Daniel R. Cox and Thomas A. Nelson

Department of Biological Sciences, Eastern Illinois University

Introduction

Beavers (Castor canadensis) are a keystone species capable of creating and maintaining critical wetland habitats. Beavers also can be a costly nuisance when their activities bring them into conflict with humans. Although, beavers are common throughout Illinois, they are not uniformly distributed across the landscape. Consequently, the Illinois Department of Natural Resources (IDNR) would like to improve its ability to monitor their relative abundance and populations trends statewide and on regional levels.

Beaver populations can be monitored in lakes, ponds, and marshes by counting colonies from the air along fixed transects. However, riverine habitats are more difficult to survey because tree cover and fluctuating water levels hide bank dens and beaver sign. Therefore, it would be beneficial to develop a better method for monitoring beaver populations in Illinois’ rivers. We tested whether various habitat measurements correlate with the relative density of beavers and could be used to indirectly estimate population size.

Objectives

1. Locate, map and quantify the distribution of colonies on the Embarras River,
2. Identify habitat characteristics that correlate with the relative abundance of beaver colonies,
3. Test the efficacy of the U.S. Fish and Wildlife Service’s habitat suitability index (HSI) model for predicting the presence and density of beavers, and
4. Develop a multiple regression habitat model for predicting the density of beavers in riverine habitats.

Methods

The Embarras River, located in central Illinois, is approximately 315 km in length. The river was partitioned into three ecological divisions based on topography, channel width and land use adjacent to the river. The number and location of colonies were mapped from Nov. 2001 to Feb. 2002. Next, all variables that may influence the quantity and quality of beaver habitat were selected a priori based on natural history information and variables used in previous beaver habitat models. These parameters were measured in the field or on aerial photographs between June and September 2002 using a stratified-random sampling scheme. A Spearman correlation analysis was conducted to determine whether pairs of habitat variables were closely related. Forward linear regression was used to determine which variables provided the greatest predictive significance.

Results and Discussion

Beaver colonies were abundant in the upper and lower divisions of the river (Figure 1). The mean colony-to-colony distance was x.x km, x.x km, and x.x km in the upper, middle and lower divisions of the river, respectively. An abundance of winter food, adequate den sites, and good water characteristics contribute to the quality of habitat along the river.

The HSI model did not produce useful estimates of beaver density on the Embarras River and is probably not useful for monitoring beavers in Illinois rivers. Correlations between HSI scores and beaver densities were low (r² =0.xx). The model was developed in the western U.S. where winter food supplies and stream gradients are limiting factors for beavers.

Since the HSI model proved to be a poor predictor of beaver density, we built a new habitat model using forward regression. A model containing four variables (stream gradient, channel width, number of tributaries/km, and percentage of the riparian zone that grassy) proved to be the best predictor of beaver density (r² = 0.46).

Regression Model to Predict Beaver Density:

Number of colonies/km = 1.492 – 0.142(stream gradient) + 0.902(# of tributaries/km) – 0.31(% grass) – 0.292(mean channel width)

Acknowledgements

We would like to thank Scott Meiners for his assistance on this project.

Funding was provided by the IDNR through Federal Aid to Wildlife Restoration Project W-135-R-3.