Study On Comparisons Of Drought Indices Ongrassland By MODIS Data

Drought is a phenomenon that water supply and demand imbalance. There are having rising lots of monitoring methods and forming a multi-application, multi-spectral, multi-temporal monitoring system after many years development and application of remote sensing technology. However, the occurrence and development of drought is a complex process with various performances. For a long time, people monitor drought with remote sensing technology and establish a number of drought-related index that attempt to express accurately wet or dry spatial situation . Every method has their advantages and disadvantages. Generally speaking, these methods are either not enough to reflect dynamic; or lack of study adaptability and the validity of the assessment.In response to this problem, this article selects NDVI, NDWI, VCI, TCI, VSWI five drought index on the basis of comparing advantage and disadvantage of a variety of remote sensing drought monitoring methods, and determines the most applicable one to grassland. the main content and conclusions are as follows:(1) Correlation analysis between NDVI and different schemes precipitation during grassland vegetation growing season. The result reveals that there is a significant correlation relation among them with different correlation coefficient and the most significant relation is 50-60 days of accumulated precipitation. Lags time differ between different types of grassland vegetation NDVI response to precipitation; in general, the time of desert grassland is later than Meadow and the typical steppe's. In addition, during different time of the growing season, there are seasonal differences in the time lag. The lag time in each month from the April to September, are, respectively, 64, 80,40,40,56 and 64 days and especially the response of vegetation growth to precipitation in June and July are more sensitive than other moths.(2) Comparison between NDVI and NDWI. there both exists lag correlation relation among precipitation. The remote Sensing index, NDVI and NDWI, response poor to the same period rainfall, affected by the long-term precipitation. NDVI respond better to precipitation than NDWI. NDVI and NDWI have both close relations with air humidity. NDWI controlled by air humidity better than the NDVI, for the more significant relevance than NDVI.(3) Comparison of other drought index with CVA technology. The result reveals that there is no consistency spatial overlapping among different monitoring results. Different index identify different drought area distribution. In the aspects of reflection band index, NDVI and NDWI produce similar results. On the contrary, the thermal infrared band index, such as TCI has different results. The drought affected areas monitored by SPI and remote sensing drought index has no consistency. Statistical analysis showed that correlation among the reflectivity indices is better than infrared or mixed drought index. Drought index in remote sensing and meteorological drought index have very poor correlation.