https://cpw.cvlcollections.org/files/original/57c812f2113a06dc147a659f4dfa230f.pdf 2273335e022db9703870dff6bbdeb277 PDF Text Text Colorado Division of Parks and Wildlife September 2013-September 2014 WILDLIFE RESEARCH REPORT State of: Cost Center: Work Package: Task No.: Colorado 3420 0660 N/A Federal Aid Project No. N/A : : : : Division of Parks and Wildlife Avian Research Greater Sage-grouse Conservation Using GPS satellite transmitters to estimate survival, detectability on leks, lek attendance, interlek movements, and breeding season habitat use of male greater sage-grouse in northwestern Colorado Period Covered: September 1, 2013 – August 31, 2014 Author: B. L. Walker Personnel: B. Holmes, B. Petch, B. deVergie All information in this report is preliminary and subject to further evaluation. Information MAY NOT BE PUBLISHED OR QUOTED without permission of the author. Manipulation of these data beyond that contained in this report is discouraged. ABSTRACT Implementing effective monitoring and mitigation strategies is crucial for conserving populations of sensitive wildlife species. Concern over the status of greater sage-grouse (Centrocercus urophasianus) populations has increased both range-wide and in Colorado due to historical population declines, range contraction, continued loss and degradation of sagebrush habitat, and recent federal listing of the species as warranted but precluded under the Endangered Species Act in 2010. Despite untested assumptions, lekcount data continue to be widely used as an index of abundance by state and federal agencies to monitor sage-grouse populations. Lek locations are also commonly used to identify and protect important sagegrouse habitat. However, the use of lek counts and lek locations to monitor and manage sage-grouse populations remains controversial because it is unknown how closely lek-count data track actual changes in male abundance from year to year, or if lek buffers are effective at reducing disturbance to male sagegrouse and their habitat during the breeding season. Colorado Parks and Wildlife deployed solar-powered GPS transmitters on male greater sage-grouse and conducted double-observer counts and resighting at leks to obtain data on male survival, lek attendance, inter-lek movements, detectability, and diurnal and nocturnal habitat use around leks during the breeding season in and near the Hiawatha Regional Energy Development project area in northwestern Colorado in spring 2011-2014. These data will allow us to evaluate whether current lek-based monitoring methods provide reliable information about sage-grouse population trends and lek buffer sizes effective for conserving greater sage-grouse. We continued monitoring 37 males with GPS transmitters from Sep 2013 - Jun 2014 to obtain an additional year of data on habitat use, lek attendance, and within and between-year inter-lek movements and lek fidelity. 1 �COLORADO PARKS AND WILDLIFE RESEARCH REPORT USING GPS SATELLITE TRANSMITTERS TO ESTIMATE SURVIVAL, DETECTABILITY ON LEKS, LEK ATTENDANCE, INTER-LEK MOVEMENTS, AND BREEDING SEASON HABITAT USE OF MALE GREATER SAGE-GROUSE IN NORTHWESTERN COLORADO BRETT L. WALKER PROJECT OBJECTIVES 1. Test the effect of GPS transmitters on male greater sage-grouse: a. Estimate and compare seasonal and annual survival rates of yearling and adult male greater sage-grouse with GPS transmitters to published and empirical estimates for leg-banded males. b. Compare fitted leg-loop size for adult vs. yearling males to assess whether yearling males will outgrow harnesses; if needed, recapture yearling males and refit harnesses in the field. c. Compare strutting display rates between GPS males and color-banded or unmarked males. d. Compare raw lek attendance rates between GPS males and color-banded males. 2. Use locations of GPS males to locate, verify, and count new leks in and around the study area. 3. Estimate the number of known and unknown leks in the study area 4. Use unreconciled double-observer lek counts and time-to-detection models with lek-count and resighting data to estimate detectability of males attending leks. 5. Develop a modified sightability model approach to estimate daily, seasonal, and annual variation in male lek attendance. 6. Use movements of GPS males to determine presence near leks in the study area and to estimate the frequency, timing, and distance of breeding-season movement among leks. 7. Estimate daily and breeding-season survival rates of GPS males. 8. Use simulations to quantify how variation in age-specific male survival, presence, detectability, lek attendance, movement, and count frequency affect lek count indices and trend estimation. 9. If possible, use mark-resight data and counts of marked and unmarked males and females at leks to generate annual estimates of age- and sex-specific population size. 10. Quantify male habitat use and movement around leks to test the effectiveness of current oil and gas lease stipulations for lek buffers. SEGMENT OBJECTIVES 1. Monitor remaining marked adult and yearling males through the end of the 2014 breeding season. 2. Compile and verify 2011-2014 data, and analyze breeding-season habitat use data INTRODUCTION Greater sage-grouse (Centrocercus urophasianus) are a conservation concern due to historical population declines, range contraction, and recent federal listing of the species as warranted but precluded under the Endangered Species Act (Connelly et al. 2004, Schroeder et al. 2004, USFWS 2010). The species continues to be threatened by ongoing anthropogenic and ecological changes to their habitat, including residential housing development, wildfire, invasive plants, pinyon-juniper encroachment, West Nile virus, agricultural conversion, and energy development (Connelly et al. 2004, CGSSC 2008, USFWS 2010). Accurately monitoring changes in sage-grouse abundance is crucial for assessing the current conservation status of populations, for quantifying responses of populations to potential stressors, and for documenting success or failure of conservation and mitigation efforts. Management strategies to protect sage-grouse habitat must also be validated to ensure they are effective at preventing unwanted impacts to populations. 2 �Greater sage-grouse populations are typically monitored and managed using data collected at leks. Each spring, male sage-grouse congregate on traditional mating grounds, or leks, to display and mate with females (Schroeder et al. 1999). Males attending leks are then counted by observers on the ground or from aircraft following standardized protocols (Jenni and Hartzler 1978, Beck and Braun 1980, Autenrieth et al. 1982, Connelly et al. 2000). Lek counts are the primary index used by all state wildlife agencies in the western U.S., including the Colorado Division of Wildlife, to monitor sage-grouse population trends (Connelly et al. 2004, CGSSC 2008, WAFWA 2008). Changes in lek size and lek persistence derived from lek count data are also used to investigate how regional and range-wide populations respond to changes in habitat and to anthropogenic stressors such as oil and gas development (e.g., Braun et al. 2002, Walker et al. 2007, Aldridge et al. 2008, Doherty at el. 2010b, Harju et al. 2010, Tack 2010). Lek locations are also used to help identify and protect important habitat in land-use planning efforts because leks are typically centrally located within breeding areas (Gibson 1996, Doherty et al. 2010c). For example, federal oil and gas lease stipulations include timing and surface occupancy restrictions on oil and gas development within specific distance buffers around sage-grouse leks to minimize disturbance to males and their habitat during the breeding season. Many state and regional “core areas” have been delineated by placing buffers around known leks that meet male count and lek density criteria (e.g., CGSSC 2008, Doherty et al. 2010a, Hagen 2010, State of Wyoming 2010). The importance of accurate and effective monitoring and management strategies is heightened in areas slated for energy development. A major threat factor in the listing decision was expanding energy development in the eastern portion of the range (USFWS 2010). Accumulated evidence suggests that sage-grouse populations show substantial declines following oil and gas development, even when standard mitigation measures are implemented (e.g., Holloran 2005, Walker et al. 2007, Doherty et al. 2008, Harju et al. 2010, Holloran et al. 2010). However, measured population responses to oil and gas development, while consistently negative, are not always of the same magnitude due to variation in: (a) the intensity of development; (b) the type of infrastructure required to develop the resource, which in turn affects the ecological processes by which impacts occur; (c) lag times between development and detection of impacts; (d) inherent differences in habitat quantity and configuration among populations subject to development; and (e) extent of overlap between development and important seasonal habitats (Harju et al. 2010). These same factors have also lead to the suggestion that it may not be appropriate to apply a onesize-fits-all protective buffer around leks based on range-wide data to local populations (Harju et al. 2010). Uncertainty about how quickly and how much sage-grouse populations will decline in response to development, and about the size of lek buffers required to minimize impacts on populations, creates potential for conflict among agencies, industry, and other stakeholders and underscores the need to test, validate, and implement scientifically defensible strategies for monitoring and managing populations in portions of greater sage-grouse range that overlap with planned energy development. Lek-based Monitoring Lek-based monitoring and management strategies are also subject to empirical criticisms and require additional research to understand their uses and limitations (Applegate 2000). Using lek-count data as an index of population size has been called into question because the quantitative relationship between lek counts and actual population size has never been established (Beck and Braun 1980; Applegate 2000; Walsh 2002; Walsh et al. 2004, 2010). The probability of detecting an individual male during a lek count (p) is the product of: (1) the probability that a male is alive (survival, palive); (2) the probability of the male being present in the survey area, given that it is alive (presence, ppresent); (3) the probability of the male attending the lek, given that it is alive and present (availability, pavail); (4) the probability of detecting the male, given that it is alive, present, and attended the lek (detectability, pdetect); and (5) the probability that the lek is counted (count probability, pcount), such that: p = palive * ppresent * pavail * pdetect * pcount (Walsh et al. 2004, Alldredge et al. 2007, Riddle et al. 2010). To understand the quantitative relationship between lek counts and male population size and to quantify how that relationship changes 3 �on an annual basis, we need daily and annual estimates of the proportion of males alive over the course of the breeding season, the proportion of those males present in the study area, the proportion of males attending leks, the proportion of males detected on lek counts, and the probability that the lek is counted, which depends on count effort. At present, too few quantitative data are available to estimate survival, presence, lek attendance, and detectability for male greater sage-grouse during the breeding season. No published studies have quantified how much annual variation occurs in the proportion of males detected or how much detectability varies among observers or with male age, weather, the observer’s distance from lek, equipment used (binoculars vs. spotting scopes), or count method (e.g., ground vs. aerial counts) (Connelly et al. 2003, Walsh et al. 2004). Male lek attendance is known to vary with age, time of day relative to sunrise, date, weather, snow depth, renesting by females, predator activity, and human disturbance, but standardization of lek-count protocols only minimizes variation associated with some of these variables (Patterson 1952, Dalke et al. 1963, Rogers 1964, Hartzler 1972, Jenni and Hartzler 1978, Beck and Braun 1980, Autenrieth et al. 1982, Emmons and Braun 1984, Ellis 1984, Dunn and Braun 1985, Connelly et al. 2000, Connelly et al. 2003, Boyko et al. 2004, Walsh et al. 2004). Past studies that have addressed male lek attendance also did not collect or report data in a consistent fashion, making generalization across studies difficult (Walsh 2002, Walsh et al. 2004). In the most rigorous studies on lek attendance for greater sage-grouse to date, Walsh et al. (2004, 2010) emphasized the importance of individual heterogeneity, age, sex, time of day, and date, but because their data on lek attendance of greater sage-grouse came from only one year in one population, they concluded that additional research was needed to quantify annual variation in lek attendance. Age-specific inter-lek movements by males have been reported in several studies, with 4-50% of males known to have attended more than one known lek during a single breeding season (Dalke et al. 1963, Gill 1965, Wallestad and Schladweiler 1974, Emmons and Braun 1984, Dunn and Braun 1985, Bradbury et al. 1989, Walsh et al. 2004), but the effect of inter-lek movements on lek count data has not been quantified. Several other factors that influence lek-count data have also never been addressed quantitatively including: disturbance by observers, predator activity (Ellis 1984), disturbance from activities associated with energy development (Braun et al. 2002), and annual variation in female attendance associated with renesting effort (Dalke et al. 1963, Eng 1963 in Walsh 2010). Methodological considerations may also affect counts. Non-random access to leks due to logistical constraints (e.g., road conditions, landowner permission) could bias population estimates derived from count data if access is correlated positively or negatively with attendance or abundance (e.g., if attendance is lower near roads). The number of counts conducted per breeding season can also influence lek-count data. Some states only record the maximum count of males at each lek in state-wide databases, and the maximum count is likely to be higher with more counts because any given count is more likely to coincide with peak attendance (Walsh et al. 2004). Despite these shortfalls, lek-count data continue to be widely used. Large decreases in lek counts or disappearance of leks over large areas over time are thought to reliably indicate population decline or range contraction (Walker et al. 2007, Aldridge et al. 2008, Doherty et al. 2010b, Harju et al. 2010). The fact that core areas have been established based largely on counts of males on leks and lek density also suggest that state and federal agencies still consider higher lek counts, on average, to represent larger populations (CGSSC 2008, Doherty et al. 2010a, Hagen 2010). This raises an important, but unresolved question. How big of a change or difference in lek counts is required to confidently and reliably infer an actual change or difference in male population size? Investigating these questions and assessing the reliability of lek-count data collected using current, standard protocols for measuring changes in actual population size over time has been identified as a range-wide research priority (Naugle and Walker 2007). There are two main options for resolving these issues. First, mark-recapture or mark-resight models using either marked birds or genetic data could be used to generate annual population estimates 4 �(Lukacs and Burnham 2005, Walsh et al. 2010), and over time, these estimates could be compared to maximum counts of males on leks to better understand the relationship between the two metrics. Using mark-resight approaches is probably the most rigorous way for generating defensible point population estimates (Clifton and Krementz 2006, Walsh et al. 2010), but they are generally too costly, and too time and resource-intensive to implement over large areas or on an annual basis (Walsh 2002, Walsh et al. 2004, Clifton and Krementz 2006, Walsh et al. 2010). A cheaper, easier method would be preferable for long-term monitoring. Another alternative would be to estimate survival, presence, lek attendance, and detectability from field data in relation to measured ecological and methodological variables, then correct lek count data to obtain annual population estimates and measures of precision. Double-observer approaches, originally developed for use with songbird point counts (Nichols et al. 2000), have recently been modified to use raw count data from independent observers to estimate detectability of males on lek counts (Riddle et al. 2010). Time-to-detection models can also be used to estimate the effects of individual-, group-, survey- and time-specific covariates on detectability (Alldredge et al. 2007). In addition, sightability models have been widely used with other species to estimate the effects of covariates on detection probability and to generate corrected population estimates from annual count data (e.g., Samuel et al. 1987, Rice et al. 2008, Walsh et al. 2009). Such models can be modified for use with lekking species to estimate the probability that individual males attend leks as a function of ecological and methodological covariates that can be measured or recorded in the field (Walsh et al. 2004). Intensive monitoring of individuals with transmitters in the field can be used to calculate daily probability of survival, presence, and lek attendance during the breeding season. Simulations would also be valuable for exploring the consequences of variation in survival, presence, lek attendance, detectability, and count probability on lek-count data. Most lek-count data are currently collected according to standardized protocols, but it may be that directing biologists to collect one or two more key covariates (e.g., distance from lek, type of optics used) would increase precision of population estimates without increasing cost. Even after following standard count protocols, it may still be beneficial, in terms of the precision of population and trend estimates, to collect and correct count data for weather, time of day, and count method using a modified sightability model. Moreover, not all variables known to influence detectability and lek attendance can be measured when collecting annual lek count data (e.g., inter-lek movement). It would be informative to use simulations based on empirical field data to quantify and illustrate how much lek-count data are likely to vary when I either do not correct for measurable covariates, cannot correct for unmeasured covariates, or both, even in the absence of actual population change. Simulations have been successfully used with other species to assess the effects of unmeasured sources of variation on count data, estimated abundance, and estimated population trends (e.g., Rice et al. 2008). Lek-based Management Lek-based management strategies are also subject to criticism. First, such strategies incorrectly assume that all lek locations are known. Several states, including Colorado, have used a combination of known lek locations, male counts at those leks, and vegetation layers to delineate priority areas for conservation of sage-grouse (e.g., “core areas”; CGSSC 2008, NRCS 2009, Doherty et al. 2010a, Hagen 2010). Each analysis used slightly different criteria and methodologies, but each assumed that all lek locations were known. This assumption is clearly violated. New leks are discovered annually, particularly in more remote portions of the species’ range where surveying is more difficult. The number of known leks monitored range-wide increased 10-fold between 1965 and 2007 due to the discovery of new leks, and the majority of leks currently monitored were discovered after 1994 (WAFWA 2008). For that reason, current core areas are geographically biased toward areas with greater survey effort (which are typically areas closer to population centers and with easier access) and the extent of core areas is most likely underestimated. Moreover, lek-based oil and gas lease stipulations can only be applied to leks 5 �whose locations are known. In combination, the presence of unknown leks and underestimation of core areas could lead to inadequate levels of protection in oil and gas fields. Although monitoring data can be adjusted to account for unknown leks using area-based sampling designs (e.g., dual-frame sampling; WAFWA 2008) or estimators that incorporate correction factors (e.g., Huggins estimators; Walsh et al. 2004), lek-based management strategies. For this reason, one of the keys to appropriately managing sagegrouse in oil and gas fields is to locate all leks within and adjacent to the field prior to leasing and development. One way to do this would be to intensively track a representative sample of males to see where they go to display in the early morning hours during the breeding season. Second, lek-based approaches for managing populations in areas with energy development have not been empirically validated. Oil and gas leases typically stipulate either no surface occupancy (NSO) or restricted surface occupancy (RSO) within certain buffer distances around leks. Historically, the Bureau of Land Management implemented a 0.25-mi. NSO or RSO buffer around leks to minimize disturbance to lekking males and to prevent degradation of the habitat males use during the breeding season, with the overall intention of minimizing long-term population declines and preventing extirpation in areas with development. However, the 0.25-mi. stipulation has no scientific basis (p. B-5, Appendix B, CGSSC 2008). More recently, a review of range-wide studies of male diurnal habitat use and movements during the lekking season suggested that a 0.6-mile buffer around leks may be more appropriate (p. B-6, Appendix B, CGSSC 2008), and this criterion is now recommended by state agencies in Colorado and Wyoming (CGSSC 2008, State of Wyoming 2010). However, the distribution of suitable habitat around leks often is not homogenous and no studies to date have empirically tested how large buffers need to be to protect habitat for males during the lekking season, so it is unclear whether a 0.6 mi. buffer is too large, adequate, or too small. Research is needed to quantify the buffer size needed by intensively tracking both day-time and night-time habitat use of individual males around leks during the breeding season without disturbing the males. Testing GPS Transmitters Recent technological advances have led to production of 22-30 g, solar-powered, global positioning system (GPS) satellite transmitters that may be well-suited for generating the data needed to resolve lek-based monitoring and management issues. Most research studies use very high frequency (VHF) transmitters attached to a neck collar to radio-track individual sage-grouse. VHF necklace collars are widely accepted as the current standard method for radio-marking (Connelly et al. 2003), and necklace collars have been widely used on males (Ellis et al. 1987, Walsh et al. 2004, Knerr 2007, Robinson 2007, Wisinski 2007, Holloran et al. 2010, Walsh et al. 2010). However, no studies to date have tested the impact of VHF collars on male (or female) survival, and field observations have generated concern whether males can safely be fitted with necklace-style VHF collars. Necklace collars may interfere with male displays by bouncing up and striking the male’s beak during strutting; they may restrict breathing or foraging when neck and breast tissue swells during the breeding season; they may prevent yearling males from swallowing as their necks grow over time, leading to suffocation or starvation; and males may become distressed and repeatedly attempt to remove the collar, thereby increasing their detectability to predators (B. Walker, pers. obs.). Lek attendance of females with necklace-mounted VHF collars did not appear to be affected (Walsh et al. 2004), but females do not display, so whether necklace collars reduce male lek attendance remains unclear. GPS transmitters have several advantages over VHF necklace collars. GPS transmitters record multiple locations per day at specific, pre-programmed times; logistical problems that prevent crews from locating birds on the ground are eliminated (e.g., weather, road conditions, truck breakdowns, technicians oversleeping, denied access, etc.); data are gathered without disturbing the bird or its flock mates; and they provide high-resolution data on survival, movement, lek attendance, and diurnal and nocturnal habitat use around leks. Collecting data of comparable resolution and accuracy using VHF collars would result in excessive disturbance to birds and be logistically impossible. However, solar cells require that 6 �transmitters be mounted dorsally so they are exposed to the sun. Because of their similarity to backpackstyle transmitters (Brander 1968, Amstrup 1980), there is concern that rump-mounted transmitters may directly or indirectly reduce survival of sage-grouse. Moreover, as with any new technology, data are also needed to assess the proportion and accuracy of GPS locations acquired, transmitter durability and longevity under field conditions, and cost effectiveness in comparison with VHF collars. Current studies with female greater sage-grouse indicate that rump-mounted leg-loop harnesses may be a viable option for attaching GPS transmitters to males as well. Satellite GPS transmitters cannot be used with necklace collars because solar cells under the neck receive insufficient sunlight to charge the battery (B. Henke, Northstar Science and Technology; C. Bykowsky, Microwave Telemetry, pers. comm.). Five separate studies are now using GPS transmitters with a rump-mounted leg-loop harness design to track female sage-grouse. Survival of females with VHF transmitters (n = 42) and GPS transmitters (n = 50) was tracked in northwestern Colorado from spring 2009 to spring 2010. VHF and GPS females had similar survival rates through October 2009, but survival of GPS females was lower from October 2009 - March 2010, resulting in lower estimates of annual survival (0.556 ± 0.073 SE for VHF vs. 0.406 ± 0.068 SE for GPS). Despite limited sample sizes, these results suggest that the ratio of transmitter size to body size reduced, the harness design should be made more flexible, transmitters should be fit less snugly, or all three. Because males are larger than females, 30-g transmitters (38 g including harness and crimps) would be proportionally less of male body mass, approximately 1.1-1.9%, depending on male age (~2000-2400 g for yearlings, ~2800-3300 g for adults; Beck and Braun 1978). Leg-loop harnesses may still cause skin irritation under the legs, particularly during males’ vigorous strutting displays. Moreover, if harnesses are fitted too tightly around the legs, or if swelling occurs around the legs prior to the breeding season (as it does around the neck and breast), this may also restrict the ability of males to display in spring. Having a GPS transmitter with a highly reflective solar cell attached dorsally may also increase detectability of males to predators or alter their distribution of body weight such that it impedes flight and makes them more susceptible to predation or targeted by visual predators. If yearling males grow during the course of the study, they may also outgrow a less flexible or snugly-fitting harness. If leg-loop harnesses impact survival of males, I would predict lower survival rates for GPS males than those published for leg-banded males. Band-recapture data suggest that survival rates of male sage-grouse vary annually and by age (0.635 ± 0.034 SE for yearlings vs. 0.368 ± 0.007 SE for adults; Zablan et al. 2003). Males with VHF collars in southwestern Montana averaged 0.34 ± 0.067 SE annual survival, but the author did not distinguish between yearling and adult males (Wisinski 2007). If harnesses hinder movement or displays of males, I would also predict reduced display rates for GPS males compared to either color-banded or unmarked males of the same age under the same conditions. This study represents a four-year investigation (2011-2014) of greater sage-grouse lek monitoring and management strategies using males deployed with GPS transmitters in the Hiawatha Regional Energy Development Project area in NW Colorado and SW Wyoming. STUDY AREA The study area covers the Hiawatha Regional Energy Development project boundary in northwestern Colorado and southwestern Wyoming and includes birds from both Colorado (Zone 1; “Cold Spring Mountain/Hiawatha”) and Wyoming (“Salt Wells”) core breeding populations (Fig. 1). This area holds a large, robust population contiguous with greater sage-grouse range in northwestern Colorado and south-central Wyoming. The maximum count of males on known leks in Colorado’s Zone 1 varies annually (in part due to variation in effort), but it is considered a stable population (CGSSC 2008, p. 259). Previous data from VHF- and GPS-marked females in this region indicate that sage-grouse typically winter in or near the Hiawatha project area and attend leks both within the project area and at higher elevations surrounding the project area. At the start of the project in fall 2010, there were nine known leks within the study area plus 13 more immediately adjacent to the study area. It is unclear what proportion of 7 �males in the population our sample will represent because not all leks are known and it is unclear how the number of males counted on known leks relates to actual population size. Research is being conducted with the support of the Wyoming Game and Fish Department and the Rock Springs (WY) and Little Snake (CO) field offices of the Bureau of Land Management. METHODS Capture and Handling Most males are captured in fall and winter habitat prior to the onset of the breeding season to prevent biasing data on lek attendance the following spring (Walsh et al. 2004). A small number males trapped early in the breeding season are used to estimate inter-lek movements and habitat use around leks. Trapping effort and GPS unit deployment follow a probability-based sampling scheme based on winter habitat identified in seasonal habitat models (Walker 2010) to ensure that males from all potential wintering areas and therefore all leks in the project area are represented in the sample. I plan to capture and attach 30-g, rump-mounted solar-powered GPS PTT satellite transmitters (Northstar Science and Technology, King George, VA) on 30 adult male sage-grouse in November-December and on 30 yearling male sage-grouse in February each year. I selected 30-g transmitters because they have larger battery capacity than 22-g models, which decreases risk of transmitter failure (or temporary failure to transmit data) under low-light conditions. GPS males will also receive individually numbered aluminum leg bands (size 16) and distinctive combinations of colored leg bands. I also plan to capture and individually colorband 30 adult males in November-December and 30 yearling male sage-grouse in February each year. I will alternate marking methods during captures to maintain equal proportions of GPS males versus colorbanded males in each portion of the study area. Trapping yearling males in February rather than in the fall will allow them to reach larger body mass prior to deploying the transmitter, thereby reducing the chance that they will outgrow the harness during the breeding season. Transmitters from birds that die may be recovered, cleaned, refurbished, and redeployed to maintain or increase sample sizes for survival analyses and or collecting mark-resight data. Capture and handling methods followed standard operating procedures established for sagegrouse (Appendix A), with two exceptions: we were approved to use Super Talon® and MagNet® handheld net guns for captures, and decisions about whether injured birds should be released or euthanized was made in the field (rather than transporting the bird to a rehabilitation center) because no known rehabilitators in Colorado currently have the facilities to care for wild sage-grouse. We used night-time spotlighting and hoop-netting (Wakkinen 1992) or hand-held net guns for all captures. I selected a sample size of 30 individuals per age class (yearling vs. adult). It is crucial to estimate parameters for each age class separately because they have different survival rates (Zablan et al. 2003) and different rates of lek attendance (Walsh et al. 2004). Sample size must also be balanced with the potential for impacts on the population should GPS transmitters have highly detrimental effects on male survival. With a sample size of 30 males in each age class, statistical power will be > 0.80 if survival of adult males is < 0.14 or > 0.62 or if survival of yearling males is < 0.39 or > 0.86. This sample size will only allow detection of relatively large differences in survival with statistical power > 0.80. However, deploying more GPS transmitters would be unethical without data regarding whether the transmitters have catastrophic effects on survival. The loss of > 30 males in any given age class in any given year in this population would likely pose an unacceptable risk to stakeholders and cooperators. A sample size of 30 should inform us whether GPS transmitters have catastrophic effects on survival. GPS Transmitter Attachment I used a rump-mount attachment for GPS transmitters based on the method B design described in Bedrosian and Craighead (2007) modified for sage-grouse (Fig. 2). Transmitters were manufactured with a medium-brown, sand-textured finish to reduce reflected light. A thin layer of neoprene (either 0.125 in. or 0.25 in. thick) was glued to the bottom side of the transmitter to ensure that contact between the 8 �transmitter material and the bird’s lower back was padded and insulated. Harness material was 0.55-cm (0.25-inch) wide, brown Teflon ribbon (Bally Ribbon Mills, Bally, PA). A 12 cm length of 0.55-cm wide elastic cord was sewn into the center of a 75-cm (36-inch) length of Teflon ribbon such that 4-6 cm of stretchy Teflon ribbon extended out from the attachment points on either side (Fig. 2). The elastic gives the harness flexibility when the bird extends its legs during take-off and when males are displaying. Yearling harnesses were sewn with more elastic (16 cm) to accommodate possible increases in body size over time. In fall 2012, we documented that transmitters sometimes failed when back feathers near the front and sides of the transmitter covered the solar panel prevented the transmitter from charging and transmitting. This led us to develop an improved design with thicker neoprene (1/4”) to raise the units higher off the back and with longer and wider neoprene padding under the front half of the transmitter to prevent feathers from covering the solar panel (Fig. 3). The transmitter sides, front, and back were painted with brown, tan, black, and white camouflage to decrease visibility to predators (Fig. 4). Harnesses were fit with birds held in a standing position in fall 2010, but in spring 2011 we switched to holding birds on their sides to improve our ability to correctly fit the harness. Transmitters were mounted on the bird’s lower back centered between the legs (as seen from behind and as seen from the side of the bird) with the antenna extending toward the rear above the tail (Fig. 4). Harnesses were fit down, around, and underneath the legs and attached to the rear loop of the transmitter using a small section of 0.55-cm (0.25inch) diameter copper tubing as a crimp (Fig. 2). Copper crimps typically quickly become tarnished with exposure to the elements, but as a precaution, crimps were also marked with black ink before release to reduce reflected light. The Teflon ribbon is trimmed at an angle and left with just enough excess ribbon on each side (~3 cm) to allow us to refit or enlarge the harness if necessary. The end of the excess ribbon is dabbed with Superglue® (Super Glue Corporation, Rancho Cucamonga, CA) to prevent fraying. The life span of the exposed Teflon ribbon has not been tested, but it was successfully used with rump-mount transmitters on female sage-grouse for >36 months without breaking or deteriorating. The life of the elastic cord is unknown. Transmitters were fitted just snugly enough to prevent birds from dropping transmitters. The GPS transmitters units were solar-powered and may last for 3-5 years, which is longer than the life span of almost all male sage-grouse (Zablan et al. 2003). All GPS units were pre-programmed to collect 8 locations per day from March-May so as to get data on early morning lek attendance (6 am, 7 am, 8 am), mid-day feeding/loafing areas (12 pm), evening feeding areas (6 pm), and night roost locations (12 am). Units were programmed to collect two locations per day at 12 pm and 12 am from June-Feb to capture basic patterns of seasonal habitat use and movements while reducing demand on the battery during low-light conditions encountered in fall and winter. We did not remove or replace GPS transmitters unless there was an indication of transmitter failure or incorrect fit. GPS transmitters recovered from mortalities were cleaned and re-deployed on additional males to maximize sample sizes and reduce the cost of transmitters. Brett Walker was trained in the initial attachment technique in the field in March 2009 by Bryan Bedrosian, who has used GPS transmitters with raptors, corvids, and sagegrouse (Craighead and Bedrosian 2009). The ARGOS system sends GPS transmitter data as a text file by email. Raw text files are then parsed using “DSDCODE” software provided by Northstar Science and Technology. This software automatically amends new locations from GPS birds to an ArcGIS shapefile for each individual. I amended the parsed data (in .dbf format) to an existing Microsoft Excel® spreadsheet of GPS bird locations and removed duplicate records, flagged date and location errors, and identified records signifying important events (e.g., mortality). Lek and Lek Attendance Definitions I defined a lek as any area within which ≥ 2 males have displayed in ≥ 2 years, which is consistent with previous state-wide and range-wide definitions (Connelly et al. 2000, CGSSC 2008). I use this definition to ensure that small leks and “satellite” leks are included, but that locations where males do 9 �not consistently display are excluded (i.e., one-time use locations). The status of a lek may be active or inactive in any given year. Leks used by displaying males at least once within the past 5 years are considered active (CGSSC 2008). Newly-discovered leks > 500 m from all other known leks will be designated as potential leks. If those locations have displaying males in ≥ 2 years, they will be classified as new leks and assigned a name based on local geography. I will delineate a “count boundary” for each known lek prior to the first count and for each new lek immediately following its discovery. The count boundary represents the specific perimeter within which males would be visible and available for counting by observers during any given count. The purpose of establishing a count boundary is to ensure that the geographic area of observation for each lek is consistent over time. This prevents the characteristics of specific leks (e.g., their area, location, topography, etc.) from changing over time. This count boundary will necessarily be larger than the outer perimeter around displaying males on any given date because: (a) observers can typically see and count males over an area larger than just the area where displaying males are found, (b) males may shift the location where they strut slightly from day to day (WAFWA 2008), and (c) observers typically adjust the location from which they count males from day to day to maximize their ability to obtain complete counts of males. It is also important to unambiguously define lek “attendance” because some males use habitat near leks, but they may or may not be within the area that can be counted by observers. I define lek “attendance” for each male as a binomial variable. Lek attendance is classified as 1 if the male is inside the count boundary (i.e., visible and available for counting by observers) at any time during the standard count period (0.5 hrs before sunrise to 1.5 hrs after sunrise) and as 0 if the male is either: (a) outside the count boundary (i.e., not visible and unavailable for counting) during the standard count period, or (b) inside the count boundary at a time other than during the standard count period. Lek attendance of GPS males should be straightforward to assess when resighters are present, but there may be some ambiguity about lek attendance for GPS males not directly observed (those that attend leks at which no observers are present). The accuracy of high-quality locations derived from GPS transmitters is typically ≤ 26 m. Only GPS males with early morning locations within 26 m of the count boundary will be considered to have attended a lek. Lek Counts and Resighting CDOW lek-count protocol instructs observers to obtain a maximum count of males by conducting repeated counts 5-10 minutes apart over a 30-minute period between 0.5 hr before and 1.0 hr after sunrise (Appendix B). Although no specific guidelines are given for exactly how far away to be, biologists and wildlife managers typically count leks from 50-400 m away, depending on topography, access, and how far away they need to stay to keep from disturbing birds at the lek. They use whichever optics are required to obtain a reliable count (binoculars or spotting scope) and whichever mode of transportation (truck, ATV, on foot) gets them close enough to the lek to count it. A truck is preferred because it reduces disturbance to birds and is logistically easier and more comfortable for conducting repeated scans. Field crews will focus on collecting count data and resighting data at only those leks attended by GPS males, most of which are likely to be within or adjacent to the study area. Observers will visit each of these leks once a week. The field crew will be divided into three groups: resighters, counters, and surveyors. Surveyors will check locations of potential new leks as needed, and if males are present, will conduct a standard 30-minute lek count. Resighters will each go to a different lek and collect resighting data on GPS and color-banded males during each 30-minute interval from 0.5 hr before local sunrise to either 1.0 hrs after sunrise or to when all birds depart the lek, whichever is later. Resighters will use a spotting scope from a portable blind placed ~50 m from the lek (Walsh et al. 2004). The goal of each resighter is to collect accurate data on the identity of all GPS and color-banded males present on the lek during each 30-minute interval. Portable blinds will be placed near leks either the night before or >1 hr before to sunrise to prevent disturbance to birds on the lek (Walsh et al. 2004). Blinds will have raptor perch deterrents installed on top to prevent aerial predators from using blinds as perches. Counters will 10 �work in pairs, and each pair will conduct a 30-minute lek count during the standard count period at two leks per day (the same leks being observed by resighters). For counters, each 30-minute visit to a lek will be divided into six 5-minute scan intervals. Counters will follow CDOW count protocols and record the maximum number of yearling and adult males and females counted during each 5-minute interval. The goal of each counter is to get an accurate count of yearling and adult males and females during each scan interval and to determine the number (and eventually, the identity) of all GPS males present on the lek. Counters will also record any birds that arrive or leave the lek during each interval. Counters will alternate between using a spotting scope and binoculars during each scan interval. Each observer will be allowed to scan the lek multiple times within each 5-minute interval because that is typically how lek counts are conducted by CDOW biologists and wildlife managers. At the end of each count, the counters will consult with the resighter by two-way radio to reconcile and confirm the identity of any GPS males observed on the lek. Observers will be systematically rotated such that each observer conducts an equal number of lek counts and resighting days with each other observer. I will only hire observers with experience conducting lek counts. All observers will be trained in standard lek-count protocols, will practice resighting prior to collecting field data, and will collect data on standardized forms. All counts will be conducted from within a realistic distance from leks, depending on topography and optics (50-400 m), and all counters will record the distance (m) to the approximate lek center using a laser rangefinder. All observers will conduct counts using the same standard make and model of 10x binoculars and 20-60x zoom spotting scopes. In winter and spring 2012, we discovered that color-bands older than one year were deteriorating and either expanding and sliding down over the metal band or breaking and falling off. Field crews conducting resighting at leks in spring 2012 reported numerous cases of incomplete or incorrect band combinations caused by color-bands breaking and falling off or expanding and slipping down over metal bands. Even more males with incorrect band combinations were recorded on leks in spring 2013. It was logistically impossible to recapture all previously color-banded males, and we wouldn’t have been able to capture sufficient numbers of new males in fall 2012 to estimate differences in return rates or survival. For that reason, we opted not to mark a separate sample of color-banded only males in fall 2012 or spring 2013 and instead marked all males with color-bands and GPS transmitters. Objective 1a: Survival comparison – I will use location and mortality data from males with GPS transmitters to estimate seasonal and annual survival rates of yearlings and adults. The null hypothesis is that male greater sage-grouse with GPS transmitters in each age class have survival rates indistinguishable from those reported for leg-banded males in the published literature. If location data from a GPS male indicate a stationary transmitter, field crews will visit all subsequent locations to determine whether it was mortality or a dropped transmitter and to recover the transmitter using a metal detector. Transmitters deployed so far have typically been recovered within 20 m of their last set of stationary location(s) (B. Walker, unpub. data). I do not anticipate estimating cause-specific mortality rates because the delay between when birds are killed, the acquisition and processing of satellite data, and when locations can be checked by field crews is typically 4-7 days, which in most cases precludes determining proximate cause of death. I will use an information-theoretic approach (Burnham and Anderson 2002) to evaluate sets of a priori candidate models describing variation in daily and seasonal survival rates of males during breeding, summer, fall, and winter. Survival analyses of GPS male data will use a continuous-time approach such as a Cox proportional hazards model (Murray 2006). Age will be a fixed effect in all seasons (adult vs. yearling), and landscape-scale habitat variables known to influence habitat selection in each season (e.g., terrain ruggedness, proportion sagebrush habitat within 1 km; Walker 2010) will be included as additional 11 �explanatory variables. During the breeding-season, daily lek attendance status will be included as an explanatory variable to quantify risk due to lek attendance. Several males with GPS transmitters disappeared without any evidence of mortality in 20112012. In some cases, we documented that transmitters had failed due to feathers on the bird’s back covering the solar panel on the transmitter (i.e., males with covered transmitters were recaptured and transmitters were removed and tested). In other cases, it was unclear whether transmitters had failed, whether the transmitters were destroyed or lost power following mortality, or whether the transmitters slipped and failed to transmit their last location. Because random censoring is an assumption of survival analysis, we attached miniature VHF mortality transmitters (Advanced Telemetry Systems, Model A2720, Isanti, MN) underneath all GPS transmitters starting in fall 2012 to test whether GPS males whose transmitters stopped transmitting data were alive or dead (Fig. 3). Objective 1b: Leg-loop size comparison – Leg loops were marked at various distances from the front attachment point using colored iridescent, permanent markers. The exact length of the leg loop from the front to the rear attachment point was recorded in the field on each leg on each bird after fitting. Means and variances of harness lengths will be compared between yearlings and adults using a standard one-sided, two-sample t-test because of the a priori expectation that yearlings will have smaller leg-loop lengths than adults. If needed, yearlings may have to be recaptured after the breeding season to refit them with adult-sized harnesses. Recapture of yearlings may be difficult because the transmitters cannot be tracked in real time. If needed, I will use location data to identify recent night roost locations of yearling males and attempt to find and capture those males by trapping in those areas. Objective 1c: Comparison of GPS and color-banded male display rates – During lek counts at which marked males are present, the resighter will record the display rate (no. struts/minute) of the GPS male nearest the observer and of the color-banded male in the same age class that is nearest the observed GPS male. The resighter will conduct three 1-minute observations per individual spaced 1 minute apart. Data from the three 1-minute observation periods will be summed. Observation periods will alternate between GPS and color-banded males, and the first bird to be observed will be randomly determined. If no color-banded males are present on the lek, the resighter will observe the nearest color-banded only or unmarked male in the same age class. The observation period will occur during at some time during the first 1.0 hr after local sunrise to ensure that light is sufficient to record behavioral data, but after resighting data have been collected. When more than one GPS male and more than one color-banded male are present, the resighter will collect on the next pair of marked males at the next earliest opportunity. Time spent fighting with other males or copulating with females will be excluded when calculating display rates. The null statistical hypothesis is that GPS males and color-banded males will show no difference in mean display rate. Comparisons will be made using a paired-sample, repeated-measures design because the dataset will include repeated observations from the same individuals over time. Objective 1d: Comparison of GPS and color-banded male lek attendance rates – The null statistical hypothesis is that GPS males and color-banded males will show no difference raw rates of season-long lek attendance. Raw lek attendance for each individual will be calculated as the proportion of the total number of 30-minute intervals during the breeding season during which each marked bird was resighted on a lek. I will then compare raw lek attendance among GPS and color-banded males separately for each age class because the two age classes will be marked at different times of year. Objective 2: Using GPS males to find new leks – Early morning locations of GPS males will be compared against locations of known leks every three days as satellite data arrive and are processed to identify potential new lek locations in and near the study area. Males that make ≥ 2 early morning visits to the same location on consecutive mornings during the breeding season will be considered to have visited a potential lek location. The surveyor will then visit those locations or they will be checked from the air at 12 �least once during the next 7 days to document whether displaying males or their sign (e.g., pellets, tracks, feathers) are present or absent, and if so, how many. If displaying males or sign are present at a newly discovered lek, then that lek will be added to the list of regularly counted leks following standard protocols, and the count boundary determined prior to the next visit. A GPS male that uses a location within the count boundary during the count period that is subsequently discovered to be a lek will be considered to have attended that lek on that date. Objective 3: Estimate no. of leks in the study area – Data from GPS males will be used in a markrecapture framework to estimate the number of leks in the study area. Visits by marked GPS males can be used to “capture” leks and subsequent visits by marked birds to that lek constitute “recaptures” of that lek. Recapture histories for individual leks can then be derived and analyzed using an appropriate markresight model (Bartmann et al. 1987, Bowden and Kufeld 1995, McClintock et al. 2008). Objective 4: Estimating detectability of males on leks – I will compare three methods for estimating detectability of males on leks. Two of the methods have only recently been published and require validation for use with lekking species (Alldredge et al. 2007, Riddle et al. 2010). The third method is included as a way to double-check an assumption of the first two methods. First, I will use an unreconciled, independent, double-observer approach to estimate detectability from lek-count data (Riddle et al. 2010). Standard double-observer and removal models require that observers match or reconcile specific individual animals that were or were not detected by each observer (Nichols et al. 2000). Because there may be as many as 80 or more males on any given lek and most of these males will be unmarked, this would be impossible to do on most lek counts. Unreconciled doubleobserver models use raw maximum counts of the number of individuals detected (in each age class) from each of two independent observers to generate a site history for each observer on each count (e.g., 13, 15) (Riddle et al. 2010). Site histories are then analyzed using the repeated-counts hierarchical model of Royle (2004) in program PRESENCE, with the difference being that, in the unreconciled double-observer model, each observer is considered an independent “visit” (Riddle et al. 2010). One of the benefits of this approach is that leks do not actually have to be visited twice, and the closed population assumption is met (Riddle et al. 2010). The method may require using a negative binomial or zero-inflated Poisson distribution in place of a Poisson distribution if data are overdispersed (Riddle et al. 2010). This estimator may become unstable when detectability is low (P. Lukacs, pers. comm.). However, I anticipate relatively high detectability because observers typically position themselves to maximize their ability to detect males attending the lek. The counting protocol outlined above (under Lek counts and resighting) results in dataset with six repeated counts from the same lek on each date for each counter for each age class of males and for females, with three of the six counts by each counter done with a spotting scope and three with binoculars. Counters will record distance to approximate lek center and presence or absence of snow cover on the count as well as predator activity and weather (temperature, wind speed, precipitation, visibility, illumination) at the end of each 5-minute interval. Predator activity will be broken into three classes (no predator detected, predator visible near lek, predator on, over, or attacking males) based on observations of potential predators of adults (eagles, hawks, falcons, owls, coyote, red fox, bobcat, mountain lion, feral dog) that, in the opinion of the counter or resighter, should have been visible to males attending the lek. Covariates in the analysis of site histories will include a random effect of lek and fixed effects of lek size (i.e., max no. of males counted), distance from lek, optics used (binoculars vs. spotting scope), predator activity, weather, and an interaction between optics and distance from the lek. Because the data consist of repeated counts from the same lek within and among days, this dependence will have to be addressed using a repeated-measures approach. 13 �Second, I will use a time-to-detection approach with resighting data from GPS males collected by counters to estimate detectability. Time-to-detection approaches use resighting data to generate capture histories for individual males detected during the count, and at least four intervals are required for modeling (Alldredge et al. 2007). In the field, counters will record the number of GPS males they detect on the lek during each of the six 5-minute scan intervals. GPS transmitters should be visible at distances at which counts are typically conducted using binoculars. Counters will then double-check with resighters by two-way radio to confirm the identity of GPS males observed on the lek. Resightings will also be checked against early morning locations of GPS males to ensure correct identification of males. I use data from counters instead of from resighters to ensure that detectability measured is representative of how counts are typically conducted. Detections by resighters are not used in detectability calculations because lek counts generally are conducted at distances > 50 m from leks. Resighting data from counters will result in a dataset of capture histories for each marked individual observed during each scan interval for each count period on each date on each lek (e.g., 101011). Capture histories will then be linked with individual-, group-, count-, and interval-specific covariates. This method assumes that males do not arrive, leave, or leave then return to the lek between intervals within each 30-minute count period (i.e., it assumes a closed population). The method has fewer assumptions and more flexibility for modeling than either traditional double-observer (Nichols et al. 2000) or removal methods (Farnsworth et al. 2002). Covariates will include a random effect of either lek or observer (but not both at the same time) and fixed effects of distance from lek, optics used (binoculars vs. spotting scope), predator activity, weather, and an interaction between optics and distance from lek. Time-to-detection models for estimating detectability will be run in program MARK, version 6.0 (Alldredge et al. 2007, White 2010). I will also estimate detectability by calculating the proportion of GPS males known to have attended a lek that were also detected by either resighting or counting observers during lek counts on that same date. This is to test the implicit assumption that all males that attend a lek are available for counting. It is possible that not all males attending a lek are necessarily visible to both observers (e.g., some may be hidden by topography). Although time-to-detection and unreconciled double-observer approaches should both theoretically account for males attending that are hidden from view, it would be good to directly test this assumption. To do this, I will compare early morning locations of males with GPS transmitters against records of individual marked GPS males observed by resighters during lek counts at approximately the same times that GPS transmitters are scheduled to record early morning locations (6 am, 7 am, and 8 am). The resighting observer will estimate individual marked bird locations by correcting observer UTM locations for direction (θ, in degrees) and distance (m) using the formulas: northingmale = northingobserver + cos(θ) * distance and eastingmale = eastingobserver + sin(θ) * distance. Resighters will record their locations in Universal Transverse Mercator (UTM) coordinates in the North American Datum 1983 using a high-sensitivity GPS unit (Garmin eTrex Vista HCx), they will estimate direction to males from true north with a declinated compass (Silva Ranger CL), and they will estimate distance to those males using a laser rangefinder (Nikon Prostaff 550). To estimate the effect of counting males from the air on detectability, I use the maximum raw count of all males combined from counters on the ground versus the maximum raw count from a counter in a fixed-wing aircraft (either the pilot or an observer) using the same unreconciled double-observer approach as above. In this case, the difference in detection probability among observers represents the difference in detectability of counting on the ground versus from the air. The comparison will be made between data recorded on the flight and data recorded over the entire 30-minute count period on the ground. This comparison is appropriate because ground counts based on data from a 30-minute count period and flight counts based on data from 3-5 minute count periods are recorded with equal weight in statewide count databases. I will attempt to conduct 40 paired lek counts per year on the ground and from fixed-wing aircraft on the same dates and at the same times. Detectability from the air may be lower because data are derived from only 2-3 passes during a brief window of time (3-5 minutes) rather than counted for an extended period of time during the morning (30 minutes) as is typical for ground counts. 14 �However, it is possible that ground-based counts could result in lower counts if topography prevents observers on the ground from detecting all males. Objective 5: Estimate age-specific lek attendance of males – I will analyze lek attendance data in two ways. First, as recommended by Walsh et al. (2004), I will develop a modified “sightability” approach to estimate lek attendance for adult and yearling males using data from GPS males. I will use early morning locations of GPS males to determine which leks (or potential leks) GPS males are attending or likely to attend. Field crews will make every effort to count and resight GPS and colorbanded males on each of those leks at least once in random order during each week-long resighting occasion throughout the season. Resighting observations will be lumped into 30-minute resighting periods starting at 0.5 hr before local sunrise for lek attendance analyses. Resighters will also collect data on covariates likely to influence lek attendance for each 30-minute interval. Covariates collected by observers at the lek will also be applied to non-attending GPS males because the focus of this analysis is on testing factors that influence presence on the lek rather than factors influencing presence at locations away from leks. Because GPS males sometimes move between leks, they may not always be present on leks they previously attended that get counted. For this reason, data for the modified “sightability” model will necessarily come from a subset of our sample of GPS males. Data from males that attend noncounted leks will be excluded from this analysis. The dependent variable is lek attendance (1 = attended lek, 0 = did not attend lek). Covariates will include a random effect of lek, fixed effects of time of day, date, snow depth, the previous day’s weather, presence or absence of females on the lek, probability of female attendance, lek size (i.e., maximum count of males), and marking type (GPS vs. color-banded), as well as fixed effects of weather variables, predator activity, and frequency of anthropogenic disturbance during the previous 30-minute interval. Logistic regression will be conducted in program R (version 2.11.0, R Development Core Team 2010). Although misidentification of color-bands combinations is a concern for resighting, comparison of color-band combinations recorded against early-morning locations should allow us to correct any misidentification of GPS males by resighters. If misidentification is a problem, new mark-recapture approaches may be available to address that issue (e.g., Link et al. 2010). Second, I will use the entire GPS male dataset to estimate lek attendance as a function of variables that can be measured without observing attending males directly. I will compare early morning locations of males with GPS transmitters against the count boundary for all known active lek locations to determine whether or not GPS males attended leks (see definition of “attending a lek” in Objective 4, above). I can then estimate daily rates of lek attendance for each male using logistic regression. Field crews will document all major weather events that could influence male attendance throughout the field season (e.g., storms, high winds). Daily lek attendance will be modeled as a function of date, current weather (temperature, wind speed, precipitation), the previous day’s weather, resighter presence, counter presence, average lek size, previous lek attendance (as a measure of reproductive effort), and probability of female attendance (estimated from counts of females at leks over the course of the season). I include observer presence because having observers count leks may cause males or females to move to another lek or to forgo lek attendance that day, yet this has never been tested. Overall lek attendance for each individual over the season will be calculated by summing the total number of days that each bird attended a lek and dividing that value by the total number of days for which each individual was alive and its early morning location was known. Detection probability (the joint probability of detectability and lek attendance), may also be estimated as part of estimating male population size (see Objective 9, below) and can be compared against the product of detectability and lek attendance estimated separately. Objective 6: Estimating probability of age-specific presence using movements of males – As outlined above, probability of presence is one of five key components of detection probability for sagegrouse males on leks that need to be estimated for running simulations. I will use location data to estimate 15 �the daily probability of presence for each GPS male for each lek within the survey area on each day of the breeding season. Data will be stored as an N x L x D matrix, where N = the number of GPS males in the sample, L = the number of leks attended by GPS males, and D = the number of days during the breeding season. Each cell in the matrix is assigned a 1 or a 0 based on the presence (1) or absence (0) of each GPS male within a certain distance of each lek on each day. A value of 1 does not denote lek attendance because males have the option of either attending or not attending the nearest lek to them on each day. From this matrix, I can calculate the raw proportion of the males in our population that were present at or near each lek we’re studying on each day. This is the daily probability of presence that will be used in simulations. Because I will be tracking the location and movement of each male and identifying all leks used by males, where GPS males move and where field crews can determine lek status will define the survey area. However, if a male moves so far outside of the study area that field crews cannot survey any leks he might be attending and it is impossible to determine whether or not he attended a lek, then he will be excluded from the dataset (all values for that male for those days will remain blank). I will also use location data from GPS males overlaid with locations of all known active leks to document the frequency, timing, and distance of inter-lek movements by yearling and adult males. Objective 7: Estimating age-specific survival during the breeding season – The purpose of estimating daily survival is to determine the proportion of males in each age class that remain alive on each date over the course of the breeding season in each year for simulations. Age will be used as the only predictor variable in this analysis (adult vs. yearling). Survival analysis will use a continuous-time Cox proportional hazards model (Murray 2006). Objective 8: Simulate lek-count data – I will use empirical data on variation in male survival, presence, lek attendance, detectability, and lek-count effort in conjunction with important covariates (e.g., time of day, date, weather, etc.) to simulate how much lek counts are likely to vary in the absence of population change when conducted according to standardized protocols. I will simulate data for the same sample of leks for which I have data on the number of males counted, as well as data on survival, presence, lek attendance, and detectability. The number of males in the simulated population will be set at a value equal to the maximum count of yearling or adult males at each lek during the period of peak attendance for each age group divided by age-specific detectability estimated during that period. I can use these data to simulate what proportion of the simulated population of adult and yearling males would actually be alive, present, and attending each lek during each time period of the morning on each day of the breeding season in each year. I would then run scenarios using this simulated dataset with realistic combinations of measured and unmeasured variables that influence detectability (e.g., time of day, optics, distance from lek, weather, and number of counts per season). Scenarios would include counts conducted: (a) under more restrictive (0.5 hrs before to 0.5 hrs after sunrise) or less restrictive (0.5 hrs before to 1.0 hrs after sunrise) time of day requirements; (b) with binoculars versus spotting scope; (c) close to leks, farther away from leks, or at various distances from leks; (d) in good versus marginal weather conditions; (e) using a varying number of counts per season from 1 to 6 on randomly selected dates at least a week apart (to mimic data contained in state databases); (f) with varying proportions of leks counted to mimic access problems encountered in the field. Simulations will be set up in program R (version 2.11.0, R Development Core Team 2010). Objective 9: Estimate age-specific population size – If sufficient data from repeated counts are available at leks within the study area, I will use mark-resight and lek count data to estimate detection probabilities and population size for yearling males, adult males, and females. Because this is an open population (many leks surround the study area), I will analyze the data using an immigration-emigration mixed logit-normal mark-resight model (Bartmann et al. 1987, Neal et al. 1993, Clifton and Krementz 2006) in program MARK (version 6.0, White 2010). Population estimates will be generated in each year of the study. Although previous authors concluded that the joint hypergeometric estimator was unsuitable 16 �for greater sage-grouse because it does not allow for individual heterogeneity in lek attendance, violation of the closed-population assumption could lead to even greater bias in population estimates. Objective 10: Test 0.6-mile lek buffer – Portions of the study area have had oil and gas development since the 1920’s (Walker 2010). However, most leks within the study area are far enough away from areas with oil and gas development that I should have sufficient data to measure how male sage-grouse use habitat around leks in the absence of disturbance related to oil and gas development. If the hypothesis that males avoid disturbance is true, I would predict a pattern of constrained habitat use around leks within or near development compared to those outside development after accounting for habitat features. This can be tested by comparing buffer distances required to protect the same proportion of the male population at leks inside and outside development after controlling for habitat and topography. I will measure distances of three off-lek locations per day (at noon, 6 pm, and midnight) for each male to the center of the lek attended that day, the lek most recently attended, the nearest active lek (as recorded in CDOW databases or by field crews), and the lek attended on the next visit. I will then calculate the proportion of off-lek locations (for each portion of the day) that fall within specific distances of dissolved buffers around the centers of known active leks to test the effectiveness of the current 0.6-mi. NSO/RSO stipulation for lek buffers and to make recommendations on the most efficient buffer size to use to protect specific proportions of the population. It may also be possible to use a kernel or bivariate normal mixture model to estimate the probability of males using the area around leks (D. Walsh, pers. comm.). I will also compare the effectiveness of conserving areas that fall within different circular buffer sizes to areas of high priority habitat of similar size already identified using VHF locations of females (Walker 2010). RESULTS AND DISCUSSION We monitored a total of 37 non-juvenile (i.e., adult or yearling) male greater sage-grouse from September 2013 through mid-June 2014 within or adjacent to the Hiawatha Regional Energy Development boundary. Data represent an additional year of data on: (a) locations of potential new leks, (b) male survival, (c) lek attendance, (d) inter-lek movements, (e) inter-annual lek fidelity, and (f) nocturnal and diurnal habitat use around leks. I am currently analyzing data on male habitat use around leks. After that is submitted for publication, I will start analyzing data on lek attendance, inter-lek movement, detectability, and survival. LITERATURE CITED Aldridge, C. L., S. E. Nielsen, H. L. Beyer, M. S. Boyce, J. W. Connelly, S. T. Knick, and M. A. Schroeder. 2008. Range-wide patterns of greater sage-grouse persistence. Diversity and Distributions 14:983-994. Alldredge, M. W., K. H. Pollock, T. R. Simons, J. A. Collazo, and S. A. Shriner. 2007. Time-of-detection method for estimating abundance from point-count surveys. Auk 124:653-664. Amstrup, S. C. 1980. A radio-collar for game birds. Journal of Wildlife Management 44:214-217. Applegate, R. D. 2000. Use and misuse of prairie chicken lek surveys. Wildlife Society Bulletin 28:457458. Autenrieth, R., W. Molini, and C. E. Braun. 1982. Sage grouse management practices. Western States Sage Grouse Committee Technical Bulletin 1. Twin Falls, Idaho, USA. Barker, R. J. 1997. Joint modeling of live-recapture, tag-resight, and tag-recovery data. Biometrics 53:666-677. 17 �Bartmann, R.M., G.C. White, L.H. Carpenter, and R.A. Garrott. 1987. Aerial mark-recapture estimates of confined mule deer in pinyon-juniper woodland. Journal of Wildlife Management 51:41-46. Beck, T. D. I., and C. E. Braun. 1978. Weights of Colorado sage-grouse. Condor 80:241-243. Beck, T. D. I., and C. E. Braun. 1980. The strutting ground count: variation, traditionalism, and management needs. Proceedings of the Annual Conference of the Western Association of Fish and Wildlife Agencies 60:558-566. Bedrosian, B., and D. Craighead. 2007. Evaluation of techniques for attaching transmitters to common raven nestlings. Northwestern Naturalist 88:1-6. Craighead, D., and B. Bedrosian. 2009. A relationship between blood lead levels of Common Ravens and the hunting season in the southern Yellowstone Ecosystem. In R. T. Watson, M. Fuller, M. Pokras, and W. G. Hunt (Eds.). Ingestion of Lead from Spent Ammunition: Implications for Wildlife and Humans. The Peregrine Fund, Boise, Idaho, USA. Boyko, A. R., R. M. Gibson, and J. R. Lucas. 2004. How predation risk affects the temporal dynamics of avian leks: greater sage-grouse versus golden eagles. American Naturalist 163: 154-165. Bowden, D. C., and R. C. Kufeld. 1995. Generalized mark-sight population size estimation applied to Colorado moose. Journal of Wildlife Management 59:840-851. Bradbury, J. W., S. L. Verhencamp, and R. M. Gibson. 1989. Dispersion of displaying male sage grouse: patterns of temporal variation. Behavioral Ecology and Sociobiology 24:1-14. Brander, R. B. 1968. A radio-package harness for gamebirds. Journal of Wildlife Management 32:630632. Braun, C. E., O. O. Oedekoven, and C. L. Aldridge. 2002. Oil and gas development in western North America: effects on sagebrush steppe avifauna with particular emphasis on sage grouse. Transactions of the North American Wildlife and Natural Resources Conference 67:337-349. Burnham, K. P. and D. R. Anderson. 2002. Model selection and inference: a practical informationtheoretic approach. Second edition. Springer-Verlag, New York, New York. Clifton, A. M., and D. G. Krementz. 2006. Estimating numbers of greater prairie-chickens using mark– resight techniques. Journal of Wildlife Management 70:479-484. Colorado Greater Sage-grouse Steering Committee (CGSSC). 2008. Colorado greater sage-grouse conservation plan. Colorado Division of Wildlife, Denver, CO, USA. Connelly, J. W., S. T. Knick, M. A. Schroeder, and S. J. Stiver. 2004. Conservation assessment of greater sage-grouse and sagebrush habitats. Western Association of Fish and Wildlife Agencies, Cheyenne, Wyoming, USA. Connelly, J. W., K. P. Reese, and M. A. Schroeder. 2003. Monitoring of greater sage-grouse habitats and populations. Station Bulletin 80, University of Idaho, Moscow, Idaho, USA. Connelly, J. W., M. A. Schroeder, A. R. Sands, and C. E. Braun. 2000. Guidelines to manage sage-grouse populations and their habitats. Wildlife Society Bulletin 28:967-985. Dalke, P. D., D. B. Pyrah, D. C. Stanton, J. E. Crawford, and E. F. Schlatterer. 1963. Ecology, productivity, and management of sage grouse in Idaho. Journal of Wildlife Management 27:811840. Doherty, K. E., D. E. Naugle, H. Copeland, A. Pocewicz, and J. Kiesecker. 2010a. Energy development and conservation tradeoffs: systematic planning for greater sage-grouse in their eastern range. Studies in Avian Biology. In press. Doherty, K. E., D. E. Naugle, and J. Evans. 2010b. A currency for offsetting energy development impacts: horse-trading sage-grouse on the open market. PLoS ONE 5(4): e10339. doi:10.1371/journal.pone.0010339. Doherty, K. E., D. E. Naugle, and B. L. Walker. 2010c. Greater sage-grouse nesting habitat: the importance of managing at multiple scales. Journal of Wildlife Management 74:1544-1553. Doherty, K. E., D. E. Naugle, B. L. Walker, and J. M. Graham. 2008. Greater sage-grouse winter habitat selection and energy development. Journal of Wildlife Management 72:187-195. Dunn, P. O., and C. E. Braun. 1985. Natal dispersion and lek fidelity of greater sage-grouse. Auk 102:621-627. 18 �Ellis, K. L. 1984. Behavior of lekking sage-grouse in response to a perched Golden Eagle. Western Birds 15:37-38. Ellis, K. L., J. R. Murphy, and G. H. Richins. 1987. Distribution of breeding male sage-grouse in northeastern Utah. Western Birds 18:117-122. Emmons, S. R., and C. E. Braun. 1984. Lek attendance of male sage grouse. Journal of Wildlife Management 48:1023–1028. Eng, R. L. 1963. Observations on the breeding biology of male sage grouse. Journal of Wildlife Management 27:841-846. Forcey, G. M., J. T. Anderson, F. K. Ammer, and R. C. Whitmore. 2006. Comparison of two doubleobserver point-count approaches for estimating breeding bird abundance. Journal of Wildlife Management 70:1674-1681. Gibson, R. M. 1996. A re-evaluation of hotspot settlement in lekking sage grouse. Animal Behaviour 52:993-1005. Gill, R. B. 1965. Distribution and abundance of a population of sage grouse in North Park, Colorado. Thesis, Colorado State University, Fort Collins, Colorado, USA. Hagen, C. 2010. Greater sage-grouse conservation assessment and strategy for Oregon: a plan to maintain and enhance populations and habitat. Unpublished report. Oregon Department of Fish and Wildlife. 184 p. Harju, S. M., M. R. Dzialak, R. C. Taylor, L. D. Hayden-Wing, and J. B. Winstead. 2010. Thresholds and time lags in the effects of energy development on greater sage-grouse populations. Journal of Wildlife Management 74:437-448. Hartzler, J. E. 1972. An analysis of sage grouse lek behavior. Ph. D. dissertation. University of Montana, Missoula, Montana, USA. Holloran, M. J. 2005. Greater sage-grouse (Centrocercus urophasianus) population response to natural gas field development in western Wyoming. Ph.D. Dissertation, University of Wyoming, Laramie, Wyoming, USA. Holloran, M. J., R. C. Kaiser, and W. A. Hubert. 2010. Yearling greater sage-grouse response to energy development in Wyoming. Journal of Wildlife Management 74:65-72. Jenni, D. A., and J. E. Hartzler. 1978. Attendance at a sage grouse lek: implications for spring censuses. Journal of Wildlife Management 42:46–52. Knerr, J. S. 2007. Greater sage-grouse ecology in western Box Elder County, Utah. M. S. thesis. Utah State University, Logan, Utah, USA. Link, W. A, J. Yoshizaki, L. L. Bailey, and K. H. Pollock. 2010. Uncovering a latent multinomial: analysis of mark–recapture data with misidentification. Biometrics 66:178-185. Lukacs, P. M., and K. P. Burnham. 2005. Review of capture-recapture methods applicable to noninvasive genetic sampling. Molecular Ecology 14:3909-3919. McClintock, B. T., G. C. White, K. P. Burnham, and M. A. Pryde. 2008. A generalized mixed effects model of abundance for mark–sight data when sampling is without replacement. Pages 273–292 in D. L.Thompson, E. G. Cooch, and M. J. Conroy, eds. Modeling demographic processes in marked populations. Springer, New York, New York, USA. Murray, D. L. 2006. On improving telemetry-based survival estimation. Journal of Wildlife Management 70: 1530-1543. Natural Resources Conservation Service (NRCS). 2009. Greater Sage-Grouse Habitat Conservation Strategy. U. S. Department of Agriculture. Unpublished report. Billings, Montana, USA. 34 p. Naugle, D. E., and B. L. Walker. 2007. A collaborative vision for integrated monitoring of greater sagegrouse populations. Pages 57-62 in K. P. Reese and T. R. Bowyer, eds. Monitoring populations of greater sage-grouse: proceedings of a symposium at Idaho State University. College of Natural Resources Experiment Station Bulletin 88. Moscow, ID. Neal, A. K., G. C. White, R. B. Gill, D. F. Reed, and J. H. Olterman. 1993. Evaluation of mark-resight model assumptions for estimating mountain sheep numbers. Journal of Wildlife Management 57:436-450. 19 �Nichols, J. T., J. E. Hines, J. R. Sauer, F. W. Fallon, J. E. Fallon, and P. J. Heglund. 2000. A doubleobserver approach for estimating detection probability and abundance from point counts. Auk 117: 393-408. Patterson, R. L. 1952. The sage grouse in Wyoming. Sage Books, Denver, CO USA. 339 p. Rice, C. G., K. J. Jenkins, and W. Chang. 2008. A sightability model for mountain goats. Journal of Wildlife Management 73:468-478. Riddle, J. D., K. H. Pollock, and T. R. Simons. 2010. An unreconciled double-observer method for estimating detection probability and abundance. Auk 127: 841-849. Robinson, J. D. 2007. Ecology of two geographically distinct populations of greater sage-grouse inhabiting Utah’s West Desert. M. S. thesis, Utah State University. Logan, Utah, USA. Rogers, G. E. 1964. Sage grouse investigations in Colorado. Colorado Game, Fish and Parks Department Technical Publication Number 16. Royle, J. A. 2004. N-mixture models for estimating population size from spatially replicated counts. Biometrics 60:108-115. Samuel, M. D., E. O. Garton, M. W. Schlegel, and R. G. Carson. 1987. Visibility bias during aerial surveys of elk in north-central Idaho. Journal of Wildlife Management 51:622-630. Schroeder, M. A., C. L. Aldridge, A. D. Apa, J. R. Bohne, C. E. Braun, S. D. Bunnell, J. W. Connelly, P. A. Deibert, S. C. Gardner, M. A. Hilliard, G. D. Kobriger, C. W. McCarthy. 2004. Distribution of Sage-grouse in North America. Condor 106:363-376. Schroeder, M. A., J. R. Young, and C. E. Braun. 1999. Sage-grouse (Centrocercus urophasianus). Account 425 in A. Poole and F. Gill, editors. The birds of North America. The Academy of Natural Sciences, Philadelphia, PA, USA. State of Wyoming. 2010. State of Wyoming Executive Department, Executive Order Number 2010-4. Office of the Governor. Cheyenne, Wyoming. 15 p. Tack, J. D. 2010. Sage-grouse and the human footprint: implications for conservation of small and declining populations. Master’s thesis. University of Montana, Missoula, MT. 96 p. United States Fish and Wildlife Service (USFWS). 2010. 12-month finding for petitions to list the greater sage-grouse (Centrocercus urophasianus) as threatened or endangered. Federal Register 75(55):13909-14014. Wakkinen, W. L., K. P. Reese, J. W. Connelly, and R. A. Fischer. 1992. An improved spotlighting technique for capturing sage-grouse. Wildlife Society Bulletin 20:425-426. Walker, B. L. 2010. Hiawatha Regional Energy Development Project and Greater Sage-grouse Conservation in Northwestern Colorado and Southwestern Wyoming. Phase I: Winter and Breeding Habitat Selection and Maps. Unpublished Interim Progress Report. Colorado Division of Wildlife, Fort Collins, CO. 26 p. Walker, B. L., D. E. Naugle, and K. E. Doherty. 2007. Greater sage-grouse population response to energy development and habitat loss. Journal of Wildlife Management 71:2644-2654. Wallestad, R. and P. Schladweiler. 1974. Breeding season movements and habitat selection of male sage grouse. Journal of Wildlife Management 38:634-637. Walsh, D. P. 2002. Population estimation techniques for greater sage-grouse. Thesis, Colorado State University, Fort Collins, USA. Walsh, D. P., G. C. White, T. E. Remington, and D. C. Bowden. 2004. Evaluation of the lek-count index for greater sage-grouse. Wildlife Society Bulletin 32:56-68. Walsh, D. P., J. R. Stiver, G. C. White, T. E. Remington, and A. D. Apa. 2010. Population estimation techniques for lekking species. Journal of Wildlife Management 74:1607-1613. Walsh, D. P., C. F. Page, H. Campa, III, S. R. Winterstein, and D. E. Beyer, Jr. 2009. Incorporating estimates of group size in sightability models for wildlife. Journal of Wildlife Management 73:136-143. Western Association of Fish and Wildlife Agencies (WAFWA). 2008. Greater sage-grouse population trends: an analysis of lek count databases 1965-2007. Sage- and Columbian Sharp-tailed Grouse Technical Committee. Unpublished report. Cheyenne, Wyoming, USA. 138 p. 20 �White, G. C. 2010. MARK, Version 6.0. http://warnercnr.colostate.edu/~gwhite/mark/mark.htm. Wisinski, C. L. 2007. Survival and summer habitat selection of male greater sage-grouse (Centrocercus urophasianus) in southwestern Montana. M. S. thesis. Montana State University, Bozeman, Montana, USA. Zablan, M. A., C. E. Braun, and G. C. White. 2003. Estimation of greater sage-grouse survival in North Park, Colorado. Journal of Wildlife Management 67:144-154. 21 �Figure 1. Hiawatha Regional Energy Development project area, Colorado and Wyoming greater sagegrouse core areas, and surrounding region showing known active, inactive, and unknown status greater sage-grouse leks as of 2013, plus new leks discovered by monitoring and checking early-morning locations of GPS males in 2011 (purple squares), 2012 (hot pink squares), and 2013 (light pink squares). 22 �Figure 2. Harness design for rump-mounted leg-loop attachment of solar-powered GPS satellite PTT transmitters to male greater sage-grouse. 23 �Figure 3. Improved harness and transmitter design for rump-mounted leg-loop attachment of solar-powered GPS satellite PTT transmitters for male greater sage-grouse. This photo also shows the underside placement of a micro-VHF mortality sensor/transmitter (with the magnet held in place with blue painter’s tape). 24 �Figure 4. Attachment, placement, and camouflage of rump-mounted, solar-powered, GPS satellite PTT transmitters for male greater sage-grouse. 25 � https://cpw.cvlcollections.org/files/original/e6ef1c2e6b12e91871e80c0b813325ce.pdf d64edaccba37c26391dd7769ea8bea61 PDF Text Text Colorado Division of Parks and Wildlife September 2014-September 2015 WILDLIFE RESEARCH REPORT State of: Cost Center: Work Package: Task No.: Colorado 3420 0660 N/A Federal Aid Project No. N/A : : : : Division of Parks and Wildlife Avian Research Greater Sage-grouse Conservation Using GPS satellite transmitters to estimate survival, detectability on leks, lek attendance, interlek movements, and breeding season habitat use of male greater sage-grouse in northwestern Colorado Period Covered: September 1, 2014 – August 31, 2015 Author: B. L. Walker Personnel: B. Holmes, B. Petch, B. deVergie All information in this report is preliminary and subject to further evaluation. Information MAY NOT BE PUBLISHED OR QUOTED without permission of the author. Manipulation of these data beyond that contained in this report is discouraged. EXTENDED ABSTRACT Implementing effective monitoring and mitigation strategies is crucial for conserving populations of sensitive wildlife species. Concern over the status of greater sage-grouse (Centrocercus urophasianus) populations has increased both range-wide and in Colorado due to historical population declines, range contraction, continued loss and degradation of sagebrush habitat, and the potential for listing the species under the Endangered Species Act. Despite untested assumptions, lek-count data continue to be widely used as an index of abundance by state and federal agencies to monitor sage-grouse populations. Lek locations are also commonly used to identify and protect important sage-grouse habitat. However, the use of lek counts and lek locations to monitor and manage sage-grouse populations is controversial because how closely lek-count data track actual changes in male abundance from year to year and how effective lek buffers are at reducing disturbance to male sage-grouse and the habitat they use during the breeding season are largely unknown. Colorado Parks and Wildlife deployed solar-powered GPS transmitters on male greater sage-grouse and conducted double-observer counts and resighting at leks to obtain data on male survival, lek attendance, inter-lek movements, detectability, and diurnal and nocturnal habitat use around leks during the breeding season in and near the Hiawatha Regional Energy Development project area in northwestern Colorado in spring from 2011-2014. These data will allow us to evaluate the reliability of current lek-based monitoring methods for providing information about sage-grouse population trends the performance of lek buffers for conserving greater sage-grouse habitat. Analyses for this project are in progress. 1 � https://cpw.cvlcollections.org/files/original/41f778f58a760245f9845db8ea515361.pdf 8a235f257b445496fb1d251b50789282 PDF Text Text Colorado Division of Parks and Wildlife September 2015-September 2016 WILDLIFE RESEARCH REPORT State of: Cost Center: Work Package: Task No.: Colorado 3420 0660 N/A Federal Aid Project No. N/A : : : : Division of Parks and Wildlife Avian Research Greater Sage-grouse Conservation Using GPS satellite transmitters to estimate survival, detectability on leks, lek attendance, interlek movements, and breeding season habitat use of male greater sage-grouse in northwestern Colorado Period Covered: September 1, 2015 – August 31, 2016 Author: B. L. Walker Personnel: B. Holmes, B. Petch, B. deVergie All information in this report is preliminary and subject to further evaluation. Information MAY NOT BE PUBLISHED OR QUOTED without permission of the author. Manipulation of these data beyond that contained in this report is discouraged. EXTENDED ABSTRACT Implementing effective monitoring and mitigation strategies is crucial for conserving populations of sensitive wildlife species. Concern over the status of greater sage-grouse (Centrocercus urophasianus) populations has increased both range-wide and in Colorado due to historical population declines, range contraction, continued loss and degradation of sagebrush habitat, and the potential for listing the species under the Endangered Species Act. Despite untested assumptions, lek-count data continue to be widely used as an index of abundance by state and federal agencies to monitor sage-grouse populations. Lek locations are also commonly used to identify and protect important sage-grouse habitat. However, the use of lek counts and lek locations to monitor and manage sage-grouse populations is controversial because how closely lek-count data track actual changes in male abundance from year to year and how effective lek buffers are at reducing disturbance to male sage-grouse and the habitat they use during the breeding season are largely unknown. Colorado Parks and Wildlife deployed solar-powered GPS transmitters on male greater sage-grouse and conducted double-observer counts and resighting at leks to obtain data on male survival, lek attendance, inter-lek movements, detectability, and diurnal and nocturnal habitat use around leks during the breeding season in and near the Hiawatha Regional Energy Development project area in northwestern Colorado in spring from 2011-2014. These data will allow us to evaluate the reliability of current lek-based monitoring methods for providing information about sage-grouse population trends and the performance of lek buffers for conserving greater sage-grouse habitat. Analyses for this project are in progress. 1 � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Reports Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Using GPS satellite transmitters to estimate survival, detectability on leks, lek attendance, inter-lek movements, and breeding season habitat use of male greater sage-grouse in northwestern Colorado Description An account of the resource Implementing effective monitoring and mitigation strategies is crucial for conserving populations of sensitive wildlife species. Concern over the status of greater sage-grouse (<em>Centrocercus urophasianus</em>) populations has increased both range-wide and in Colorado due to historical population declines, range contraction, continued loss and degradation of sagebrush habitat, and recent federal listing of the species as warranted but precluded under the Endangered Species Act in 2010. Despite untested assumptions, lek-count data continue to be widely used as an index of abundance by state and federal agencies to monitor sage-grouse populations. Lek locations are also commonly used to identify and protect important sage-grouse habitat. However, the use of lek counts and lek locations to monitor and manage sage-grouse populations remains controversial because it is unknown how closely lek-count data track actual changes in male abundance from year to year, or if lek buffers are effective at reducing disturbance to male sage-grouse and their habitat during the breeding season. Creator An entity primarily responsible for making the resource Walker, Brett L. Subject The topic of the resource Greater sage-grouse <em>Centrocercus urophasianus</em> Northwestern Colorado Wildlife management Extent The size or duration of the resource. various pages Date Created Date of creation of the resource. 2014-2016 Rights Information about rights held in and over the resource <a href="http://rightsstatements.org/vocab/NoC-NC/1.0/">No Copyright - Non-Commercial Use Only</a> Type The nature or genre of the resource Text Format The file format, physical medium, or dimensions of the resource application/pdf Language A language of the resource English Is Part Of A related resource in which the described resource is physically or logically included. Cost Center 3420 Avian Research. Work Package 0660 Greater Sage-grouse conservation