A range-wide occupancy estimate and habitat model for the endangered
Point Arena mountain beaver (Aplodontia rufa nigra)
William J. Zielinski,* Fredrick V. Schlexer, Jeffrey R. Dunk, Matthew J. Lau, and James J. Graham
USDA Forest Service, Pacific Southwest Research Station, Redwood Sciences Laboratory, 1700 Bayview Dr., Arcata, CA 95521,
USA (WJZ, FVS)
Department of Environmental Science and Management, Humboldt State University, 1 Harpst Street, Arcata, CA 95521, USA (JRD, JJG)
Department of Wildlife, Humboldt State University, 1 Harpst Street, Arcata, CA 95521, USA (MJL) * Correspondent: firstname.lastname@example.org
The mountain beaver (Aplodontia rufa) is notably the most primitive North American rodent with a restricted distribution in the Pacific Northwest based on its physiological limits to heat stress and water needs. The Point Arena subspecies (A. r. nigra) is federally listed as endangered and is 1 of 2 subspecies that have extremely small and disjunct distributions, putting the taxon at risk from habitat loss and warming global temperatures. We sought to understand their range-wide distribution, to predict the environmental features that best describe their occurrence, and to establish a foundation for monitoring their distribution. We randomly sampled the occurrence of Point Arena mountain beavers (PAMB) from the portion of their geographic range that was accessible (public lands plus private lands where permission was granted). We surveyed 127, 25-ha sample units for their distinctive burrows and estimated the probability of detecting burrows, if they were present, at > 90% per visit. Using this information, we estimated occupancy across the accessible portion of the range to be 26.2%. Range-wide estimates of occupancy, combined with strategically selected locations where abundance and survival can be estimated noninvasively, may comprise a realistic monitoring program for this taxon. We also used the detection and nondetection locations to develop a habitat suitability model by relating these locations to remotely sensed predictors. We evaluated 53 a priori candidate habitat suitability models and the bestfitting model included gentle slopes, low terrain roughness indices, and the high density of rivers and streams. Selecting the probability value that best separated the sample units into suitable and nonsuitable habitat, we estimated a total of 70.5 km2 of suitable habit, or approximately 40.4% of the original geographic range. New detections significantly expanded the known geographic range, moderating concerns about habitat loss, including that predicted by climate change. A substantial number of suitable areas were predicted to occur outside the current range. The identification of high suitability areas allows management agencies to prioritize areas for PAMB conservation planning, evaluate human impacts on habitat, and evaluate how a changing climate may affect distribution.
Key words: Aplodontia, California, habitat, modeling, mountain beaver, occupancy, Point Arena
Published by Oxford University Press on behalf of the American Society of Mammalogists. This work is written by (a) US
Government employee(s) and is in the public domain in the US.
The mountain beaver (Aplodontia rufa) is the most primitive living rodent, with a distribution that is limited to the cool and moist environments in the Pacific region of North America (Feldhamer et al. 2003). Its restricted distribution has been explained by a limited ability to concentrate urine, thereby requiring abundant surface water or succulent vegetation in their diet (Nungesser and Pfeiffer 1965). The mountain beaver also lacks adequate mechanisms to avoid heat stress (Johnson 1971), which may also restrict the species to cool environments in the Pacific
Northwest. It is semi-fossorial, inhabiting complex burrow systems that frequently occur in moist seeps that host a dense tangle of vegetation (Dalquest 1948). The distribution of the species is relatively limited, but 2 of the 7 subspecies have extremely limited and disjunct distributions: the Point Reyes mountain beaver (A. r. phaea) and the Point Arena mountain beaver (PAMB;
A. r. nigra—Feldhamer et al. 2003). Both occupy ranges that are small (< 100 km2) and that have probably been shrinking since the Eocene due to drying of the regional climate as well as vegetation and topographic changes (Shotwell 1958).
The PAMB is listed as endangered under the United States
Endangered Species Act due to habitat loss from development and livestock grazing (Steele and Litman 1998; USFWS 2009).
Journal of Mammalogy, 96(2):380–393, 2015
ZIELINSKI ET AL.—MOUNTAIN BEAVER OCCUPANCY AND HABITAT 381
Despite its endangered status, there is little known about its ecology (e.g., Camp 1918; Fitts 1996; Fitts et al. 2002; Billig and Douglas 2007; Zielinski et al. 2010). The PAMB range includes 2 very different ecosystems: forests in mountains in the eastern portion and shrub and grasslands in the coastal terraces of the western portion of the range (Fig. 1). As a result, it is of conservation interest to understand the environmental features that are associated with their occurrence in these 2 very different realms. The small size of the known range, the vulnerability of the taxon to a warming climate (USFWS 2009), and the diversity of vegetation types that the PAMB appears to occupy within its range all demand a better understanding of the features within the range that favor their occurrence.
Recovery goals for the PAMB specify minimum standards for the number and size of populations and for monitoring trends in populations and geographic range (Steele and Litman 1998; USFWS 2009), yet quantitative information about population status is limited to abundance estimates within or immediately adjacent to one of the few protected areas within the subspecies’ range: Manchester
State Park (Northen and Fitts 1998; Zielinski et al. 2013a). Markrecapture methods were used to estimate population size for 2 locations within the park, and both had low but stable numbers from 2006 to 2009 (Zielinski et al. 2013a). At other historic locations throughout the subspecies’ range, recent searches found no evidence of their burrows (F. V. Schlexer, pers. obs.). However, whether new sites had been colonized over the same time period is