Research Article: Modeling intrinsic potential for beaver (Castor canadensis) habitat to inform restoration and climate change adaptation

Date Published: February 28, 2018

Publisher: Public Library of Science

Author(s): Benjamin J. Dittbrenner, Michael M. Pollock, Jason W. Schilling, Julian D. Olden, Joshua J. Lawler, Christian E. Torgersen, Ulrike Gertrud Munderloh.

http://doi.org/10.1371/journal.pone.0192538

Abstract

Through their dam-building activities and subsequent water storage, beaver have the potential to restore riparian ecosystems and offset some of the predicted effects of climate change by modulating streamflow. Thus, it is not surprising that reintroducing beaver to watersheds from which they have been extirpated is an often-used restoration and climate-adaptation strategy. Identifying sites for reintroduction, however, requires detailed information about habitat factors—information that is not often available at broad spatial scales. Here we explore the potential for beaver relocation throughout the Snohomish River Basin in Washington, USA with a model that identifies some of the basic building blocks of beaver habitat suitability and does so by relying solely on remotely sensed data. More specifically, we developed a generalized intrinsic potential model that draws on remotely sensed measures of stream gradient, stream width, and valley width to identify where beaver could become established if suitable vegetation were to be present. Thus, the model serves as a preliminary screening tool that can be applied over relatively large extents. We applied the model to 5,019 stream km and assessed the ability of the model to correctly predict beaver habitat by surveying for beavers in 352 stream reaches. To further assess the potential for relocation, we assessed land ownership, use, and land cover in the landscape surrounding stream reaches with varying levels of intrinsic potential. Model results showed that 33% of streams had moderate or high intrinsic potential for beaver habitat. We found that no site that was classified as having low intrinsic potential had any sign of beavers and that beaver were absent from nearly three quarters of potentially suitable sites, indicating that there are factors preventing the local population from occupying these areas. Of the riparian areas around streams with high intrinsic potential for beaver, 38% are on public lands and 17% are on large tracts of privately-owned timber land. Thus, although there are a large number of areas that could be suitable for relocation and restoration using beavers, current land use patterns may substantially limit feasibility in these areas.

Partial Text

North American beaver (Castor canadensis) have long been recognized as ecosystem engineers, creating diverse and resilient wetland and riverine systems [1, 2]. Prior to near extirpation in the early 1900s due to over-trapping and habitat conversion [3], beavers and beaver-created wetland complexes were a ubiquitous component of riparian systems [4]. Many species depend upon these systems due to the high geomorphic complexity, aquatic thermal variability, and habitat diversity that they aford. For example, the decline in populations of some aquatic species, including Pacific Coho salmon (Oncorhynchus kisutch), have been partially attributed to the loss of beaver ponds [5], a feature that salmonids have evolved with since at least the Pleistocene [6].

Beaver reintroduction and relocation hold much potential for habitat restoration and for addressing the impacts of climate change [43]. Identifying where to perform those introductions over large areas, however, remains a conservation challenge. Here, we demonstrate how models of remotely sensed intrinsic habitat potential identify potential habitat with high confidence. This approach offers a straightforward method for developing accurate estimates of the potential for beaver habitat using readily available data. The accuracy of this model makes it particularly useful for identifying sites that are suitable for beaver relocation and beaver-assisted restoration.

 

Source:

http://doi.org/10.1371/journal.pone.0192538

 

0 0 vote
Article Rating
Subscribe
Notify of
guest
0 Comments
Inline Feedbacks
View all comments