Exploring the use of remote sensing tools and geospatial technologies applied to the multidimensional food security problem

[Objective] The aim of this study was to analyze which role remote sensing technologies can play to study multidimensional factors influencing the Food and Nutrition Security (FNS), in Cordoba Argentina. [Methodology] The study area includes the city of Cordoba, Argentina. Epidemiological data on the prevalence of underweight, overweight, and obesity (malnutrition) in 2013 were obtained from 23 primary health care centers in the city. The environmental conditions of the surroundings of the health centers were explored within a radius of 1000m. SPOT 5 images were classified using spectral and spatial features and we show how a non-supervised classification can give information to describe the social dimension and economic access to food. In addition, a multivariate stepwise linear regression was performed to examine the relation between the prevalence of malnutrition and the environmental and spatial variables, derived from the SPOT image, proposed. [Results] The results of the unsupervised image classification show the difference in the spectral-spatial pattern of neighborhoods showing how a simple satellite image classification can become a useful discrimination tool. Multiple regression analyses with adjusted R2 of 0.70 and 0.6435 respectively are obtained for undernutrition, and overweight, and obesity. On the basis of the obtained models, continuous maps of prevalence are built. [Conclusions] The method proposed in this work can discriminate socially different areas related to FNS. It is innovative and necessary to take advantage of remote sensing tools and geospatial technologies, in our region, applied to FNS.