A Rapid Method for Estimating Crop Losses Using Space Imagery

Abstract

A method based on the use of space imagery is proposed for a rapid estimate of the losses inflicted on agricultural crops by floodwaters forming as a result of downpours. Images registered by the MSU-E instrument aboard the Resurs-O1 satellite, having adequate spatial, temporal, and radiometric resolution, were examined. Digital processing is based on the use of cluster analysis (ISODATA) in which pixel-by-pixel matched images of different survey periods (before and after the flood) are processed. The method involves use of a new method for detecting changes in the condition of agricultural crops based on an analysis of the spectral reflectance of natural features before and after a flood. The normalized vegetation index (NDVI) is used in determining the condition of agricultural crops prior to flood waters. Areas of agricultural fields subjected to flooding in Slovakia in 1997 are determined and the material sustained by summer flooding in that year are calculated.