The Response of Net Primary Production to Climate Change: A Case Study in the 400 mm Annual Precipitation Fluctuation Zone in China.

The regions in China that intersect the 400 mm annual precipitation line are especially ecologically sensitive and extremely vulnerable to anthropogenic activities. However, in the context of climate change, the response of vegetation Net Primary Production (NPP) in this region has not been scientifically studied in depth. NPP suffers from the comprehensive effect of multiple climatic factors, and how to eliminate the effect of interfering variables in the correlation analysis of NPP and target variables (temperature or precipitation) is the major challenge in the study of NPP influencing factors. The correlation coefficient between NPP and target variable was calculated by ignoring other variables that also had a large impact on NPP. This increased the uncertainty of research results. Therefore, in this study, the second-order partial correlation analysis method was used to analyze the correlation between NPP and target variables by controlling other variables. This can effectively decrease the uncertainty of analysis results. In this paper, the univariate linear regression, coefficient of variation, and Hurst index estimation were used to study the spatial and temporal variations in NPP and analyze whether the NPP seasonal and annual variability will persist into the future. The results show the following: (i) The spatial distribution of NPP correlated with precipitation and had a gradually decreasing trend from southeast to northwest. From 2000 to 2015, the NPP in the study area had a general upward trend, with a small variation in its range. (ii) Areas with negative partial correlation coefficients between NPP and precipitation are consistent with the areas with more abundant water resources. The partial correlation coefficient between the NPP and the Land Surface Temperature (LST) was positive for 52.64% of the total study area. Finally, the prediction of the persistence of NPP variation into the future showed significant differences on varying time scales. On an annual scale, NPP was predicted to persist for 46% of the study area. On a seasonal scale, NPP in autumn was predicted to account for 49.92%, followed by spring (25.67%), summer (13.40%), and winter (6.75%).