Testing a high-resolution satellite interpretation technique for crop area monitoring in developing countries

Abstract

District-level crop area (CA) is a highly uncertain term in food production equations, which are used to allocate food aid and implement appropriate food security initiatives. Remote sensing studies typically overestimate CA and production, as subsistence plots are exaggerated at coarser resolution, which leads to overoptimistic food reports. In this study, medium-resolution (MR) Landsat 7 Enhanced Thematic Mapper Plus (ETM+) images were manually classified for Niger and corrected using CA estimates derived from high-resolution (HR) sample image, topographic and socioeconomic data. A logistic model with smoothing splines was used to compute the block-average (0.1°) probability of an area being cropped. Livelihood zones and elevation explained 75% of the deviance in CA, while MR did not add explanatory power. The model overestimates CA when compared to the national inventory, possibly because of temporal changes in intercropping and the exclusion of some staple crops in the national inventory.