Albedo measurement for remote sensing of crop yields

Abstract

ANY remote sensing programme designed to estimate regional agricultural productivity has three major requirements. First, the crops present must be identified; second, the amount of land planted to each type must be determined, and third, the yields of the crops per unit land area must be evaluated. The large area crop inventory experiment (LACIE) uses measurements of reflected solar radiation to fulfil the first two requirements in a true remote sensing context, but to establish yield, LACIE relies on statistical and climatological models largely dependent on ground-based observations¹. Here we report the basis for a crop yield prediction programme that is not subject to this restriction. Working with wheat, we have shown that the intensity of reflected solar radiation can be used to evaluate crop yields per unit land area. Linked to a satellite system, this could make the worldwide surveillance of agricultural production a technological reality that could greatly stabilise erratic fluctuations in world grain trade of the type which have recently disturbed the world economy.