Making inference from wildlife collision data: inferring predator absence from prey strikes

Area of Science:

• Ecology
• Wildlife Management
• Conservation Biology

Background:

• Wildlife collision data offer insights but are hampered by sampling uncertainties.
• Ecological inference from such data requires robust statistical methods to address biases.

Purpose of the Study:

• To develop a novel statistical approach for ecological inference from wildlife collision data.
• To estimate predator population size and extinction probability using predator-prey collision data.
• To test hypotheses regarding species distribution and abundance, exemplified by red fox incursions.

Main Methods:

• Conditioning on a second prey species within the same collision data.
• Utilizing biologically realistic numerical response functions for predator-prey dynamics.
• Applying statistical conditioning to account for unmeasured variables affecting collision incidence.

Main Results:

• A coherent numerical response relationship was established between predators and prey.
• The method was applied to estimate the probability of red fox presence in Tasmania.
• A predictive probability of 0.001 for observing no fox strikes after 15 lagomorph strikes was calculated.

Conclusions:

• Sufficient evidence exists to reject the null hypothesis of a widespread red fox population in Tasmania.
• The developed method provides a powerful tool for ecological inference from collision data.
• The approach has broad potential applications in wildlife ecology and conservation.