Spatial-temporal Variation Of Grassland Vegetation Dynamics And Its Response To Climate Change In China's Temperate Zone

Grassland is one of the most important terrestrial ecosystems in the world,and it is bearing great risks and challenges due to climate change.As an important grassland resource in China,the grassland in China's temperate zone is located in both climate sensitive and ecological fragile districts,making the study of spatial-temporal variation of grassland vegetation dynamics and its response to climate change a key factor to the adaptation to climate change,the rational utilization of grassland resources,and the promotion of sustainable development of livestock husbandry.Based on the data of GIMMS NDVI 3g remote sensing,meteorological raster,and actual measurement,both the extraction of phenological parameters and the CASA light use efficiency model were used in this study,to obtain NDVI,phenological phase,and net primary productivity,reflecting the characteristics of vegetation dynamics.With the help of the Standardized Precipitation Evapotranspiration Index(SPEI),the spatial-temporal variation of climate resources and drought was analyzed in thesis,also,the response of grassland vegetation to drought events at different time scales and influencing factors was revealed.The characteristics of spatial-temporal changes of grassland phenophase and NPP,together with their responses to climate change at different time scales were also explored.The discussion on the relationship between productivity and phenological changes lays a foundation of a theoretical support for the possible changes and response rules of grassland vegetation dynamics in the future climate scenarios.The main results are listed below:(1)The temperature of grassland increased dramatically in China's temperate zone from 1982 to 2015,among which the minimum temperature increased most significantly.Annual precipitation and solar radiation fluctuated greatly,showing no significant decreasing trend.In terms of the seasonal variations,the spring was getting warmer and wetter,while the summer was getting warmer and drier,the radiation decreased in autumn and no significant trends in winter.The region was getting drier during the past 34 years,but not obviously,and generally becoming wet in spring,autumn and winter.The drought spatial scale within this region were larger during spring and autumn,which also showed an increasing trend during winter.The frequency of light drought was the highest in the whole region.(2)The average annual NDVI in the study area showed a significant increased trend,whose largest increase occurred in summer.Spatially,the growth rate of the lower NDVI area of grassland was significant,while the growth rate of the higher NDVI area showed an obvious decline.There was a significant positive correlation between the NDVI and the SPEI.The short-term drought events were likely to affect grassland growth in summer and autumn,but received a slow response in spring.Precipitation had a significant effect on the correlation between the NDVI and the SPEI,the less precipitation,the stronger responded from the grassland to drought,meanwhile,water and radiation conditions could affect the response speed.(3)According to the results from the remote sensing extraction of phenological phase,the start ofseason(SOS)of the 17%grassland vegetation was significantly advanced.The areas with obvious delayed in the end of season(EOS)and significant extension in the length of season(LOS)were 14%and 17%respectively.The LOS of 61%area was determined by the SOS.The SOS was negatively correlated with short-term temperature and long-term cumulative precipitation,meanwhile,the EOS was closely related to short-term temperature and cumulative precipitation.Temperature was the main factor controlling the change of SOS in different pre-season time periods.The SOS was negatively correlated with the temperature in the current month in most areas.Pre-seasonal cumulative precipitation mainly affected the change of EOS,and the region that had a positive correlation between precipitation and EOS in the previous month was largest.(4)The grassland NPP showed a significant increase in 26%of the region in the results from the CASA model,but the interannual growth was not significant.The correlation between annual/seasonal mean temperature and NPP was not obvious.Water profit and loss was the main factor affecting NPP in different types of grassland vegetation,especially in summer.The SPEI showed a stronger correlation with NPP than precipitation,indicating that the SPEI considering of hydrothermal conditions can reflect the spatial-temporal pattern of grassland vegetation productivity in a better way.(5)The analyzed results of the relationship between grassland productivity and phenological phase showed that there was no significant correlation between the LOS and annual NPP.The regions with prolonged LOS and increased NPP were mainly affected by temperature.Both of the shortening of LOS and the increasing of NPP occurred in areas where hydro-thermal coupling affected.Due to the decrease of precipitation,the prolonged or shortened LOS in regions where precipitation was significantly controlled was not conducive to grassland productivity accumulation.Spatially,the grassland productivity increased first and then decreased with its LOS,meanwhile,the higher the temperature was,the later the turning point was.The NPP was more easily affected by water stress in the area where the LOS became longer with less precipitation.The NPP in the region with both the least and the most precipitation had less fluctuation in response to the LOS.Most of the grassland vegetation SOS advanced was conducive to the increase of spring NPP,the delay of EOS promoted the accumulation of autumn NPP,and the performances of different grassland types were similar to those of the whole region.