Utilizing the time-to-event framework to estimate elk abundance over a large spatial scale in the Klamath Mountains of California

Abstract

In northern California, Roosevelt (Cervus canadensis roosevelti) and Rocky Mountain (C. c. nelsoni) elk occupy a wide variety of habitats over a large extent, including the Marble Mountains Elk Management Unit (MM EMU). Dense forest canopy and steep, mountainous terrain present significant challenges for monitoring elk populations using traditional aerial and ground-based methods. These constraints have resulted in inadequate spatial and temporal research and monitoring. To address the need for comprehensive and reliable elk abundance estimates, we implemented a landscape-level camera trap study within the MM EMU. We deployed 180 cameras and applied a time-to-event model to estimate elk abundance. This method uses the movement rate, area in front of each camera, and leverages the latency time to detection for a given species. Analysis yielded an estimate of 1,415 (95% CI: 1,044–1,919) elk across the management unit. Here, we present the use of a recently developed method to non-invasively estimate the abundance of unmarked elk at the landscape-level in northern California. Implementation of this method can provide reliable information to aid management decisions for the continued recreational, ecological, and economic benefits of elk and wildlife in general.

Keywords

• bundance
• california
• landscape-level monitoring
• non-invasive
• rocky mountain elk
• roosevelt elk
• camera traps
• time-to-event
• unmarked