Integrating robotics into wildlife conservation: testing improvements to predator deterrents through movement

Authors

Stewart W. Breck, USDA/APHIS/WS National Wildlife Research CenterFollow
Jeffrey T. Schultz, USDA-WS-National Wildlife Research Center, Fort Collins
David Prause, Colorado State University
Cameron Krebs, Krebs Livestock, Ione, Oregon
Anthony J. Giordano, Colorado State University
Byron Boots, University of Washington

Date of this Version

2023

Citation

Breck SW, Schultz JT, Prause D, Krebs C, Giordano AJ, Boots B. 2023. Integrating robotics into wildlife conservation: testing improvements to predator deterrents through movement. PeerJ 11:e15491 DOI 10.7717/peerj.15491

Comments

U.S. government work

Abstract

Background Agricultural and pastoral landscapes can provide important habitat for wildlife conservation, but sharing these landscapes with wildlife can create conflict that is costly and requires managing. Livestock predation is a good example of the challenges involving coexistence with wildlife across shared landscapes. Integrating new technology into agricultural practices could help minimize human-wildlife conflict. In this study, we used concepts from the fields of robotics (i.e., automated movement and adaptiveness) and agricultural practices (i.e., managing livestock risk to predation) to explore how integration of these concepts could aid the development of more effective predator deterrents.

Methods We used a colony of captive coyotes as a model system, and simulated predation events with meat baits inside and outside of protected zones. Inside the protected zones we used a remote-controlled vehicle with a state-of-the art, commercially available predator deterrent (i.e., Foxlight) mounted on the top and used this to test three treatments: (1) light only (i.e., without movement or adaptiveness), (2) predetermined movement (i.e., with movement and without adaptiveness), and (3) adaptive movement (i.e., with both movement and adaptiveness). We measured the time it took for coyotes to eat the baits and analyzed the data with a time-to-event survival strategy.

Results Survival of baits was consistently higher inside the protected zone, and the three movement treatments incrementally increased survival time over baseline except for the light only treatment in the nonprotected zone. Incorporating predetermined movement essentially doubled the efficacy of the light only treatment both inside and outside the protected zone. Incorporating adaptive movement exponentially increased survival time both inside and outside the protected zone. Our findings provide compelling evidence that incorporating existing robotics capabilities (predetermined and adaptive movement) could greatly enhance protection of agricultural resources and aid in the development of nonlethal tools for managing wildlife. Our findings also demonstrate the importance of marrying agricultural practices (e.g., spatial management of livestock at night) with new technology to improve the efficacy of wildlife deterrents.