On building a general framework for assessing food security risk under probabilistic socioeconomic scenarios

Food security is a key issue in sustainability studies. In this work, a general framework for providing detailed probabilistic socioeconomic scenarios as well as predictions concerning food security is proposed. Our methodology builds (a) on the Bayesian probabilistic version of world population prediction model and (b) on the dependencies of food needs and food system capacities on key drivers, such as population, gross domestic product (GDP) and other socioeconomic and climate indicators. In this perspective, the concept of the recently developed convex risk measures involving model uncertainty is employed for the construction of a risk assessment framework in the context of food security. The proposed method provides within and across the various probabilistic scenarios predictions and evaluations for food security risk. Our methodology is illustrated by studying food security and quantifying the occurring risk in Egypt and Ethiopia up to the year 2050, in the combined context of the Shared Socioeconomic Pathways (SSPs) and the Representative Concentration Pathways (RCPs).