https://cpw.cvlcollections.org/files/original/186fd3247fb606ab7d05f9dd327542d8.pdf 05779bdb970ba099661a51a115561e82 PDF Text Text WTPT 2011 Seglund WHITE-TAILED PTARMIGAN OCCUPANCY REPORT 2011 AMY SEGLUND WILDLIFE CONSERVATION COORDINATOR COLORADO PARKS AND WILDLIFE 1 �WTPT 2011 Seglund Introduction In 2010, the white-tailed ptarmigan (Lagopus leucura; WTPT) was petitioned to be listed as threatened under the Endangered Species Act. The petitioners cited climate change and its impacts to a species dependent on alpine habitats as the main threat to the viability of WTPT populations. Colorado supports the largest population of WTPT in the lower 48 states. Thus, if the species does become listed, Colorado Parks and Wildlife (CPW) will be responsible for the bulk of the protection and management of the species, yet understanding of the species distribution and demographic parameters is limited. To develop a baseline distribution and inventory of the species, statewide occupancy surveys were conducted in 2011. These surveys will be used to respond to the petition to list the WTPT and to monitor populations as climate change and other disturbances impact the viability of the species over time. Objectives for 2011 Occupancy Surveys 1. Assess status of the WTPT statewide using occupancy sampling. The baseline data collected from surveys will help inform future monitoring efforts by supplying derived estimates of detection probabilities and percent occupancy which can be used to generate a monitoring program with sufficient precision for adequate management of the species. 2. Use field collected data to test accuracy of the current Predicted Range Model to aid in site-selection for a long-term monitoring program, assess potential habitat and connectivity of populations and investigate potential threats to the species. 3. Collect feathers to assess genetic variability and connectivity of populations using DNA from samples. Methods The primary parameter of interest in the monitoring program was the number of occupied units measured within the predicted summer range of the WTPT. The predicted range was defined using the following criteria: all areas > 3292 m in elevation and Colorado GAP vegetation types that included Mixed Tundra, Meadow Tundra, Prostrate Shrub Tundra, Bare Ground Tundra, Exposed Rock, Shrub Dominated Wetland/Riparian, and Graminoid/Forb Dominated Wetland (Hoffman 2006). Colorado GAP data was used instead of SWReGAP because information was needed for both the states of Colorado and Wyoming during development of the model. Once the model was created, CPW District Wildlife Managers (DWM) and area biologists made edits in Species Activity Mapping sessions to help refine the information. The sampling frame used to select a population of 60 plots for sampling statewide was derived from overlaying a grid of 5.3 x 4.6 km plots across the range of the WTPT. Only plots that contained a minimum of 50% suitable habitat within their boundaries based on the predicted range model were selected for sampling. Within each of these larger plots, a cluster of 4 plots (cluster; 200 m x 200 m; 100 meters distance apart) were randomly positioned within modeled 2 �WTPT 2011 Seglund habitat. The reasoning for the different spatial scale surveys was to investigate the size of plot needed to achieve an adequate occupancy rate for future monitoring, identify protocols that can be replicated, and to evaluate spatial variation measurements for the species that can be locally abundant but rare across its entire range. The sampling frame was divided geographically into three strata with 20 plots per strata; sites north of Interstate 70, sites between Interstate70 and Highway 50, and sites south of Highway 50. Based on long term weather station data, the northern strata has been shown to receive more consistent snow fall and less consistent summer monsoonal moisture and conversely, the southern strata has had less consistent winter moisture and more consistent summer monsoons. Two independent surveys were completed per plot to estimate the detection probability for WTPT. Estimation of this parameter is critical for occupancy modeling. Because detection probabilities may be a function of covariates, a number of variables considered important to detection probability estimation were collected on plots. These covariates included: 1) time of day survey was conducted; 2) time spent surveying for WTPT on plots; 3) area covered during survey efforts; 4) the number of days the survey was conducted; 5) weather data; 6) bird detection technique, and 7) habitat type detection occurred in. Large Plots: For WTPT occupancy surveys of large plots, only suitable habitat within the plot boundary was surveyed. Suitable habitat is considered alpine at or above treeline (Braun 1971, Braun et al 1993 as cited in Hoffman 2006). The most important vegetation features used by WTPT are presence of willows (Salix spp.), rocky areas (large rocks of 30 cm diameter comprising 50% of ground cover), moist alpine meadows and snow fields (Choate 1963, Braun 1971, Frederick and Gutierrez 1992 as cited in Hoffman 2006). Willows are important because they serve as the primary food source from late fall to early summer whereas rocky areas and dry tundra lying near snow fields and other moist areas become the most commonly used sites from mid-summer to early fall. WTPT use high rocky windswept ridges and mountain tops where slopes are moderate to low angle. Cliffs and steep slopes are rarely used other than for escape routes. To identify potential habitat to survey, observers reviewed the predicted range model overlaid on satellite imagery in the Google Earth files provided. This helped guide observers to the location of patches to survey. In addition, observers used topographic maps and aerial photos to help decipher potential habitat and to determine access to the habitat within plots. Not all potential habitat within plots was accessible. The maximum amount of time the observer was allowed to survey suitable habitat within a plot was 12 hours (this amount of time did not include the time it took to hike to suitable habitat patches in a plot or the time it took to get to a plot). This 12 hour time limit could occur over several days, but could not extend past a consecutive 4 day time period. Surveys in suitable habitat were thought to be most productive in early morning hours (6-10am), but surveys could be completed anytime of the day. Each survey effort was conducted by a pair of observers. This was to maintain consistency among surveys and for safety. 3 �WTPT 2011 Seglund To increase our detection probability for a cryptic species, playbacks were used to attract birds within plots during surveys. Playbacks of chicks have been shown to double the number of females detected in study areas (Braun et al. 1973). Playbacks of chick calls assist in detection of females with chicks or females with recently failed reproductive attempts. Females will respond to chick distress calls by clucking, making themselves visible by coming into the open, jumping onto a boulder, or by running/flying to the call source (Braun et al. 1973). Females readily respond to playbacks from mid-July to late August/early September when chicks range in age from 2 to 7 weeks of age (Braun et al. 1973, Gregory Wann pers. comm. 2011). Male territorial calls were used to attract male birds. Males will readily respond to territorial calls by responding with their own call or flying to the source of the call. A positive detection of a WTPT was either a visual of a bird or an auditory response of a male call. Positive detections did not include scat or feathers found on the plot. When a WTPD was detected, a UTM coordinate was collected from the location and the number of birds, sex of birds (if it could be determined) and age (chick or adult) of birds observed were recorded. Photos were taken at sites where WTPTs were detected in order to help characterize the site. In addition, information on surrounding vegetation, presence of rocks, talus and permanent snow fields within 100 m of detections were collected as well as evidence of human disturbance. Disturbances noted included presence of hiking trails, dirt roads, paved roads, mining operations, reservoirs or other human recreational activities such as camping. A survey of a plot was considered complete once all suitable habitats within the large plot had been surveyed, the time limit (12 hours) or day limit for surveying was reached, or when a positive detection of WTPT occurred on the plot. Small Plots: Three sets of a cluster of 4 small plots were randomly positioned within the large plot boundary. The observer assigned to a plot selected whichever of these available clusters to sample based on available habitat and accessibility. Cluster plot sampling versus single plot sampling was conducted to boost sample size for small plots surveys and increase power to detect differences in proportion of smaller plots occupied. All plots contained in a cluster were surveyed as an independent sample. Cluster plot boundaries were located with GPS units that contained coordinates of the corner points of a plot. The cluster plots were surveyed in the same manner as the large plots with playback surveys being conducted. The survey of cluster plots occurred within the same week that occupancy surveys were conducted within the corresponding larger plot. Surveys of cluster plots were completed within a 24 hour sample period. The same information regarding surrounding vegetation and disturbances that were recorded in the larger plot survey effort were recorded for cluster plots with detections and for plots without detections. Helicopter Survey: Helicopter surveys were conducted at 8 large plots in the central strata; small plots were not surveyed. This was done to examine if aerial surveys could be incorporated into the occupancy sampling to access difficult plots and to determine if aerial surveys could be an effective technique to use for future long-term monitoring efforts. 4 �WTPT 2011 Seglund Non-Targets: Observers were asked to collect information on non-target species of interest including brown-capped rosy finches (Leucosticte australis) and pika (Ochotona princeps). Information collected was limited to a UTM coordinate of location detected. For pika, only one UTM per colony or talus area was recorded, points were not recorded for every animal encountered. For brown-capped rosy finches, a UTM coordinate of location of observation and number and sex of animals seen were recorded. In addition, any predator species present while surveying suitable habitat in a plot were recorded. This included both avian and mammalian predators. Feather Collection: Though presence of WTPT feathers was not recorded as a positive detection of a WTPT on a plot, it was asked that observers look for feathers in habitats and collect them if they were located. CPW is working with US Geological Survey (USGS) Science Center in Fort Collins Colorado to: 1) document the distribution of genetic variation across all locations sampled during occupancy surveys, 2) evaluate levels of connectivity among the three different strata sampled in Colorado and 3) examine levels of heterozygosity among individuals. From feather samples collected, DNA will be extracted and each individual will be genotyped using ~20 microsatellite markers. The microsatellite markers are the same markers that have been used to genotype WTPT at two long-term study sites at Mt. Evans and in Rocky Mountain National Park. Thus, the data generated will be directly comparable to data collected elsewhere in the state and throughout the species’ range. Results Large plots: Fifty-nine of the 60 large plots available for sampling were visited, with 55 of the plots having two visits. Of the 59 plots visited, 57 were occupied (Figures 1-4). The occupancy rate for stratum 1 was 0.974 (SE 0.062) while stratum 2 and 3 both had an occupancy of 1.00 (Table 1). Because our occupancy rate was almost 100%, the implication is that our sample unit was too large for the species of interest and may not be adequate for long-term monitoring. Additional work is needed to identify the appropriate plot size for occupancy surveys. The probability of detecting a WTPT was surprisingly high mostly aided by the use of playbacks. However, we did find variations in the group of observers conducting surveys. The calculated probability of detection of a WTPT for hired technicians was 0.747 (SE .046) and was 0.441 (SE 0.072) for biologists and DWMs that conducted surveys. Discrepancies between calculated detection probabilities may have been due to experience conducting surveys or the multi-tasking of efforts required of biologists and DWMs versus technicians. The amount of effort expended by each group and the time of day the survey occurred did not appear to impact detection probabilities. The number of birds detected on large plots from July 13-October 4 was 103 males, 64 females, 135 chicks, and 76 unknown sex. The largest group of birds recorded during surveys was a flock of 30 birds. Due to the project success at detecting hens and chicks, we calculated a reproductive estimate for all plots sampled in August. This was based on the assumption that evaluating average brood size in late August is a good measure of chick survival (Braun and Roger 1971). In the 2011 survey effort in August we counted 29 hens with at least one chick, 9 hens with no 5 �WTPT 2011 Seglund chicks, and a total of 90 chicks for a reproductive success estimate of 2.37. This estimate is similar to what has been recorded at the two long-term study sites in Colorado (G. Wann pers. comm. 2012). Alpine was the dominate habitat type birds were located in based on field observations where data were collected. Ninety-four percent of the sites had either a talus or rock/cliff component within 100 meters of an observation (Table 2). Only 6 sites did not have rock or talus detected near an observation. Of the rock component at sites, the average coverage fell within the category of 25-50%. Human disturbance was recorded at 38 sites with 27 observations near hiking trails, 9 near dirt 4x4 roads, 7 near old mines, and 2 near Off Highway Vehicle only roads. The average slope and elevation for bird detections were derived from 30 m USGS DEM models in ArcGIS. These models found that the average slope angle that birds were located was 19˚ (Min: 3˚, Max: 40˚) and mainly at elevations between 3505-3962 m (Tables 3-4). Small Plots: Two-hundred and fifteen small plots were visited with 16 being occupied (Figures 1-4). Eight plots were occupied in strata 1 out of 71 plots visited, 5 plots were occupied in strata 2 out of 64 plots visited, and 3 plots were occupied in strata 3 out of 80 plots visited. None of the small plots recorded more than a single detection, so the assumption of a closed population may have been violated. This violation could have occurred due to the amount of time that was allowed to elapse between surveys at plots (sometimes the two survey efforts spanned more than a month of time) or because WTPT have a larger home range than the defined plot size. Additional surveys would need to be conducted to evaluate the probability of detecting WTPT at small plots and surveys would need to follow a more strict timing protocol. Overall though, it appears that the estimated occupancy rate for small plots is too low to be effective for long-term monitoring. More work is needed to determine an appropriate plot size for monitoring WTPT. Helicopter Surveys: Eight plots were surveyed with 7 being surveyed twice. Five of the plots surveyed had positive detections of birds with only one plot recording detections on both survey efforts (Figure 3). Of the plots surveyed from the ground and air, three plots had detections on the ground with no detections from the air. In addition to the detections in the large plots surveyed, helicopter surveys detected an additional 10 WTPT flying between plots (Figure 3). Ground surveys proved to have a higher detection probability and a higher occupancy. Though the detection probability and occupancy were lower for helicopter surveys, incorporating this sampling would allow for more habitat to be surveyed increasing the sample size of the survey effort. Predicted Range Model: Using data collected from detections of WTPT in 2011, we worked to refine the current WTPT model developed by Hoffman in 2006 to build a summer Predicted Range Model (PRM) for Colorado. We used SWReGap vegetation values in the PRM versus the GAP data set used in the Hoffman model. Vegetation values used in the PRM included dry tundra (45 WTPT detections fell in this value), Alpine Bedrock and Scree (43 WTPT detections fell in this value), Alpine Fell Field (15 WTPT detections fell in this value), Subalpine-Montane Riparian Shrubland (6 WTPT detections fell in this value), and Alpine-Montane Wet Meadow (2 6 �WTPT 2011 Seglund WTPT detections fell in this value). We also incorporated slope into the PRM of < 40˚ (the Hoffman model did not incorporate slope) and elevations of > 3500 m. The results of the new refined model showed little difference from the Hoffman model originally used for the occupancy surveys (Table 5; Figure 6). The new PRM encompassed 7075.482 km2 or 2.6% of the state compared to the model developed by Hoffman which covered 7583.903 km2 or 2.8% of the state (Table 5). The dominant land manager within the PRM is the US Forest Service (USFS) with > 90% management authority and with > 55% of the model designated as USFS wilderness. Private lands encompass < 1% of the PRM. Non-targets: Pikas were detected at 452 locations during WTPT surveys. Pikas were very abundant within the plots and areas surveyed. Brown-capped rosy finches were detected at 17 locations during surveys (Figure 5). Other species detected that were of interest were yellow-bellied marmots (Marmota flaviventris), moose (Alces alces), and big horn sheep (Orvis canadensis). Predators were detected at 60% of plots surveyed with common ravens (Corvus corax) being the most common predator detected on 36 plots, golden eagles (Aquila chrysaetos) on 7 plots, bald eagle (Haliaeetus leucocephalus) on 1 plot, sharp-shinned hawks (Accipiter striatus) on 2 plots, prairie falcons (Falco mexicanus) on 7 plots, peregrine falcon (Falco peregrinus) on 1 plot, redtailed hawks (Buteo jamaicensis) on 8 plots, northern harriers (Circus cyaneus) on 2 plots, Cooper’s hawk (Accipiter cooperii) on 1 plot, unknown raptors on 5 plots, unknown falcon on one plot, unknown weasels on 2 plots, and coyotes (Canis latrans) on 10 plots. Feather Collection: Observers collected 174 feather samples during surveys (Figure 5). The feathers are currently being analyzed at the Fort Collins Science Center. Summary In response to the petition to list the WTPT under the ESA, CPW conducted statewide occupancy surveys to develop a baseline distribution of the WTPT. These surveys demonstrated that WTPT are widely distributed across the state in suitable habitats and have high detection probabilities. However, these surveys were not designed to provide information on the demographics of individual populations (i.e., increasing or decreasing). Being able to incorporate demographic parameters into surveys would be useful in estimating population numbers, reproductive output, and allow for a better understanding of the effects climate and other impacts may be having on populations. It would also allow CPW to develop better management strategies to maintain viable populations and avoid further ESA listing scenarios. In addition, the 2 different plot sizes used in 2011 to sample occupancy were found to be too large of sampling unit (almost 100% occupancy) or too small (occupancy close to 0) to effectively model and evaluate occupancy rates through time. Further information is needed to determine an appropriate size of plot to sample for statewide occupancy surveys. 7 �WTPT 2011 Seglund Figure 1. Location of large plots used for occupancy surveys in Colorado in 2011. 8 �WTPT 2011 Seglund Figure 2. Result of WTPT occupancy surveys completed in the northern strata of Colorado in 2011. 9 �WTPT 2011 Seglund Figure 3. Result of WTPT occupancy surveys completed in the central strata of Colorado in 2011. 10 �WTPT 2011 Seglund Figure 4. Result of WTPT occupancy surveys completed in the southern strata of Colorado in 2011. 11 �WTPT 2011 Seglund Table 1. Model selection results for estimation of occupancy rates for WTPT at large plots in Colorado. Delta AICc Model Num. AICc AICc Weights Likelihood Par Deviance 186.6044 0 0.99448 1 3 180.1681 197.242 10.6376 0.00487 0.0049 2 193.0277 201.3175 14.7131 0.00063 0.0006 4 192.5768 209.1671 22.5627 0.00001 0 2 204.9528 220.2107 33.6063 0 0 1 218.1405 223.8977 37.2933 0 0 3 217.4614 Model {p(tech) psi(strat)} {p(.) psi(strat)} {p(strat) psi(strat)} {p(tech) psi(.)} {p(.) psi(.)} {p(strat) psi(.)} Table 2. Habitat and disturbance variables present within 100 m of WTPT detections. Habitat w/in 100 m of WTPT Detection Alpine Rock/Cliff Talus Subalpine Krummholz Forbs Cushion Plant Snow Willow Disturbance Sites Percent (%) 91/97 59/97 75/97 4/97 20/97 92/94 58/90 56/97 52/96 38/97 94 61 82 0.04 21 98 64 58 54 40 Table 3. Elevation measured using ArcGIS at sites where WTPT were detected. Meters <3505 3505-3657 3657-3810 3810-3962 3962-4114 >4114 Number of Sites 6 26 49 24 6 1 Table 4. Slope measured using ArcGIS at sites where WTPT were detected. Degrees 0-10 10-20 20-30 30-40 Number of Sites 18 50 32 12 12 �WTPT 2011 Seglund Table 5. Differences in area coverage of the new refined Predicted Range Model developed in 2011 with the Hoffman (2006) non-winter habitat model. WTPT 2012 Model (SWReGAP based) Owner Sq Km % of % of state model NGO 0.014 0.0002% 0.000005% CPW 0.0183 0.0003% 0.000007% Land Trust 0.55 0.0087% 0.000204% Joint City/County County City SLB NPS BLM Private USFS 0.69 0.0110% 0.000256% 2.04 0.0324% 13.43 0.2135% 23.24 0.3695% 309.35 4.9189% 311.64 4.9553% 455.85 7.2484% 5958.66 94.7473% 0.000757% 0.004981% 0.008620% 0.114742% 0.115591% 0.169081% 2.210142% WTPT Hoffman's Non-Winter Model Owner Sq Km 0.019 0 0.126 % of model 0.0003% 0.0000% 0.0020% NGO CPW Land Trust % of state 0.000007% 0.000000% 0.000047% Joint City/County 1.15 0.0183% 0.000427% County City SLB NPS BLM Private USFS 1.98 18.13 32.09 387.94 334.73 426.328 6381.41 0.0315% 0.2883% 0.5103% 6.1685% 5.3225% 6.7789% 101.4694% 0.000734% 0.006725% 0.011903% 0.143892% 0.124156% 0.158130% 2.366946% 13 �WTPT 2011 Seglund Figure 5. Locations where WTPT feathers were collected. 14 �WTPT 2011 Seglund Figure 6. Locations where brown-capped rosy finches were detected during WTPT surveys. 15 �WTPT 2011 Seglund Literature Cited: Braun, C.E. 1971. Habitat requirements of Colorado white-tailed ptarmigan. Proceedings of the Western Association of State Game and Fish Commissioners 51:284-292. Braun, C.E. and G.E. Rodgers. 1971. The white-tailed ptarmigan in Colorado. State of Colorado-Division of Game, Fish and Parks. Tech. Pub. 27. Braun, C.E., R.K. Schmidt, Jr., and G.E. Rogers. 1973. Census of Colorado white-tailed ptarmigan with tape-recorded calls. Journal of Wildlife Management 37:90-93. Braun, C.E., K. Martin, and L.A. Robb. 1993. White-tailed ptarmigan (Lagopus leucurus). In: A. Poole and F. Gill, editors. The birds of North America, Number 68. The Academy of Natural Sciences, Philadelphia, Pennsylvania, and The American Ornithologists’ Union, Washington, D.C., USA. Choate, T.S. 1963. Habitat and population dynamics of white-tailed ptarmigan in Montana. Journal of Wildlife Management 27:684-699. Frederick, G.P. and R.J. Gutierrez. 1992. Habitat use and population characteristics of the whitetailed ptarmigan in the Sierra Nevada, California. Condor 94:889-902. Hoffman, R.W. (2006, April 4). White-tailed Ptarmigan (Lagopus leucura): a technical conservation assessment. [Online]. USDA Forest Service, Rocky Mountain Region. Available: http://www.fs.fed.us/r2/projects/scp/assessments/whitetailedptarmigan.pdf [date of access]. 16 � 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 White-tailed ptarmigan occupancy report Description An account of the resource In 2010, the white-tailed ptarmigan (<em>Lagopus leucura</em>; WTPT) was petitioned to be listed as threatened under the Endangered Species Act. The petitioners cited climate change and its impacts to a species dependent on alpine habitats as the main threat to the viability of WTPT populations. Colorado supports the largest population of WTPT in the lower 48 states. Thus, if the species does become listed, Colorado Parks and Wildlife (CPW) will be responsible for the bulk of the protection and management of the species, yet understanding of the species distribution and demographic parameters is limited.<br /><br />To develop a baseline distribution and inventory of the species, statewide occupancy surveys were conducted in 2011. These surveys will be used to respond to the petition to list the WTPT and to monitor populations as climate change and other disturbances impact the viability of the species over time. Creator An entity primarily responsible for making the resource Seglund, Amy Subject The topic of the resource White-tailed ptarmigan <em>Lagopus leucura</em> Endangered Species Act Occupancy surveys Extent The size or duration of the resource. 16 pages Date Created Date of creation of the resource. 2011 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 Publisher An entity responsible for making the resource available Colorado Parks and Wildlife