A Coupling Model of Net Primary Productivity Pattern Simulation and Prediction

  Objectives  Net primary productivity (NPP) not only directly reflects the production of vegetation communities, but also is a major factor in determining ecosystem carbon sources/sinks and regulating ecological processes.

  Methods  We propose a new approach which couples the CASA(Carnegie-Ames-Stanford approach)model with the CA-Markov model to simulate and predict NPP at the pixel scale, thus reflecting the spatiotemporal distribution characteristics of NPP and predicting the possible directions of changes in NPP. Using the Weihe River basin as the study area, we estimated NPP in 2000, 2005, 2010 and 2015 by the CASA model based on NDVI(normalized difference vegetation index) MOD13Q1 data, meteorological data and vegetation distribution data, analyzed the spatio-temporal distribution characteristics of NPP under the changes pattern of climate fators and topographic factors. The CA-Markov model is coupled to simulate the changes of NPP in 2020 and predict the changes of NPP in 2025, and 2030.

  Results  The results show that the coupling of CASA and CA-Markov model has good applicability. Comparing the estimate and predicted values of NPP in 2015, the Kappa coefficient reaches 0.877 6 which indicates that the coupled model has high accuracy and good applicability for NPP prediction. And in the next 10 years, the NPP classes keep transforming to higher coverage areas which will mainly distribute above Zhangjiashan of Jinghe River and Xianyang to Tongguan of Weihe River.

  Conclusions  This study have important practical significance for understanding the spatial and temporal evolution characteristics and mechanisms of basin vegetation and promoting the ecological security in basin.