Development of a non-invasive method for species and sex identification of rare forest carnivores using footprint identification technology

Abstract

Many wildlife species have sex specific habitat requirements. Due to the unique requirements for birthing and raising offspring, female reproductive habitat is often a limiting factor for a population and has been identified as a priority for conservation. Therefore, the ability to detect where females persist on the landscape and identify these potential reproductive areas is essential in creating effective conservation strategies. Here we describe the development of a non-invasive method to identify species and sex based on track images collected at track plate stations using footprint identification technology (FIT). We developed this technique using data from the southern Sierra Nevada fisher (Pekania pennanti) population and a co-occurring species of conservation concern the Pacific marten (Martes caurina). We coupled track plate footprints with non-invasive genetic samples and camera trap images to create a reference dataset of known species for fisher and marten and known sex for fisher. We used FIT to geo-reference 167 marten and 367 fisher tracks (34 males, 27 females) using 7 landmark points and then extracted 124 morphometric variables (distances, angles, and areas) for use in identifying species and sex for fisher using linear discriminant analyses. Using a single variable, we found species classification accuracy >99% in distinguishing fisher from marten. For fisher sex identification our most parsimonious model consisting of only 2 variables achieved an accuracy of 94.0% for the training set and 89.4% for the test set. We also report a method to quantify classification uncertainty for each track. This method provides a rapid, cost effective, entirely non-invasive method to accurately identify sex that can easily be implemented in field studies.