Estimating habitat selection when GPS fix success is less than 100%.

Area of Science:

• Wildlife Ecology
• Animal Movement Ecology
• Conservation Biology

Background:

• Global Positioning System (GPS) data are crucial for understanding animal habitat selection.
• GPS fix success can be reduced by environmental factors, leading to data gaps.
• Ignoring these data gaps can bias habitat selection models and inferences.

Purpose of the Study:

• To develop a statistical model that corrects for habitat-induced GPS data loss.
• To improve the accuracy of animal habitat selection models when locations are missing.
• To provide a robust method for analyzing animal movement data with incomplete GPS fixes.

Main Methods:

• A novel resource selection function (RSF) was developed to model GPS fix success.
• The model formulation is analogous to occupancy models with imperfect detection.
• The method was demonstrated using mule deer (Odocoileus hemionus) GPS data and simulations.

Main Results:

• The proposed habitat selection model effectively accounts for missing GPS locations.
• Simulations showed the model remained largely unbiased with up to 50% of locations missing.
• The approach allows for more reliable habitat selection estimates despite data loss.

Conclusions:

• Accurate animal habitat selection analysis requires accounting for GPS data loss.
• This new RSF model offers a robust solution for incomplete GPS datasets.
• The method enhances the reliability of wildlife research and conservation strategies.