Mark-recapture and mark-resight methods for estimating abundance with remote cameras: a carnivore case study
Robert S Alonso1, Brett T McClintock2, Lisa M Lyren3
1Department of Fish, Wildlife, and Conservation Biology, Colorado State University, Fort Collins, Colorado, United States of America; Western Ecological Research Center, Biological Resources Discipline, United States Geological Survey, Thousand Oaks, California, United States of America.
Plos One
|March 31, 2015
Summary
A new hybrid mark-resight model improved bobcat population size estimates. This method offers greater precision than traditional mark-recapture and mark-resight techniques for uniquely identifiable carnivore populations.
Area of Science:
- Wildlife ecology
- Population dynamics
- Conservation biology
Background:
- Estimating carnivore abundance is challenging, often relying on non-invasive methods like camera traps.
- Traditional methods include mark-recapture for uniquely patterned species and mark-resight for others.
Purpose of the Study:
- To compare traditional mark-recapture and mark-resight methods with a novel "hybrid" mark-resight model for bobcat (Lynx rufus) population estimation.
- To assess the precision of different abundance estimation models.
Main Methods:
- Deployed 30 cameras in urban southern California and physically marked 27 bobcats with GPS collars.
- Utilized unique pelage patterns for individual identification alongside physical markings.
- Applied traditional mark-recapture, traditional mark-resight, and a newly developed hybrid mark-resight model.
Main Results:
- All methods yielded similar abundance estimates, but precision varied significantly.
- Traditional methods produced imprecise estimates with confidence intervals exceeding 100% of point estimates.
- The hybrid mark-resight model achieved higher precision, with confidence intervals not exceeding 57%.
Conclusions:
- The hybrid mark-resight model offers a more precise alternative for estimating populations of uniquely identifiable species.
- This method effectively integrates data from both camera traps and physically marked individuals.
- The developed estimator is particularly valuable for species challenging to sample with camera traps alone.
