Main content area

Modeling interannual variability of global soil respiration from climate and soil properties

Chen, Shutao, Huang, Yao, Zou, Jianwen, Shen, Qirong, Hu, Zhenghua, Qin, Yanmei, Chen, Haishan, Pan, Genxing
Agricultural and forest meteorology 2010 v.150 no.4 pp. 590-605
agricultural land, grasslands, forests, tundra, ecosystems, soil respiration, air temperature, precipitation, soil organic carbon, nitrogen, soil fertility, pH, simulation models, prediction
To develop a model describing the dependence of annual soil respiration on climate and soil properties, we compiled 657 published annual soil respiration (R s ) measurements that were assembled from 147 sites globally, representing croplands, grasslands, forests and tundra ecosystems. Each of these annual soil respiration data was then aggregated with the appropriate mean air temperature (T) and annual precipitation (P) data derived from geographically referenced datasets and with soil properties gathered from the original literature. Partial correlation analyses showed that global annual R s significantly related to annual mean temperature, annual precipitation, and topsoil (0-20cm) organic carbon (SOC) storage, while topsoil total nitrogen (SN) and pH did not show a direct and clear relationship with R s across ecosystems. While we employed the T&P-model that used temperature and annual precipitation to globally predict annual soil respiration, it was able to explain 41%, 57%, and 31% of the variability of soil respiration for croplands, grasslands and forests, respectively. However, the residuals were significantly related to SOC for croplands and grasslands. Thus, we developed a T&P&C-model that includes SOC as an additional predictor of annual R s . This extended but still simple model performed better than the T&P-model and explained 69%, 89%, and 47% of the interannual and intersite variability of R s with a mean absolute error of 0.11, 0.18 and 0.28kgCm⁻² yr⁻¹ for croplands, grasslands and forests, respectively. Overall, the modeling efficiency of the T&P&C-model was nearly 60% across ecosystems. Globally, the mean turnover time of topsoil carbon (SOC/R s ) was highly comparable among croplands, grasslands and forests, equivalent to 6.1-6.3 years. Therefore, better estimates of global annual soil respiration would be obtained with the new model driven by climate and soil properties together. We expect significant improvements of global annual soil respiration predictions given that measurements of soil respiration coupling with soil properties and site productivities are widely taken across ecosystems over the world.