| Under the background of global warming,the problem of drought is becoming more and more serious,which has a serious negative impact on industrial and agricultural production,people ’s life and economic development.Shandong Province is located in the eastern coastal areas of China and is a major grain production province in China.However,due to the complex terrain and changeable climate,drought events occur frequently.Therefore,this paper obtains normalized difference vegetation index(NDVI)and land surface temperature(LST)based on daily Moderate-resolution Imaging Spectroradiometer(MODIS)products from 2011 to 2020.Firstly,the slope and intercept changes of dry and wet edges in NDVI-LST feature space constructed by daily product data and 8-day synthetic products are compared.The daily feature space with more reasonable slope and intercept changes of dry and wet edges is selected to invert the Temperature Vegetation Dryness Index(TVDI),and Global Land Data Assimilation System(GLDAS)soil moisture data is used for verification.Then,the distribution of TVDI in different terrains and ground objects at different time scales is analyzed by combining DEM data and land use data.Finally,the gravity center migration model,Theil-Sen trend degree,and Mann-Kendall test were used to analyze the spatial and temporal variation characteristics of drought in Shandong Province in the past 10 years,and the Hurst index method was used to predict the future drought intensity.The main research results are as follows:(1)The wet and dry edges in the NDVI-LST feature space constructed by daily data can form a triangular model that conforms to the concept of TVDI,which is suitable for the calculation of TVDI in climate-changing areas.There is no obvious change in the slope of the wet and dry edges.The dry edge intercept increases from 281 K on the 16 th day to about 326 K on the 164 th day,then decreases to about 283 K on the 345 th day,and finally fluctuates between 279 K and 283 K from the 350 th day to the 359 th day.The wet edge intercept and the dry edge intercept have the same change trend and are about 10 K lower.The proportion of TVDI synthesized according to different time scales and 10 cm soil moisture passing the extremely significant test was 86%,showing a significant negative correlation.TVDI is sensitive to precipitation,but there is also a certain lag.(2)In Shandong Province,there was a significant negative correlation between TVDI and altitude,topographic relief and slope,that is,with the increase of topographic factor level,the degree of drought will gradually weaken,but there is no obvious correlation with slope aspect.From different time scales,the degree of drought in spring had a strong correlation with topographic relief,and the degree of drought in other seasons had a strong correlation with altitude.Among different types of ground objects,the drought degree of cultivated land and construction land in low altitude areas was relatively large,accounting for 61.12%and 62.23% respectively,while the drought area of forest land was relatively small,only 12.88%.(3)From 2011 to 2020,the drought degree of Shandong Province generally showed a downward trend.The drought distribution mainly showed that the southwest and southern plains were relatively dry,and the central mountainous areas and southeast hilly areas were in normal or humid conditions.At the seasonal scale,the drought degree from large to small was spring > winter >autumn > summer;among the prefecture-level cities,Heze had the fastest increase rate of TVDI,which was 1.34%/a,and Qingdao had the fastest decline rate,which was-1.44%/a.(4)The center of drought in Shandong Province is mainly distributed in the mountainous areas of Zibo City,Jinan City and Tai’an City,showing a northeast-southwest distribution pattern.Except that the center of drought in spring migrates to the northeast,the center of drought in other time scales moves to the southwest.The Sen trend value of TVDI showed an increasing trend in46.54 % of the area,mainly distributed in the southwest region,and a decreasing trend in 53.46%,distributed in the central mountainous and southeast hilly areas.The summer,autumn,winter and inter-annual droughts were mainly non-significant wetting,accounting for more than 60 % of the area,while the sum of non-significant drying and significant drying was less than 28%,mainly concentrated in the southwest and southern regions.The future drought changes showed that the area of inter-annual,summer,autumn and winter future drought weakening trend accounts for more than 41.40%,while the spring drought area is mainly increasing,mainly in the southwest region. |