Abstract
Drought is one of the most important and frequent natural hazards to agriculture production in North China Plain. To
improve agriculture water management, accurate drought monitoring information is needed. This study proposed a
method for comprehensive drought monitoring by combining a meteorological index and three satellite drought indices
of TM data together. SPI (Standard Precipitation Index), the meteorological drought index, is used to measure
precipitation deficiency. Three satellite drought indices (Temperature Vegetation Drought Index, Land Surface Water
Index, Modified Perpendicular Drought Index) are used to evaluate agricultural drought risk by exploring data from
various channels (VIS, NIR, SWIR, TIR). Considering disparities in data ranges of different drought indices,
normalization is implemented before combination. First, SPI is normalized to 0 — 100 given that its normal range is -4
- +4. Then, the three satellite drought indices are normalized to 0 - 100 according to the maximum and minimum
values in the image, and aggregated using weighted average method (the result is denoted as ADI, Aggregated drought
index). Finally, weighed geometric mean of SPI and ADI are calculated (the result is denoted as DIcombined). A case study
in North China plain using three TM images acquired during April-May 2007 show that the method proposed in this
study is effective. In spatial domain, DIcombined demonstrates dramatically more details than SPI; in temporal domain,
DIcombined shows more reasonable drought development trajectory than satellite indices that are derived from independent
TM images.
