Use of RADARSAT-2 polarimetric SAR images for drought code mapping over Canadian Prairie grasslands

Abstract

To enable fire danger mapping, the Canadian Fire Weather Index System’s drought code (DC) component needs to be determined. This study reports test results on the use of CBand polarimetric synthetic aperture radar images to assess DC levels over Canadian Prairies grasslands. Forty-five RADARSAT-2 images acquired for the 2008-2012 fire seasons were analyzed to determine those corresponding to the extreme wet, extreme dry, and in-season wet DC conditions. These were used to compute a number of polarimetric parameters which were compared to identify which vary most significantly over the DC conditions within eight vegetation classes in the study area. The results show statistically significant differences for a number of parameters demonstrating their utility to distinguish between DC conditions independent of seasonal vegetation changes. The current study will enable further research to develop algorithms predicting DC values which would permit remote estimation of fire danger conditions on the Canadian Prairies grasslands.