Recognizing successional recovery gradients following release from disturbance is important for sustainable land management. To evaluate the recovery and transitions of vegetation post disturbance there needs to be a chronology of community succession. This study in northwestern Montana evaluated the composition of recognized plant communities and proposes successional recovery pathways of the native grassland following release from livestock grazing. Initial surveys were completed 15 years after grazing ended. Inventories of 16 herbaceous vegetation communities indicate that litter depth was the only statistically significant factor related to community composition (P = 0.03) and, importantly, was highly correlated with non-native species canopy cover (r = 0.904). The lack of litter was correlated with the persistence of three climax species (r = 0.736). Comparisons with published grassland community descriptions indicated the intermontane grasslands in this refuge are part of two Northern Fescue Grassland habitat types (Fescue campestris/F. idahoensis and F. idahoensis/Carex filifolia) rather than communities common to the Palouse Prairie. There are two possible transitional pathways within these grassland communities. These pathways appear to be related to local soil moisture regimes. Moister sites transition from the degraded (Poa compressa/ Thinopyrum intermedium) community to the lower seral (T. intermedium) community type, to the mid-seral (C. filifolia) community type and finally to the climax (F. campestris) community type. The drier pathway transitions from the degraded (P. compressa/T. intermedium) community to the lower seral (P. compressa/F. campestris) community type, then to the mid-seral (Moss) community and finally to the climax (F. idahoensis) community type.

Understanding predator-prey dynamics is a fundamental principle of ecology and an ideal component for management decisions. Across North America, the impact of cougars (Puma concolor) on their prey varies regionally. To document the relationships between cougars, bighorn sheep (Ovis canadensis canadensis), mule deer (Odocoileus hemionus), and feral horses (Equus caballus) on the Bighorn Canyon National Recreation Area and the Pryor Mountains, we deployed GPS collars on 6 cougars (the total number residing on the study area), and visited their clusters to determine predation rates and foraging patterns. We examined the composition of cougar kills by species, mule deer sex-age classes, prey size classes, season, and cougar sex. As a measure of selection, we examined the composition of prey killed relative to the composition of ungulates obtained during an aerial survey. We found mule deer were the primary prey, while bighorn sheep constituted secondary prey. While cougars selected for bighorn sheep, this was attributable to a single cougar. Among mule deer, female cougars killed more does and male cougars killed more bucks. Family groups had the highest predation rates (i.e., the shortest time intervals between kills), while adult males had the lowest rate. During the study, cougars were not depredating any feral horses in the area. Maintaining predator and prey numbers will require agencies to monitor and man-age all fauna within this complex ecosystem. Habitat manipulations may be necessary to increase populations of deer and bighorn sheep, while continued management of feral horses will be required to reduce competition with native ungulates.

Six channel-spanning boulder weirs with other associated structural configurations were constructed along mainstem of Mosby Creek in 2011 to improve fish habitat. The structures were studied to evaluate their influence on water temperatures and their stability against hydraulic forces across a range of boulder sizes in different configurations. Boulders were configured as weirs, clusters, duos, barbs, and individuals, with mean volumes 1.4, 0.45, 0.42, 0.83, and 0.63 m³, respectively. Bankfull widths ranged from 9.1 to 12.2 m, with stream gradients between 0.8 and 1.2%. After summer construction, 671 boulders were surveyed and then resurveyed following winter flows. There were five high flow events during the first winter, with two above the 2-year recurrence interval. The weirs captured approximately 1200 m³ of alluvium after the first winter. Only minor movement of boulders was observed, with only seven moving > 10 m, and all weirs were intact. Boulders of similar density (mass/volume) in these flow conditions should be stable once boulder volume exceeds 0.5 m³. Three noteworthy findings were seen in the temperature analysis. Lower absolute maxima temperatures were observed directly below weirs. There was no change upstream or downstream of weirs, signifying that thermal change is local. Thirdly, maximum minima were similar following alluvial deposition, indicating little effect of daylight heat gain. While created hyporheic zones may have little stream-wide influence, small areas of cool water are thermal refugia for fish, particularly in systems where current temperatures can stress salmonids. Boulder weirs show promise as a method that develops such habitats.

As remote sensing mapping products representing important habitat components increase in availability, detail, and accuracy, new opportunities arise to map predicted habitat for forest species of conservation concern. One such species is the Lewis's woodpecker which is a species of conservation interest in multiple states and throughout its range in British Columbia, Canada with habitat loss and degradation proposed as contributing factors in the species decline. The purpose of this study was to model and map Lewis's woodpecker nesting habitat using primary (directly derived) and secondary (modeled using ground calibration data) remote sensing products, as well as contribute to the knowledge of nesting ecology for the species in a source habitat type. We found promise in the utility of lidar and Landsat time series modeled snag and shrub products in conjunction with additional lidar structure and topographic metrics for predicting nesting habitat for the Lewis's woodpecker across a post-fire landscape. Our selected model had a percent correctly classified rate of 83.3% and contained additional variables (variability in canopy cover and topographic slope) to those included in the Fish and Wildlife Service habitat suitability model (HSM) for the species (canopy cover, shrub cover, and density of large snags). We were able to apply the remote sensing products to map realized habitat relationships for this species of conservation concern in an identified source habitat type, providing a potential resource for local scale conservation and management efforts and adding to the regional knowledge of habitat selection for the Lewis's woodpecker.

Basic assumptions about the ecological effects of dam removal remain untested hypotheses. In this work, I move to address this by assessing the effectiveness of dam removal as a tool to restore anadromous Pacific salmonids' (Oncorhynchus spp.) access to upstream habitat. I compiled all available post-removal data from all cases (N = 40) where anadromous salmonid restoration was a project goal in California, Oregon, and Washington from 1999 to 2011. In cases without empirical data, I contacted biologists and river managers and collected anecdotal information. I analyzed full barriers (N = 16), where upstream passage was not possible for fish, separately from partial barriers (N = 27), where fish were able to pass upstream of the dam. In three cases, dams were passable to some species but not others. In cases of full barrier removal, formerly excluded anadromous salmonids were observed to have recolonized upstream reaches in 9 of 16 cases. In cases of partial barrier removal, results were uneven; they are discussed on a case-by-case basis. There were no useful data on 14 of 27 partial barrier removals, and there was only anecdotal evidence of success in 9 of 27 cases. Dam removal remains a promising restoration action for anadromous Oncorhynchus species, but monitoring, especially in cases of partial barrier removal, has largely been inadequate to determine whether these fish expand into newly accessible habitat after dam removal.

We examined the mercury hazard posed to selected piscivorous wildlife in Glacier National Park (GNP), Montana. Logging Lake was our focal site where we estimated the dietary mercury concentrations of wildlife (common loon [Gavia immer], American mink [Neovison vison], river otter [Lontra canadensis], and belted kingfisher [Megaceryle alcyon]) by assuming that fishes were consumed in proportion to their relative abundances. To evaluate if Logging Lake provided a suitable baseline for our study, we made geographic comparisons of fish mercury levels and investigated the distribution and abundance of high mercury fishes within GNP. We complimented our assessment by examining selenium:mercury molar ratios in fishes from Logging Lake and Saint Mary Lake. Our results suggest fish consumption does not imperil wildlife from Logging Lake based on published thresholds for adverse mercury effects, but some hazard may exist particularly if there is strong feeding selectivity for the most contaminated species, northern pikeminnow (Ptychocheilus oregonensis). The geographic comparisons of fish mercury levels, together with the distribution and abundance of high mercury fishes within GNP, suggest that Logging Lake provided a relatively protective baseline among our study lakes. Risk may be further reduced by the molar excess of selenium relative to mercury, particularly in the smaller fishes typically consumed by GNP wildlife. Our findings contrast with studies from northeastern US and southeastern Canada where greater mercury hazard to wildlife exists. An emergent finding from our research is that waterborne concentrations of methylmercury may provide limited insight into regional differences in fish mercury levels.

The reintroduction of fire to Pacific Northwest prairies has been useful for removing non-native shrubs and supporting habitat for fire-adapted plant and animal species. However, very little is known about the influence of fire on prairie bryophyte and lichen communities. In this study, we investigated the effects of fire on ground-dwelling bryophytes and lichens by estimating standing biomass, cover, mat depth and functional group diversity in burned and unburned plots at five prairie sites located in the south Puget Sound bioregion of Washington State. After accounting for differences among sites, the bryophyte and lichen ground layer in burned plots had about 18% the biomass and about 60% the cover of unburned plots. All ground layer functional groups had lower average biomass in burned plots, except for ephemeral mosses, which had greater biomass. Forage lichens and nitrogen-fixing lichens were absent from all five burned plots. Of particular concern are the regionally rare, state-listed reindeer forage lichens (Cladonia ciliata var. ciliata, Cladonia ciliata var. tenuis and Cladonia portentosa ssp. pacifica), which occur at three of the five prairies we surveyed. Our results indicate that some lichen populations could be extirpated from these sites if they are not considered in prescribed burn management plans. We suggest some options that could maintain ground layer integrity while balancing other management objectives in the south Puget Sound prairies.

Recent declines of anadromous salmonids in the Pacific Northwest have prompted a need to understand the factors limiting populations across habitats and life stages. In the weeks following emergence from the spawning gravel, juvenile salmonids have limited swimming capacity and can be particularly vulnerable to predation by piscivores, such as sculpins (Cottus sp.) which are widespread in Pacific Northwest streams. The objective of this study was to investigate the extent to which sculpin prey on newly emerged steelhead (Oncorhynchus mykiss) in a threatened population in Idaho. Three species of sculpin were present in the watershed, including Paiute (C. beldingii), mottled (C. bairdii), and torrent (C. rhotheus) sculpin. Gut content analyses of 360 sculpin showed that invertebrates were the dominant food source (> 85% of all sculpin preyed mainly on invertebrates), and piscivory was rare (< 2%). None of the samples contained steelhead alevins. We conclude that there was no indication of predation during emergence in our study system, but note that future studies should incorporate stable isotope analyses or directly investigate the extent to which sculpin prey on salmonid eggs.

The federally endangered Taylor's checkerspot (Euphydryas editha taylori) is an increasingly rare prairie butterfly in the Willamette Valley-Puget Trough-Georgia Basin ecoregional complex. Since the arrival of European settlers, several factors have reduced available native host plants for E. e. taylori larvae. The most common host is now Plantago lanceolata, an exotic species long prevalent in the area. Local biologists have observed that the known native hosts have low available biomass relative to P. lanceolata, and may senesce too early to support the animals throughout their entire larval feeding cycle. Federally threatened Castilleja levisecta, which often grows larger and persists longer than currently-utilized native hosts, may have been important historically but its range does not now overlap with E. e. taylori. Previous work with other E. editha ssp. has shown that oviposition preference is: 1) heritable and may provide clues to which hosts were used historically, and 2) has been correlated with larval success so might indicate which hosts would be most effective for population restoration. We undertook an oviposition preference experiment to determine which potential hosts were preferred by E. e. taylori among P. lanceolata, C. levisecta, and the more common congener, C. hispida. The two Castilleja spp. were preferred equally, but C. levisecta was preferred over P. lanceolata. If further research confirms the suitability of C. levisecta as a host for E. e. taylori, restoration efforts for the two species could be united, and the effectiveness of both might be synergistically increased.