Just Accepted

This article has been peer reviewed and accepted for publication. It is in production and has not been edited, so may differ from the final published form.

Improving the efficiency of aerial surveys for monitoring North American beaver population dynamics

Madeline Kenyon, Catherine Dennison , Viorel Popescu

Abstract

Context: The North American beaver (Castor canadensis) was extirpated from much of its range in the US in the 1800s due to fur trapping and land use change. However, the species has recolonized much of its former range, including the US state of Ohio. Since 2013, the Ohio Division of Wildlife (ODOW) has monitored trends in beaver colony density via aerial surveys of 40×40 km plots classified as low, medium, or high suitability based on the amount of wetland. Nonetheless, the current classification system may miss important correlates of beaver colony density. Aims: Our study aimed to (1) identify predictors of beaver colony density (number of colonies inferred from aerial counts of lodges) across Ohio, and (2) develop a model-based classification system to improve the efficacy of monitoring efforts. Methods: We used an aerial survey dataset of 54 40 x 40 km plots collected by ODOW annually between 2013 and 2020, along with a suite of environmental, anthropogenic, and climate variables in a mixed effects modeling framework to predict beaver colony density in Ohio. Results: Beaver colony density was positively associated with wetland and reclaimed surface mine areas and inversely associated with the proportion of agricultural lands. There was a negative interaction between wetland and surface mines; in general, beaver colony density increased with wetland and surface mine area. However, in plots with wetland area >1000 ha, beaver colony density was weakly negatively associated with surface mine area. Using median and interquartile ranges of model-averaged predicted beaver colony density, we developed a new classification of low, moderate and high suitability for both the survey plots and the entire state of Ohio. We found that eastern Ohio had high suitability, while the central and western parts of the state had lower suitability for C. canadensis. Conclusions: Our approach to identifying predictors for beaver colony density at broad spatial scales highlights the importance of reclaimed surface mines and wetlands for beaver populations, while the model-based habitat classification provides ODOW additional information for monitoring and beaver management decisions.

WR23105  Accepted 10 September 2024

© CSIRO 2024