Forest is a critical component of the earth’s surface, covering about 30% of the land area (e.g., FAO, 2001). Forest-cover changes, especially those of anthropogenic origin, have a wide impact on critical environmental processes including energy balance, water cycle, and biogeochemical processes. Understanding such changes, as well as their causes, requires that the changes be quanti‡ed. Reliable and up-to-date information on forest and forest change is required not only for resource management and ecological applications but also for addressing many pressing issues ranging from local to global scales, including carbon assessment, ecosystem dynamics, sustainability, and the vulnerability of natural and human systems (Band, 1993; Houghton, 1998; Lal, 1995; Pandey, 2002; Schimel, 1995). With its ability to obtain repeated observations of the earth’s surface, satellite remote sensing is a primary data source for forest change monitoring.