Spatio-temporal variability of soil respiration of forest ecosystems in China: influencing factors and evaluation model.

Understanding the influencing factors of the spatio-temporal variability of soil respiration (R (s)) across different ecosystems as well as the evaluation model of R (s) is critical to the accurate prediction of future changes in carbon exchange between ecosystems and the atmosphere. R (s) data from 50 different forest ecosystems in China were summarized and the influences of environmental variables on the spatio-temporal variability of R (s) were analyzed. The results showed that both the mean annual air temperature and precipitation were weakly correlated with annual R (s), but strongly with soil carbon turnover rate. R (s) at a reference temperature of 0°C was only significantly and positively correlated with soil organic carbon (SOC) density at a depth of 20 cm. We tested a global-scale R (s) model which predicted monthly mean R (s) (R (s,monthly)) from air temperature and precipitation. Both the original model and the reparameterized model poorly explained the monthly variability of R (s) and failed to capture the inter-site variability of R (s). However, the residual of R (s,monthly) was strongly correlated with SOC density. Thus, a modified empirical model (TPS model) was proposed, which included SOC density as an additional predictor of R (s). The TPS model explained monthly and inter-site variability of R (s) for 56% and 25%, respectively. Moreover, the simulated annual R (s) of TPS model was significantly correlated with the measured value. The TPS model driven by three variables easy to be obtained provides a new tool for R (s) prediction, although a site-specific calibration is needed for using at a different region.