Research On The Spatiotemporal Distribution Characteristics Of Potential Natural Vegetation In China In The Future Based On CSCS Model And Climate Big Dat

Understanding the response of potential vegetation to future climate changes would substantially contribute to the scientific assessment of vegetation–climate interactions.The Comprehensive Sequential Classification System(CSCS)based on the relationships of climate,soil,and vegetation enhance our knowledge of the relationship between climate and potential natural vegetation.However,these benefits have been largely constrained by acquisition of meteorological data resources and the station interpolation technology required.Therefore,the previous efforts on the Potential Natural Vegetation(PNV)in China based on the CSCS model were limited to a local area,a specific period or a single future climate scenario.The CSCS model combined with climate data covering three future scenarios and long time series spanning 1970-2080s,the value of research on PNV in the whole territory of China has not been implemented.In addition,detailed comparisons between the CSCS model,the Holdridge Life Zone(HLZ)model,and a map of actual vegetation at a national scale have only been carried minimally.Here,we used the 1970-2000 and RCP2.6,RCP4.5,RCP8.5 climate scenarios dataset in the 2030s,2050s,2070s,and 2080s to analyze the spatio-temporal distribution patterns,change trend,shift direction and distance,and succession tendency of PNV in China based on the CSCS model.On this basis,combined with the change times of PNV types in China,five sensitivity levels were proposed.The main innovations and results are as follows:(1)The CSCS model was compared with the HLZ model,which has become a critical tool for simulating biome at multiple spatial scales,and the 1:1,000,000 vegetation distribution map in China,which validates the accuracy of CSCS in simulating potential vegetation in China.The results show that on a national global scale there are good agreements among three maps produced by the CSCS model,by the globally well-accepted HLZ model,and 1:1,000,000 vegetation distribution map in China.Comparing with HLZ model,the CSCS model can not only successfully predict the spatial distribution of grassland,forest and desert at the national scale,but also perform well in grassland.(2)Based on the CSCS model,the spatial distribution characteristics of PNV in China at the national scale and nine agricultural regions under the baseline scenario(1970-2000)were discussed respectively.Overall,the potential vegetation types in China are expected to increase from 36 in 1970-2000 to 39(2030s,2050s for RCP2.6 and 2030s for RCP4.5)or 40(others).Under the baseline scenario,China has not yet developed the climatic and environmental condition for the growth and development of six potential vegetation,including warm-arid warm subtropical semidesert(ⅤB).From the perspective of the nine major agricultural regions,the region with the most abundant PNV types(25types)would appear in the Qinghai-Tibet Plateau.Among them,the cold,frigid-perhumid rain tundra,alpine meadow(ⅠF)are the most widely distributed PNV types,accounting for 59.86%of the Qinghai-Tibet Platea area.(3)The dynamics of potential vegetation in China in the next 80 years under three scenarios were predicted,including area change trend,shift distance and direction of super-classes based on the centroidal shift model,and succession.ⅠF are the most distributed vegetation types,with an area of more than 78.45×10~4 km~2,whereas there are no climate conditions suitable for tropical–extra-arid tropical desert in China.Some plants would benefit from climate changes to a certain extent.And by the 2080s,the area of tropical-semiarid savanna(ⅦC)and tropical–subhumid tropical xerophytic forest(ⅦD)would increase significantly,and the increasing rate is expected to reach an average of 183%and 134%,respectively,relative to the baseline climate scenario.Similarly,a continuous expansion of more than 18.81×10~4 km~2 and northward shift of more than 124.93 km in tropical forest would occur across all three scenarios;however,some potential vegetation would experience a severe and irreversible dieback caused by the consistently rising temperature and altered rainfall patterns in the future.More than1.33%of arid temperate zone desert(ⅢA)would be decreasing.The most severe succession pattern occurred in the steppe to semi-desert(28.93×10~4 km~2)under the RCP8.5 scenario among all types of succession from grasslands to deserts,accounting for about 97.31%of the loss rate of grassland.(4)Based on the changes of PNV types during 1970-2080s,five sensitivity levels of potential vegetation in China under three climate scenarios were proposed.Five sensitivity levels present an interphase distribution;The sensitive areas are relatively concentrated,while the non-sensitive areas are relatively scattered.In the RCP2.6scenario,the extremely sensitive area with the smallest distribution area would be15.81×10~4km~2,accounting for 1.65%of the total territory,and only scattered in northeast Plain,North China Plain and Inner Mongolia Plateau.