| Grassland is an important terrestrial ecosystem,which not only creates the material basis for agricultural and forestry production and human life,but also improves the ecological space through water circulation and energy balance.Sanjiangyuan is located in the hinterland of Qinghai-Tibet Plateau,and its fragile and sensitive alpine grassland is affected by climate change and human activities,which shows a trend of grassland degradation.Since the implementation of the ecological restoration project in 2000,some parts of the grassland have also shown a trend of restoration.Clarifying the spatial and temporal distribution characteristics of climate change and grassland change in Sanjiangyuan can effectively reveal the regional hydrothermal effect of grassland degradation and restoration in the alpine region,so as to provide a scientific basis for the evaluation of ecological restoration projects.Therefore,this paper took Sanjiangyuan region as an example,divided the surface dryness and wetness based on the moisture index,used the Land use/Land Cover data of 2000,2010 and 2018 to determine the degradation/restoration of grassland,evaluated landscape patterns with Fragstats 4.2software,and predicted future grassland greenness based on the Hurst index.Finally,the study investigated the local surface temperature changes of grassland degradation and restoration mechanisms based on biogeophysical indicators such as albedo,net radiation,and evapotranspiration.The main results of this study are as follows:…(1)From 2000 to 2018,the air temperature in the Sanjiangyuan region showed a significant increasing trend with a warming propensity rate of 0.92℃/10a with a spatial distribution of high in the east and low in the west,and the warming trendency slowed down after 2010.The precipitation showed an insignificant increasing trend with an increasing propensity rate of 20.42mm/10a,and a decreasing distribution from southeast to northwest.The Potential Evapotranspiration showed an insignificant decreasing trend with a variation propensity rate of-2.44mm/10a,and the spatial distribution was higher in the east than in the west.The moisture index showed an insignificant increasing trend with a propensity rate of 3.58(-)/10a.There were dry periods(2000-2004,2013-2018)and wet periods(2005-2012).In terms of spatiality,the distribution was decreasing southeast-northwest,and the dry and wet dividing line was more consistent with the400mm precipitation line.The fitting analysis found that the precipitation explained 81%of the moisture index,showing a significant positive correlation.(2)In the study area,grassland covered the largest area in Land use/Land cover.From 2000 to 2018,the land transfer direction was mainly from unused land to low cover grassland and medium cover grassland,with the transfer area of 27653 km~2 and 12141km~2 respectively.The overall landscape pattern of Sanjiangyuan tended to be stable in the past.The stability of the protected area was higher than that of the non-protected area.However,the patch aggregation index in the Yangtze River Source Park was only 11.28,much lower than that of the Lancang River and Yellow River Source Parks.Among the patch types,the fragmentation of grassland and unused land was relatively high and heavily influenced by human activities and ecological changes.However,the number and density of patches with low coverage of grassland and unused land decreased in dynamic trends,and the fragmentation degree of patches showed a reducing trend.The Normalized Difference Vegetation index(NDVI)showed a decreasing distribution from southeast to northwest,and the mean value of NDVI was as follows:Huangnan Tibetan Autonomous Prefecture>Guoluo Tibetan Autonomous Prefecture>Hainan Tibetan Autonomous Prefecture>Yushu Tibetan Autonomous Prefecture>Tanggula Mountain Town.The NDVI showed a significant increasing trend in almost 20 years,with a change tendency of 0.01(-)/10a.(3)The biogeophysical process parameters of the Sanjiangyuan fluctuated on an interannual basis,and there were regional differences in space.The surface albedo was high in the east and low in the west,with a mean value of 0.22,showing a significant decreasing trend.The net radiation was high in the east and low in the west,with an average of 2180.31 MJ/m2,showing an insignificant decreasing trend,and the low value appeared in 2005 and the high value appeared in 2013.The actual evapotranspiration showed a southeast to northwest decreasing distribution,similar to the potential evapotranspiration,with an average annual value of 348.33mm,only accounting for half of the potential evapotranspiration,with an insignificant decreasing trend and a low value in 2016.The local surface temperature was high in the east and low in the west,with an average annual value of 13.94℃,showing an insignificant decreasing trend,contrary to the trend of a significant increase in air temperature.Moreover,there was a relatively obvious low temperature in 2005 and 2012,which was consistent with the change of high humidity index in the same period.(4)During the change of alpine grassland cover,the biogeophysical effects of constant/degraded/restored grassland were consistent,but the temperature effects were slightly different.Among them,NDVI showed a positive effect,and the increase range was restored grassland>degraded grassland>unchanged grassland.Albedo showed a negative effect,and the decreasing range was restored grassland>degraded grassland>unchanged grassland.The net radiation showed a negative effect,and the reduction range was degraded grassland>unchanged grassland>restored grassland.The actual evapotranspiration showed a positive effect,and the increase range was restored grassland>degraded grassland>unchanged grassland.Both restored grassland and degraded grassland showed cooling effects.The cooling effect of restored grassland on surface temperature was 0.2℃higher than degraded grassland,while the cooling effect of constant grassland was not obvious.(5)Since the implementation of ecological restoration project,the growth of grassland has improved.The whole grassland showed cooling effect and mostly distributed in semi-arid area.The random forest model can explain 67.70%of land surface temperature in the whole region,and the important variables were precipitation,net radiation,moisture index and potential evapotranspiration.The precipitation and moisture index of each region showed a significant negative correlation with the change of land surface temperature,and the dry and wet of the region offset part of the annual warming effect,so there was no obvious correlation between the two and land surface temperature.In the future,NDVI of 55.40%grassland would decrease,mainly distributed in Yushu,Guoluo and the central and southern part of Huangnan Tibetan Autonomous Prefecture.About 41.84%of grassland NDVI may be on the rise,mainly distributed in Yushu and the eastern part of Hainan Tibetan Autonomous Prefecture.The NDVI of 0.81%grassland may not be sustainable,and the sustainability of the rest grassland was not obvious.Based on the grassland cover level of Land use/Land cover,this paper determined the invariance/degradation/restoration of grassland,quantified the dynamic changes of surface water and heat in the alpine grassland of Sanjiangyuan driven by climate and human activities,and preliminarily revealed the geobiophysical effects of alpineIt was found that since the grassland restoration project in 2000,the stability of landscape pattern has increased,the surface moisture index has increased overall,the local surface temperature has decreased as a whole,which affirmed the contribution of ecological restoration project to alleviating global warming.In the future,it is necessary to further promote the ecological complex protection and special treatment of"mountains,rivers,forests,fields,lakes,grass,sand and ice".In addition,there is a certain regional heterogeneity in the thermal response of grassland surface water,which requires further exploration of land surface energy redistribution and the process of how radiation is converted into latent and sensible heat in future studies. |
PDF Full Text Request