OPTIMIZING REMOTE SENSING OF NATIVE GRASSLANDS FOR MONITORING AND CHANGE DETECTION

Almost concurrent imagery from Landsat-5 and Radarsat-2 are examined separately and in combination to maximize the accuracy of a simple classification of a typical multi-use grassland region in western Canada. Almost all classifications were of sufficient accuracy to be used in an operational sense. Landsat seven band classification was the most accurate, but was deemed less likely to be useful as an operational tool due to its relatively long revisit period and its weather sensitivity. The Radarsat Constellation Mission satellites, when fully deployed, will provide sufficiently accurate classifications with a very short revisit period.