We know little about how ecosystems of different complexity will respond to
global warming(1-5). Microcosms permit experimental control over species c
omposition and rates of environmental change. Here we show using microcosm
experiments that extinction risk in warming environments depends on trophic
position but remains unaffected by biodiversity. Warmed communities dispro
portionately lose top predators and herbivores, and become increasingly dom
inated by autotrophs and bacterivores. Changes in the relative distribution
of organisms among trophically defined functional groups lead to differenc
es in ecosystem function beyond those expected from temperature-dependent p
hysiological rates. Diverse communities retain more species than depauperat
e ones, as predicted by the insurance hypothesis, which suggests that high
biodiversity buffers against the effects of environmental variation because
tolerant species are more likely to be found(6,7). Studies of single troph
ic levels clearly show that warming can affect the distribution and abundan
ce of species(2,4,5), but complex responses generated in entire food webs g
reatly complicate inferences based on single functional groups.