As cropland management and land use shifted towards more intensive practices, global land degradation increased drastically. Understanding relationships between ecological and socioeconomic drivers of soil and landscape degradation within these landscapes in economically dynamic contexts such as the Mediterranean region, requires multi-target and multi-scalar approaches covering long-term periods. This study provides an original approach for identifying desertification risk drivers and sustainable land management strategies within Italian agro-forest districts. An Environmental Sensitivity Area (ESA) approach, based on four thematic indicators (climate, soil, vegetation and land-use) and a composite index of desertification risk (ESAI), was used to evaluate changes in soil vulnerability and landscape degradation between the years 1960 and 2010. A multivariate model was developed to identify the most relevant drivers causing changes in land susceptibility at the district scale. Larger districts, and those with a higher proportion of their total surface area classified as agro-forest, had a significantly lower increase in land susceptibility to degradation during the 50 years when compared with the remaining districts. We conclude that preserving economic viability and ecological connectivity of traditional, extensive agricultural systems is a key measure to mitigate the desertification risk in the Mediterranean region.

Climate change and human activities in particular are important causes of the possible variations in Mediterranean basin forest health conditions. Over the last decades, deciduous oak-forest mortality has been a recurrent problem in central and southern Italy. Despite the perception of increasingly visible damage in oak forests in drought sites, the role of various environmental factors in their decline is not completely clear. Among the modern methods of monitoring terrestrial ecosystems, remote sensing is of prime importance thanks to its ability to provide synoptic information on large areas with a high frequency of acquisition. This paper reports the preliminary results regarding a replicable and low cost monitoring tool planned to quantify forest health conditions based on the application of the Normalized Difference Vegetation Index (NDVI), using the diachronic images provided by the Sentinel-2 satellite. The study area is represented by a peri-urban forest of natural Mediterranean deciduous oaks, characterized by a high variability in the composition of the species and in the silvicultural structures. In order to monitor the health conditions of a specific forest canopy cover with remote sensing data, it is necessary to classify the forest canopy cover in advance to separate it from other species and from the Mediterranean scrub. This is due to the spatial distribution of vegetation and the high rate of biodiversity in the Mediterranean natural environment. To achieve this, Light Detection and Ranging (LiDAR) data, forest management data and field sampling data were analyzed. The main results of this research show a widespread decline in oak health conditions over the observed period (2015–2017). Specifically, for the studied area, thanks to the specific localization of the oak canopy cover, we detected a high potential concerning the Sentinel-2 data application in monitoring forest health conditions by NDVI application.