Analysis Of Drought Characteristics In Yellow River Basin Based On Multi-Source Remote Sensing Data

The annual precipitation in the Yellow River Basin is small,and most areas are prone to severe drought.Observation data from surface meteorological stations are important basic data for hydrological analysis.However,in the face of the large area of the Yellow River Basin,the complex terrain,the uneven distribution of meteorological stations,and even the lack of meteorological stations in individual areas,it is easy to cause large errors in the analysis of precipitation..This paper analyzes the accuracy of satellite precipitation data in two dimensions,time and space,and verifies it with measured data from meteorological stations.Based on the analysis of the main factors affecting remote sensing data,the significant factors were screened by stepwise regression method,and the optimal relationship model was established based on this.TRMM3B43 precipitation data was down-scaled from 0.25 ° × 0.25 ° to 1km resolution..Then,it is verified that the TRMM satellite data based on satellite radar rain measurement can effectively make up for the lack of ground observation data,and the precipitation index index algorithm(TRMM-Z index)is used to construct the remote sensing precipitation drought index.At the same time,multi-source remote sensing data products(vegetation Index MODIS13A3,surface temperature MOD11C3,etc.)verify the effectiveness of the TRMM-Z index in monitoring drought.Finally,based on the SPEI index,temperature condition index TCI,and vegetation health index VHI,and combined with drought examples,a comprehensive multi-exponential drought analysis was performed.The main conclusions are as follows:(1)Applicability analysis of TRMM satellite precipitation data: Based on the measured precipitation data of 92 meteorological stations in the Yellow River Basin,the correlation analysis of remotely sensed precipitation data from annual,quarterly,and monthly scales was performed using correlation coefficients and relative Error as an indicator.Studies have shown that TRMM data has relatively high accuracy,the correlation coefficients of 92 meteorological observation stations are above 0.9,and satellite precipitation and measured precipitation have good consistency in time and space.(2)Statistical downscaling research of TRMM satellite precipitation data: Analyze the impact of rainfall quality on satellite precipitation data,explore the influence factors of satellite monitoring from various aspects such as latitude,longitude,elevation,and NDVI,and find the TRMM data by stepwise regression by month The optimal linear relationship between the significant influencing factors was obtained,and the optimal linear regression model was obtained.All the regression equations passed the significance test.The kriging interpolation method was used to remove the residual error and downscale it.(3)Use the Z index algorithm of precipitation to construct a remote monitoring index of drought monitoring index TRMM-Z index.Use the rainfall data from the weather station to calculate the SPEI value and compare and analyze it.Through the analysis of the two,you can know that TRMM-Z is monitoring drought events Time,have a certain characterization ability.However,in many sites,the results of the TRMM-Z index are not completely consistent with the SPEI,and there are some cases of weakening the drought level.(4)Prediction analysis of agricultural drought: further analysis of agricultural drought,inspection of multiple drought indicators,verification through drought examples,temperature state index TCI and vegetation health index VHI,as third-party indicators for measuring drought,suitable for drought monitoring and early warning.The research results can provide reference and data support for the Yellow River Basin to deal with drought and disaster prevention and early warning.