Analysis Of The Drought In Hunan Province Based On Temperature Vegetation Drought Index

As a frequent natural disaster in recent years,drought has seriously affected the balance of ecological environment and the sustainable development of human economy and society.It's an important issue that puzzles the society.As an major grain producing area in China,Hunan is located in the central hinterland of China.The climate is characterized by subtropical monsoon climate.The seasonal distribution of precipitation is uneven,and the interannual variation is obviously.However,because of the economic conditions and other factors,the meteorological stations are limited in distribution and the sites are unevenly distributed.The emergence of remote sensing technology provides favorable conditions for drought monitoring.It is widely accepted by scholars because of its all-weather,multidate and wide coverage.Make full use of the technical advantages of remote sensing and establish a suitable remote sensing monitoring model,which has a positive meaning for the reasonable analysis of the drought.Based on the relevant data of remote sensing drought analysis at home and abroad,the study takes Hunan Province as the study area.Using MODIS surface temperature product MOD11C3 and vegetation index product MOD13A2 to extract NDVI/EVI and LST to construct the Temperature Vegetation Drought Index(TVDI);By fitting NDVI-LST and EVI-LST feature space to dry edge and wet edge,the correlation coefficients were used to explore two kinds of exponential fitting effects;.Through the correlation analysis between NDVI-LST and EVI-LST index and measured rainfall and temperature in meteorological stations,further verify the applicability of EVI-LST drought index in Hunan province.The drought classification map of the study area is calculated by using the fitting coefficient of the EVI-LST characteristic space and the distribution of TVDI in different land use types(construction land,cultivated land and woodland).The drought conditions of the study area from 2005 to 2015 were analyzed in time and space.The analysis was carried out by using linear regression analysis from 2005 to 2015 by EVI-LST.The main results and conclusions are as follows:(1)Using IDL programming languages in ENVI software,output LST-NDVI and LST-EVI feature space scatterplot and dry>wet edge fitting parameters,establish the characteristics of the LST-NDVI and LST-EVI space.To realize the TVDI index inversion of Hunan province from 2005 to 2015.(2)Using precipitation observational data of Hunan province from 2005 to 2015,and using precipitation anomaly.Morlet wavelet analysis of 10 weather stations in every month to analysis season precipitation characteristics and cycle pattern.Mann-Kendall method was used to analysis precipitation mutation characteristics and trend of precipitation change.The results showed that the OF curve showed that the threshold was exceeded between 2006 and 2007,and the threshold of significance level was reached to 1.28,with the significance test of 0.1.Precipitation in hunan province has significant seasonal characteristics,and the cumulative distance is between-1 and 1,and there is an alternate period of rainfall between 3 and 15.(3)Based on the linear regression analysis method,the trend of the increase and decrease in Hunan province from 2005 to 2015 is "reducing-increasing-decreasing"from northwest to southeast.The mean change trend of TVDI is-0.0014,and TVDI showed a decrease of 51.39%.(4)The results showed that:the wetland?forest land distribution in the area of TVDI value is low and high demand for water.Building land is drought than other kind of serious,it's TVDI value is highest.Farmland and grasslands' distribution is secondary and lower demand for water,giving priority to semiarid.