State of the art
Abiotic forcing of aquatic ecosystems acts at different spatial and temporal scales (Pinel-Alloul et al. 1988; Giller et al. 1994). Not only physical forcing but also nutrients, a key factor determining the growth, biomass and community structure in lakes, are distributed heterogeneously. The continuum of spatial variation in physical and chemical properties affects primary production and trophic interactions and thus leads to spatial and temporal patterns in plankton community (Neill 1994). The coupling between spatially heterogeneous abiotic forcing and ecological processes modifies or introduces additional spatial heterogeneity. Spatial heterogeneity has prominent effects on ecosystem productivity and population and community dynamics (Serra et al. 2007; Ryan et al. 2010). Changing conditions cause different response patterns (e.g. regime shifts) of the ecosystem depending on the abiotic conditions and the community structure of the ecosystem (Genkai-Kato et al. 2012; Hilt 2015). The prediction of key drivers that cause different response patterns under consideration of spatial heterogeneities will increase the understanding of lake ecosystems.
Palaeolimnological approaches provide information on the past occurrence and relative abundance of e.g. different algal groups (pigments) and macrophytes (propagules) (Battarbee et al. 2005). These sediment records can be used in combination with observational data sets to determine the different response patterns (reversibility, resilience, regime shifts) of the ecosystem to e.g. spatially heterogeneous nutrient distribution between lake sub-basins. Combined hydrodynamic-ecosystem models are a powerful tool with predictive capability for physical and biological processes under changing environmental conditions, e.g. spatial and temporal variation in phytoplankton abundance and community structure (Tsiaras et al. 2008; Hillmer et al. 2008) as well as growth, recolonization and dispersal of macrophytes (van Nes et al. 2003).
The applicants have investigated large- and small-scale as well as short- and long-term processes that cause spatio-temporal heterogeneity between the lake littoral zone and the pelagic zone (Hofmann et al. 2010, 2011). Light, internal waves, and predator avoidance were identified as important criteria for the distribution of phytoplankton and zooplankton in lakes (Lorke et al. 2008; Hingsamer et al. 2014). We also studied horizontal mixing and dispersion (Peeters and Hofmann 2015) an important process affecting horizontal patchiness. Other field investigations and model approaches have shown short-term and long-term effects of temperature and vertical mixing on the abundance and temporal and spatial growth of plankton (Peeters et al. 2007a; b; Straile et al. 2015b). A modelling study considering 28 years of re-oligotrophication in Lake Constance demonstrated that bottom-up control of phytoplankton became increasingly important compared to top-down control (Kerimoglu et al. 2013). We are able to distinguish between different phytoplankton groups in-situ by using a multi-spectral fluorescence and acoustic backscatter probes (Hofmann and Peeters 2013). The project will take advantage of unique data sets on phytoplankton (biweekly sampling interval at three stations of Lower Lake Constance (LLC); and repeated, horizontally resolved macrophyte distributions conducted in 1967, 1978, 1993, and between 2006 and 2010 covering the change in trophic state of LLC.
Proposed project and role within the RTG
The project will assess the consequences of gradients in nutrient concentrations and of trophic change (oligotrophic to eutrophic and back to oligotrophic) for the spatial distribution of phytoplankton functional groups and abundance under consideration of lake hydrodynamics. This way, the project will add a spatial understanding, identify key processes structuring the phytoplankton distribution, and response patterns during trophic perturbations. The doctoral researches will conduct simulations with a coupled 3D hydrodynamic and aquatic ecosystem model in LLC which is subdivided into three basins with different histories of nutrient loading. The model approaches will be combined with pigment analysis on spatially resolved sediment cores (long-term change in phytoplankton pigment diversity and abundance) and water samples (seasonal change) as well as with data obtained from seasonal resolved field surveys and long-term monitoring programs from all three sub-basins of LLC providing information on the present and past occurrence, relative abundance, and distribution of phytoplankton and macrophytes as well as information on abiotic parameters and other ecological state variables. This setup will be used to test the following hypothesis: 1) Phytoplankton resilience is maintained by changes in diversity of phytoplankton expressed in changes in pigment diversity and possibly changing overall resource use efficiency – pigment diversity is higher under oligotrophic compared to eutrophic conditions; 2) The pattern of the spatial distribution of summer phytoplankton between sub-basins of LLC depends on the nutrient load in the inflows rather than on the trophic state thus leading to a resilience of the spatial patterns during eutrophication and oligotrophication, respectively; 3) The spatiotemporal differences in phytoplankton communities between lake sub-basins may however be sensitive to the trophic state, because changes in the composition of the macrophyte community growing on the sill between the basins (e.g. between Gnadensee and Zellersee) reduce the horizontal exchange across the sill and thus alter inter-basin mixing of nutrients. The growth of macrophyte Potamogeton, favoured under eutrophic conditions, causes large hydraulic resistance that may essentially disconnect Gnadensee from Zellersee, and thus result in an alternate state compared to oligotrophic conditions; 4) Changes in the macrophyte community may also lead to resilience in the summer phytoplankton biomass with respect to trophic state, i.e. during eutrophic conditions Potamogenton outcompetes Characeens and produces increased macrophyte biomass which leads to a reduction of nutrient and light availability for the growth of phytoplankton especially in the littoral zones. During oligotrophic conditions this relationship is reversed.
The numerical-model based studies conducted in B1 and A4 have similar demands on model parametrization and profit from the exchange and comparison of model results on phytoplankton dynamics using a 1D or 3D approach. Projects A5 and A4 cross-fertilize in the analysis of the long-term monitoring data of phytoplankton and macrophytes in LLC. The projects A1, B2, and C3 profit from jointly conducted sampling, dating, and the analysis of sediment cores. The temporal change in phytoplankton biomass and pigment dynamics during trophic change will be set in relation to the diversity dynamics of phytoplankton investigated in A1. Seasonally resolved data on phytoplankton community composition (from routine sampling) and pigments (from sediment traps) can be used as reference for the metagenomics analysis in A2.