Impact of farmland fragmentation on rainfed crop allocation in Mediterranean landscapes: A case study of the Lebna watershed in Cap Bon, Tunisia

Mediterranean agricultural landscapes provide ecosystem services and disservices that are driven by land use pattern dynamics, the latter of which results from the crop spatiotemporal distribution. Farmland fragmentation is known to be a driver of crop management and farm performance. However, existing studies on farmland fragmentation have not addressed the impact of farmland fragmentation and the subsequent neighbouring interactions on decision-making about annual crop allocation. Therefore, the current study aims to explore how much farmland fragmentation can drive the decisions made by farmer about annual crop allocation by characterizing and quantifying the influences of both crop sequences and neighbouring crops at the field scale. We addressed this issue within the Lebna watershed (210 km(2) size, located on the Cap Bon Peninsula in Tunisia), which is typified by a hilly topography, rainfed mixed farming (cereals, fodder, legumes, spices) and livestock (cattle, sheep and goat), and strong farmland fragmentation. The experimental phases consisted of conducting 30 farmer interviews and collecting data regarding the field spatial distribution for a total number of 360 fields in 2015 and 355 fields in 2016, hereafter called farmer fields. We also recorded (1) crop types for the farmer fields in 2015 and 2016 and (2) land uses (including crop types) in the neighbouring pieces of land in 2016. Data analysis relied on differentiating the farmer fields between isolated and non-isolated fields. Isolation/non-isolation depended upon farmland fragmentation and field dispersion relative to the other farm fields and to the farmstead. Using univariate and linear discriminant analysis on both crop sequences and neighbouring crops, data analysis revealed a significant effect of farmland fragmentation on farmer decision-making regarding crop allocation. When fields are isolated (fragmented farmlands), farmers implement with some of their neighbours, collective rules of crop allocation that permit the management of common constraints, such as the lack of roads to access fields. The landscape subsequently depicts aggregates of fields with the same type of crop. When fields are non-isolated (aggregated farmlands close to the farmsteads), the allocation constraints are mainly related to the cropping systems, with a strong impact from the crop sequences. Overall, these results indicate that to improve our understanding of crop allocation drivers at the landscape level, it is not only sufficient to address rules and drivers at the field and farm scales but it is also necessary to account for the collective contexts in which farmers operate.