As the global climate changes and the populations of Niger, Mali and Burkina-Faso (abbreviated to NMB-F) grow, these countries face daunting challenges to ensuring adequate local food availability. Landlocked, and with a large portion of their total land area located securely within the boundaries of the Sahara desert, this cluster of countries already has limited arable land and faces losing farmland as the climate changes. In this research we quantify the relationships between food availability and food insecurity through the use of several NASA products -- the Moderate Resolution Imaging Spectroradiometer (MODIS) landcover product, Normalized Difference Vegetation Index (NDVI), the Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) and a measure of soil moisture from an existing Land Data Assimilation System (LDAS) project - and population/health survey data with a particular focus on household food insecurity. Similarities in climatic and topographic gradients in this tri-country region indicate the land cover land use change (LCLUC) differences reflect social and political forces. These differences lead to variations in human health outcomes reflected at the household level. We propose a three-stage analysis plan, building on the natural experiment setting provided by these countries that share many climate and population characteristics but differ on agricultural production or land use strategies. The three stages are as follows: 1) community-level estimates, based on MODIS data, of locally available food 2) country-specific models examining the relationship between food insecurity and locally available food, and 3) country-specific human-health impacts of climate change as outcomes of changes in expected agricultural output. Methods To accomplish our research objectives we have developed a series of methods. In Stage 1 we rely on NASA’s MODIS land cover product to identify broad areas with agricultural potential. We will then construct a 250 meter grid to identify the cultivated areas at a fine scale. The fine scaled interpretations allow for the estimation of cultivated area at the community-level. We will then build a cultivated area prediction model for locations not covered by the very high resolution data using -- rainfall, slope and elevation (DEM). We will also include both NDVI and a measure of soil moisture from an ongoing NASA funded LDAS project. Both NDVI and the LDAS-based measure of soil moisture provide information at a fine scale supporting our goal of producing local estimates of cropping as a direct measure of the amount of locally available food. This strategy will be applied separately to each of the three countries to produce country-specific estimates of community-level local food availability. In Stage 2 we incorporate results from Stage 1 into regression models to examine food availability and the relationship between several health outcomes related to food insecurity. The regression models are constructed around the hierarchy of the data -households are nested within communities. In this way we can examine the impact of community-level food availability on household malnutrition. In Stage 3 we incorporate climate data into the models created in the previous stage and estimate small-scale food availability given different climate scenarios. One of the primary drivers of individual-level malnutrition and household-level food insecurity is food availability. In this project we will construct estimates of local food availability using remotely sensed imagery. We will then examine the relationship between micro-level malnutrition outcomes and food availability, ultimately expanding scientific understanding of household food insecurity. The research results will be relevant to the scientific community interested in micro-level food insecurity in a context of climate change.