Using high resolution satellite imagery to estimate cropped area in Guatemala and Haiti

Guatemala and Haiti are two of the most food insecure nations in the Western Hemisphere. Measurements of food availability and access are instrumental in developing targeted hunger reduction strategies yet no estimates of cropped area (a critical input in the calculation of food production) at either a national or sub-national-level exist. The purpose of this research is to produce estimates of cropped area for Guatemala and Haiti using an area frame sampling approach and very high resolution (∼1 m) satellite imagery. Related research has combined livelihood data with topographic information to construct cropped area estimates in other settings using generalized additive models. We expand this approach with the inclusion of specific population variables in place of the livelihood data. We produce estimates of cropped area for the two countries and sub-national units and our results highlight the significance and complexity of incorporating explicit population characteristics into models of cropped area.

► We use satellite imagery to estimate cropped area in two developing countries.
► We include geophysical and population variables into models to predict cropped area.
► Population variables aid in prediction in some cases.
► Correlates and predictors of cropped area depend on the setting.