Coal fires detectable in low resolution MODIS data

A ratio image of MODIS band 20 versus band 32 was generated and input into a fully automatic algorithm for the detection of subtle thermal anomalies. This algorithm was developed at DFD. It could be shown that the hottest coal fires in the Wuda and Ruqigou study areas can be extracted from the MODIS data. This applies also to the formerly unknown coal fires of Hulusitai, Shitanjing and southern Haibowan, which were formerly detected in ETM+ thermal data and validated in the field in September 2003 and June 2004.
The results of the MODIS analysis will be presented in a book about coal fires, with contributions from the international coal fire community still to be compiled this year.