We characterized the microbes that are attached to deep-water moss in Crater Lake, Oregon. Moss leaves are densely colonized with a microbial community that differs from that found in the water column. Most notably, the cyanobacterial epiphytes on Crater Lake's moss are absent from the water column. Algal and cyanobacterial moss epiphytes were surveyed and identified using microscopy and gene sequencing. We found that diatoms and filamentous algae took up the greatest surface area on moss, with cyanobacterial epiphytes present in fewer numbers. Heterotrophic bacteria were more abundant on moss leaves than eukaryotes but occupied far less surface area. Eukaryotic large subunit rRNA gene sequences obtained from moss DNA extracts verified the presence of several different genera of algae associated with moss. In addition, sequence analysis revealed the presence of symbiotic fungi and protozoa. To identify prokaryotic nitrogen fixers on moss, nitrogenase and nitrogenase-like gene sequences were obtained from moss epiphytes. All of the nitrogenase genes associated with moss epiphytes were cyanobacterial sequences. In contrast, no nitrogenase sequences were detected in the water column. In addition, we cultured and identified four different genera of cyanobacteria from Crater Lake moss leaves. Given that Crater Lake has virtually undetectable levels of nitrate in the zone where moss grows most abundantly, the nitrogen fixing ability of cyanobacterial epiphytes is likely to provide a growth advantage to cyanobacteria and to nearby organisms.

A multi-faceted management approach on a group by group basis may be the most effective strategy for managing human-elk conflict within the North Cascades elk herd. Consideration of spatial distribution is important when developing management strategies designed to benefit or manipulate elk habitat use. We analyzed data from 23 GPS collared female elk (Cervus elaphus) in known conflict areas and adjacent upland forests to investigate spatial distribution, site fidelity, and home range overlap in northwest Washington. We identified 21 non-migratory sub-herds associated with our collared elk with discrete home ranges, year-round site fidelity and predictable core use areas. Home range areas ranged from 1.34–29.79 km² with 50% core use areas ranging from 0.02–1.67 km². We used seasonal median centers as indicators of site fidelity and found that all but one of the 21 groups had seasonal and total median centers < 3 km apart within their 95% home ranges. Home ranges showed minimal overlap between groups with Utilization Distribution Overlap Index values of zero or < 1 for 20 of the 21 groups. Groups that did overlap or had adjacent range boundaries showed incursion tracks suggesting that attempts were made to access occupied habitat. While the herd continues to recover, we recommend managing elk in conflict zones at the sub-herd level using a combination of techniques to manipulate behavior, block resource access, and control group size with the overall goal of influencing spatial distribution without removing elk from the landscape altogether.

Washington State's only endemic fish, Olympic mudminnow (Novumbra hubbsi), was listed as state “Sensitive” in 1999 due to a small range and concern about population declines and threats, including impacts from nonnative competitors and predators. We investigated the trophic ecology of Olympic mudminnow to identify habitat and foraging specialization across three sites in Lake Ozette (Olympic National Park, Washington), and evaluate evidence for resource interactions with sympatric fish species. Stable isotopes of carbon and nitrogen from dominant co-occurring fish—yellow perch (Percaflavescens), prickly sculpin (Cottus asper), and northern pikeminnow (Ptychocheilus oregonensis)—were evaluated for evidence of competition with and predation on Olympic mudminnow using trophic niche overlap and stable isotope mixing models. Evidence from co-occurrence and isotope analyses supported spatial and trophic segregation of Olympic mudminnow, and reduced likelihood of competitive and predatory relationships. Of the sympatric fish examined, prickly sculpin may be the most likely to experience biotic interactions with Olympic mudminnow. These results are consistent with other research demonstrating strong habitat and microhabitat specialization by Olympic mudminnow and suggest that access to these favored habitats may be an important mediating factor in defining the potential for biotic interactions with other species, including nonnative fishes.

Pools provided by beaver (Castor canadensis) contribute to critical habitat requirements of salmonids in fluvial systems of the Pacific Northwest, therefore more land managers are interested in managing watersheds that include beavers or engaging in beaver-related restoration projects. We evaluated the utility of applying an existing beaver habitat suitability model to better understand beaver dam site characteristics in coastal Oregon, identify optimum dam site locations, and guide future beaver-related restoration efforts. We used a combination of t-tests, Wilcoxon rank sum tests, and a stepwise discriminant function analysis to examine stream habitat associations with field data collected at known and predicted dam sites at reach and pool/riffle levels. We found bank-full width, valley floor width, and channel gradient performed well in predicting dam locations across the Alsea River Basin. Known dam sites had wider valley floors, shallower shoreline slopes, and fewer larger, deeper pools than predicted sites. Overall, our results suggest the beaver habitat suitability model combined with a digital elevation model can be used to guide where beaver dams may occur within the Alsea River Basin, yet they do not capture fine scale habitat associations that may lead to a settling response in beavers. For example, presence of large deep pools may be necessary for beavers to escape predation before and during dam building. Results from our study may be used to prioritize potential dam sites in other coastal basins that have similar geomorphic characteristics.

Many ground squirrels in western USA persist in degraded grass-dominated habitats where diet choices are more limited than prior to European settlement. This study provided a rare opportunity to investigate ground squirrel diets in relatively undisturbed habitats with a wide range of native diet options. We studied the diet of the northern Idaho ground squirrel, Urocitellus brunneus, a Threatened species under the US Endangered Species Act. It currently has a small, fragmented distribution of about 20 km² in west-central Idaho. We collected fecal pellets over three years at five study sites with high vascular plant diversity (93 to 133 species) and analyzed them using a microhistological technique. Over all sites and seasons, diets were diverse (74 diet items); 45 species of forbs contributed 86.3% of the observed diet, whereas nine species of graminoids contributed 7.8%. Forb foliage was most utilized, followed by underground parts (roots, bulbs, etc.), and then flowers, whereas seeds were a minor component. Principal components analysis showed that season and age class contributed more to diet variation than sex or site. Kernel regression tracked changes in forb utilization over the active season from foliage to underground parts, and then to flowers and seeds. These data suggest that a prudent restoration strategy for ground squirrel habitat should focus on establishing and maintaining local native forb diversity, especially species with heavily utilized underground parts, flowers, and/or seeds. Legumes and composites are especially important. However, local ecotypes of short-statured grasses such as Poa secunda may also be useful.