Identifying small mammals' hairs collected with a non-invasive low-cost method on Staten Island, New York.

In the absence of sufficient manpower or time, hair tubes represent a humane and cost-effective method for monitoring small mammal presence. We describe a methodology for morphological identification of hairs collected in hair tubes placed in urban yards of Staten Island using existing literature and a collection of reference hairs. Hair tubes targeted mice (Peromyscus leucopus and Mus musculus), meadow voles (Microtus pennsylvanicus), Eastern moles (Scalopus aquaticus), chipmunks (Tamias striatus), and northern short-tailed shrews (Blarina brevicauda). We used Factor Analysis of Mixed Data and a subsequent classification model to evaluate our hair classification algorithm based on hair medullary appearance, hair thickness, and hair length. We observed a high correspondence between model predictions of the hair identity based on reference hair and our identification methodology for all analyzed target species (85% of hair samples were in agreement), except chipmunks. Hair morphology proved to be a valuable technique for monitoring the distribution of small mammals in urban yards, with high acceptability among homeowners. However, detailed guides for undertaking such studies are still rare. We present an overview of our methodology for future use in similar studies.