Search

Atmospheric acid deposition of sulfate and nitrate has declined markedly in the northeastern United States due to emissions controls. We investigated long-term trends in soil water (1984–2011) and stream water (1982–2011) chemistry along an elevation gradient of a forested watershed to evaluate the progress of recovery of drainage waters from acidic deposition at the Hubbard Brook Experimental Forest in the White Mountains of New Hampshire, USA. We found slowed losses of base cations from soil and decreased mobilization of dissolved inorganic aluminum. Stream water pH at the watershed outlet increased at a rate of 0.01 units yr−1, and the acid neutralizing capacity (ANC) gained 0.88 µeq L−1 yr−1. Dissolved organic carbon generally decreased in stream water and soil solutions, contrary to trends observed at many North American and European sites. We compared whole-year hydrochemical trends with those during snowmelt, which is the highest-flow and lowest ANC period of the year, indicative of episodic acidification. Stream water during snowmelt had long-term trends of increasing ANC and pH at a rate very similar to the whole-year record, with closely related steady decreases in sulfate. A more rapid decline in stream water nitrate during snowmelt compared with the whole-year trend may be due, in part, to the marked decrease in atmospheric nitrate deposition during the last decade. The similarity between the whole-year trends and those of the snowmelt period is an important finding that demonstrates a consistency between recovery from chronic acidification during base flow and abatement of snowmelt acidification.

[1] In many industrialized regions of the world, atmospherically deposited sulfur derived from industrial, nonpoint air pollution sources reduces stream water quality and results in acidic conditions that threaten aquatic resources. Accurate maps of predicted stream water acidity are an essential aid to managers who must identify acid-sensitive streams, potentially affected biota, and create resource protection strategies. In this study, we developed correlative models to predict the acid neutralizing capacity (ANC) of streams across the southern Appalachian Mountain region, USA. Models were developed using stream water chemistry data from 933 sampled locations and continuous maps of pertinent environmental and climatic predictors. Environmental predictors were averaged across the upslope contributing area for each sampled stream location and submitted to both statistical and machine-learning regression models. Predictor variables represented key aspects of the contributing geology, soils, climate, topography, and acidic deposition. To reduce model error rates, we employed hurdle modeling to screen out well-buffered sites and predict continuous ANC for the remainder of the stream network. Models predicted acid-sensitive streams in forested watersheds with small contributing areas, siliceous lithologies, cool and moist environments, low clay content soils, and moderate or higher dry sulfur deposition. Our results confirmed findings from other studies and further identified several influential climatic variables and variable interactions. Model predictions indicated that one quarter of the total stream network was sensitive to additional sulfur inputs (i.e., ANC

An integrated biogeochemical model (PnET-BGC) was formulated to simulate chemical transformations of vegetation, soil, and drainage water in northern forest ecosystems. The model operates on a monthly time step and depicts the major biogeochemical processes, such as forest canopy element transformations, hydrology, soil organic matter dynamics, nitrogen cycling, geochemical weathering, and chemical equilibrium reactions involving solid and solution phases. The model was evaluated against soil and stream data at the Hubbard Brook Experimental Forest, New Hampshire. Model predictions of concentrations and fluxes of major elements generally agreed reasonably well with measured values, as estimated by normalized mean error and normalized mean absolute error. Model output of soil base saturation and stream acid neutralizing capacity were sensitive to parameter values of soil partial pressure of carbon dioxide, soil mass, soil cation exchange capacity, and soil selectivity coefficients of calcium and aluminum. PnET-BGC can be used as a tool to evaluate the response of soil and water chemistry of forest ecosystems to disturbances such as clear-cutting, climatic events, and atmospheric deposition.

Seven high-altitude headwater catchments were studied from 1990 to 1994 to evaluate susceptibility to episodic acid neutralizing capacity (ANC) depression. Dilution (decreasing base cation concentrations) was the primary factor in ANC depression during snowmelt, accounting for 75 to 97% of the ANC reduction. In lakes where acidification (increasing anion concentrations) was noted, nitrate and sulfate were equally important during the first half of snowmelt, while sulfate dominated the latter half. A linear model, based on the relationship between minimum and fall-overturn ANC for the lakes in our study, estimated that none of the 114 lakes sampled during the 1985 EPA Western Lakes Survey had been episodically acidified (ANC < 0). Modifications of the model were used to predict that approximately 6 and 10% of Sierran lakes will become episodically acidified with increases in nitrate and sulfate deposition of 50 and 150%, respectively. No lakes will be chronically acidified with these depositional increases.

Variations in acid-neutralizing capacity (ANC) in 13 streams in the Adirondack, Catskill, and Northern Appalachian Plateau regions of the northeast United States were related to discharge, time of year, and seasonal variations in cation and anion concentrations using periodic regression analysis, ANC varied significantly with both discharge and time of year in 12 streams. Generation of ANC seasonal variations, being dependent upon the precise timing and magnitude of seasonal variations in cation and anion concentrations, was unique to each stream. Greatest seasonal ANC variations occurred in streams where seasonal variations in major anion and cation concentrations were completely out of phase. Maximum errors that could occur because of extrapolation of ANC data from one time of year to another were equal to or greater than maximum errors due to extrapolation of ANC from one discharge to another.

Acidification of groundwater in a noncalcareous sandy aquifer at Klosterhede, Denmark, is the result of acid rain deposition. In the 4- to 5-m-thick unsaturated zone the pH ranges from 4.2 to 4.9 with Al concentrations of up to 0.8 mmol L{sup {minus}1}. The groundwater at the top of the saturated zone still has a pH below 5. Deposition of sea salt affects the solute profiles, and its importance varies both spatially from the forest margin to the inner part of the forest and temporally through seasonal variations in infiltration and dry deposition. As a result, pulses of high solute concentrations travel downward through the unsaturated zone. The cation exchange capacity (CEC) of the sediments ranges between 0.2 and 1 meq 100 g{sup {minus}1}, and in the acidified zone, base saturation is around 17%. The pore waters are close to equilibrium with gibbsite, supersaturated for kaolinite, and strongly undersaturated for other silicate minerals. Mass balance calculations on increases in dissolved silica over depth suggest that the buffering effect of silicate weathering is small. Buffering processes and solute transport were modeled with the code PHREEQM. Simulation of pre-acid rain weathering indicates that this process operates on a timescale of thousands of years,more » yielding minimum pH values near 5.2 and a base saturation of greater than 70%. The present leaching of Al{sup 3+} rich acid water from the soil yields acidification rates of 7 and 10 cm yr{sup {minus}1} for weathering of a naturally weathered and a pristine profile, respectively. Simulation of infiltration of sea-salt pulses indicates that the cation distribution quickly becomes attenuated by the exchanger composition. However, due to coupling of gibbsite equilibrium with ion exchange processes, downward traveling pulses with high solute concentrations will cause pH variations throughout the unsaturated zone by precipitation and dissolution of gibbsite. 67 refs., 16 figs., 3 tabs.« less

A two-component mixing model of acid neutralizing capacity (ANC) is proposed for explaining two observed features related to the episodic acidification of surface waters during snowmelt periods: (1) maximum episodic declines in ANC are largest in high ANC systems and increase linearly with antecedent ANC and (2) relative depressions in ANC attributable to increases in nitric acid concentrations are larger in low ANC systems, while relative depressions in ANC attributable to dilution of base cations are larger in high ANC systems. Conceptually, the model represents the physical mixing of two hydrochemical end-members within a surface water environment, although the physical sources of water in the model are undefined. The model is shown to explain 55–72% of the total variation of these characteristics among various surface water systems within the Catskill and Adirondack mountain regions of New York. In addition, the model also explains 11–47% of the relative depression in ANC attributable to natural organic acidity in surface waters in these regions. The model is subsequently linked to an empirical equilibrium acidification model for predicting the long-term episodic acidification response of Adirondack lakes during snowmelt periods. Model predictions suggest that percentage decreases in sulfuric acid concentrations of the magnitude mandated by the 1990 Clean Air Act Amendments (40%) will not restore to positive values the ANC of all Adirondack lakes which are currently acidic (ANC < 0) during spring snowmelt periods. Long-term increases in nitric acid concentrations may counterbalance the expected increases in ANC attributable to reductions in sulfur deposition.

[1] In the 300 Area of a U(VI)-contaminated aquifer at Hanford, Washington, USA, inorganic carbon and major cations, which have large impacts on U(VI) transport, change on an hourly and seasonal basis near the Columbia River. Batch and column experiments were conducted to investigate the factors controlling U(VI) adsorption/desorption by changing chemical conditions over time. Low alkalinity and low Ca concentrations (Columbia River water) enhanced adsorption and reduced aqueous concentrations. Conversely, high alkalinity and high Ca concentrations (Hanford groundwater) reduced adsorption and increased aqueous concentrations of U(VI). An equilibrium surface complexation model calibrated using laboratory batch experiments accounted for the decrease in U(VI) adsorption observed with increasing (bi)carbonate concentrations and other aqueous chemical conditions. In the column experiment, alternating pulses of river and groundwater caused swings in aqueous U(VI) concentration. A multispecies multirate surface complexation reactive transport model simulated most of the major U(VI) changes in two column experiments. The modeling results also indicated that U(VI) transport in the studied sediment could be simulated by using a single kinetic rate without loss of accuracy in the simulations. Moreover, the capability of the model to predict U(VI) transport in Hanford groundwater under transient chemical conditions depends significantly on the knowledge of real-time change of local groundwater chemistry.

Geochemical reactions and hydrologic regime both affect runoff chemistry in many catchments. Each factor can obscure the effects of the other. This complicates efforts to interpret and predict catchments' geochemical and hydrological behavior. Here we present simple empirical techniques for separating effects of discharge fluctuations and effects of changes in acid anion concentrations in catchment monitoring data. We illustrate these techniques with data from central Ontario. Statistically subtracting acid anion effects reveals that important water quality variables (e.g., sum of base cations, acid neutralizing capacity, H+, Ali) are simple functions of the logarithm of catchment discharge. Likewise, correcting for discharge effects reveals that these water quality variables are roughly linear functions of runoff acid anion concentrations. The linear functions linking acid anion concentrations to water quality can be reliably predicted from bulk catchment runoff chemistry. Therefore acid anions' effects on water quality can be reliably predicted, even if they are obscured by discharge fluctuations in catchment monitoring data.

Alkalinity and salinity are linearly correlated in the waters of the most western part of the Mediterranean Sea, Nevertheless, the specific alkalinity of the saltier Mediterranean waters is higher than that of the less saline water of Atlantic origin. This may be explained by alkalinity input from rivers and the Black Sea. In the hypothesis of a steady state, the relationship between alkalinity and salinity permits calculation of the net flux of alkalinity between the Mediterranean Sea and the Atlantic Ocean, provided the water budget of the Mediterranean is known. The sources (rivers and the Black Sea) and sinks (seafloor) of alkalinity in the Mediterranean balance the net alkalinity transport into the Atlantic, if the water deficit of the Mediterranean is 2.2 x 10(12) m3 yr-1 or 88 cm yr-1 per unit area. The budget of carbon in the Mediterranean Sea is more difficult to assess as the sources and sinks are numerous and not well-documented. There is particularly a lack of data on the seasonal variations in the inorganic and organic carbon concentrations in the region of the strait of Gibraltar and on the partial pressure of CO2 in surface water of the Mediterranean Sea. It appears nevertheless that there is an important flux of inorganic carbon from the Mediterranean to the Atlantic due to the supply of carbon by the rivers and the Black Sea and to the transformation of 40% of the organic carbon entering the Mediterranean in inorganic carbon. Moreover, the Mediterranean is likely a sink for the atmospheric carbon.

Mineral acids and bases influence pH largely through their effects on acid-neutralizing capacity (ANC). This influence becomes particularly significant as ANC approaches zero. Analysis of data collected by the Adirondack Lakes Survey Corporation (ALSC) from 1469 lakes throughout the Adirondack region indicates that variations in ANC in these lakes correlate well with base cation concentrations (CB), but not with the sum of mineral acid anion concentrations (CA). This is because (CA) is relatively constant across the Adirondacks, whereas CB varies widely. Processes that supply base cations to solution are ion-specific. Sodium and silica concentrations are well correlated, indicating a common source, mineral weathering. Calcium and magnesium also covary but do not correlate well with silica. This indicates that ion exchange is a significant source of these cations in the absence of carbonate minerals. Iron and manganese concentrations are elevated in the lower waters of some lakes due to reducing conditions. This leads to an ephemeral increase in CB and ANC. When the lakes mix and oxic conditions are restored, these ions largely precipitate from solution. Sulfate is the dominant mineral acid anion in ALSC lakes. Sulfate concentrations are lowest in seepage lakes, commonly about 40 μeq/L less than in drainage lakes. This is due in part to the longer hydraulic detention time in seepage lakes, which allows slow sulfate reduction reactions more time to decrease lake sulfate concentration. Nitrate typically influences ANC during events such as snowmelt. Chloride concentrations are generally low, except in lakes impacted by road salt.

Past approaches for evaluating the effects of organic acids on the acid-base characteristics of surface waters have typically treated them solely as weak acids. Analysis of data collected by the Adirondack Lakes Survey Corporation (ALSC) from 1469 lakes throughout the Adirondack region shows that this approach is not valid. While the data indicate that natural organics contain a continuum of acid functional groups, many of which display weak acid characteristics, a significant fraction of the organic acid is strong (pKa < 3). Dissolved organic carbon (DOC) contributes 4.5–5 μeq/mg DOC of strong acid to solution. The associated anions make a negative contribution to Gran acid-neutralizing capacity (ANC). Because organic anions can produce negative Gran ANC values, the common practice of considering negative values of Gran ANC evidence of acidification solely by mineral acids is not valid. The strength of organic acids also influences the observed deviation between Gran ANC values and ANC values calculated as the difference between base cation and mineral acid anion concentrations (CB – CA). Ninety percent of the deviation is due to the presence of strong organics while the remaining 10% is due to DOC-induced curvature in the F1 Gran function. Organic acids can also strongly influence pH. Their largest effects were found in the 0–50 μeq/L Gran ANC range where they depressed pH by up to 1.5 units. In addition, a method for predicting changes in pH in response to changes in mineral acidity, DOC, or both without having to rely on inferred thermodynamic constants and the uncertainties associated with them has been developed. Using the predictive method, the response of representative lakes from four sensitive lake classes to a 15-μeq/L decrease in mineral acidity ranged from +0.17 to +0.38 pH units. If concurrent increases in DOC are considered, the pH changes would be even smaller.

[1] An abundance of exposed bedrock, sparse soil and vegetation, and fast hydrologic flushing rates make aquatic ecosystems in Yosemite National Park susceptible to nutrient enrichment and episodic acidification due to atmospheric deposition of nitrogen (N) and sulfur (S). In this study, multiple linear regression (MLR) models were created to estimate fall-season nitrate and acid neutralizing capacity (ANC) in surface water in Yosemite wilderness. Input data included estimated winter N deposition, fall-season surface-water chemistry measurements at 52 sites, and basin characteristics derived from geographic information system layers of topography, geology, and vegetation. The MLR models accounted for 84% and 70% of the variance in surface-water nitrate and ANC, respectively. Explanatory variables (and the sign of their coefficients) for nitrate included elevation (positive) and the abundance of neoglacial and talus deposits (positive), unvegetated terrain (positive), alluvium (negative), and riparian (negative) areas in the basins. Explanatory variables for ANC included basin area (positive) and the abundance of metamorphic rocks (positive), unvegetated terrain (negative), water (negative), and winter N deposition (negative) in the basins. The MLR equations were applied to 1407 stream reaches delineated in the National Hydrography Data Set for Yosemite, and maps of predicted surface-water nitrate and ANC concentrations were created. Predicted surface-water nitrate concentrations were highest in small, high-elevation cirques, and concentrations declined downstream. Predicted ANC concentrations showed the opposite pattern, except in high-elevation areas underlain by metamorphic rocks along the Sierran Crest, which had relatively high predicted ANC (>200 μeq L−1). Maps were created to show where basin characteristics predispose aquatic resources to nutrient enrichment and acidification effects from N and S deposition. The maps can be used to help guide development of water-quality programs designed to monitor and protect natural resources in national parks.

[1] In this study, we examine the impact of increasing concentrations of dissolved organic carbon (DOC) on the recovery from acidification for 66 lakes in Southern Sweden. The lakes are small, weakly buffered, and have all been affected by acidifying deposition. A majority of the lakes (∼75%) showed an increase in DOC concentrations between 1990 and 2008. The method used in this study was to model pH in 2008 from DOC, acid neutralizing capacity, pCO2 (partial carbon dioxide pressure), and Al speciation, using both the DOC observed in 2008 and the “unelevated” DOC of 1990. Furthermore, we consider the indirect effects of increasing DOC on acidity, i.e., the ancillary effects from DOC on pCO2, Al transport and speciation, and release of base cations (BCs). Our results indicate that the DOC increase in the latest decades has retarded the recovery from acidification by 0.13 pH units (median for all lakes) and by more than 1 unit for individual lakes. The effects of elevated pCO2 and BC concentrations accompanying the DOC increase compensated for each other for the average lake, whereas the effects of higher Al transport on pH were minor. The estimate of the amount of BCs released with the organic anions is however uncertain, and further studies on this topic are needed.

The model of acidification of groundwater in catchments (MAGIC) was applied to data from the Panola Mountain Research Watershed located near Atlanta, Georgia, in the southern Piedmont physiographic province of the United States. In contrast to past applications, MAGIC was calibrated directly to two soil environments using information gained from an independent mixing model rather than to the stream water. A third soil environment, the deeper soil horizons which provide base flow, was identified from the mixing model, but its chemistry was assumed to be invariant during the forecast period, and hence it was not included in MAGIC. This mixing model identified soil environments that appear to be important in determining present stream water chemistry, determined the hydrologic routing parameters necessary for the model, and permitted predictions of stream water chemistry during future storm events. Annual average stream water acid neutralizing capacity (ANC) was predicted to decrease during the next 50 years in a manner largely consistent with a regional analysis of the nearby southern Blue Ridge province performed by the U.S. Environmental Protection Agency. The two upper soil layers, however, were predicted to acidify substantially within the next 20 years so that the stream water may become unsuitable for sensitive aquatic biota for much of the year, despite a positive annual average ANC. This approach, which combines a predictive model and a mixing model, provides testable hypotheses because actual soil environments are modeled; field experiments to test the processes contained in the model are suggested.

In late 1988, parallel studies of Lake Five-O (pH 5.14) in the Florida panhandle and Lake Barco (pH 4.50) in north central Florida were initiated to develop hydrologic and major ion budgets of these lakes as part of an overall effort to improve understanding of the hydrologic, depositional, and biogeochemical factors that control acid-neutralizing capacity (ANC) in seepage lakes. Preliminary findings from these studies indicate that earlier perceptions of lake hydrology and mechanisms of ANC regulation in Florida seepage lakes may have to be revised. The traditional perspective of seepage lakes in the Florida panhandle views these systems as dominated by precipitation inputs and that ANC regulation is due largely to in-lake processes. Our results for Lake Five-O show modest to steep hydraulic gradients almost entirely around the lake. In addition, the horizontal hydraulic conductivity of the surficial aquifer is high (8–74 m day−1), indicating that large quantities of groundwater flow into Lake Five-O. Calculations of net groundwater flow from hydrologic budgets also indicate that groundwater may comprise at least 38 to 46% of the total inflow. For Lake Barco, net flow estimates of the minimum groundwater inflow range from 5 to 14% of total inflow. Enrichment factor and ion flux calculations for Lake Five-O and Lake Barco indicate that terrestrial as well as in-lake processes contribute significantly to ANC regulation. The extent that terrestrial processes contribute to ANC generation is directly related to the magnitude of groundwater inflow as well as the degree of ion enrichment or depletion that occurs in the surficial aquifer. Net ANC generation in both study lakes was dominated by anion retention (NO3− and SO42−). Where previous studies concluded that in-lake reduction was the primary sink for SO42−, our preliminary calculations show that adsorption of SO42− within the watershed is perhaps twice as important as in-lake reduction as a source of ANC. Net base cation enrichment in both lakes was negligible.

Snowpack runoff contributions to the hydrochemistry of an alpine catchment in the Sierra Nevada were evaluated in 1986 and 1987 by analyzing snowpack, meltwater, and stream water samples for major inorganic ions, conductance, acid neutralizing capacity (ANC), and silicate. An ionic pulse in meltwater with initial concentrations twofold to twelvefold greater than the snowpack average, varying with site and with ion, was measured in lysimeters placed at the base of the snowpack. Maximum concentrations of ions in meltwater were inversely related to the rate of snowmelt; melt-freeze cycles increased the concentration of solutes in meltwater. Hydrogen concentration in meltwater was buffered by ANC produced in part by the dissolution of particulates. The anionic pulse in meltwater was observed in stream waters during the first 30 days of snowpack runoff, with NO3− concentrations in stream waters at this time about 1.6-fold greater than the average NO3− concentration for the time period of snowpack runoff, Cl− about 1.5-fold greater and SO42− about 1.3-fold greater. Maximum H+ concentration during snowpack runoff (increase of 170% over winter concentrations) occurred near maximum discharge. ANC minima occurred at maximum discharge as a result of dilution, with a decrease from winter concentrations of 70% in 1986 and 60% in 1987. Interactions between snowpack runoff and soils were important to the chemistry of stream water. Eighty to ninety percent of the H+ stored in the snowpack was consumed before it reached the base of the catchment. Soils were a sink for NH4+ from snowpack meltwater, with less than 1% of the NH4+ released from snowpack storage exported from the basin as NH4+. Sulfate concentrations in stream waters were less variable than NO3− or Cl− concentrations; sorption processes in soils were a likely cause for the regulation of SO42− concentrations.

To assess the regional acid-base status of streams in the mid-Atlantic and southeastern United States, spring base flow chemistry was surveyed in a probability sample of 500 stream reaches representing a population of 64,300 reaches (224,000 km). Approximately half of the streams had acid-neutralizing capacity (ANC) ≤ 200 μeq L−1. Acidic (ANC ≤ 0) streams were located in the highlands of the Mid-Atlantic region (southern New York to southern Virginia, 2330 km), in coastal lowlands of the Mid-Atlantic (2600 km), and in Florida (462 km). Acidic streams were rare (less than 1%) in the highlands of the Southeast. Inorganic monomeric aluminum (Alim) concentrations were highest in acidic streams of the Mid-Atlantic Highlands where over 70% of the acidic streams had Alim greater than 100 μg L−l, a concentration above which deleterious biological effects have frequently been reported. Dissolved organic carbon concentrations were much higher in lowland coastal streams, compared with inland streams. Our data support a hypothesis that atmospheric sources and watershed retention control regional patterns in streamwater sulfate concentrations. Most stream watersheds retain the vast majority of the total nitrogen loading from wet deposition. The data suggest, however, that some deposition nitrogen may be reaching streams in the Northern Appalachians. These results show that acidic surface waters are found outside the glaciated northeastern portions of the United States and that watershed sulfate retention is not sufficient to prevent acidic conditions in some Mid-Atlantic Highlands streams.

The authors examined anion composition in National Stream Survey (NSS) data in order to evaluate the most probably sources of current acidity in acidic and low acid-neutralizing capacity (ANC) streams in the eastern US. Acidic streams that had almost no organic influence (less than 10% of total anions) and sulfate and nitrate concentrations indicative of evaporative concentration of atmospheric deposition were classified as acidic due to acidic deposition. These acidic streams were located in small (

Two indices are presented for assessing the expected responsiveness of lakes to changes in external acid inputs. The first is the ratio of the sum of acidic anion to sum of base cation equivalents (CA/CB), which is a familiar index of lake response to past inputs; the second quantity is buffer intensity (β = −dCA/dpH), which serves as an index of lake pH sensitivity to changes in current inputs. Because the indices account for both capacity and intensity of pH buffering, the approach provides greater information on lake responsiveness to acidification than the more customary value, acid neutralizing capacity. The approach is applied to water quality data from the Environmental Protection Agency's Eastern Lake Survey and preliminary paleolimnological data from the PIRLA project (paleolimnological investigation of recent lake acidification). These applications suggest that the approach may yield a useful tool with which to evaluate existing data from sites considered for monitoring the effects of changing acid inputs on lake systems.

[1] The Adirondack region of New York has high inputs of acidic deposition and large numbers of acidic lakes. The biogeochemical model, PnET-BGC, was applied to 44 statistically representative Environmental Monitoring and Assessment Program (EMAP) lake watersheds in the Adirondacks. Model simulations help provide an understanding of historical effects of acidic deposition on soils and lake waters. Model simulations indicate that median annual concentrations of SO42− and NO3− in the 44 EMAP lakes were 15.9 μeq/L and 3.8 μeq/L, respectively, in 1850, compared to the median current measured values of 88.8 μeq/L and 20.0 μeq/L. Simulated median values of pH, acid-neutralizing capacity (ANC), and soil percent base saturation were 6.63, 67.7 μeq/L, and 12.3%, respectively, in 1850, compared to the median current measured values of 5.95, 27.8 μeq/L, and 7.9%. The estimated historical surface water acidification was greatest in lakes having low ANC below values of 100 μeq/L. This pattern of historical acidification is in agreement with a previous paleolimnological investigation.

Cokriging is used to predict stream chemistry at unsampled locations with the use of spatial and intervariable correlation. The technique is used in this study to predict the acid neutralizing capacity (ANC) of streams in the Southern Blue Ridge Province (SBRP). ANC measurements between pairs of streams surveyed in this region were found to be spatially correlated over distances up to around 40 km. Predictions were improved by including elevation in the analysis to represent the combined influence of elevational gradients in climate, geology, soils, hydrology, and vegetation on stream ANC. The cokriging analysis identified specific stream reaches predicted to be most sensitive to acidification and located areas of high uncertainty. Stream ANC levels below 50 μeq/L were predicted for one-fifth of the upper nodes associated with digitized headwater reaches in the SBRP. The majority of these were located in the higher elevations of the Great Smoky Mountains National Park, in the vicinity of Mount Mitchell, and in the Blue Ridge Mountains in southern North Carolina.

Maximum values of acid neutralizing capacity (ANC) in Adirondack, New York lake outlets generally occur during summer and autumn. During spring snowmelt, transport of acidic water through acid-sensitive watersheds causes depression of upper lake water ANC. In some systems lake outlet ANC reaches negative values. We examined outlet water chemistry from 11 Adirondack lakes during 1986 and 1987 snowmelts. In these lakes, SO42− concentrations were diluted during snowmelt and did not depress ANC. For lakes with high baseline ANC values, springtime ANC depressions were primarily accompanied by basic cation dilution. For lakes with low baseline ANC, NO3− increases dominated ANC depressions. Lakes with intermediate baseline ANC were affected by both processes and exhibited larger ANC depressions. Ammonium dilution only affected wetland systems. A model predicting a linear relationship between outlet water ANC minima and autumn ANC was inappropriate. To assess watershed response to episodic acidification, hydrologic flow paths must be considered.

This comparative biogeochemical analysis focused on the patterns and processes of aluminum cycling in two small watersheds, one in the west-central Adirondacks of New York and the other on Cumberland Plateau of eastern Tennessee. Despite shared similarities in soil acidity, soil exchangeable aluminum concentrations, and elevated inputs of acidic deposition, the northern and southern sites exhibited strong differences in aqueous aluminum chemistry and transport. Soil and stream drainage waters in the northern watershed were more acidic, and contained higher concentrations of base cations, sulfate, nitrate, and organic carbon than waters in the southern ecosystem. Mean concentrations of biologically active labile inorganic aluminum, Ali, ranged from 17 to 46 μmol L−1 in soil solutions and stream water in the northern drainage basin, and from 0 to 2 μmol L−1 in the southern system. The major differences in aluminum chemistry and transport between the two watersheds were related to different patterns of alkalinity generation and mobile anion transport in these contrasting systems. In the northern watershed, atmospheric inputs of acidity were partially neutralized through the release of mixed cations (Ca, Al, Mg, K, and Na) from soils and detritus. Because of the high mobility of sulfate and nitrate in the northern watershed, there was significant transport of Al through the soil profile and into stream water. At the southern watershed, soil sulfate adsorption, biological retention of nitrate, and base cation release were the major sources of acid neutralizing capacity for soil drainage waters and surface waters. Because these processes resulted in both the release of alkalinity and the removal of mobile strong acid anions, concentrations of labile Al remained low throughout most of the drainage profile in the southern watershed. At the ecosystem level, results indicated that aluminum cycling at the southern site was controlled more by biotic transfers (e.g., plant uptake and litterfall), while solution transfers (e.g., soil leaching) dominated aluminum cycling at the northern site.

From December 1982 to May 1987 the quality and quantity of snowfall and snowpack were measured at four stations in a small (176 ha) gauged watershed adjacent to Lake Superior's south shore. Annual precipitation was less than and winter precipitation greater than observed east and south of the Lake Superior basin. Snowfall averaged 37% of annual precipitation but was quite variable. Snowfall significantly increased (from 21.1 to 35.7 cm, P < 0.001) with 150 m increase in elevation. Hydrogen and SO42− were the dominant ions in precipitation and snowpack while Ca2+ and HCO3− dominated stream water. No significant time trends in winter precipitation chemistry were found. Snowpack ionic retention rarely matched precipitation input. Generally, midwinter freeze-thaw periods resulted in elevated concentrations of solutes in snow meltwater and greatly reduced snowpack ion loads prior to peak stream water discharge. All ions exhibited pulses in snowmelt each year. Preferential elution from the snowpack of SO42− and K+ was observed. Stream water H+, NH4+ and most NO3− pulses were associated with midwinter thaws. Stream water SO42− pulses were rare. For those years exhibiting the most pronounced spring increase in stream H+, NO3− and SO42− concentration, the magnitude and rate of concentration change was as great in a year of small peak snowpack ion load as it was following a more rapid ion loss from a larger snowpack load. This suggests that the pattern of snowpack melt and especially the pathway the meltwater follows to the stream may be important factors in determining whether stream water chemical pulses occur.

A first-order second-moment uncertainty analysis has been applied to two lakes in the Adirondack Park, New York, to assess the long-term response of lakes to acid deposition. Uncertainty due to parameter error and initial condition error was considered. Because the enhanced trickle-down (ETD) model is calibrated with only 3 years of field data and is used to simulate a 50-year period, the uncertainty in the lake alkalinity prediction is relatively large. When a best estimate of parameter uncertainty is used, the annual average alkalinity is predicted to be -11 + or - 28 microeq/L for Lake Woods and 142 + or - 139 microeq/L for Lake Panther after 50 years. Hydrologic parameters and chemical weathering rate constants contributed most to the uncertainty of the simulations. Results indicate that the uncertainty in long-range predictions of lake alkalinity increased significantly over a 5- to 10-year period and then reached a steady state.

[1] Acid rain affects headwater streams by temporarily reducing the acid-neutralizing capacity (ANC) of the water, a process termed episodic acidification. The increase in acidic components in stream water can have deleterious effects on the aquatic biota. Although acidic deposition is uniform across Shenandoah National Park (SNP) in north central Virginia, the stream water quality response during rain events varies substantially. This response is a function of the catchment's underlying geology and topography. Geologic and topographic data for SNP's 231 catchments are readily available; however, long-term measurements (tens of years) of ANC and accompanying discharge are not and would be prohibitively expensive to collect. Transfer function time series models were developed to predict hourly ANC from discharge for five SNP catchments with long-term water-quality and discharge records. Hourly ANC predictions over short time periods (≤1 week) were averaged, and distributions of the recurrence intervals of annual water-year minimum ANC values were model-simulated for periods of 6, 24, 72, and 168 hours. The distributions were extrapolated to the rest of the SNP catchments on the basis of catchment geology and topography. On the basis of the models, large numbers of SNP streams have 6- to 168-hour periods of low-ANC values, which may stress resident fish populations. Smaller catchments are more vulnerable to episodic acidification than larger catchments underlain by the same bedrock. Catchments with similar topography and size are more vulnerable if underlain by less basaltic/carbonate bedrock. Many catchments are predicted to have successive years of low-ANC values potentially sufficient to extirpate some species.

[1] Seasonal patterns are evident in surface water chemistry draining forested headwaters in the northeastern United States. This variation in surface water chemistry is largely driven by seasonal fluctuations in hydrologic flow paths and biological activity. Especially during spring snowmelt, high-flow conditions are characterized by high concentrations of NO3‒, naturally occurring organic anions and aluminum species, and depression in surface water pH and acid neutralizing capacity (ANC). Under extreme conditions, decreases in pH and ANC and increases in aluminum can have adverse effects on aquatic biota. As a result, there is a critical need to be able to simulate seasonal variations in surface water acid-base chemistry. Previously, a single soil layer biogeochemical model (PnET-BGC) was found to be inadequate to simulate seasonal variations in stream chemistry draining acid-sensitive forest watersheds. In order to better simulate the seasonal variations in the acid-base chemistry of surface waters, a two-layer model (PnET-BGC2) was formulated and applied to a northern forest ecosystem. End-member mixing analysis was used to better understand hydrologic flow paths contributing to temporal patterns in stream chemistry and to parameterize the model. The resulting two-layer model is generally able to reproduce the seasonal variations in surface water runoff, concentrations of base cations, SO42‒, NO3‒, pH, and ANC.

[1] We evaluated the sensitivity of The Loch, a subalpine lake in Rocky Mountain National Park in Colorado, to acidification in response to increased atmospheric loading of sulfur (S) and nitrogen (N) using the Model of Acidification of Groundwater in Catchments (MAGIC). Lake water acid-base chemistry was moderately sensitive to changes in both S and N deposition. However, the loads of S deposition that would drive chronic lake water acid neutralizing capacity (ANC) to below 0 or 20 μeq L−1 were estimated to be 11 and 8 kg S ha−1 yr−1, respectively, assuming constant future N deposition at current levels. Comparable loads for N deposition, assuming constant future S deposition, were estimated to be 21 and 12 kg N ha−1 yr−1, respectively. Modeling results for Andrews Creek, an alpine tributary to The Loch, suggested critical loads for surface water acidification that averaged about one third lower. Surface water ANC = 50 μeq L−1 was projected to occur in 50 years in The Loch if S or N deposition increased by a moderate amount (

[1] Episodic concentration-discharge (c-Q) plots are a popular tool for interpreting the hydrochemical response of small, forested catchments. Application of the method involves assuming an underlying conceptual model of runoff processes and comparing observed c-Q looping patterns with those predicted by the model. We analyzed and interpreted c-Q plots of acid-neutralizing capacity (ANC) for 133 storms collected over a 7-year period from three catchments in Shenandoah National Park, Virginia. Because of their underlying lithologies the catchments represent a gradient in both hydrologic and geochemical behavior, ranging from a flashy, acidic, poorly buffered catchment to a moderate, neutral, well-buffered catchment. The relative frequency of observed anticlockwise c-Q loops in each catchment decreased along this gradient. Discriminant function analysis indicated that prestorm base flow ANC was an important predictor of loop rotation direction; however, the strength of the predictive relationship decreased along the same gradient. The trends were consistent with several equally plausible three-component mixing models. Uncertainty regarding end-member timing and relative volume and possible time variation in end-member concentrations were key factors precluding identification of a unique model. The inconclusive results obtained on this large data set suggest that identification of underlying runoff mechanisms on the basis of a small number of c-Q plots without additional supporting evidence is likely to be misleading.

[1] We constructed a model-based evaluation of current and future effects of atmospheric S and N deposition on aquatic resources in the eight-state southern Appalachian Mountains region. Modeling was conducted with the MAGIC model for about 40 to 50 sites within each of three physiographic provinces, stratified by stream water acid neutralizing capacity (ANC) class. Simulations were based on assumed constant future atmospheric deposition at 1995 levels and on three regional strategies of emissions controls provided by the Southern Appalachian Mountains Initiative (SAMI), based on the Urban to Regional Multiscale One-Atmosphere model. The National Stream Survey statistical frame was used to estimate the number and percentage of stream reaches in the region that were projected to change their chemistry in response to the emissions control strategies. There was a small decline in the estimated length of projected acidic (ANC ≤ 0) streams in 2040 from the least to the most restrictive emissions control strategy, but there was little difference in projected stream length in the other ANC classes as a consequence of adopting one or another strategy. However, projections of continued future acidification were substantially larger under a scenario in which S and N deposition were held constant into the future at 1995 levels. Model results suggested that the percent of potentially acid-sensitive streams having chemistry that is chronically unsuitable for brook trout would increase slightly between 1995 and 2040 under all except the most restrictive emissions control strategy.

The Emerald Lake Watershed study was organized to investigate the effects of acidic deposition on high-elevation watersheds and surface waters of the Sierra Nevada, California. Some of the results of this comprehensive study of aquatic and terrestrial ecosystems at a small, headwater basin are presented in four papers in this series. The watershed study site is in Sequoia National Park, on the western slope of the Sierra Nevada. This glacial cirque is located in the upper Marble Fork of the Kaweah River. This 120-ha watershed ranges from Alta Peak (3416 m) down to Emerald Lake (2400 m). Most of the watershed surface area is exposed granite and granodiorite rocks, with limited coverage (about 20%) by thin, acidic soils. The hydrology of the basin is dominated by snowmelt runoff during March–June. Emerald Lake, a glacial tarn, is 2.72 ha in area, with a maximum depth of 10.5 m. Surface waters are poorly buffered and dominated by calcium and bicarbonate. Most of the yearly precipitation falls as dilute snow (pH 5.2–5.4), with acidic rain storms sampled during May–October.

18and12,500km 2 .Soilexchangeablecationcontentexplainsmostofthevarianceinstream concentrations of cations, dissolved silicon (Si), and acid neutralizing capacity (ANC) in both forested and deforested basins based on regression analysis (R 2 range 0.61–0.81), though the mechanism underlying the relationship is unknown. We use the relationship between soil exchangeable cation content and stream solute concentrations in forested catchments to estimate the predisturbance and postdisturbance signal concentrations of stream solutes for deforested watersheds. Signal concentrations of potassium (K), sodium (Na), and chloride (Cl) increase with deforestation extent for streams on gneiss, granite, and sedimentary rock in both the dry and wet seasons after the effect of soil type has been statistically removed. Sulfate (SO4) signal concentrations increase with deforestation extent in the dry season only. For catchments >40% deforested but with no urban populations, the ratios of disturbed to predisturbance stream concentrations range from 1.2 to 3.1 (K), 0.8 to 2.2 (Na), 0.7 to 2.7 (SO4), and 2.4 to 14.6 (Cl) in the dry season and 1.2 to 2.2 (K) and 0.6 to 3.4 (Cl) in the wet season. Urban population density strongly affects Cl and SO4 concentrations in both seasons and Na concentrations in the dry season; the urban signal comprises 40–58% of the dry season Cl signal and 83–89% of the wet season Cl signal for watersheds with >250 persons per km 2 . Streams on mica-schist, mafic rock and carbonate shalehavehigherconcentrationsofCa,Mg,Si,andANCthanwatershedsongneiss,granite, tertiary sediments or sandstone for a given soil exchangeable cation content in both seasons. Simple mass balance calculations suggest that supplemental cattle salts could comprise a significant fraction of the Na and Cl signals in the wet season but would not significantly impact wet season concentrations of Ca, Mg, or K. INDEX TERMS: 1045 Geochemistry: Lowtemperature geochemistry; 1806 Hydrology: Chemistry of fresh water; 1871 Hydrology: Surface water quality; 1886 Hydrology: Weathering (1625); KEYWORDS: stream chemistry, Amazon, deforestation, urbanization, biogeochemistry

Integrated snowpack samples and lakewater samples were collected from a low-alkalinity watershed in the North Cascades in 1984 and 1985 and analyzed for inorganic chemical parameters. Mean concentrations of NO3− and SO4−2 from North Cascades snowpack were near 4.5 and 5.5 μeq L−1, respectively. Mean pH was 5.2–5.3. During melting, anion levels in the snowpack decreased and pH increased. By July, mean NO3− and SO4−2 concentrations were 0.32 and 1.73 μeq L−1, respectively, and mean pH was 5.48. Acid neutralizing capacity at the lake outlet during snowmelt was diluted to 50% of April premelt values. During snowmelt runoff there was no apparent neutralization of bicarbonate by strong acids. By comparing lakewater NO3− concentrations to snowpack concentrations, over 75% of the NO3− in the lake could be attributed to a snowpack source. An estimated 30% of the SO4−2 in the lake was derived from the snowpack.

Most alpine and subalpine lakes of the Sierra Nevada (California) are very dilute, weakly buffered calcium bicarbonate waters. Seventy-three lakes selected to lie in basins representative of Sierran bedrocks were sampled in the summers of 1981 and 1982 and have a median alkalinity of 50 μeq 1−1 (range 5–1137) and a pH of 7 (range 5.7–9.4). The alkalinity of Sierran lakes cannot be predicted from altitude, basin area, or lake area. The concentrations of sulfate, calcium, or alkalinity in lake waters are positively correlated with the percent areal coverage of volcanic or calcareous rocks in the basins. The lakes are not acidified but are extremely sensitive to increases in acidity of atmospheric precipitation.

Chemical requirements and costs are presented for neutralization of 663 Adirondack lakes with acid neutralizing capacity (ANC) values of

By considering the input/output of hydrogen ion with respect to the aquifer, including nonconservative transport through the use of the retardation factor, the effect of a delay in the arrival of acid to the lake is demonstrated. A two-dimensional, finite difference flow and random walk transport model, neglecting dispersion, was used to simulate the migration of the acid water through the ground water system. Also, the temporal changes in lake pH were examined by modeling the lake with both a lumped parameter model and an equilibrium chemistry model. Thus, the lake model mixes inflow from the ground water to the lake, precipitation on lake surface, and lake water, with the consideration of equilibrium chemical reactions. From their work they concluded that ground water inflow can mitigate the effect of acid deposition on lake watersheds, as well as the fact that the low ionic strength of the lakes was an important buffer. The scope of this comment is confined only to the issue of modeling the buffering capability of the aquifer. The purpose is to show that an alternate and simpler formulation for the ground water model is also appropriate for the case study.

Studies of individual lakes and streams have documented the occurrence of transient, short-term acidification of surface waters during hydrologic events, but a regional assessment of episodic chemical effects has not been made. Application of a two-box mixing model, together with regional chemistry and deposition data, indicates that acidic episodes (acid neutralizing capacity

In-lake alkalinity generation (IAG) is important in regulation of alkalinity in lakes with long residence times, particularly seepage lakes. An IAG model based on input/output modeling concepts is presented that describes budgets for each ion involved in alkalinity regulation by a single differential equation that includes inputs, outputs, and a first-order sink term. These equations are linked to an alkalinity balance equation that includes inputs, outputs, IAG (by sulfate and nitrate reduction), and internal alkalinity consumption (by ammonium assimilation). Calibration using published lake budgets shows that rate constants are generally similar among soft water lakes (kSO4 ≈ 0.5 m/yr; kNO3 ≈ 1.3 yr−1; kNH4 ≈ 1.5 yr−1). Sensitivity analysis shows that predicted alkalinity is sensitive to water/residence time, but less sensitive to modest changes in rate constants. The model reflects the homeostatic nature of internal alkalinity generation, in which internal alkalinity production increases with increasing acid input and decreases with decreasing acid inputs of HNO3 or H2SO4.

Streamwater samples were collected from watersheds in Jamieson Creek, British Columbia, and the Hubbard Brook Experimental Forest, New Hampshire, to compare the chemistry of drainage waters which are unimpacted and impacted by acidic deposition. Qualitative and quantitative differences in the chemistry of headwater streams were evident. Principal among these were a shift from acidification by organic acids to strong acids, increased concentrations of aqueous Al, and a change in Al speciation from largely organic to inorganic forms. In both watersheds, streamwater acidity is ultimately neutralized by the release of basic cations.

Major ion chemistry data collected as part of the Environmental Protection Agency (EPA) Eastern Lake Survey was examined to evaluate the mechanisms and extent of alkalinity regulation in 37 undisturbed, softwater lakes in Florida. Comparison of major ion-Cl ratios in atmospheric deposition and in lakewater shows the reactions resulting in retention of sulfate and nitrate are the dominant sources of alkalinity; production of organic acids and ammonium retention are the major alkalinity-consuming processes. Based on average reactions, enrichment of major cations accounted for only 12% of net alkalinity generation in the study lakes. In general, calcium and potassium were depleted in low ANC lakes, presumably by in-lake sinks, and were enriched in most higher ANC lakes by groundwater inputs. Differences in alkalinity among these lakes reflect hydrologic factors and the proximity of clay and carbonate deposits to the lake bed. Overall, net alkalinity generation nearly balanced H+ predicted from evaporative concentration of atmospheric acid inputs; this close balance suggests that the alkalinity status of these lakes is very sensitive to changes in atmospheric loadings and groundwater alkalinity inputs.

Seventy lakes in the Mt. Zirkel Wilderness Area, Colorado, were sampled for alkalinity, and 20 of these lakes were sampled for major anions. Analysis of the data indicates that most of the lakes are extremely sensitive to acidification; about half of the lakes have alkalinities less than 100 ..mu..eq/L. A linear regression of alkalinity against lake altitude accounts for 55% of the variance in alkalinity among the lakes, whereas a regression of alkalinity against dominant geologic unit (exclusive of altitude) accounts for 63% of the alkalinity variance. A conceptual model of lake acidification allows estimation of the quantity of acidification that may have occurred. Acidity associated with strong acid anions probably neutralized no more than 9 ..mu..eg/L of alkalinity. This estimate is much smaller than a previous estimate for nearby Colorado lakes.

Of the 10 papers presented in the ASLO special symposium, Regional Effects of Acid Deposition on Surface Waters, seven were regional case studies of lakes (four papers) or streams (three papers). These papers stressed the importance of hydrologic settings and flow paths in regulating pH and the acid neutralizing capacity (ANC). Three papers dealt with special topics: fishery models, paleolimnological reconstructions, and an overview of lake-watershed-airshed interactions. Geographically, data were derived primarily from a triangular section of the eastern and central United States (from Maine to Minnesota and Florida) corresponding to regions 1, 2, and 3 of the National Surface Water Survey.