Impacts And Predictions Of Future Scenarios On The Growth Of Vegetation By Vapor Pressure Deficit In China

Vapor Pressure Deficit（VPD）describes the difference between saturated water vapor pressure and actual water vapor pressure in air at a certain temperature,and is a good indicator of atmospheric water demand,which plays a key role in determining plant physiological functions and is important for ecosystem structure and function as well as global water resources.We analyzed the spatial and temporal pattern changes of vegetation and VPD in China from 2000 to 2020,and studied the influence of VPD on vegetation and its contribution.We also constructed a vegetation change prediction model based on multiple linear regression method using environmental factors such as VPD,temperature and CO₂ concentration as factors,and predicted the change of vegetation in China from 2025 to 2100 by combining CMIP6 model data.It is important to explore the impact of VPD on vegetation in China and to carry out prediction studies for vegetation change for a deeper understanding of Chinese ecosystems.The main conclusions of this paper are as follows.（1）The VPDs estimated in this paper have good accuracy,where VPD-ERA has higher accuracy and is more similar in all aspects to the VPD products with high confidence,and the VPD-CRU results are low overall.Compared with the simulated global VPD products,the VPD-ERA estimated in this paper combining topography and enhancement factors in the study area is more applicable.（2）From 2000 to 2020,the annual mean VPD in China showed a fluctuating upward trend with a rate of 0.012h Pa/yr（P<0.01）.Its spatial distribution shows the characteristics of the highest in the northwest arid region;the second highest in the east;and the lowest in the Qinghai-Tibet Plateau and northeast region.The multi-year average VPD values ranged from 0 to 21.24 h Pa.The annual average VPD changed abruptly in 2005,and the average VPD values from 2006 to 2020 were 1.43 to 1.64 times of those from 2000 to2005,while the growth rates of NIRv,LAI and GPP decreased after 2005,indicating that the increase of VPD may have a suppressive effect on vegetation growth.In addition,the three vegetation indices combined with the spatial distribution of VPD indicate that excessive VPD does inhibit vegetation growth,but VPD and vegetation growth are not simply negatively correlated.（3）From the relationship between VPD and vegetation growth,when the VPD values were in the range of 4.06～4.82h Pa,the growth rates of the three vegetation indices reached their peaks,which were 2.31E-4/yr,2.22E-4/yr,and 2.90 g C·m^-2·a^-1,respectively;when VPD was less than the value of this interval,VPD showed a significant positive correlation with vegetation growth rate;when VPD When VPD was larger than the interval value but lower than 9h Pa,VPD and vegetation growth rate showed a significant negative correlation;when VPD exceeded 9h Pa,the correlation between vegetation growth rate and VPD was significantly weakened.Vegetation growth in eastern China was mostly negatively correlated with VPD.The reason for the change of vegetation growth rate is not only related to the magnitude of VPD growth,but also has a close relationship with the base value of VPD and vegetation characteristics in the region.（4）CO₂ showed the most significant positive contribution to vegetation,with mean values of 0.32,0.30,and 0.29 for the three types of vegetation data;VPD showed a significant negative contribution,with mean values of-0.10,-0.084,and-0.089 for the three vegetation data,respectively.without considering other factors,the decrease in vegetation due to the increase in VPD partially offset about 30%of the fertilization effect of CO₂.（5）The vegetation change prediction model constructed based on multiple regression equations in this study can predict future vegetation dynamics more accurately.The future vegetation change scenarios are very similar to the CO₂ factor,which is related to the significant positive contribution of CO₂ to vegetation.the mean values of the range rate of NIRv are 60.91%and 115.40%for the SSP2-4.5 and SSP5-8.5 scenarios,respectively;the mean values of the range rate of LAI are 74.61%and 148.16%,respectively;the mean values of the range rate of GPP are 90.13%and 172.73%for GPP.The regions with significant vegetation growth under both scenarios are mainly located in the central northeast,the Loess Plateau,the western mountainous region and the southeastern coastal region.The regions with a decrease in total vegetation are mainly located in the southern region of the Tibetan Plateau and the desert region in the northwest.