Hierarchical regulation of pelagic invertebrates in lakes of the northern Great Plains: a novel model for interdecadal effects of future climate change on lakes

Endorheic lakes of the northern Great Plains encompass a wide range of environmental parameters (e.g., salinity, pH, DOC, Ca, nutrients, depth) that vary 1000-fold among sites and through the past 2000 years due to variation in basin hydrology and evaporative forcing. However, while many environmental parameters are known to individually influence zooplankton diversity and taxonomic composition, relatively little is known of the hierarchical relationships among potential controls or of how regulatory mechanisms may change in response to climate variation on diverse scales. To address these issues, we surveyed 70 lakes within a 100 000 km2 prairie region to simulate the magnitude of environmental change expected to occur over 100-1000 years and to quantify the unique and interactive effects of diverse environmental parameters in regulating pelagic invertebrate community structure at that scale. Multivariate analyses showed that salinity was the principal correlate of changes in invertebrate composition among lakes, with a sequential loss of taxa between salinities of 4 and 50 g total dissolved solids L-1 until one to two species predominated in highly saline systems. In contrast, changes in the concentrations of Ca2+ and other mineral nutrients exerted secondary controls of invertebrate assemblages independent of salinity, whereas lake depth provided a tertiary regulatory mechanism structuring species composition. In contrast to these large-scale hierarchical patterns, seasonal surveys (May, July, September) of a subset of 21 lakes in each of 2003-2005 revealed that annual meteorological variation had no measurable effect on pelagic invertebrates, despite large differences in temperature, precipitation, and evaporation arising from regional droughts. Together these findings show that pelagic invertebrate communities in saline lakes are resilient to interannual variability in climate, but suggest that lakes of the northern Great Plains may provide a sensitive model to forecast centennial effects of future climate change.