Operational modelling of peri-urban farmland for public action in Mediterranean context

This paper presents an analytical framework for peri-urban farmland characterisation that is operational for public action. Based on in-depth analysis of a Mediterranean local case study using surveys, on-site landscape reading, remote sensing analysis and interviews, we classed peri-urban farming into spatial units of peri-urban agriculture (USAPU). The classification obtained over seven municipalities was later used to train a fractional regression model, which was then tested on the rest of this French département (similar to NUTS-3 level), to predict the presence and actual proportion of each USAPU in the total agricultural land of each municipality. Furthermore, we drew up categories of municipality according to USAPU distribution that open perspectives for public action on peri-urban farming. We discussed whether such a model could be used as an instrument for decision-making on food planning, rather than simply for fairly reliable future predictions. This work is the starting point for the development of a methodology characterising complex peri-urban areas, simple to handle and hence operational for policy-makers and planners.