USGS Science⌕ Search

SEARCH · USGS Science

Results for “CHANCE”

Search indexed USGS publications on groundwater, aquifers, geologic maps, mineral resources and earthquakes. Explore source records by subject and place.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3Linked to original sources
USGS1988

Guidance for modeling causes and effects in environmental problem solving

Environmental problems are difficult to solve because their causes and effects are not easily understood. When attempts are made to analyze causes and effects, the principal challenge is organization of information into a framework that is logical, technically defensible, and easy to understand and communicate. When decisionmakers attempt to solve complex problems before an adequate cause and effect analysis is performed there are serious risks. These risks include: greater reliance on subjective reasoning, lessened chance for scoping an effective problem solving approach, impaired recognition of the need for supplemental information to attain understanding, increased chance for making unsound decisions, and lessened chance for gaining approval and financial support for a program/ Cause and effect relationships can be modeled. This type of modeling has been applied to various environmental problems, including cumulative impact assessment (Dames and Moore 1981; Meehan and Weber 1985; Williamson et al. 1987; Raley et al. 1988) and evaluation of effects of quarrying (Sheate 1986). This guidance for field users was written because of the current interest in documenting cause-effect logic as a part of ecological problem solving. Principal literature sources relating to the modeling approach are: Riggs and Inouye (1975a, b), Erickson (1981), and United States Office of Personnel Management (1986).

Report↗
USGS2015

Regional implications of new chronostratigraphic and paleogeographic data from the Early Permian Darwin Basin, east-central California

The Darwin Basin developed in response to episodic subsidence of the western margin of the Cordilleran continental shelf from Late Pennsylvanian (Gzhelian) to Early Permian (late Artinskian) time. Subsidence of the basin was initiated in response to continental truncation farther to the west and was later augmented by thrust emplacement of the Last Chance allochthon. This deep-water basin was filled by voluminous fine-grained siliciclastic turbidites and coarse-grained limestone-gravity-flow deposits. Most of this sediment was derived from the Bird Spring carbonate shelf and cratonal platform to the northeast or east, but some came from an offshore tectonic ridge (Conglomerate Mesa Uplift) to the west that formed at the toe of the Last Chance allochthon. At one point in the late Artinskian the influx of extrabasinal sediment was temporarily cut off, resulting in deposition of a unique black limestone that allows precise correlation throughout the basin. Deep-water sedimentation in the Darwin Basin ended by Kungurian time when complex shallow-water to continental sedimentary facies spread across the region. Major expansion of the Darwin Basin occurred soon after the middle Sakmarian emplacement of the Last Chance allochthon. This tectonic event was approximately coeval with deformation in northeastern Nevada that formed the deep-water Dry Mountain Trough. We herein interpret the two basins to have been structurally continuous. Deposition of the unique black limestone is interpreted to mark a eustatic sea level rise that also can be recognized in Lower Permian sections in east-central Nevada and central Arizona.

California↗
USGS2016

Using maximum entropy to predict suitable habitat for the endangered dwarf wedgemussel in the Maryland Coastal Plain

Species distribution modelling can be useful for the conservation of rare and endangered species. Freshwater mussel declines have thinned species ranges producing spatially fragmented distributions across large areas. Spatial fragmentation in combination with a complex life history and heterogeneous environment makes predictive modelling difficult. A machine learning approach (maximum entropy) was used to model occurrences and suitable habitat for the federally endangered dwarf wedgemussel, Alasmidonta heterodon , in Maryland's Coastal Plain catchments. Landscape-scale predictors (e.g. land cover, land use, soil characteristics, geology, flow characteristics, and climate) were used to predict the suitability of individual stream segments for A. heterodon . The best model contained variables at three scales: minimum elevation (segment scale), percentage Tertiary deposits, low intensity development, and woody wetlands (sub-catchment), and percentage low intensity development, pasture/hay agriculture, and average depth to the water table (catchment). Despite a very small sample size owing to the rarity of A. heterodon , cross-validated prediction accuracy was 91%. Most predicted suitable segments occur in catchments not known to contain A. heterodon , which provides opportunities for new discoveries or population restoration. These model predictions can guide surveys toward the streams with the best chance of containing the species or, alternatively, away from those streams with little chance of containing A. heterodon . Developed reaches had low predicted suitability for A. heterodon in the Coastal Plain. Urban and exurban sprawl continues to modify stream ecosystems in the region, underscoring the need to preserve existing populations and to discover and protect new populations.

Maryland↗
USGS2019

Predicting persistence of Rio Grande Cutthroat Trout populations in an uncertain future

The Rio Grande Cutthroat Trout Oncorhynchus clarkii virginalis (RGCT ) occupies just 12% of its ancestral range. As the southernmost subspecies of Cutthroat Trout, we expect a warming climate to bring additional stressors to RGCT populations, such as increased stream temperatures, reduced streamflows, and increased incidence of wildfire. We developed a Bayesian network (BN ) model using site‐specific data, empirical research, and expert knowledge to estimate the probability of persistence for each of the 121 remaining RGCT conservation populations and to rank the severity of the threats they face. These inputs quantified the genetic risks (e.g., inbreeding risk and hybridization risk), population demographics (disease risk, habitat suitability, and survival), and probability of stochastic disturbances (stream drying risk and wildfire risk) in an uncertain future. We also created stream temperature and base flow discharge models coupled with regionally downscaled climate projections to predict future abiotic conditions at short‐term (2040s) and long‐term (2080s) time horizons. In the absence of active management, we predicted a decrease in the average probability of population persistence from 0.53 (current) to 0.31 (2040s) and 0.26 (2080s). Only 11% of these populations were predicted to have a greater than 75% chance of persisting to the 2080s. Threat of invasion by nonnative trout had the strongest effect on population persistence. Of the 78 populations that are already invaded or lacking complete barriers, 60% were estimated to be extirpated by 2080 and the remainder averaged only a 10% chance of persistence. In contrast, the effects of increased stream temperatures were predicted to affect the future persistence of only 9% of the 121 RGCT populations remaining, as most have been restricted to high‐elevation habitats that are cold enough to buffer against some stream warming. Our BN model provides a framework for evaluating threats and will be useful to guide management actions that are likely to provide the most benefit for long‐term conservation.

New Mexico, Colorado↗
USGS2021

Habitat associations of black basses in a reservoir system

The habitat associations of three species of black bass Micropterus spp. were examined in six habitat types (i.e., sediment, gravel, rock, riprap, brush, and aquatic plants) along a cascade of 10 reservoirs in the Tennessee River. We tested whether habitat selection differed among the three species and whether species’ co-occurrence depended on habitat type. We found that some species occurred in some habitats in proportion to habitat availability (some at higher frequencies and some at frequencies lower than availability) and that juveniles and adults exhibited similar occurrence patterns. Our habitat selection results largely corroborate previous descriptions of black bass habitat associations and generally track preference for lithic habitats, as reported in native streams. We expected the different black bass species to show negative co-occurrence to avoid competitive interactions. Nevertheless, we found that with few exceptions, adults co-occurred in habitats mostly as expected by chance and juveniles co-occurred more often than was expected by chance. Our findings imply that environmental filtering, rather than competitive interactions that dominate in natural environments, may be the dominant mechanism shaping black bass assemblages in reservoirs of the Tennessee River. The observed patterns of habitat selection and co-occurrence further suggest that the conservation and management of black bass assemblages in reservoirs can be supported through habitat management activities. Protecting and enhancing the remaining lithic habitat in the reservoirs as well as recovering habitat that is blanketed by sediment could provide desirable environments for all black bass species.

Tennessee River basin↗
USGS2018

A tool for efficient, model-independent management optimization under uncertainty

To fill a need for risk-based environmental management optimization, we have developed PESTPP-OPT, a model-independent tool for resource management optimization under uncertainty. PESTPP-OPT solves a sequential linear programming (SLP) problem and also implements (optional) efficient, “on-the-fly” (without user intervention) first-order, second-moment (FOSM) uncertainty techniques to estimate model-derived constraint uncertainty. Combined with a user-specified risk value, the constraint uncertainty estimates are used to form chance-constraints for the SLP solution process, so that any optimal solution includes contributions from model input and observation uncertainty. In this way, a “single answer” that includes uncertainty is yielded from the modeling analysis. PESTPP-OPT uses the familiar PEST/PEST++ model interface protocols, which makes it widely applicable to many modeling analyses. The use of PESTPP-OPT is demonstrated with a synthetic, integrated surface-water/groundwater model. The function and implications of chance constraints for this synthetic model are discussed.

Environmental Modelling and Software↗
USGS2004

Seismic hazard maps of Mexico, the Caribbean, and Central and South America

The growth of megacities in seismically active regions around the world often includes the construction of seismically unsafe buildings and infrastructures due to an insufficient knowledge of existing seismic hazard and/or economic constraints. Minimization of the loss of life, property damage, and social and economic disruption due to earthquakes depends on reliable estimates of seismic hazard. We have produced a suite of seismic hazard estimates for Mexico, the Caribbean, and Central and South America. One of the preliminary maps in this suite served as the basis for the Caribbean and Central and South America portion of the Global Seismic Hazard Map (GSHM) published in 1999, which depicted peak ground acceleration (pga) with a 10% chance of exceedance in 50 years for rock sites. Herein we present maps depicting pga and 0.2 and 1.0 s spectral accelerations (SA) with 50%, 10%, and 2% chances of exceedance in 50 years for rock sites. The seismicity catalog used in the generation of these maps adds 3 more years of data to those used to calculate the GSH Map. Different attenuation functions (consistent with those used to calculate the U.S. and Canadian maps) were used as well. These nine maps are designed to assist in global risk mitigation by providing a general seismic hazard framework and serving as a resource for any national or regional agency to help focus further detailed studies required for regional/local needs. The largest seismic hazard values in Mexico, the Caribbean, and Central and South America generally occur in areas that have been, or are likely to be, the sites of the largest plate boundary earthquakes. High hazard values occur in areas where shallow-to-intermediate seismicity occurs frequently. ?? 2004 Elsevier B.V. All rights reserved.

Tectonophysics↗
USGS2010

The earthquake prediction experiment at Parkfield, California

Since 1985, a focused earthquake prediction experiment has been in progress along the San Andreas fault near the town of Parkfield in central California. Parkfield has experienced six moderate earthquakes since 1857 at average intervals of 22 years, the most recent a magnitude 6 event in 1966. The probability of another moderate earthquake soon appears high, but studies assigning it a 95% chance of occurring before 1993 now appear to have been over-simplified. The identification of a Parkfield fault “segment” was initially based on geometric features in the surface trace of the San Andreas fault, but more recent microearthquake studies have demonstrated that those features do not extend to seismogenic depths. On the other hand, geodetic measurements are consistent with the existence of a “locked” patch on the fault beneath Parkfield that has presently accumulated a slip deficit equal to the slip in the 1966 earthquake. A magnitude 4.7 earthquake in October 1992 brought the Parkfield experiment to its highest level of alert, with a 72-hour public warning that there was a 37% chance of a magnitude 6 event. However, this warning proved to be a false alarm. Most data collected at Parkfield indicate that strain is accumulating at a constant rate on this part of the San Andreas fault, but some interesting departures from this behavior have been recorded. Here we outline the scientific arguments bearing on when the next Parkfield earthquake is likely to occur and summarize geophysical observations to date.

California↗
USGS2002

Puerto Rican parrots and potential limitations of the metapopulation approach to species conservation

Population viability analyses for a number of endangered species have incorporated a metapopulation approach. The risk assessments of these viability analyses have indicated that some extant populations should be subdivided into numerous subgroups with exchange of individuals among them in order to reduce the chance of catastrophic loss of the species. However, routine application of a policy of extensive subdivision may have detrimental consequences for certain endangered species. We examine the Puerto Rican Parrot as a case history in which this policy is ill-advised. In 1989, a population viability analysis was conducted for the parrot. The document recommended subdivision of the existing small captive flock into three groups. One of these captive flocks would consist of individuals transferred to a multi-species facility in the continental United States. Subsequently, individuals from this facility would be exchanged with the insular captive population(s) and the relict wild flock. For two reasons, implementation of this recommendation might have led to serious repercussions. First, this parrot, like many endangered species, has gone through a genetic bottleneck and may have a heightened susceptibility to disease. Multi-species facilities are a high-risk environment favoring the transmission of pathogens, especially when the facilities are located outside the natural ranges of a particular species. Second, the parrot is a K-selected species for which mate selection is idiosyncratic. This type of species often proves difficult to breed in captivity in small groups. Part of the problem in mate selection may be reduced by a policy allowing frequent transfers of individuals among facilities, but such movements increase the chances of spreading disease in the metapopulation. Thus, population viability analyses need to acknowledge that proliferation of captive subgroups accompanied by exchanges of individuals can in themselves carry substantial risks that must be weighed against the presumed benefits of subdivision.

Puerto Rico↗
USGS2019

Understanding and mitigating bee drownings in open feeders

Whereas open feeders are relatively inexpensive and are easily set up and maintained, they do present a drowning hazard to bees. We observed that bees feeding off the walls of the open container fell into the sugar water because of their incarnation, interactions with other bees and when shooed off the feeder walls while removing the feeder for cleaning. Twigs, angled laths and utility screen perches permitted bees to exploit more of the sugar water surface area and provided drowning bees a platform for self-rescue. Because angled laths and utility screen perches extended over the entire feeder, they offered greater feeding surface area and increased the chances that a drowning bee would quickly encounter the perches for self-rescue than twig perches. Additionally, bees were less likely to fall into the sugar water when removing the angled lath and utility screen perches from the feeders than when removing twigs. Our anecdotal observations identified three characteristics of perches that can mitigate for the drowning hazard. Perches need to: 1. Allow bees to use a greater surface area of the sugar water to reduce crowding while feeding from the container walls. 2. Encompass most of the feeder to improve the chances that drowning bees will encounter the perch and be able to rescue themselves. 3. Allow bees to quickly extricate themselves from the sugar water to minimize sugar crystallization on the bees. The information presented herein provides a practical window into the factors that lead to bee drownings and the type of mitigation that is required. It is of note that we used the three perch types to develop perch characteristics for mitigating drowning hazards. Novice researchers can apply these principles to customize perches based on their specific needs.

Bee World↗
USGS2026

Compounding of 100-year coastal floods by rainfall in an urban environment

Coastal and pluvial flooding are both becoming more prevalent and severe due to climate change and urbanization in floodplains. The co-occurrence of these flood drivers is generally assumed to exacerbate the resulting flood impacts, a result referred to as compound flooding. However, few observational or modeling studies have investigated the circumstances under which this occurs. Here, we study the impacts of these combined flood drivers and evaluate the implicit hypothesis of official flood maps, which is that rainfall has a negligible impact on the flood depth and flooded area due to a 100 year coastal flood. A coastal system model, configured to capture coastal and pluvial flood drivers, is used. We evaluate the flooding for different urban landform types, including coastal landfill (human-made land), convergent areas (topographic depressions) and other urban terrain, within a model domain covering the Jamaica Bay watershed of New York City. A scenario-based strategy is adopted with a 100 year coastal flood as a control simulation, to which we add a set of realistic scenarios of rainfall data from historical tropical cyclones. We also apply a joint probability analysis framework with historical data to evaluate the probability of these compound coastal-pluvial scenarios. Results reveal cases where the pluvial driver compounds the coastal flood through expansion of the flood zone, with a 17% chance of rainfall increasing the flood area by 6%–38%, and a 5% chance of an increase of 61%–73%. It is rare that floods are significantly deepened but when deepening occurs, it is more common for the convergent zone than for the coastal landfill. These findings quantitatively assess the potential of the pluvial driver to exacerbate flooding, which may influence emergency management strategies such as evacuation plans, shelter arrangements, and related preparedness measures.

New York↗
USGS2020

Panmixia in a sea ice-associated marine mammal: evaluating genetic structure of the Pacific walrus (Odobenus rosmarus divergens) at multiple spatial scales

The kin structure of a species at relatively fine spatial scales impacts broad-scale patterns in genetic structure at the population level. However, kin structure rarely has been elucidated for migratory marine mammals. The Pacific walrus ( Odobenus rosmarus divergens ) exhibits migratory behavior linked to seasonal patterns in sea ice dynamics. Consequently, information on the spatial genetic structure of the subspecies, including kin structure, could aid wildlife managers in designing future studies to evaluate the impacts of sea ice loss on the subspecies. We sampled 8,303 individual walruses over a 5-year period and used 114 single-nucleotide polymorphisms to examine both broad-scale patterns in genetic structure and fine-scale patterns in relatedness. We did not detect any evidence of genetic structure at broad spatial scales, with low F ST values (≤ 0.001) across all pairs of putative aggregations. To evaluate kin structure at fine spatial scales, we defined a walrus group as a cluster of resting individuals that were less than one walrus body length apart. We found weak evidence of kin structure at fine spatial scales, with 3.72% of groups exhibiting mean relatedness values greater than expected by chance, and a significantly higher overall observed mean value of relatedness within groups than expected by chance. Thus, the high spatiotemporal variation in the distribution of resources in the Pacific Arctic environment likely has favored a gregarious social system in Pacific walruses, with unrelated animals forming temporary associations.

Bering Sea↗
USGS2024

Duckling survival increased with availability of flooded wetland habitat and decreased with salinity concentrations in a brackish marsh

Waterfowl population recruitment is sensitive to duckling survival. We quantified predator types and survival rates for Anas platyrhynchos (Mallard) and Mareca strepera (Gadwall) ducklings in one of the largest brackish water marshes in western North America (Suisun Marsh, California) using 556 radio-tagged ducklings from 284 broods tracked during the 2016 to 2019 breeding seasons. Overall, 78% of ducklings died and 84% of mortalities occurred < 7 days after hatch. After hatching in upland fields, survival was greater for broods that hatched closer to flooded wetlands; broods had a ≥ 75% chance of surviving the move from the nest to water when nests were located ≤ 140 m from the nearest wetland and ≤ 50% chance of surviving when nests were located ≥ 970 m from the nearest wetland. Predation accounted for 91% of mortalities and was attributed to mammals (27.6%), birds (22.0%), snakes (4.4%), and unknown predators (46.0%). Anas platyrhynchos survival to fledging (54 days) was only 3.2% and 0.9% during 2 drier years and 11.7% and 16.7% during 2 wetter years. Mareca strepera survival to fledging was 9.4% to 11.2% among years. Daily survival rates for ducklings generally increased with the amount of flooded wetlands within 0.5 km ( A. platyrhynchos ) and 1.0 km ( M. strepera ) of the nest at hatch. Additionally, survival rates increased with duckling age and body mass at hatch for both species and decreased with hatch date for A. platyrhynchos but not M. strepera , which may be partially due to the earlier onset of A. platyrhynchos nesting. For ducklings that survived the initial move to water, survival rates were negatively correlated with salinity and this effect was more pronounced for younger ducklings. Anas platyrhynchos survival to 7 days post hatch decreased by 9.1% (wetter year) to 31.4% (drier year) when ducklings were in 12 ppt water (99th quantile of cumulative salinity concentrations experienced by ducklings) versus 0.5 ppt water. Mareca strepera survival to 7 days decreased by 7.4% when ducklings were in 12 ppt vs. 0.5 ppt water. Our results suggest that maintaining a network of low salinity wetlands within 1 km of upland nesting sites would likely improve duckling survival rates, especially during the critical 7-day period after hatch.

California↗
USGS2011

Quantifying the hydrological responses to climate change in an intact forested small watershed in southern China

Responses of hydrological processes to climate change are key components in the Intergovernmental Panel for Climate Change (IPCC) assessment. Understanding these responses is critical for developing appropriate mitigation and adaptation strategies for sustainable water resources management and protection of public safety. However, these responses are not well understood and little long-term evidence exists. Herein, we show how climate change, specifically increased air temperature and storm intensity, can affect soil moisture dynamics and hydrological variables based on both long-term observation and model simulations using the Soil and Water Assessment Tool (SWAT) in an intact forested watershed (the Dinghushan Biosphere Reserve) in Southern China. Our results show that, although total annual precipitation changed little from 1950 to 2009, soil moisture decreased significantly. A significant decline was also found in the monthly 7-day low flow from 2000 to 2009. However, the maximum daily streamflow in the wet season and unconfined groundwater tables have significantly increased during the same 10-year period. The significant decreasing trends on soil moisture and low flow variables suggest that the study watershed is moving towards drought-like condition. Our analysis indicates that the intensification of rainfall storms and the increasing number of annual no-rain days were responsible for the increasing chance of both droughts and floods. We conclude that climate change has indeed induced more extreme hydrological events (e.g. droughts and floods) in this watershed and perhaps other areas of Southern China. This study also demonstrated usefulness of our research methodology and its possible applications on quantifying the impacts of climate change on hydrology in any other watersheds where long-term data are available and human disturbance is negligible.

Guangdong Province↗
USGS2017

Quantifying long-term population growth rates of threatened bull trout: challenges, lessons learned, and opportunities

Temporal symmetry models (TSM) represent advances in the analytical application of mark–recapture data to population status assessments. For a population of char, we employed 10 years of active and passive mark–recapture data to quantify population growth rates using different data sources and analytical approaches. Estimates of adult population growth rate were 1.01 (95% confidence interval = 0.84–1.20) using a temporal symmetry model ( λ TSM ), 0.96 (0.68–1.34) based on logistic regressions of annual snorkel data ( λ A ), and 0.92 (0.77–1.11) from redd counts ( λ R ). Top-performing TSMs included an increasing time trend in recruitment ( f ) and changes in capture probability ( p ). There was only a 1% chance the population decreased ≥50%, and a 10% chance it decreased ≥30% ( λ MCMC ; based on Bayesian Markov chain Monte Carlo procedure). Size structure was stable; however, the adult population was dominated by small adults, and over the study period there was a decline in the contribution of large adults to total biomass. Juvenile condition decreased with increasing adult densities. Utilization of these different information sources provided a robust weight-of-evidence approach to identifying population status and potential mechanisms driving changes in population growth rates.

Oregon↗
USGS2011

Oil and gas resource potential north of the Arctic Circle

The US Geological Survey recently assessed the potential for undiscovered conventional petroleum in the Arctic. Using a new map compilation of sedimentary elements, the area north of the Arctic Circle was subdivided into 70 assessment units, 48 of which were quantitatively assessed. The Circum-Arctic Resource Appraisal (CARA) was a geologically based, probabilistic study that relied mainly on burial history analysis and analogue modelling to estimate sizes and numbers of undiscovered oil and gas accumulations. The results of the CARA suggest the Arctic is gas-prone with an estimated 770–2990 trillion cubic feet of undiscovered conventional natural gas, most of which is in Russian territory. On an energy-equivalent basis, the quantity of natural gas is more than three times the quantity of oil and the largest undiscovered gas field is expected to be about 10 times the size of the largest undiscovered oil field. In addition to gas, the gas accumulations may contain an estimated 39 billion barrels of liquids. The South Kara Sea is the most prospective gas assessment unit, but giant gas fields containing more than 6 trillion cubic feet of recoverable gas are possible at a 50% chance in 10 assessment units. Sixty per cent of the estimated undiscovered oil resource is in just six assessment units, of which the Alaska Platform, with 31% of the resource, is the most prospective. Overall, the Arctic is estimated to contain between 44 and 157 billion barrels of recoverable oil. Billion barrel oil fields are possible at a 50% chance in seven assessment units. Undiscovered oil resources could be significant to the Arctic nations, but are probably not sufficient to shift the world oil balance away from the Middle East.

Geological Society Memoir↗
USGS2020

Assessing the value of removing earthquake-hazard-related epistemic uncertainties, exemplified using average annual loss in California

To aid in setting scientific research priorities, we assess the potential value of removing each of the epistemic uncertainties currently represented in the US Geological Survey California seismic-hazard model, using average annual loss (AAL) as the risk metric of interest. Given all the uncertainties, represented with logic-tree branches, we find a mean AAL of $3.94 billion. The modal value is 17.5% lower than the mean, and there is a 78% chance that the true AAL value is more than 10% away from the mean, and a 5% chance that it is a factor 2.1 greater or lower than the mean. We quantify the extent to which resolving each uncertainty improves the AAL estimate. The most influential branch is one that adds additional epistemic uncertainty to ground motion models, but others are found to be influential as well, such as the rate of M ≥ 5 events throughout the region. We discuss the broader implications of our findings, and note that the time dependence caused by spatiotemporal clustering can be much more influential on AAL than the epistemic uncertainties explored here.

California↗
USGS2021

Habitat selection by wolves and mountain lions during summer in western Montana

In the Northern Rockies of the United States, predators like wolves ( Canis lupus ) and mountain lions ( Puma concolor ) have been implicated in fluctuations or declines in populations of game species like elk ( Cervus canadensis ) and mule deer ( Odocoileus hemionus ). In particular, local distributions of these predators may affect ungulate behavior, use of space, and dynamics. Our goal was to develop generalizable predictions of habitat selection by wolves and mountain lions across western Montana. We hypothesized both predator species would select habitat that maximized their chances of encountering and killing ungulates and that minimized their chances of encountering humans. We assessed habitat selection by these predators during summer using within-home range (3 rd order) resource selection functions (RSFs) in multiple study areas throughout western Montana, and tested how generalizable RSF predictions were by applying them to out-of-sample telemetry data from separate study areas. Selection for vegetation cover-types varied substantially among wolves in different study areas. Nonetheless, our predictions of 3 rd order selection by wolves were highly generalizable across different study areas. Wolves consistently selected simple topography where ungulate prey may be more susceptible to their cursorial hunting mode. Topographic features may serve as better proxies of predation risk by wolves than vegetation cover-types. Predictions of mountain lion distribution were less generalizable. Use of rugged terrain by mountain lions varied across ecosystem-types, likely because mountain lions targeted the habitats of different prey species in each study area. Our findings suggest that features that facilitate the hunting mode of a predator (i.e. simple topography for cursorial predators and hiding cover for stalking predators) may be more generalizable predictors of their habitat selection than features associated with local prey densities.

Montana↗