Progress report for SW22-934
Project Information
Wildlife-livestock conflicts, such as depredation by predators, challenge the livelihoods of livestock producers (hereafter, ranchers). Protecting livestock from predators is a complex endeavor, and successful predator conflict mitigation practices require both an analysis of the efficacy of various practices and collaborative information sharing across invested stakeholders. Ranchers typically use an integrated management approach—primarily focusing on animal husbandry and ranch management, while also deploying mitigation practices to reduce depredation risks and lethal techniques when mitigation practices fail, and lethal control is authorized. Mitigation practices include human presence (e.g., range riders), deterrents (e.g., fladry), livestock management, and habitat manipulation, but there is limited scientific information on which practices are most effective and under what scenarios they succeed or fail. Ranchers also lack adequate resources to apply mitigation practices or share knowledge gained through experience. Through a diverse partnership of ranchers, scientists, conservation groups, and agencies with decades of experience with landowner collaborative strategies and predator conflict mitigation practices, we evaluated the effectiveness of range riding across western landscapes with grizzly bears and wolves, hosted opportunities for ranchers to exchange information about mitigation practices, and will disseminate information from our research and exchanges via scientific papers, extension articles, and traditional and novel education and outreach programming. We therefore pursued a co-production approach in what is, to our knowledge, the most comprehensive project of its kind to date.
We focused on range riding because this practice is of high utility to ranchers, yet riding strategies vary widely. How and what works best is unclear, inhibiting adoption by more ranchers. Our research described and quantified the types and efficacy of range riding strategies, while rancher-to-rancher exchanges provided opportunities for foundational knowledge from years of rancher experience to be distributed more broadly. Published products will add scientific credibility to the findings and reinforce learning from exchanges through multiple and highly accessible outlets. We provided diverse outreach venues to facilitate new and improved use of predator conflict mitigation practices and add new users; we anticipated ranchers already using these practices would fine-tune their application for increased efficacy, while others would incorporate these practices into their management for the first time. Results of our study helped to create adaptive and integrative predator conflict mitigation practices disseminated to 600+ ranchers across 7+ states.
We used an iterative process to ensure successful implementation to improve sustainable agricultural practices. Information was continuously communicated among team members about the research, outreach, and rancher-to-rancher exchanges, which facilitated incremental changes in co-production processes and in how the practices are implemented by ranchers. This project has and will continue to create transformative change in agricultural sustainability by supporting a community of practice to research range riding across diverse social and ecological contexts in different grazing scenarios. We expect this framework to be applied to other mitigation practices and develop mechanisms for sustainable funding of such practices through NRCS-administered Farm Bill programs. Because we continuously coordinated with NRCS personnel and ranchers about how our research can be applied to decision-making surrounding funding for ranchers, our work helped to establish best practices for predator conflict mitigation that significantly improve sustainable agricultural production through incentivizing the adoption of proactive strategies. Importantly, the work resulted in $22 million for the implementation of the practices.
While some analyses for the research components are still ongoing, so far our preliminary work has yielded these key takeaways:
- Cattle may be adapting their use of high forage availability areas in response to predator activity as a means to reduce predation risk. Whether this stress response is significant enough to impact livestock economics (e.g., reduced weaning weights, reduced cow body condition score, reduced pregnancy rates) will require more research.
- Riders may not meaningfully reduce predator-induced stress and/or pressure in cattle to improve the use of high availability foraging areas, but finer scale analyses (spatiotemporally) may better assess livestock response to both rider and predator activity.
- Rider deployment strategies should remain context-specific and adaptive rather than prescriptive.
- Support the conflict mitigation approaches livestock producers advocate for - cultural and operational context, as well as uncertainty reduction, adaptive management, data collection, and improved coordination/communication are equally important effectiveness metrics for nonlethal tool adoption as effectiveness in reducing direct loss.
For the education and outreach portion, we have learned the following lessons:
- Building communities of practice amongst diverse stakeholders in conflict reduction can help support information exchange relevant to practice adoption for a host of practices, including range riding
- Engagement of broad networks with effective, science and land-stewards centered communications can support increased knowledge of range riding and its application, and support cross-pollination of ideas within networks, building momentum for practice implementation.
- Landowners and livestock producers maintain knowledge of the land and stewardship practices that are not often captured in scientific research, or elevated for peer-to-peer learning. Incorporating this knowledge is both important for representative applied science, and for diffusion and implementation of practices such as range riding.
- Improve ranch profitability through range riding, a predator conflict mitigation practice that is highly adaptable across diverse ranching operations.
- Coproduce research to evaluate the cost and effectiveness of different range riding strategies across at least seven western states.
- Incorporate data collected by researchers, ranchers, and local landowner groups to accomplish a robust evaluation of conflict reduction strategies.
- Expand and integrate effective range riding strategies with adaptive conflict mitigation programs through rancher-to-rancher knowledge exchanges to support an enhanced quality of life for ranchers, livestock, and wildlife.
- Elevate conservation planning and natural resources management within predator-occupied regions through co-interpretation and dissemination of project results on range riding and other conflict mitigation practices.
- Synthesize research using metrics relevant to livestock production to indicate the value of range riding and other conflict mitigation practices to ranchers.
- Provide data to NRCS to inform the development of new or modified conservation practices to incentivize broad adoption of conflict mitigation techniques.
- Disseminate and amplify the collective experience and knowledge gained through this project by providing highly relevant content through a combination of traditional outreach programming (workshops, seminars) and novel outreach products, including audio, print, and digital platforms.
Cooperators
- - Technical Advisor
- - Producer
- - Producer
- - Producer
Research
The goal of the project was to reduce the financial and social burden of expanding predator populations through the evaluation of range riding practices and information sharing among ranchers about their experience with all predator conflict mitigation practices, leading to more resilient ranches and connected landscapes.
We accomplished the following goals through four project objectives:
- Improve ranch profitability through range riding, a predator conflict mitigation practice that is highly adaptable across diverse ranching operations.
- Expand and integrate effective range riding strategies with adaptive conflict mitigation programs through rancher-to-rancher knowledge exchanges to support an enhanced quality of life for ranchers, livestock, and wildlife.
- Elevate conservation planning and natural resources management within predator-occupied regions through co-interpretation and dissemination of project results on range riding and other conflict mitigation practices; and
- Disseminate and amplify the collective experience and knowledge gained through this project by providing highly relevant content through a combination of traditional outreach programming (workshops, seminars) and novel outreach products, including audio, print, and digital platforms.
Objective 1 focuses on research, while objectives 2-4 are primarily concerned with education and outreach, so they are described in detail in the educational section below. However, all the results and information we learned from reaching our research objectives were applied to the educational objectives. Through objective 3, we worked to ensure the information was accurate and readily available to incorporate into objectives 2 and 4 using an iterative process. As we gained information from objective 1, we tried to incorporate it into the outreach materials, disseminate it at workshops, and discuss it at rancher-to-rancher exchanges. We used feedback from ranchers and other stakeholders about the clarity of information, the relevancy of these findings to their practices, and what other information was needed.
There were several steps to accomplishing our research objective. First, we met with individual ranchers and landowner groups. As part of our existing Conflict on Workinglands Conservation Innovation Grant (CoW-CIG), we met frequently with ranchers and other stakeholders from across the West to understand what value riding provided to them, how best to measure that value, and what methods would be useful and feasible to measure that value. We examined the influence of varied rider activity on 1) annual conflict (combined predator induced injuries and depredation on cattle), 2) annual herd return rates, 3) annual herd pregnancy rates, and 4) chemical indicators of stress in cattle herds (cortisol and thyroid function) in an effort to improve rancher profitability through reducing conflict, enhancing stewardship through an improved understanding of how a rider can reduce cattle stress that leads to indirect losses, and improving overall quality of life for ranchers and their communities by helping to empower operational decision-making. Funding from the Montana Western SARE allowed us to build on that research with the following objectives:
Objective 1. Research Objectives (1-2) and Hypotheses (a. - cvi.)
- Identify whether varied range riding activity alters behavioral indicators of stress in cattle.
As rider intensity, time spent within proximity of the herd, and time riding at dusk, dawn, or at night increase,
a. Seasonal herd vigilance will decrease, and
b. Cattle use of high availability foraging areas will increase.
Although not a deliverable for this grant, the larger Conflict on Workinglands Conservation Innovation Grant (CoW-CIG) also examined the following;
As rider intensity, time spent within proximity of the herd, and time riding at dusk, dawn, or at night increase,
ci. direct losses (depredations and injuries) will decrease,
cii. return rates will increase,
ciii. pregnancy rates will increase,
civ. average herd cortisol levels will decrease,
cv. average herd thyroid levels will increase, and
cvi. average herd body condition scores (cows) will increase.
While we will not be reporting results for all of these hypotheses, we list them here to provide context for data collected, and will share preliminary findings on our cortisol and thyroid analyses (Research Objectives 1civ. and 1cv.).
2. Conduct unstructured interviews with livestock producers to capture the unique operational, environmental, and economic context driving decision making regarding range riding, and to provide important context to our qualitative findings.
Research Objective 1 - Study Area and Data Sources
Study Area
To examine the influence of varied rider activity on cattle behavior, we collected several data streams on rider activity, habitat features and forage availability, predator spatial use, and cattle behavioral activity across multiple operations in Montana, Washington, Oregon, Arizona, and New Mexico from 2022-2024 (Figure 1). In total, we had eight operations participating in 2022, 12 in 2023, and 13 in 2024. All Montana operations are in areas with wolf and grizzly bear populations, whereas operations in the other states were only in areas with wolf populations. Field seasons started in January and wrapped up by early December, depending on the operation location.
Data Sources
Range Rider Data
In grazing seasons 2022, 2023, and 2024, participating range riders completed rider data sheets. We encouraged daily data collection, but to accommodate the time constraints of range riders, we also allowed for weekly data collection (Nickerson_Daily_Rider, Nickerson_Weekly_Rider). Several riders in Washington and Montana also recorded their riding tracks with a GPS unit. Range riders were trained each spring on data collection for both GPS units and data sheets that were provided at the start of the season. Range Rider GPS tracks will be used to compare rider landscape use and proximity to cattle location data (recorded by the rider) and carnivore location data when available.
Rider sheets (n = 1505) were collected during the 2022, 2023, and 2024 grazing seasons and were cleaned and coded for:
- Rider frequency: total rides per ranch-season,
- Mean ride duration: ride duration averaged across all sampled rides per ranch-season,
- Mean proportion of ride time at night: proportion of ride duration from local dusk to dawn, averaged across all sampled rides per ranch-season,
- Predator effort: proportion of total recorded rider time devoted to predator-related tasks defined as checking game cameras, locating predators via telemetry, and hazing, divided by the total duration of each sampled ride, averaged across all rides per ranch-season,
- Livestock effort: proportion of total recorded rider time devoted to cattle-related tasks defined as checking on livestock, providing salt/mineral, fixing fence or other infrastructure, and/or moving the herd, divided by the total duration of each sampled ride, averaged across all rides per ranch-season,
- Mode of travel: proportion of total sampled rides per ranch-season for truck, atv/side by side, horse, and foot, and
- Use of dogs: proportion of total sampled rides per ranch-season where dogs were used (yes/no).
Environmental Data
Environmental data was collected for all three grazing seasons through our producer partners and existing open-sourced data:
- Pasture and allotment shapefiles for each operation were provided by the USFS or participating livestock producers.
- To calculate available forage biomass for each herd, we wanted to restrict availability to represent only areas on allotments that cows were likely to graze. We used distance-to-water and slope as restrictive parameters.
- Surface water availability was mapped using the U.S. Geological Survey National Hydrography Dataset (NHD - USGS National Map, 2024). We retained feature classes representing perennial water sources (NHDArea and NHD Waterbody), spring and tank locations (NHDPoint), but excluded flowlines and stream segments because these features frequently represent ephemeral or intermittent drainages that may not be seasonally reliable. Minimum distance to any water source was retained via Euclidean distance rasters generated in ArcGIS Pro version 3.5 (ESRI, Redlands CA). Cattle grazing areas were then classified across all ranches by limiting grazing areas greater than or equal to 1.5 miles (2414 meters) as is commonly cited in the literature (Valentine, 1947; Hart et al., 1993; Bailey, 2005).
- Elevation data were obtained from the LANDFIRE US_220 Elevation dataset at 30-m spatial resolution (LANDFIRE, 2023). The percent slope was then derived for all operations using the slope tool in ArcGIS. Areas with slopes greater than or equal to 30% were classified as unlikely to be utilized by cattle based on the peer-reviewed literature (Bailey, 2005).
The percent of each allotment rendered unavailable by both distance-to-water and slope constraints was low (~7% on one allotment, but an unclassified water tank was likely present, and no more than 4% across all other allotments. The difference in slopes between the restricted raster and non-restricted was only 0.02-1.2%); therefore, forage availability metrics were calculated for all acres of an allotment, and not restricted by distance-to-water or slope.
Available forage biomass (kg/ha represented by 16-day incremental production estimates at 30-m resolution) within pasture boundaries was quantified using remotely sensed vegetation products (herbaceous production rasters) accessed through the Rangelands Analysis Platform (RAP; Allred et al., 2021). For each area of interest, herbaceous production rasters were extracted and averaged to a single AOI-level production value for each month of the grazing season (mean standing herbaceous biomass). We estimated forage availability at three spatiotemporal scales:
-
-
- Mean standing herbaceous biomass per ranch-month was estimated for each month overlapping dates when cattle were present within each operation’s pastures/allotment(s) following rotation,
- Mean standing herbaceous biomass per ranch-season was calculated by aggregating monthly estimates for the entirety of the grazing season, and
- Mean standing herbaceous biomass per camera-month was calculated using a 30-m buffer at the location of each remote game camera, estimating monthly forage availability, then averaging values across all cameras deployed within each camera cluster for a monthly cluster average.
-
3. Openness metrics were derived from the Rangeland Analysis Platform 10-m vegetation cover product and defined as the inverse of woody vegetation cover (100 − percent tree and shrub cover - RAP). We estimated:
-
-
- Allotment-level openness by averaging values across constituent pastures, and
- Game camera–level openness as the mean value within a 30-m buffer surrounding each camera location.
-
4. Proportion of days per ranch-season with stress-producing temperatures for cattle (Nakajima et al., 2019; Kim et al., 2023; Nielsen et al., 2025) were calculated by summing the proportion of days per ranch season exposed to stress- causing heat exposure and stress-causing cold exposure. For heat exposure, we calculated the daily temperature-humidity index (THI) where daily mean air temperature (degrees Celsius) and mean dewpoint temperature (degrees Celsius) were obtained from PRISM daily climate products dataset (PRISM Climate Group, 2024) and summarized across active pastures. We set our threshold for THI at 22.22 degrees Celsius as is commonly used in the cattle literature (West, 2003). For cold exposure, we calculated effective temperature computed from daily minimum temperatures (degrees Celsius) and wind speed (dm/h) via the nClimGrid_Daily dataset (NOAA National Centers for Environmental Information, 2024). We set our threshold for wind equivalent temperature at or below 0 degrees Celsius, as some studies have found even higher temperatures to result in measurable chemical stress responses in cattle (Nakajima et al., 2019; Kim et al., 2023).
Predator Data
We collected data on predator locations using three methods: 1) rider observations recorded via rider data sheets (see attachments), 2) data provided by state, federal, and tribal wildlife agencies through data sharing agreements (both wolf and grizzly bear GPS-collar data and wolf density data), and 3) game cameras deployed on grazing allotments/pastures.
- Rider datasheets were cleaned and species detections were summarized by ranch–year. Formal training in wildlife track and sign identification was not implemented until fall 2024 due to workshop scheduling; consequently, rider observations were excluded from predator activity metrics to minimize observer-related bias.
- GPS-collar data for wolves and grizzly bears provided by state, federal, and tribal agencies were spatially filtered to retain only locations within actively grazed pastures, thereby aligning predator exposure metrics spatially and temporally with areas used by cattle. Filtered locations were aggregated across pastures within each grazing season to produce a ranch–year index of predator spatial use/activity rather than an estimate of abundance (total collar locations).
- Wolf densities were provided by agencies or collected via public annual reports. Density data on grizzly bears were not available.
- We deployed 30 motion-activated cameras per operation in 2-3 clusters consisting of 10–15 cameras each (Figure 2). Grid locations were selected in collaboration with producers and range riders to target areas of consistently high cattle use during active grazing periods. Cameras within each cluster were arranged in a radial configuration centered on focal cattle-use areas (e.g., high-quality pastures, water tanks, salt licks, etc.). Cameras nearest the center were spaced approximately 10–30 m from the center point of interest. Additional cameras were positioned along expanding radii (2-4 radii, depending on local landscape conditions) at approximately 100 m from the center point at the first radii point, 250 at the second, 475 at the third, and 700 at the fourth, 925 at the fifth, and up to 1,150 m at the sixth, depending on the number of radii. This approach increased spatial coverage while reducing the probability of failing to detect predator presence within heavily used areas by cattle. Camera groups were relocated throughout the grazing season to correspond with pasture rotations and maintain alignment with active cattle pastures. Cameras were configured with an 18.29m detection distance and a 30-s delay between triggers to conserve battery life and capacity while reducing repeated captures of the same individuals. Because cameras were intentionally placed in areas of high cattle activity (no grid or random design), predator activity inference is limited to within camera groups rather than landscape-wide.
In 2022, we collected data from 240 game cameras across eight ranches, 390 cameras on 13 ranches in 2023, and 420 cameras across 14 ranches in 2024. Camera trap images were organized by unique camera deployment and processed using SpeciesNet (Google, 2026), designed to combine object detection with the species classifier MegaDetector (Beery et al., 2019). SpeciesNet is well supported in recent peer-reviewed literature (Clarfeld et al., 2025), and version 5a used for our analysis has a reported accuracy of 97-98% (Beery et al., 2019). SpeciesNet was run in batch mode on a GPU-enabled high-performance computing environment to generate detection bounding boxes, class probabilities, and predicted labels for each image. To improve accuracy, we used geographic filters, used human review on images with low confidence assignments, and manually validated a small subset of images across predicted classes.
Herd/Operational Data
Operational-level covariates were collected to inform conflict models, both as response variables and as predictors to disentangle the potential effect of both predator and rider activity. Data were collected during the summer of 2025 in collaboration with participating livestock producers and range riders. The following ranch-year metrics were compiled:
- Producer reported predator-caused conflicts by ranch-year: summed injuries and depredations,
- Wildlife agency confirmed predator-caused conflicts by ranch-year: summed injuries and depredations,
- Producer reported return rate by ranch-year,
- Producer reported average pregnancy rate by ranch-year,
- Producer reported average weaning weights by ranch-year,
- Producer reported herd size by ranch-year (used to calculate herd density using aggregated pasture size in acres),
- Producer reported grazing season length by ranch-year (i.e., total days),
- Whether cattle had access to salt on a regular basis by ranch-year (yes/no),
- Whether cattle had access to mineral on a regular basis by ranch-year (yes/no),
- Whether cattle were vaccinated by ranch-year (yes/no),
- Whether operations use low-stress livestock handling techniques by ranch-year (yes/no),
- Whether an operation used artificial insemination or bulls by ranch-year (AI/bulls),
- Whether an operation used growth hormones in calves by ranch-year (yes/no),
- Whether other nonlethal tools or lethal removal was used during the grazing season by ranch-year (yes/no),
- Average calf age at both allotment turn-out and weaning by ranch-year (in months),
- Herd composition by ranch-year (categorical),
- Herd breed by ranch-year (categorical),
- Producer-reported average herd calving date by ranch-year, and
- Producer-reported average herd weaning date by ranch-year.
Although recorded, we did not find significant variation across ranches regarding access to salt (all regular basis), access to mineral (all regular basis), vaccination protocol (all use), herd breed (all Angus or Angus mix), herd composition (all cow-calf pairs with replacement heifers), or use of low-stress handling techniques (all reported yes). Therefore, these metrics were not used in the models but were standardized across operations.
Cattle Behavioral Data
To explore the influence of varied riding behavior on behavioral indicators of stress in livestock, and the potential influence of behavioral stress on chemical and physiological indicators of stress in cattle, we obtained two metrics of the cattle: 1) landscape-use behavior and 2) vigilance.
- Cattle landscape-use behavior was quantified using data from range rider data sheets and deployed motion-activated cameras (see above for camera design and forage availability calculation methods using RAP). Cameras were then classified into low, medium, or high-quality foraging areas based on forage biomass tertiles (lbs/acre) calculated separately for each allotment within each 16-day RAP interval (RAP; Allred et al., 2021). Thus, forage classifications were relative to conditions within a given ranch and time period rather than applied uniformly across the study area. To characterize forage conditions at a broader spatial scale more aligned with average cow traveling speeds while grazing, forage biomass values were averaged across cameras within each camera group to generate a group-level measure of forage availability by date. Camera groups were subsequently classified into low, medium, and high forage categories using the same tertile-based approach.
Camera trap images were processed as described above for cattle classification among other species using SpeciesNet (Google, 2026). Cattle use by camera was organized by date (dates spanning all days at least one camera was active in each camera group), and photos were coded for the total number of cows recorded on that date. Dates with no photos of cows were scored zero.
In 2023, we collected cattle location data by deploying 219 VHF collars/ear tags across four ranches. VHF collars allowed range riders to more easily locate and record cattle locations on rider data sheets, which we hoped would allow for improved cattle location data, but also a comparison between riders with and without VHF assistance. We prioritized collaring lead cows at the three operations, with a minimum of 20% of each herd receiving a collar or ear tag. Cattle were collared at the start of the season with help from ranchers, riders, and their employees. We had a collar/tag failure rate of about 30% by the time units were removed in October and November of 2023. However, we found from communication with range riders that finding cows with at least 20% herd coverage was more than sufficient (continuously hearing 3-6 cows at once). For that reason, we only collared 10% of each herd in 2024 to simplify our process and stay within the number of remaining functional units. We deployed about 150 VHF collars/tags on three of the same ranches last year, adding a new producer in Washington state. We did not place collars on cattle in Oregon last year. Collars and tags were removed during the months of October and November with a similar failure rate of about 40%.
Unfortunately, rider telemetry data were excluded from analyses of cattle foraging distributions because location records from rider data sheets were incomplete and reflected opportunistic rider movements rather than systematic searches for cattle. Therefore, telemetry-derived locations were spatially biased toward established riding routes and were not considered representative of cattle forage selection. However, future analyses will use telemetry data to evaluate whether access to cattle telemetry reduced rider response times to cattle groups and increased the duration of time a rider was present near livestock relative to riders without telemetry.
2. Cattle vigilance data was collected by our game cameras and also organized by unique camera deployment. Vigilance is the amount of time cattle spend moving or on the lookout for predators rather than eating, ruminating, or resting, all of which contribute to weight gain, reproductive success, and improved body scores (Laundré et al., 2001). We defined vigilance as head up above the ground. Non-vigilant behavior was defined as head at the ground (foraging/grazing - see Figure 3 for example photos). We were unable to use more fine-scale behavioral distinction as planned. We generated a behavior training dataset from our SpeciesNet cattle detections and annotated them manually to fit our definition criteria.
Unfortunately, our previously proposed method for classifying cattle vigilance behavior (LabGym - Hu et al., 2023) did not end up meeting our classification confidence threshold. As of December 2025, we are now building our own AI behavior classifier from scratch. This has slowed our process significantly, but once photos are coded, vigilance scores will be organized by unique camera deployment, and the proportion of total cows scored vigilant out of total cows detected will be analyzed.
By modeling both herd vigilance and herd landscape use/foraging behavior as a function of varied rider and predator activity, these data will allow us to answer the following research questions:
- Does rider activity influence the proximity of predators to cattle?
- Does rider activity influence vigilance in cattle?
- Does rider activity influence the quality of foraging areas used by cattle?
Cattle Biochemical Data
Although not a deliverable for this grant, we evaluated whether predator activity and varied range rider behavior influenced endocrine indicators of physiological stress in cattle. Endocrine measures, when interpreted alongside behavioral observations, provide a more comprehensive assessment of animal physiological condition than behavior alone (Boonstra et al., 2005; MacDougall-Shackleton et al., 2019). Cortisol reflects activation of the hypothalamic-pituitary-adrenal axis (HPA) and is associated with energy expenditure and perceived risk (Gerlach, 2015). In contrast, thyroid hormones, regulated by the hypothalamic-pituitary-thyroid axis (HPT), reflect metabolic activity, nutrient intake, and processes influencing body condition and reproduction (Gy et al., 2002). Interpretation of glucocorticoids can be challenging because cortisol responds to a variety of ecological and physiological stimuli not exclusively linked to stress (Gerlach, 2014), and prolonged stressors may producer muted glucocorticoid responses in some mammals (Caceres et al., 2023; Chen et al., 2015). However, sustained vigilance and reduced foraging may still manifest in lower thyroid concentrations even when cortisol responses are muted (Gy et al., 2002; Pryce et al., 2001). Jointly evaluating cortisol, thyroid hormones, and vigilance behavior allows a more robust assessment of predator-induced physiological effects than any single metric alone.
Sample Design
We collected tail hair samples because hair provides an integrated measure of endocrine activity over extended periods, and tail hair grows longer than body hard on cattle, reflecting cumulative rather than acute physiological responses. We sampled at least 30% of each participating herd to capture both intra- and inter-herd variation. Sampling occurred twice annually: in the spring immediately before cattle turnout onto grazing allotments, and in the fall following herd return at the end of the grazing season. The initial spring sample for each ranch served as a baseline for pre-grazing endocrine levels. To ensure fall samples reflected hormone accumulation during the grazing period only, tail hair was trimmed each subsequent spring (without analysis) to remove prior growth. Therefore, each herd contributed one baseline spring sample and one fall sample per ranch-year (1-3 years depending on participation). Tail hair samples (~5 cm) were trimmed using electric clippers and submitted to the Endocrinology Laboratory at the Smithsonian Conservation Biology Institute (Front Royal, Virginia) for hormone extraction and analysis. Hormone concentrations were quantified from hair and reported as ng/g hair sample.[RN4]
Hormone Extraction and Assay Procedures
Hair cortisol and thyroid extraction followed established protocols adapted from Meyer et al. (2014). Samples were washed in ethanol to remove external contaminants such as dirt, sweat, and sebum, then dried and ground into a fine powder to disrupt the keratin matrix and improve extraction efficiency. Approximately 0.10 ± 0.02g of powdered hair was incubated in 2 ml of 100% methanol for 24 hours under continuous agitation. Samples were then centrifuged for 5 minutes, the supernatant collected, evaporated under air, and reconstituted in Assay Buffer Concentrate X053 (Arbor Assays, Ann Arbor, MI) before storage at −20 °C until analysis.
Cortisol concentrations were quantified using an in-house enzyme immunoassay (EIA; R4866, Munro, University of California, Davis, CA; 1:85 [C.J. Munro, University of California, Davis]) validated in multiple wildlife studies (Young et al., 2004; Maly et al., 2018; Putman et al., 2019; Fazio et al., 2020). Samples were run in duplicates. Inter-assay variation was <15% and intra-assay variation was <10%. Assay sensitivity was 0.78 pg/ml. Serial dilutions of fecal extracts produced a displacement curve parallel to the standard curve, and recovery of added standards averaged 94% (y = 1.1691x -3.3064, R2 = 0.9943).
Triiodothyronine (T3) concentrations were quantified using a commercial T3 ELISA Kit (K056-H1, Arbor Assays, Ann Arbor, MI). Samples were run in duplicates. Inter-assay variation was <20% and intra-assay variation was <10%. Assay sensitivity was 37.4 pg/ml. Serial dilution of fecal extracts yielded a displacement curve parallel to the standard curve, and recovery of added standards was 113% (Y= 1.0077x + 510.86, R2 = 0.9961).
Research Objective 1 - Models and Statistical Methods
Our objective was to identify whether varied range riding activity alters behavioral indicators of stress in cattle. Our research hypotheses were that as rider intensity, time spent within proximity of the herd, and time riding from dusk to dawn increased, we would also see a 1) decrease in seasonal herd vigilance and 2) cattle use of high availability foraging areas will increase.
Model 1a. - Herd Vigilance
Vigilance data has not been analyzed due to the needed change in AI coding software. We plan to model herd vigilance as the number of vigilant cattle relative to the total number of cattle observed using a Bayesian beta-binomial generalized linear mixed model. Vigilance behavior in cattle (similar to wild ungulates) is facilitated socially, meaning vigilant behavior in one animal increases the likelihood of vigilance among nearby animals. This contagion violates the independence assumption inherent to binomial and Poisson distributions for count data. A beta binomial likelihood accommodates for within-group dependence, facilitating improved estimation and parameter certainty. Using the brms package (version 2.23, Bürkner, 2017), which implements Hamiltonian Monte Carlo sampling via Stan (Carpenter et al., 2017) via program R (version 4.5.2, R Core Team, 2026), we will model vigilance counts for unit i (ranch-year) as:
where yijkry is the number of vigilant cattle per game camera deployment, ni is the total number of cattle observed per unique game camera deployment, 𝝅iijkry is the probability of vigilance on the logit scale, and theta is the dispersion parameter for the binomial process. We modeled random intercepts for camera, ranch, cluster, and year to account for the hierarchical structure and repeated observations (normally distributed with a mean of zero and estimated variance). Our fixed effects, informed by our hypothesis, include:
- Predator activity (indexed as total number of wolf and/or grizzly bear photos per unique camera deployment),
- Rider frequency (indexed as the total number of rides per ranch-season,
- Mean ride duration (indexed as the averaged ride duration across all sampled rides per ranch-season,
- Mean proportion night riding (indexed as the proportion of each sampled ride taking place between dusk and dawn, averaged over all sampled rides per ranch-season,
- Mean proportion of cattle effort (indexed as the proportion of each sampled ride focused on cattle related activities (see definition for “cattle related” above) averaged over all sampled rides per ranch year, and
- Landscape openness (indexed as a percent per ranch year across all utilized pastures).
All continuous predictors will be standardized to improve coefficient interpretation and model convergence. Predictors will be evaluated for collinearity using Pearson's Correlation Tests and a threshold of |r| ≥ 0.7. We will specify weakly informative priors to regularize parameter estimates while allowing our data to inform our posterior distributions, and run sensitivity tests to ensure posterior estimates are not artificially driven by priors. Convergence will be assessed using the Gelman-Rubin diagnostic, with values less than 1.01 indicating adequate convergence (Gelman & Rubin, 1992; Vehtari et al., 2021). Effective sample size (ESS) and traceplots will be used to evaluate sampling efficiency and visually inspect chain mixing, and we will ensure sufficient exploration of the posterior distribution by the lack of divergent transitions. Model fit will be evaluated using posterior predictive checks to compare simulated datasets from the posterior predictive distribution to observed data (Gelman et al., 2013).
Model 1b. - Forage Selection
Because rider telemetry data were incomplete and spatially biased toward customary riding routes, cattle location information from rider datasheets were not considered representative of cattle forage selection and space use, and therefore not used for analyses. Instead, we used our systematically deployed game cameras spanning gradients of forage availability to quantify cattle space use and evaluate how predator and rider activity influence cattle selection. Range cattle, like wild ungulates, exhibit spatial aggregation driven by forage quality and social dynamics, so we used count data as a proxy for herd subgroup clustering behavior. We modeled cattle selection of camera clusters with varied forage availability as the daily count of cattle detected at each unique camera deployment, aggregated across all cameras in that cluster to better reflect spatial scales relevant to cattle grazing movement. These preliminary results used data from our 2024 grazing season (n=14 total operations).
Average forage biomass (kg/ha) per month was derived from the Rangeland Analysis Platform (RAP) across 30-m cells (Allred et al., 2021) and assigned to each unique camera deployment using a 30-m buffer. As a sensitivity test, variability across a 30m, 150-m, and 300-m buffer size was tested showing insignificant variation. Forage biomass values per camera were then averaged across all cameras in each camera cluster for a monthly cluster forage availability measurement. Cluster forage biomass values were then classified into low or high categories based on reference distributions calculated separately for each allotment within each month (summarized across RAP’s 16-day intervals). Cluster forage biomass values above the mean value per ranch-month were classified as high availability and cluster forage biomass values below the ranch-month mean were classified as low availability. This approach ensured camera cluster forage classifications were relative to local conditions (both spatial and temporal) rather than imposed uniformly across the study area. For example, a “low” camera group forage classification is interpreted as low relative to the allotment-wide seasonal production range within that monthly window, not relative to other camera clusters or operations. Daily cattle counts were derived from photographs on dates when at least one camera within a group was active. Days with no cattle detections were recorded as zero to preserve the observation process. To ensure accurate representation of sampling effort, we removed any dates a camera was considered inactive indicating malfunction or theft. Each camera’s last photo was used to determine the date a camera became inactive. 74.4% of cameras recorded photos within two days of being pulled from the field, suggesting minimal data loss due to inactive days.
We modeled cattle detections per camera-day as an index of cattle space use/intensity (total number of cattle detected per active camera, per day aggregated across all photos recorded that day). Because we could not identify individual cattle, detections may represent repeated observations of the same individuals/groups. Therefore, our response is an index of cattle activity/intensity of use, and not a measure of abundance. We fitted a Bayesian generalized linear mixed model with a negative binomial likelihood implemented in the brms package (version 2.23, Bürkner, 2017) in program R (version 4.5.2; R Core Team, 2026). Remote game camera count data are often overdispersed, and a negative binomial with log-link can accommodate overdispersion better than a Poisson distribution. We modeled cattle counts for unit i (ranch) as:
where yijkrt is the total number of cattle i detected at camera j from cluster k on day t on ranch r, and μijkrt is the expected cattle count. ϕ is the dispersion parameter.
To account for repeated measures and site-specific differences in baseline detection rates, we included a random intercept for unique camera deployment (j). We also included random intercepts for camera cluster and ranch to account for hierarchical structure and repeated observations across ranches and cameras (normally distributed with a mean of zero and estimated variance). Our fixed effects, informed by our hypothesis, include:
- Predator activity (indexed as the total number of wolf and/or grizzly bear photos per unique camera deployment),
- Rider (indexed as both 1) total rides per ranch-season, and 2) a binary representing whether or not the herd had a rider present that date (yes/no),
- Forage availability classification for each camera cluster (indexed as low or high).
We ran two sub-models - one with all 14 ranches using total ride frequency as our rider metric, and one with 11 of the ranches where rider activity by date could be analyzed (rider present that day or not). All continuous predictors were standardized to improve coefficient interpretation and model convergence. Predictors were evaluated for collinearity using Pearson's Correlation Tests and a threshold of |r| ≥ 0.7 with no high correlations detected. We ran four chains of 5,000 iterations each, with 2,500 warmup iterations per chain, yielding 10,000 post-warmup draws. No thinning was applied (thin = 1). We specified weakly informative priors to regularize parameter estimates while allowing our data to inform our posterior distributions and ran sensitivity tests to ensure posterior estimates were not artificially driven by priors (normal(0, 1) for fixed effects and random intercept standard deviations). Convergence was assessed by inspecting traceplots for adequate mixing and using the Gelman-Rubin diagnostic with R-hat values less than 1.01 indicating adequate convergence (Gelman & Rubin, 1992; Vehtari et al., 2021) and by inspecting effective sample size (Bulk_ESS and Tail_ESS).
Models 1civ. and 1cv. - Cortisol and Thyroid
Our preliminary analyses on hormone concentrations used only inferential tests and not fully specified multivariable models (see Results section).
Research Objective 2 - Qualitative Interview Methods
Conflict reduction tools are often created and evaluated without the direct involvement of ranchers. This can result in tools being researched at inappropriate temporal or spatial scales, or testing within a limited scope that does not account for the diverse, complex, and sometimes limiting relationships between an operation’s social, ecological, and economic dynamics (Naugle et al., 2020; Chambers et al., 2021; Hyde et al., 2022). In turn, this can result in tools or solutions that are not feasible to deploy or maintain, are cost-prohibitive, or simply ineffective in certain contexts (Chambers et al., 2021). The coproduction of our research questions and methods has ensured our approach to quantitative analyses is relevant to operational processes. By also collecting qualitative data, we helped ensure contextual understanding of our quantitative findings from our project partners, utilizing range riding regularly.
Recruitment
We conducted semistructured interviews (Interview Questions CIG) with producers and range riders who participated in our research using criterion-based purposive sampling (Campbell et al., 2020). We interviewed 20 participants; nine interviewees were from Montana residents, seven from Washington, two from Arizona, and one from both Oregon and New Mexico. Eight interviewees identified primarily as a producer, seven as a range rider, and five as occupying both identities/roles (i.e., livestock producers who ride their own herds regularly). All operations have active wolf populations and operations in Montana, and one operation in Washington, also have active grizzly bear populations. All interviews were conducted with participants’ informed consent and adhered to the ethical standards of the Utah State University Institutional Review Board (protocol #12545).
Interview Guide
We designed a semi-structured interview guide to elicit rancher and range rider perspectives on range riding as a management tool. Interview questions prompted participants to reflect on why they chose to participate in range riding, whether they perceived range riders as proficient in various conflict mitigation strategies, how to deploy riders effectively, historical and present predator conflict context, perspectives on range rider programs, and perspectives on communication and coordination within range rider efforts and/or range rider programs. Interviews were conducted in person between April 2021 and December 2024 and were audio recorded for transcription and analysis. Consistent with semi-structured interviewing practice, participants were encouraged to lead conversations toward topics they felt most relevant to their own experience (Tisdell & Merriam , 2025).
Analysis
We conducted a hybrid deductive-inductive thematic content analysis, a method used to systematically code qualitative data by identifying recurring patterns or themes relevant to research objectives (Saldaña, 2016; Clarke & Braun, 2017). Deductive analysis applies existing theoretical concepts or research to guide the coding process, whereas inductive analysis allows additional themes, patterns, and concepts to emerge directly from the data through repeated transcript review (Clarke & Braun, 2017; Fereday & Muir-Cochrane, 2006). Our thematic content analysis had three phases. In phase one, our team reviewed the Gonzalez et al. codebook (2024) developed for an evaluation of state-funded range rider programs in Washington State to understand definitions of categories and themes. Prior to formal coding, we developed an initial codebook by adapting the Gonzalez et al. (2024) codebook to reflect the present dataset and research objectives. Drawing on researcher immersion and contextual familiarity with the system, we retained applicable codes and added new codes to capture themes not included in the Gonzalez et al. (2024) framework (Fereday & Muir-Cochrane, 2006; Braun & Clarke, 2021). In phase two of analysis, two researchers used the initial codebook to independently code four transcripts representative of the varying stakeholders that participated in interviews. Through this inductive process, researchers finalized the codebook by identifying new codes, adjusting codes, and reconciling discrepancies in coding through consensus. In the third phase of analysis, coders used the final codebook to complete coding all 20 interviews and reviewed one another's work to meet consensus. Interviews were transcribed using Otter.ai software (Otter.ai, Inc.) and coded using Dedoose software (Version 10.34; SocioCultural Research Consultants, LLC).
Over time, as operations, participating producers, predator populations, and extenuating circumstances have evolved over the last four years, so have our research objectives to accommodate. To note, the primary research objectives to help achieve our project’s consistent goals are:
- Evaluate the effectiveness of different intensities and styles of riding at reducing behavioral and chemical indicators of stress in grazing livestock.
- Through interviews with ranchers, provide the context and detail necessary to understand decision making around range riding as it relates to operational management and protocols, ecosystem resiliency, and economic sustainability.
Other project objectives are primarily concerned with education and outreach, so they are described in detail in the educational section below. However, all the results and information we learned from reaching our research objectives were applied to the educational objectives. Through objective 3, we worked to ensure the information was accurate and readily available to incorporate into objectives 2 and 4 using an iterative process. As we gained information from objective 1, we tried to incorporate it into the outreach materials, disseminate it at workshops, and discuss it at rancher-to-rancher exchanges. We used feedback from ranchers and other stakeholders about the clarity of information, the relevancy of these findings to their practices, and what other information was needed.
Over time, as operations, participating producers, predator populations, and extenuating circumstances have evolved over the last four years, so have our research objectives to accommodate. To note, the primary research objectives to help achieve our project’s consistent goals are:
- Evaluate the effectiveness of different intensities and styles of riding at reducing behavioral and chemical indicators of stress in grazing livestock.
- Through interviews with ranchers, provide the context and detail necessary to understand decision making around range riding as it relates to operational management and protocols, ecosystem resiliency, and economic sustainability.
Results Objective 1: Research Findings
Research Objective 1a. - Cattle Vigilance Model
Results forthcoming (see above).
Research Objective 1b. - Cattle Forage Selection Model
Our remote game cameras accumulated 47,054 camera days for the 2024 grazing season. Of the 95 unique clusters across 14 different ranches, 34% were considered high forage availability, and 66% were considered low. For the maximum, minimum, mean, and standard deviation values for cluster mean forage biomass, total predators, total rides, and proportion of season days with a rider, see Table 1.
Model 1 - Total Rides
To model cattle forage selection based on predator activity and total rider frequency (rides per ranch-season), our model included 47,054 observations across 886 unique camera deployments, 95 unique clusters, and 14 ranches. All parameters demonstrated convergence (R^ = 1.00) with an adequate effective sample size indicating stable posterior estimates (Bulk_ESS > 1,500, Tail_ESS > 2,000). No divergent transitions were observed. Our shape parameter was estimated at 0.14 (95% Bayesian Credible Interval (BCI): 0.13, 0.15), indicating significant overdispersion and supporting our use of a negative binomial distribution. Posterior predictive checks indicated that the model adequately reproduced the distribution of observed counts.
Random Effects Model 1
Substantial variation in cattle counts occurred at the camera level (σ = 2.84, 95% BCI: 2.55, 3.18), with moderate variability among clusters (σ = 1.67, 95% BCI: 1.25,2.13) and ranches (σ = 0.60, 95% BCI: 0.04, 1.40).
Fixed Effects Model 1
Cattle counts were significantly lower in low forage availability clusters relative to high availability (β = -0.92, 95% BCI: -1.75, -0.11). Expected counts of cattle were approximately 60% lower in low forage areas. Neither predator activity (β = 0.07, 95% BCI: -0.08, 0.22) nor rider activity (β = 0.35, 95% BCI: -0.38 to 1.09) were strongly associated with cattle counts, although increased predator activity was positively associated with cattle counts in this model. Interaction terms between predator activity and forage category (β = -0.07, 95% BCI: -0.26, 0.12) and between rider frequency and forage category (β = 0.20, 95% BCI: -0.53, 0.95) were small and not meaningful, providing little evidence that predators or riders altered forage selection patterns. (Figure 4).
Model 2 - Daily Rider Presence
To model cattle forage selection based on predator activity and whether or not a rider was active each day (rider binary), our model included 35,670 observations across 606 unique camera deployments, 66 unique clusters, and 11 ranches. All parameters demonstrated convergence (R^ = 1.00) with an adequate effective sample size indicating stable posterior estimates (Bulk_ESS > 1,500, Tail_ESS > 2,000). No divergent transitions were observed. Our shape parameter was estimated at 0.13 (95% Credible Interval (BCI): 0.11, 0.14), indicating significant overdispersion and supporting our use of a negative binomial distribution. Posterior predictive checks indicated that the model adequately reproduced the distribution of observed counts.
Random Effects Model 2
Substantial variation in cattle counts occurred at the camera level (σ = 3.06, 95% BCI: 2.67, 3.49), with moderate variability among clusters (σ = 1.32, 95% BCI: 0.73, 1.95) and ranches (σ = 0.99, 95% BCI: 0.17, 1.86).
Fixed Effects Model 2
Cattle counts were significantly lower in low availability forage clusters relative to high availability forage clusters, but the effect was only moderately meaningful (β = -0.80, 95% BCI: -1.80, 0.23). Expected counts of cattle were approximately 55% lower in low forage areas. Increased predator counts had a moderately meaningful negative effect on cattle counts (β = -0.21, 95% BCI: -0.63, 0.13). A one SD increase in predator count (0.13) was associated with an expected 19% decrease in cattle counts. Days without a rider were associated with slightly lower cattle counts overall, but the effect was not meaningful (β = −0.15, 95% BCI: −0.52, 0.21). Our interaction term between forage availability and rider presence was not meaningful (β = 0.11, 95% BCI: -0.35, 0.57), suggesting limited evidence that rider presence impacted forage selection in cattle. However, our forage availability and predator activity interaction was moderately meaningful (β = 0.24, 95% BCI: -0.14, 0.68). A one SD increase in predator count was associated with an expected 24% increase in cattle use of low availability forage areas, suggesting predator activity may influence cattle to select for lower quality foraging areas (Figure 5).
Research Objective 1civ. and 1cv. - Exploratory Analyses of Cortisol and Thyroid
In 2022, we collected n = 319 cattle hair samples, n = 1090 in 2023, and n = 747 in 2024. Of these samples, 2152 were viable for cortisol analysis, and 719 were viable for thyroid. Although our cattle vigilance behavioral data is not yet ready to compare with our cortisol, thyroid and body condition score results, we conducted a preliminary analysis of tail hair samples collected during the 2023 grazing season (spring and fall) from three geographically and operationally similar ranches: Ranch O (n = 162), Ranch A (n = 27), and Ranch I (n = 33). Ranches were selected to minimize environmental and management variation that may independently influence endocrine markers (Heimbürge et al., 2019; Moya et al., 2013; Bristow & Holmes, 2007). We evaluated differences in cortisol and thyroid hormone concentrations across 1) season (spring vs. fall) and 2) ranch.
Cortisol
A two-way ANOVA (Fisher, 1925) revealed a significant main effect of ranch on cortisol concentrations (p < 0.001) and a significant ranch × season interaction (p = 0.02), but no overall main effect of season (p = 0.94). Ranch explained approximately 20% of the variance in cortisol concentrations, whereas season explained less than 1% (Figure 6).
Post hoc Tukey’s HSD tests (Tukey, 1949) indicated no significant difference between Ranch I and Ranch A (p = 0.27). However, Ranch O exhibited significantly higher cortisol concentrations than both Ranch A (p < 0.001) and Ranch I (p < 0.001). Season remained non-significant in post hoc comparisons (p = 0.95). The significant interaction reflected consistently higher cortisol concentrations at Ranch O relative to Ranch I in both spring and fall (p < 0.001 for both seasons), and higher concentrations relative to Ranch A in autumn only (p < 0.001).
Paired t-tests (Student, 1908) were conducted to assess seasonal differences within ranches (Figure 7). Ranch O showed no significant seasonal difference (p = 0.23; mean difference = −0.04 ng/ml). Ranch A exhibited a significant seasonal increase (p < 0.01; mean difference = 0.25 ng/ml). Ranch I showed no seasonal difference (p = 0.95; mean difference = 0.00 ng/ml).
Thyroid Hormone (T3)
Because thyroid data were non-normally distributed, we used a Kruskal–Wallis test (Kruskal & Wallis, 1952). Ranch had a significant effect on thyroid concentrations (p = 0.05), whereas season did not (p = 0.34; Figure 8). Dunn’s post hoc tests (Dunn, 1964) indicated significant differences between Ranch A and Ranch I (p = 0.02), and between Ranch O and Ranch I (p = 0.02), but no difference between Ranch O and Ranch A (p = 0.43).
Wilcoxon signed-rank tests (Wilcoxon, 1945) were used to assess seasonal differences within ranches (Figure 9). Ranch O (p < 0.001, V = 13) and Ranch A (p < 0.001, V = 102) exhibited significant seasonal differences in thyroid concentrations, whereas Ranch I did not (p = 0.78, V = 70).
Research Objective 2 - Qualitative Interviews
Across our 20 participants, our analysis resulted in three categories of perceptions: 1) range riding approaches (6 themes, 10 unique codes), 2) rider skills, knowledge, and characteristics (2 themes, 2 unique codes), and 3) range riding context (1 theme, 1 unique code). See Table 2 for each theme and code, along with their definitions, organized within their corresponding category.
We coded specific perceptions across our three categories a total of 653 times (Table 2). Our most frequently coded category was range riding approaches (n=391), followed by range riding context (n=321), and rider skills, knowledge, and characteristics (n=150). Every participant shared at least one perspective on all six categories. The following results explore the top three themes within each category.
Range Riding Approaches
Participants most frequently discussed how to apply/deploy range riders (Table 2) with participants describing perspectives on optimal riding frequency, timing, responsibilities, communication expectations, and overall effectiveness.
Rider Frequency
Most participants shared that frequent riding - ideally near daily - was important for effectiveness. However, the right frequency was considered context-dependent, influenced by allotment size, terrain, accessibility, and current predator conflict. Interviewees emphasized the importance of the rider learning the landscape, wildlife, and livestock patterns in order to effectively detect shifts that may signal emerging conflict, and adapt frequency accordingly. Several participants mentioned that on large rugged allotments, complete coverage in a single day was often unrealistic; thus, how often a rider rides was generally considered more important than for how long they ride.
When to Ride
Participants expressed mixed views on night riding. Several participants acknowledged that riding when predators were most active (e.g., dawn to dusk) may improve riding’s effectiveness, but felt this posed more safety risks - particularly with the risks associated with riding horses in the dark, not being able to see well enough to be effective, and the potential threat of encountering a grizzly bear at night outweighing potential benefits. One producer described their experience trying to range ride at night:
“Total waste of money in my own opinion, I think it razzled the cows more than we helped that night, so I never went out again. And as far as taking a horse out…it's not even logical.”
Other participants felt night riding provided a unique opportunity to detect abnormal behavior in cattle that would normally be bedded at that time. Camping in the herd, rather than actively riding, was frequently mentioned as a way to reduce response and commute times.
What Riders Should Focus on While Riding
Participants shared frequently that riders should prioritize monitoring cattle and locating as many animals as possible. Activities like checking cows for illness, injury, and stress were mentioned as essential, and having knowledge in livestock was seen as necessary to do those tasks well:
“You have to understand cattle...people that don't understand and don't know how to read cattle and cannot comprehend what is going on…make it worse.”
Some participants shared that operational or “cowboy” tasks such as fixing fence, putting out salt, or doctoring injured or sick animals were integral to effective range riding, while others viewed these tasks as unnecessary to a rider’s core goals:
“...that's why I say all riders should be cowboys. Because what the hell good does it do to have this guy just out there riding around and not see that the fence is down? It all goes together. And besides that, if he doesn't see that the fence is down, maybe there's seven cows over here that he doesn't even know…if it's a cowboy, he's going to see that…He's going to say, ‘Oh, I must have some cows over here…I better go see where they are and put them back wherever they belong’.”
Participants also shared that riders should monitor predator activity by tracking their sign (e.g., tracks, scats, behavior), detecting depredations, managing game cameras, and hazing predators when necessary.However, some participants believed that excessive time spent away from the herd could be detrimental as changes in cattle behavior are often the earliest indicator or predator conflict.
Participants frequently mentioned using creative ways to "mix up" their rider activities and patrol routes to take advantage of predator fear of humans. One rider noted that wolves may become familiar with rider presence and scents over time, thus it is essential riders change their approaches and behaviors regularly to ensure they stay “...something (for the wolves) to worry about.” Participants also mentioned that riders should be adaptable, observant, and curious:
“...I don't think like, trotting through pasture and covering every patch is going to be quite as informative as…taking your time and…just observing what's around you."
Perceptions on Rider Effectiveness:
Participants described varied definitions of range riding effectiveness. Most participants questioned whether riders can reduce predator depredation, particularly from highly motivated predators:
“They (the riders) are not going to have any effect if that wolf hits or if that bear hits cattle ever…We cannot be that omnipotent, or whatever you want to say…that's giving us way too much credit. Yeah, we're not going to change their habits.”
Others felt their ability to reduce depredation was apparent:
“There's no doubt in my mind that it works to a point…I mean, anytime you got eyes and ears out there, you're looking at something."
“...they had a lot of losses, and as soon as we started riding, they didn't have an issue again last year.”
Several respondents suggested that predator hunting and trapping seasons may improve rider effectiveness by reinforcing predator fear of humans:
“Certainly the fact that we shoot our wolves here and trap our wolves, I think trapping plays a big part, and maybe a lot more than people realize...”
Many participants emphasized that direct loss/depredation reduction may be difficult to quantify, and may not fully capture the value of a range rider:
“No, yeah, I don't feel like we prevent depredations at all. I feel like we very effectively, though, keep ranchers in a (calm) mental state…and keep them on the landscape…at the end of the day…what is like, the ultimate goal? Is it preventing depredations or is it keeping ranchers?...I feel like we are very effective at supporting the ranchers, making them feel like they're not alone…reducing their anxiety. They can sleep better at night.”
“And my measure of effectiveness is if the landowners are happy with it, if they thought it was worth it, then it was effective. It doesn't matter if every cow died, if they (the rancher) thought that he (the rider) did something up there that was helpful, then it was effective.”
“Maybe I would have had 10 calves killed, 14 calves killed in 2015 instead of 5, because the range riders did a good job.”
Interviewees also highlighted the importance of communication and collaboration between riders, producers, and wildlife agencies as part of their effectiveness, particularly concerning timely lethal removal in cases of chronic depredation. Riders were viewed as critical information gatherers that support agency decision making, but riders are not solely responsible for resolving persistent conflict. Most preferred daily communication, with one producer noting:
“I'd rather have too much communication than not enough.”
Participants discussed whether or not riders are effective at displacing conflict spatially. Most felt any spatial displacement was localized and temporary, but that information sharing across operations was a significant benefit.
Opinions varied regarding whether riders meaningfully reduce predator-induced stress in livestock that leads to indirect losses (e.g., reduced pregnancy and return rates, increased illness rates, decreased weaning weights). Some participants believed livestock stress was at least alleviated while riders were present with the herd.
Participants acknowledged that learning the livestock and the landscape takes time, and thus rider turnover was one of the largest hindrances to long term effectiveness:
“The golden ticket for a range rider is finding somebody who can stay for a long amount of time. It's hard to learn country. It's hard to learn the different herds and their different behaviors. It's time consuming as a producer to train these people and teach these people.”
Access to Predator Collar Location Data
Access to predator collar location data was the second most coded perception (Table 2), where participants described the potential benefits, limitations, and challenges obtaining data from agencies. GPS collars were considered useful for identifying broader predator movement patterns, however, data delays often limited utility for rapid response. Alternatively, VHF collars were viewed as more useful for real-time response, though limitations included terrain constraints.
“And it (predator location information) gave us detailed information where they were at…And it was wonderful, because if they weren't in the area, then I felt like I could go about my business. If they were in the area, and then I'd go out, check and see what was going on. So that was really helpful.”
“I can't track the ones I can't find, right? I mean, being able to zero in and find specific wolves (VHF) is more beneficial to me…I don't need to know where the wolves were two weeks ago. I need to know where they were two hours ago.”
Participants described barriers to accessing timely collar data from wildlife agencies (primarily due to poaching concerns), and emphasized the role location sharing by agency staff with producers and riders played in building trust and partnership:
“I think from a long term relationship…I think it would be more functional for the department to monitor collar data and be transparent with those things…I think it's a great way for landowners and agency folks to kind of keep breaking down some of those barriers that exist”.
Others argued that skilled riders should not need collar data to be effective:
“...if your rider’s out there, and he's listening, looking and reading the signs on the ground. He knows whether they (the wolves) are there or not there.”
The drawbacks to using dogs (cons dogs) was the third most cited perception (Table 2). While dogs may provide protection from grizzly bears, participants raised concerns about dogs attracting bears and causing potential human mortality, and increasing stress in livestock, particularly in areas with elevated wolf activity:
“When I have wolves in my cattle, I cannot use my herd dogs at all…I have to go in by myself. The cattle are so scared and stressed out that they run everything into the dirt, even horses.”
Range Riding Context
This category had one specific theme: Why range riding and/or main goals for having a rider (Table 2). Common primary goals for employing a rider included livestock monitoring, minimizing depredation, providing presence, improving response times, early detection of potential depredations, information gathering, and protecting a producer’s economic investment:
“I feel better having (rider) and his feedback, and his presence on the landscape, helps me feel like my cattle are having a better opportunity at success.”
“So you know what's going on. So if you do have a depredation you can actually find it, otherwise you're going to have 20 of them before you ever know you had one. That's why I think they're important.”
“Ranching is labor intensive, and we already have too much work to do, and having to babysit predators is just one more thing that we have to try to fit in. And if we're trying to do a good job keeping predators at bay and keeping our livestock alive, then we don't have time to do the rest of our work. And because we didn't ask for these predators to be here, somebody else needs to manage them so that we can manage our livestock.”
One rider stated:
“I think it's time to bring range riders back. They're necessary again. And I see my job as an insurance policy…”
Perceived Additional Benefits
Perceptions of range riding effectiveness were often linked to perceived additional benefits that motivate a producer to deploy a rider. Additional benefits included wildfire detection, operational support through holistic risk management, relationship building, and maintaining the viability of livestock production:
“...ultimately, you're helping everybody, you know? By me being out trying to keep the depredations down, it’s helping the environmental groups with their goal of not killing (predators). Helping the rancher with keeping the cows safe.”
“The benefit for me personally, would be coming out with 100% success rate of our calves, but (in addition) you're building relationships…that's a really important part of their job is building that relationship piece.”
“And I think if you want to keep open space and not see things turn into housing developments you need to keep producers on the landscape. And part of keeping producers on the landscape is keeping their operations viable. And to keep them viable, you can't be having 40% open cows, you can't be losing 20% of your herd to depredations… There's a reason that the elk spend their time on my ground. There's a reason that the grizzly bears spend their time on my ground.”
Rider Skills, Knowledge, and Characteristics
Similar to our findings under the how to apply/deploy range riders theme, stockmanship, track and sign identification, and communication/documentation were seen as essential skills:
“They have to be able to read a cow. Know if a cow is stressed out, know, if a cow is hurting, know if a cow's about to die.”
“...being able to track - know whether you're tracking wolf tracks, or whether you're tracking cougar tracks or cow tracks, you know. Being able to take in and decipher the difference on a cow track, or yearling track, or bull track, and figure out…what's going on.”
“If you don't communicate, we're gonna be losing cattle and they're gonna be dying, and then it's bringing in more predators, because we have dead cattle for no reason. So (the rider) has to be able to communicate to you (the producer), that is a big asset.”
Participants described additional skills that contribute to a rider’s effectiveness including backcountry competence, landscape literacy, observational awareness and knowing what’s abnormal or unusual, and game camera proficiency. While participants noted many skills can be learned, structured training opportunities were requested.
Personal characteristics was the second most coded perception (Table 2), where participants described trustworthiness, integrity, patience, work ethic, and eagerness to learn. Trust was noted by all participants as extremely important:
“If somebody doesn't know, well then don't bullshit me. Just say ‘I don't know’.”
“I think the trust is huge in that aspect to where they (the producers) trust the fact that they can call you (the rider), they know that you're going to be in there when you say you're going to be in there, and if they ask you to do something, you're going to do it. You know, there's been times for all of us (riders) that we have dropped what we're doing and go if we can. And I think that is a big deal to them (the producers)...there's days you drive out of there that you are just exhausted, you know…You've got to have a lot of go, really.”
“(Range riding) is like being an airplane pilot, hour upon hour of extreme boredom with brief intermissions of sheer terror…You got to be able to stand being alone. Be comfortable with yourself. Be confident with yourself .”
Teamwork, humility, and professionalism (particularly considering the contentious nature of predator-livestock conflict) were considered more important than any one technical skill.
Citations:
See citations at insert below.
Research outcomes
Research Objective 1a - Discussion and Recommendations Cattle Vigilance
Discussion and implications for sustainable agriculture forthcoming (see above)
Research Objective 1B - Discussion and Recommendations Forage Selection
Discussion
Our research objectives were to test whether cattle were selecting for high availability forage areas across sites with active game cameras and whether rider and predator activity impacted that selection. Our first model evaluated rider effect as total rides per ranch across all ranches, and our second model evaluated rider presence each day of the season (rider vs. no rider) on ranches where that scale of data was available.
The majority of variation in our models was attributed to unique camera deployment. This suggests substantial camera-to-camera differences in cattle counts that may reflect unmeasured site characteristics. In addition to forage availability and quality, cattle selection preference can also be motivated by preferred plant species, distance to water and salt/mineral, herding pressure, and slope/terrain ruggedness (Bailey et al., 2005; McDonald et al., 2019; Ashworth et al., 2024). While all cameras were set in locations within a ≤ 30% slope (ruggedness factor) and ≤ 1 mile distance to water, unique plant species and nutritional content was not evaluated and may have influenced camera cluster selection (Bailey et al., 2005; McDonald et al., 2019; Ashworth et al., 2024). Explicitly measuring heterogeneous camera location microhabitat variables (e.g. water tank or salt lick, visibility, forage quality/nutritional content, etc.) could clarify whether rider and predator activity interact with local environmental conditions at each camera site (Acciaro et al., 2022).
Both models found an effect of cattle selecting for high availability forage areas (although the effect in the second model was only moderately meaningful). Our findings are consistent with a substantial body of literature demonstrating that grazing cattle preferentially select areas of higher forage availability and quality (Bailey et al., 2005; Teague & Barnes, 2017; McDonald et al., 2019).
We did not find support for our hypotheses that 1) increased rides per season and 2) days with a rider would affect cattle selection on high quality foraging areas. While Zengeya et al. (2015) were able to detect an impact of herders on cattle forage selection and travel speed, the majority of riders in our study do not physically move cattle regularly, instead prioritizing monitoring over management. This distinction implies that a rider effect in our models would be less direct like a herder’s effect, and more indirectly human presence mitigating predator-induced stress, allowing cattle to select for higher availability forage areas. Although some research warns that human presence with rangeland livestock should not be assumed to reduce stress depending on the familiarity and proximity of the person (Creamer & Horback, 2021), the majority of our riders had been riding their herds for 2-10+ years, and those with less time had been consistent in their herds for an entire season. We suspect a lack of rider effect from our modeling effort represents either a true lack of impact on forage selection, or may point to a need to look at cattle response to riders at a finer scale. Measures such as total rides per season or daily rider presence may be too coarse to capture meaningful relationships. More refined measures such as GPS-based analyses of rider proximity to cattle and predators, or duration of time riders spend with livestock each ride in close enough proximity to monitor their stress, condition, and behavior, may better capture the scale at which riders influence cattle selection.
Rider frequency or rider daily presence may not be the best metrics for evaluating rider conflict-reducing effort. Ride frequency and presence on allotment may not accurately represent the amount of time riders spend in close proximity with livestock, as riders may have long commutes to and from an allotment or be focused on other activities away from the herd (e.g., fixing fence, checking remote game cameras, trailing or hazing predators, or looking for livestock depredations). Future research exploring rider effect on predator-induced stress and forage selection should focus on measuring the
In our second model, we found a moderate negative association between increased predator activity and 1) cattle counts and 2) reduced selection of high availability foraging areas. These findings are consistent with research suggesting that cattle may adjust their behavior to avoid predation risk. For example, Laporte et al., (2010) found correlation between increased wolf activity and cattle grouping behavior and decreased travel sinuosity, and Clark et al., (2017) found that wolf activity was associated with decreased travel distances in cattle that may impact access to forage. However, evidence across studies remains mixed. Oakleaf et al. (2003) found no relationship between wolf spatial overlap and calf movement patterns or group size, and Weise et al. (2019) reported no effect of lion predation risk on cattle forage area selection. In contrast, Muhly et al. (2010) found cattle reduced their selection of high quality foraging areas following wolf presence, although the effect took place after wolves left the area, potentially signaling that anti-predation behaviors in domesticated ungulates has been bred out. While uncertainty remains, our findings provide moderate support for the hypothesis that cattle may alter forage selection patterns in response to predation risk.
Similar to range rider effects, cattle responses to predator activity may occur at finer spatiotemporal scales than those captured by our daily, camera-based models, potentially obscuring the magnitude of predator effects. Kohl et al. (2018) found that elk populations in Yellowstone adjusted their foraging behavior within high-risk areas to periods of reduced wolf activity, maintaining nutritional intake while reducing predation risk. A similar dynamic may occur in cattle where individuals or groups temporarily vacate high availability forage areas when predators are active, but return when risk subsides, having a minimal impact on daily cattle counts at cameras. While brief predator avoidance may not significantly influence nutritional intake, pairing these results with our forthcoming vigilance and physiological stress analyses will help clarify whether predator and/or rider activity influence cattle stress responses not fully captured by forage selection at the daily scale (Kluever et al., 2008; Smit & Kuijper, 2024).
Finally, riders are often deployed in direct response to elevated predator risk both spatially and temporally. This may mean that rider effort is correlated with conflict risk, potentially introducing confounding and reverse causality (are riders reducing predator induced stress and improving cattle use of high availability foraging areas, or are riders responding to situations where predator-induced stress is already high?). Due to the potential for delayed effect, our first cattle selection model may better reflect ecological processes by more generally measuring rider effort (total rides per season) compared to the second model (rider present that date). Future research should model rider activity as a dynamic process linked to the time and location of known predator conflict events to better explore reactive vs preventative rider effects.
Our next steps for this research include adding the 2022 and 2023 seasons to our analyses, and combining predator collar and camera data into a probability of use/occupancy model.
Future Research and Recommendations for Agriculture
- Our findings suggest cattle may be adapting their use of high forage availability areas in response to predator activity as a means to reduce predation risk. Whether this stress response is significant enough to impact livestock economics (e.g., reduced weaning weights, reduced cow body condition score, reduced pregnancy rates) will require more research.
- If predator avoidance is having subtle but cumulative nutritional impacts on livestock, strategic placement of attractants (e.g., salt/mineral) and temporary/virtual fencing should be experimentally evaluated as methods to help reduce localized spatial overlap between livestock and high-risk predator areas like den sites or rendezvous sites.
- Riders may not meaningfully reduce predator-induced stress and/or pressure in cattle to improve the use of high availability foraging areas.
- Having riders collect data on cattle locations and/or deploying GPS collars on a subset of cattle would support more research on the impacts of rider and predator activity on cattle spatial use.
- Future research should model rider activity (e.g., total rides per season, day with vs. without a rider, etc.) as a dynamic process linked to the time and location of predator conflict events to better explore reactive vs preventative rider effects.
- Combining forage selection with behavioral indicators of stress (e.g., vigilance) and physiological indicators (e.g., cortisol and thyroid hormone levels, body condition scores) may better assess livestock responses to both predator and rider activity.
- Future research should include more microhabit variables at each camera location (e.g., distance to salt/mineral or water, visibility, forage quality/nutritional content) to better understand how local factors influence predator and rider effects.
Research Objective 1civ and 1cv - Discussion and Recommendations Chemical Indicators of Livestock Stress
Discussion
Our preliminary analysis revealed significant differences in cortisol and thyroid concentrations among ranches, but no consistent main effect of season. Ranch O exhibited higher cortisol concentrations than Ranch I in both spring and fall, and higher than Ranch A in autumn. Seasonal differences in cortisol were detected only on Ranch A. Thyroid concentrations were significantly lower on Ranch I relative to Ranches A and O, and seasonal differences were observed on Ranches A and O but not Ranch I.
These results should be interpreted cautiously. In this exploratory analysis with only ranch and season as variables, each variable likely represents multiple underlying drivers of hormone concentrations, including forage availability, predator and rider activity, herd density, and more. Future models will control for this variability by including additional ranch, predator, landscape, and rider covariates to more directly evaluate the influence of predator and rider activity on hormone concentrations.
We did not expect strong seasonal differences because spring baseline samples likely contained residual endocrine signals from previous grazing periods. In contrast, ranch-level differences were anticipated given variation in environmental and management conditions.
Unexpectedly, Ranch O exhibited the highest cortisol concentrations despite historically lower wolf conflict relative to Ranches A and I. One potential explanation is that Ranch O operates on substantially larger allotments, requiring greater travel distances for forage and water, which may elevate cortisol via increased energetic expenditure through activation of the hypothalamic–pituitary–adrenal (HPA) axis (Gerlach, 2015). Alternatively, cattle experiencing chronic predator exposure may exhibit amuted cortisol responses, as documented in other mammals exposed to persistent stressors (Chen et al., 2015: Cáceres et al., 2023). If so, lower cortisol concentrations on Ranches A and I could reflect physiological adaptation rather than reduced stress. Integration with vigilance, body condition, and reproductive metrics will be critical to distinguish between these possibilities (Pryce et al., 2001; Gy et al., 2002).
Ranch A was the only ranch with higher spring than fall cortisol concentrations. Because this ranch experiences year-round wolf presence, spring samples may reflect cumulative exposure beyond the focal grazing season. This highlights the importance of considering annual predator dynamics rather than restricting inference to grazing months alone.
Thyroid hormones primarily reflect metabolic activity and nutrient availability (Gy et al., 2002). Lower thyroid concentrations on Ranch I, which has experienced higher historical wolf conflict, may indicate reduced nutritional intake potentially associated with altered grazing behavior which is partially supported by our cattle forage selection models. However, all three herds receive supplemental feeding and salt/mineral during the non-grazing season, complicating simple seasonal interpretations. The contrasting seasonal thyroid patterns between Ranches A and O further underscore the need to explicitly model forage availability and predator activity.
Overall, these preliminary findings suggest that ranch-level variation in hormone concentrations likely reflects a complex interaction among nutritional, ranch-level, and predation drivers rather than predator presence alone. Pairing these findings with cattle behavior and predator activity data (vigilance) will help clarify chemical concentration drivers.
Future Research and Recommendations for Agriculture
- Our forthcoming research will 1) incorporate predator and rider activity into the two chemical data models described in the methods section (cortisol and thyroid), 2) compare results from chemical models to cattle vigilance and body condition score findings for a more comprehensive exploration of predator-induced stress outcomes and range riding’s ability to mitigate those outcomes, and 3) incorporate all ranches from all years for a more robust analysis. This will help us identify whether herds are experiencing muted cortisol responses, or instead predator and rider activity do not have a meaningful effect on seasonal hormone concentrations. Future research should resist using cortisol hormone concentrations alone to measure metabolic stress.
- Future research should consider monitoring nutritional intake and individual movement (e.g., accelerometers) in addition to forage availability, predator activity, and rider activity since these, too can meaningfully affect metabolic stress.
Research Objective 2 - Discussion and Recommendations Qualitative Interviews
Discussion
Our objective was to interview project partners to capture participants’ perceptions related to the complexities of range riding as a management tool and to provide critical context to our quantitative results. Participants commonly reported effective communication between range riders and producers, along with the added value of increased monitoring of livestock, wildlife, and landscape, as primary motivations for using a range rider. Participants frequently expressed that effectiveness varied depending on a rider’s skills and knowledge of livestock, their ability to work in the backcountry, and the characteristics of individual predators (e.g., boldness). Our qualitative findings contribute to a growing body of literature on range riding’s role in predator conflict mitigation, overall operational success and sustainability, and how to improve range riding’s effectiveness for best outcomes under varied ecological, political, and operational contexts.
Context Dependent Range Riding Approaches
Our findings suggest that rider deployment strategies should remain context-specific and adaptive rather than prescriptive. Standardized protocols are unlikely to hold across diverse operational and ecological contexts. For example, although near-daily riding was generally perceived as optimal, participants highlighted several operational and landscape constraints that should shape rider schedules and priorities such as pasture/allotment accessibility, ability to “get eyes” on cattle, and recent predator activity and/or conflict. These considerations align with other research outlining how logistical and operational contexts should drive rider effort (Smith et al., 2025). An operation with mild terrain, low tree cover, and ample road access may lend easily to locating and checking the majority of cows in a herd, whereas an allotment with little access, rough terrain, and high tree cover may need increased riding effort. Night riding may provide more opportunities to haze wolves, but may present unacceptable safety risks in grizzly bear country. Wolves in states with active wolf hunting and trapping seasons may respond more sensitively to rider activities and hazing efforts, requiring less pressure for a desirable behavioral change. We recommend riders prioritize learning the unique landscape, livestock, and wildlife in the areas they ride to determine their best approach while staying flexible and dynamic to changing ranch and/or conflict conditions and producer goals.
Livestock Focus as a Priority
Similar to the findings of Smith et al., (2025) and Bogezi et al. (2021), our participants emphasized that effective range riding requires knowledge and skills in livestock husbandry. Core responsibilities of a rider described by respondents included locating livestock, detecting signs of conflict (injuries and depredation), checking for signs of stress or illness (that may eventually transform into conflict concerns), and learning livestock landscape use and behavioral patterns. Range riders were described primarily as information-gatherers whose value extends far beyond predator conflict deterrence. By providing real time information on the herd, range riders support producer decision making that is efficient and thorough, and help wildlife agencies respond to conflict before conflict becomes chronic.
Unlike the findings of Parks & Messmer (2016), our participants did not distinguish between predator and cattle-focused range riding when discussing approaches, emphasizing that riders should prioritize cattle monitoring, but maintain a sufficiently diverse skillset to monitor and manage predators simultaneously. Even riders hired primarily to focus on predators mentioned that herd monitoring was a priority in their approach because livestock behavior should signal a rider that predator conflict is afoot. Additionally, rider activities traditionally thought of as “predator focused” are easily transferable to livestock monitoring - aging and identifying tracks and scat can be applied to livestock and may help a rider know how quickly livestock were moving or whether predators and livestock overlapped spatially, and remote game cameras can provide critical information on livestock behavior and health in addition to predator monitoring.
Our findings suggest that while predator-related skills/knowledge are important, livestock husbandry skills/knowledge may be the most important competency for range riders and many skills are transferable across livestock and predators. However, like rider approaches, which skills are most desirable is context specific. For example, a ranch with ample support from ranch hands or cowboys may prefer a rider more focused on predator behavior. For this reason, we recommend anyone hiring riders other than the ranch owner/manager (e.g., government agencies or NGOs) should include producers in the process to ensure rider fit and operation-specific goals and conditions are addressed (Naugle et al., 2020; Ostermann‐Miyashita et al., 2025).
Important Skills and Training
The diversity of skillsets beneficial to rider effectiveness makes finding qualified riders difficult, especially considering that trustworthiness, communication, a strong work ethic and a curiosity to learn were also requested.
Structured training for range riders could improve not only the pool of qualified riders, but also overall effectiveness and the ability of riders to coordinate with other riders, agencies, and producers. Based on these findings, we recommend that training include livestock monitoring and signs of stress and conflict, livestock and wildlife track and sign identification, predator behavior and biology, and documentation and communication protocols. Incorporating agency staff and producers into shared training may also contribute to relationship and trust building by fostering shared knowledge and consistency. For example, if producers, riders, and agency staff are responding together to a suspected depredation, knowing that all parties have the same training may improve investigation quality and trust in agency depredation conclusions.
Riding’s Effectiveness Beyond Direct Predator Conflict
Although many participants expressed uncertainty regarding range riding’s effectiveness at reducing direct predator conflict, all participants acknowledged the additional, sometimes indirect benefits that riders provide to an operation. These findings align with previous work exploring how perceived responsiveness and beliefs about intervention effectiveness can matter as much as technical outcomes in wolf conflict reduction approaches like range riding (Smith et al., 2025; Martin et al., 2025; Kantor et al., 2026). Management outcomes depend on both measured and perceived benefits (Miller et al., 2016; Ostermann‐Miyashita et al., 2025). A mitigation approach or tool’s effectiveness at reducing biological losses is meaningless if producers are not inclined to adopt or maintain it (Volski et al., 2021). Understanding what about a tool or approach drives producers to adopt/maintain its use is critical, and additional/indirect benefits should not be underestimated (Volski et al., 2021).
Based on these findings, we recommend that evaluations of range riding’s effectiveness should extend beyond direct conflict metrics like depredation and injury to include broader social and operational benefits (i.e. indirect conflict losses like reduced pregnancy rates or weaning weights, increased physiological indicators of stress, producer reported satisfaction and adoption rates, and improved communication and coordination rates with wildlife management agencies and the agricultural community).
Predator Location Sharing and Trust
There were mixed views regarding the utility of predator location data for range riding. However, similar to the findings of Shawler (2025), participants noted frequently that wildlife agencies willing to share predator location information displayed goodwill and strengthened trust, especially during the early stages of predator recolonization and conflict. Trust determines whether partnerships between producers and agencies can function, including whether critical information like depredations are shared/reported, whether agency policies are seen as legitimate, and the social escalation of conflict (Young et al., 2015; Morehouse et al., 2020; Dietsch et al., 2021; Marsden et al., 2026).
We recommend that agencies communicate transparently about both predator location information and the limitations of location data. While we recognize that 1) providing predator location information daily to all operations may be unsustainable for agency staff, and 2) the inaccuracies and time delays inherent to predator location data from GPS and VHF collars introduce complications to overrelying on that data, finding creative data sharing solutions that contribute to trust-building and collaboration may benefit resource management projects beyond current predator conflict needs.
Connection to Quantitative Findings
“Range riding” as a term has popularized recently with government agencies, academics, and NGOs alike (Parks & Messmer, 2016; Bogezi et al., 2021; Louchouarn & Treves, 2023; Anderson et al., 2024; Nickerson et al., 2024; Martin et al., 2025), but range riding as a practice is as old as livestock husbandry itself. The persistence of range riding, especially among several technological advancements in other conflict mitigation approaches is significant: producers are hesitant to adopt tools they do not believe to be effective, or do not fit into their cultural and operational context (Greiner et al., 2009; Scasta et al., 2017). Despite this persistence, range riding has yet to be systematically evaluated under different ecological and operational contexts. This report provides the first transdisciplinary exploration of range riding’s effectiveness at reducing predator-livestock conflict.
While our qualitative results emphasize the perceived value of range riding, our preliminary quantitative analyses did not find meaningful associations between range riding effort (ride frequency, duration, and time spent riding at night) and reductions in total conflict (depredation and injury), or between riding effort and cattle selection for high availability forage selection, improved pregnancy rates, or improved return rates (publications forthcoming). Like other nonlethal tools/approaches, we suspect that range riding is being deployed reactively in response to high predator presence or conflict. However, we are unable to make inference on the effectiveness of range riding directly because of the lack of an experimental control group (e.g., where no mitigation was practiced). Such a control group would be unethical since it could expose livestock producers to significant livelihood losses and also jeopardize relationships between livestock producers and practitioners.
Participants in our qualitative study admitted to this very ambiguity in riding’s ability to reduce predator conflict directly, yet praised riding’s contribution to uncertainty reduction, adaptive management, data collection, and improved trust, coordination, and communication. Participant descriptions of best approaches to deploying riders, skills necessary for success, and goals they hope to achieve through riding are not just for conflict reduction, but support a holistic approach to livestock production on remote and rugged rangelands. This finding is critical as these contributions may do more for long-term predator-livestock coexistence even if direct conflict reduction is minimal. In this sense, range riding’s primary contribution may be rooted not in conflict reduction directly, but in the enhancement of resiliency and responsiveness in the broader ecological system. Evaluations that focus exclusively on biological loss metrics may overlook riding’s culturally and operationally aligned contributions to livestock production and healthy rangelands.
Study Limitations and Future Research
Because interviews were only conducted with participants in our larger evaluation on range riding’s effectiveness, our findings are limited to a specific group of producers who were willing to engage in range riding and proactive conflict mitigation approaches. This may have introduced bias and our findings may not adequately represent the diversity of operations using or interested in range riding. Our representation across states and participant roles (rider, producer, or both) was also not proportional, and may not adequately represent variations on state-level policies and ecological contexts. Future research should aim to include broader perspectives (including those from producers not currently utilizing range riding).
In addition to our interviews, the experiential knowledge of producers and riders were collected during panel discussions, breakout groups, and inter-stakeholder group discussion at our range rider workshops across the West. Findings from these workshops reflected similar findings from our interviews. Regarding what skills/traits are needed by a rider to be effective, producers stated that riders need to 1) have a great work ethic and flexible schedule, 2) be trusted by the local community, 3) listen well, be curious to learn and keen to observe, 4) know the specific terrain and livestock herd (with recognition that this takes on average 1-2 full years), 5) have livestock skills and prioritize stockmanship, not just wildlife skills, 6) have excellent wildlife track and sign skills, 7) want to come back year after year, and the most requested skill/trait, and 8) good communication skills and the ability to maintain and build trust with a diverse set of stakeholders.
Common concerns included 1) how do we keep range rider programs funded so that good riders want to come back each year, 2) how do we familiarize riders with more than one allotment so they can cover for one another more effectively, but not in a way that prevents each rider from learning their herd and allotment, 3) how do we better deploy riders by knowing where conflict is/will be concentrated, 4) whether producers should be allowed to ride their own herds as part of a funded range rider program, 5) how do we provide work for riders off season, and 6) how do we find riders with the diverse skillset required to be effective?
Collectively, participants at each workshop defined good range riding and range rider programming as 1) happy livestock, 2) good pregnancy rates, 3) healthy and sustainable range forage, 4) riders who know their allotment and their herd so well that they can recognize problems quickly, 5) shared wolf/predator location information, 6) improved relationships, and 7) improved carcass detection rates.
Aspects that participants believed would improve riding’s effectiveness included 1) access to thermal, 2) access to wolf collar data and/or VHF telemetry (mixed – some participants see access to wolf collar data as a distraction and not that useful), and 3) better property access (especially for agency riders). Both workshops agreed that how a rider should spend their time and what an “effective rider” looks like will always be context specific to each ranch/allotment, regardless of how we try to cookie-cutter define what a “range rider” is. Flexibility is key.
When asked what metrics should be used to distribute range riders across a landscape with many producers requesting riders and only so many riders (think NRCS ranking protocols/questions), participants collectively agreed that 1) history of conflict, 2) current conflict, 3) known wolf presence, and 4) rancher acceptance/willingness should be considered (not in order of importance). How many riders were needed to effectively cover what number of head/acres was a lingering question.
Participants agreed that some kind of range rider accreditation or training program was needed at a west-wide scale. This presented a challenge in both groups, as ranchers expressed concern about being passed more responsibility/burden related to carnivore conflict to “train” riders that will not be riding their allotments, while also recognizing that the best training would have to happen on allotment with the producer due to the context-dependent nature of range riding. A combination of off-site initial training through a training program followed by on-ranch training with the producer, ranch employee, or other riders was suggested. Partnering with universities, particularly Extension, was mentioned as a possible way forward for the initial accreditation/training program.
Future Research and Recommendations for Agriculture
- Rider deployment strategies should remain context-specific and adaptive rather than prescriptive.
- Livestock husbandry experience/knowledge may be the most important competency for range riders. Anyone hiring riders outside of the ranch owner/manager (e.g., government agencies or NGOs) should include producers in the process to ensure rider fit and operation specific goals and conditions are being addressed.
- Structured training for range riders could improve not only the pool of qualified riders, but also rider overall effectiveness and the ability of riders to coordinate effectively with other riders, agencies, and producers.
- Evaluations of range riding’s effectiveness should extend beyond direct conflict metrics to include the broader social and operational benefits, and effectiveness should include coordination with agencies for timely conflict mitigation (e.g., swift lethal removal when necessary).
- Agencies should communicate transparently about both predator location information and the limitations of location data. Finding creative data sharing solutions that contribute to trust-building and collaboration may benefit predator management beyond current conflict needs.
- Support the conflict mitigation approaches livestock producers advocate for - cultural and operational context, as well as uncertainty reduction, adaptive management, data collection, and improved coordination/communication are equally important effectiveness metrics for nonlethal tool adoption as effectiveness in reducing direct loss.
Lastly, in late 2023, our collective efforts resulted in 22 million dollars becoming available through NRCS for range riding efforts in Oregon, New Mexico, Colorado, Montana, and Arizona. Our broader CoW-CIG team has been working to make information available to landowners on how to qualify and apply for these funds (see Education and Outreach for more on our peer-to-peer range rider workshops).
Citations
see link below
Education and Outreach
Participation summary:
Through the duration of this grant, we measured success through progress towards and achievement of the following grant deliverables including:
- Annual structured meeting;
- Regional workshops, clinics, and/or annual webinar series;
- New podcast and video series;
- Development of a “Toolkit” for Livestock Producers; and
- Multimedia distribution to sustain and amplify these activities.
Within the following section, we offer a summary of these proposed deliverables, methods of evaluation, and a brief summary of what has been achieved to date to be expanded within the results section.
1: We will hold an annual meeting for each year of the project and a final meeting at the end of the project as opportunities to present results and engage the broad spectrum of project participants (individual ranchers, landowner collaboratives, NRCS specialist, extension specialist, USDA-WS, and state wildlife agencies). Metrics: Participation - 200 stakeholders annually.
Annual Meeting: Held meetings in 2022, 2023, 2024 and 2025 that convened diverse stakeholders, engaging the broad spectrum of participants within the effort to research the effectiveness of range riding, share resources, and identify durable cost-sharing opportunities.
2: We plan to conduct yearly regional workshops and several range riding clinics for the three years of the project. These workshops will focus on synthesizing the latest research and producer-led presentations around “lessons learned” from the field surrounding range riding practices and/or other predator conflict mitigation practices (Objective 2). Metric: Surveys, participation.
Twelve regional workshops were held by partner organizations Western Landowners Alliance and The Heart of the Rockies in Arizona, Montana, Colorado, Oregon and Wyoming focused on range riding and other nonlethal tools to reduce predator conflicts.
3: To increase the educational reach into communities, five additional tools have been noted as important components of the Extension toolbox: blogs, wikis, Facebook, YouTube, and podcasts [27]. Although the traditional forms of Extension program delivery will continue to play an important role, podcasts are poised to be one of the most effective forms of Extension education because information can be effectively distributed to global audiences without the need for in-person contact. Podcasts have a similar niche as webpages in that you only have to build it once, and the audience is limitless [26]. Metric: Building capacity and publishing/advertising season 1 of WWU
Within the first project year of this grant, the team was scheduled to release one season of the Working Wild U Podcast, as well as convene a media campaign to widely distribute the podcast. That season, ‘Wolves in the West’ was released at the end of 2022 and through the spring of 2023, with 77,113 accumulated views.
4: As we near project completion in year 3, we will synthesize results based on the evaluation of the effectiveness and costs of range riding and other predator conflict mitigation practices (Objective 1) and best management practices developed through rancher-to-rancher stakeholder learning at annual workshops (Objective 2). This synthesis will be used to design a user-friendly toolkit to guide and facilitate producer adoption of the most effective implementation approaches to predator conflict mitigation practices based on ranch-specific goals, capacities, and resource conditions. Metric: Completion of Guide.
Guided by livestock producers and other dedicated people working daily to reduce, manage, and mitigate predation risk on working wild landscapes, the range riding, carcass management and electric fencing toolkits highlight decades of experience compiled into three documents that make up the Producer Toolkit for Predation Risk Management. This Toolkit is a compilation of direct experience and knowledge of risk assessment, range riding, carcass management, and various types of electric fencing.
5: We will use multimedia to distribute information about outreach activities 1-4 to help amplify and scale outreach efforts in order to engage a larger community of predator conflict mitigation practice users and practitioners in an effort to improve the likelihood of adoption (Objectives 4). Metric: E Newsletters reaching 1,300 Subscribers, On Land Magazine Reaching 10,000 Land Stewards. Delivered through Gov Delivery Email (100,000 subscribers).
Western Landowners Alliance Working Wild Challenge quarterly newsletters highlighting stories, opportunities, and relevant news about what it means to manage working lands while sharing space with wildlife were sent to an average list of 3,500 people. Bi-annual Western Landowners Alliance On Land magazines were distributed to an average of 3,000 readers, including landowners, livestock producers, and members of the general public.
- Annual structured meeting: Metrics - Participation
CRC Annual meeting
On October 11th - 12th, 2022 the CRC meeting was convened at the Buffalo Bill Center of the West in Cody, WY. There, 30 participants representing livestock producers groups, state and federal wildlife management agencies, NRCS staff, and researchers convened to chart a path forward for the CRC maintain it’s role as a radical-center messenger, expand policy work, and continue knowledge exchange essential to sharing best practices for practices to reduce conflicts such as range riding.
Participation: 30
Qualitative Insights: When participants were asked why they keep participating in the Conflict Reduction Consortium.
- “It is one of the few meeting places that makes progress on tough issues by bringing often conflicting or different views into a well-facilitated space.”
- “I think we accomplish a great amount of collaboration and outreach with CRC, and I hope that we can keep the momentum going, and I want to be a positive driver of that movement.”
- “It's a great opportunity for shared learning. I believe it's important [for people in academia, NGOs, and agencies] to be connected to people on the ground dealing with the issues first hand, and the CRC has been a great way to do that.”
- Working together is synergistic and ties the western ranchers together. I believe that the CRC can become the most effective west wide resource for finding and implementing conflict mitigation ideas.”
- “Because it is one of the few meeting places that makes progress on tough issues by bringing often conflicting or different views into a well-facilitated space. That rarely happens well outside of the CRC. It is inspiring.”
Together with developing a 3-5 year work-plan for the CRC, this meeting re-affirmed the importance of the community of practice, trust, and collective potential for this group to further policies and practices to support wildlife-livestock conflict reduction in the West.
Convening on Collaboration and Conflict Prevention: On June 14th and 15th, 2023, more than one hundred individuals representing landowners, agricultural producers, Tribes, state and federal agencies, and nonprofit organizations from across Montana and the West gathered in Missoula, Montana, to explore solutions that would increase funding, technical assistance, and coordination to prevent conflicts between carnivores and agricultural producers, while supporting the economic viability of working lands that provide important space for wildlife.
Conflict Prevention/Coordination meeting
Participants: 105
As a result of this meeting, momentum is building to further align state and Tribal agency capacities with federal technical and financial assistance to support coordinated landowner and agency implemented conflict prevention practices to reduce conflicts between agricultural operations and wildlife for the long term.
Priority needs and opportunities highlighted by participants across the workshop, included:
- Increased coordination across partners and agencies to foster collaboration, information sharing and learning, and the most efficient use of resources. This is best accomplished by somebody who is paid to fill that role, and most likely within an agency and thus capable to work peer-to-peer with the diversity of state, federal and Tribal agencies. That said, existing forums for working across watersheds such as the Locally-Led Conflict Reduction Partnership and the Conflict Reduction Consortium are filling an important role and should continue.
- Increased public and private financial resources to lower the burden on those agricultural producers, Tribes, and locally-led partnerships who need to decrease their time fundraising for conflict prevention measures so they can increase their time implementing solutions. Sustainability of these funding resources is also essential to increase participation and succeed in achieving the long-term goals of working lands and healthy wildlife populations—both of which are critical to rural economies.
- Increased technical assistance to agricultural producers, Tribes, and community-based organizations interested in implementing conflict prevention measures. Technical assistance should include electric fencing technicians working with local landowners to secure attractants, support to set up and deliver carcass removal and composting programs, guard dog experts to inform landowners about options for using dogs to protect livestock and other attractants such as grain storage facilities, and range riders to increase monitoring of livestock on open range.
- Support from state and federal leaders as well as communities for investing in locally and Tribally-led programs. To succeed in preventing conflicts, it is critical that there is leadership and support from the top of government all the way down to individual community members and residents.
- Scientific monitoring and research to support the growing use of these tools, increase our understanding of best practices, and demonstrate success. In addition, social science research would increase our understanding of how producers and rural residents view these tools
We have received very positive feedback from landowners and producers regarding this workshop, which can be summarized via these quotes:
- “I just wanted to write you a note and thank you for the invitation to the workshop. I usually avoid that sort of thing, if I can, but must confess that I am glad that I attended. I learned some things, got a few new ideas, and got to see some people that I haven't seen in many years. If it comes around next year, I hope that I will be welcome to attend.”
- “This workshop just felt different. I can’t explain it, but I felt like everyone there was trying to address the same challenges, rather than trying to prove the other is wrong. It gives me hope that something good will come out of it.
Conflict on Working Lands Conservation Innovation Grant Annual Meeting and Celebration
Participants: 55
Envisioning new paths forward to expand upon the Conflict on Working Lands Conservation Innovation Grant (CoW-CIG), The Cow-CIG and Western SARE team gathered in Missoula Montana on February 27th and 28th 2024 to review lessons learned, celebrate successes, and brainstorm next steps to continue and expand partnerships to streamline research and delivery of predation risk management practices. 55 agency, NGO, rancher, and range rider partners attended both in person and over Zoom to provide feedback on the research/coproduction process of the CoW-CIG and Western SARE grants, education materials including the three documents that make up the producer toolkit and to help inform and design the scope of work for the funds awarded to Western Landowners Alliance (WLA) and Heart of the Rockies (part of the 22 million awarded through the RCPP), and continue building relationships/trust important for successful implementation of tools such as Range Riding.
Priority needs and opportunities highlighted by workshop participants include:
- Effective and efficient delivery of conflict prevention grants will require an increased level of coordination between all stakeholders – state and federal agencies, Tribes, producers, communities, and NGOs. This coordination is also critical to ensure that these new investments complement and are additive to the existing contributions of many supporting conflict prevention work.
- Explore and pursue additional state and federal funding sources for place- based collaborative groups providing important coordination for predation risk management practices
- Seek to simplify the process of collecting data from livestock producers, and ensure that we are using the data to solve issues faced by the landowners/support production.
- Pursue state-level coordinators where RCPP funds are implemented to build relationships and help to organize projects at the state level.
- Seek expanded funding for USDA-led conflict prevention, using the example of feral swine funding in the last Farm bill as a model for expanded investment.
Results from administered surveys indicated that attending livestock producers received tremendous value from this workshop.
- Of the seven producers queried, all indicated an overall “Excellent” rating for the meeting.
- When asked how likely they would use the some aspect of the project a participant shared: “Will continue to develop relations with folks met and work on sharing resources and knowledge, especially with on-the-ground tool implementation.”
- Further, all participants responded “strongly agree” across the board on the questions the instructor(s): Stimulated me to learn, presented information that will help me, related program content to real-life situations, stimulated me to think how to use the information, demonstrated enthusiasm, and showed respect for all persons attending the program.
Montana Conflict Reduction Consortium Annual Meeting
Participants: 42
The annual meeting of the Montana Conflict Reduction Consortium gathered members of the Western SARE team in Dillon Montana to set collective priorities for supporting the economic viability of working lands that provide habitat in Montana. The morning included a work-session to define work items for the upcoming meeting, while the afternoon offered opportunities for individuals and communities interested in forming or expanding place-based groups an opportunity to connect with landowners, livestock producers, facilitators, program managers and funders who have formed and lead place-based efforts within their own communities.
Work priorities and opportunities highlighted during the workshop include:
- Create shared understanding of core functions, program and funding needs regardless of listing status and potential changes to existing systems and programs.
- Support economic viability of working lands sharing space with grizzly bears through developing and funding a grizzly bear habitat lease
- Improve and fund depredation compensation in Montana
- Build local capacity for conflict reduction through supporting the formation of place-based collaboratives in Montana.
- Strengthen shared understanding of range riding as a conflict mitigation tool
2. Regional workshops, clinics, and/or annual webinar series: Metrics: Capacity Built.
- Colorado (35 participants): Western Landowners Alliance worked as a bridging organization to connect producers in North Park Colorado who were experiencing consistent wolf conflicts for the first time in over 100 years, with Cat Urbigkit, a rancher, writer, and range riding expert. Cat shared her experience employing game cameras for carnivore monitoring to inform her grazing rotation patterning as well as to inform when and where to apply predator deterrents.
- Wyoming (30 participants): Comprising one portion of the CRC annual meeting, workshop attendees visited a ranch in the nearby South Fork of the Shoshone, and learned about problems and potential solutions for carnivore conflicts from the perspective of the ranch manager. Further, a panel highlighted landowner perspectives from managers and livestock producers in the greater Cody area, some of whom shared challenges and successes of managing range riding operations.
- Montana (105 Participants): During the convening on collaboration and conflict prevention, two panel sessions highlighted the role landowner and agricultural producer-led organizations are playing in conflict prevention while another session focused on Tribal and agency conflict prevention work, with an emphasis on existing and upcoming opportunities for increased involvement and investments. All three panels showcased the broad suite of partners that are working together to address carnivore conflict challenges through shared learning and implementing effective practices. Participants in the first panel discussed creative partnerships that have formed to reduce carnivore access to attractants, including through carcass pickup, electric fencing and mats, and bear-resistant garbage programs. Speakers in the second panel session highlighted traditional practices such as range riding and guard dogs that are being used to reduce conflicts on open range, spoke to the challenges of both starting and sustaining these practices, and raised the importance of working with producers to identify and address research needs.
- Arizona (66 participants): The Western Landowners Alliance and the Farm Bureau hosted a producer-only meeting in Springerville, Arizona focussed to identify problems as it relates to public lands grazing management and Mexican wolf-livestock conflicts. The producers quickly coalesced around a vision of establishing an expansive range riding program (10-20 range riders) throughout the Gila National Forest supported by a NRCS Regional Conservation Partnership Program. The application for this grant has been submitted and the project team will be hearing back about this opportunity in December. If secured, there will be a direct need for a series of range riding trainings and workshops in the region.
- Colorado (180 participants): Together with the Holy Cross Cattlemen's’ Association and partners, WLA hosted an evening of conversations and presentations in Rifle, Colorado, to help producers prepare for impending wolf reintroduction. This meeting, attended by over 180 livestock producers, offered an opportunity to connect with a panel of local livestock producers, learn from Northern Rockies land stewards experienced with living and working with wolves, and engage state and federal wildlife managers charged with managing wolves and reducing conflicts with livestock.
- Wyoming (50 participants) In November, WLA hosted a landowner resource event in Pinedale, Wyoming, to raise awareness about the importance of working lands and the tools that support them. Gathering around 50 people, the event featured a landowner panel where participants shared the challenges of maintaining working lands while reducing conflicts with carnivores and their experiences using various tools and programs. Following the landowner panel, agencies and WLA provided short presentations about the tools available through the USDA/WY Big Game Conservation Partnership, upcoming resource opportunities for conflict prevention, and invasive weed treatment opportunities.
- Oregon (60 participants) and Arizona (82 participants) peer to peer learning workshops: In partnership with Western Landowners Alliance, Ph.D. student Rae Nickerson organized and facilitated two peer-to-peer workshops in La Grande, Oregon and Eager, Arizona during autumn 2024 (education and outreach goal 1). Rae brought a team of producer and rider panelists from her research to each workshop. Day one of these workshops included presentations and discussion on important range rider related content including a presentation from NRCS staff on how the new RCPP funds will be available and distributed. Day two was a wildlife track and sign certification opportunity offered to producers and range riders for free in partnership with Cyber Tracker North America. SARE surveys were collected by Western Landowners Alliance at both workshops. Our Range Rider Toolkit was shared broadly at all workshops.
- Colorado January (128 participants) and April (65 participants) Range Riding Training: Rae provided training for range riders in Colorado in partnership with Western Landowners Alliance, Colorado Parks and Wildlife and Colorado Department of Agriculture. The training, drawing on Nickerson’s West- wide study on range riding effectiveness and featuring panels with experts, disseminated skills in risk evaluation, decision-making, communication and track and sign interpretation. The January one day training, which was open to the public, drew significant attendance at 128 range riders, producers and agency representatives. The April training was convened specifically to train Colorado Parks and Wildlife’s and Colorado Department of Agriculture’s newly hired range riders, and on top of the classroom portion included a two-day field evaluation focused on track and sign skills. The Range Riding Toolkit was shared broadly.
- Washington (21 participants) Track and Sign Evaluations: In May 2025, Rae organized and hosted two additional and free wildlife track and sign certification opportunities for riders and producers in northeast Washington.
- Montana (36 participants) Range Riding Training: In November 2025, Rae organized and facilitated a peer to peer learning workshop in partnership with Western Landowners Alliance in Dillon, Montana. Rae brought a team of producer and rider panelists from her research. Day one of this workshop included presentations and discussion on important range rider related content Day two was a wildlife track and sign certification opportunity offered to producers and range riders for free in partnership with Cyber Tracker North America. The Range Rider Toolkit was shared broadly.
Arizona and Oregon Peer-to-Peer learning workshop evaluations
SARE surveys were collected at both of our peer-to-peer range rider workshops in La Grande, Oregon and Eager, Arizona last fall. Out of n=53 total survey respondents, 100% agreed that the event improved their awareness of the topics covered. Ninety-six percent agreed that the event provided new knowledge and was current, up to date, and relevant, and 70% that the event provided new skills. Seventy-two percent of those surveyed believed that the event modified their opinions and attitudes.
Seventy-two percent of respondents strongly agreed that the instructors presented information that would help them, and between 80 and 94% of respondents strongly agreed that the instructors stimulated them to learn, related program content to real-life situations, stimulated them to think about how to use the information presented, allowed for questions and interaction with participants, demonstrated enthusiasm, answered questions clearly, and showed respect for all persons attending the program.
Overall, the events received an 81% excellent rating with a 19% good rating. Respondents together reported that they will share what they learned from the workshops with at least 852 new people. Some of the reported “best” things about the sessions included (1) getting to know the different organizations and stakeholders in person, especially hearing their diverse experience, knowledge, and ideas, (2) the culture of respect, honesty, and team work maintained at the workshop through hard conversations and different perspectives/points of view, (3) having all agencies present with a stake in the game, (4) community development, (5) the panel discussions and break-out discussions, (6) provided hope and optimism that there can be effective solutions to the current conflict, (7) learning about new NRCS and other funding options for range riding and conflict, (8) networking with fellow producers and/or range riders, and (9) the collaboration and relationship building.
Constructive feedback for future events included (1) choosing a less busy time of year for producers, (2) to include more information on when range riding doesn’t work (under what contexts), and (3) to provide more workshops like these that are longer with more time for panel discussion, mingling, and questions.
Of the producers and landowners who took the survey (n=29), 76% said they were likely to adopt one or more of the practices mentioned, 56% said doing so would improve their operations diversification of strategies, 30% reported that using the practice would reduce the need for off farm purchasing (less relevant to range riding), 79% reported some aspect of the range riding would increase networking with other producers, and 63% reported that range riding would add value into their overall operation.
Of the professionals and practitioners who took our survey (n=40), 80% reported that they would use some aspect of the training educationally or as a participant in the future, 86% said they would participate in making range riding an available resource, 74% reported they would use some aspect of the workshop as a professional development tool for their peers, and 95% reported that they would use some aspect of the workshop to improve the advice/counsel they provide to producers. When asked to describe how they are likely to use some aspect of these workshops for an educational purpose, the professional and practitioner responses included (1) at their job with the USFS, (2) with their peers and encourage participation at future events by other producers, (3) letting people know specifically about what resources are available for conflict mitigation, and (4) helping to educate non-ranchers on the intricacies of wolf-livestock conflict.
Colorado Range Riding Trainings
Western SARE surveys were adapted to include questions important to co-hosts Colorado Parks and Wildlife and Colorado Department of Agriculture. Of N=59 responses, 74.6% rated the workshop as “excellent”, and 23.7% rated it as “good”. Notably, 100% of respondents shared that the workshop improved their awareness of the topic covered, while 98% indicated that it provided new knowledge. Seventy six percent of respondents believed that the workshop modified their opinion and attitudes with 91.5% noted it provided new skills.
Some of the reported “most helpful” things about the sessions included, (1) the knowledge and adeptness of the moderator, Rae, (2) the attack signatures of different predators, (3) low-stress stockmanship, (4) discussions with the panelists, (5) and seeing real world examples of how weather, age and external factors impact animal track and sign. The “best” aspects of the meeting were, (1) the instructors, (2) the panel's conversations, and (3) the track and sign evaluation. Some respondents highlighted “all aspects” of the meeting as being valuable. Highlighting the impact of this material. Of N=53 that responded, all indicated that the material was conveyed clearly and in a way that made sense to them.
3. New podcast and video series; Metrics: Developed capacity and publishing and dissemination of season 1.
Listener survey data related to Season 1 of Working Wild U: Wolves in the West shows that we provided significant value to our target audiences: ranchers, wildlife managers, and urban wildlife enthusiasts, informing the discussion around wolves. The following metrics indicate the success of the first season
- The podcast achieved over 35,000 downloads to date in all 50 US states, with top states being CO, MT, CA, and WA, surpassing listenership goals
- 91% of people surveyed said the show improved their awareness of the topics covered and 89% of people surveyed said the show provided them with new knowledge.
- 83% of natural resource professionals and practitioners surveyed said they intend to use some aspect of this project as an educational resource and when advising others on this issue.
- 60% of people surveyed said the show modified their opinions and/or attitudes around these controversial topics and 44% of people surveyed said the show provided them with new skills to address similar issues.
- Multiple reporters and producers covering wolves, such as Kylie Mohr for the Deseret News, a film producer working on a documentary about wolves in the West, shared that they listened to the entire season as background for their coverage of the issue, which demonstrates the value of the show's radical center perspective on what is too often framed as a polarizing, lose-lose issue.
- We received the Gold Award from the Association of Natural Resource Professionals (ANRP), Podcast or Radio Category as part of the The Natural Resources University (NRU) Podcast Network.
- We also earned 74 five star reviews on Apple Podcasts (79 total reviews, 4.8/5 stars)
4. Development of a “Toolkit” for Livestock Producers;
The primary pathway for integrating the collective experience and knowledge of partners in this project was through the producer toolkit for conflict reduction. Guided by livestock producers and other dedicated people working on a daily basis to reduce, manage and mitigate predation risk on working wild landscapes, the toolkit highlights decades of experience compiled into three documents and is a compilation of direct experience and knowledge of risk assessment, range riding, carcass management, and various types of electric fencing.
Developed through Technical Advisory Committees, these documents were co-produced amongst livestock producers, researchers, and non-profit and state agency representatives. Each document provides a comprehensive overview of range riding, carcass management, and fencing, and conveys context specific application through the risk assessment framework as well as the six principles of predation risk management mentioned above. Of note, the Range Riding Toolkit provides a concise definition of the practice, (see Figure 2), as well as best practices for implementation– two important contributions that have not been well described in other producer-facing documents prior. The Carcass Management toolkit provides a novel contribution in categorizing and describing the four phases of carcass management: 1. Finding and securing a carcass; 2. Temporary or permanent ranch facility; 3. Transportation; and 4. Community carcass management facility. The Fencing Toolkit offers an overview of four widely used types of fence – turbo-fladry, electric night pens, 4 and 5-wire fences, and electric drive over “unwelcome mats” – as well as information to guide their context specific implementation. The documents also include case studies highlighting on-the ground application of each practice in different contexts throughout the West. In order to disseminate information and highlight the producers whose knowledge led the way for forming this document, the project team hosted a webinar attended by over 200 individuals from all seven states within the project area.
An addition, place-based collaborative groups provide a way to coordinate community-scale action to address wildlife-livestock conflicts, and processes to lift landowner and livestock producer needs, while finding areas of agreement and shared purpose to meet a variety of resource challenges. These groups, many of which are landowner-led, may include all or part of a particular community and offer a way to meaningfully engage state and federal wildlife agencies, non-profit organizations and other stakeholder groups within a community-level decision-making process. Informed by interviews and co-produced with existing place-based groups, This team developed the “Road-Map to Place Based Collaboration for Conflict Reduction”, a hands-on guide for developing landowner-led, place-based collaborative groups with a focus on reducing wildlife-livestock conflicts. While the regulatory context, stakeholders, wildlife, and landscape will vary, this 10-step guide outlines a process and provides examples to aid landowners and practitioners in developing community-led solutions to address wildlife-livestock conflicts. Four case studies provide on-the-ground examples of how place-based collaborative groups have formed and organized to address conflicts and support landowner and wildlife needs.
5. Multimedia distribution to sustain and amplify these activities.
Throughout the past years, we have engaged in communications and media campaigns to increase the profile of our education and outreach efforts through newsletters, print Magazines, and photos and videos communicated through social media.
Newsletters
- Western Landowners Alliance Working Wild Challenge newsletters highlighting stories, opportunities, and relevant news about what it means to manage working lands while sharing space with wildlife were sent to an average list of 3,500 people, with an average open rate of 44%. The industry standard open rate is only 28.5%, according to Google.
On Land Magazines
- The Western Landowners Alliance On Land magazine shares stories of land stewardship across the West and reaches over 3,000 individuals through subscription and store sales. This reach is furthered by On-land online
Working Wild University Social Media
- All social content for WWU S1 was designed to spark curiosity, directing viewers to listen to the full episode. By maintaining a regular cadence as episodes were being released, we were able to maintain an ongoing social buzz around the show which helped drive our listenership.
- WWU IG reels alone amassed 77,337 views in the past year, plus another 5,125 views on WWU-related posts on WLA IG
- WWU IG grew from 0 to 4,703 followers since 2023
- WWU TikTok amassed 11,881 views in the past year
Education and Outreach Products and Activities
Podcasts
- Working Wild University Podcast
- 13 episodes and 2 bonus episodes included within Working Wild U Season 1
- Standouts
- 13 episodes and 2 bonus episodes included within Working Wild U Season 1
- Working Wild U Season 1 - Wolves in the West
- Working Wild U - Episode 11 - Old World Tools to New World Technology
- Place-Based Collaboratives and Conflict Reduction with Matt Collins
Zoom Meetings
- 26 Conflict reduction Consortium Zoom meetings
- Monthly Conflict Reduction Consortium meetings created opportunities for knowledge exchange and galvanized opportunities for collective action for expanding funding opportunities for range riding.
- 28 Practitioners Calls
- 12 Technical Advisory Committee Meetings
Conveneings
- 16 in person meetings (Colorado, Wyoming, Montana, Arizona, Oregon, Washington)
Conflict on Working Lands outreach products
- CIG project summary
- CIG research summary
- Range Riding Toolkit
- Carcass Management toolkit
- Fencing Toolkit
- Road-Map to Place Based Collaboration for Conflict Reduction
Journal Articles
- Hyde, M., Breck, S. W., Few, A., Beaver, J., Schrecengost, J., Stone, J., ... & Young, J. K. (2022). Multidisciplinary engagement for fencing research informs efficacy and rancher-to-researcher knowledge exchange. Frontiers in Conservation Science, 3, 938054.
- Gonzalez, M., Wilson, S., Nickerson, R., Hyde, M., Purdy, C., Young, J., Breck, S., Crooks, K. (2024). Assessment of the Northeast Washington Wolf-Livestock Management Program – 2021 – 2023 Seasons. [Final report].
Note: while not directly supported with funding by the WSARE, both articles were an outcome of our more broadly funded research and support our Student SARE Education and Outreach goals. Research articles from this project are ongoing and will be submitted as analyses are completed.
Earned Media
- Wolf Monitoring that Works For Ranchers
- Seeing Red: Montana Ranchers and The Line Between Conflict and Coexistence
- Montana State University Extension launches podcast on wolf reintroduction, ranching
- A new wolf-focused podcast wants to create a mutual understanding of wildlife and working land issues
- Composting a Recipe for Conflict Reduction
- The Balancing Act: Public Wildlife on Private Lands
- Save Ranching and Wildlife - Invest in Relationships
- Research Roundup: Is An Ounce of Prevention Worth a Pound of Cure?
- Western Landowners Welcome Historic USDA Working Lands Investment
- An official Grizzly In the Big Hole
- 22 million to help ranchers steward habitat and reduce conflicts with large carnivores
Stories
- Surprising benefits of range riding at Alderspring Ranch
- Acting odd on the range to change carnivore behavior
- Wolf Monitoring that Works For Ranchers
- Western Landowners Welcome Historic USDA Working Lands Investment
- 22 million to help ranchers steward habitat and reduce conflicts with large carnivores
Webinars, talks and presentations
Webinars
- Conflict on Working Lands Classic CIG Grant Review Webinar
- Carcass management
- Range Riding
- Fencing/Fladry
- Producer Toolkit Webinar
Presentations
- Collins, M. and Breck, S.W. 2024. Reducing Risk on the Range: Co-Production and Cost Sharing for Managing Carnivore-Livestock Conflicts. Colorado Parks and Wildlife Annual Meeting.
- Collins, M. 2022 Place-Based Collaboratives for Minimizing Human-Carnivore Conflict: Collective Factors Driving Success Throughout the American West. International Wolf Symposium.
- Collins, M. 2023 Place-Based Collaboratives for Minimizing Human-Carnivore Conflict: Collective Factors Driving Success Throughout the American West. The Wildlife Society Colorado Chapter Annual Meeting.
- Nickerson, R. 2024. Effective Range Riding Workshop. Conflict on Working Lands Conservation Innovation Grant team. La Grande, Oregon.
- Nickerson, R. 2024. Effective Range Riding Workshop. Conflict on Working Lands Conservation Innovation Grant team. Eager, Arizona.
- Nickerson, R. 2025. Range Rider Training. Colorado Department of Agriculture and Colorado Parks and Wildlife.
- Nickerson, R. 2025. Evaluating the Effectiveness of Range Riding at Reducing Conflicts with Large Carnivores. Ecology Center Graduate Student Research Symposium. Utah State University.
- Nickerson, R. 2025. Evaluating the Effectiveness of Range Riding at Reducing Conflicts with Large Carnivores. The Wildlife Society International Conference. Alberta, Canada.
- Nickerson, R. 2024. Effective Range Riding Workshop. Conflict on Working Lands Conservation Innovation Grant team. Dillon, Montana.
Workshop / field days
- Colorado North Park Stockgrowers Meeting (35 participants):
- Wyoming Conflict Reduction Consortium annual meeting (30 participants)
- Montana Collaboration and Conflict Prevention Meeting (105 participants)
- Arizona Range Riding Meeting (66 participants)
- Oregon Range Riding Knowledge Exchange (60 participants)
- Eager Arizona Range Riding Knowledge Exchange (82 participants)
- Colorado Preparing for Wolves Meeting (180 participants)
- Wyoming Landowner Resources Event (50 participants)
- Colorado One Day Range Riding Training (128 participants)
- Colorado Four Day Range Riding Training (65 participants)
- Washington Track and Sign Evaluations (21 participants)
- Montana Range Riding Training (36 participants)
Other Educational Activities
Participants
- ~ 100,000 Including WWU listeners and Social Media Views
Number of farmers/ranchers who participated in education and outreach activities
- ~5000 including WWU listeners
Number of other agricultural stakeholders who participated in educational activities
- ~10,000 including WWU listeners
Number of farmers who intend/plan to change their practice(s)
- A letter to NRCS indicating producer interest in establishing farm-bill funding for range-riding demonstrates that 151 producers from around the West are interested in practice adoption.
Number of farmers/ranchers who changed or adopted a practice, if applicable
- Undetermined. With cost-sharing availability for range riding, carcass management, and fencing through NRCS EQIP on the horizon, there will be widespread opportunity for novel practice adoption across the West. This is a significant accomplishment, with wi
Education and Outreach Outcomes
Through the first stages of this project we have learned the following lessons:
- Building communities of practice amongst diverse stakeholders in conflict reduction can help support information exchange
- Engagement of broad networks with effective, science and land-stewards centered communications can support increased knowledge of range riding and it’s application, and support cross-pollination of ideas within closet networks, building momentum for practice implementation.
- Landowners and livestock producers maintain knowledge of the land and stewardship practices that are not often captured in scientific research, or elevated for peer-to-peer learning. Incorporating this knowledge is both important for representative applied science, and for diffusion and implementation of practices such as range riding.
Lastly, success leads to success, using engaged producers that part of the project design from start to finish recruits more producers.
- Identification of predators sign and activity and ways to adopt to this
- Benefits of non-lethal conflict prevention tools for wildlife, livestock, and rangeland production in the presence of large carnivores
- Risk assessment and landscape stratification
- Conflict management and the Conservation Planning Process
- What does Range Riding encompass and is it right for you
- How predator species, native prey base, and other factors are influenced by range riding
- Best practices, considerations, and success stories of using Range Riding
Potential role and opportunity for NRCS in non-lethal large predator-livestock conflict prevention
Existing and current research on how these non-lethal tools benefit wildlife and livestock
Benefits of non-lethal conflict prevention tools for wildlife, livestock, and rangeland production in the presence of large carnivores
Conflict management and the Conservation Planning Process
Non-lethal tools and considerations surrounding mitigating conflict with wolves
Identification of predator sign and activity