Spatio-temporal Distribution Characteristics Of The CSCS-based Potential Natural Vegetation In China

Based on the comprehensive and sequential classification system (CSCS), the simulated potential natural vegetation (PNV) can be used to reflect the response of terrestrial ecosystems to the global climate change, which plays an important role in the succession and spatio-temporal distribution dynamics of grassland vegetation types. In this study, combined with meteorological data and the CSCS model, we simulated the current distribution of PNV in China. On this basis, the spatio-temporal distribution of the existing natural vegetation in China was analyzed using remote sensing data. Furthermore, the dynamic change of the normalized difference vegetation index (NDVI) of various existing natural vegetation types and its response to climate change were studied. In order to study the impact of human activities to natural vegetation, the concept of "human occupancy" was employed, and the predicted climate data was used to analyze the distribution of potential vegetation in China under different climate scenarios. The results are as follows:1) The types of PNV in China include tundra and alpine steppe, frigid desert, semi-desert, steppe, temperate humid grassland, temperate forest, sub-tropical forest, tropical forest, warm desert and savanna. In overall, from2001to2010the area of the existing natural vegetation had an increasing trend.2) From2001to2010, the peak value of monthly average NDVI of the existing natural vegetation appeared between July and September. The annual average NDVI of existing natural vegetation in extreme natural environment (i.e., severe cold, torridity or drought) was tended to decrease. While the annual average NDVI of the existing natural vegetation in the subtropics and the tropics was tended to increase. The annual average NDVI of the rest vegetation types distributed in the middle of the above zones changed gently.3) The NDVI of tundra and alpine steppe, semi-desert, steppe, temperate humid grassland and temperate forest had significant positive correlations with temperature and precipitation; while the NDVI of savanna had strong negative correlation relationships with temperature and precipitation. There was a significant positive correlation between temperate and the NDVI of frigid desert and tropical forest, and a significant negative correlation between precipitation and the NDVI. The NDVI of sub-tropical forest and warm desert had a strong negative correlation with temperature, but it had a significant positive correlation with precipitation.4) The area of natural vegetation occupied by human beings in China increased gradually from2001to2004, and then decreased year by year from2004to2010. The impacts of human activities to PNV as following:The intensity of human occupation to frigid desert had a slight increasing trend, while the intensities of human occupation to the steppe, temperate humid grassland, temperate forest, sub-tropical forest and savanna tended to decrease slowly. The intensities of human occupation to semi-desert, tropical forest and warm desert were not obvious, and the tundra and alpine steppe was influenced slightly.5) The distribution of PNV varied under four different future climate scenarios. The tundra and alpine steppe decreased under all of the scenarios. The frigid desert and the sub-tropical forest increased firstly and then decreased under the scenario of RCP2.6while they tended to decrease under the rest three scenarios. The semi-desert and savanna increased firstly and then decreased under the scenario of RCP2.6while they tended to increase under the rest three scenarios. The steppe tended to increase under the scenario of RCP4.5while it increased firstly and then decreased under the rest three scenarios. The temperate humid grassland and temperate forest changed variously under the four scenarios. The tropical forest and the warm desert showed an increasing tendency under all of the scenarios.6) In the future three periods (i.e., in2030s、2050s and2080s), under the four scenarios of RCP2、RCP4.5、RCP6.0and RCP8.5, the center of tundra and alpine steppe will tend to move to the west or northwest. As for the frigid desert, the center will move to the east under the scenarios of RCP4.5and RCP6.0, no clear movement direction under the scenario of RCP2.6, and from the east to the south under the scenario of RCP8.5. The center of semi-desert and the warm desert will move in varied directions under each scenario. The steppe will move to the southwest under the scenarios of RCP4.5, RCP6.0and RCP8.5, while under RCP2.6it will move to the southwest, northwest and northeast successively. Savanna will move from the northeast to the southwest. The temperate humid grassland will move to the northwest from2010s to2030s under the four scenarios. The sub-tropical forest will move from the north to the northeast under the scenarios of RCP2.6and RCP4.5, and it will move to the west and northeast under the scenarios of RCP6.0and RCP8.5, respectively. Under the scenario of RCP2.6, the tropical forest will move to the southwest. northwest and southeast, and it will move to the east under the rest three scenarios.