Interannual variability in soil respiration from terrestrial ecosystems in China and its response to climate change.

Soil respiration is an important process in terrestrial carbon cycle. Concerning terrestrial ecosystems in China, quantifying the spatiotemporal pattern of soil respiration at the regional scale is critical in providing a theoretical basis for evaluating carbon budget. In this study, we used an empirically based, semi-mechanistic model including climate and soil properties to estimate annual soil respiration from terrestrial ecosystems in China from 1970 to 2009. We further analyzed the relationship between interannual variability in soil respiration and climatic factors (air temperature and precipitation). Results indicated that the distribution of annual soil respiration showed clear spatial patterns. The highest and lowest annual soil respiration rates appeared in southeastern China and northwestern China, respectively, which was in accordance with the spatial patterns of mean annual air temperature and annual precipitation. Although the mean annual air temperature in northwestern China was higher than that in some regions of northeastern china, a greater topsoil organic carbon storage in northeastern China might result in the higher annual soil respiration in this region. By contrast, lower temperature, less precipitation and smaller topsoil organic carbon pool incurred the lowest annual soil respiration in northwestern China. Annual soil respiration from terrestrial ecosystems in China varied from 4.58 to 5.19 Pg C a between 1970 and 2009. During this time period, on average, annual soil respiration was estimated to be 4.83 Pg C a. Annual soil respiration in China accounted for 4.93%-6.01% of the global annual soil CO emission. The interannual variability in soil respiration depended on the interannual variability in precipitation and mean air temperature. In order to reduce the uncertainty in estimating annual soil respiration at regional scale, more in situ measurements of soil respiration and relevant factors (e.g. climate, soil and vegetation) should be made simultaneously and historical soil property data sets should also be established. [ABSTRACT FROM AUTHOR]