Assessing anthropogenic impact on boreal lakes with historical fish species distribution data and hydrogeochemical modeling.
Bottom Line: Quantifying the effects of human activity on the natural environment is dependent on credible estimates of reference conditions to define the state of the environment before the onset of adverse human impacts.In 2010, MAGIC predicted chemical recovery in 50% of the lakes, however roach only recolonized in five lakes after 1990, showing a lag between chemical and biological recovery.Based on our results, we show how the conceptual model can be used to understand and prioritize management of physico-chemical and ecological effects of anthropogenic stressors on surface water quality.
Affiliation: Department of Aquatic Sciences and Assessment, Swedish University of Agricultural Sciences, Box 7050, Uppsala, SE-750 07, Sweden.Show MeSH
Related in: MedlinePlus
Mentions: Data on presence/absence of roach were analyzed to determine if this species was present historically and if/when it went extinct during the acidification period. Observations were grouped into three periods: (i) ‘beginning of acidification period’ (before 1960), (ii) ‘heavy acidification period’ (1960–1990), and (iii)‘recovery period’ (1990–2012). Only lakes where roach was present during period 1 and with additional data from period 2 were analyzed further. For many lakes, there were records from several different surveys in each period, which sometimes provided contradictory information, perhaps due to low-density roach populations which were undetected in some gillnet surveys. Thus, we excluded all lakes with single observations during periods 1 and 2, and used probabilistic criteria when classifying lakes as acidified or not. Specifically, we classified a lake as acidified if the proportion of surveys that reported roach was ≥75% before 1960 and ≤25% between 1960 and 1990. A lake was classified as nonacidified if more than 90% of the surveys reported roach during both periods. Applying these criteria produced a dataset of 267 lakes, from which 121 were classified as acidified and 146 as nonacidified. Of the total number of lakes that were classified according to the criteria for roach, 85 lakes were also modeled using MAGIC pH reconstruction and those were the lakes used in this study. The 85 lakes covered the natural spatial distribution of roach in Sweden including the southern half and the northern east coast (Fig.1).
Affiliation: Department of Aquatic Sciences and Assessment, Swedish University of Agricultural Sciences, Box 7050, Uppsala, SE-750 07, Sweden.