High resolution predictive modelling of potential natural vegetation under recent site conditions and future climate scenarios: Case study Bavaria

Potential natural vegetation (PNV) is a widely used concept to characterise complex site conditions of vegetation types. We developed a model to simulate the distribution of 26 PNV units (forest types) under recent and future climate change conditions (15 scenarios) in the Federal State of Bavaria, SE of Germany (70,550 km(2)) with an extraordinarily fine spatial resolution (50 m x 50 m; 28.1 million grid cells). We used MaxEnt and a set of 146 high-resolution site maps. Calibration of the model is based on 7,504 sampling points of the national forest inventory of Germany. Future scenarios include temperature increase only (+1 K, +2 K, +3 K, +4 K, +6 K) and with precipitation increase (+10%) or decrease (-10%). The model describes the current distribution of natural vegetation types very precisely. For the study area, all future scenarios show drastic changes in site conditions: The constellations of site conditions, which are absent in Bavaria at present, will increase from 36% in the +2 K to 94% in the +4 K scenario. Forestry as well as agriculture and land use management will then find themselves victims of global climate change caused by new site situations. With the model, we provide a tool that enables politicians at all administrative levels as well as practitioners on their operational level to identify climate change hotspots down to the local level by identifying areas with minor or stronger changes in site conditions to be expected under different climate change scenarios. Moreover, everybody can check how global change will influence his backyard.