Understanding variations and seasonal characteristics of net primary production under two types of climate change scenarios in China using the LPJ model

The approach of conditional nonlinear optimal perturbation related to parameter (CNOP-P) is employed to provide a possible climate scenario and to study the impact of climate change on the simulated net primary production (NPP) in China within a state-of-the-art Lund-Potsdam-Jena dynamic global vegetation model (LPJ DGVM). The CNOP-P, as a type of climate perturbation to bring variation in climatology and climate variability of the reference climate condition, causes the maximal impact on the simulated NPP in China. A linear climate perturbation that induces variation in climatology, as another possible climate scenario, is also applied to explore the role of variation in climate variability in the simulated NPP. It is shown that NPP decreases in northern China and increases in northeastern and southern China when the temperature changes as a result of a CNOP-P-type temperature change scenario. A similar magnitude of change in the spatial pattern variations of NPP is caused by the CNOP-P-type and the linear temperature change scenarios in northern and northeastern China, but not in southern China. The impact of the CNOP-P-type temperature change scenario on magnitude of change of NPP is more intense than that of the linear temperature change scenario. The numerical results also show that in southern China, the change in NPP caused by the CNOP-P-type temperature change scenario compared with the reference simulated NPP is sensitive. However, this sensitivity is not observed under the linear temperature change scenario. The seasonal simulations indicate that the differences between the variations in NPP due to the two types of temperature change scenarios principally stem from the variations in summer and autumn in southern China under the LPJ model. These numerical results imply that NPP is sensitive to the variation in temperature variability. The results influenced by the CNOP-P-type precipitation change scenario are similar to those under the linear precipitation change scenario, which cause the increasing NPP in arid and semi-arid regions of the northern China. The above findings indicate that the CNOP-P approach is a useful tool for exploring the nonlinear response of NPP to climate variability.