TY - JOUR
AU1 - Severson, John P
AU2 - Coates, Peter S
AU3 - Prochazka, Brian G
AU4 - Ricca, Mark A
AU5 - Casazza, Michael L
AU6 - Delehanty, David J
AB - Abstract Reliable demographic estimates hinge on the assumption that marking animals does not alter their behavior, reproduction, or survival. Violations can bias inference and are especially egregious for species of high conservation concern. Global positioning system (GPS) devices represent a recent technological advancement that has contributed greatly to avian ecological studies compared with traditionally used very high frequency (VHF) radio transmitters, but may affect demographic rates differently than VHF transmitters. We compared survival between VHF (necklace attachment) and GPS (rump-mounted attachment) devices from >1,100 Greater Sage-Grouse (Centrocercus urophasianus), a species of high conservation concern, across multiple populations within California and Nevada. We found lower survival for GPS-marked compared to VHF-marked sage-grouse across most sex, age, and seasonal comparisons. Estimates of annual survival for GPS-marked sage-grouse were 0.55–0.86 times that of VHF-marked birds with considerable variation among sex and age classes. Differences in survival could be attributed to features associated with GPS devices, including greater weight, position of attachment (e.g., rump-mount harness), and a semi-reflective solar panel. In a post hoc analysis, we evaluated additive and interactive effects between device type (GPS vs. VHF) and transmitter mass as a proportion of body mass (PBM). While the device type effect alone was the best model, the PBM interaction also had support. For GPS devices, survival decreased with increasing PBM, whereas PBM effects were not found for VHF. We attributed differences in PBM effect to placement of transmitters on sage-grouse, as weight of GPS devices was positioned rearward. This information can help managers and researchers weigh costs and benefits of GPS-based monitoring. Our results indicate demographic data collected from GPS devices should be interpreted with caution, and use of these devices should be tailored to specific ecological questions. Future research aimed at investigating behavioral impacts and GPS designs that reduce adverse impacts on survival would be beneficial. Resumen Las estimaciones demográficas confiables se basan en el supuesto de que la marcación de los animales no altera su comportamiento, reproducción o supervivencia. La violación de estos supuestos puede sesgar las inferencias y son especialmente preocupantes para especies de alto valor de conservación. Los dispositivos de los sistemas de posicionamiento global (GPS por sus siglas en inglés) representan un avance tecnológico reciente que ha contribuido enormemente en los estudios de ecología de aves en comparación con los radio-transmisores de muy alta frecuencia (VHF por sus siglas en inglés) usados tradicionalmente, pero pueden afectar las tasas demográficas de modo diferentes a los transmisores VHF. Comparamos la supervivencia entre dispositivos VHF (agarrados al cuello) y GPS (agarrados a la rabadilla) de >1100 individuos de Centrocercus urophasianus, una especie de alto valor de conservación, a lo largo de múltiples poblaciones en California y Nevada. Encontramos menor supervivencia de individuos marcados con GPS en comparación con individuos marcados con VHF para la mayoría de las comparaciones de sexos, edades y estaciones. Las estimaciones de supervivencia anual de los individuos marcados con GPS fueron 0.55–0.86 veces que la de los individuos marcados con VHF con variaciones considerables entre sexos y clases de edad. Las diferencias en supervivencia podrían atribuirse a características asociadas con los dispositivos GPS, incluyendo mayor peso, posición de agarre (e.g., arnés montado en la rabadilla) y panel solar semi-reflectante. En un análisis post hoc, evaluamos los efectos aditivos e interactivos entre los tipos de dispositivo (GPS vs. VHF) y el peso de los transmisores como una proporción de la masa corporal (PMC). Mientras que el efecto del tipo de dispositivo representó el mejor modelo, la interacción de la PMC también obtuvo apoyo. Para los dispositivos GPS, la supervivencia disminuyó con un aumento de la PMC, mientras que no se encontraron efectos de la PMC para VHF. Atribuimos las diferencias en el efecto de la PMC al emplazamiento de los transmisores en los individuos, ya que el peso de los dispositivos GPS se ubicó en la parte trasera. Esta información puede ayudar a los gestores e investigadores a sopesar los costos y beneficios del monitoreo con GPS. Nuestros resultados indican que los datos demográficos colectados con dispositivos GPS deberían ser interpretados con cuidado, y el uso de estos dispositivos debería estar vinculado a preguntas ecológicas específicas. Serían beneficiosas investigaciones futuras dirigidas a estudiar los efectos en el comportamiento y los diseños de GPS que reduzcan los impactos adversos en la supervivencia. INTRODUCTION Radio telemetry revolutionized the study of wildlife ecology by improving capacity to estimate demographic rates as well as measure movement and patterns of marked individuals without excessive disturbance or need to recapture (Millspaugh and Marzluff 2001, Thorn et al. 2005, Murray 2006). A critical assumption of these analyses is that marked individuals share similar hazard (i.e. risk of mortality) to unmarked individuals such that their fates can be used to estimate population-level mortality rates (Murray and Fuller 2000, Murray 2006, Millspaugh et al. 2012). Concerns about direct and indirect effects of affixing tracking devices to animals were raised soon after deployment began (Marshall 1963, Boag 1972), and concerns persist (Murray and Fuller 2000, Withey et al. 2001, Guthery and Lusk 2004, Barron et al. 2010). While investigators who deploy tracking devices seek to minimize unintended experimental effects and comply with the ethics of animal welfare (Millspaugh et al. 2012), it is understood that placing tracking devices on animals is not a neutral action. Therefore, a key issue is whether marking animals with tracking devices yields scientifically meaningful information that is obtained ethically for animal welfare. In avian studies, an important question is the effect of tracking devices on survivorship. Increased mortality rate of radio-tagged individuals could lead to biased inferences further confounding our understanding of population trajectories and resulting in incorrect management decisions (Warner and Etter 1983, Paton et al. 1991, Winterstein et al. 2001). Also, different rates of death between marked and unmarked individuals could lead to potential biases in interpretation of population-level causes of mortality (Gilmer et al. 1974, Marks and Marks 1987). This could occur, for example, if bearing a tracking device increases the likelihood of being detected by a predator or decreases ability to escape predators. Furthermore, such effects could increase the efficiency of some predators relative to others such as greater susceptibility to ground-based predators for marked individuals (Marks and Marks 1987). Effects other than direct mortality may also be important such as increased energy demands associated with carrying a tracking device or the attachment method that might alter habitat use, movement, and other behaviors that introduce bias into ecological and management inferences (Boag 1972, Pietz et al. 1993, Gibson et al. 2013). While very high frequency (VHF) radio transmitters have been in widespread use for several decades, global positioning system (GPS) tracking devices have been developed more recently (Cagnacci et al. 2010, Tomkiewicz et al. 2010) and increased use is likely due to their capacity to further revolutionize studies of movement and space use (Cagnacci et al. 2010, Tomkiewicz et al. 2010). Global positioning systems can record large amounts of data with precise locations, which facilitates high-resolution analyses of exposure to factors that influence survival and reproduction as animals move through their environment (Cagnacci et al. 2010). The combination of GPS tracking devices with additional components that improve performance and usability, such as VHF units or platform transmitter terminals (PTT) for ground tracking and remote data collection, can increase overall weight of a GPS tracking device, potentially leading to negative effects. Additionally, each type of tracking device has attributes such as size, shape, and reflectance that could influence the performance of the marked individual. The characteristics of the study animals should also be considered such as their size, fusiform shape, and the potential effect on their energy balances attributed to tracking devices (Kenward 2001). Modifications used to lighten GPS devices, such as solar panels for battery charging, might also reduce crypsis and make the animals more visible to predators (Marks and Marks 1987, Burger et al. 1991). We assessed the effects of GPS and VHF tracking devices on Greater Sage-Grouse (Centrocercus urophasianus; hereafter, sage-grouse) mortality. Sage-grouse are relatively large birds that are studied throughout their range using these tracking devices. Sage-grouse present an especially relevant problem because they have experienced substantial declines in abundance and distribution subsequent to Euro-American settlement (Schroeder et al. 2004). Unbiased estimates of population trends of sage-grouse are important because they are often considered indicators of the ecological health of the sagebrush biome (Bureau of Land Management 2015, Coates et al. 2017) and are at the center of national conservation policy for sagebrush ecosystems (Connelly et al. 2011a, U.S. Fish and Wildlife Service 2015). Quantifying potential estimation bias in survival or differential mortality on sage-grouse imposed by different tracking devices would be valuable for ecological inference. VHF transmitters have been used frequently to monitor sage-grouse populations since the 1960s (Brander 1968), but GPS devices are being increasingly used (Dzialak et al. 2011, Prochazka et al. 2017). Attachment of VHF transmitters has evolved from backpack attachment (Brander 1968) to necklace-style attachment (Kenward 2001) following evidence linking grouse mortality to backpack attachments (Herzog 1979, Small and Rusch 1985, Marcstrom et al. 1989). However, early developers of neck-mounted poncho designs warned against fitting them to male lekking grouse, largely because of possible interference with the air sacs (Pyrah 1970, Amstrup 1980). More recent research has observed negative effects of necklaces on male sage-grouse (Gibson et al. 2013, Fremgen et al. 2017), suggesting backpack or rump-mount designs should be examined further. Currently, GPS devices are typically attached using a rump-mount design with Teflon ribbon looped around the legs that places the GPS on the lower back of the bird (Bedrosian and Craighead 2007). GPS devices are often heavier than VHF transmitters due to additional hardware (e.g., PTT and VHF or UHF), and solar-powered GPS devices have a semi-reflective solar panel intentionally exposed skyward that helps limit battery weight. Recognizing any measurement biases among tracking devices will improve the reliability of population-level inferences. Our objective was to quantify differences in survival between sage-grouse outfitted with VHF and GPS devices using a treatment–control experimental design (Murray and Fuller 2000) across multiple sites in sagebrush ecosystems of the Great Basin in the western United States. While the ideal experimental control would be unmarked or possibly leg-banded sage-grouse (Zablan et al. 2003, Hagen et al. 2018), we considered VHF units as a baseline control because the majority of sage-grouse demographic data are collected with these units and estimating demographic rates from unmarked animals is logistically challenging. We accounted for differences between the effects of tracking device types according to age, sex, season, and body mass. We hypothesized that GPS-marked sage-grouse would have decreased survival relative to VHF devices, and effects would be particularly pronounced for smaller sage-grouse because of factors associated with GPS devices such as greater weight and attachment style. We also sought to estimate effects of using data derived from sage-grouse outfitted with GPS tracking devices to assess possible population-level inferences on demography, and to offer general guidelines to researchers and managers for best uses of these devices and future research that would be beneficial. METHODS Study Area We monitored sage-grouse survival at 14 sites throughout the distribution of sage-grouse in California (n = 3) and Nevada (n = 11; Figure 1; Supplementary Material Table S1). Four sites were within the Bi-State Distinct Population Segment (Oyler-McCance et al. 2014), and 10 sites were within the primary sage-grouse range. Sage-grouse were marked and monitored as part of larger studies of sage-grouse ecology and their response to environmental stressors. Elevation across all sites averaged 2,064 m and ranged from 1,278 m to 3,369 m. Long-term annual precipitation averaged 32.6 cm across all sites and ranged from 24.2 cm to 60.9 cm (PRISM Climate Group 2015). The dominant vegetation was sagebrush (primarily Artemisia tridentata wyomingensis, A. t. vaseyana, A. t. tridentata, A. arbuscula, and A. nova). Other important shrub species included rabbitbrush (Ericameria spp., Chrysothamnus spp.), antelope bitterbrush (Purshia tridentata), snowberry (Symphoricarpos spp.), and saltbrush (Atriplex spp.). Pinyon (primarily Pinus monophylla) and juniper (primarily Juniperus osteosperma) woodlands were common at mid- to high elevation. FIGURE 1. Open in new tabDownload slide Locations of study sites in California and Nevada where Greater Sage-Grouse were marked using GPS and VHF devices (site abbreviations and associated data are defined in Supplementary Material Table S1). Red dashed square in inset shows study area location in western United States. Red polygons show sage-grouse distribution (Schroeder et al. 2004). FIGURE 1. Open in new tabDownload slide Locations of study sites in California and Nevada where Greater Sage-Grouse were marked using GPS and VHF devices (site abbreviations and associated data are defined in Supplementary Material Table S1). Red dashed square in inset shows study area location in western United States. Red polygons show sage-grouse distribution (Schroeder et al. 2004). Study Animal and Tracking Devices Sage-grouse are gallinaceous birds that live in open, sagebrush plains of western North America across 11 states and 2 provinces (Connelly et al. 2011b). They are strong fliers but tend to spend most of their time on the ground (Schroeder et al. 1999) and are susceptible to predation by various meso-carnivores and raptors (Hagen 2011). Males typically weigh 1,700–2,900 g and have featherless esophageal air sacs on the neck that inflate during breeding displays on leks. Females generally weigh 1,000–1,800 g and do not have air sacs (Schroeder et al. 1999). The VHF radio transmitters we used in this study were model A4060 avian necklace from Advanced Telemetry Systems (Isanti, Minnesota, USA; Figure 2), a common device in contemporary sage-grouse research (Blomberg et al. 2013a, 2013b; Smith et al. 2014). The VHF devices were charcoal gray with dimensions of 16 × 26 × 22 mm and a 30 cm black antenna. VHF devices weighed 22 g from the factory and had an expected battery life of ~869 days with pulse rate of 40 pulses per minute (ppm) and a pulse width of 20 milliseconds. Devices were equipped with a mortality sensor that caused the pulse rate to double (80 ppm) when the bird had not moved in 8 hr. The transmitter was attached around the neck of the bird with a steel cable inside a black plastic tube and secured with copper ferrule crimps (Figure 2). The transmitter was fit as loosely as possible around the neck while still not being able to slip over the head (generally a gap ~1–2 cm). The antenna would normally point straight up behind the bird, but we bent the antenna near the transmitter so that it pointed rearward and laid nearly flat against the back. With all attachment hardware, final transmitter packages averaged 24.5 g (range: 24–25 g). If the transmitter mass:body mass recommendations of 2%, 3%, or 5% are used (Kenward 2001, Withey et al. 2001, Fair et al. 2010), minimum bird weights for transmitter attachment are 1,225 g, 817 g, or 490 g, respectively. FIGURE 2. Open in new tabDownload slide Sage-grouse VHF device (A) shown with all attachment hardware weighs 24.5 ± 0.5 g. The black plastic tube goes around the neck and is attached to the transmitter and adjusted with steel cable and black copper ferrule crimps. GPS device (B) shown with all attachment hardware weighs 37.3 ± 1.8 g. Solar panel is black surface facing the camera, and small VHF transmitter for ground tracking is on the side of the device. The GPS device is glued to a foam neoprene pad and painted camouflage. Elastic ribbon is sewn into the Teflon ribbon (left side of the picture) to allow greater range of movement. The device is placed on the bird’s lower back (i.e. rump) with the antennae facing the rear. The Teflon ribbon loops around both legs and is adjusted and fixed with aluminum ferrule crimps, and the excess ribbon is removed. Comparison of devices with all hardware (C and D). The upper antenna in (D) is for the PTT for data transfer, and the lower antenna is for the VHF used for ground tracking. The GPS antenna, for acquiring high-resolution locations remotely, is the protruding square on the left side of the device. FIGURE 2. Open in new tabDownload slide Sage-grouse VHF device (A) shown with all attachment hardware weighs 24.5 ± 0.5 g. The black plastic tube goes around the neck and is attached to the transmitter and adjusted with steel cable and black copper ferrule crimps. GPS device (B) shown with all attachment hardware weighs 37.3 ± 1.8 g. Solar panel is black surface facing the camera, and small VHF transmitter for ground tracking is on the side of the device. The GPS device is glued to a foam neoprene pad and painted camouflage. Elastic ribbon is sewn into the Teflon ribbon (left side of the picture) to allow greater range of movement. The device is placed on the bird’s lower back (i.e. rump) with the antennae facing the rear. The Teflon ribbon loops around both legs and is adjusted and fixed with aluminum ferrule crimps, and the excess ribbon is removed. Comparison of devices with all hardware (C and D). The upper antenna in (D) is for the PTT for data transfer, and the lower antenna is for the VHF used for ground tracking. The GPS antenna, for acquiring high-resolution locations remotely, is the protruding square on the left side of the device. The GPS devices contained a PTT for data transfer, were solar powered, and weighed either 22 g or 30 g from the factory (GeoTrak, Apex, North Carolina, USA). They were painted camouflage to blend in with the bird’s back feathers, but the solar panel covering the majority of the upper surface was blue or black and semi-reflective (Figure 2). We used a rump-mount design to attach GPS devices, which is now the typical technique for GPS deployment in sage-grouse research and monitoring projects (Bedrosian and Craighead 2007). Device dimensions were 62 × 21 × 16 mm, in addition to a 20 cm gray antenna for the PTT. A small VHF transmitter weighing ~3 g was attached to the side of the device to allow ground tracking and had a 22 cm black antenna (Figure 2). The GPS device was attached to a foam neoprene pad to add comfort for the bird and reduce abrasions. Brown Teflon ribbon was used to construct leg loops for rump-mount attachment to the bird. Elastic ribbon was sewn into a portion of the Teflon to allow more freedom of movement and compensation for bird growth. The device was placed on the lower back (i.e. rump), the leg loops were fit as loosely as possible while assuring the device would not slip off, and then the ribbon was secured with aluminum ferrule crimps and cyanoacrylate glue. The VHF antenna pointed straight back along the tail feathers, but the PTT antenna pointed rearward at ~35–40° above the VHF antenna. With the small VHF transmitter, neoprene pad, and all attachment hardware, the final package weight increased to an average of 37.3 g (range: 35–40 g). If the device mass:body mass recommendations of 2%, 3%, or 5% are used (Kenward 2001, Withey et al. 2001, Fair et al. 2010), sage-grouse receiving GPS devices would need to weigh 1,865 g, 1,243 g, or 746 g, respectively. We note that we could not standardize VHF and GPS devices to similar masses and dimensions owing to production differences among device manufacturers over the relatively long duration of our study. Field Methods We captured sage-grouse during the spring and fall 2012–2017 using the spotlight method (Giesen et al. 1982, Wakkinen et al. 1992), determined the sex and age of captured birds (Braun and Schroeder 2015), and measured their weights to the nearest gram with digital scales. Ages in the fall included juvenile (hatch year), yearling (second year), and adult (>second year), while ages in the spring were yearling and adult. Captured adult, second year, and hatch year males weighed an average of 1,960 ± 213 g (mean ± SD), 1,538 ± 322 g, and 1,447 ± 219 g, respectively, in the fall; and adults and second year males weighed 2,487 ± 199 g and 2,206 ± 200 g, respectively, in the spring. Adult, second year, and hatch year females weighed 1,152 ± 84 g, 1,087 ± 103 g, and 992 ± 143 g, respectively, in the fall; and adults and second year females weighed 1,408 ± 137 g and 1,311 ± 153 g, respectively, in the spring. Captured birds were fitted with an aluminum butt-end leg band and were randomly assigned for marking with a GPS or VHF device. We used the 5% criterion of the bird’s weight as a cutoff for installation of both device types (Kenward 2001, Fair et al. 2010). VHF-marked birds were occasionally recaptured and fitted with a GPS, which was accounted for as a time-dependent variable in the modeling process (see Survival Analysis). GPS devices recorded multiple locations (~4–13) daily, and location data were downloaded weekly from CLS-America (Lanham, Maryland, USA). We attempted to locate the device in the field to determine the bird’s fate when consecutive location data indicated a lack of movement over a period of >1 day. We listened for VHF-marked birds at weekly minimum intervals during the spring and summer and 1–2 mo intervals during the fall and winter to determine survival status using the mortality sensor. Presumed mortalities that could not be located in the field, missing birds, and failed devices were right censored because definitive fate could not be determined. Right censoring was assumed to be random and was not expected to bias the results. Survival Analysis We used Bayesian shared frailty models based on their ability to independently account for intraclass correlation via random effects and estimate frailties (i.e. mortality risk) across multiple groups/classes (Gutierrez 2002, Halstead et al. 2012) including age, sex, and device type (R code and vague prior specifications for all analyses are provided in Supplementary Material Appendix S1 and S2, respectively). We parameterized 2 separate models to estimate the difference in seasonal survival for male and female birds (sex model) and juvenile, yearling, and adult birds (age model) according to the device types (i.e. VHF, GPS) fitted to each bird. The age model was restricted to female birds only because of limited sample sizes for VHF-marked males across all 3 age classes. Random intercept structures for both models comprised additive effects of individual bird, site, and year. Survival was modeled as a continuous process observed at discrete intervals (1 week). Because sage-grouse survival is known to vary across their annual life cycle (Moynahan et al. 2006, Blomberg et al. 2013a, 2013b), we fit a separate log hazard ratio for 4 distinct seasons: Spring = March 15 to June 14; Summer = June 15 to September 14; Fall = September 15 to December 14; Winter = December 15 to March 14. However, because we were not interested in the main effect of each season, but rather the interaction of sex/age with season, we specified an interaction between the 2, which we refer to in our model as λ. The frailty model was expressed as UHhijkl=exp(λkl+βklG+κh+ηi+ζj) denoting a change in unit hazard (UH) with expected change of magnitude β when G (a binary variable for GPS) equaled 1, in essence treating β as the offset log hazard to λ. Subscripts h, i, j, k, and l reference bird, site, year, age/sex (depending upon model), and season, respectively. Age, season, year, and device type were allowed to vary with time. Estimating age and device type as time-varying covariates allowed individuals to graduate into a higher age class with time or switch between VHF and GPS devices. Individuals graduated into the next age class on March 15 (approximate start of lek activity) each year. Adults could not graduate into a higher age class. Juveniles were not available for capture before fall and, therefore, we could not estimate hazards for spring and summer for this age class. Using the UH from 13 one-week intervals, we derived a cumulative hazard (CH) as CHwhijkl=∑T=13w=1UH1:w,hijkl where subscript w references week, and the survival parameter (S) took the form Swhijkl=e−CHwhijkl However, the duration of the study spanned 289 weeks (March 11, 2012, to September 25, 2017). Because individuals entered into the study (i.e. were captured and marked) at different times we chose a study start date based on the date on which the first individual was captured (March 11, 2012) and transformed all other start dates to weeks since the beginning of the study. We did not test suites of competing models because our goal was to estimate survival for the aforementioned sex, age, and seasonal groups; moreover, examination of posterior distributions can identify differences among groups. We ran 3 MCMC chains of 20,000 iterations each following a burn-in period of 50,000 iterations and thinned by a factor of 5. Convergence (i.e. a stationary posterior distribution) was assessed visually using history plots and the R̂ statistic (Gelman 2014). We did not find a lack of convergence among any of the parameters monitored (maximum R̂ = 1.1). Posterior probability distributions for each model procedure were estimated using R 3.1.1 (R Core Team 2014) and the package RJAGS (Plummer 2015). Posterior distributions are displayed in figures, described in the text and tables as the median (i.e. 50th percentile) and 95% credible interval (CI), and are reported in relation to priors used for each parameter (Supplementary Material Table S2). We derived annual survival using a separate model in which season was removed, and we report the median and 95% CI. As a quick way to estimate potential effects to population growth rate, we multiplied the female annual survival by a previously published range-wide sensitivity value of 1.7 (Taylor et al. 2012) to evaluate how differences in annual female survival fitted with VHF or GPS devices may bias population-level inference. Using this sensitivity value, for example, a 1.0% increase in female survival would yield a 1.7% increase in annual per capita population growth. Additionally, we calculated the cumulative survival over a 9 yr period (i.e. a likely maximum life span for sage-grouse; Schroeder et al. 1999) to assess long-term survivorship of birds marked with GPS versus VHF devices. For female cumulative survival, we began the calculation with juveniles trapped in the fall and used the seasonal age-structured survival estimates to project through time. For male cumulative survival, we also started the calculation in the fall but did not use an age structure as this was not in the model. In a post hoc analysis, we compared multiple models representing specific hypotheses of mechanisms of the effect of tracking devices on survival (Table 1; Supplemental Material Figure S1). We calculated device weight as a percent of bird body mass (PBM) to help elucidate weight effects. The models included (1) device type only, which we hypothesized would represent an effect of the solar panel or of the different attachment types; (2) PBM only, which we hypothesized would represent an effect of weight without a specific difference between the devices; (3) device type plus PBM additive effect, which we hypothesized would represent an effect of weight along with the solar panel or attachment; and (4) a device type by PBM interaction, which we hypothesized would represent a weight effect different for GPS than for VHF. For the latter hypothesis, we predicted that heavier weights placed on the rear of the bird (GPS type) are more likely to reduce survival. If data supported this hypothesis, then our goal was to try to identify rough thresholds of PBM where device type effects on mortality diverged. This threshold could be used by researchers to identify individuals too light to be marked with GPS devices based on a simple ratio of the actual device mass to bird mass. For this analysis, we restricted the dataset to adult female sage-grouse only, largely to prevent potential confounding effects on differences in survival among transmitter types as a function of PBM. This is because males are substantially heavier than females and yearlings are typically lighter than adults. Adult females represented the largest sample size in our study. Most research is focused on females, and variation in population growth is strongly sensitive to variation in female survival (Taylor et al. 2012). Sample sizes were 31 and 42 for GPS, and 146 and 120 for VHF, during spring and fall, respectively. We also restricted this post hoc assessment to the first 60 days post-capture, so the assumption of a relatively constant body mass could be met given that we did not know changes in individual body mass over time. Model inferences in PBM were at values where devices were represented (1.5–3.0%) so that predictions were in-sample. We accounted for biological variation in PBM related to circannual cycles by fitting month of capture as a random effect. We first attempted to fit this model in a shared frailty framework, but owing to the truncated period, few mortality events occurred leading to non-convergence of MCMC chains. Therefore, we fit bird fate (survived or died) as a binary response variable within a logistic regression model that converged. We ran the model on 3 chains of 25,000 iterations each, following a burn-in of 100,000 iterations. All chains were then thinned by a factor of 5. The models were compared using the Watanabe-Akaike information criterion (WAIC; Watanabe 2010), and we considered models with WAIC < 2 from the null model to have support and WAIC < 1 from the top model (lowest WAIC) to be highly competitive. TABLE 1. Comparison of models affecting survival to 60 days post-marking by device type (VHF or GPS) and weight of device as a percent of body mass (PBM) for female Greater Sage-Grouse in Nevada and California. WAIC = Watanabe-Akaike information criterion. Model . Hypothesized mechanism . Penalty . Deviance . WAIC . ΔWAIC . Device only Solar panel or Attachment 4.28 164.24 173.54 0.00 Device * PBM Device mass + Mass placement 4.93 163.19 173.89 0.36 Device + PBM Device mass + Solar panel or Attachment 4.86 163.89 174.42 0.89 PBM Only Device mass 5.00 166.71 177.69 4.15 Null No effect 3.49 183.31 190.81 17.28 Model . Hypothesized mechanism . Penalty . Deviance . WAIC . ΔWAIC . Device only Solar panel or Attachment 4.28 164.24 173.54 0.00 Device * PBM Device mass + Mass placement 4.93 163.19 173.89 0.36 Device + PBM Device mass + Solar panel or Attachment 4.86 163.89 174.42 0.89 PBM Only Device mass 5.00 166.71 177.69 4.15 Null No effect 3.49 183.31 190.81 17.28 Open in new tab TABLE 1. Comparison of models affecting survival to 60 days post-marking by device type (VHF or GPS) and weight of device as a percent of body mass (PBM) for female Greater Sage-Grouse in Nevada and California. WAIC = Watanabe-Akaike information criterion. Model . Hypothesized mechanism . Penalty . Deviance . WAIC . ΔWAIC . Device only Solar panel or Attachment 4.28 164.24 173.54 0.00 Device * PBM Device mass + Mass placement 4.93 163.19 173.89 0.36 Device + PBM Device mass + Solar panel or Attachment 4.86 163.89 174.42 0.89 PBM Only Device mass 5.00 166.71 177.69 4.15 Null No effect 3.49 183.31 190.81 17.28 Model . Hypothesized mechanism . Penalty . Deviance . WAIC . ΔWAIC . Device only Solar panel or Attachment 4.28 164.24 173.54 0.00 Device * PBM Device mass + Mass placement 4.93 163.19 173.89 0.36 Device + PBM Device mass + Solar panel or Attachment 4.86 163.89 174.42 0.89 PBM Only Device mass 5.00 166.71 177.69 4.15 Null No effect 3.49 183.31 190.81 17.28 Open in new tab RESULTS We attached VHF transmitters to 821 female and 52 male sage-grouse and attached GPS devices to 234 female and 125 male sage-grouse from 2012 to 2017 (Supplementary Material Table S1). Sample sizes by age at capture for females were juveniles = 260, yearlings = 292, and adults = 503. Sample sizes by age at capture for males were juveniles = 28, yearlings = 29, and adults = 120. We recorded 316 mortalities of VHF-marked birds and 261 mortalities of GPS-marked birds. The remaining birds were right censored because either the bird disappeared, the device failed, or the bird survived to the end of the study. Several differences in survival related to device type were evident for males and females across seasons. Except for males in spring, the median hazard ratio was >1.0 for both sexes in all seasons indicating greater risk for birds marked with GPS devices (Supplementary Material Table S3). The 95% CI for hazard ratio distributions did not overlap 1.0 for females in all seasons or males in summer and fall (Supplementary Material Figure S2). Median seasonal survival estimates were 1.08–1.19 times greater for females marked with VHF devices than those marked with GPS devices (Figure 3). Seasonal survival estimates for males marked with VHF devices ranged from 0.98 to 1.32 times that of individuals marked with GPS devices. Posterior distributions by device type were largely nonoverlapping except for males in spring and winter. FIGURE 3. Open in new tabDownload slide Posterior distribution of survival estimates for female (A) and male (B) sage-grouse marked with VHF compared to GPS devices during each season. Polygons are posterior distributions using sex–season combinations as group-factor variables. Solid vertical bars are mean values during specific seasons. Vertical dashed lines are overall mean values. FIGURE 3. Open in new tabDownload slide Posterior distribution of survival estimates for female (A) and male (B) sage-grouse marked with VHF compared to GPS devices during each season. Polygons are posterior distributions using sex–season combinations as group-factor variables. Solid vertical bars are mean values during specific seasons. Vertical dashed lines are overall mean values. Device type also appeared to impact female survival across several age classes and seasons. Except for yearlings in the summer, the median hazard ratio was greater than 1.0 for all ages in all seasons indicating greater risk for birds marked with GPS devices (Supplementary Material Table S3). The 95% CI for hazard ratio distributions did not overlap 1.0 for adults in all seasons, yearlings in the winter, and juveniles in the fall (Supplementary Material Figure S2, Table S3). Median seasonal survival estimates were 1.09–1.25 times greater for adult females, 1.02–1.29 times greater for juvenile females, and ranged from 0.97 to 1.11 times that of yearling females marked with VHF devices than those marked with GPS devices (Figure 4). Posterior distributions by device type were largely nonoverlapping except for yearling females in spring, summer, and fall, and juvenile females in winter. FIGURE 4. Open in new tabDownload slide Posterior distribution of survival estimates for adult female (A), yearling female (B), and juvenile female (C) sage-grouse marked with VHF compared to GPS devices during each season. Polygons are posterior distributions. Solid vertical bars are mean values during specific seasons. Vertical dashed lines are overall mean values. FIGURE 4. Open in new tabDownload slide Posterior distribution of survival estimates for adult female (A), yearling female (B), and juvenile female (C) sage-grouse marked with VHF compared to GPS devices during each season. Polygons are posterior distributions. Solid vertical bars are mean values during specific seasons. Vertical dashed lines are overall mean values. Median annual survival estimates were higher for VHF-marked birds than GPS-marked birds for both sexes and ages, although posterior distributions by device overlapped for yearlings (Figure 5). The median annual estimate for GPS-marked females was 0.55 and 0.85 times that of VHF-marked females within the adult and yearling age classes, respectively. The median annual estimate calculated across all age classes, for GPS-marked males and females was 0.58 and 0.61 times that of VHF-marked males and females, respectively. When we multiplied the difference in female survival estimates by the previously published sensitivity for female survival (i.e. 1.7), potential change in annual growth rate was estimated to be 46% lower for a GPS-marked population than a VHF-marked population, given all individuals are marked with either GPS or VHF within a population. FIGURE 5. Open in new tabDownload slide Annual survival estimates for GPS-marked sage-grouse compared to VHF-marked sage-grouse. Polygons are posterior distributions. Solid vertical bars are mean values for specific sex or age classes. Vertical dashed lines are overall mean values. FIGURE 5. Open in new tabDownload slide Annual survival estimates for GPS-marked sage-grouse compared to VHF-marked sage-grouse. Polygons are posterior distributions. Solid vertical bars are mean values for specific sex or age classes. Vertical dashed lines are overall mean values. When we calculated cumulative survival over the long term using the age-structured seasonal model, females marked with VHF devices were expected to survive ~3.2 seasons (9.6 mo) longer than those marked with GPS devices (Supplemental Material Figure S3). With the seasonal survival model, VHF-marked males were expected to live ~3.5 seasons (10.5 mo) longer than those marked with GPS devices. In the post hoc analysis, the most-supported model had device type only, but all models performed better than the null model by WAIC > 13 (Table 1). The device type by PBM interaction and the device type plus PBM additive effect were WAIC < 1 of the top model. The interaction model had the lowest deviance. In general, VHF-marked females survived at higher rates than GPS-marked females, and PBM had more of a negative relationship for GPS-marked females (Figure 6). FIGURE 6. Open in new tabDownload slide Results of post hoc analysis of interaction between device type and device weight as a percent of adult female sage-grouse body mass (PBM) on survival to 60 days post-capture (A) and the difference between birds marked with VHF and GPS devices (B). Polygons are posterior distributions. Vertical red line is zero difference in survival. FIGURE 6. Open in new tabDownload slide Results of post hoc analysis of interaction between device type and device weight as a percent of adult female sage-grouse body mass (PBM) on survival to 60 days post-capture (A) and the difference between birds marked with VHF and GPS devices (B). Polygons are posterior distributions. Vertical red line is zero difference in survival. DISCUSSION Advances in technology and reductions in device mass have encouraged the use of GPS devices to study smaller taxa, such as sage-grouse, because they yield finer temporal and spatial resolution data that can be remotely acquired. However, our results demonstrated that marking sage-grouse with currently available GPS devices increased mortality (i.e. sage-grouse marked with GPS devices had lower survival across sexes, ages, and seasons than those marked with VHF transmitters). Consequently, demographic rates derived from samples of sage-grouse carrying relatively heavy rump-mounted GPS devices may result in biased parameter estimates for survival, which has ramifications for estimates of population growth that could trigger unnecessary management action. While our experimental design did not allow identification of the exact mechanism of the marking effect, our post hoc analysis supported a possible combination of attachment type, reflective solar panel, device weight as percentage of body mass, and device positioning as the likely cause of the decreased survival. Our results represented an important first step in understanding the possible mechanisms driving increased mortality risk and tradeoffs associated with using GPS devices to study a species of conservation concern such as the sage-grouse that has substantial management and research attention. These results will help researchers weigh advantages and disadvantages of using different kinds of tracking devices in future studies. Device Design and Mechanisms of Effects Multiple design factors can influence the probability and magnitude of a tracking device negatively impacting a study animal (Murray and Fuller 2000, Kenward 2001, Withey et al. 2001, Barron et al. 2010). Size and shape can influence body movements, aerial drag, and likelihood of entanglement with vegetation (Karl and Clout 1987, Pennycuick et al. 2012). Device mass can influence takeoff speed, cruising speed, flight distance, and agility (Gessaman and Nagy 1988, Pennycuick et al. 2012). Color and sheen of devices can potentially attract or deter predators and affect concealment (Marks and Marks 1987, Burger et al. 1991). Attachment method, materials, and placement on the body can influence body movement, aerial drag, concealment, sound (e.g., “antenna slap”), and feather wear or abrasions (Gilmer et al. 1974, Herzog 1979, Hines and Zwickel 1985, Marks and Marks 1987). In our study, the major design differences between the 2 devices were attachment type (VHF: necklace; GPS: rump-mount harness), mass (VHF: 24.5 g; GPS: 37.3 g), placement of device (VHF: front; GPS: rear), and color (VHF: charcoal gray; GPS: camouflage with semi-reflective blue/black solar panel). Design characteristics of the tracking device likely have more impact on marked birds than mass when devices are less than 5% of body weight (Barron et al. 2010). In our post hoc analysis, we found that device type affected survival of adult females, which indicated a mass-independent effect such as the solar panel or the attachment type. The interaction model also received support, which indicated that increased device mass can disproportionately impact GPS-marked sage-grouse compared to VHF-marked birds, possibly as a result of placing the additional mass in a suboptimal location. Placing mass rearward likely impinges on flight balance and disrupts aerodynamics leading to decreased lift and increased drag (Gessaman and Nagy 1988, Pennycuick et al. 2012). Conversely, gallinaceous birds routinely experience mass and volume fluctuations near the neck and chest as a result of filling and evacuating the crop during daily foraging, possibly making the neck a more natural location to carry additional weight. Therefore, our results provide some support for previously published hypotheses on effects of device placement on grouse (Amstrup 1980, Small and Rusch 1985). Other studies have observed negative effects of transmitters on galliform birds based on design, attachment type, and placement. In studies that directly compared effects of necklace-style collars to backpacks, collars often have had less impact than backpacks (Small and Rusch 1985, Marcstrom et al. 1989). A comprehensive review reported negative effects of all radio-transmitter designs combined occurred in 47% of studies, whereas negative effects of backpack transmitters alone occurred in 68% of studies (Withey et al. 2001). In a comparison of Ring-necked Pheasants (Phasianus colchicus) marked with backpacks, necklaces, and leg bands only, Marcstrom et al. (1989) observed the lowest survival with backpacks and no difference between necklaces and leg bands. Additionally, although we did not evaluate impacts of the solar panels directly, the support for the device-only model indicated that the solar panel may be a major causative factor. Burger et al. (1991) observed decreased survival in Greater Prairie-Chickens (Tympanuchus cupido) with solar-powered VHF transmitters with a reflective surface compared to non-solar devices. It was possible that solar panels on GPS devices in our study contributed to higher mortality through a reduction of crypsis. Similar treatment–control survival studies using GPS rump-mounted designs with different solar designs for sage-grouse would be highly beneficial. Possible Effects on Behavior Tracking devices have been reported to have negative impacts on animal behavior that may drive indirect negative effects on survival and reproduction (Hines and Zwickel 1985, White and Garrott 1990), and different attachment methods may have both advantages and disadvantages. Specific to sage-grouse, Gibson et al. (2013) observed decreased lek attendance, while Fremgen et al. (2017) observed altered vocalizations for male sage-grouse marked with VHF necklaces compared to unmarked or leg-banded males. Both studies suggested the differences in behavior could be due to the necklace interfering with the esophageal air sacs, which supported previous recommendations to not place collars on male grouse having air sacs (Pyrah 1970, Amstrup 1980). Nevertheless, collars have been commonly fitted to male sage-grouse since the 1980s. Thus, Fremgen et al. (2017) suggested rump-mounted telemetry devices should be evaluated as an alternative method for attaching transmitters to male sage-grouse. However, our results suggest a higher survival cost to both males and females fitted with GPS devices, and higher hazard for adult females as GPS proportion of body mass increases. Hence, while rump mounts may have less impact on some behaviors than necklaces, they do not appear to be an obviously superior attachment method, although we concede that more research is needed on the influences of rump-mount attachment on survival and behavior relative to the characteristics of various devices. Information on movement and space use may be valuable despite higher mortality rates for radio-tagged birds than untagged birds (Marks and Marks 1987). GPS devices have proven to be even more useful than VHF units for providing precise movement analyses for sage-grouse, for example quantifying fine scale incubation recess movements (Dudko et al. 2019), movement patterns relative to vegetative features (Prochazka et al. 2017), and lek attendance rates of individual males (Wann et al. 2019). Fedy et al. (2012) did not observe differences in movement rates between VHF- and GPS-marked sage-grouse but suggested further study. Small and Rusch (1985) compared VHF devices attached to Ruffed Grouse (Bonasa umbellus) either as backpacks or as collars, and while they observed greater survival with collars than with backpacks, they did not observe differences in movement. Similarly, Kesler et al. (2014) determined that while survival of waterfowl marked with satellite devices could be impacted indirectly through reduced seasonal weight gain, movements of marked birds were similar to those of unmarked birds. Nevertheless, no one has estimated the effect of GPS devices on sage-grouse movement behavior, altered body condition, or predation risk. In addition, device-induced effects on resource selection and subsequent survival could manifest during critical life history stages such as nesting and brood-rearing. Hence, linked selection–survival analyses might erroneously indicate maladaptive selection for resources with elevated mortality risk owing to device rather than true ecological effects. Caveats and Research Considerations Survival estimates from GPS-marked birds had wider posterior distributions compared to those from VHF, which could be explained by fewer samples of GPS relative to VHF across sites. Still, our results were informed by >350 GPS-marked sage-grouse, and we know of no similar comparisons published with sample sizes this large. The lack of previously published studies on GPS effects on sage-grouse may stem from a “file drawer effect” (i.e. not publishing nonsignificant results; Sterling 1959, Rosenthal 1979, Barron et al. 2010) owing to small sample sizes and large variability. Moreover, tracking devices with various attachment designs have been found to have negative effects in some studies and no effects in others (Caizergues and Ellison 1998, Withey et al. 2001), and debate over the biological significance of possible bias remains contentious (Guthery and Lusk 2004, Folk et al. 2007). Hence, more treatment–control studies, such as ours, are necessary to draw firm inference about GPS effects on sage-grouse. It is important that researchers publish parameter estimates for VHF- and GPS-marked birds from the same study, regardless of statistical significance, so those estimates can be used in future meta-analyses (Barron et al. 2010). In our study, the consistency of differences between GPS and VHF devices across most sex, age, and seasonal estimates provided compelling evidence for a general trend of decreased survival of GPS-marked sage-grouse. The limited cases where strong differences did not occur (i.e. posterior distributions overlapped by device type) can be explained from a life history perspective. For example, overlapping distributions for yearling compared to adult females (Figure 5) may relate to fitness costs of reproduction, whereby adult females are more likely to reproduce successively and consequently have poorer body condition post-reproduction (Blomberg et al. 2013b). The lack of device type effects for males during spring and winter likely relates to conspicuous courtship displays on leks and grouping behavior, respectively, that may result in relatively equal chances of predation. Although our results reveal significant reductions in survival associated with GPS devices, we offer the following caveats. First, lower survival probabilities estimated for GPS-marked sage-grouse in our study indicate that the welfare of individual marked birds needs to be evaluated relative to research objectives. For example, researchers may be enticed by the ease of remotely monitoring GPS-marked sage-grouse and the corresponding increase in location data, but if birds suffer higher mortality, then this would need ethical consideration and an understanding of potential lowering of data quality. Even low PBM is correlated with lower mortality for birds carrying GPS devices, so benefits of research must be weighed against increased risk to individuals. We note, however, that these impacts in the low sample sizes typical of GPS sage-grouse studies likely have a minimal effect on the overall population. Regardless, future design and attachment methods would need testing to evaluate improved efficacy of these devices. Second, we assumed that VHF-derived estimates represented baseline survival because they account for a majority of demographic estimates for sage-grouse (Taylor et al. 2012), but we could not directly account for any possible bias associated with VHF use. Data from less-invasive methods such as direct observation (Baumgardt et al. 2017), banding studies (Zablan et al. 2003, Hagen et al. 2018), geolocators (Raybuck et al. 2017), and noninvasive sampling of biomarkers such as genetic assays (Baumgardt et al. 2013, Cross et al. 2017, Row et al. 2018) and stable isotopes (Hobson 2005) could yield parameter estimation with little to no marking-induced biases. Although these methods should incur little to no decrease in longer-term survivorship, they also yield coarser-grain data or require multiple resight or recapture occasions that might affect detection probabilities. Therefore, there are gradients of animal welfare and quality and quantity of data that these noninvasive methods, as well as VHF and GPS tracking devices, fall within that researchers can use to weigh costs and benefits. Third, advantages and disadvantages of currently available models of VHF and GPS devices for sage-grouse represent additional research design tradeoffs. For example, necklace-style VHF device attachment is simpler and less expensive than GPS devices, allowing for larger sample sizes, which may provide more precise and less biased demographic estimates. However, VHF device data collection is more labor intensive and has lower temporal resolution, leading to coarser results. VHF application is also less feasible and data become more limited as terrain ruggedness increases and technician skill decreases. Data can also be lost entirely if birds move to new areas where VHF signals cannot be heard. In contrast, GPS devices require minimal labor subsequent to initial deployment, and yield data when human access is limited or when birds make large movements to areas outside of normal monitoring boundaries. Directions for Future Research Future research on tracking devices to determine precise mechanisms of effects is needed to inform improvements in future device designs. Additionally, PBM-based guidelines for use following sage-grouse capture can be developed and improved through experimental evaluation of designs. For example, fitting rump-mounted VHF devices, fitting otherwise similar devices both with and without solar panels, fitting collars and backpacks of similar mass, and fitting collars or backpacks with a variety of masses. Analyses of potential differential device effects as PBM changes should also be explored across sex and age classes with adequate data. Researchers could also attempt to recapture VHF- and GPS-marked individuals to measure changes in body condition or stress hormones relative to the original capture. Additionally, more research is needed on changes to relative probability of different causes of mortality, so inference into mortality factors is not biased. Directly assessing differences in sage-grouse behavior as a function of device type would also help researchers weigh the tradeoffs between tracking devices. Useful behavioral analyses include determining differences in resource selection parameters, movement rates, home range size, lek attendance, and incubation behavior using nest cameras. Researchers studying survival using both GPS and VHF devices to investigate environmental covariate effects on survival might also consider including tracking device type in model structures to account for differences while estimating other effects. If differences are observed, GPS-marked birds may need to be dropped from the analysis. Conversely, if GPS devices or attachment techniques could be designed that reduce impacts, they would provide greater access to fine-resolution location data that allow investigation of specific behavioral movement patterns and space use, which would otherwise not be possible with typically sparse VHF data (Prochazka et al. 2017, Dudko et al. 2019, Wann et al. 2019). Therefore, we caution that advances in telemetry technology should not guide the research question; rather, the question would be better answered with appropriate technology and caveats taken into account. ACKNOWLEDGMENTS We thank numerous field technicians for their devoted efforts toward field collection, especially K. Andrle, J. Dudko, R. Kelble, and S. Mathews. We thank E. Blomberg for input on rationale for hypotheses and predictions. K. LeSage (GeoTrak) provided expert technical advice on GPS-PTT configuration. Constructive reviews of previous manuscripts were provided by 2 anonymous reviewers, R. Gutiérrez, and T. Kimball. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. Funding Statement: Field research was supported by multiple federal, state, university, and private partners that include Bureau of Land Management, U.S. Fish and Wildlife Service, U.S. Forest Service, Nevada Department of Wildlife, California Department of Fish and Wildlife, Great Basin Bird Observatory, Idaho State University, Ormat Technologies, and GRP Minerals. Ethics statement: All activities within this manuscript have followed appropriate U.S. Geological Survey, Western Ecological Research Center animal care and use committee guidelines (WERC-2015-02, approved 12/15/2015). Author Contributions: P.S.C., M.L.C., B.G.P., and D.J.D. conceived idea and design. P.S.C. and B.G.P. collected the data. B.G.P. and P.S.C. analyzed the data. J.P.S., P.S.C., M.A.R., B.G.P., D.J.D., and M.L.C. wrote and edited the manuscript. P.S.C., M.L.C., M.A.R., and D.J.D. contributed resources and funding. LITERATURE CITED Amstrup , S. C . ( 1980 ). A radio-collar for game birds . The Journal of Wildlife Management 44 : 214 – 217 . Google Scholar Crossref Search ADS WorldCat Barron , D. G. , J. D. Brawn, and P. J. Weatherhead ( 2010 ). Meta-analysis of transmitter effects on avian behaviour and ecology . Methods in Ecology and Evolution 1 : 180 – 187 . Google Scholar Crossref Search ADS WorldCat Baumgardt , J. A. , C. S. Goldberg, K. P. Reese, J. W. Connelly, D. D. Musil, E. O. Garton, and L. P. Waits ( 2013 ). A method for estimating population sex ratio for sage-grouse using noninvasive genetic samples . Molecular Ecology Resources 13 : 393 – 402 . Google Scholar Crossref Search ADS PubMed WorldCat Baumgardt , J. A. , K. P. Reese, J. W. Connelly, D. D. Musil, and E. O. Garton ( 2017 ). Visibility bias for sage-grouse lek counts . Wildlife Society Bulletin 41 : 461 – 470 . Google Scholar Crossref Search ADS WorldCat Bedrosian , B. , and D. Craighead ( 2007 ). Evaluation of techniques for attaching transmitters to Common Raven nestlings . Northwestern Naturalist 88 : 1 – 6 . Google Scholar Crossref Search ADS WorldCat Blomberg , E. J. , D. Gibson, J. S. Sedinger, M. L. Casazza, and P. S. Coates ( 2013a ). Intraseasonal variation in survival and probable causes of mortality in Greater Sage-Grouse Centrocercus urophasianus . Wildlife Biology 19 : 347 – 357 . Google Scholar Crossref Search ADS WorldCat Blomberg , E. J. , J. S. Sedinger, D. V. Nonne, and M. T. Atamian ( 2013b ). Seasonal reproductive costs contribute to reduced survival of female Greater Sage-Grouse . Journal of Avian Biology 44 : 149 – 158 . Google Scholar Crossref Search ADS WorldCat Boag , D. A . ( 1972 ). Effect of radio packages on behavior of captive Red Grouse . The Journal of Wildlife Management 36 : 511 – 518 . Google Scholar Crossref Search ADS WorldCat Brander , R. B . ( 1968 ). A radio-package harness for game birds . The Journal of Wildlife Management 32 : 630 – 632 . Google Scholar Crossref Search ADS WorldCat Braun , C. E. , and M. A. Schroeder ( 2015 ). Age and sex identification from wings of sage-grouse . Wildlife Society Bulletin 39 : 182 – 187 . Google Scholar Crossref Search ADS WorldCat Bureau of Land Management ( 2015 ). Notice of availability of the record of decision and approved resource management plan amendments for the Great Basin Region Greater Sage-Grouse Sub-Regions of Idaho and Southwestern Montana; Nevada and Northeastern California; Oregon; and Utah . Federal Register 80 : 57633 – 57635 . OpenURL Placeholder Text WorldCat Burger , L. W. , M. R. Ryan, D. P. Jones, and A. P. Wywialowski ( 1991 ). Radio transmitters bias estimation of movements and survival . The Journal of Wildlife Management 55 : 693 – 697 . Google Scholar Crossref Search ADS WorldCat Cagnacci , F. , L. Boitani, R. A. Powell, and M. S. Boyce ( 2010 ). Animal ecology meets GPS-based radiotelemetry: A perfect storm of opportunities and challenges . Philosophical Transactions of the Royal Society B 365 : 2157 – 2162 . Google Scholar Crossref Search ADS WorldCat Caizergues , A. , and L. N. Ellison ( 1998 ). Impact of radio-tracking on Black Grouse Tetrao tetrix reproductive success in the French Alps . Wildlife Biology 4 : 205 – 212 . Google Scholar Crossref Search ADS WorldCat Coates , P. S. , B. G. Prochazka, M. A. Ricca, G. T. Wann, C. L. Aldridge, S. E. Hanser, K. E. Doherty, M. S. O’Donnell, D. R. Edmunds, and S. P. Espinosa ( 2017 ). Hierarchical population monitoring of Greater Sage-Grouse (Centrocercus urophasianus) in Nevada and California—Identifying populations for management at the appropriate spatial scale. US Geological Survey Open-File Report No. 2017-1089 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Connelly , J. W. , S. T. Knick, C. E. Braun, W. L. Baker, E. A. Beever, T. Christiansen, K. E. Doherty, E. O. Garton, S. E. Hanser, D. H. Johnson, et al. ( 2011a ). Conservation of Greater Sage-Grouse: A synthesis of current trends and future management. In Greater Sage-Grouse: Ecology and Conservation of a Landscape Species and its Habitat (S. T. Knick, and J. W. Connelly, Editors) . Studies in Avian Biology 38 : 549 – 563 . Google Scholar OpenURL Placeholder Text WorldCat Connelly , J. W. , E. T. Rinkes, and C. E. Braun ( 2011b ). Characteristics of Greater Sage-Grouse habitats. In Greater Sage-Grouse: Ecology and Conservation of a Landscape Species and its Habitat (S. T. Knick and J. W. Connelly, Editors) . Studies in Avian Biology 38 : 69 – 83 . Google Scholar OpenURL Placeholder Text WorldCat Cross , T. B. , D. E. Naugle, J. C. Carlson, and M. K. Schwartz ( 2017 ). Genetic mark recapture identifies long-distance breeding dispersal in Greater Sage-Grouse (Centrocercus urophasianus) . The Condor: Ornithological Applications 119 : 155 – 166 . Google Scholar OpenURL Placeholder Text WorldCat Dudko , J. E. , P. S. Coates, and D. J. Delehanty ( 2019 ). Movements of female sage grouse Centrocercus urophasianus during incubation recess . Ibis 161 : 222 – 229 . Google Scholar Crossref Search ADS WorldCat Dzialak , M. R. , C. V. Olson, S. M. Harju, S. L. Webb, J. P. Mudd, J. B. Winstead, and L. D. Hayden-Wing ( 2011 ). Identifying and prioritizing Greater Sage-Grouse nesting and brood-rearing habitat for conservation in human-modified landscapes . PLOS One 6 : e26273 . Google Scholar Crossref Search ADS PubMed WorldCat Fair , J. M. , E. Paul, and J. Jones (Editors) ( 2010 ). Guidelines to the use of wild birds in research, third edition . Ornithological Council, Washington, D.C. https://birdnet.org/info-for-ornithologists/guidelines-to-the-use-of-wild-birds-in-research/ Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Fedy , B. C. , C. L. Aldridge, K. E. Doherty, M. O’Donnell, J. L. Beck, B. Bedrosian, M. J. Holloran, G. D. Johnson, N. W. Kaczor, C. P. Kirol, et al. ( 2012 ). Interseasonal movements of Greater Sage-Grouse, migratory behavior, and an assessment of the core regions concept in Wyoming . The Journal of Wildlife Management 76 : 1062 – 1071 . Google Scholar Crossref Search ADS WorldCat Folk , T. H. , J. B. Grand, W. E. Palmer, J. P. Carroll, D. C. Sisson, T. M. Terhune, S. D. Wellendorf, and H. L. Stribling ( 2007 ). Estimates of survival from radiotelemetry: A response to Guthery and Lusk . The Journal of Wildlife Management 71 : 1027 – 1033 . Google Scholar Crossref Search ADS WorldCat Fremgen , M. R. , D. Gibson, R. L. Ehrlich, A. H. Krakauer, J. S. Forbey, E. J. Blomberg, J. S. Sedinger, and G. L. Patricelli ( 2017 ). Necklace-style radio-transmitters are associated with changes in display vocalizations of male Greater Sage-Grouse . Wildlife Biology 2017: wlb.00236 . Google Scholar OpenURL Placeholder Text WorldCat Gelman , A . ( 2014 ). Bayesian Data Analysis, third edition . CRC Press, Boca Raton, FL, USA . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gessaman , J. A. , and K. A. Nagy ( 1988 ). Transmitter load affects the flight speed and metabolism of homing pigeons . The Condor 90 : 662 – 668 . Google Scholar Crossref Search ADS WorldCat Gibson , D. , E. J. Blomberg, G. L. Patricelli, A. H. Krakauer, M. T. Atamian, and J. S. Sedinger ( 2013 ). Effects of radio collars on survival and lekking behavior of male Greater Sage-Grouse . The Condor 115 : 769 – 776 . Google Scholar Crossref Search ADS WorldCat Giesen , K. M. , T. J. Schoenberg, and C. E. Braun ( 1982 ). Methods for trapping Sage Grouse in Colorado . Wildlife Society Bulletin 10 : 224 – 231 . Google Scholar OpenURL Placeholder Text WorldCat Gilmer , D. S. , I. J. Ball, L. M. Cowardin, and J. H. Riechmann ( 1974 ). Effects of radio packages on wild ducks . The Journal of Wildlife Management 38 : 243 – 252 . Google Scholar Crossref Search ADS WorldCat Guthery , F. S. , and J. J. Lusk ( 2004 ). Radiotelemetry studies: Are we radio-handicapping Northern Bobwhites? Wildlife Society Bulletin 32 : 194 – 201 . Google Scholar Crossref Search ADS WorldCat Gutierrez , R. G . ( 2002 ). Parametric frailty and shared frailty survival models . Stata Journal 2 : 22 – 44 . Google Scholar Crossref Search ADS WorldCat Hagen , C. A . ( 2011 ). Predation on Greater Sage-Grouse. In Greater Sage-Grouse: Ecology and Conservation of a Landscape Species and its Habitat (S. T. Knick and J. W. Connelly, Editors) . Studies in Avian Biology 38 : 95 – 100 . Google Scholar OpenURL Placeholder Text WorldCat Hagen , C. A. , J. E. Sedinger, and C. E. Braun ( 2018 ). Estimating sex-ratio, survival, and harvest susceptibility in Greater Sage-Grouse: Making the most of hunter harvests . Wildlife Biology 2018 : wlb.00362 . Google Scholar Crossref Search ADS WorldCat Halstead , B. J. , G. D. Wylie, P. S. Coates, P. Valcarcel, and M. L. Casazza ( 2012 ). Bayesian shared frailty models for regional inference about wildlife survival: Regional inference for wildlife survival . Animal Conservation 15 : 117 – 124 . Google Scholar Crossref Search ADS WorldCat Herzog , P. W . ( 1979 ). Effects of radio-marking on behavior, movements, and survival of Spruce Grouse . The Journal of Wildlife Management 43 : 316 – 323 . Google Scholar Crossref Search ADS WorldCat Hines , J. E. , and F. C. Zwickel ( 1985 ). Influence of radio packages on young Blue Grouse . The Journal of Wildlife Management 49 : 1050 – 1054 . Google Scholar Crossref Search ADS WorldCat Hobson , K. A . ( 2005 ). Using stable isotopes to trace long-distance dispersal in birds and other taxa . Diversity and Distributions 11 : 157 – 164 . Google Scholar Crossref Search ADS WorldCat Karl , B. J. , and M. N. Clout ( 1987 ). An improved radio transmitter harness with a weak link to prevent snagging . Journal of Field Ornithology 58 : 73 – 77 . Google Scholar OpenURL Placeholder Text WorldCat Kenward , R. E . ( 2001 ). A Manual for Wildlife Radio Tagging, revised edition . Academic Press, San Diego, CA, USA . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kesler , D. C. , A. H. Raedeke, J. R. Foggia, W. S. Beatty, E. B. Webb, D. D. Humburg, and L. W. Naylor ( 2014 ). Effects of satellite transmitters on captive and wild Mallards: Effects of satellite telemetry harness . Wildlife Society Bulletin 38 : 557 – 565 . Google Scholar Crossref Search ADS WorldCat Marcstrom , V. , R. E. Kenward, and M. Karlbom ( 1989 ). Survival of Ring-Necked Pheasants with backpacks, necklaces, and leg bands . The Journal of Wildlife Management 53 : 808 – 810 . Google Scholar Crossref Search ADS WorldCat Marks , J. S. , and V. S. Marks ( 1987 ). Influence of radio collars on survival of Sharp-tailed Grouse . The Journal of Wildlife Management 51 : 468 – 471 . Google Scholar Crossref Search ADS WorldCat Marshall , W. H . ( 1963 ). Studies of movements, behavior and activities of Ruffed Grouse using radio telemetry techniques . University of Minnesota Progress Report, University of Minnesota , St. Paul, MN, USA . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Millspaugh , J. J. , and J. M. Marzluff ( 2001 ). Radio-tracking and animal populations: Past trends and future needs . In Radio Tracking and Animal Populations ( J. J. Millspaugh and J. M. Marzluff, Editors). Academic Press , San Diego, CA, USA . pp. 383 – 393 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Millspaugh , J. J. , D. C. Kesler, R. W. Kays, R. A. Gitzen, J. H. Schulz, C. T. Rota, C. M. Bodinof, J. L. Belant, and B. J. Keller ( 2012 ). Wildlife radiotelemetry and remote monitoring . In The Wildlife Techniques Manual: Research Volume 1, seventh edition ( N. J. Silvy, Editor). Johns Hopkins University Press , Baltimore, MD, USA . pp. 258 – 283 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Moynahan , B. J. , M. S. Lindberg, and J. W. Thomas ( 2006 ). Factors contributing to process variance in annual survival of female Greater Sage-Grouse in Montana . Ecological Applications 16 : 1529 – 1538 . Google Scholar Crossref Search ADS PubMed WorldCat Murray , D. L . ( 2006 ). On improving telemetry-based survival estimation . The Journal of Wildlife Management 70 : 1530 – 1543 . Google Scholar Crossref Search ADS WorldCat Murray , D. L. , and M. R. Fuller ( 2000 ). A critical review of the effects of marking on the biology of vertebrates . In Research Techniques in Animal Ecology: Controversies and Consequences ( L. Boitani and T. K. Fuller, Editors). Columbia University Press , New York, NY, USA . pp. 15 – 64 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Oyler-McCance , S. J. , M. L. Casazza, J. A. Fike, and P. S. Coates ( 2014 ). Hierarchical spatial genetic structure in a distinct population segment of Greater Sage-Grouse . Conservation Genetics 15 : 1299 – 1311 . Google Scholar Crossref Search ADS WorldCat Paton , P. W. C. , C. J. Zabel, D. L. Neal, G. N. Steger, N. G. Tilghman, and B. R. Noon ( 1991 ). Effects of radio tags on Spotted Owls . The Journal of Wildlife Management 55 : 617 – 622 . Google Scholar Crossref Search ADS WorldCat Pennycuick , C. J. , P. L. F. Fast, N. Ballersätdt, and N. Rattenborg ( 2012 ). The effect of an external transmitter on the drag coefficient of a bird’s body, and hence on migration range, and energy reserves after migration . Journal of Ornithology 153 : 633 – 644 . Google Scholar Crossref Search ADS WorldCat Pietz , P. J. , G. L. Krapu, R. J. Greenwood, and J. T. Lokemoen ( 1993 ). Effects of harness transmitters on behavior and reproduction of wild Mallards . The Journal of Wildlife Management 57 : 696 – 703 . Google Scholar Crossref Search ADS WorldCat Plummer , M . ( 2015 ). rjags: Bayesian graphical models using MCMC. R package version 3–15 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC PRISM Climate Group ( 2015 ). PRISM climate data . Oregon State University, Corvallis, OR, USA . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Prochazka , B. G. , P. S. Coates, M. A. Ricca, M. L. Casazza, K. B. Gustafson, and J. M. Hull ( 2017 ). Encounters with pinyon-juniper influence riskier movements in Greater Sage-Grouse across the Great Basin . Rangeland Ecology & Management 70 : 39 – 49 . Google Scholar Crossref Search ADS WorldCat Pyrah , D . ( 1970 ). Poncho markers for game birds . The Journal of Wildlife Management 34 : 466 – 467 . Google Scholar Crossref Search ADS WorldCat Raybuck , D. W. , J. L. Larkin, S. H. Stoleson, and T. J. Boves ( 2017 ). Mixed effects of geolocators on reproduction and survival of Cerulean Warblers, a canopy-dwelling, long-distance migrant . The Condor: Ornithological Applications 119 : 289 – 297 . Google Scholar OpenURL Placeholder Text WorldCat R Core Team ( 2014 ). R: A language and environment for statistical computing . R Foundation for Statistical Computing, Vienna, Austria . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rosenthal , R . ( 1979 ). The file drawer problem and tolerance for null results . Psychological Bulletin 86 : 638 – 641 . Google Scholar Crossref Search ADS WorldCat Row , J. R. , K. E. Doherty, T. B. Cross, M. K. Schwartz, S. J. Oyler-McCance, D. E. Naugle, S. T. Knick, and B. C. Fedy ( 2018 ). Quantifying functional connectivity: The role of breeding habitat, abundance, and landscape features on range-wide gene flow in sage-grouse . Evolutionary Applications 11 : 1305 – 1321 . Google Scholar Crossref Search ADS PubMed WorldCat 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, et al. ( 2004 ). Distribution of sage-grouse in North America . The Condor 106 : 363 – 376 . Google Scholar Crossref Search ADS WorldCat Schroeder , M. A. , J. R. Young, and C. E. Braun ( 1999 ). Greater Sage-Grouse (Centrocercus urophasianus) . In The Birds of North America (A. F. Poole and F. B. Gill, Editors). Cornell Lab of Ornithology, Ithaca, NY, USA. https://doi.org/10.2173/bna.425 Google Scholar OpenURL Placeholder Text WorldCat Small , R. J. , and D. H. Rusch ( 1985 ). Backpacks vs. ponchos: Survival and movements of radio-marked Ruffed Grouse . Wildlife Society Bulletin 13 : 163 – 165 . Google Scholar OpenURL Placeholder Text WorldCat Smith , K. T. , C. P. Kirol, J. L. Beck, and F. C. Blomquist ( 2014 ). Prioritizing winter habitat quality for Greater Sage-Grouse in a landscape influenced by energy development . Ecosphere 5 : 1 – 20 . Google Scholar Crossref Search ADS WorldCat Sterling , T. D . ( 1959 ). Publication decisions and their possible effects on inferences drawn from tests of significance—or vice versa . Journal of the American Statistical Association 54 : 30 – 34 . Google Scholar OpenURL Placeholder Text WorldCat Taylor , R. L. , B. L. Walker, D. E. Naugle, and L. S. Mills ( 2012 ). Managing multiple vital rates to maximize Greater Sage-Grouse population growth . The Journal of Wildlife Management 76 : 336 – 347 . Google Scholar Crossref Search ADS WorldCat Thorn , T. D. , R. B. Emery, D. W. Howerter, J. H. Devries, and B. L. Joynt ( 2005 ). Use of radio-telemetry to test for investigator effects on nesting Mallards, Anas platyrhynchos . The Canadian Field-Naturalist 119 : 541 – 545 . Google Scholar Crossref Search ADS WorldCat Tomkiewicz , S. M. , M. R. Fuller, J. G. Kie, and K. K. Bates ( 2010 ). Global positioning system and associated technologies in animal behaviour and ecological research . Philosophical Transactions of the Royal Society B 365 : 2163 – 2176 . Google Scholar Crossref Search ADS WorldCat U.S. Fish and Wildlife Service ( 2015 ). 50 CFR Part 17 endangered and threatened wildlife and plants; 12-month findings for petitions to list the Greater Sage-Grouse (Centrocercus urophasianus) as threatened or endangered . Federal Register 80:59858–59942. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 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 . Google Scholar OpenURL Placeholder Text WorldCat Wann , G. T. , P. S. Coates, B. G. Prochazka, J. P. Severson, A. P. Monroe, and C. L. Aldridge ( 2019 ). Assessing lek attendance of male Greater Sage-Grouse using fine-resolution GPS data: Implications for population monitoring of lek mating grouse . Population Ecology 61 : 183 – 197 . Google Scholar Crossref Search ADS WorldCat Warner , R. E. , and S. L. Etter ( 1983 ). Reproduction and survival of radio-marked hen Ring-necked Pheasants in Illinois . The Journal of Wildlife Management 47 : 369 – 375 . Google Scholar Crossref Search ADS WorldCat Watanabe , S . ( 2010 ). Asymptotic equivalence of Bayes cross validation and widely applicable information criterion in singular learning theory . Journal of Machine Learning Research 11 : 3571 – 3594 . Google Scholar OpenURL Placeholder Text WorldCat White , G. C. , and R. A. Garrott ( 1990 ). Analysis of Wildlife Radio-Tracking Data . Academic Press, San Diego, CA, USA . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Winterstein , S. R. , K. H. Pollock, and C. M. Bunck ( 2001 ). Analysis of survival data from radiotelemetry studies . In Radio Tracking and Animal Populations ( J. J. Millspaugh and J. M. Marzluff, Editors). Academic Press , San Diego, CA, USA . pp. 351 – 380 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Withey , J. C. , T. D. Bloxton, and J. M. Marzluff ( 2001 ). Effects of tagging and location error in wildlife radiotelemetry studies . In Radio Tracking and Animal Populations ( J. J. Millspaugh and J. M. Marzluff, Editors). Academic Press , San Diego, CA, USA . pp. 43 – 75 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zablan , M. A. , C. E. Braun, and G. C. White ( 2003 ). Estimation of Greater Sage-Grouse survival in North Park, Colorado . The Journal of Wildlife Management 67 : 144 – 154 . Google Scholar Crossref Search ADS WorldCat Published by Oxford University Press for the American Ornithological Society 2019. This work is written by (a) US Government employee(s) and is in the public domain in the US. Published by Oxford University Press for the American Ornithological Society 2019.
TI - Global positioning system tracking devices can decrease Greater Sage-Grouse survival
JF - Ornithological Applications
DO - 10.1093/condor/duz032
DA - 2019-08-26
UR - https://www.deepdyve.com/lp/oxford-university-press/global-positioning-system-tracking-devices-can-decrease-greater-sage-PBS3I3pTRI
VL - 121
IS - 3
DP - DeepDyve
ER -