Temperature mean and variance alter phytoplankton biomass and biodiversity in a long-term microcosm experiment

Global predictions of increasing mean temperature and its variance result in concerns about the fate of biodiversity under future climatic conditions. On the local scale of interacting species, the response of diversity to warming will depend much on the change in interspecific interactions such as competition or predation. Using a phytoplankton model community, we conducted a 2 x 2 x 2 factorial long-term microcosm experiment (16 months) with an underlying seasonal temperature profile. We manipulated the mean temperature (T-mean), the temperature variance (T-var), and the presence of consumers (ciliates) and monitored treatment effects on algal biomass, species composition and species richness. Temperature effects on algal biomass depended on the seasonal development and consumer presence. Algal biomass decreased with increasing T-mean at consumer absence, but increased at consumer presence. Overall, algal biomass decreased with increasing T-var. Consumer presence reduced algal biomass from summer to winter, but then ciliates and the consumer effect disappeared. Almost all treatment combinations collapsed to monocultures after 16 months, but extinction occurred faster at higher T-mean (especially at consumer absence) and slower at consumer presence (especially at higher T-mean). Contrasting our predictions, increasing T-var reduced richness and increased extinction rate. The treatment effects on biomass and richness were not independent, as algal species richness and biomass were positively correlated. Moreover, accelerated loss of species was consistently correlated to higher temporal variability in biomass. In conclusion, altered temperature regimes strongly affected algal biomass and diversity by interdependently altering competitive and consumer interactions.