Measurements of H2O2, aldehydes and organic acids in Los Angeles rainwater: Their sources and deposition rates

Rainwater samples were collected in Los Angeles, during 1985–1991 to determine concentration levels, sources and deposition rates of atmospheric H₂O₂, aldehydes and organic acids, in addition to major cations, anions and pH. Volume-weighted mean concentrations of H₂O₂, aldehydes (formaldehyde + acetaldehyde + glyoxal + methylglyoxal) and organic acids (formic acid + acetic acid) in rain collected at Westwood were 4.4., 3.9 and 16.5 μM, respectively, during the 6-year study period. Monocarboxylic organic acids were estimated to account for 27% (2–80%) of total free acidity (as on overall average) in rain collected at Westwood, whereas sulfuric acid and nitric acid accounted for 39% and 34% of the total acidity, respectively. Concentrations of aldehydes were strongly dependent on precipitation volume and decreased with increasing precipitation volume, whereas H₂O₂ and organic acids were only weakly dependent on precipitation volume. These results indicate that concentrations of aldehydes in rain are mainly controlled by dilution, whereas H₂O₂ and organic acid concentrations are controlled by other factors, such as decomposition of H₂O₂ by reacting with S(IV) and continuous aqueous formation/decomposition of organic acids by reactions involving aldehydes, dissolved OH radicals and H₂O₂. Principal component analyses indicate that aldehydes in rainwater mainly originate from gases and aerosols derived from anthropogenic sources, whereas the sources of H₂O₂ and organic acids in rain do not correlate with anthropogenic sources or marine and continental sources. There is good agreement between reported gas-phase concentrations of H₂O₂, aldehydes and organic acids in Los Angeles and calculated equilibrium concentrations of these chemical species from their rainwater concentrations and Henry's law constants. Temporal variations of concentrations of chemical species indicate that H₂O₂, aldehydes and organic acids were highest in the early afternoon. Summer rains contained the highest concentration of these chemical species, suggesting the photochemical activities during rain storms significantly affect their concentration levels. Estimation of annual rate of wet and dry depositions of H₂O₂, aldehydes and organic acids for the period studied, indicates that 84% of H₂O₂, 97% of aldehydes and 94% of organic acids, respectively, are annually scavenged from the atmosphere, by dry deposition, which is the dominant process for removal of these atmospheric pollutants in Los Angeles.     

Atmospheric H2O2 field measurements in a tropical environment: Bahia,Brazil

Concentrations of H₂O₂ in atmospheric gas and liquid phases were determined for the first time in the tropical Southern Hemisphere. Measurements were carried out in the Salvador area, Bahia, 13°S, 38°W) both at the seaside and 500 m away from it as well as at 270 km inland, in a rural area, during March–April 1988. Gaseous H₂O₂ was collected by cryogenic and rain by wet only sampling. Analyses were performed with the peroxyoxalate chemiluminescence method, employing a compact field apparatus. The measured gas phase concentrations ranged from 0.2 to 3.9 ppbv compared to 0.01-0.6 ppbv obtained from measurements with similar methodology in Dortmund (51°30'N, 7°30'E), F.R.G, during summer. The corresponding concentration ranges in rainwater are 0.9-6.8 ppmw (Bahia) and 0.1–2.2 ppmw (Dortmund, summertime). Gas/liquid H₂O₂ equilibrium during rain events is only attained at precipitation rates below 1 mm h⁻¹.     

Response of ozone to changes in hydrocarbon and nitrogen oxide concentrations in outdoor smog chambers filled with Los Angeles air

During the summer portion of the 1987 Southern California Air Quality Study (SCAQS), outdoor smog chamber experiments were performed on Los Angeles air to determine the response of maximum ozone levels, O₃(max), to changes in the initial concentrations of hydrocarbons, HC, and nitrogen oxides, NO_x. These captive-air experiments were conducted in downtown Los Angeles and in the downwind suburb of Claremont. Typically, eight chambers were filled with LA air in the morning. In some chambers the initial HC and/or NO_x concentrations were changed by 25% to 50% by adding various combinations of a mixture of HC, clean air, or NO_x. The O₃ concentration in each chamber was monitored throughout the day to determine O₃(max).An empirical mathematical model for O₃(max) was developed from regression fits to the initial HC and NO_x concentrations and to the average daily temperature at both sites. This is the first time that a mathematical expression for the O₃-precursor relationship and the positive effect of temperature on O₃(max) have been quantified using captive-air experiments. An ozone isopleth diagram prepared from the empirical model was qualitatively similar to those prepared from photochemical mechanisms. This constitutes the first solely empirical corroboration of the O₃ contour shape for Los Angeles.To comply with the Federal Ozone Standard in LA, O₃(max) must be reduced by approximately 50%. Several strategies for reducing O₃(max) by 50% were evaluated using the empirical model. For the average initial conditions that we measured in LA, the most efficient strategy is one that reduces HC by 55–75%, depending on the ambient HC/NO_x ratio. Any accompanying reduction in NO_x would be counter-productive to the benefits of HC reductions. In fact, reducing HC and NO_x simultaneously requires larger percentage reductions for both than the reduction required when HC alone is reduced. The HC-reduction strategy is the most efficient on average, but no single strategy is the optimum every day.     

A captive-air irradiation study of the response of nitric acid and peroxyacetyl nitrate to ozone control strategies in Los Angeles

Outdoor smog chamber experiments were used to study the sensitivity of the yields of two important nitrogen-containing pollutants, nitric acid (HNO₃) and peroxyacetyl nitrate (PAN) to changes in nonmethane hydrocarbon (HC) and nitrogen oxide (NO_x) concentrations in Los Angeles. The experiments were conducted at two sites in the Los Angeles Basin using eight chambers filled with morning Los Angeles air on 33 days. At least one chamber was unchanged and served as a control, while the initial HC and/or NO_x concentrations were changed by 25–50% in up to seven chambers to simulate O₃ control strategies and to broaden the range of HC - NO_x conditions studied. Empirical models that predict the maximum yields of HNO₃ and PAN were used to determine the response of these pollutants to three possible ozone control strategies. All three strategies (reductions in HC, NO_x or both HC and NO_x) reduced PAN while only NO_x reductions decreased HNO₃. However, reducing NO_x increased the HC reductions required to attain lower O₃ levels. Thus, there is a conflict between the O₃ and HNO₃ control strategies.     

Transported acid aerosols measured in southern Ontario

During the period 29 June 1986–9 August 1986, a field health study assessing the acute health effects of air pollutants on children was conducted at a summer girls' camp on the northern shore of Lake Erie in SW Ontario. Continuous air pollution measurements of SO₂, O₃, NO_x, particulate sulfates, light scattering, and meteorological measurements including temperature, dew point, and wind speed and direction were made. Twelve-hour integrated samples of size fractioned particles were also obtained using dichotomous samplers and Harvard impactors equipped with an ammonia denuder for subsequent hydrogen ion determination. Particulate samples were analyzed for trace elements by X-ray fluorescence and Neutron Activation, and for organic and elemental carbon by a thermal/optical technique. The measured aerosol was periodically very acidic with observed 12-h averaged H⁺ concentrations in the range < 10–560 nmoles m⁻³. The aerosol H⁺ appeared to represent the net strong acidity after H₂SO₄ reaction with NH₃(g). Average daytime concentrations were higher than night-time for aerosol H⁺, sulfate, fine mass and ozone. Prolonged episodes of atmospheric acidity, sulfate, and ozone were associated with air masses arriving at the measurement site from the west and from the southwest over Lake Erie. Sulfate concentrations measured at the lakeshore camp were more than twice those measured at inland sites during extreme pollution episodes. The concentration gradient observed with onshore flow was potentially due to enhanced deposition near the lakeshore caused by discontinuities in the meteorological fields in this region.     

Atmospheric pollutant deposition to high-elevation ecosystems

Current knowledge regarding deposition of atmospheric pollutants to mountain ecosystem is reviewed focusing on the mountains of eastern North America. Despite a general paucity of published data on the subject, some generalization emerge. Wet deposition (i.e. precipitation input) of SO₄²⁻, NO₃⁻, H⁺ and Pb tends to increase with elevation, primarily because of the orographic increase in precipitation amount. Cloud water deposition of these substances can be very significant for mountain forests, but is highly variable spatially because of its strong dependence on wind speed, cloud characteristics, and vegetation canopy structure, which are all heterogeneously distributed. Dry deposition has not been quantified sufficiently to draw empirical generalizations, but the processes involved are discussed with regard to expected elevational trends. Based on the few studies in which total annual deposition (wet, dry, plus cloud water inputs for an entire year) has been measured, it appears that some high-elevation sites in the Appalachian Mountains receive substantially more SO₄²⁻, NO₃⁺ deposition than do typical low-elevation sites. The amount of elevational increase depends largely on the amount of cloud water deposition at the mountain site. Data from two clusters of sites in the northern Appalachians indicate that total deposition of SO₄²⁻, NO₃⁻, and H⁺ to mountaintop sites is typically 3–7 times greater than deposition to nearby lowland sites. Similarly, some studies of Pb accumulation in organic soil horizons suggest a two- to four-fold increase from lowlands to mountaintops. Deposition in mountain areas can be highly variable over short distances because of the patchiness of meteorological conditions and vegetation canopy characteristics, and also because exposed trees and forest edges can receive deposition loads much higher than the landscape average. Night-time and early-morning O₃ concentrations are greater at high-elevation than at low-elevation sites. Daytime O₃ levels are equal or slightly higher at high-elevation sites. Additional studies are suggested which would allow better characterization of pollutant exposure along elevational gradients.     

A numerical experiment on the relative importance of H2O2 O3 in aqueous conversion of SO2 to SO42-

The importance of the three major aqueous reactions thought to be responsible for the in-cloud conversion of SO₂ to SO₄^2- was studied using the acidic deposition and oxidants model by supressing each reaction individually and all reactions simultaneously. The reactions are the oxidation of SO₂ by H₂O₂, or O₃ and catalytic oxidation by O₂ in the presence of Fe and Mn. The model simulations were 19–24 April 1981. It was found that SO₄^2- precipitation concentrations were generally more sensitive to H₂O₂ oxidation than to O₃ oxidation. The contribution of catalytic oxidation of SO₂ in the presence of Fe and Mn is insignificant everywhere and at all times. The contributions of H₂O₂ oxidation to SO₄^2- in precipitation is strongest in light precipitation areas while O₃ oxidation can be greater than H₂O₂ oxidation in heavy precipitation areas. The effect of supressing one reaction is mitigated by compensation through another mechanism. This is seem from the significant difference observed in the effects when individual suppressions were added together and when all reactions were suppressed simultaneously. From this, it is estimated that the contribution of aqueous oxidation of SO₂ to SO₄^2- in precipitation is approximately 50–80 per cent. Further simulations show that the relationship between SO₂ emissions and SO₄^2- production in the aqueous-phase through the oxidation reaction with O₃ is always non-linear in view of the pH dependence of the reaction rates.     

Gas- and aqueous-phase formic and acetic acids at a tropical cloud forest site

Atmospheric gas-phase and aqueous-phase (dew and fog) formic and acetic acids were measured over a cloud forest in Venezuela. The gaseous acids showed diurnal cycles, with higher mixing ratios during daytime. Higher concentrations were observed during the dry season (HCOOH 1.7 +/−0.5 ppb; CH₃COOH 1.4+/−0.6 ppb) in comparison with the rainy season (HCOOH 0.79+/−0.24 ppb; CH₃COOH 0.54+/−0.20 ppb). Liquid-phase concentrations in dew and fog are of the same order and range from 8.1 to 69.5 μM for HCOOH and 4.3 to 15.3 μM for CH₃COOH. The field-observed Henry's Law coefficients, calculated from the simultaneous measurements of gas- and liquid-phase acids, do not show a significant trend with the pH of the solution, in contrast to theoretical considerations. Dry deposition velocities to the nighttime dew are 1.1+/−0.6 and 0.68+/−0.42 cm s⁻¹ for formic and acetic acids, respectively. A loss of 0.054 ppb HCOOH and 0.022 ppb CH₃COOH from the atmospheric boundary layer to the dew is produced nightly.     

Processes affecting concentrations of aerosol strong acidity at sites in eastern England

Concentrations of aerosol strong acidity and related species have been measured at sites in eastern England using a sampler in which ammonia is pre-separated by a denuder. High concentrations occurred at a coastal site and were associated with air advected over the North Sea. At inland sites, ammonia concentrations were higher and the aerosol was more substantially neutralized. Daytime concentrations of aerosol H⁺ exceeded those measured at night, despite higher daytime levels of ammonia, presumably due to more effective production of H₂SO₄ during daytime hours. Concentrations of acidic aerosols were within the range 0–178 neq m⁻³, well below those observed at many eastern North American sites with lower concentrations of ammonia.     

Field studies of the SO2/aqueous S(IV) equilibrium in clouds

Concentrations of S(IV) were measured in cloudwater at Great Dun Fell and compared with theoretical HSO₃⁻ assuming equilibrium between aqueous and gaseous phases in cloud. Detectable concentrations of S(IV) in the range of 1 × 10⁻⁶ to 17.2 × 10⁻⁶ mol dm⁻³ were observed only in samples which contained low H₂O₂ concentrations, generally <1 × 10⁻⁶ mol dm⁻³. Concentrations of S(IV) were below the detection limit of 1 × 10⁻⁶ mol dm⁻³ in samples which contained hign H₂O₂ levels (1 × 10⁻⁶−80 × 10⁻⁶ mol dm⁻³) confirming that either SO₂ or H₂O₂ acts as the limiting reagent in the oxidation of SO₂ in cloudwater.Equilibrium HSO₃⁻ concentrations were estimated from the measured cloudwater pH, the gas phase SO₂ concentration and the ambient temperature and found to be on average about 5 times lower than the measured S(IV) concentrations. The possible role of formaldehyde in stabilizing S(IV) in cloudwater is discussed. The kinetic data available in the literature suggest that the complexation reaction between S(IV) and HCHO is too slow to account for the observed difference between measured and calculated S(IV) concentrations over the typical lifetime of clouds in our study.S(IV) accounted for up to 10% of the SO₄²⁻ measured in stored cloudwater samples.     

Studies on the formation of H2O2 in the ozonolysis of alkenes

The formation of H₂O₂ in the reactions of ozone with alkenes, isoprene and some terpenes has been studied with tunable diode laser absorption spectroscopy. The measured yields of H₂O₂ were found to be considerably enhanced in the presence of water vapour. H₂O₂ is thought to be formed in the ozonolysis of the alkene with O₃ by direct reaction of an intermediate with water vapour. The yield of H₂O₂ relative to the reacted alkene in the ozonolysis of trans-2-butene in the presence of water vapour was also studied with long path FTIR spectroscopy. Irrespective of the analytical methods and reaction conditions applied, the H₂O₂ yields in the reaction of O₃ with the different alkenes in the presence of water vapour were found to be in the range of a few per cent or less. Under the assumption that the reactive species forming H₂O₂ in the ozonolysis is the Criegee biradical, the overall rate constants for the reactions of some biradicals with water vapour were measured relative to the rate constant of the biradical with SO₂. For the H₂COO biradical a rate constant of (5.8 ± 2.5) × 10⁻¹⁷ cm³ s⁻¹ was determined and for the (CH₃)₂COO biradical (2.9 ± 1.5) × 10⁻¹⁷ cm³ s⁻¹; in the latter case with the assumption that (CH₃)₂COO reacts with SO₂ as fast as CH₂COO.     

Sulfur dioxide,hydrogen sulfide,total reduced sulfur,chlorinated hydrocarbons and photochemical oxidants in southern California museums

Indoor and outdoor concentrations of the air pollutants ozone, NO₂, SO₂, H₂S, total reduced sulfur (TRS), peroxyacetyl nitrate (PAN), methyl chloroform and tetrachloroethylene, have been measured at three southern California museums. Indoor maxima were 175 ppb for NO₂, 77 ppb for O₃, 0.7 ppb for PAN, 1.2 ppb for C₂Cl₄, >6.3 ppb for CH₃CCl₃, 2.5 ppb for SO₂, 1.4 ppb for TRS, and 46 ppt for H₂S. Indoor levels and indoor/outdoor (I/O) ratios for the chlorinated hydrocarbons pointed out to indoor sources. Outdoor and indoor levels of SO₂ and TRS were low at all three museums, but I/O ratios for SO₂ were high and averaged 0.89. H₂S concentrations were low, 16–46 ppt at one museum and less than 6 ppt at the other two museums. I/O ratios for the air pollutants with outdoor sources (ozone, PAN and NO₂) showed substantial variations, from low values of 0.02–0.33 at locations without influx of outdoor air to high values of 0.85–0.88 at locations experiencing high influx of outdoor air. Of the 10 institutions we have surveyed in southern California to date, eight exhibit high I/O ratios, e.g. 0.60–1.00 for PAN. Of the four museums surveyed that were equipped with HVAC and chemical filtration, only two yielded the expected low I/O ratios.     

Intensive studies of Sierra Nevada cloudwater chemistry and its relationship to precursor aerosol and gas concentrations

Measurements of inorganic aerosol and gas phase species are presented for three sites in central California during a 4 day period in April 1988. The measurement sites were located along an east-west transect at Visalia, Ash Mountain, and Lower Kaweah, with elevations of 90, 550 and 1900 m, respectively. Aerosol compositions were nearly neutral at all locations, however large concentrations of NH₃ at Visalia contributed significant excess alkalinity to the air mass sampled there. Concentrations of all major species were observed to decrease with elevation during most of the sampling periods. Concentrations at the upper two sites exhibited diurnal fluctuations, with peaks in the late afternoon, consistent with the transport of pollutants from San Joaquin Valley sources by daytime upslope winds. Concentrations of most of these species reached a maximum at the elevated sites on 28 April, as a weak cold front approached, reducing the atmospheric stability over the valley floor. Concentrations at Visalia on this day were somewhat lower than those observed earlier in the week.Clouds intercepting the mountain slopes on 28 April were sampled at two locations. The coudwater pH at both sites was observed to fall throughout the event, dropping as low as 4.34. Precursor concentrations of aerosol NO₃⁻, SO₄^2- and NH₄⁺, and gas phase HNO₃ and NH₃, were sufficient to account for the observed cloudwater loadings of NO₃⁻, SO₄^2- and NH₄⁺. In-cloud measurements made near the cloud base indicated a considerable S(IV) oxidation potential in the form of H₂O₂, but only low S(IV) concentrations. Cloudwater concentrations of formic acid were approximately three times acetic acid concentrations. Carbonyl concentrations were dominated by formaldehyde and glyoxal.     

Heterogeneous sulfate production in an urban fog

Heterogeneous production of sulfate in an urban fog has been investigated using data collected during the SCAQS program in the Los Angeles area, for the period of 10–11 December 1987. Fog was observed during the night of 10 December and the early morning hours of 11 December near the coast of southern California. Measurements at several sites (Hawthorne, downtown Los Angeles, etc.) indicated a significant increase in sulfate concentration during the afternoon of 11 December. Trajectory analysis suggests that these high sulfate concentrations were associated with the arrival at the receptor sites of air parcels that passed through the fog layer the previous night. To quantify the contribution of aqueous-phase processes to the above sulfate levels, a detailed trajectory model was employed to simulate the gas-phase processes during that episode. The model, using the available information about SO₂ and sulfate emissions, initial conditions and meteorology, successfully explained the sulfate levels in air parcels that did not pass through the fog layer, but underestimated by as much as 2.5 the sulfate levels of the trajectories through the fog. Sensitivity/uncertainty analysis indicated that the presence of sulfate beyond that attributable to gasphase chemistry (around 10μg⁻³) cannot be attributed to uncertainties in the model parameters (e.g. initial conditions, emissions, mixing heights, deposition velocities). The episode was then simulated using a full gas- and aqueous-phase chemistry model and the analysis indicated that heterogeneous sulfate formation in fog droplets under the conditions of the episode can indeed explain the observed sulfate.     

Spatial and temporal patterns in summertime sulfate aerosol acidity and neutralization within a metropolitan area

A study of sulfate aerosol acidity in Metropolitan Toronto was conducted during the summer of 1986. Fine-fraction aerosol (<2.5-μm) were collected using Teflon membrane filters and analyzed for major ionic species (H⁺, NH⁺₄, NO⁻₃, SO²⁻₄). Samples were collected for 6 weeks at three study sites: one in the Center City and the others 13 km (WNW) and 20 km (NE) away. There were very strong correlations among the three sites with respect to measured aerosol species (r² > 0.9 for 24-h data). However, spatial variations in the magnitude of aerosol acidity were observed during sulfate episodes. For example, the peak concentrations for all sites occurred on 25–26 July 1986. While the 24-h data for sulfate were quite uniform at the three sites (34, 34 and 35 μg m⁻³), H⁺ concentrations were 9.4, 8.3 and 6.0 μg m⁻³ (as H₂SO₄) for the NE, WNW and Center City sites, respectively. For most of the summertime episodes, the downtown area also had lower aerosol acidity compared to the two sites in suburban areas.     

Measurement of hydrogen peroxide and organic hydroperoxide concentrations during autumn in Beijing, China

Gaseous peroxides play important roles in atmospheric chemistry. To understand the pathways of the formation and removal of peroxides, atmospheric peroxide concentrations and their controlling factors were measured from 7:00 to 20:00 in September, October, and November 2013 at a heavily trafficked residential site in Beijing, China, with average concentrations of hydrogen peroxide (H₂O₂) and methyl hydroperoxide (MHP) at 0.55 ppb and 0.063 ppb, respectively. H₂O₂ concentrations were higher in the afternoon and lower in the morning and evening, while MHP concentrations did not exhibit a regular diurnal pattern. Both H₂O₂ and MHP concentrations increased at dusk in most cases. Both peroxides displayed monthly variations with higher concentrations in September. These results suggested that photochemical activity was the main controlling factor on variations of H₂O₂ concentrations during the measurement period. Increasing concentrations of volatile organic compounds emitted by motor vehicles were important contributors to H₂O₂ and MHP enrichment. High levels of H₂O₂ and MHP concentrations which occurred during the measurement period probably resulted from the transport of a polluted air mass with high water vapor content passing over the Bohai Bay, China.     

Removal of elemental mercury by photocatalytic oxidation over La2O3/Bi2O3 composite

La₂O₃/Bi₂O₃ photocatalysts were prepared by impregnation of Bi₂O₃ with an aqueous solution of lanthanum precursor followed by calcination at different temperatures. The composite materials were used for the first time for the photocatalytic removal of Hg⁰ from a simulated flue gas under UV light irradiation. The results showed that the sample containing 6 wt.% La₂O₃ and calcined at 500°C has the highest dispersion of the active sites, which was promoted by the strong interaction with the support (i.e., the formation of Bi-O-La species). Since they are fully accessible on the surface, the material also exhibits excellent optical properties while the heterojunction formed in La₂O₃/Bi₂O₃ promotes the separation and migration of photoelectron-hole pairs and thus Hg⁰ oxidation efficiency is enhanced. The effects of the various factors (e.g., the reaction temperature and composition of the simulated flue gas (i.e., O₂, NO, H₂O, and SO₂)) on the efficiency of the Hg⁰ photocatalytic oxidation were investigated. The results demonstrated that O₂ and SO₂ enhanced the efficiency of the reaction while the reaction temperature, NO, and H₂O had an inhibitory effect.     

Field investigations on the snow chemistry in central and southern california—II. Carbonyls and carboxylic acids

Snow samples from central and southern California were collected during the winter of 1987–1988 from there storms and analyzed for carbonyl compounds and carboxylic acids. Approximately 90% of the samples contained total aldehyde concentrations up to 40 μM. Formaldehyde and acetaldehyde were the dominant aldehydes observed; secondary aldehydes included glyoxal, methylglyoxal, and benzaldehyde. The highest aldehyde concentrations were observed in snow collected in areas where deciduous and coniferous forests are widespread. However, these aldehydes can be attributed also in part to primary and secondary products of anthropogenic activities. Formic and acetic were analyzed in all measured samples with concentrations ranging from 0.5 to 4.9 μM for HCOOH and from <0.3 to 13.4 μM for CH₃COOH. Maximum contribution of organic acids to precipitation-free acidity, calculated by assuming that the only sources of the measured formate and acetate were their respective acid forms, averaged 43.1% for samples with a pH⩽5. A consistent correlation between NH₄⁺ and acetate was found. [CH₃COOH] exceeded [HCOOH] in about 50% of the samples with the highest levels for CH₃COOH measured in cores collected from lower elevated locations adjacent to the Los Angeles basin. Results presented in this paper suggest that dry deposition and/or scavenging of carbonyl compounds and organic acids to snow may be important sinks for these compounds.     

Cloud chemistry at Mt Rigi,Switzerland: Dependence on drop size and relationship to precipitation chemistry

The chemical composition of winter and spring cloud water sampled at 1620 masl elevation on Mt Rigi in central Switzerland was dominated by NO₃⁻, SO₄²⁻, NH₄⁺ and H⁺. A wide range of concentration levels was observed, with maxima of 3700, 1800 and 4600 micronormal for NO₃⁻, SO₄²⁻ and NH₄⁺, respectively. Concentrations at a lower elevation (1030 masl) site on the mountain were higher due to lower cloud liquid water contents and higher pollutant levels at that site. The lowest pH observed was 2.95; large concentrations of NH₃ in the region prevented pH values from falling even lower. A comparison of simultaneously sampled cloud water and precipitation revealed much higher concentrations for most species in the cloud water, except in one case of extreme precipitation riming when the concentrations in the two phases converged. An exception to the pattern was H⁺; at times the precipitation was more acidic than the cloud water. The chemical composition of the cloud drops varied with drop size. Drops smaller than 10 μm diameter were enriched in NO₃⁻, SO₄²⁻ and NH₄⁺ relative to larger drops. Since the larger drops are the ones most effeciently captured by snow crystals, knowledge of their composition is essential to understanding the chemical implications of accretional growth of precipitation.