Phytoplankton responses to meteorological and hydrological forcing at decadal to seasonal time scales

One of the challenges for predicting global change effects on aquatic ecosystems is the vague understanding of the mechanisms of multiple controlling factors affecting phytoplankton dynamics at different time scales. Here we distinguish between hydrometeorological forcing of phytoplankton dynamics at time scales from days to decades based on a 54-year monthly phytoplankton time series from a large shallow Lake Võrtsjärv (58 160N, 26 020E) in Estonia, combined with daily data on forcing factors— thermal-, wind-, light- and water-level regimes. By using variance partitioning with linear mixed effect modelling (LME), we found a continuum from the large dominant K-selected filamentous cyanobacteria with strongest decadal scale variation (8–30%) to r-selected phytoflagellates with large stochastic variability (80–96%). External forcing revealed strong seasonal variation (up to 80%), while specifically water level and wind speed had a robust decadal variation (8% and 20%, respectively). The effect of external variables was proportionally manifested in the time scales of phytoplankton variation. Temperature, with a clear seasonal variation, had no impact on the dominant cold tolerant filamentous cyanobacteria in Lake Võrtsjärv. We found the LME as a reliable method for resolving the temporal cross-scale problem. It yielded quantitative results that matched our intuitive understanding of the dynamics of different variables. Supplementary Information The online version of this article (https://doi.org/10.1007/s10750-021-04594-x) contains supplementary material, which is available to authorised users. This study was funded by MANTEL ITN (Management of climatic extreme events in lakes and reservoirs for the protection of ecosystem services) through European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 722518 and by the Estonian Research Council grants (PRG1266 and PRG1167). We would also like to thank Estonian Environment Agency for the long-term data used on this study. This study was funded by MANTEL ITN (Management of climatic extreme events in lakes and reservoirs for the protection of ecosystem services) through European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 722518 and by the Estonian Research Council grants (PRG1266 and PRG1167). We would also like to thank Estonian Environment Agency for the long-term data used on this study.