Detecting acid precipitation impacts on lake water quality

Loftis, Jim C.; Taylor, Charles H.

doi:10.1007/bf01874958

Cited by 13 publications

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Trends in acid-neutralizing capacity (ANC), base cations, sulfate, and nitrate were analyzed using the nonparametric Seasonal Kendall test (SKT; Hirsch et al 1982). The SKT has become something of a standard for detecting site-specific trends in water quality data, largely because it can accommodate the nonnormality, missing and censored data, and seasonality that are common in data of this type, but it is nevertheless a powerful (in a statistical sense) trend test (Loftis and Taylor 1989). Because the SKT is sensitive to serial correlation, we use a modified form of the test described by Hirsch and Slack (1984), which uses a covariance term to correct the seasonal Z scores for serial dependence.…”

Section: Methodsmentioning

confidence: 99%

Can Site-Specific Trends be Extrapolated to a Region? An Acidification Example for the Northeast

Stoddard

Driscoll

Kahl

et al. 1998

Ecological Applications

View full text Add to dashboard Cite

In the absence of true regional data on changes in the acid/base status of lakes in the northeastern United States, we explore the possibility of using site-specific trends information from a judgment sample of lakes to assess the efficacy of the Clean Air Act Amendments. A meta-analytical technique is used to combine trends results from 44 Long-Term Monitoring (LTM) lakes in the Northeast for the period 1982-1994, with the goal of producing estimates of overall trends in the region. The lakes are subdivided into subpopulations (High ANC, Intermediate Till Drainage, Thin Till Drainage and Perched Seepage lakes) on the basis of their expected response to changes in acidic deposition, and they appear to represent the most acid-sensitive of these lake classes well. While the overall tendencies in the trends are as expected (e.g., most of the recovery is observed in the most sensitive subpopulations), there is significant trend heterogeneity among the lakes within most of the subpopulations; this heterogeneity prohibits the summarizing of trends at the regional level (i.e., for all of the Northeast). This heterogeneity is explained by differences in the responses of lakes in two subregions (Adirondacks vs. New England), and we present trends results separately for each subpopulation within these two subregions. All subpopulations in both subregions showed decreasing trends in sulfate concentrations, probably a reflection of decreasing trends in sulfur deposition in the region. Few trends in nitrate concentrations were observed. Recovery (as evidenced by increasing trends in acid-neutralizing capacity) was evident in Thin Till and Intermediate Till Drainage lakes in New England, but not in the Adirondacks. Most groups of lakes exhibited downward trends in base cations (⌺[Ca 2ϩ ϩ Mg 2ϩ ϩ Na ϩ ϩ K ϩ ]); the magnitudes of these trends were always greater in Adirondack lakes than in similar New England lakes. This suggests that the depletion of soil cation pools in the Adirondacks may be responsible for some of the differences in recovery between Adirondack and New England lakes. While export of base cations may be the key difference producing different trends results in the two subregions, the site-specific nature of the trends, and their possible lack of regional representation, should be considered in interpreting the overall results.

show abstract

Section: Methodsmentioning

confidence: 99%

Can Site-Specific Trends be Extrapolated to a Region? An Acidification Example for the Northeast

Stoddard

Driscoll

Kahl

et al. 1998

Ecological Applications

View full text Add to dashboard Cite

show abstract

“…Hyman et al (1983) approached this problem by converting periodic changes in ichthyoplankton density to a linear form (Le., grouping samples into "day" and "night" categories to account for diel cycles) and by using a sinusoidal regression model to generate residuals free of periodic components. In contrast, Loftis and Taylor (1989) used a "deseasonalization" technique to detrend lake water quality data, This technique could also be used to remove seasonal or diel trends from ichthyoplankton data.…”

Section: Introductionmentioning

confidence: 99%