Uncertainties in the temperature sensitivity of decomposition in tropical and subtropical ecosystems: Implications for models

Tropical ecosystems play a central role in the global carbon cycle. Large changes in tropical temperature over geologic time and the significant responses of tropical ecosystems to shorter-term variations such as El Nino/La Nina argue for a robust understanding of the temperature sensitivity of tropical decomposition. To examine the responsiveness of heterotrophic respiration to temperature, we measured rates of heterotrophic respiration from a wide range of tropical soils in a series ol laboratory incubations. Under conditions of optimal soil water and nonlimiting substrate availability, heterotrophic respiration rose exponentially with rising temperature. The mean Q(10) measured across all temperature ranges in these short-term incubations was 2.37, but there was significant variation in eros across sites. The source of this variation could not be explained by soil carbon or nitrogen content, soil texture, site climate, or lignin to nitrogen ratio. At the beginning of the incubation, heterotrophic respiration increased exponentially with temperature for all sites, despite the fact that the fluxes differed by an order of magnitude. When substrate availability became limiting later in the incubation, the temperature response changed, and heterotrophic response declined above 35 degreesC. The documented changes in temperature sensitivity with substrate availability argue for using temperature relationships developed under optimal conditions of substrate availability for models which include temperature regulation of heterotrophic respiration. To evaluate the significance of this natural variation in temperature control over decomposition, we used the Century ecosystem model gridded for the areas between the tropics of Cancer and Capricorn. These simulations used the mean and upper and lower confidence limits of the normalized exponential temperature response of our experimental studies. We found that systems with the lowest temperature sensitivity accumulated a total of 70 Pg more carbon in soil organic carbon and respired 5.5 Pg yr(-1) less carbon compared to the systems with the highest sensitivity.