Adirondack Lakes Study:
An Interpretive Analysis of Fish Communities and Water Chemistry (1984-1987)

INTRODUCTION

The Adirondacks is a large, forested upland and mountainous region in northern New York. The majority of surface waters in the region are dilute, with low levels of acid neutralizing capacity (ANC). At the same time, the region receives relatively large inputs of precipitation (100-150 cm annually) and acidic deposition (e.g., typical pH is about 4.1 to 4.2). As a result, much of the debate regarding the effects of acidic deposition on aquatic ecosystems has focused on the Adirondacks.

The Adirondack Lakes Survey Corporation (ALSC) was formed as a cooperative effort of the New York State Department of Environmental Conservation (NYSDEC) and the Empire State Electric Energy Research Corporation (ESEERCO) to better characterize the chemical and biological status of Adirondack lakes. Between 1984 and 1987, the ALSC surveyed 1469 lakes within the Adirondack ecological zone (Figure 1). This unique cooperative effort resulted in the acquisition of an unparalleled, extensive physical, chemical and biological data base on 52% of the ponded waters in the Adirondack region. The results from this survey are described in the ALSC summary report, Adirondack Lakes Study 1984-1987: An Evaluation of Fish Communities and Water Chemistry.

As a follow up to the survey, the ALSC sponsored a series of interpretive analyses of the ALSC data base. The primary objectives of these analyses were as follows:

• evaluate the influence of mineral acids (e.g., from acidic deposition) and nonmineral acids (e.g., natural organic acids) on lake pH levels

• classify Adirondack lakes according to lake and watershed features expected to influence their responsiveness to changes in acidic deposition

• evaluate the sensitivity of Adirondack lakes to changes in environmental conditions, such as changes in mineral acids or dissolved organic carbon (DOC) concentrations

• identify lake characteristics important in explaining the observed present-day status of fish communities in Adirondack lakes, in particular, the relative importance of lake acidity

• evaluate changes that have occurred over time in Adirondack fish communities and probable causes for these trends by using the available historical data on fish communities in the Adirondacks and the ALSC data base

• determine the degree to which the existing fish resource might be at risk from continued acidic deposition or might recover if acidity levels were reduced

The basic approach examined relationships observed in the ALSC data base among watershed characteristics, lake chemistry and fish status. These analyses built upon the understanding of important processes influencing lake chemistry and fish communities derived from earlier, mechanistic research projects in the Adirondacks, such as the Integrated Lake/Watershed Acidification Study (ILWAS) and the Regional Integrated Lake/Watershed Acidifications Study (RILWAS). Key advantages of the ALSC data base include its size, which allows for in-depth analyses for particular lake types of data subsets and the associated regional perspective provided by the survey.

SURFACE WATER CHEMISTRY

Lakes with low pH and low ANC are prevalent in the Adirondacks. For example, 26% of the waters surveyed by the ALSC had air-equilibrated pH <5.0, representing 8% of the lake surface area. Negative ANC values (indicating the occurrence of strong acids in the system) were measured in 26% of the lakes sampled, 11% of the lake area. Waters with low pH and ANC were found throughout the region, but were concentrated in the western and southwestern Adirondacks, in particular in the Oswegatchie-Black and Mohawk-Hudson watersheds (see Figure 1). This area is characterized by large numbers of small, high elevation lakes and receives the highest levels of precipitation in the region.

Influence of Mineral Acids

Acid neutralizing capacity is a measure of the net strong base in solution, or net strong acid if its value is negative. As a result, additions of strong acids or bases to surface waters have a direct effect on ANC. Depending on the initial ANC of the solution, this addition may also significantly influence pH.

Sulfate is the dominant mineral acid anion in ALSC lakes, indicating that sulfuric acid is the major source of mineral acidity. Some portion of this sulfate is attributable to acidic deposition. In the absence of all modifying processes other than evaporation, current levels of sulfate deposition in the Adirondacks would produce concentrations of sulfate from about 100 to 150 µeq/L in surface waters. Most drainage lakes (that is, lakes with an outlet) in the Adirondacks have sulfate concentrations between about 80 and 140 µeq/L. The retention of sulfate in Adirondack soils is minimal and thus, sulfate retention would have little effect on lake sulfate concentrations. In-lake and wetlands processes, however, may modify sulfate concentrations substantially, especially in systems with longer water residence times. Seepage systems (lakes with no outlet), for example, even with more evaporation, commonly have sulfate concentrations that are about 40 µeq/L lower than those in drainage lakes, most likely as a result of in-lake processes.

In contrast to sulfate, nitrate and chloride concentrations measured by the ALSC were generally low (except for chloride concentrations in lakes impacted by road salt). Other studies, however, have shown that higher concentrations of nitrate occur during hydrologic events, such as snowmelt and result in periodic decreases in ANC and pH.

The relative contributions of mineral and nonmineral acids to hydrogen ion (H⁺) concentration in Adirondack lakes with low pH were estimated based on the concentrations of mineral acid anions, base cations and DOC as measured by the ALSC. Of the 256 drainage lakes sampled by the ALSC with pH <5.0, mineral acids were estimated to support more than half the H⁺ present in 48% of the lakes, and over 90% of the H⁺ present in 83 (32%) of these lakes. In seepage lakes, on the other hand, mineral acids are apparently less important. Of the 92 seepage lakes sampled by the ALSC with pH <5.0, mineral acids were estimated to support more than half of the H⁺ present in 33% of the samples and over 90% of H⁺ in 13% of these lakes.

Influence of Non-mineral Acids

Other important acid-base systems in Adirondack waters include organic acids (products of the decay of plant and animal matter) and carbon dioxide. These acids can have a significant effect on lake pH levels. Results from the ALSC survey indicate that volatile acids (largely carbon dioxide) have the greatest effect on pH levels in waters with ANC > 50 µeq/L, depressing the lake pH by 0.3 to 1 pH units below the levels expected for air equilibration. Nonvolatile acids (primarily organic acids) have the greatest effect on pH levels in waters with ANC between about zero and 50 µeq/L, where pH depressions attributed to organic acids range from about 0.5 to 2.5 pH units.

In the past, organic acids were generally thought to act as weak acids capable of decreasing pH but not ANC. The ALSC data, however, indicate that in Adirondack waters dissolved organics also have strong acid components, as evidenced by the strong relationship between DOC and the anion deficit (i.e., the charge imbalance) in lakes with very high H⁺ concentrations (pH 4.0-4.5). Analysis of the ALSC data shows that organic acids contribute 4.5-5.0 µeq/L of strong acid to solution per milligram of DOC. The implication of this finding is most important. As a result of their strong acid component, organic acids, by themselves, can produce negative values for ANC. Thus, the common assumption that negative ANC values are necessarily indicative of acidification from mineral acids is incorrect. These observations were based on empirical data collected during extensive field investigations and no confirming laboratory experiments were conducted.

Finally, as a corollary to the statements regarding the importance of mineral acids, organic acids were estimated to support more than half the H⁺ present in 52% of the drainage lakes sampled with pH <5.0, and over 90% of the H⁺ in 41% of these lakes. In seepage lakes with pH <5.0, organic acids were estimated to support more than half of the H⁺ present in 67% of the 92 lakes sampled and more than 90% in 43% of these lakes.

Lake Classification

The acid-base status of surface waters is ultimately a function of the relative contributions of acids and bases. These acids are derived from both atmospheric deposition and natural processes, such as organic matter decay and the oxidation of organic nitrogen. The bases result from reactions occurring within the watershed soils (e.g., weathering and cation exchange reactions) and from atmospheric deposition. The degree to which the incoming precipitation interacts with the watershed soils and the routing of the water through these soils, are major determinants of the base supply rate. In watersheds with thick soils, acids - whether natural or from acidic deposition - will be largely neutralized by bases within the soils. In contrast, in watersheds with thin soils, much of the incoming precipitation passes only through the shallow organic horizons and into the lakes and streams. Such systems would generally be expected to be more responsive to increases or decreases in atmospheric acid inputs. Some seepage lakes, in particular "mounded" seepage lakes that are separated by relatively impermeable soil layers from the groundwater system, receive almost all of their water as direct precipitation, with little to no interaction with the soil. Mounded seepage lakes, might be expected to be quite responsive to changes in atmospheric inputs.

A classification scheme was developed to group the ALSC lakes according to these influential watershed features based on their chemical and hydrological characteristics measured during the ALSC survey. Thirty-nine percent of the lakes classified fell within the four classes expected to have the smallest amount of neutralization by the watershed: (1) mounded seepage lakes with low levels of DOC (3%); (2) mounded seepage lakes with high DOCs (3%); (3) drainage lakes in watersheds with thin till and with low DOC (19%); and (4) drainage lakes in thin-tilled watersheds with high DOC (14%).

Predicted Changes in pH

One of the objectives of the analysis of the ALSC water chemistry data was to determine how the H⁺ concentration (or pH) might change in response to changes in mineral acid inputs of DOC. Because of the large size of the ALSC data base, it was possible to conduct these sensitivity analyses based solely on the ALSC data.

The U.S. Environmental Protection Agency's Direct/Delayed Response Project (DDRP) involved the application of three dynamic simulation models to predict changes in ANC as a result of a 30% decrease in atmospheric deposition. All three models predicted increases of about 5 to 15 µeq/L for Adirondack lakes. Analyses were conducted, to determine the response of selected representative ALSC lakes from the four lowest ANC classes (high-and low-DOC thin-tilled drainage lakes and high-and low-DOC mounded seepage lakes) to a 15 µeq/L increase in ANC. The predicted increases in pH, assuming no change in DOC, ranged from 0.17 to 0.38 pH units. If increases in DOC occur simultaneously, as a limited number of field observations indicate, the changes in pH would be smaller. For example, an increase in DOC of about 2-4 mg/L would be sufficient to entirely offset the increase in pH resulting from the 15 µeq/L increase in ANC. Whether or not a DOC increase might occur in the Adirondacks with a decrease in deposition is not known. Some additional research is needed to answer this question.

FISH COMMUNITIES

Fish were caught in 76% of the lakes surveyed by the ALSC. No fish were caught in 24%, representing 7% of the lake area. Lakes with fish occurred throughout the Adirondacks, while fishless lakes were concentrated in the western and southwestern watersheds (Oswegatchie-Black and Mohawk-Hudson). Fifty-three fish species were collected in total. The number of species caught per lake (i.e., species richness) ranged from zero (fishless) to 13, with a median of one species per lake in the Oswegatchie-Black and Mohawk-Hudson watersheds, as compared to median values of 4-5 species per lake in the other three major Adirondack watersheds.

Characteristics of Lakes Without Fish

Lakes without fish have a diversity of characteristics, reflecting the large number of factors that determine the suitability of a lake for fish survival and reproduction. Many of the fishless lakes were small (nearly 70% were <4 ha) and shallow (60% had a mean depth <2m). Fishless lakes were also prevalent at high elevations, no fish were caught in 65% of the lakes surveyed at >600 m. The majority of lakes without fish had relatively low pH and ANC levels: 77% had pH <5.0; 92% had pH <5.5; and 77% had ANC <0 µeq/L.

The factors responsible for the absence of fish vary among lakes. Forty-six of the lakes surveyed by the ALSC were classified as bogs. No fish were caught in any of these lakes. The absence of fish from bogs probably reflects the combined effects of low concentrations of dissolved oxygen, poor spawning substrate and naturally low pH levels. Of the remaining 300 fishless lakes, 270 had pH levels below 5.5 and may potentially be fishless as a result of high levels of acidity. About half of these lakes (54%) had a high estimated fraction (>50%) of the H⁺ supported by organics, while the rest had less than half of the H⁺ supported organics and appear to be dominated instead by mineral acids. Not all of the low pH fishless lakes, would necessarily be able to support fish in the absence of high acidity. For example, 31 of the low pH lakes that were dominated by mineral acids were small (<4 ha) and occurred at elevations above 600 m. Some portion of these small, high elevation lakes would be expected to lack fish regardless of the lake pH (perhaps about 40% based on comparison to similar lakes with pH >5.5 that were fishless) because of low levels of dissolved oxygen during winter, freezing or inaccessibility to fish. Based on these analyses, the estimated number of surveyed waters for which mineral acids (and acidic deposition) appear to be the primary cause for the lake of fish is 100-113, or about 30% of the fishless lakes sampled by the ALSC.

Lake Characteristics Associated with Patterns of Fish Species Distribution

Changes in fish species composition and lower species richness may occur at more moderate environmental conditions than those that result in the total absence of fish from a lake. Factors associated with the presence and absence of selected fish species and with among-lake variations in species richness were examined using bivariate and multivariate regression, both for the Adirondack region as a whole and for individual drainage systems.

As expected, a large number of factors were significantly associated with fish status. For example, stocking, biological variables (e.g., the occurrence of potential predators or competitors), and silica (an index of groundwater inflow) were of particular importance in explaining patterns of brook trout presence/absence. Similar analyses for creek chub identified lake characteristics such as the presence of predators, small lake size, the absence of inlets and outlets, and lower flushing rates as significantly associated with the absence of creek chub. In each analysis, one or more variables related to surface water acid-base chemistry, generally lake pH, was also selected as a significant predictor of the occurrence of brook trout and creek chub. Thus, even after accounting as much as possible for the effects of the many other factors that influence fish, lakes without these fish species had lower pH than did lakes with fish present. Likewise, within individual drainage systems in the western Adirondacks, where low pH lakes are common, all of the fish species and groups of species examined exhibited uniformly lower frequencies of occurrence in waters with lower pH. These results support the conclusion that lake pH (and/or other acidity-related variables) plays an important role in restricting the distribution of fish species from at least some Adirondack lakes that would otherwise be suitable habitat.

In nonacidic lakes, species richness is highly correlated with lake area; larger lakes provide more divers habitat and generally support more fish species. Lakes with higher pH also tend to support more fish species than do lakes with lower pH. The relative effects of lake area and pH on species richness were quantified using a regression model and applying it to selected drainage systems in the western and eastern Adirondacks. In general, lakes in the western Adirondacks support fewer species than do lakes in the eastern Adirondacks. These differences appear to be largely attributable to differences in lake pH. Model applications suggest that the species-impoverished, acidic lakes prevalent in the western Adirondacks have the potential (based on the surface area/species richness relationship) to support more diverse fish communities, comparable to those present in the eastern Adirondacks, if pH levels in these western lakes were to increase significantly. Although other factors (e.g., productivity, access for fish immigration, disturbance) may locally modify the potential species richness for individual lakes, these other variables were not found to be strongly influential on a regional scale.

Historical Changes in Fish Communities

Changes in fish communities over time also provide insight into the potential effects of acid-base chemistry on fish resources. Survey data collected since the late 1920's by the NYSDEC were examined to assess the historical occurrence of fish in lakes sampled by the ALSC and probable causes for observed fish community and population trends.

Only 17 of the lakes that are now fishless were surveyed in the initial statewide biological surveys in 1929-34. At that time, about half of these lakes (9 lakes) had one or more fish species caught or reported to occur. The other half have apparently always been fishless, although it is possible that some lakes lost fish communities prior to these earliest surveys.

Most of the lakes sampled in 1929-34 were larger lakes at lower elevations, which today have relatively high pH. Fish communities in these lakes would be less likely to have been impacted by acidic deposition, compared to Adirondack lakes as a whole. To partially compensate for this bias, historical surveys conducted after 1929-34 but before 1970 were also considered in evaluating fish trends. Unfortunately, many of these surveys were not intended to comprehensively sample the fish community. Trends through time were assessed only for the subset of lakes with confirmed occurrence of a species in the past (i.e., caught in a survey prior to 1970). This expanded data set is also somewhat biased towards larger, lower-elevation lakes with higher pH, but to a lesser degree than for the subset of lakes surveyed in 1929-34. For this reason, results from these analyses provide evidence for fish population losses but are not suitable for regional estimates of the extent of such losses.

All available data, except for information on acid-base chemistry, were reviewed for each lake to identify apparent fish population losses and probable causes for observed losses. The numbers of lakes classified as having lost fish populations apparently as a result of acidification varied between zero (for rainbow trout, brown trout, largemouth bass and northern pike) to 44 (for brook trout, representing 11% of the lakes surveyed by the ALSC in which brook trout were caught prior to 1970). Acid-sensitive minnow species had apparently disappeared as a result of acidification from 19% of the lakes with evidence for historical occurrence; common shiner from 15% of the lakes; creek chub and lake trout from 10%; and white sucker from 6% of the lakes with confirmed historical occurrence. Among species variations in the proportion of fish populations lost are a function of several factors: (1) the sensitivity of the species to acidic waters; (2) the degree to which the species tends to naturally occur in those lakes more susceptible to acidification; and (3) the susceptibility of the species to capture during the historical and ALSC surveys. Of the 858 lakes evaluated for historical trends (i.e., all lakes with at least some historical information), 140 (16%) had lost one or more fish populations apparently as a result of acidification. Lakes considered to have lost fish populations as a result of acidification consistently had lower present-day levels of pH than did lakes still supporting the fish species.

Acidification is not the only factor or necessarily the dominant factor influencing Adirondack fish communities. The numbers of lakes for which fish population losses were attributed to factors other than acidification actually exceeded those for which acidification was rated as the primary cause. Common alternative explanations that could be identified with the available data include lake reclamation with rotenone, changes in stocking policy and the introduction of potential competitors or predators. In contrast to acidification, however, these other factors often affected only some fish species within the fish community (resulting in species replacements rather than a net decline in species richness) or resulted directly from fishery management practices designed to improve fishing opportunities in the region.

Fisheries Resource at Risk

As noted above, lakes classified as mounded seepage lakes or thin-tilled drainage lakes (for both high and low DOC) would be expected to be more responsive to changes in atmospheric inputs of strong acids. Thirty-nine percent of the lakes sampled by the ALSC were classified into these groups. Of the lakes in which the ALSC caught fish, only 27% were mounded seepage or thin-tilled drainage lakes. Twenty-five percent of the lakes with brook trout fall within these two lake classes. Other sport fish species, occur relatively infrequently in these lake classes: 10% of the lakes with lake trout caught by the ALSC; 6% of the lakes with smallmouth bass or largemouth bass; 3% of the lakes with rainbow trout or brown trout; and 3% of the lakes with northern pike. The majority of lakes currently supporting sport fish would appear to be relatively insensitive to further acidification, i.e., they were classified as drainage lakes in watersheds with intermediate or thick till or influenced by carbonate minerals, or as flow-through seepage lakes.

With a reduction in lake acidity, some portion of these lakes would gradually recover fish populations. Quantitative regional analyses of fish recovery were outside of the scope of this project. However, the historical trend analyses provide an index of the types of species likely to be recovered with improvements in water quality. Of the fish populations rated as having been lost apparently as a result of acidification, brook trout accounted for 28%, white sucker for 14%, creek, chub 13%, common shiner 11%, brown bullhead 9%, pumpkinseed 7%, northern redbelly dace 6%, golden shiner 5%, and lake trout, smallmouth bass, yellow perch, and lake whitefish, <5% each. The majority of lakes affected tend to be smaller lakes at higher elevations, which historically supported fairly simple fish communities.

Brook trout is the most important indigenous sport fish in Adirondack waters and was the second most common species caught by the ALSC (occurring in 52% of the lakes with fish present). Anecdotal accounts of brook trout occurrence in the 1800's indicate that the species was also widespread historically. Although native to the Adirondacks, brook trout have been introduced into many waters where they must be currently maintained by stocking because of insufficient spawning substrate. Brook trout typically spawn in tributaries and over areas of groundwater upwelling, and high levels of silica were found to be associated with the potential for self-sustaining brook trout populations. Concentrations of silica measured by the ALSC were used, to estimate the potential for brook trout natural reproduction within waters that currently have low pH (<5.0) and no brook trout. Relatively few of these lakes (10-15%) were classified, based on their silica concentration, as having the potential to support lake-spawning populations of brook trout. Thus, even if acidity levels were reduced, it would appear that under present-day conditions (e.g., of fishing pressure) many of these lakes would require supplement stocking to support brook trout.

Finally, perhaps 50-70% of the lakes that are currently fishless appear to always have been fishless, or are fishless primarily for reasons other than mineral acidity, and would likely remain fishless even with reductions in acidic deposition.

Print This Page