Validation Of Drought Mitigation Product With Remote Sensing Method

China is a large agricultural country that affected by the monsoon and climate, with the characteristics of drought-prone which caused significant losses to crops. An urgent problem in the drought field that is to find an efficient and timely way for monitoring the drought.Remote sensing have the features of high temporal resolution, high spectral resolution and easily obtained and so on. Usually we don’t need high spatial resolution remote sensing images for monitoring because drought is a large area case.Therefore, the use of remote sensing methods formonitoring drought, will greatly enhance the level of fight a drought, service at reduceagricultural disaster.Drought product needs to be validated before the application.RS Validation is the use of direct or indirect test, to testing and evaluation the radiationaccuracy, geometric accuracy and spectral accuracy of proposed test sensing data, to testedand evaluated the precision of the application and inversion of remote sensing products.Thefocus of this paper is use of remote sensing drought indices, and appropriate improvements,the inversion of drought products, while the use of meteorological drought index for the sameperiod to monitor the same region, to establish the validation between the same dimensionand different dimensions, and to analyze the advantages, disadvantages and applicability ofthe three indices.In this paper, the North China Plain to the study area, respectively, based on remotesensing index-Vegetation Health Index (VHI) and Modify NDMI and meteorological index-Standardized Precipitation Index (SPI) to monitor changes in the drought years of2000-2012in the region and analyze the effects of elevation and vegetation cover types ofdrought.Finally, complete the process of validation and the applicability of the three indices.The content and conclusions of the study are as follows:（1） Study the monitoring principles of remote drought index VHI, and based on MODISremote sensing data inversion VHI from2001to2012, Use VHI evaluation of the North China Plain drought conditions, and access to the calendar year drought verifies droughtmonitoring VHI index.（2） Study the monitoring principles of remote drought index NDMI, and improved NDMI,named modify NDMI, Improved index takes into account the impact on them over the years,and based on MODIS remote sensing data inversion MNDMI, finally evaluation of the NorthChina Plain drought conditions from2001to2012.（3）Based on DEM remote sensing analyzed elevation distribution of the North China Plain,and classified according0-25,25-50,50-100,100-500elevation, using IDL to RGB display theimage, the final analysis of the effects of different elevations on the drought situation.（4） Based on MOD12Q1data,Extraction of the11-year vegetation coverage of the NorthChina Plain, The11-year vegetation types of the area remains unchanged assign the originalvegetation type, the vegetation types more than a year to change assign zero. Finally, analysisof the impact of the drought situation in different vegetation.（5）Study the monitoring principles of drought index SPI, and using IDL achieve algorithm.Based on Rainfall data from1960to2012,SPI values were calculated for52years betweenJanuary. extract SPI values from2001to2012years to validate different dimension.（6） Extract Nine pixel VHI and MNDMI values at the same position with themeteorological station. The VHI is assumed to be validated remote sensing products WithMNDMI and SPI as a test product, achieve validation process, analyze test results, determinethe applicability of advantages and disadvantages of the three drought indices.The results show that: VHI and MNDMI more sensitive to moisture, suitable forreal-time monitoring of drought; SPI suitable for analyzing time series of precipitationrelationship with drought; MNDMI drought monitoring rule is consistent with the VHI, butaccording to the classification described in this article, the monitoring results are not a goodmatch with the VHI, Should be combined with more field data to explore more suitableMNDMI drought classification; Different dimensions of validation results were significantlylower than the results between same dimensions, but drought monitoring rule is consistent.