Spatial And Temporal Changes Of Drought In Hubei Province Based On GIS And RS

In recent years,with global environmental problems such as climate warming becoming increasingly prominent,extreme disasters such as drought are more frequent,and major drought events will lead to food shortages,drinking water difficulties and other disasters,which have a serious impact on national and social stability.In this paper,the Normalized Vegetation Index(NDVI)and the Enhanced Vegetation Index(EVI)were extracted from MOD13A3 and the Land Surface Temperature(LST)was extracted from MOD11A2,which were used to construct the Temperature Vegetation Drought Index(TVDI)based on the two vegetation indices respectively.By comparing the fitting effects of NDVI-LST and EVI-LST on the dry and wet edges of the feature space,selecting TVDI with better fitting effect,conducting correlation analysis between temperature,precipitation,evaporation and standardized precipitation index(SPI)of meteorological station data and TVDI,and exploring the distribution characteristics of TVDI of different land use types(farmland,woodland,grassland and construction land)to analyze the applicability of TVDI in drought monitoring in Hubei province.Finally,the temporal and spatial variations of Hubei Province over the years are studied by using a variety of time series analysis methods.The following results and conclusions have been achieved in this paper:(1)TVDI index based on NDVI is easy to be saturated in high vegetation coverage area,and its spatial fitting is inaccurate,and NDVI-LST has no better effect than EVI-LST in fitting triangular relationship of feature space.Therefore,TVDI index based on EVI-LST is more applicable in Hubei Province.This paper establishes feature space and retrieves the TVDI index of Hubei Province from 2008 to 2017 through EVI-LST.(2)The applicability of the TVDI index was analyzed by using temperature,precipitation,evaporation,SPI of different time scales and different land use types.The results showed that the correlation between TVDI and temperature was very significant,and the correlation between TVDI and precipitation and evaporation was low,but the correlation between TVDI and precipitation and evaporation was better in autumn.In the correlation analysis of SPI at all scales,TVDI is consistent with the drought information obtained by SPI in January and March.According to the research of the TVDI distribution characteristics of different land use types,in the year of drought,farmland and construction land suffered the most serious damage,followed by grassland.However,the drought damage of woodland is relatively less,for it has a good ability of water conservation,which is beneficial to soil water conservation and inhibiting the development of drought.(3)Various time series analysis methods were used.Firstly,the SPI index from 1961 to 2017 was used to analyze the change of drought period and amplitude.It was found that there were three main periods of drought in Hubei Province in recent 57 years: 5,10 and 22,and the dry-wet transition periods were 34,17 and 8 times,respectively.By using TVDI index to analyze the spatial and temporal changes of drought in different months from 2008 to 2017 in Hubei Province,it is found that the drought fluctuation in southeastern Hubei is relatively small and that in northwestern Hubei is relatively large,but the overall trend of change in Shen Long Jia forest area is not obvious,while there is an increasing trend of drought in January,March,April,July and September in Hubei Province,and there is a decreasing trend in May,June,August and November.Under the condition of significant a=0.05,drought in January,April,July and September had an increasing trend,while drought in May,June,August and November had a decreasing trend.