Eaton.math.rpi.edu

Subscriber access provided by RENSSELAER POLYTECH INST Chemical Response of Lakes in the Adirondack
Region of New York to Declines in Acidic Deposition
Charles T. Driscoll, Kimberley M. Driscoll, Karen M. Roy, and Myron J. Mitchell Environ. Sci. Technol., 2003, 37 (10), 2036-2042• DOI: 10.1021/es020924h • Publication Date (Web): 11 April 2003
Downloaded from http://pubs.acs.org on March 24, 2009
More About This Article
Additional resources and features associated with this article are available within the HTML version: Links to the 17 articles that cite this article, as of the time of this article download Links to articles and content related to this article Copyright permission to reproduce figures and/or text from this article Environmental Science & Technology is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036 Environ. Sci. Technol. 2003, 37, 2036-2042
Chemical Response of Lakes in the
deposition (5). There are approximately 2770 lakes in theAdirondacks (>2000 m2 surface area). A survey of 1469 lakes Adirondack Region of New York to
during 1984-1987 found that 27% of these lakes werechronically acidic (acid-neutralizing capacity (ANC) <0 Declines in Acidic Deposition
µequiv L-1), and an additional 21% had summer ANC valuesbetween 0 and 50 µequiv L-1 and could experience hydrologicevents, which decrease ANC values near or below 0 µequiv C H A R L E S T . D R I S C O L L , * , † K I M B E R L E Y M . D R I S C O L L , † There have been marked changes in emissions of SO and atmospheric deposition of sulfate (SO 2-) and hydrogen ion (H+) in the United States since the early 1970s. Following Department of Civil and Environmental Engineering, the 1970 Amendments of the Clean Air Act (CAAA), emissions 220 Hinds Hall, Syracuse University, Syracuse, New York 13244, Adirondack Lakes Survey 2 in the United States peaked in 1973 at 28.8 Tg/yr and have declined 38% since that time (9). In contrast, emissions Corporation, NYSDEC, Ray Brook, New York 12977, andDepartment of Forest Biology, SUNY-ESF, 210 Illick Hall, of NOx in the United States peaked in 1990 (21.8 Tg/yr), but values have remained relatively constant since 1980. The1990 CAAA was the first legislation in the United States tospecifically address acidic deposition. Through Title IV ofthe 1990 CAAA, there will be a 13.2 Tg/yr cap in emissions Long-term changes in the chemistry of wet deposition of SO2 by 2010, in addition to resulting in a 1.8 Tg/yr reduction and lake water were investigated in the Adirondack Region in emissions of NOx from utilities than would be expected of New York. Marked decreases in concentrations of without the legislation. However, there is no cap on annualemissions of NO x. Therefore, emissions may increase with and H+ in wet deposition have occurred at two sites future increases in U.S. population and energy consumption.
since the late 1970s. These decreases are consistent Despite widespread acclaim of the cost-effectiveness of the with long-term declines in emissions of sulfur dioxide (SO2) 1990 CAAA (2), there have been several reports of severe in the eastern United States. Changes in wet NO - acidification of soil due to accelerated losses of calcium (Ca2+) deposition and nitrogen oxides (NOx) emissions have and magnesium (Mg2+) and limited recovery of acidic surface been minor over the same interval. Virtually all Adirondack Lakes have shown marked decreases in concentrations The Clean Air Act is due for reauthorization (as of 2000).
In addition, there is currently considerable debate over U.S.
, which coincide with decreases in atmospheric S air pollution and energy policies. In this regard, it is a useful in several Adirondack lakes. As atmospheric N deposition time to examine the most recent patterns in the recovery oflakes in the Adirondack Region in response to U.S. emissions has not changed over this period, the mechanism control programs. The Adirondack Long-Term Monitoring contributing to this apparent increase in lake/watershed Program (ALTM) was established in 1982 to assess seasonal N retention is not evident. Decreases in concentrations of and long-term patterns in the chemistry of lakes in the Adirondack Region of New York. The program was initiated neutralizing capacity (ANC) and pH and resulted in a shift with 17 lakes. It was expanded in 1992 with an additional 35 in the speciation of monomeric Al from toxic inorganic lakes for a total of 52 sites to improve representation of classes species toward less toxic organic forms in some lakes.
of lakes across the Adirondacks (Table 1). Here we report for Nevertheless, many lakes continue to exhibit pH values and the first time trends in the acid-base status for the entire concentrations of inorganic monomeric Al that are group of ALTM lakes and classes of ALTM lakes relative to critical to aquatic biota. Extrapolation of rates of ANC changes in wet deposition. Moreover, we extrapolated thesetrends to estimate time to chemical recovery of Adirondack increase suggests that the time frame of chemical recovery of Adirondack Lakes will be several decades if currentdecreases in acidic deposition are maintained.
Methods
In the Adirondacks, wet deposition has been monitored at
Introduction
two sites (the Huntington Forest (HF), 43°58′ N, 74°13′ W,and Whiteface Mountain (WM), 44°24′ N, 73°52′ W) as part The Adirondack Region of New York probably exhibits the of the National Atmospheric Deposition Program (NADP) most severe ecological impacts from acidic deposition of since 1978. The weekly precipitation collections are measured any region in North America (1). This large forested area for major ions using NADP protocols (13).
(24 000 km2) has long been an indicator of the response of The 52 ALTM lakes are sampled monthly, and the collected forest and aquatic ecosystems to U.S. policy on atmospheric waters are measured for major solutes (14, 15). The water- emissions of sulfur dioxide (SO2) and nitrogen oxides (NOx) sheds surrounding ALTM lakes are largely forested, with (2-4). Because of bedrock geology and generally shallow predominantly hardwood or mixed vegetation. One of the surficial deposits, the Adirondack Region is characterized by original ALTM lakes (Barnes Pond) and three of the recent soils with low pools of available nutrient cations and a large group of ALTM lakes (Woods Lake, Little Simon Pond, and number of lakes that are acidic or sensitive to acidic Little Clear Pond) were limed (i.e., calcium carbonateaddition) in the 1980s to mitigate surface water acidification * Corresponding author telephone: (315)443-3434; fax: (315)443- and therefore have been excluded from this analysis.
1243; e-mail: ctdrisco@syr.edu.
There is considerable variability in the response of lake ‡ Adirondack Lakes Survey Corporation.
ecosystems to acidic deposition. As a result, we previously have developed a classification system for the acid-base 2036 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 10, 2003
TABLE 1. Lakes of Adirondack Lake Term Monitoring (ALTM) Program, Including the Date Monitoring Was Initiated, Location,
Lake/Watershed Class, and Characteristics
surficial
location
a Record start 1982. b Record start 1992.
status of Adirondack Lakes, largely based on characteristics sheds containing calcite or with more than 25% of the of surficial geology and hydrologic flow paths (15, 16).
watershed with thick deposits of glacial till or stratified drift Drainage lakes situated in watersheds with predominantly are insensitive to acidic deposition. Two of the original and shallow deposits of glacial till (thin till watersheds; <5% of three of the entire group of ALTM lakes are in the thick till the watershed containing thick, i.e., >3 m depth, deposits of drainage class, and one of the original and two of the entire glacial till) are very sensitive to acidic deposition and are group have calcite in the watershed, for a total of five lakes typically chronically acidic (ANC <0 µequiv L-1). Eight of the considered insensitive to acidic deposition. Adirondack Lakes original and 26 of the entire group of ALTM lakes are in the also include mounded seepage lakes, which receive most of thin till drainage class. Lakes located in watersheds with their water directly from precipitation. One of the original intermediate deposits of glacial till (5-25% of watershed area and five of the entire group of ALTM lakes are mounded contains thick deposits of glacial till) generally have positive seepage lakes. In contrast, groundwater flow-through seepage but low ANC values and are susceptible to short-term lakes largely receive water from groundwater inflows and acidification associated with snowmelt or storm events. Four are relatively insensitive to acidic deposition. The ALTM of the original and 12 of the entire group ALTM lakes are in program does not have lakes in this class. Wetlands are an the medium till drainage class. Drainage lakes with water- important feature of the Adirondack landscape. Wetlands VOL. 37, NO. 10, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 2037
(-1.53 to -2.50 µequiv L-1 yr-1) across the region and strongly TABLE 2. Slopes of Significant (at p < 0.05) Changes in
suggests that decreases in SO2 emissions and atmospheric Concentration of Solutes in Wet Deposition at Huntington
deposition are responsible for this change. The rate of Forest and Whiteface Mountain (in µequiv L-1 yr-1) from
decrease was more rapid in lakes in the thin till drainage 1978 to 2000a
class (mean value -2.27 µequiv L-1 yr-1) as compared to the medium and thick till drainage classes (mean value -1.77µequiv L-1 yr-1). This difference may reflect less attenuation of atmospheric S deposition in shallow surficial materials.
This pattern suggests that lakes in the most sensitive and a CB is the sum of basic cations. Nonsignificant trends are indicated impacted drainage class (i.e., thin till) are the most responsive to controls in SO2 emissions. Similar decreases in concentra-tions of SO 2- were evident for the entire 48 ALTM lakes supply naturally occurring organic acids to surface waters.
sampled since 1992. Forty-four of the 48 lakes studied showed Each of the Adirondack lake classes is designated as a significant decrease in concentrations of SO 2- containing high or low concentrations of naturally occurring (One lake showed a significant increase in SO 2- organic acids on the basis of the concentrations of dissolved organic carbon (DOC; > or < 500 µmol of C/L, respectively; interval (1992-2000) was more variable (-4.93 to -0.80 µequiv L-1 yr-1) than observed for the lakes with the longer The nonparametric seasonal Kendall Tau (SKT) test was record, the mean rate of decline for those lakes with significant used to detect monotonic trends (generally increasing or decreasing trends was greater (-2.57 µequiv L-1 yr-1) than decreasing over time) in solute concentrations in precipita- that observed for the longer period (-2.06 µequiv L-1 yr-1).
tion and lake water (17). The tests were run for precipitation Similarly, the original 16 ALTM show a greater rate of SO 2- chemistry at HF and WM, the original 16 ALTM lakes (1982- decline since 1992 (-2.67 µequiv L-1 yr-1) than observed for 2000) that were not limed, and the entire 48 ALTM lakes that were not limed (1992-2000). The SKT test is a robust time- series procedure for data that are nonnormal and character- observed for the Adirondacks spans the range of values ized by seasonal patterns. This approach corrects data with reported previously for eastern North America. Mattson et moderate levels of serial correlation. We used p < 0.1 as our al. (20) observed that the average rate of SO 2- 300 streams in Massachusetts (-1.8 µequiv L-1 yr-1) wassimilar to our observed values for the Adirondacks. Stoddard Results and Discussion
et al. (11) conducted a regional analysis of trends in surface Trends in Atmospheric Deposition. Long-term changes in
water chemistry with respect to changes in atmospheric the chemistry of precipitation have been evident in recent deposition from the early 1980s to 1995. For the Adirondack years across the eastern United States (12, 18). NADP sites and Catskill Regions of New York, they reported lower rates in the Adirondacks have shown similar changes in the chemical composition of wet deposition (Table 2). Both HF particularly for the 1990s (-0.9 µequiv L-1 yr-1). Moreover, and WM have exhibited declines in concentrations of most major solutes, such that over the last 22 years decreases in compared to the early 1990s. They observed decreases in the sum of concentrations of strong acid anions (SO 2- + surface water concentrations of SO 2- in the 1980s and 1990s throughout eastern North America, including Maine/Atlantic Cl-) have greatly exceeded decreases in concentration Canada, Vermont/Quebec, South/Central Ontario, and Mid- western North America, with these regions all showing greater concentrations of hydrogen ion (H+). For HF, the pH of rates of decline in the 1990s than the 1980s.
precipitation has increased from 4.18 in 1979-1981 to 4.5 in 1998-2000. Similarly, the pH of precipitation at WM has not been consistent over the record. For the first time since increased from 4.1 (1979-1981) to 4.5 (1998-2000).
monitoring was initiated in 1982, many of the ALTM lakes The most conspicuous change in precipitation chemistry showed significant decreases in concentrations of NO - over the last 20 years has been marked decreases in the original ALTM lakes, 8 of the 16 sites exhibited a significant decrease in NO3 (p < 0.1; mean value -0.44 µequiv L-1 yr-1, by reductions in emissions of SO2 that have occurred over range -0.21 to -0.66 µequiv L-1 yr-1). Only the mounded the same period. Annual volume-weighted concentration of seepage lake (Little Echo Pond) had a small but significant at HF (r 2 ) 0.38) and WM (r 2 ) 0.58) were positively increase in concentrations of NO3 (0.01 µequiv L-1 yr-1; p 0.06). These trends of decreases in concentrations of NO3 area for the northeastern United States (Maine, Vermont, in the Adirondack Lakes are different than patterns reported New Hampshire, Massachusetts, Connecticut, Rhode Island, previously for the same lakes. Driscoll and Van Dreason (14) New York, New Jersey, Delaware, Maryland, Virginia, Penn- conducted time-series for the original 16 ALTM lakes from sylvania, Ohio, Indiana, Michigan, North Carolina, West 1982 to 1991 and reported that many (9 of 16) had a pattern Virginia, Illinois, Kentucky, and Tennessee based on 21-h back-trajectory analysis; 19). Unlike SO NO3 generally offset a pattern of decreasing SO4 relationship between emissions of NOx and precipitation trations, resulting in no change or in some cases decreases 3 . This lack of a relationship may reflect in surface water ANC (5 of 16 lakes) at that time. There was the fact that emissions of NOx have changed little over the some speculation in this and other papers (4, 21, 22) that long-term increases in NO3 indicate that forest watersheds Trends in Lake Sulfate and Nitrate. As observed for
are approaching a condition of N saturation with respect to patterns of wet deposition, there have been marked changes atmospheric inputs of N. This condition is thought to occur in the chemical composition of Adirondack Lakes in recent under elevated atmospheric N deposition and decreasing years. All of the original ALTM lakes have shown significant watershed retention of N associated with relatively mature (p < 0.05) decreases in concentrations of SO 2- forest ecosystems with a history of limited land disturbance with a mean rate of decline of 2.06 µequiv L-1 yr-1 (e.g., Figures (e.g., not previously in agriculture or severely burned). More 1 and 2). The range of this decline was remarkably uniform recently, Driscoll et al. (15) conducted a time-series analysis 2038 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 10, 2003
FIGURE 2. Mean rates of change in solute concentration in 16
lakes of the Adirondack Long-Term Monitoring (ALTM) program
from 1982 to 2000. Minimum, mean, and maximum changes in
concentrations and number of lakes showing significant trends are
shown. All values are in µequiv L-1 yr-1, except for concentrations
of inorganic monomeric aluminum (Ali), which is expressed in µmol
L-1 yr-1.
fertilization effect from increases in atmospheric CO2. Recentexperiments have shown that increases in atmospheric CO2cause increases in plant growth and N accumulation (25,26), possibly resulting in decreases in losses of NO - drainage waters. It seems likely that climate could be a strongdriver controlling N retention and loss in Adirondackwatersheds (e.g., ref 27). Note that long-term declines in lakeconcentrations of NO - FIGURE 1. Concentrations of SO 2-
(a), NO3 (b), acid-neutralizing
drainage lakes, which may be suggestive of climatic controls.
capacity (ANC; c), pH (d), dissolved organic carbon (DOC; e), and
Lawrence et al. (28) evaluated long-term patterns in the monomeric Al (f) in Big Moose Lake. A significant trend is indicated
hydrology of the Adirondack Region, showing generally high by a line.
discharge during the early 1990s and conditions of lowdischarge in the late 1990s. This trend in hydrologic condi- of the original 16 ALTM lakes over the period 1982-1997, tions may be partially responsible for the observed declines finding essentially no long-term trends in lake NO - in lake concentrations of NO - through the 1990s.
15-yr interval, the increases in lake NO - 1980s had changed such that trends were no longer signifi- As concentrations of both SO4 and NO3 were decreasing cant. Our most recent analysis shows this pattern of NO - concentrations has essentially completely reversed from decreases as well. For the original 16 ALTM lakes, all sites previous analyses (e.g., Figure 1). This trend is confirmed from time-series analysis for the complete group of ALTM with a mean value of -2.31 µequiv L-1 yr-1. For the entire lakes for the interval 1992-2000. Sixteen of the 48 lakes set of ALTM lakes, 41 of the 48 sites showed significant showed significant changes in concentrations of NO - NO3 (p < 0.1) with a mean value of 0.1), with 15 showing decreasing trends. Although 8 yr is a 3.26 µequiv L-1 yr-1. One lake (East Copperas Pond) relatively short period to conduct time-series analysis, particularly for a solute that is so inherently variable and Trends in Lake Basic Cations. In soil-water systems,
concentrations of basic cations generally respond to changes the 1990s was generally a period of decreasing concentrations in concentrations of strong acid anions (e.g., SO 2- + through the displacement of cations from cation-exchange It is not clear why some Adirondack watersheds are sites (29). We have observed a near stoichiometric cor- retaining N to a greater extent than was observed in the 1980s.
in CB (Figure 2). For the original 16 ALTM lakes, all exhibited would be expected if the Adirondacks were approaching a significant declines in CB (p < 0.05; mean rate -2.32 µequiv condition of N saturation. As discussed above, there has not L-1 yr-1) except the mounded seepage lake Little Echo Pond.
been any appreciable change in emissions of NOx or The rate of CB decline was somewhat greater for lakes in the medium and thick till drainage classes (mean value -2.52 in 1982. Using the model PnET-CN to gain insight, Aber and µequiv L-1 yr-1) as compared to the acidic lakes in the thin co-workers (23, 24) observed that long-term (∼30 yr; from till drainage class (mean value -2.22 µequiv L-1 yr-1). This 1963 to the early 1990s) patterns in stream NO - difference can be attributed to high rates of decline in Hubbard Brook Experimental Forest in New Hampshire were concentrations of inorganic monomeric Al and H+ that have largely explained by long-term climatic patterns and minor occurred in the low ANC lakes, which help balance the decline disturbance events (e.g., insect defoliation). However, the NO3 (see below). Note that all of the individual basic cations (i.e., Ca2+, Mg2+, Na+, and K+) had highly the 1990s could not be explained by climate. Using the model, significant decreasing trends, except for Na+ concentrations these investigators speculated that the long-term decreased in Little Echo Pond. Although all individual basic cations was associated with increased retention of N due a have shown decreasing concentrations, the overall decrease VOL. 37, NO. 10, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 2039
in CB was largely due to decreases in Ca2+ (mean rate ofdecline -1.30 µequiv L-1 yr-1). Since 1992, 26 of the 48 ALTMlakes have also shown significant decreases in CB (p < 0.1)coinciding with decreases in SO 2- + medium and thick till lakes (mean value -4.64 µequiv L-1yr-1) showing a greater rate of decline than the thin tilldrainage lakes (mean value -2.96 µequiv L-1 yr-1).
Inputs of CB to forest watersheds is largely derived from weathering supply and atmospheric deposition, with weath-ering generally dominating (30). In addition to these inputs,surface waters losses of CB may originate from changes inecosystem pools, such as net mineralization of soil organicmatter or the net displacement from the soil exchangecomplex. It has generally been assumed that the declines inconcentrations of CB observed in low ANC waters are due todecreases in the leaching of exchangeable cations corre-sponding to declines in SO 2-+ FIGURE 3. Time for lakes to reach acid-neutralizing capacity (ANC)
values of 50 µequiv L-1 as a function of ANC value in the year 2000.
NO3 are consistent with this process being the major These values are extrapolated assuming a linear rate of change
mechanism responsible for declines in lake concentrations based in slope of ANC change from time-series analysis. The
of CB (29, 31). However, patterns in the original 16 ALTM extrapolation was done for two intervals, 1982-2000 and 1992-
lakes may suggest that there also may be a long-term decline 2000. Six lakes were evaluated for the longer record, and 28 lakes
in weathering inputs to these watersheds. Because of limited were evaluated for the shorter record. Note that rates of ANC
interactions with vegetation and soil exchange surfaces, increase were generally greater when calculated over the later
investigators have used concentrations of Na+ and H4SiO4 as interval, so the time to reach 50 µequiv L-1 is shorter. Lakes with
indicators of weathering inputs (10, 31, 32). We observed g50 µequiv L-1 in 2000 or not showing a positive trend in ANC are
small but significant decreases in lake concentrations of both not represented here.
Na+ (15 of 16 lakes (p < 0.1), with a mean rate of decline of0.36 µmol L-1 yr-1) and H4SiO4 (7 of 16 lakes exhibited had mean ANC values <50 µequiv L-1; including 10 lakes decreases (p < 0.1) with a mean rate of decline of -0.96 µmol with ANC values <0 µequiv L-1.
There is considerable policy interest in rates of ANC Trends in Lake ANC and pH. Of particular interest is the
increase in response to decreases in acidic deposition to long-term change in ANC of Adirondack Lakes. Previous quantify the time scale of recovery of surface water acidifica- studies, including those involving ALTM lakes, have shown tion. To date, acidification models have been used to estimate little response of ANC to decreases in acidic deposition or changes in surface waters chemistry in response to antici- pated future emissions and atmospheric deposition and rates (4, 10-12, 14, 15). Our analyses indicate that 7 of the 16 of chemical recovery (e.g., refs 12 and 33). However, estimates original ALTM lakes have had significant increases in ANC of the rate and extent of recovery are tenuous because of (p < 0.1; Figure 2) from 1982 to 2000. One lake, West Pond, uncertainty in (i) future atmospheric emissions of SO2, NOx, significantly decreased in ANC. Note that 23% of the NH3, and other materials (e.g., CB) and relationships between watershed area of West Pond is wetlands and that lake water changes in emissions and deposition; (ii) the response of is characterized by elevated concentrations of dissolved watershed processes to changes in atmospheric deposition organic carbon (DOC; mean value 667 µmol of C /L). The (e.g., cation exchange, weathering, mineralization of soil S mean rate of ANC increase for those lakes showing a and N pools); and (iii) climatic and land disturbances that significant increasing trend was 0.78 µequiv L-1 yr-1, with a may occur in the future and alter the acid-base status of soil range from 0.42 to 1.54 µequiv L-1 yr-1. Most of the lakes (i.e., and surface waters. To try to provide bounds on the time 5) showing significant increases in ANC were in the thin till scale of chemical recovery of Adirondack Lakes, we used drainage class. Note that the mounded seepage lake, Little linear rates of ANC increase obtained from time-series Echo Pond, which receives water largely from direct pre- analysis to extrapolate the time it would take for lakes with cipitation inputs, had by far the greatest rate of ANC increase ANC <50 µequiv L-1 to reach a value of 50 µequiv L-1. The (1.54 µequiv L-1 yr-1) of all the sites studied. For the entire results of this extrapolation suggest that lakes with low ANC group, 29 of the 48 ALTM lakes had significant trends of values that are susceptible to episodic acidification (0-50 increasing ANC (p < 0.1) for the period 1992-2000. Twenty- µequiv L-1) will reach the 50 µequiv L-1 values over the period one of the 26 thin till drainage lakes exhibited increases in ranging from a few years to approximately 50 yr (Figure 3).
ANC. This pattern of increasing ANC has never been For lakes that are chronically acidic (ANC <0 µequiv L-1), the previously reported for large numbers of Adirondack Lakes.
time period to reach an ANC of 50 µequiv L-1 was estimated The mean rate of ANC increase for lakes showing a significant between 25 and 100 yr. Note that of the original ALTM lakes, trend over the 1992-2000 interval was 1.60 µequiv L-1 yr-1.
10 had ANC values <50 µequiv L-1 and that 6 of these This recent increase in ANC can be attributed to the fact that exhibited a significant trend of increasing ANC. Of the entire and NO3 concentrations have been decreasing, group of ALTM lakes, 39 had ANC values <50 µequiv L-1, and resulting in a marked rate of decline in the sum of strong of these, 28 showed a significant increase in ANC. Hence, a acid anions. If watershed retention of N should decrease in large fraction (∼30-40%) of ALTM lakes with ANC values the future, as observed in the 1980s, then NO - 50 µequiv L-1 have shown no change in ANC in recent could increase and limit increases in ANC. Despite recent years or were decreasing in ANC. This coarse calculation improvements, ANC values remain at levels of concern for must be considered with caution. It assumes the ANC aquatic biota in the majority of lakes in the study. An ANC increases are maintained at a constant linear rate for a period value of 50 µequiv L-1 has been used as a threshold to indicate that extends to recovery (i.e., 50 µequiv L-1). Indeed data chemical conditions under which aquatic organisms are from the original ALTM sites show that changes in ANC values largely protected from the effects of surface water acidification have been variable over the measurement period (see Figure from atmospheric deposition (12). In 2000, 34 of the 48 lakes 1). For example, Big Moose Lake is chronically acidic. The 2040 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 10, 2003
rate of ANC increase over the monitoring period (1982- the 16 lakes. One lake had a significant decrease in estimated 2000) was 0.79 µequiv L-1 yr-1. Assuming a linear extrapolation concentrations of organic anions. The mean increase in of this rate, Big Moose Lake is expected to reach ANC ) 50 organic anion concentration for lakes with significant µequiv L-1 in 46 yr. In contrast, the rate of ANC increase for increasing trends was 0.56 µequiv L-1, with values ranging Big Moose Lake over the more recent 1992-2000 period was from 0.37 to 1.03 µequiv L-1. Note, this rate of increase in 1.44 µequiv L-1 yr-1, resulting in a time to reach ANC ) 50 estimated concentrations of organic anions is considerably µequiv L-1 of 25 yr. Despite problems and uncertainty lower than the observed declines in SO 2- + associated with these estimates, it appears that at current The organic acid model also suggests that some of the rates of change in acidic deposition it will be several decades functional groups associated with naturally occurring organic before chronically acidic lakes in the Adirondacks will solutes are strongly acidic. The mean increase in DOC approach chemical conditions to alleviate acidification stress concentration would result in a 0.36 µequiv L-1 yr-1 loss in to aquatic biota. This length of time to reach chemical ANC because of the dissociation of strongly acidic organic recovery is comparable to estimates based on model functional groups. This rate of ANC loss for those lakes calculations for chronically acidic surface waters in the exhibiting DOC increases is clearly significant in comparison Northeast (12, 33). Note that, once chemical stress is to observed rates of ANC increase for the region.
mitigated, there will be additional delays in the recovery of An important consequence of acidic deposition is the aquatic biota (12, 34).
mobilization of Al from soil resulting in elevated concentra- Our analysis showed significant (p < 0.1) decreases in tions of inorganic species in surface waters that may be toxic concentrations of H+ in 9 of the 16 original ALTM lakes. One to aquatic biota (40, 41). Paradoxically, 9 of the 16 original lake, West Pond, exhibited a significant increase in H+. Not ALTM lakes had increases in concentrations of monomeric surprising, rates of H+ decrease were highly variable. Lakes Al (Alm), despite decreases in concentrations of SO 2- + that are chronically acidic or have low ANC values in the thin with one lake (Big Moose Lake) exhibiting decreasing till drainage class such as Big Moose Lake (-0.33 µequiv L-1 concentrations. This unexpected result was due to increases yr-1), Constable Pond (-0.20 µequiv L-1 yr-1), and Squash in concentrations of the organic fraction of monomeric Al Pond (-0.58 µequiv L-1 yr-1); the perched seepage lake Little (Alo) in 13 of the 16 lakes (mean rate of increase 0.07 µmol Echo Pond (-0.94 µequiv L-1 yr-1) had the highest rates of L-1 yr-1). Eight of the 16 lakes showed significant trends in H+ decrease. Over the shorter record, 18 of 48 lakes had concentrations of inorganic monomeric Al (Ali). Three lakes significant decreases in H+, while two lakes (West Pond and in the thin till drainage class with low values of ANC (Big Sunday Pond) showed a significant increase in H+. Similarly, Moose Lake, Darts’ Lake, and Otter Lake) showed the highest pH was shown to be increasing in small (mean 0.01 pH unit/ rates of Ali decrease (-0.14, -0.09, and -0.04 µmol L-1 yr-1, yr) but significant increments in 8 of the 16 original lakes, respectively). West Pond had significant increases in Ali (0.07 with West Pond pH decreasing at 0.02 pH unit/yr. Since 1992, µmol L-1 yr-1), consistent with observed decreases in pH a significant increase has been evident in 20 of 48 ALTM and ANC. The other four lakes showed low rates of change lakes (p < 0.1), with two lakes decreasing. Note, however, in in concentrations of Ali (i.e., <0.02 µmol L-1 yr-1). Over the 2000, 23 lakes still had mean pH values <5.5, including 13 more recent period, the entire group of ALTM lakes exhibited a somewhat different pattern. Twenty of the 48 lakes showed Trends in Lake Dissolved Organic Carbon and Alumi-
a significant change in concentrations of Alm; one lake with num Speciation. One of the more intriguing patterns
increasing concentrations and 19 with decreasing concen- observed in this time-series investigation was changes in trations. As with the longer record of original ALTM lakes, concentrations of DOC. Eight of the original ALTM lakes six sites exhibited increases in Alo. Twenty-eight lakes had exhibited changes in DOC, with concentrations increasing decreases in Ali, with a mean rate of decline of -0.31 µmol in seven lakes. The mean rate of DOC increase in those lakes L-1 yr-1. As observed for the longer record, the thin till showing significant increases was 6.6 µmol of C L-1 yr-1. In drainage class of lakes with chronically acidic conditions general, the rate of DOC increase was more rapid at higher exhibited the highest rates of decreases in Ali. The marked lake DOC concentrations (increase in DOC (in µmol of C L-1 extent and rate of decreases in concentrations of Ali in yr-1) ) 0.015 × DOC (in µmol of C/L) - 1.7; r 2 ) 0.97). Seven Adirondack Lakes in the 1990s is consistent with the high of the 48 lakes showed increases in DOC concentrations over NO3 decrease. These trends in Al chemistry the shorter interval. Although our observed pattern of in ALTM lakes clearly show a shift in speciation from toxic increases in DOC in some Adirondack Lakes in response to inorganic form toward less toxic organic forms with decreases decreases in acidic deposition is tentative, if real, it may have in atmospheric deposition and increases in DOC concentra- important ecological implications. Krug and Frink (35) tions. However in 2000, 16 out of 48 lakes showed mean Ali hypothesized that acidic deposition resulted in a shift in the concentrations above 2 µmol/L, a value identified as toxic nature of the acidity of surface waters, from acidity derived to aquatic organisms, including juvenile forms of Adirondack from naturally occurring organic acids to largely strong inorganic acids. Since that time, there has been considerable Acknowledgments
debate and discussion over the role of naturally occurringorganic acids in the acidification of surface waters and how Support for this study was provided by the New York State organic solutes change in response to changes in acidic Energy Research and Development Authority (NYSERDA), deposition (36-39). If DOC is a surrogate for naturally the New York State Department of Environmental Conser- occurring organic acids, increases in DOC should offset, to vation, and the U.S. Environmental Protection Agency. We some extent, increases in pH and ANC in surface waters that thank T. Butler for his help compiling emissions data. This result from decreases in acidic deposition. Increases in DOC manuscript has not been subjected to agency review, and no should also increase the concentration of organic ligands official endorsement by any agency should be inferred.
that are available to complex potentially toxic inorganic Literature Cited
species of aluminum (Al; see below). To bound the extent ofthis process, we applied a triprotic organic acid analogue (1) Landers, D. H.; Overton, W. S.; Linthurst, R. A.; Brakke, D. F.
model (39) developed for Adirondack Lakes to times-series Environ. Sci. Technol. 1988, 22, 128-135.
(2) National Acid Precipitation and Assessment Program. NAPAP data for the 16 original ALTM lakes. These results show that Biennial Report to Congress: An Integrated Assessment; National increases in pH and DOC in some lakes resulted in significant Acid Precipitation and Assessment Program: Silver Spring, MD, increases in calculated organic anion concentrations in 9 of VOL. 37, NO. 10, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 2041
(3) U.S. Environmental Protection Agency, Office of Air and (22) Aber, J. D.; McDowell, W. H.; Nadelhoffer, K. J.; Magill, A.; Radiation, Acid Rain Division. Acid Deposition Standard Fea- Berntson, G.; Kamakea, M.; McNulty, S. G.; Currie, W.; Rustad, sibility Study Report to Congress; U.S. Government Printing L.; Fernandez, I. BioScience 1998, 48, 921-934.
Office: Washington, DC, 1995; EPA-430-R-95-001a.
(23) Aber, J. D.; Driscoll, C. T. Global Biogeochem. Cycles 1997, 11,
(4) Acid Rain: Emissions Trends and Effects in the Eastern United States; Report to Congressional Requesters; United States (24) Aber, J. D.; Ollinger, S. V.; Driscoll, C. T.; Likens, G. E.; Holmes, General Accounting Office: Washington, DC, 2000; GAO/RCED- R. T.; Freuder, R. J.; Goodale, C. L. Ecosystems 2002, 5, 648-658.
(25) Ollinger, S. V.; Aber, J. D.; Reich, P. B.; Freuder, R. J. Global (5) Driscoll, C. T.; Newton, R. M.; Gubala, C. P.; Baker, J. P.; Change Biol. 2002, 8, 545-562.
Christensen, S. In Acidic Deposition and Aquatic Ecosystems: (26) Oren, R.; Ellsworth, D. S.; Johnsen, K. H.; Philips, N.; Ewers, B.
Regional Case Studies; Charles, D. F., Ed.; Springer-Verlag: New E.; Maier, C.; Schafer, K. V. R.; McCerthy, H.; Hendry, G.; McNulty, S. G.; Katul, G. G. Nature 2001, 411, 469-472.
(6) Kretser, W. J.; Gallager, J.; Nicolette, J. An Evaluation of Fish Communities and Water Chemistry; Adirondack Lakes Survey (27) Mitchell, M. J.; Driscoll, C. T.; Kahl, J. S.; Likens, G. E.; Murdoch, P. S.; Pardo, L. H. Environ. Sci. Technol. 1996, 30, 2609-2612.
(7) Larsen, D. P.; Thornton, K. W.; Urquhart, N. S.; Paulsen, S. G.
(28) Lawrence, G. B.; Momen, B.; Roy, K. J. Environ. Qual. (in review).
Environ. Monit. Assess. 1994, 32, 101-134.
(29) Reuss, J. O.; Johnson, D. W. Acid Deposition and the Acidification (8) Stevens, D. L. J. Environ. Manage. 1994, 42, 1-29.
of Soils and Waters; Ecological Studies 59; Springer-Verlag: New (9) U.S. Environmental Protection Agency. National Air Pollutant Emission Trends 1900-1998; U.S. Government Printing Office: (30) Johnson, D. W.; Lindberg, S. E. Atmospheric Deposition and Washington, DC, 2000; EPA-454-R-00-002.
Forest Nutrient Cycling; Ecological Studies 91; Springer-Verlag: (10) Likens, G. E.; Driscoll, C. T.; Buso, D. C. Science 1996, 272, 244-
(31) Kirchner, D. W. Geochim. Cosmochim. Acta 1992, 56, 2311-
(11) Stoddard, J. L.; Jefferies, D. S.; Lukewille, A.; Clair, T. A.; Dillon, P. J.; et al. Nature 1999, 410, 575-578.
(32) Peters, N. E.; Driscoll, C. T. Biogeochemistry 1987, 3, 163-180.
(12) Driscoll, C. T.; Lawrence, G. B.; Bulger, A. J.; Butler, T. J.; Cronan, (33) Gbondo-Tugbawa, S. S.; Driscoll, C. T. Atmos. Environ. 2002,
C. S.; Eagar, C.; Lambert, K. F.; Likens G. E.; Stoddard, J. L.; Weathers, K. C. BioScience 2001, 51, 180-198.
(34) Gunn, J. M.; Mills, K. H. Ecology 1998, 6, 390-397.
(13) Bigelow, D. S.; Dossett, S. P. Instruction Manual; NADP/NTN Site Operation; NADP Program Coordination Office: Ft. Collins, (35) Krug, E. C.; Frink, C. F. Science 1983, 221, 520-525.
(36) Cumming, B. F.; Smol, J. P.; Kingston, J. C.; Charles, D. F.; Birks, (14) Driscoll, C. T.; van Dreason, R. Water Air Soil Pollut. 1993, 67,
H. J. B.; Camburn, K. E.; Dixit, S. S.; Uutala, A. J.; Sele, A. R. Can. J. Fish. Aquat. Sci. 1992, 48, 128-141.
(15) Driscoll, C. T.; Postek, K. M.; Mateti, D.; Sequeira, K.; Aber, J.
(37) Wright, R. F. Water Air Soil Pollut. 1989, 46, 251-259.
D.; Kretser, W. J.; Mitchell, M. J.; Raynal, D. J. Environ. Sci. Policy (38) Schindler, D. W.; Curtis, P. J. Biogeochemistry 1997, 36, 1-8.
1998, 1, 185-198.
(39) Driscoll, C. T.; Lehtinen, W. D.; Sullivan, T. J. Water Resour. Res. (16) Baker, J. P.; Gherini, S. A.; Christensen, S. W.; Munson, R. K.; 1994, 30, 297-306.
Driscoll, C. T.; Newton, R. M.; Gallagher, J.; Reckhow, K. H.; (40) Cronan, C. S.; Schofield, C. L. Environ. Sci. Technol. 1990, 24,
Schofield, C. L. Adirondack Lakes Survey: An Interpretive Analysis of Fish Communities and Water Chemistry, 1984-87; Adirondack (41) Driscoll, C. T.; Postek, K. M. In The Environmental Chemistry Lakes Survey Corporation: Ray Brook, NY, 1990.
of Aluminum; Sposito, G., Ed.; Lewis: Chelsea, MI, 1995; pp (17) Hirsch, R. M.; Slack, J. R. Water Resour. Res. 1984, 20, 727-732.
(18) Lynch, J. A.; Bowersox, V. C.; Grimm, J. W. Atmos. Environ. 2000,
(42) Wigington, P. J., Jr.; Baker, J. P.; DeWalle, D. R.; Kretser, W. A.; Murdoch, P. S.; Simonin, H. A.; Van Sickle, J.; McDowell, M. K.; (19) Butler, T. J.; Likens, G. E.; Stunder, B. J. Atmos. Environ. 2001,
Peck, D. V.; Barchet, W. R. Ecol. Appl. 1996, 6, 374-388.
(20) Mattson, M. D.; Godfrey, P. J.; Walk, M. F.; Kerr, P. A.; Zajicek, O. T. Water Air Soil Pollut. 1997, 96, 211-232.
Received for review September 5, 2002. Revised manuscript (21) Stoddard, J. L. In Environmental Chemistry of Lakes and received February 18, 2003. Accepted February 26, 2003. Reservoirs; Baker, L., Ed.; American Chemical Society: Wash-ington, DC, 1994; pp 223-284.
2042 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 10, 2003

Source: http://eaton.math.rpi.edu/faculty/schwendeman/Workshop/OxfordGradCamp2011/Refs/Chemical%20Response%20of%20Lakes%20in%20the%20Adirondack%20Region%20of%20New%20York%20to%20Declines%20in%20Acidic%20Deposition.pdf