A Desert Tortoise–Common Raven Viable Conflict Threshold

Authors

• Kerry L. Holcomb, U.S. Fish and Wildlife ServiceFollow
• Peter S. Coates, USGS
• Brian G. Prochazka, USGS
• Timothy A. Shields, Hardshell Labs, Inc.
• William I. Boarman, Hardshell Labs, Inc.

Date of this Version

Winter 2021

Document Type

Article

Citation

Human–Wildlife Interactions (Winter 2021) 15(3): article 14 

doi: 10.26077/eeca-1eec 

Special topic: Raven management

Comments

United States government work. Public domain

Abstract

Since 1966, common raven (Corvus corax; raven) abundance has increased throughout much of this species’ Holarctic distribution, fueled by an ever-expanding supply of anthropogenic resource subsidies (e.g., water, food, shelter, and nesting substrate) to ecoregion specific raven population carrying capacities. Consequently, ravens are implicated in declines of both avian and reptilian species of conservation concern, including the California (USA) endangered and federally threatened Mojave desert tortoise (Gopherus agassizii; desert tortoise). While ravens are a natural predator of desert tortoises, the inter-generational stability of desert tortoise populations is expected to be compromised as annual juvenile survival is suppressed below 0.77 through a combination of raven depredation and other sources of mortality. To estimate the extent to which raven depredation suppresses desert tortoise recruitment within the Mojave Desert of California, we collected data from 274 variable-radius point counts, 78 desert tortoise decoy stations, and 8 control stations during the spring of 2020. Additionally, we complied a geodatabase of previously active raven nests, observed between 2013 and 2020. Raven density estimates from 4 monitoring areas ranged between 0.63 (eastern most) and 2.44 (western most) raven km^-2 (95% CI: 0.35–1.14 and 1.33–4.48, respectively). We used a Bayesian shared frailty model to estimate the effects of raven density and distance to the nearest previously active raven nest on the annual “survival” of juvenile desert tortoise decoys (75-mm Midline Carapace Length), which we then converted into survival estimates for 0- to 10-year-old desert tortoises by adjusting exposure to reflect natural activity patterns. At the 1.72-km median distance from the nearest previously active raven nest, the estimated annual survival of desert tortoises decreased as raven density increased, ranging among conservation areas from 0.774 (eastern most) to 0.733 (western most). Accordingly, our model predicts that desert tortoise populations exposed to raven densities in excess of 0.89 raven km^-2, at a distance < 1.72 km from a previously active raven nest, are not expected to exhibit inter-generational population stability, as excess additive juvenile mortality is expected to exceed the natural history limits of desert tortoise populations. These results also demonstrate that estimates of raven density, distance to the nearest previously active raven nest, and decoy “survival” rates can inform development of a desert tortoise–raven viable conflict threshold.