COS 253-4 - Macroecological response to temperature: scaling up from species to ecosystem level

Abstract: Temperature affects metabolic rates of every species on Earth because of the thermodynamics of chemical reactions. Metabolic rates in turn affect many other biological rates, including population-level rates like birth and death rates, and evolutionary rates due to the effects of temperature on mutation rates. Temperature furthermore causes variation in basal metabolic rates between species living in different environments, as has been hypothesized by the Metabolic Theory of Ecology. However, understanding how temperature affects ecosystem-level characteristics such as stability, diversity, and abundance remains a major challenge. Many factors, such as differing adaptations to temperature and ecological interactions between species within foodwebs, make it difficult to predict or measure community-level thermal response. To isolate the possible mechanisms of large-scale ecosystem thermal responses and the factors that could obscure them, we modify the Tangled Nature Model, an evolving ecosystem model of foodweb interactions and resource constraints, to include species-specific thermal response curves for birth, death, and mutations. We allow numerous randomly generated ecological communities to evolve in silico, and analyse the statistics of these model outputs. We test a case where all species have the same thermal response curve, one where thermal optima can evolve through mutations, and one where evolving species-specific thermal response curves are scaled by the exponential curve predicted by the Metabolic Theory of Ecology. These experiments disentangle the effects of temperature, adaptation, thermodynamic constraints on adaptation, and interspecific interactions. Community survival probability is found to generally correspond to the underlying thermal response curves. The frequency of ecosystem reconfiguration events, and associated bursts in speciation, shows similar dependence on temperature. In contrast, the abundance and diversity of those ecosystems that do survive show thermal dependence initially, but ultimately they converge to similar values across a wide range of temperatures. This work distinguishes between possible mechanisms of thermal effects on macroecological patterns, and suggests that observed macroecological patterns such as latitudinal variation in diversity could arise from differential community survival or from thermal dependence of mutation. The tools developed here can be further developed to investigate how the effects of temperature scale from the individual level up to the ecosystem level.