The European wildcat (Felis silvestris) is increasingly expanding its range again in Germany. The species depends on structurally rich habitats that provide sufficient cover and are as well connected as possible. Contrary to earlier assumptions, suitable habitats may occur not only in largely undisturbed areas with low anthropogenic influence, but also in urban and densely populated landscapes, provided that sufficient suitable structures are available. The aim of this thesis was to conduct a BRT-based habitat modelling approach using telemetry data from five wildcats in the Leipzig floodplain forest, with potential transferability to other anthropogenically influenced forest habitats. The influence of distance to cover structures, settlements, roads and open land on habitat suitability was analysed. In addition, sex-specific and seasonal differences in habitat use were examined.
Boosted Regression Tree-based habitat modelling enables a robust analysis of complex habitat rela-tionships. The method captures nonlinear relationships and interactions between environmental varia-bles and, at the same time, allows the most important predictors to be prioritised based on their relative influence. By transferring the model results into GIS, spatially explicit habitat suitability maps can be created, making suitable habitats, potential connectivity areas and deficit areas visible. These maps provide a sound basis for conservation assessments and the planning of management measures.
The modelling results show that wildcat habitat use is largely shaped by anthropogenic structures and the availability of cover. In the overall model, distance to settlement structures had the highest relative influence on the modelled probability of occurrence, accounting for 58.9%. Habitat suitability in-creased markedly with increasing distance from settlement areas, particularly from a distance of ap-proximately 1,340 m. Areas close to settlements therefore had a clearly limiting effect on predicted habitat use.
Distance to forest and woody structures was the second most important predictor, with a relative in-fluence of 23.4%. The highest probability of occurrence was reached in the immediate vicinity of cover-providing structures. With increasing distance, particularly between approximately 20 and 150 m, habitat suitability declined markedly. Roads had a lower, but nevertheless ecologically relevant, influence, accounting for 10.8%. The probability of occurrence was reduced near roads and increased with greater distance. Distance to open land had the lowest relative influence, at 7.0%, and showed no clearly linear relationship. This indicates that open land is mainly used in relation to adjacent cover structures.
The modelled suitability areas in Saxony are mainly concentrated in larger, contiguous and compara-tively undisturbed forest landscapes in the Vogtland, the Ore Mountains, Saxon Switzerland, Upper Lusatia and the heathland areas of northern Saxony.
The sex-specific models show that suitable habitats for females are spatially more restricted than those for males. While 6.3% of the state area was calculated as highly suitable (> 90%) for males, this
12
proportion was only 1.9% for females. The proportion of unsuitable areas was also considerably higher for females. Seasonal differences were mainly reflected in the increased importance of proximity to forest during the period from February to May, whereas distance to settlements dominated during the period from June to September.
Overall, this thesis demonstrates that the wildcat shows a certain degree of flexibility in habitat selec-tion, but that this flexibility is clearly linked to structural richness, cover, low disturbance and connec-tivity. For the Leipzig floodplain forest, this results in particular conservation relevance. Despite its urban location, it represents a suitable habitat but depends on the preservation of low-disturbance core areas and functional connecting corridors. The results can therefore serve as a basis for biotope net-work planning, corridor development and targeted conservation measures for the wildcat in Saxony.