Verlag der Österreichischen Akademie der Wissenschaften

Abstract:
Austria is one of the few countries in Europe that has not been recolonized by stable populations of wolves, yet many dispersing individuals have been observed. Understanding spatial and temporal patterns of recolonization can help prepare management agencies for conflict that may arise and allow for adaptive management, yet characterizations of the recolonization process are lacking in most areas where it is occurring. Here, a geospatial application of an agent-based model was explored as a potential tool in characterizing spatial and temporal patterns of wolf recolonization in Austria. Sub-models for wolf appearance in Austria, dispersal through a habitat-suitability model, mating, pack formation and death were developed in the Agent Analyst programming environment and parameterized with literature-derived values. Model outputs included total wolf numbers, wolf presence locations, and the number and location of packs formed. Throughout model runs, wolf presence locations were predictably focused near known neighbouring populations; yet different parameterizations resulted in varied larger-scale movement patterns. About half of all runs resulted in pack formation, predominantly near the Slovenian and Italian borders, indicating that it is possible for the model to predict recolonization. Although the approach has high uncertainty, it can lend insight into recolonization and can be refined through the collection of more empirical data and the application of further research into wolf decision-making.