| Summary: | In this study, eleven mega predators, coyote (Canis latrans), wolf (Canis lupus), fox (Vulpes vulpes), arctic fox (Vulpes lagopus), black bear (Ursus americanus), brown bear (Ursus arctos), polar bear (Ursus maritimus), wolverine (Gulo gulo), marten (Martes americana), lynx (Lynx canadensis) and golden eagle (Aquila chrysaetos) were selected to represent an Ecosystem Unit entitled “Mega Predator”. The most influential factors affecting this Ecosystem Unit were determined using a machine learning algorithm (TreeNet) and a Geographic Information System (GIS). Public available range layers were corrected for errors and detectability using occupancy model, and several ‘robust’ hotspots of the predator community were identified. Anthropogenic variables, such as proximity to railways, together with regionalized IPCC climate variables (precipitation and temperature), Alaska SNAP data and spatial variables (e.g. distance to coast) proved to be the main predictors. A second predictive TreeNet model based on climate data forecasting the next 100 years was also performed to assess the resilience of these predators. The results indicate that the Ecosystem Unit “Mega Predator” shall undergo extreme changes in the next decades, commencing in 30 years or less. The TreeNet model points to a complete shattering of the current mega predator community food chain within the next century as a direct consequence of climate change alone. Owing to the fact that IPCC models are underestimates and other factors co-occur, the findings displayed herewith are consequently underestimates. The results of the first TreeNet model and the second predictive model were used to find the optimal potential protected areas for the predator community. This prioritization search was performed with the program MARXAN. Results of the MARXAN Model indicate that the main importance of protected areas for predators lies in the Brooks Range of Northern Alaska. This study could serve as a first (digital) platform and a first step to provide a basis for ...
|
|---|