Analysis of spatial and temporal patterns of net primary production and their climate controls in China from 1982 to 2010

Ecosystem net primary production (NPP) represents vegetation biomass increment after accounting for autotrophic respiration and is recognized as an important component of the terrestrial carbon cycle. In this study, the spatial and temporal patterns of NPP and their climate controls in China's ecosystems for the period of 1982-2010 were analyzed by using a remote sensing-based carbon model (i.e., the Carnegie-Ames-Stanford Approach, CASA) and multiple statistical methods. Validation against NPP observations from 335 forests sites showed good performance of CASA over the study region, with an overall coefficient of determination (R-2) of 0.73 and root mean square error (RMSE) of 132.9 g Cm-2 yr(-1). Spatially, we found that the spatial pattern of China's NPP showed gradients decreasing from the southeast toward northwest, which could be mainly explained by the spatial variability in annual precipitation. Temporally, China's NPP showed a significant increasing trend at both the national and biome levels during 1982-2010, with an annual increase of 0.011 Pg C or 0.42%. However, the increasing trends in NPP were not continuous throughout the 29-year period at the national scale. On the other hand, it showed three periods where the trends changed, which was likely being caused by a shift in climate conditions and extensive drought. Air temperature was found to be the dominant climatic factor that controlled the interannual variability in NPP throughout the country except for arid and semi-arid regions in the middle-north and northwest parts of China, where the interannual variations in NPP were mainly explained by changes in precipitation. Similar results were also obtained at the seasonal scale that changes in NPP were generally controlled by that in air temperature except for summertime, in which higher NPP were favored by higher summer precipitation, whereas summer temperature was negatively correlated with NPP. At the monthly scale, NPP responded to change in temperature more rapidly than that in precipitation. However, temperature appeared to control NPP only in humid and semi-humid regions. For monthly NPP-precipitation relationship, the strongest positive relations were observed when NPP lagged behind precipitation by 1-3 months. However, deviating from the common hypothesis that plant in drier areas should respond to water availability more rapidly than in other regions, our analysis revealed a relative large time lag between monthly NPP and precipitation in arid ecosystems (i.e., 3 months). This result suggests that there may be a more complex mechanism of local water redistribution that controls vegetation water use (e.g., use of water from previous year precipitation) in extremely arid ecosystems. (C) 2015 Elsevier B.V. All rights reserved.