Assessing site occupancy of Mohave ground squirrels: Implications for conservation

Abstract

To better understand the distribution and habitat relationships of the Mohave ground squirrel (Xerospermophilus mohavensis), a California state threatened species, I investigated site occupancy of squirrels using remote‐sensing wildlife cameras (i.e., camera traps). In 2012, I conducted camera trap surveys at 96 randomly selected sites within and adjacent to the Desert Tortoise Research Natural Area, in the western Mojave Desert, California. In 1,105 camera trap days, there were 91 detections of squirrels at 30 out of 96 sites. I used single‐season, single‐species occupancy models to explore several survey and site variables hypothesized to influence detection probability (p) and site occupancy (ψ) of the species. Detection probability of squirrels varied with survey date and between 2 study areas, peaking in early April (p = 0.23, SE = 0.05 and p = 0.13, SE = 0.05, respectively) but decreasing approximately 4‐fold by the end of the study season in mid‐May. Site occupancy was positively related to the length of ephemeral stream channel(s) within a site, increasing from 0.24 (SE = 0.07) to 0.89 (SE = 0.10) with the range of stream channel lengths observed. Vegetation transects conducted at the study sites in 2014 indicated higher percent canopy cover, species richness, and diversity of perennial shrubs along ephemeral stream channels than in the surrounding upland habitat. This study provides the first estimates of site occupancy and detection probability of Mohave ground squirrels using the non‐invasive method of camera trapping. The findings have implications for design and timing of future studies, and highlight the importance of protecting ephemeral stream channels for Mohave ground squirrel conservation. © 2015 The Wildlife Society.

A range‐wide model of contemporary, omnidirectional connectivity for the threatened Mojave desert tortoise

“…Importantly, our highresolution (1 m) desert wash model provides a novel method and variable for future habitat modeling in arid systems. Desert washes are important landscape features that enhance plant dispersal and provide key habitat for many wildlife species in these systems (Heaton et al 2006, Gaddis et al 2016, Kotschwar Logan 2016.…”

“…Importantly, our highresolution (1 m) desert wash model provides a novel method and variable for future habitat modeling in arid systems. Desert washes are important landscape features that enhance plant dispersal and provide key habitat for many wildlife species in these systems (Heaton et al 2006, Gaddis et al 2016, Kotschwar Logan 2016. Although we lacked data with which to evaluate our range-wide connectivity model in a comprehensive manner, we demonstrated a localized evaluation using existing relocation data, within and proximate to the Ivanpah Valley.…”

A narrative review on the use of camera traps and machine learning in wildlife research

“…Camera trapping is an efficient way to monitor animal populations remotely and provide real-time observations (Parker et al, 2020). One of the most used applications is the occupancy and distribution of species, in particular rare, endangered, or elusive species (Alexiou et al, 2022;Bisi et al, 2019;Johnson et al, 2020;Kely et al, 2021;Li et al, 2018;Kotschwar Logan, 2016;Murphy et al, 2018;Parker et al, 2020) (Table 1). Tra-ditionally, these were done by visual or auditory surveys, track counts, scat analysis, detection dogs, driven counts, and trapping (Parker et al, 2020).…”