The U.S. radiometric calibration procedure for the reflective bands of the Landsat-5 Thematic Mapper was updated in May 2003. This update was based on a model of the performance of the instrument developed from its response to the best-behaved internal calibration lamp and from a cross calibration with Landsat-7 ETM+ that occurred in June 1999. Since this update was performed, there have been continued attempts to validate the model. These validations have relied primarily upon data acquired over deserts of the world. These studies have been limited by the amount of data available over any one site for the 22-year life of the mission. Initial attempts over the desert Southwest of the United States were inconclusive, though they were suggestive of additional degradation occurring in the shorter wavelength channels. More recently, significant holdings from European Space Agency of data over North Africa have been made available for analysis. The North Africa test area results to date for one site in Libya are considerably less noisy than the North American datasets. They indicate an exponential-like decay of about 19%, 16%, 8% and 4% for TM bands 1, 2, 3 and 4, with the degradation, at least in bands 1 and 2 occurring throughout the mission. The current model shows changes of roughly the same magnitude, but with the change occurring more rapidly so that nearly all the change is completed in 4 years. These results are generally consistent with independent work going on outside of this effort. Additional sites are being analyzed as data become available.