摘要：The objectives of this study were to evaluate several spectral indices for estimating soil tillage practices and determining crop residue cover using multispectral and Hyperspectral satellite images. For this purpose, Landsat satellite and EO-1 Hyperion imaging spectrometer data were acquired over agricultural fields in central Ohio, near Columbus, in April and May. According to the calculated mean normalized difference tillage index data (NDTI), in the period 2003 to 2014, area of conservation tillage fields increased at a rate of 8.41 %, while no-tillage fields decreased at a rate of 14.45 %. Comparing normalized difference index (NDI) data from 2003 and in 2014 it was determined that a reduction of 10% vegetated areas occurred. Normalized difference senescent vegetation index (NSDVI) values did not evidence significant change over a nearly 10 year period. The best determination of tillage practices was obtained using the cellulose absorption index (CAI) value calculated from Hyperspectral data. By combining information on crop classification with indexes, similar results were obtained for crop distribution over a ten year period.
关键词：Remote sensing ; hyperspectral ; tillage ; index
