Procedures for wheat yield prediction using Landsat MSS and IRS-1A data

Abstract

This study describes procedures for analysis of data from Landsat MSS and Indian Remote Sensing Satellite IRS-IA for pre-harvest wheat yield prediction. A linear yield-spectral index model has been developed wherein radiance values have been used instead of digital counts in order to take care of data from different satellites/sensors. The spectral ratios have been normalized using a crop growth profile so that all the values correspond to maximum vegetative growth. Using these procedures district-wise grain yield prediction of wheat for the Haryana state in India for the 1988-89 growing season was carried out. Yield estimates from Landsat MSS and IRS-IA LISS-I data were in good agreement with one another. The maximum deviation of estimated yield by IRS- 1A data from that of Bureau of Economics and Statistics (BES) of the government was 18 percent.