OH radical can be rapidly removed from boundary-layer air by either dry deposition (B.B. Hicks, National Oceanic and Atmospheric Administration, private communication, 1982; Huebert 1982) or by wet deposition (Levine and Schwartz 1982). If the mechanisms suggested by Rodhe et al. significantly influence the concentrations of SO4 and NO] in precipitation, the ratio of dissolved sulfate to nitrate is expected to increase as the SO2 oxidation becomes more efficient with increasing distance from the source region. The occurrence of the gradient of ratios does not, however, unambiguously demonstrate the importance of this mechanism, since other factors (such as very rapid dry deposition of nitric acid vapor in relation to that of SO] aerosol) might yield a similar effect if gas- phase chemistry plays a dominant role. However, the absence of the gradient of ratios would be difficult to reconcile with the proposed model. Molar ratios of sulfate to nitrate in precipitation in Europe calculated from the data of Rodhe et al. (1981) do indeed increase with increasing distance from the source region in summer (Figure 4.16 and Table 4.7), when the influence of gas- phase chemistry is expected to be greatest. The annual average of the molar ratio of sulfate to nitrate in precipitation is relatively constant through- out the area of the northeastern United States and eastern Canada in regions in which the pH precipitation is low but increases in extreme northern Canada (Figure 4.17). High values of the ratios in northeastern Canada

128 / 10°W ~0° 10°E Cot N/ / I '60~ ~ I at? it, flat ~ Z = ~ W -sat ~ FIGURE 4.16 Locations ofmonitonng stations used in the data analysis of Table 4.7. SOURCE: Rodhe et al. (1981~. and far north Europe (region 5 in Figure 4.16) may be affected by local smelting facilities. Enhancements of the ratio occur in areas of concentrated point-source emissions, e.g., the Ohio and Tennessee River Valleys (Figure 4.18) and along coastal areas, possibly because of sea-salt sulfate. A lower ratio occurs near the Great Lakes, a region with a relatively low ratio of SO2 to NOx emissions. That the ratio is relatively constant over a broad region appears inconsistent with the exis

129 TABLE 4.7 Sulfate, Nitrate, and the Molar Ratio of Sulfate to Nitrate in Precipitation in Europe Group of Stationsa 1 2 3 4 5 Sulfate (,umoles/liter) 1955-1959 62.5 50.0 28.1 25.0 15.6 197~1974 90.6 56.3 50.0 40.6 40.6 Increase (Jo) 45 13 78 63 160 Nitrate (~moles/liter) 195~1959 35.0 22.9 13.6 7.1 4.3 197~1974 60.7 51.4 25.0 21.4 7.1 Increase (%) 73 125 84 200 67 Molar ratio 1955-1959 1.8 2.2 2.1 3.5 3.6 197~1974 1.5 1.1 2.0 1.9 5.7 SOURCE: Rodhe et al. (1981). aGeographic locations of the groups of stations are indicated on Figure 4.16. fence of a strong nonlinearity due to photochemistry and demonstrates the apparent effectiveness of atmospheric processes to lead to thorough mixing of pollutants over a large spatial scale (up to 1,000 km in linear dimension). When the ratio of sulfate to nitrate in precipitation more complex. bute to the ratio). - is considered on a monthly basis, the pattern becomes Average ratios have been calculated for the MAP3S stations (MAP3S/RAINE Research Community 1982) for two seasonal periods, April to September and October to March, for the period 1977 to 1981. The seasonal and annual ratios are given in Table 4.8. The ratio tends to be larger in summer than in winter, with the seasonal dif- ference diminishing at lower latitudes (possibly because the intensity of sunlight varies less with season) or in maritime environments (where sea-salt sulfate may contri The seasonal change may reflect a less event conversion of SC2 than of NOk during the cooler months than during the warmer months, when photochemistry is more active and transport is more vigorous. Incorporation of nitrate into precipitation is much less sensitive to seasonal change. Emissions of SO2 are relatively constant throughout the year (16 percent increase in winter) (Electric Power Research Institute 1981). Seasonal variations in emissions, therefore, cannot account for the seasonal change in deposition ratios. The deposition ratios, however, are relatively invariable with respect to distance from major source regions, particularly during the warm season (1.4

130 ~1.2 ~ ~ ~ 4.1'' i'''/\ it_ \ ~' \ ~ 1.d ~ x, - ~ `~T '-' _-it \~1.2 ~ ~ ~1 .0 41.2 0.8 ~ :-'9 ,. o FIGURE 4.17 Average molar ratio of sulfate to nitrate in precipitation in eastern North America in 1980. SOURCE: B. Heikes, National Center for Atmospheric Re- search, personal communication (1983). Based on data from U.S./Canada Work Group #2 (1982).

131 it, 7W °~ ~ ~67.5 ~1.14 ) Hi' r / If 1 ~8 0.74 ~ I ~ _ 4)~~0.49 --in 1 05 1,'-87 ~-5' ~ ~ - 0.81 ~ 0.34 /J: ~ ~ ~ 0.92 rat -~ 0~72 ~ 1.24 <~1.21 if o A___ __ ^7 1 1.64 1 __-~ ~ _ FIGURE 4.18 Average molar ratio of sulfur to nitrogen oxides in emissions in eastern North America in 1980. SOURCE: B. Heikes, National Center for Atmospheric Re- search, personal communication (1983~. Based on data from U.S./Canada Work Group #3B (1982).

132 TABLE 4.8 Molar Ratios of Sulfate to Nitrate in Precipitation in the United States, 1977-198 1a April- October- Annual ;~ Whiteface Mt., N.Y. 1.4 (29) 0.65 (26) 1.08 Ithaca, N.Y. 1.4 (25) 0.62 (21) 1.07 University Park, Pa. 1.3 (27) 0.74 (29) 1.04 Charlottesville, Va. 1.3 (27) 0.81 (21) 1.16 Urbana, Ill. 1.4 (15) 0.94 (13) 1.25 Brookhaven, N.Y. 1.12 (17) 1.0 (17) 1.10 Lewes, De1. 1.4 (18) 0.93 (18) 1.16 Oxford, Ohio 1.5 (16) 1.2 (19) 1.37 Average 1.4 + 0.1 0.86 + 0.19 SOURCE: Adapted from MAP3S/RAINE Research Con~Tnunity (1982). aNumbers in parentheses are the number of months of data in each sample. + 0.1) when gas-phase photochemistry would be expected to exert its maximum impact on rain composition in accordance with the model of Rodhe et al. (1981). In neither season do the ratios significantly increase with distance from the most highly industrialized areas, suggesting that retardation of sulfate formation by NOx may be a minor factor at any time of the year in eastern North America. Accounting for the surprising relative invariability of the ratio despite the geographic variability of conditions requires considerable speculation. Perhaps the factor most responsible for uniformity is atmospheric mixing. Much of the frontal and convective precipitation in the eastern United States is from clouds forming in moist southwesterly air. Air-parcel trajectories have shown that this air is frequently adverted several hundreds of kilometers before precipitation occurs (Lazrus et al. 1982, Raynor and Hayes 1982b). During this transit, the air incorporates and integrates the emissions from all the sources in its path. Another possibility is an unknown coupling of the NON and SO2 chemical reaction pathways, which could tend to maintain a relatively unchanging deposition ratio of sulfate to nitrate, although there is no theoretical or experimental evidence of such a coupling. _ . . It is surprising to note in addition that the annually averaged molar ratios in precipitation at the MAP3S, NADP, and CANSAP sites in eastern North America resemble the molar ratios of SO2 to NOx emissions averaged on an annual basis by state or province (Figure 4.18). The emission ratios are somewhat higher than the deposition

133 ratios in the Ohio and Tennessee River Valleys. Simi- larly, the emission ratios are somewhat lower than the low deposition ratios immediately north of Lake Ontario. The smaller spatial variability of the deposition ratios compared with the emission ratios near large point sources is probably related to a rapid rate of atmospheric mixing relative to the rates of wet-deposition processes. Alter- natively, a preferential dry deposition of SO2 relative to that of NO2 may occur in the near-source region, or the differences may indeed represent a suppression of SO2 oxidation within a short distance from the emission sources. However, the general similarity of wet- deposition and emission ratios on an annual average basis suggests that the incorporation of sulfuric acid in rain is not retarded with respect to the conversion and incorporation of nitric acid in rain. A small body of data suggests that the historical trends in the molar deposition ratios of sulfate to nitrate correspond to the trend in molar SO KNOX emission ratios. The historical trends of the molar deposition ratio SO=~NO3 at Hubbard Brook and the molar ratios of so2 to NCk in emissions in the eastern United States by EPA Region are plotted in Figure 4.19. Also shown in the figure is the result of a linear- regression analysis of the Hubbard Brook deposition ratio against time. Between 1965-1966 and 1979-1980, the linear regression gives a slope of -0.039 (standard error 0.006) and an intercept of 1.444 (standard error 0.054). The significance level of the slope is 0.0001. Another perspective is provided by comparing the annually averaged regional wet-deposition rates with regional emissions. Integrating wet deposition over the northeast quadrant of the United States [using the isopleths of Pack (1980) based on MAP3S and SURE data} Golomb (1983) found sulfate deposition to be 4.3 x 1016 (+20 percent) moles/year and total nitrate deposition to be 3.9 x 101° (+25 percent) moles/year. Golomb calculated regional emissions from U.S./Canada (1981) to be 2.2 x 1011 moles/year of SCk and 1.8 x 1011 moles/year of NOX. The emissions were similarly distributed geographically. Golomb also indicates that 19 + 6 percent of the sulfur and 20 + 7 percent of the nitrogen emitted in this region are wet deposited there. It is interesting to note that Golomb found the molar ratio of sulfate to nitrate wet deposited annually in this region to be 1.1, compared with an annually averaged molar ratio of emissions for SO2 and NOX of

134 2.0 1 .9 _ 1.8 1.6 ., _ 1.1 . _ ___ _ _' 1 .0 = 0.9 _ 0.8 0.7 0.6 1960 1965 1970 IV · Hubbard Brook data Linear regression of Hubbard Brook data --Emissions 1 a Excludes Vermont. \\ -j'' -~ ~- 1975 1 980 Year FIGURE 4.19 Molar ratio of SO2 to NOX in emissions in the eastern United States by EPA Region and molar ratio of sulfate to nitrate in deposition at Hubbard Brook, New Hampshire, 1960-1978. SOURCE: Adapted from Tables 4.4 and 4.5.

135 1.2. The similarity of the ratios suggests that, whatever specific processes are taking place, there is a tendency on an annual basis for sulfate and nitrate to be deposited in the northeast quadrant of the United States with no significant loss of one component relative to the other. Two inferences result from this observation. On an annual basis there is no indication that within the region the oxidation of SO2 is retarded relative to the oxidation of NOx. Laboratory evidence, field obser- vations, and photochemical theory predict a relatively rapid gas-phase oxidation of NOx to HNO3, especially in summer (see Appendix A). If the rate of SO2 oxidation (as a result of both gas-phase and aqueous reactions) is comparable with that of NOy, then no significant retardation of sulfate formation is likely to occur within the eastern region of high acidity as a result of limited availability of gaseous or aqueous oxidants. On the basis of Golomb's calculation and our own analysis, we conclude, therefore, that (1) on an annual average basis there is no ~ ~. , , ~ _ . . evidence that the hypothesis of noane en a'. Amp applies in the northeastern United States and further that (2) the net oxidation of SO2 leading to sulfate in precipitation is as efficient as the net oxidation of NOk leading to nitrate in precipitation in this region. The first conclusion relates to a specific mechanism; the second to the core of the critical issue of non- linearity. If SC2 oxidation were limited by the availability of oxidant, it would be possible for SO2 emissions to overwhelm, or saturate, the capacity of the atmosphere to oxidize all or most of the SO2 within the region of eastern North America. Oxidation over the ocean and in remote land regions would presumably convert the remaining SO2. Since a large fraction of emitted S ~ could conceivably be in excess of the oxidant available in North America, a significant reduction in SO2 might not yield a significant reduction in wet deposition of sulfate in North America, including, of course, the regions deemed sensitive to acid deposition. The second conclusion, however, suggests that the oxidative capacity is not saturated in the northeast. Furthermore, the similarity of deposition ratios in eastern Canada both to those in the eastern United States and to emission ratios upwind also tend to support this implication. The indirect evidence agrees with the small amount of reliable North American historical data, which

136 / suggest that there is no evidence of a strong nonlin- earity in the relationship between SO4 deposition and SO2 emissions. The inference to be drawn from these data is that if the average ratio of SO2 to NOx in emissions in the region of high acidity covered by the data (consisting of the northeast quadrant of the United States, bounded by the southwest corner of Tennessee, the northwest corner of Illinois, New Hampshire, and the southeastern corner of North Carolina) were changed by changing SO2 emissions while keeping NOx emissions constant, that change should be reflected in a similar change in the ratio of sulfate to nitrate in wet deposition, all other emissions and conditions remaining unchanged. If dry deposition is linearly proportional to emissions, as suggested in the section on principal-component analysis, then the annual average ratio in bulk deposition in the area should also respond to changes in the emission ratio. The analysis is limited in its applicability to circumstances in which the spatial distribution of emissions is unchanged. Without greater confidence in the results of deterministic models, we cannot judge the consequences of emission reductions in a smaller region, such as the Midwest, for deposition in another region, such as the Adirondacks or southern Ontario (Chapter 3). It is now necessary to consider the uncertainties inherent in the application of these observations to the problem of nonlinearity. Our present inability to model the processes linking air composition with the composition of precipitation is revealed by a comparison of wet-deposition ratios with the molar ratios of SOx to NOx observed in air by the SURE network (Table 4.9). Precipitation scavenges airborne material both in and below clouds at some time after the injection of emissions into the atmosphere. Molar ratios in precipitation are expected therefore to reflect concentrations in ambient air. The ratios observed in ambient air at ground level in the SURE network, however, are not similar to those either in precipitation or in emissions. In ambient air at ground level, SOx is depleted with respect to NOx in the sum of gas and particle phases but is enriched in particles alone. The depletion of S°k in ambient air relative to NOx may be related to inherent nonlinearities in the processes involved, to differences in dry-deposition rates, to differences in the composition of air at ground and cloud 'evels, or to the comparison of ambient

137 TABLE 4.9 Molar Ratios of SOx to NOX in the Region of the SURE Experiment in 1977-1978a Emissions Ambient Airb Precipitation 1.2 (all sources) Gas and particles 0.66 0.73 to 1.3 2.2 (utilities)C Gas only 0.42 Particles only 24.0 SOURCE: Adapted from Mueller and Hidy (1983). aRatio taken as SO2/NO2. Sated at ground level. CWeighted heavily toward emissions injected at altitudes at or above 300 m. concentration data that are averaged over all meteoro- logical conditions with precipitation data that reflect only those conditions specifically favoring precipitation. As indicated earlier, the ratio of sulfate to nitrate in precipitation varies regularly throughout the year, being high in summer and low in winter (see Table 4.8). It is unclear how the specific nrn~=cmc -~1 1 ;~ _ = ~= ~_ _ ~ ~ ~ ~ ~ ~ ^ ~ l y Renege variations average out annually +^ h" =^ 1ln;.F~-m spatially and so similar to the emission ratios over much of the area when ambient atmospheric conditions apparently are so variable. Conceivably the uniformity of the annual SO ANON ratio in precipitation in the northeastern United States could reflect some as yet unidentified stoichiometric linkage between the atmospheric chemistry of SO2 and NO2. For example, the reaction of HOSO2 radicals formed in reactions of HO with SO2 could lead to HOSO2O2 and HOSO2O radicals, which could ultimately react preferentially with NC2 to form a one-to-one, S to N compound such as HOST ONO2 (see Chapter 3 and Appendix A). . This compound wood an principle hydrolyze in cloud water to form equal amounts of H2SO4 and HNO3 acids. However, this reaction mechanism would have to be the dominant ~, rim of on ~ _ HA x~ ~ ~ _ . . a..- ,~3 ~ ·~nca~n one ratio or moles of SO4 to moles of NC] near unity, as observe in nr"~;~;' =~;~- : eastern United States. rat A- -- ^~ ~= There is no laboratory or field evidence of the significant participation of nitryl sulfuric acid (HOSC2ONC2) or other similar compounds in acid development; in fact, the existing limited evidence appears to discount the possible importance of such a product as an intermediate in the HO-SC:-NOk reaction system (see Appendix A). From theoretical considerations (e.g., Chapter 2 and Appendix A), we expect that ambient concentrations of oxidants (such as H2O2, C3, and HO) will have

138 maximum values in summer and minimum values in winter. It is expected, therefore, that in winter in areas at higher latitudes, aqueous oxidation is slow because of low concentrations of oxidants in cloud water. Hence, if a limitation of oxidant were to lead to nonlinearity in the relationship between emissions and deposition, that effect would be most likely to be observed in winter. The evidence that the relationship between average annual emissions and deposition is not strongly nonlinear may result from the fact that most of the total annual deposition of acid occurs during the warm months. Comparison of emission ratios and deposition ratios constitutes indirect evidence for the absence of a serious nonlinearity even though we lack an adequate understanding of the intermediate chemical and meteoro- logical processes to predict, either qualitatively or quantitatively, the deposition ratios from the emission ratios. It is our opinion that the necessary data and theoretical understanding to model the effects of emissions from distant sources on the composition of precipitation reliably and accurately will not be available in the near future. The evidence based on the empirical observations appears incompatible with a seriously nonlinear system in terms of our current knowledge. Unquestionably, however, it would be preferable to reinforce such conclusions with a more complete understanding of atmospheric chemistry and meteorology under conditions prevalent in eastern North America. Additional uncertainties are introduced by the inherent variabilities of natural processes (e.g., annual rates of precipitation), imprecision of measurements, and errors in determining emission rates. For example, the ratios of emissions and wet depositions in the Northeast are uncertain by a factor of about 30 percent. The concen- trations of sulfate and nitrate in the MAP3S study are estimated to be uncertain by a factor of 15 to 20 percent. A significant reduction in the uncertainties related to natural variability, measurement imprecision, or establishment of emission rates is unlikely in the near future. FINDINGS AND CONCLUSIONS Although data from which to assess the relationships between emissions and deposition are relatively sparse in

139 North America, analysis of the available empirical data provides insight into the nature of the atmospheric processes involved in acid deposition. Nonlinearity Data indicate that the variability in ambient concentra- tions of SO2 vapor and SO4 particles do not necessarily correlate with the variability in SO2 emissions but are predominantly controlled by meteoro- logical conditions. Variations in sulfate particle concentrations tend to correlate with variations in SO2 vapor concentrations in rural areas of the Northeast. Regression on principal components and empirical orthogonal-function analysis suggest that if other factors were constant, ambient concentrations of SO2 and soi would be determined largely by patterns of SC2 emissions. Direct evidence of a strongly nonlinear relationship between wet deposition of sulfate and SO2 emissions is limited to extensive historical data in Europe. The continuous historical record of reasonably reliable data in North America, at Hubbard Brook, indicates no evidence for a strongly nonlinear relationship between annual depositions and annual emissions in the Northeast. Indirect evidence based on patterns of the ratio of SO4 to Ned in annual deposition do not support the hypothesis of a strongly nonlinear relationship between SO2 emissions and sulfate deposition in eastern North America. Differences in the relationship between emis- sions and deposition in Europe and eastern North America may be the consequence of differences in meteorology, latitude, or other factors, such as the spatial dis- tribution of sources, between the two regions. On the basis of currently available empirical data and within the limits of uncertainty associated with the data and with estimating emissions, we therefore conclude that there is no evidence for a strong nonlinearity in the relationships between long-term average emissions and depositions in eastern North America. The conclusion that nonlinearity is probably not sig- nificant for annual average deposition in eastern North America is clouded by three types of uncertainties. First, direct evidence based on long-term time-series data is severely limited in North America to only ten stations, all of which have collected bulk deposition

140 data. The data from only one site, Hubbard Brook, are considered to be reasonably reliable. Therefore we have relied on the historical record at only one station combined with the indirect evidence provided by data on sulfate and nitrate deposition compared with SO2 and NOX emissions. Second, there remain uncertainties in our detailed understanding of the meteorological, physi cal, and chemical processes that relate emissions to deposition. Third, the unknown influences natural vari- ability in the composition and occurrence of precipi- tation, imprecision in sampling and analysis, and uncertainties in estimation of emissions further limit confidence in this conclusion. Influence of Local and Distant Sources Both observational and theoretical evidence exists for the long-range transport of pollutants leading to acid deposition. It is apparent that any receptor site will be influenced to one degree or another by both local and distant sources. The issue of concern is the extent of this zone of influence for sensitive ecological areas, including the relative contributions to deposition of nearby and distant sources. In the case of the Hubbard Brook data, the trends in concentrations of sulfate and nitrate appear to reflect general trends in emissions. Analyses of the trajectories of precipitating systems delivering acidic precipitation to Whiteface Mountain and Ithaca in New York and south central Ontario in Canada indicate that most of the acidity in precipitation at these sites--as well as most of the precipitation--is associated with air masses that have passed over source regions to the south and southwest. The spatial distribution of the annual average molar ratios of pollutants in emissions and deposition suggest that atmospheric processes in eastern North America lead to a thorough mixing of pollutants over a wide geographic area, making it difficult to distinguish between the effects of distant and local sources. On the basis of currently available empirical data, we cannot in general determine the relative importance for the net deposition of acids in specific locations of long-range transport from distant sources or more direct influences of local sources. We regard the problem of relating emissions from a given region to depositions in

141 a given receptor region to be of primary importance and recommend that high priority be given to research relevant to its solution. The SURE data have indicated a relatively small zone of influence (of the order of 300 to 600 km) on ambient sulfate concentrations in general, with occasional long-range influence during ducting situations involving southwesterly flows of air. Similarly, warm frontal precipitation may involve cloud formation in air parcels adverted for hundreds of kilometers from a southwesterly direction. The relative contributions of such long-range effects and of more local regional effects are currently unknown and cannot be reliably estimated using currently available models. REFERENCES Allard, D.W., I.H. Tombach, H. Mayrsohn, and C.V. Mathai. 1982. Aerosol measurements: western regional air quality studies. Presented at the 75th Annual Meeting of the Air Pollution Control Association, New Orleans. Alpert, D.J., and P.K. Hopke. 1981. A determination of the sources of airborne particles collected during the regional air pollution study. Atmos. Environ. 15:675-688. Altshuller, A.P. 1980. Seasonal and episodic trends in sulfate concentrations (1963-1978) in the eastern United States. Environ. Sci. Technol. 14:1337-1348. Altshuller, A.P., and R.A. Linthurst, ed. 1982. Critical Assessment Document: The Acidic Deposition Phenomenon and Its Effects. Draft. Prepared for the U.S. Environmental Protection Agency. Raleigh, N.C.: The North Carolina State University Acid Precipitation Program. Atkinson, R., A.C. Lloyd, and L. Winges. 1982. An updated chemical mechanism for hydrocarbon/NO~/SO2 photooxidations suitable for inclusion in atmospheric simulation models. Atmos. Environ. 16:1341-1356. Beleley, D.A., E. Kuh, and R.E. Welsch. 1980. Regression Diagnostics: Identifying Influential Data and Sources of Collinearity. Pp. 85-191. New York: John Wiley and Sons. Chamberlain, J., H. Foley, D. Hammer, G. MacDonald, D. Rothaus, and M. Ruderman. 1981. The Physics and Chemistry of Acid Precipitation. Technical Report JSR-81-25. Arlington, Va.: SRI International.

142 - Core, J.E., P.L. Hanrahan, and J.A. Cooper. 1981. Air particulate control strategy development: a new approach using chemical mass balance methods. In Atmospheric Aerosol: Source/Air Quality Relationships, E.S. Macias and P.K. Hopke, eds. ACS Symposium Series No. 167. Washington, D.C.: American Chemical Society. Cunningham, W.C., and W.H. Zoller. 1981. The chemical composition of remote area aerosols. J. Aerosol Sci. 12:367-384. Department of Environmental Conservation. 1976. New York State Air Quality Report. DAR-77-1. Albany, N.Y.: New York State Department of Environmental Conservation. Department of Environmental Conservation. 1978. New York State Air Quality Report. DAR-79-1. Albany, N.Y.: New York State Department of Environmental Conservation. Department of Environmental Conservation. 1981. New York State Air Quality Report. DAR-82-1. Albany, N.Y.: New York State Department of Environmental Conservation. Dzubay, T.G. 1980. Chemical element balance method applied to dichotomous sampler data. In Aerosols: Anthropogenic and Natural, Sources and Transport, T.J. Kneip and P.J. Lioy, eds. Annal 338. New York: New York Academy of Sciences. Eichenlaub, V.L. 1979. Weather and Climate in the Great Lakes Region. South Bend, Ind.: The University of Notre Dame Press. Electric Power Research Institute. 1981. EPRI Sulfate Regional Experiment: Results and Implications. EPRI EA-2165-SY-LD. Palo Alto, Calif. Fisher, B.E.A. 1982. The transport and removal of sulphur dioxide in a rain system. Atmos. Environ. 16:775-784. Friedlander, S.K. 1973. Chemical element balances and identification of air pollution sources. Environ. Sci. Technol. 7: 235-24 0 . Galloway, J.N., and G.E. Likens. 1981. Acid precipitation: the importance of nitric acid. Atmos. Environ. 15:1081-1086. Gartrell, G. Jr., and S.K. Friedlander. 1975. Relating particulate pollution to sources: the 1972 California aerosol characterization study. Atmos. Environ. 9:279-300. Golomb, D. 1983. Acid deposition-precursor emission relationship in the northeastern U.S.A.: the effectiveness of regional emission reduction. Atmos. Environ. In press.

143 Gordon, G.E. 1980. Receptor models. Environ. Sci. Technol. 14:792-800. Granat, L. 1978. Sulfate in precipitation as observed by the European atmospheric chemistry network. Atmos. Environ. 12:413-424. Hansen, Do., and G.M. Hidy. 1982. Review of questions regarding rain acidity data. Atmos. Environ. 16:2107-2126. Hegg, D.A., and P.V. Hobbs. 1981. Cloud water chemistry and the production of sulfates in clol,~n Admix Environ. 15:1597-1604. ~ ~ · ~ ~_~11~ - Hegg, D.A., and P.V. Hobbs. 1982. Measurements of sulfate production in natural clouds. Atmos. Environ. 16:2663-2668. Heidam, N.Z. 1981. On the origin of the Arctic aerosol: a statistical approach. Atmos. Environ. 15:1421-1428. Heidam, N.Z. 1982. Atmospheric aerosol factor models, mass and missing data. Atmos. Environ. 16:1923-1932. Henderson, R.G., and K. Weingartner. 1982. Trajectory analysis of MAP3S precipitation chemistry data at Ithaca, New York. Atmos. Environ. 16:1657-1655. Henry, R.C., and G.M. Hidy. 1979. Multivariate analysis of sulfate and other air quality variables by principal components. Part I. Annual data from Los Angeles and New York. Atmos. Environ. 13:1581-1596. Henry, R.C., and G.M. Hidy. 1982. Multivariate analysis of particulate sulfate and other air quality variables by principal components. Part II. Salt Lake City and St. Louis. Atmos. Environ. 16:929-943. Henry, R.C., G.M. Hidy, P.K. Mueller, and K.K. Warren. 1980. Assessing sources, dispersion, chemical transformation and deposition of atmospheric sulfur over eastern North America by advanced Multivariate data analysis. Paper presented at the 5th International Clean Air Congress, Buenos Aires. International Union of Air Pollution Prevention Associations. Hidy, G.M., P.K. Mueller, and E.Y. Tong. 1978. Spatial and temporal distributions of airborne sulfate in parts of the United States. Atmos. Environ. 12:735-752. Huebert, B.J. 1982. Measurements of the dry-deposition flux of nitric acid vapor to grasslands and forest. Paper presented at the 4th International Conference on Precipitation Scavenging Dry Depositions and Resuspension, November 29 to December 3, 1982. Santa Monica, Calif.

144 Junge, C.E. 1958. The distribution of ammonia and nitrate in rain water over the United States. T'ans. ATn. Geophys. Union 39:241-248. Junge, C.E., and R.T. Werby. less. The concentrations of sulfur dioxide, sodium, potassium, calcium and sulphate in rainwater over the United States. Je Meteorol. 15:417-425. Kallend, A.S., A.R. Marsh, J.H. Pickles, and M.V. Proctor. 1983. Acidity of rain in Europe. Atmos. Environ. 17:127-137. King, W.J., and F.M. Vukovich. 1982. Some dynamic aspects of extended pollution episodes. Atmos. Environ. 16:1171-1182. Kowalazyk, G.S., G.E. Gordon, and S.W. Rheingrover. 1982. . . . Identification of atmospheric particulate sources in Washington, D.C., using chemical element balances. Environ. Sci. Technol. 16:79-90. Kurtz, J., and W.A. Schneider. 1981. An analysis of acidic precipitation in south-central Ontario using air parcel trajectories. Atmos. Environ. 15:1111-1116. Lazrus, A.L., P.L. Haagenson, G.L. Kok, B.J. Huebert, C.W. Kreitzberg, G.E. Likens, V.A. Mohnen, W.E. Wilson, and J.W. Winchester. 1983. Acidity in air and water in a case of warm frontal precipitation. Atmos. Environ. 17:581-591. Levine, S.Z., and S.E. Schwartz. 1982. In-cloud and below-cloud scavenging of nitric acid vapor. Atmos. Environ. 16:1725-1734. Likens, G.E., and T.J. Butler. 1981. Recent acidification of precipitation in North America. Atmos. Environ. 15:1103-1110. Likens, G.E., F. H. Bormann, and J.S. Eaton. 1980. Variations in precipitation and streamwater chemistry at the Hubbard Brook Experimental Forest during 1964-1977. Pp. 443-464 in Effects of Acid Precipitation on Terrestrial Ecosystems. T.C. Hutchinson and M. Havas, eds. New York: Plenum Lioy, P.J., and M.T. Morandi. 1982. Source~related winter and swnmer variations in S02, SO4 and vanadium in New York City for 1972-1974. Atmos. Environ. 16:1543-1550. MAP3S/RAINE Research Community. 1982. The MAP3S/RAINE Precipitation Chemistry Network: statistical overview for the period 1976-1980. Atmos. Environ. 16:1603-1631. Mardia, K.V., J.T. Kent, and J.M. Bibby. 1979. Multivariate Analysis. Pp. 213-246. London: Academic Press.

145 Miles, L.J., and K.J. Yost. 1982. Quality analysis of USGS precipitation chemistry data for New York. Atmos. Environ. 16:2889-2898. Mueller, P.K., and G.M. Hidy (principal investigators). 1983. The Sulfate Regional Experiment: Report of Findings. Report EA-1901~3). Palo Alto, Calif.: Electric Power Research Institute. Mueller, P.K., G.M. Hidy, K. Warren, T.F. Lavary, and R.L. Baskett. 1980. The occurrence of atmospheric aerosols in the northeastern United States. In Aerosols: Anthropogenic and Natural, Sources and Transport. T.J. Kneip and P. J. Lioy, eds. Annal 338, pp. 463-482. New York: New York Academy of Sciences. National Research Council. 1980. Controlling Airborne Particles. Washington, D.C.: National Academy Press. Niemann, B.L. 1982a. Proceedings of the American Chemical Society, Division of Environmental Chemistry, Symposium on Acid Precipitation, Las Vegas, Nevada, 29 March 1982. In press. Niemann, B.L. 1982b. Regional Relationships between H+, SO4, N ~ , ~ , Wet Depositions. Washington, D.C.: Office of Technology Assessment. Niemann, B.L. 1982c. The 1980 Data Set for Further Evaluation of Regional Air Quality/Acid Denosition Simulation Models. Washington, D.C.: U.S. _ _= ~ Environmental Protection Agency. Pack, D.H. 1980. Precipitation chemistry patterns: a two network data set. Science 208. 1143-1145. Peters, N.E., R. Schroeder, and D. Troutman. 1982. Temporal trends in the acidity of precipitation and surface waters of New York. U.S. Geol. Surv. Water Supply Paper 2188. Peterson, J.T. 1970. Distribution of sulfur dioxide over metropolitan St. Louis, as described by empirical eigenvectors and its relation to meteorological parameters. Atmos. Environ. 4:501-518. Pitchford, A.M., W. Malm, R. Flocchini, T. Cahill, and E. Walther 1981. Regional analysis of factors affecting visual air quality. Atmos. Environ. 15:2043-2054. Rahn, K.A. 1981a. Relative importance of North American and Eurasia as sources of Arctic aerosol. Atmos. Environ. 15:1447-1456. Rahn, K.A. 1981b. The Mn/V ratio as a tracer of large-scale sources of pollution aerosol for the Arctic. Atmos. Environ. 15:1457-1464. Raynor, G.S., and J.V. Hayes. 1982a. Variation in chemical wet deposition with meteorological conditions. Atmos. Environ. 16:1647-16S6.

146 Raynor, G.S., and J.V. Hayes. 1982b. Effects of varying air trajectories on spatial and temporal precipitation chemistry patterns. Water, Air, Soil Pollut. 18:173-190. Rodhe, H., P. Crutzen, and A. Vanderpol. 1981. Formation of sulfuric and nitric acid in the atmosphere during long-range transport. Tellus 33:132-141. Semonin, R.G. 1981. State Water Survey Division, Atmospheric Chemistry Section at the University of Illinois, SWS Contract Report, Vol. 252, p. 56. Shaw, R.W., and H. Rodhe. 1983. Nonphotochemical oxidation of SO2 in regionally polluted air during winter. Atmos. Environ. In press. Snedecor, G.W., and W.G. Cochran. 1967. Statistical Methods. 6th ed. Ames, Iowa: Iowa State University Press. Stensland, G.J., and R.G. Semonin. 1982. Another interpretation of the pH trend in the United States. Bull. Am. Meteorol. Soc. 63:1277-1284. Tanner, R.L., and B.P. Leaderer. 1982. Seasonal variations in the composition of ambient sulfur-containing aerosols in the New York area Atmos. Environ. 16:569-580. U.S./Canada. 1981. Memorandum of Intent on Transboundary Air Pollution, Interim Working Paper, 2-14. Washington, D.C.: U.S. Department of State. U.S./Canada Work Group t2. 1982. Atmospheric Sciences and Analysis. Final Report. H.L. Ferguson and L. Machta, cochairmen. Washington, D.C.: Protection Agency. U.S. Environmental U.S/Canada Work Group #3B. 1982. Emissions, Costs and Engineering Assessment. Final Report. M.E. Rivers and K.W. Riegel, Cochairmen. Washington, D.C.: U.S. Environmental Protection Agency. Vukovich, F.M., W. Bach, B. Crissman, and W. King. 1977. On the relationship between high ozone in the rural surface layer and high pressure systems. Atmos. Environ. 11:967-984. White, W., S. Heisler, R. Henry, and G.M. Hidy. 1978. The same-day impact of power plant emissions on sulfate levels in the Los Angeles air basin. Atmos. Environ. 12:779-784. Wilson, J.W., V.A. Mohnen, and J.A. Kadlecek. 1982. Wet deposition variability as observed by MAP3S. Atmos. Environ. 16:1667-1676.

147 Wisniewski, J., and J. Kinsman. 1982. An overview of acid rain monitoring activities in North America. Bull. Am. Meteorol. Soc. 63:598-618. Wolff, G.T., N.A. Kelly, and M.A. Ferman. 1981. On the sources of summertime haze in the eastern United States. Science 211:703-704. Wolff, G.T., N.A. Kelly, and M.A. Ferman. 1982. Source regions of summertime ozone and haze episodes in the eastern U.S.A. Water, Air, Soil Pollut. 18:65-82.

Next: 5. Research Needs »
Acid Deposition: Atmospheric Processes in Eastern North America Get This Book
×
Buy Paperback | $100.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!