Comparison of earth observing-1 ali and landsat etm+ for crop identification and yield prediction in mexico

Abstract

This paper presents a comparison of Earth Observing 1 (EO-1) Advanced Land Imager (ALI) and Landsat-7 Enhanced Thematic Mapper Plus (ETM+) images collected over an agricultural region in northwest Mexico. Across 115 fields with a range of cover types, radiance measurements collected by ALI were within 3% of ETM+ for all five common bands. Crop discrimination was significantly improved with ALI compared to ETM+, with an increase from 85% to 95% accuracy for distinguishing maize from wheat fields. This improvement was attributed to the greater SNR in ALI, as well as the unique information content of ALI band 4p (0.84-0.89 /spl mu/m), which may be due to sensitivity to canopy water content. Yield predictions from reflectance-calibrated data did not reveal significant differences between the sensors. The greatest distinction between ALI and ETM+ was observed in the panchromatic band, with ALI providing more detailed information on inter- and intrafield radiance differences, which show promise for precision agriculture applications. We conclude that ALI meets or exceeds ETM+ performance for agricultural applications evaluated here, thus providing a plausible option for continuity of the valuable Landsat record.