Table 1.
Predicted influence of selection processes on trait community structure, assuming that the community includes only one habitat.
Figure 1.
Spring bloom dynamics in Lake Zurich over the period of study (45 days between March 23 and May 6, 2009).
(A) water temperature (black solid line) and conductivity (grey dashed line); (B) free available phosphorus (black solid line) and nitrates (grey dashed line); (C) depth of the Chl-a maximum (the sampled community, black solid line) and oxygen levels (grey dashed line); (D) Chl-a concentration (black solid line) and dissolved organic carbon (grey dashed line).
Figure 2.
Spring bloom dynamics in Lake Zurich over the period of study (45 days between March 23 and May 6, 2009) at the depth of Chl-a maximum (Fig. 1).
(A) Total biovolume of phytoplankton cells measured by flow-cytometry (black solid line) and phytoplankton concentration obtained by microscopic counts (grey line). (B) Concentration of pelagic herbivore ciliates (black line) and counts for Cladocerans (○) and Copepods (⋄). Richness (C) and pairwise Bray-Curtis dissimilarity (calculated for adjacent time points, D) of phytoplankton groups derived by flow-cytometry analysis compared to taxonomic groups obtained by microscopy.
Table 2.
Results of the two-step fourth corner analysis (Pearson’s correlation coefficients) performed on Cytobuoy-derived phytoplankton traits and environmental variables.
Figure 3.
Summary of trait-based tests for community assembly at each sampling date of our study period: a) first principal component of Cytobuoy-derived phytoplankton traits (Table S1); b) size of phytoplankton particles; c) Chl-a particle fill; d) phycocyanin particle Fill (see Methods and Table 2).
Dots indicate statistically significant deviations from null-model expectations (Wilcoxon p-value >0.05): red = SDNDr (even spacing of traits); green and blue = distribution range and mean, respectively (environmental filtering). The grey shaded area emphasizes the period of herbivore ciliate grazing (phase 2 in the text).
Figure 4.
Correlations between average values of hypothesized habitat-occupancy and species-interaction traits for the 22 groups derived in this study by flow-cytometry.
A) relationship between average size and PC3 (phytoplankton FL); B) relationship between PC1 (phytoplankton morphology) and Chl-a pigment fill. For PC loadings see Table S1. Dashed lines represent linear regression models (all fits and coefficients were significant at p<0.05). The triangular data point in A and B represents a group characterized by large P. rubescens filaments (data not shown), which appeared to be an outlier in all trait correlations and was excluded from the corresponding regression models.