Changing Chemical Composition of Precipitation in Wilmington, North

Science Program, University of North Carolina Wilmington,. Wilmington, North ... like many other coastal regions has experienced explosive population ...
0 downloads 0 Views 141KB Size
Download PDF
Environ. Sci. Technol. 2006, 40, 5675-5680

Changing Chemical Composition of Precipitation in Wilmington, North Carolina, U.S.A.: Implications for the Continental U.S.A. JOAN D. WILLEY,* ROBERT J. KIEBER, AND G. BROOKS AVERY, JR. Department of Chemistry and Biochemistry and Marine Science Program, University of North Carolina Wilmington, Wilmington, North Carolina 28403-5932

The H+(aq) concentration in Wilmington, NC, precipitation has decreased by approximately 50% during the preceding two decades, similar to trends seen nationwide. The decrease in acidity is important because solution pH plays a key role in atmospheric reactions, and because the change is so large. This study presents the first long-range study of dissolved organic carbon (DOC) levels in precipitation which demonstrates that DOC concentrations have decreased by approximately half in Wilmington, NC, precipitation. The concentrations of H+(aq) and DOC are highly correlated primarily because small organic acids contribute to both DOC and H+(aq) in precipitation. Ammonium ion concentrations in precipitation have increased due to increased agricultural activities, and this also affects precipitation pH. The reduction of SO2 emissions in 1995 imposed by the Clean Air Act Amendment, better control of emissions of volatile organic compounds, and the increase in ammonia emissions all contribute to the decreasing H+(aq) in precipitation nationwide. These compositional changes in precipitation have many environmental implications, such as decreased acid deposition to lakes, changing speciation for trace metals in precipitation, increased ammonium deposition to coastal waters, and decreased DOC transport to the open ocean.

Introduction The concentration of sulfate and free hydrogen ions have decreased significantly in precipitation in the eastern United States during the past decade, as indicated by data from the National Atmospheric Deposition Program/National Trends Network (NADP/NTN) and from the Clean Air Status and Trends Network (CASTNet) (1, 2), most likely because of the 1990 Clean Air Act Amendment, which required improvements in emissions by 1995 (2). Precipitation collected on an event basis on the campus of the University of North Carolina at Wilmington (UNCW) since 1985 provides additional information for interpreting trends because many of the same components are measured in both programs (Cl-, SO42-, NO3-, H+, and NH4+), while more supporting data are available for Wilmington precipitation including dissolved organic carbon (DOC) and certain organic acids. Wilmington like many other coastal regions has experienced explosive * Corresponding author phone: (910)962-3459; fax: (910)962-3013; e-mail: [email protected]. 10.1021/es060638w CCC: $33.50 Published on Web 08/19/2006

 2006 American Chemical Society

population growth, a tripling of concentrated animal feeding operations, extensive deforestation, and increased hurricane and tropical storm activity since the onset of sampling. Extensive analyses were performed on Wilmington precipitation in 1988-1990 and 2001-2003, and so these two time periods are compared to look for temporal changes. Dissolved organic carbon is a ubiquitous rainwater component that has been measured in Wilmington, NC, for over a decade (3, 4). It is a major constituent of both marine and continental precipitation where it is present in concentrations greater than that of nitric and sulfuric acids combined. DOC is linked to the concentrations of hydrogen ions in rainwater because formic, acetic, and oxalic acids contribute to both DOC and hydrogen ion concentration in Wilmington precipitation (5) as well as to precipitation in both urban (6, 7) and remote areas of the world (8, 9). Understanding fluctuations in the concentrations of these organic acids is therefore useful in interpreting long-term variations in both the hydrogen ion and DOC concentrations in precipitation.

Experimental Section Sample Collection. Precipitation samples were collected on an event basis at the University of North Carolina at Wilmington in a large open area away from buildings (34o13.9′ N, 77o52.7′ W) approximately 8.5 km from the Atlantic Ocean. Due to the close proximity of the collection site to the laboratory, analyses were initiated within minutes of collection, which reduced the possibility of compositional changes between the time of collection and analysis. Since 1998, real time precipitation maps have been used to define the end of specific precipitation events to initiate the sampling process. Aerochem-Metrics (ACM) Model 301 automatic sensing wet/dry precipitation collectors were used to collect event precipitation samples. One ACM containing a 4 L muffled Pyrex glass beaker was used to collect samples for dissolved organic carbon analysis throughout this study. Samples for other analyses were collected from the high-density polyethylene (HDPE) container until 1997 when trace metal clean procedures were implemented. Since 1997, the remaining three ACM collectors have consisted of a HDPE funnel connected by Tygon FEP-lined tubing to a 2.2 L Teflon bottle. Storm Classification. Prior to 2000, storms were classified as continental (cold fronts, continental low-pressure systems), coastal (hurricanes, tropical storms, some Gulf Coast low-pressure systems, winter warm fronts, and El Nin ˜o precipitation), local thunderstorms, or mixed (stationary fronts) based on visual examination of daily surface weather maps for the week preceding the precipitation event. Since 2000 precipitation events have been categorized using airmass back-trajectories (version 4 of the Hybrid Single Particle Lagrangian Integrated Trajectory Model (HYSPLIT) developed at the National Oceanic Atmospheric Administration - Air Resources Laboratory). Trajectories were generated for a 72 h hind-cast starting at the 500 m level. Trajectories were used to identify end member continental and marine storms; many storms were classified as mixed (10).

Materials and Methods Inorganic Ions and pH. Anions (Cl-, NO3-, and SO42-) have been measured since 1986 using suppressed ion chromatography with a relative standard deviation (RSD) e 5%. Nonseasalt sulfate (NSS) was calculated using chloride concentrations assuming a constant seawater ratio of chloride VOL. 40, NO. 18, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5675

to sulfate (11). pH was determined using a Ross electrode calibrated with low ionic strength 4.10 and 6.97 buffers. Ionic strength adjuster (pHix from Orion Research Incorporated, Boston, MA) was added to each sample. Measurements were (0.01 pH units in the pH 4 range and (0.03 for samples with pH of 5 or above. Dissolved Organic Carbon. Dissolved organic carbon (DOC) was determined by high-temperature combustion using a Shimadzu TOC 5000 total organic carbon analyzer equipped with an ASI 5000 autosampler (4). Each sample was injected 4 times with a RSD e 3%. Results are reported as µM C L-1, where 1 µM C L-1 ) 12 µg C L-1. Organic Acids. Between 1987 and 1990, formic and acetic acids were measured by suppressed isocratic ion-exclusion chromatography using a Dionex HPICE AS5 column (9, 12). Between 1996 and 1998 formic, acetic, and oxalic acids were measured using suppressed gradient ion chromatography using a Dionex IonPacR AS11 4 mm analytical column (13). In 2000, the concentration of oxalic acid was determined using suppressed isocratic ion chromatography with a Dionex AS14A column. Preparation of standards and calibration procedures were similar for each sampling interval. The RSD varied from 3% to 15% over this time period. Hydrogen Peroxide. Hydrogen peroxide has been analyzed since 1992 using a fluorescence decay technique (14), with RSD ) 2%. Ammonium. During 1988-1990, ammonium was determined by suppressed ion chromatography with a RSD of 6%. During 2001-2003 study, NH4+ was determined using a modified version of a fluorometric method using o-phthaldialdehyde (OPA) with standard additions (RSD of 8%) (15).

FIGURE 1. A. Annual volume-weighted average hydrogen ion concentration in µM in Wilmington, NC, precipitation plotted versus year from 1985 through 2005. The linear regression line shown has a correlation coefficient of 0.665 (p < 0.001). The decreasing trend is significant at p < 0.005 (Mann-Kendall Trend Analysis (16)). The error bars represent (1 volume-weighted standard deviation. B. Annual precipitation amount in millimeters received in Wilmington, NC, plotted versus year from 1985 through 2005.

Results and Discussion The hydrogen ion concentration has decreased in Wilmington, NC, precipitation by approximately half over the preceding two decades in both coastal and continental precipitation (Figures 1A and 2) (Mann-Kendall Trend Analysis (16) p < 0.005 for each of the three data sets). The loss in H+ observed at the Wilmington site is not the result of dilution due to increases in precipitation amounts as there is no correlation between precipitation amount and annual hydrogen ion concentrations during this time period (n ) 21, r ) 0.177, p > 0.1), and the annual precipitation amount has not changed over this time period (n ) 21, r ) 0.060, p > 0.1, Figure 1B). The dissolved organic carbon (DOC) concentration in precipitation from Wilmington, NC, southeastern United States, also decreased more than 50% between 1995 and 2005 (Table 1 and Figure 3; Mann-Kendall Trend Analysis (16) p < 0.005). This data set is significant because it represents the only detailed time series study of DOC concentrations in precipitation. The concentrations of H+ and DOC are positively correlated (n ) 603, r ) 0.640, p < 0.001) in Wilmington precipitation (Figure 4). The intersect does not go through the origin because other components such as nitric, sulfuric, and carbonic acids are also important contributors of hydrogen ions in these precipitation samples. The change in precipitation hydrogen ion concentration is not limited to the Wilmington site. Hydrogen ion concentrations have also decreased at the 162 National Atmospheric Deposition Program (NADP/NTN) sites in the contiguous USA, and the 18 sites in the SEUSA that had complete data for both time periods (Table 2). The SEUSA sites include NADP/NTN sites in Al, GA, MS, NC, SC, and TN; this is the southeast region as described earlier (2) with the exception that FL is not included because of its much greater marine influence. The change in H+ with time can be explained in NADP/NTN precipitation by proton balance calculations using changes in nitrate, nonseasalt sulfate (NSS) 5676

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 18, 2006

FIGURE 2. Annual volume-weighted average hydrogen ion concentration in µM in Wilmington, NC, precipitation plotted versus year from 1985 through 2005 for continental (filled circles) and coastal (open circles) precipitation. The linear regression line shown for continental precipitation has a correlation coefficient of 0.704 (p < 0.001) and for coastal precipitation 0.547 (p < 0.02). The decreasing trends are both significant at p < 0.005 (Mann-Kendall Trend Analysis (16)). The error bars represent (1 volume-weighted standard deviation. (11), and ammonium concentrations using the following relationship:

∆H+ ) ∆NO3- + ∆NSS2- - ∆NH4+

(1)

The units for eq 1 are µeq L-1 as H+ where 1 µM NO3- is 1 µeq L-1 and 1 µM NSS is 2 µeq L-1. Nonseasalt sulfate in precipitation is the sulfate that does not come from sea spray. The signs for nitrate and NSS changes are positive because these ions are emitted to the atmosphere as acid precursors, whereas ammonia is a base that consumes protons. Chloride may have a small effect on proton balance calculations

TABLE 1. Comparison of Volume-Weighted Average Concentrations (µM) for Precipitation Collected at UNCW for the Periods of 1988-1990 and 2001-2003 (except Where Noted) and the Percentage Change and Change in µeq H+ L-1 as H+ Equiv between the Two Time Periodsf

H+ -

NO3 NSS NH4+ DOC oxalic acid sum H2O2 Cl-

1988-1990 µM

2001-2003 µM

change %

change µeq H+ L-1

31.0 pH ) 4.51 12.8 12.8 9.0 114b 1.8c

16.1 pH ) 4.79 12.5 12.1 11.7a 64 0.97d

-48

-14.9

-2 -5 +30 -44 -46

-0.3 -1.4 -2.7 -9.4 -1.3 -15.1 NR -0.07

9.6e 33.9

11.2 50.3

+17 +48

a 2002-2003. b 1995-1998. c 1996-1997. d 2000. e 1992-1994. The change in µeq H+ L-1 for dissolved organic carbon (DOC) assumes that 15% of the DOC is formic acid and 10% is acetic acid. The chloride change in µeq H+ L-1 was calculated assuming alkalinity change from seaspray predominates over the change in HCl in this coastal location. NSS ) nonseasalt sulfate, and NR ) not relevant. f

TABLE 2. Volume-Weighted Annual Averages (µM) for the Time Periods Indicated for (A) the 162 Contiguous USA and (B) 18 SEUSA NADP/NTN Sites That Had Complete Data for Both Time Periodsa 1988-1990 µM

2001-2003 µM

change %

change µeq H+ L-1

H+ field H+ lab difference NO3NSS NH4+ Cl-

25.4 20.8 4.6 16.6 14.4 13.5 8.5

(A) All USA 16.1 14.4 1.7 16.3 10.4 16.6 7.6

-37 -20 -63 -2 -28 +22 -11

-9.3 -6.4 -2.9 -0.3 -8.0 -3.1 < 0.1

H+ field H+ lab difference NO3NSS NH4+ Cl-

29.4 26.0 3.4 13.0 30.0 9.7 7.6

(B) SEUSA 20.4 19.1 1.3 12.1 22.7 11.5 6.3

-31 -27 -62 -7 -24 +19 -17

-9.0 -6.9 -2.1 -0.9 -7.3 -1.8 < 0.1

a The change between time periods is given in percent and in µeq H+ L-1. The chloride change in µeq H+ L-1 was calculated assuming alkalinity change is minimal and HCl is e10% of the total chloride. NSS ) nonseasalt sulfate.

FIGURE 3. Annual volume-weighted average concentration of dissolved organic carbon (DOC) in µM in Wilmington, NC, precipitation plotted versus year from 1995 through 2005. The linear regression line shown has a correlation coefficient of 0.771 (p < 0.01). The decreasing trend is significant at p < 0.005 (Mann-Kendall Trend Analysis (16)). The error bars represent (1 volume-weighted standard deviation.

FIGURE 4. Precipitation DOC in µM plotted versus hydrogen ion concentration in µM for Wilmington, NC, precipitation collected between 1995 and 2005 (n ) 603, R ) 0.640, p < 0.001). because some of the chloride (10%) may be emitted as HCl(g) (Table 2) and some as seaspray which contributes alkalinity (Table 1). The relationship between the decrease in hydrogen ion and the changes in the ions described above from the NADP/ NTN sites is presented in Figure 5. The data presented in Figure 5 indicate the change in hydrogen ion for the NADP/ NTN sites results from changes in nitric and sulfuric acids and ammonium ion concentrations. The sites in Figure 5

FIGURE 5. Measured change in volume-weighted hydrogen ion concentration (µM) between the time intervals 1988-1990 and 20012003 plotted versus ∆NO3- + ∆NSS - ∆NH4+ (µeq L-1) for the same time period for NADP/NTN and Wilmington sites (W). Site W1 was below the regression line for the NADP/NTN sites until the effects of organic acid changes were removed using eq 2 (horizontal arrow from W1 to W2). The NADP/NTN sites were Cl ) Clinton, NC, R ) Raleigh, NC, JC ) Jordan Creek, NC, L ) Lewiston, NC, K ) Kennedy Space Station, FL (the only NADP/NTN coastal site in the SE USA), SEUS ) average of 18 NADP/NTN sites in the southeast United States, and USA ) average of all 162 NADP/NTN sites; laboratory pH data was used. The line shown is the regression line for the NADP/NTN and Wilmington sites (excluding W1); the slope (1.35) is not significantly different than 1.0 (p < 0.05), the intercept (-0.23) is not significantly different than zero (p < 0.05), and the correlation coefficient is 0.980 (p < 0.001). include all NADP/NTN sites in eastern North Carolina (including Clinton in the center of the hog producing region), the Kennedy Space Station in Florida (the only NADP/NTN coastal location in the southeastern United States with complete data for the relevant time periods), the average value for the 162 contiguous NADP/NTN sites, the 18 SEUSA sites, and Wilmington. The analogous loss in hydrogen ion for Wilmington, NC, precipitation cannot be accounted for by eq 1, and the data do not fall on the line for the NADP/NTN data (the “W1” in VOL. 40, NO. 18, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5677

Figure 5). The change in inorganic ions depicted in Figure 5 for Wilmington, NC precipitation during the preceding 15 years is much smaller than the change in hydrogen ion concentrations for the same time period even though ammonium concentrations increased by 30% (Table 1). This ammonium increase likely results from the tripling of the swine population in eastern North Carolina since 1990 to over 10 million, with storage of hog waste in open lagoons (NC Department of Environment and Natural Resources). The hydrogen ion concentration change cannot be explained by changes in hydrogen peroxide concentrations, the major oxidant for sulfur dioxide to sulfuric acid (17, 18), which increased only slightly over this study time (Table 1). The increase in the addition of alkalinity from seasalt, as indicated by increased precipitation chloride concentrations, is also insufficient to explain the hydrogen ion decrease (Table 1). Chloride increased in Wilmington precipitation because during the first sampling period there was only one hurricane, whereas during the second, precipitation was received from six tropical storms or hurricanes, and these marine storms transport large quantities of seaspray via precipitation more than 100 km inland (19). The hydrogen ion concentrations decreased in Wilmington precipitation accompanied by decreases in DOC concentrations (Table 1 and Figure 3). Organic acids, primarily formic and acetic acid, make up approximately 25% of the DOC, and they contribute approximately 30% of the hydrogen ion concentration in Wilmington precipitation (13). Oxalic acid also contributes to DOC (4%) and hydrogen ion concentrations (8%) in Wilmington precipitation (5). The concentration of oxalic acid has decreased by 46% in Wilmington precipitation between 1996 and 2000; however, the ratio of oxalic acid to DOC has remained constant at 3% (Table 1). Formic and acetic acids do not affect laboratory pH measurements in NADP/NTN samples because these measurements were done weeks after collection during which time formic and acetic acids are lost due to biological degradation (5, 9, 12, 20). Formic and acetic acids do however affect pH measurements in Wilmington rain because these measurements have usually been made within a day after cessation of the rain event. The 50 µM decrease in DOC in Table 1 (114-64 µM) corresponds to a decrease of approximately 7 µΜ formic and 5 µM acetic acid using the proportions reported for Wilmington, NC, during 1995-1998, which was in the middle of our study and was the only time period when both organic acids and DOC were measured together (12). At precipitation pH, formic acid is almost completely dissociated (pKa ) 3.75), oxalic acid is roughly 75% dissociated (pKa1 ) 1.23 and pKa2 ) 4.19), and acetic acid is approximately half dissociated (pKa ) 4.75), resulting in an equivalent loss of hydrogen ion of approximately 10 µM which agrees well with the observed loss of hydrogen ion observed in precipitation since 1995 (Figure 1A). Because organic acids are contributing to the hydrogen ion concentration when the Wilmington precipitation pH is analyzed, they must be included in the proton balance considerations as follows, using units of H+ equivalents L-1:

∆H+ ) ∆NO3- + ∆NSS2- - ∆NH4+ + ∆formate + ∆acetate + ∆oxalate (2) When organic acids are taken into account, the ∆H+ change in Wilmington precipitation falls on the line with the NADP/ NTN sites, as indicated by the “W2” at the end of the horizontal arrow in Figure 5. The change in precipitation composition over time presented in Table 1 is not limited to Wilmington, NC. A similar change can be inferred indirectly for the contiguous USA by utilizing the NADP/NTN data. The concentration of 5678

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 18, 2006

hydrogen ion in precipitation at NADP/NTN sites was measured twice, once in the field upon collection (done on a weekly basis) and then several weeks later at a central analytical laboratory where other analyses are also performed. The laboratory pH is measured under more controlled conditions by the same analysts, whereas field measurements are made by many analysts and hence will be less reliable. Formic and acetic acids are readily degraded biologically in precipitation within days (5, 9, 12, 20, 21), so the lab pH values have lost the effect of these organic acids on pH and hence do not accurately reflect the pH of the rain when it fell. The field measurements have a variable loss of formic and acetic acids depending upon the number of days before the sample was collected and the ambient temperature. The biological consumption of these organic acids results in a loss of H+ (higher pH) in the laboratory measurements relative to those made in the field. The comparison between field and lab pH measurements while not perfect may be the only way to infer changes in organic acid content in rain over time, since there are no published long-term measurements for small organic acids in rain. The existence of formic and acetic acids in rain was in fact first detected by a comparison of lab and field pH measurements over 20 years ago (9). During 1988-1990 the volume-weighted average difference in field and lab measured hydrogen ion concentrations at these 162 NADP/NTN sites was 4.6 µM (Table 2), which corresponds to an organic acid content of approximately 6.1 µM (higher because not all of the acetic acid is completely dissociated). During 2001-2003 this difference decreased to 1.7 µM indicating an organic acid concentration of 2.3 µM (Table 2). This calculation is based on approximately 100 000 data points, which tends to average random errors in the field pH measurements. There is no spatial pattern in the distribution of this change, which ranged from a gain of 7 µM H+ to a loss of 13 µM with a median loss of 3 µM H+ for the 162 NADP/NTN sites. Similar changes were obtained when data for the 18 SEUSA sites were compiled (Table 2). A decrease of 3 µM H+ corresponds to a precipitation change of approximately 5 µM formic plus acetic acid carbon or an average minimum loss of 20 µM DOC. Oxalic acid does not have to be considered in this assessment because it is much more stable than formic and acetic acids in rain (4) and so was present during both field and laboratory pH measurements. The change in DOC in Wilmington precipitation is therefore not a local effect; this DOC concentration decrease has most likely occurred nationwide. The reasons behind the decreasing DOC concentrations presented in Figure 3 are most likely complex and many fold including better control of emissions of volatile organic carbons (VOCs) through modifications of gasoline composition, improvements in catalytic converter technology, and possibly local deforestation. One possibility for the decrease in DOC levels may involve the introduction of reformulated gasoline (RFG) which contains 2% oxygen and was designed to reduce emissions of volatile organic compounds (VOCs) and toxic pollutants (including benzene). Implementation of RFG has been mandated since 1996 in the northeastern United States which is important to the present study because Wilmington, NC, receives a large percentage (36%) of its rain from storms with a NE trajectory (Kieber et al., 2005). The US Environmental Protection Agency (EPA) estimates that use of reformulated gasoline has resulted in a 27% decrease in VOCs nationwide since 1996. Approximately 25% of atmospheric VOCs are formic, acetic, or oxalic acid precursors, including ethane, ethene, acetylene, formaldehyde, acetaldehyde, and acetone (22). Using the EPA estimate of VOCs from vehicles of 7.4 × 109 kg in 1996, 25% formic, acetic, and oxalic acid precursors, an estimated 50% oxidation and scavenging rate for these highly soluble gases (22), 0.9 m average annual precipitation amount (NADP/NTN data),

and a 27% decrease in VOCs, the decrease of formic, acetic, and oxalic acid concentrations in the U.S. precipitation would be approximately 3 µM since 1996, which is consistent with the changes observed in precipitation (Table 2). Although there are many uncertainties in this calculation and it most likely cannot account for all the decrease in DOC levels presented in Figure 3, it does provides an important order of magnitude estimate of the effect of RFGs on the composition of precipitation. Formic and acetic acid concentrations have varied during a 10-year period in Wilmington, NC, precipitation, most likely as a result of increasing traffic (13). The concentrations of acetic acid and formic acid in precipitation increased 92% and 35%, respectively, between 1987-1989 and 1996-1998, while traffic increased approximately 70% (13). Traffic is known to be a source of small organic acids, especially acetic acid, to the atmosphere in this region (23) as well as in other areas (24, 25). The 50 µM decrease in DOC observed in Wilmington precipitation between 1995 and 2004 in the current study suggests that acetic acid concentrations may have decreased to become closer to the concentrations observed in the 1987-1989 study. Implications. The profound changes in the chemical composition of precipitation in the continental United States over the last 15-20 years have significant environmental repercussions. The recent DOC decrease has significant implications to the global cycling of carbon particularly as it relates to marine systems because precipitation contributes significantly to open ocean DOC (4) and labile DOC to the coastal ocean (26, 27). More ammonium is now being transported in precipitation nationwide, potentially contributing to eutrophication in both freshwater and saltwater systems. The pH increase from approximately 4.51 to 4.79 (almost a 40% decrease in H+ concentration) means the equilibrium concentration of Fe3+free(aq) is only 15% as much as previously, based upon the solubility of the amorphous hydroxide (log Ksp ) -38). This is important because iron is a catalyst for many atmospheric redox reactions, including the oxidation of SO2 to H2SO4. Aluminum toxicity has been implicated in loss of freshwater fish in acidified lakes. The precipitation pH region of change is critical because the concentration of Al3+free(aq) in equilibrium with Al(OH)3(s) amorphous (Ksp ) -33.5) has decreased from approximately 9 µM to 1 µM, and fish toxicity estimates vary from 3 to 15 µM for the hydrated aluminum free ion. These trace metals occur in precipitation from the solubilization of wind blown soil dust; they can also be acquired in runoff water after contact with soils. Shallow water marine corals that use aragonite CaCO3 for their structures may benefit from the present day diminished pulses of acid from rainstorms, especially since these organisms are currently stressed because of increasing CO2 levels in the atmosphere. Similar decreases in hydrogen ion deposition are important for poorly buffered freshwater lakes, especially those in the northeast United States and eastern Canada which are downwind of significant pollutant sources. For example, a 1 m2 column of water in a 10 m deep lake with no alkalinity has 20 mmol H+(aq). Previously precipitation was adding 45 mmol H+(aq) m2- yr-1. Currently this amount is 23 mmol H+(aq) m-2 yr-1, versus a background for pH 5 precipitation of 15 mmol H+(aq) m-2 yr-1. These precipitation changes are important to the chemical reactivity of the troposphere and to the chemical composition of surface waters because solution pH plays such a key role in a wide range of processes, and the changes observed are of such a large magnitude.

Acknowledgments This research has been supported by eight different NSF grants (ATM-RUI) over this 20-year period. Sixty-four undergraduate students, 25 master’s students, and 3 postdoc-

toral fellows contributed to this project. Their individual contributions made this comprehensive work possible.

Literature Cited (1) Baumgardner, R. E., Jr.; Lavery, T. F.; Rogers, M. C.; Isil, S. S. Estimates of the atmospheric deposition of sulfur and nitrogen species: Clean air status and trends network, 1990-2000. Environ. Sci. Technol. 2002, 36, 2614-2629. (2) Lynch, J. A.; Bowersox, V. C.; Grimm, J. W. Acid rain reduced in eastern United States. Environ. Sci. Technol. 2000, 34, 940949. (3) Kieber, R. J.; Peake, B.; Willey, J. D.; Avery, G. B. Dissolved organic carbon and organic acids in coastal New Zealand rainwater. Atmos. Environ. 2002, 36, 3557-3563. (4) Willey, J. D.; Kieber, R. J.; Eyman, M. S.; Avery, G. B. Rainwater dissolved organic carbon: Concentrations and global flux. Global Biogeochem. Cycles 2000, 14, 139-148. (5) Tang, Y. Organic acids in North Carolina rainwater, Masters Thesis, University of North Carolina at Wilmington: Wilmington, NC, 1998; p 57. (6) Kawamura, K.; Steinberg, S.; Kaplan, I. R. Concentrations of monocarboxylic and dicarboxylic acids and aldehydes in southern California wet precipitations: Comparison of urban and nonurban samples and compositional changes during scavenging. Atmos. Environ. 1996, 30, 1035-1052. (7) Sakugawa, H.; Kaplan, I. R.; Shepard, L. S. Measurements of H2O2, aldehydes and organic acids in Los Angeles rainwater: their sources and deposition rates. Atmos. Environ. 1993, 27B, 203-219. (8) Andrea, M. O.; Talbot, R. W.; Andrea, T. W.; Harriss, R. C. Formic and acetic acids over the central Amazon region, Brazil. J. Geophys. Res. 1988, 93, 1616-1624. (9) Keene, W. C.; Galloway, J. N.; Holden, J. D. Measurement of weak organic acidity in precipitation from remote areas of the world. J. Geophys. Res. 1983, 88, 5122-5130. (10) Kieber, R. J.; Long, M. S.; Willey, J. D. Factors influencing nitrogen speciation in coastal rainwater. J. Atmos. Chem. 2005, 52, 8199. (11) Willey, J. D.; Kiefer, R. H. A contrast in winter rainwater composition: maritime versus continental rain in eastern North Carolina. Mon. Weather Rev. 1990, 118, 488-494. (12) Avery, G. B.; Willey, J. D.; Wilson, C. A. Formic and acetic acids in coastal North Carolina rainwater. Environ. Sci. Technol. 1991, 25, 1875-1879. (13) Avery, G. B.; Tang, Y.; Kieber, R. J.; Willey, J. D. Impact of recent urbanization on formic and acetic acid concentrations in coastal North Carolina rainwater. Atmos. Environ. 2001, 35, 3353-3359. (14) Kieber, R. J.; Helz, R. G. Two method verification of hydrogen peroxide determinations in natural waters. Anal. Chem. 1986, 58, 2312-2315. (15) Holmes, R. M.; Aminot, A.; Kerouel, R.; Hooker, B. A.; Peterson, B. J. A simple and precise method for measuring ammonium in marine and freshwater ecosystems. Can. J. Fish. Aq. Sci. 1999, 56, 1801-1808. (16) Helsel, D. R.; Hirsch, R. M. Statistical Methods in Water Resources, 3rd ed.; Elsevier Science Pub. Co.: 2000. (17) Calvert, J. G.; Lazarus, A.; Kok, G. L.; Heikes, B. G.; Walega, J. G.; Lind, J.; Cantrell, C. A. Chemical mechanisms of acid generation in the troposphere. Nature 1985, 317, 27-35. (18) Calvert, J. G.; Stockwell, W. R. Acid generation in the troposphere by gas phase chemistry. Environ. Sci. Technol. 1983, 17, 428443. (19) Willey, J. D.; Kiefer, R. H. Atmospheric deposition in southeastern North Carolina: Composition and quantity. J. Elisha Mitchell Sci. Soc. 1993, 109, 1-19. (20) Galloway, J. N.; Likens, G. E.; Keene, W. C.; Miller, J. M. The composition of precipitation in remote areas of the world. J. Geophys. Res. 1982, 87, 8771-8786. (21) Herlihy, L. J.; Galloway, J. N.; Mills, A. L. Bacterial utilization of formic and acetic acid in rainwater. Atmos. Environ. 1987, 21, 2397-2402. (22) Seinfeld, J. H.; Pandis, S. N. Atmospheric Chemistry and Physics; John Wiley and Sons Inc.: New York, 1998. (23) Willey, J. D.; Wilson, C. A. Formic and acetic acids in atmospheric condensate in Wilmington, North Carolina. J. Atmos. Chem. 1993, 16, 123-133. VOL. 40, NO. 18, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5679

(24) Grosjean, D. Organic acids in southern California air: Ambient concentrations, mobile source emissions, in situ formation and removal processes. Environ. Sci. Technol. 1989, 23, 1506-1514. (25) Kawamura, K.; Kaplan, I. R. Determination of organic acids (C1C10) in the atmosphere, motor exhausts and engine oils. Environ. Sci. Technol. 1985, 19, 1082-1086. (26) Avery, G. B.; Kieber, R. J.; Willey, J. D.; Shank, G. C.; Whitehead, R. F. Impact of hurricanes on the flux of rainwater and Cape Fear river water dissolved organic carbon to Long Bay, southeastern United States. Global Biogeochem. Cycles 2004, 18, 30153020.

5680

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 18, 2006

(27) Avery, G. B.; Willey, J. D.; Kieber, R. J.; Shank, G. C.; Whitehead, R. F. Flux and bioavailability of Cape Fear River and rainwater dissolved organic carbon to Long Bay, southeastern United States. Global Biogeochem. Cycles 2003, 17, 1402-1407.

Received for review March 17, 2006. Revised manuscript received June 29, 2006. Accepted July 21, 2006. ES060638W