Spatio-Temporal Variation Of Alpine Vegetation Phenology And Its Effect On Vegetation Water Use Efficiency Over Qinghai-Tibet Plateau

As a critical ecological and environmental problem facing the world,climate warming has seriously affected the process of material exchange and energy flow in various ecosystems.Vegetation phenology is the primary indicator and sensor of climate and environmental change and plays a crucial role in regulating the ecosystem’s material exchange and energy flow.The Qinghai-Tibet Plateau(QTP)is sensitive to global warming;it is high altitude and extreme hydrothermal conditions create a fragile ecosystem.In the past few decades,under the background of climate warming,the ecological environment of the QTP has undergone remarkable changes,which have exerted significant influence on the sustainable development of the local and surrounding areas.Therefore,it is significant to study the temporal and spatial variation characteristics of alpine vegetation phenology over the QTP and its response to climate change and analyze its impact on water use efficiency for assessing the response degree of fragile ecosystems to climate change.In this study,we used three phenological information extraction methods to extract the Start of Growth(SOG)and End of Growth(EOG)of alpine vegetation on the QTP based on the MOD13Q1 Normalized Difference Vegetation Index(NDVI)dataset from 2001～2020.We selected the most suitable phenological dataset by evaluating its pixel loss rates and spatial stability.Then we used Sen trend and partial correlation analysis to explore the spatio-temporal variation characteristics of alpine vegetation phenology and its response to temperature and precipitation.Finally,we analyzed the spatio-temporal variation of Water Use Efficiency(WUE)on the QTP in spring,summer,autumn,and throughout the year based on the GLASS Gross Primary Productivity(GPP)and Evapotranspiration(ET)product dataset and assessed the response mechanism and extent of phenology and WUE in alpine vegetation.The main conclusions are as follows:(1)The S-G filtering and dynamic threshold methods are the best combinations to extract the phenology of alpine vegetation on the QTP.The pixel loss rates of SOG and EOG in the SG phenological data were the lowest and the most stable.The CV variation coefficients were primarily concentrated in [0,2],which was more accurate and stable than A-G and D-L.(2)The phenological information of alpine vegetation on the QTP in the recent 20 years is different in space.The SOG gradually delayed from southeast to northwest,and most of the mean values were concentrated in 120～140 days.EOG gradually delayed from northeast to southwest,and most of the mean values were concentrated in 260～280 days.The interannual variation rates of SOG and EOG in an alpine meadow,alpine steppe and alpine shrub were similar,with SOG advancing by 0～1 day/year and EOG delaying by 0～1 day/year.The change rates of SOG and EOG of different vegetation types were bounded by elevation.Below the boundary,the SOG change rate was delayed,and EOG advanced;above the boundary,the SOG change rate was advanced,and EOG delayed.(3)The spatial distribution of alpine vegetation phenology response characteristics to temperature and precipitation was significantly different.In different seasons,the increase in summer temperature had an advanced effect on the SOG,especially the alpine shrub and alpine meadow.Autumn temperature increase had a particular delay effect on EOG,mainly affecting the alpine steppe.The increase in precipitation in spring mainly advanced the SOG,while the increase in autumn delayed the EOG,mainly affecting the alpine meadow.(4)In the eastern and central QTP,WUE increased with the advance of SOG in spring.The advanced SOG of most alpine vegetation in the QTP may decrease WUE in summer and autumn.With the delay of EOG,WUE decreased in summer in most areas of the QTP and increased in autumn in an alpine meadow and alpine shrub.