Variable temperature effects between heterotrophic stream processes and organisms

Abstract

Abstract Temperature is known to stimulate metabolism with cascading effects on multiple biological processes. These effects may, however, vary across processes, types of organisms or levels of biological organisation. They can also vary with nutrient availability, with potentially stronger temperature effects when nutrients are not limiting. This context dependence of temperature effects on processes challenges our ability to anticipate their consequences on ecosystems in a changing world. In headwater streams, the decomposition of allochthonous leaf litter, driven by both microbial decomposers and invertebrates, is known to respond to both temperature and nutrient availability. These food webs are highly tractable and a useful model system to investigate the variations of temperature effects on processes across types of organisms (microbes versus invertebrates), resource availability levels (nutrient concentration), and levels of biological organisation (from individual to ecosystem). In a microcosm experiment, we measured the effects of temperature and nitrogen availability (four levels each) on respiration rates of litter-consuming microbes and invertebrates and their decomposition activity in different contexts of food web complexity. The latter included one treatment without invertebrate detritivore (microbial decomposers only), three single invertebrate taxa (Gammarus, Potamophylax, and Sericostoma) treatments, and one mixed invertebrate taxa treatment (three‐species altogether). Microbial processes increased nearly exponentially with temperature (Arrhenius model, activation energy (± 95% confidence interval) = 0.56 ± 0.53 and 1.00 ± 0.23 eV for litter decomposition and respiration), while invertebrate‐driven processes increased (activation energy from 0.47–1.15 eV) up to a maximal value at an intermediate temperature (c. 11–15°C depending on species and process), above which process rates decreased. By contrast, litter consumption in mixed invertebrate species treatments was not significantly influenced by temperature, because of a negative effect of species mixing occurring above 12°C. Nitrogen had a weaker influence, only slightly stimulating litter consumption by mixed‐species invertebrates, which limited the scope for synergies with temperature effects. Our results raise issues about how aquatic litter consumers meet their energy requirements at high temperature and suggest that a general consequence of warming could be loss of carbon through mineralisation in headwater stream food webs. In several aspects, our results deviate from expectations based on universal relationships between temperature and individual metabolism (e.g. metabolic theory of ecology), suggesting that we may need to develop less simplistic assumptions to predict the consequence of warming on ecosystem processes.