- Department of Aquatic Sciences and Assessment, Swedish University of Agricultural Sciences
- South China Institute of Environmental Science (SCIES)
Zhao, Xuemin; Drakare, Stina; Johnson, Richard K.
Phytoplankton is a functionally important taxonomic group commonly used in the assessment of lakes. Using data from minimally disturbed lakes (n = 151) and relatively easily obtained environmental variables, we modeled the probability of taxon occurrence and taxon-specific biomass (mg L-1) of 183 phytoplankton taxa using variables characterizing spatial context, ecosystem size and land cover. Data from lakes putatively affected by one or more pressures (n = 162) were used to evaluate responses to anthropogenic stress. Longitude and latitude were the two best predictors of phytoplankton biomass, followed by the percent of catchment area classified as coniferous forest area and altitude. Model accuracy was highest for relatively predominant taxa, and taxa with relatively wide distributions did not improve model accuracy. Assemblage similarity (1-BC index) and the phytoplankton trophic index (PTI) calculated using predicted and observed phytoplankton assemblages were used to evaluate model performance. Significant changes in phytoplankton assemblages were associated with agricultural and forestry pressures. Encouragingly the PTI index responded as predicted to agricultural land use but not forestry, implying that PTI was a robust indicator of agricultural effects and that forestry effects on assemblage composition were not related to increased nutrients. Combined, our findings showed that phytoplankton assemblages can be reliably predicted using a few, often readily available, environmental variables, and that predictions of assemblage composition and taxon-specific biomass and biological metrics based on phytoplankton taxon-specific biomass are reliable indicators of anthropogenic pressures, supporting their use in lake management.
Taxon-specific modelling; Random forest; Environmental pressures; Lakes; Phytoplankton; Biomass
2019, Volume: 97, pages: 447-456
Publisher: ELSEVIER SCIENCE BV
SDG6 Clean water and sanitation