Testing the temperature dependence of stocks and fluxes in an aquatic food web

Background/Question/Methods

Understanding how ecosystem structure and function depends on environmental temperature is critical to preserving, utilizing and sustaining ecosystems. Ecosystem function is based in part on food web structure and function, which vary with the body temperatures of constituent organisms. Increasing evidence suggests a fundamentally weaker response to warming for primary production relative to heterotrophic metabolism. For food webs dominated by ectotherms, environmental temperature should therefore change the flux of energy and materials in food webs, and potentially lead to changes in food web structure in a predictable way. Theory predicts that fluxes through the food web increase with warming but standing stocks at higher trophic levels decline as metabolic rates increase. Important questions remain: Does this prediction apply to food webs with three trophic levels? Can the functional responses of flux or standing stocks can be modified by changes in body size? We tested these questions by warming planktonic freshwater food webs for 11 weeks in summer 2011 at an outdoor facility at the University of British Columbia. In a regression design, 30 cylindrical 300-L mesocosms spanned a temperature range of about 10 C. In early June, clean water was inoculated with identical planktonic assemblages (phytoplankton, zooplankton and notonectids).

Results/Conclusions

As predicted, ecosystem-level respiration rates (oxygen flux) increased and NPP declined with increasing temperature. After one week, NPP increased with warming, but subsequent weekly estimates of the temperature-NPP relationship are consistently negative. Standing stocks of primary producers (Chlorophyll A concentration) initially declined with warming, despite the increase primary productivity, possibly due to increased zooplankton abundance. However, after the first week, chlorophyll concentration remained low with a weak or absent relationship to temperature. Preliminary analysis of zooplankton abundance indicates no trend with temperature at week 11, but patterns suggest population dynamics at different phases. Predators persisted at cooler temperatures, but died at higher temperatures.

Few studies have reported simultaneous data on stocks and fluxes sufficient to test hypotheses for how metabolic scaling affects food web properties. We find that fluxes (O2) vary with temperature as predicted, and primary producer standing stock is invariant with temperature (after initial dynamics), consistent with a basic temperature-dependent consumer-prey model. Data on fluxes and stocks at all trophic levels (including nutrients) allow revision of theory for how temperature affects food webs. Strong theory is required to support sustainable use of ecosystems in a changing climate.