The impact of lossy compression on classification of remotely–sensed imagery data is examined. The impact of compression is assessed for two types of classifications: unsupervised classification via thematic map for small-footprint imagery, and supervised classification via spectral unmixing for large-footprint imagery data. An overview of viable classification and spectral unmixing procedures are given. The criteria for measuring the impact of compression are defined. Simulation results using NOAA's AVHRR (1.1 km footprint), and LANDSAT 5 TM (30 m footprint) test imagery, show that the impact of compression is insignificant for compression ratios of less than 8. It is argued that the effective impact of compression is reduced due to the presence of other sources of inaccuracy in the original data and its relevant prediction models.