Using landsat to identify thunderstorm damage in agricultural regions

Abstract

During 12 and 18 August 1999, severe thunderstorms produced damaging winds and hail that caused an estimated $50 million in damage to agriculture in west-central Illinois. Landsat-7 imagery was obtained to determine the areal extent of damage and produce a crop damage dollar estimate. The normalized difference vegetation index (NDVI) was calculated for images taken "before" and "after" the severe thunderstorm events in order to examine the changes in NDVI, or vegetation vigor. A differenced image was also produced by subtracting the NDVI of the two images. Landsat-7 data was found to be useful for identifying the areal extent of severe thunderstorm damage in west-central Illinois. In comparing the detection of damage produced by high winds and hail, it was found that hall damage was considerably easier to identify. This is due to the fact that large hail typically destroys the crops while high winds blow over corn plants that can remain rooted and survive. Additionally, county estimates of dollar losses in crops were produced and compare favorably with estimates contained in Storm Data. Findings suggest, however, that Storm Data reports are inadequate for attempting to determine the areal extent of damage due to the difficulties in drive-by, ground-based estimation. Storm Data is primarily useful for locating the general area and extent of storm damage when reports and loss estimates were able to be obtained by the local National Weather Service Office.