Male Rio Grande Turkey Survival and Movements in the Texas Panhandle and Southwestern Kansas

Research Article Male Rio Grande Turkey Survival and Movements in the Texas Panhandle and Southwestern Kansas DERRICK P. HOLDSTOCK, 1 Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, TX 79409, USA MARK C. WALLACE, Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, TX 79409, USA WARREN B. BALLARD, 2 Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, TX 79409, USA JOHN H. BRUNJES, Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, TX 79409, USA RICHARD S. PHILLIPS, Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, TX 79409, USA BRIAN L. SPEARS, 3 Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, TX 79409, USA STEPHEN J. DEMASO, Texas Parks and Wildlife Department, Austin, TX 78744, USA JACK D. JERNIGAN, 4 Texas Parks and Wildlife Department, Matador Wildlife Management Area, Paducah, TX 79248, USA ROGER D. APPLEGATE, Kansas Department of Wildlife and Parks, Emporia, KS 66801-1525, USA PHILLIP S. GIPSON, Kansas Cooperative Fish and Wildlife Research Unit, Kansas State University, Manhattan, KS 66506, USA Abstract Wildlife managers depend on accurate information regarding wild turkey (Meleagris gallopavo) survival patterns to properly manage turkey populations. Survival patterns of male Rio Grande wild turkeys (M. g. intermedia) have not been studied intensively. Wildlife managers in the Texas Panhandle, USA, and southwestern Kansas, USA, suspected that turkey populations were declining. From January 2000 through August 2002, we studied survival and movement of 107 juvenile male and 115 adult male radiomarked Rio Grande wild turkeys on 4 study sites in the Texas Panhandle and southwestern Kansas. We predicted that males would experience lowest survival during spring and that there would be no difference in survival between age classes. We also predicted that greater male movement rates would lead to lower survival rates. Juvenile males had a higher annual survival rate (0.597 [95% CI hereafter: 0.478 0.716]) than adults (0.364 [0.257 0.472]). Juvenile male survival did not differ among seasons, with survival rates of 0.813 (0.736 0.891), 0.904 (0.837 0.972), and 0.917 (0.838 0.996) for spring, summer, and autumn, respectively. Adult male turkey survival was higher during summer (0.915 [0.859 0.972]) than during spring (0.725 [0.651 0.799]), autumn (0.671 [0.536 0.807]), and winter (0.792 [0.732 0.851]). Males had lower survival rates during seasons when long-distance movements were common. The annual survival rate for turkeys that moved to new core-use areas (0.383 [0.282 0.484]) was lower than that for turkeys that did not (0.535 [0.460 0.609]). Also, survival rates increased with time since relocation of core-use areas. Hunting accounted for 18.5% of all mortalities. However, most (80.7%) mortality was attributed to natural causes, mostly mammalian predation. We suspected most predation was the result of coyotes (Canis latrans) and bobcats (Lynx rufus). Managers in the northern portion of the natural range of Rio Grande wild turkeys should be aware of the presence of natural mortality factors that are evident in lightly hunted populations. Managers interested in increasing the survival of male Rio Grande wild turkeys should concentrate on efforts that will provide needed resources in close proximity to roosts. (JOURNAL OF WILDLIFE MANAGEMENT 70(4):904 913; 2006) Key words Kansas, Meleagris gallopavo intermedia, movement, Rio Grande, seasonal survival, Texas Panhandle, wild turkey. Effects of movements have not been correlated to male Rio Grande wild turkey survival. This is because of the absence of data on male Rio Grande wild turkey survival and movements. Knowledge of male turkey survival has been based on eastern wild turkey (M. g. silvestris) studies that have documented lowest male survival during spring (Little et al. 1990, Godwin et al. 1991, Paisley et al. 1996, Vangilder 1996). Movement studies have been limited to discussions of home ranges (McMahon and Johnson 1980, Kelley et al. 1988) or movements by individuals or flocks (Clark 1985, Lambert et al. 1990, Godwin et al. 1994) with the majority of research done on eastern turkeys. Movements that may affect turkey survival can be described at 2 scales. These are localized daily movements, such as those 1 Present address: Texas Parks and Wildlife Department, Gene Howe Wildlife Management Area, Canadian, TX 79014, USA 2 E-mail: wballard@ttacs.ttu.edu 3 Present address: U.S. Fish and Wildlife Service, Upper Columbia Fish and Wildlife Office, Spokane,WA99206,USA 4 Present address: Texas Parks and Wildlife Department, Pat Mayes Wildlife Management Area, Paris, TX 75462, USA associated with feeding and pursuit of within-flock mates during spring; and long-distance movements, resulting in the use of a different roost from the previous night. These long-distance movements can be further divided into those that remain within a core-use area and those that result in the establishment of a new core-use area. Little data exists on the effects daily movements may have on survival. Longer movements, specifically dispersal, have been linked to lower survival in other species (Schwartz and Franzmann 1992, Stenseth and Lidicker 1992, Byrom and Krebs 1999, Schauster et al. 2002). Dispersal refers to a 1-way movement where an animal leaves its natal home range and does not return (Stenseth and Lidicker 1992). Although we knew nothing of movements of turkeys prior to capture, the definition of dispersal is otherwise similar to movement to a new core-use area. Risks associated with dispersals include increased predation both during the movement and after establishment in the new habitat (Shields 1987, Stenseth and Lidicker 1992), increased energy use because of lower efficiency at finding food and shelter (Shields 1987, Stenseth and Lidicker 1992), increased susceptibility to local 904 The Journal of Wildlife Management 70(4)

Table 1. Turkey-hunting seasons in counties containing our turkey study sites, 2000 2002. Data are from Colorado Division of Wildlife (2002), Jefferson (2002), and Kansas Department of Wildlife and Parks (2002). Study site Counties Limit Length Approx. dates Spring Kansas Morton, Kans.; Stevens, Kans. 1 bearded 40 d Apr mid-may Baca, Colo. 1 bearded 6 weeks Mid-Apr May Matador a Cottle, Tex.; Motley, Tex. 4 bearded b 5 weeks Early Apr mid-may Salt Fork Donley, Tex.; Collingsworth, Tex. 4 bearded b 5 weeks Early Apr mid-may Gene Howe a Hemphill, Tex. 4 bearded b 5 weeks Early Apr mid-may Autumn archery Kansas Morton, Kans.; Stevens, Kans. N/A N/A N/A Baca, Colo. N/A N/A N/A Matador a Cottle, Tex.; Motley, Tex. 4 either sex b 4 weeks Late Sep late Oct c Salt Fork Donley, Tex.; Collingsworth, Tex. 4 either sex b 4 weeks Late Sep late Oct c Gene Howe a Hemphill, Tex. 4 either sex b 4 weeks Late Sep late Oct c Autumn gun Kansas Morton, Kans.; Stevens, Kans. N/A N/A N/A Baca, Colo. 1 either sex 5 weeks Early Sep early Oct Matador a Cottle, Tex.; Motley, Tex. 4 either sex b 9 weeks Early Nov early Jan c Salt Fork Donley, Tex.; Collingsworth, Tex. 4 either sex b 9 weeks Early Nov early Jan c Gene Howe a Hemphill, Tex. 4 either sex b 9 weeks Early Nov early Jan c a The Matador and Gene Howe Wildlife Management Areas had seasons different from the county-wide seasons. b The Texas limit was 4 turkeys over all seasons. c The Texas autumn seasons were concurrent with the white-tailed deer (Odocoileus virginianus) autumn seasons. Much autumn turkey harvest was, therefore, opportunistic. diseases (Shields 1987), and social resistance from resident populations (Gaines and McClenaghan 1980). Turkeys that move to new core-use areas may face similar risks. Male turkey movements during spring could be driven by location of hens and social pressure between dominant and subordinate males (Watts 1968, Godwin et al. 1990, Hurst et al. 1991). Watts (1968) reported that dominant males mated early in the breeding period when hen flocks were still in winter concentrations, but groups of less-dominant males followed female groups to find mates later during spring as hen groups split up. If hen groups were unreceptive, males would move to new hen groups (Watts 1968). We studied survival and movement patterns of male Rio Grande wild turkeys at 1 Kansas, USA, and 3 Texas, USA, study sites from January 2000 through August 2002 to determine if survival patterns were similar to those reported for other subspecies and if survival rates were related to movements. We tested the hypothesis that male Rio Grande wild turkey survival rates were affected by movement rates. We predicted that males would experience lowest survival during spring and that, based on other research (Godwin et al. 1991, Lint et al. 1996, Paisley et al. 1996), there would be no difference in survival between age classes. We also predicted that higher movement rates would lead to lower survival rates. Based upon suspected differences in population trends among the study sites (Ballard et al. 2001), we also tested for differences in survival rates among study sites. Study Area We studied male Rio Grande wild turkey survival at 1 Kansas, USA, and 3 Texas, USA, study sites. The Kansas site was centered on the Cimarron National Grasslands in Morton County and adjacent Stevens County, Kansas, USA and Baca County, Colorado, USA. The 3 Texas sites were centered on the Matador Wildlife Management Area (Matador) in Cottle and Motley Counties, the Salt Fork of the Red River (Salt Fork) in Donley and Collingsworth Counties, and the Gene Howe Wildlife Management Area (Gene Howe) in Hemphill County. Spears et al. (2002) provided a general description of the vegetation communities at each of the 4 study sites. We determined boundaries of each study site by turkey movements rather than by political designations. We calculated a 99% fixed-kernel home range (Worton 1989) using all turkey locations to define the functional study areas. They were approximately 25,801 ha, 9,798 ha, 6,656 ha, and 5,237 ha for Kansas, Matador, Salt Fork, and Gene Howe, respectively. Land uses at the 4 study sites included production of cattle, cotton, wheat, and grain sorghum. Oil production also occurred at the Kansas and Gene Howe sites. Grazing occurred at varying intensities on most public and private land throughout the study, both in uplands and in riparian areas. Turkeys, however, always had access to rested pastures or underutilized areas within pastures for cover. Turkey hunting was allowed only by special permit on the Matador Wildlife Management Area. Five permits were issued each spring beginning in 2001 for special bearded-only hunts (D. Dvorak, Texas Parks and Wildlife Department, personal communication). Each year, 4 weekend hunts took place on the Gene Howe Wildlife Management Area with 66 hunters participating; 22 during spring 2000, 23 in 2001, and 21 in 2002 (B. Simpson, Texas Parks and Wildlife Department, personal communication). State harvest seasons and limits were in effect on lands adjacent to the wildlife management areas (Table 1). Methods Capture We captured turkeys with drop-nets (Glazener et al. 1964), rocket-nets (Schemnitz 1994), or walk-in traps (Davis 1994) on sites baited with whole-kernel corn or milo from mid-january through late March 2000, mid-january through mid-march 2001, Holdstock et al. Turkey Survival, Movements 905

and late January through early March 2002. We classified turkeys as juveniles (;0.5 years) or adults (1.5 years) based on standard methods (Pelham and Dickson 1992). We considered a turkey an adult starting 1 January of their second year (;1.5 years). We recorded sex and weight, and we fitted turkeys with a backpackstyle radiotransmitter (AVM Instruments, Livermore, California [Kans.]; Advanced Telemetry Systems, Isanti, Minnesota [Matador, Salt Fork, and Gene Howe]). We also fitted each turkey with a butt-end aluminum leg band (National Band and Tag Company, Newport, Kentucky). Transmitters, with a 4.5-hour (Kans.) or 8-hour (Matador, Salt Fork, and Gene Howe) mortality switch, weighed 120 g, and we attached transmitters using nylon over-braid harnesses. We censored turkeys surviving,14 days postcapture because of potential capture-related mortality (Spraker et al. 1987, Nicholson et al. 2000). Our research was approved by the Texas Tech University Animal Care and Use Committee (Protocol numbers 99917 and 01173B). Survival Analyses We collected turkey survival data from 15 January 2000 through 31 August 2002. We located turkeys 1 3 times per week during late winter, spring, and summer and approximately once every 1 2 weeks throughout the remainder of the year using a dual, 4- element, yagi null-peak antenna system that was truck-mounted (Advanced Telemetry Systems; Samuel and Fuller 1994) or a handheld 3-element yagi antenna (AF Antronics, White Heath, Illinois). Triangulations usually consisted of 3 sequential bearings that we made from fixed stations along roads. We stratified triangulation attempts into 4 time blocks (morning, midday, evening, and roost) so that each bird was relocated at different times of day over the course of each season (Otis and White 1999). Roost period was from dusk until dawn. Other time blocks were of equal length and varied as length of daylight changed. We defined seasons as spring (1 Mar to 31 May), summer (1 Jun to 31 Aug), autumn (1 Sep to 30 Nov), and winter (1 Dec to 28 or 29 Feb) based on seasonal differences in male activities. We did not estimate juvenile survival rates for winter because juveniles were reclassified as adults midway through winter and we did not capture new cohorts of juveniles until late January or February. When mortality signals were detected, we visited the site and attempted to determine the cause of death based on available evidence. We used a Trimble Geoexplorer 2 or Geoexplorer 3 (Trimble Navigation Limited, Sunnyvale, California) Global Positioning System to record the coordinates of all telemetry stations and visual locations. We used a base station within 250 km of each study site for differential correction. We used the maximumlikelihood estimator method (Lenth 1981) in Locate II (Nams 1990) or LOAS (Ecological Software Solutions 1999) to calculate locations from raw bearings. We triangulated test transmitters in known locations to estimate system biases for antenna calibration and bearing standard deviations (range across years and study sites: 7.75 10.598) for calculation of error ellipses around locations (Ecological Software Solutions 1999). Biases were minimal because we calibrated antennas weekly and whenever elements became damaged. We used error ellipses as a tool to identify potential data entry or data-collection errors. However, regardless of the size of the error ellipse, each azimuth was individually investigated. We discarded the location if we suspected that the turkey had likely moved a considerable distance between azimuths or that the estimated location was not within the possible range of the receiver (3.2 km on flat terrain) for 1 azimuths. We used the Kaplan-Meier product-limit estimator modified for staggered entry (Pollock et al. 1989) to estimate survival rates of telemetered turkeys. We set intervals at 0.5 month because these were the smallest units valid for autumn when sampling was less frequent. We compared survival rates using 2-tailed Z-tests and tested for differences in survival patterns using log-rank tests (Pollock et al. 1989). We conducted all tests at a ¼ 0.05. Where multiple pair-wise comparisons were made, we adjusted alpha levels to maintain the experiment-wise error rate of a ¼ 0.05 (Neu et al. 1974). We compared survival rates that included mortality from hunting and crude survival rates (Kurzejeski et al. 1987) that did not include harvest. We compared years and study sites and pooled where appropriate. Movement Analyses We collected turkey movement data from 15 January 2000 through 31 August 2002. We did not estimate movement using distances between successive locations (Fig. 1A), as has been done in the past. Instead we estimated movement 1 of 2 other ways. The first estimate of movement we used was the average distance of day locations from the nearest known roost. We used this as a measure of minimum daily movements. We calculated the minimum distance each turkey must have traveled on a given day by measuring the distance between each day location and the nearest known roost (Fig. 1B). The sampling unit was a particular season at a particular study site. The second estimate of movement we used was the average distance between successive roost locations. We assigned the nearest roost to each day location because it was the roost that was most likely used the previous night. By measuring the distances between successive roosts, we assessed long-distance movements where a turkey moved from an area typically associated with one roost area to an area typically associated with a different roost (Fig. 1C). For the example in Fig. 1C, locations 1 through 7 each have Roost A as their estimated roost, while locations 8 through 10 have Roost B as their estimated roost. All movements in this example would have a value of 0 m except the distance moved between locations 7 and 8, where it is likely a roost change took place. Again, the sampling unit was a particular season at a particular study site. We tested for seasonal and age-class differences in this and the previous movement estimate with Mann Whitney tests, and we used simple linear regression to test both of these movement estimates for correlation with seasonal survival rates. In addition, we examined the effects on survival rate of movements that resulted in a relocation of the turkey s core-use area. We grouped turkeys by whether they remained in one coreuse area during the study or moved to a new core-use area. We defined the relocation of a core-use area as a 1-way movement 1.5 times the length of the longest axis of the turkey s previous core-use area, resulting in either the establishment of a new coreuse area or death. Occasionally, turkeys made such long-distance movements immediately after capture. Considering the period of time it took to familiarize turkeys with a bait site, we assumed that 906 The Journal of Wildlife Management 70(4)

Figure 1. Examples of 3 movement estimates using fictitious data that were representative of locations for male Rio Grande wild turkeys in the Texas Panhandle and southwestern Kansas, USA, from Jan 2000 through Aug 2002. Examples include the method other studies have used to estimate overall turkey movements by calculating the average distance between successive locations (A), minimum daily movements that we estimated by calculating the average distance between day locations and the nearest known roost area within seasons and age classes (B), and long-distance movements that we estimated by calculating the distance between successive estimated roost locations within seasons and age classes (C). newly captured turkeys had a precapture core-use area that included the bait site. If a turkey made a long-distance move immediately after capture, the modified definition for the relocation of a core-use area was a 1-way movement from a capture location 1.5 times the length of the longest axis of the turkey s new core-use area. In the case that a turkey made a longdistance movement immediately after capture but died before establishing a new core-use area, the modified definition for the Holdstock et al. Turkey Survival, Movements 907

Table 2. Fates of radiomarked male Rio Grande wild turkeys in the Texas Panhandle and southwestern Kansas, USA, from Jan 2000 through Aug 2002. Age Study site Censored a Natural b Harvest Road kill Lost c Radio failed d Alive e Juvenile Gene Howe 6 9 0 0 3 1 5 f Kansas 0 4 2 0 0 0 0 f Matador 13 12 2 0 2 3 7 f Salt Fork 2 10 3 0 1 1 5 f Combined 21 35 7 0 6 5 17 f Adult Gene Howe 7 18 6 1 4 3 11 Kansas 4 9 2 0 7 2 0 Matador 5 23 4 0 0 1 16 Salt Fork 2 24 6 0 1 1 12 Combined 18 74 18 1 12 7 39 Total 39 109 25 1 18 12 56 a We censored all turkeys that lived,14 days postcapture because of potential capture-related injuries. b Natural mortality includes all mortality not related to hunting or other human causes. c Lost turkeys were those that could not be located for unknown reasons. d These turkeys were confirmed to have experienced radio failure or shed the transmitter. e Turkeys alive as of 31 Aug 2002. f Juvenile males that lived through 31 Dec of their capture year in 2000 or 2001 (n ¼ 37) were reclassified as adults. relocation of a core-use area was a 1-way movement from a capture location 1.5 times the average size of the core-use areas of others captured at the same time. This method ignores season and relates mortality to the individual turkey and whether or not it moved to a new coreuse area. For this estimate, we conducted a Kaplan-Meier survival analysis comparing survival rates and patterns of turkeys that relocated their core-use area to those that did not. We used simple linear regression to test whether the risk of death was reduced as the time since the relocation of a core-use area increased. We conducted all tests at a ¼ 0.05. Results We fitted 128 juvenile and 132 adult male turkeys with radiotransmitters. One hundred seven juvenile and 115 adult males remained, after censoring 39 turkeys that survived,14 days postcapture. Fifty six (25.2%) male turkeys survived, 135 (60.8%) died, 12 (5.4%) experienced transmitter failure or detachment, and 18 (8.1%) had unknown fates by the conclusion of our study on 31 August 2002 (Table 2). We censored turkeys whose transmitters failed or detached and turkeys with unknown fates from analyses at the beginning of the next encounter period. We re-entered turkeys if we found or recaptured them. Of the 135 that died, 109 (80.7%) died of natural causes (unrelated to hunting or other human causes), 25 (18.5%) were harvested, and 1 (0.7%) died from a vehicle collision (Table 2). Data collection at the Kansas site ended on 30 June 2001, when all males were either dead or their transmitters had ceased operation. Survival Analyses Hunting accounted for 18.5% of all mortalities. However, most (80.7%) mortality was attributed to natural causes. The natural mortality we observed among males was attributed primarily to mammalian predation. The 2 most abundant large mammalian predators were coyotes (Canis latrans) and bobcats (Lynx rufus). Mortalities where an entire carcass containing a discernable bite mark was found (n ¼ 11 of 109 natural mortalities) were more accurately classified than those where only piles of feathers were found. However, because evidence from coyote predation could not be distinguished from evidence from coyote scavenging, predator species was not identified. Coyote sign from scavenging may have led us to underestimate the effects of other potential causes of mortality, such as disease and emaciation, or other mammalian predators such as bobcats. Annual and seasonal survival rates with and without hunting did not differ. To assess other mortality patterns, we censored hunting-related mortalities. Survival rates and patterns did not differ among years by season, age class, or study site. Consequently, years were pooled. With years pooled, survival rates and patterns did not differ among study sites by age class or season. Consequently, study sites were also pooled. The minimum sample size for any group during any time interval was 28. Juveniles had a higher annual survival rate than adults (0.597 [95% CI hereafter: 0.478 0.716] vs. 0.364 [0.257 0.472]; P ¼ 0.004). Survival rates did not differ between juveniles and adults during spring (0.813 [0.736 0.891] vs. 0.725 [0.651 0.799]; P ¼ 0.108) or summer (0.904 [0.837 0.972] vs. 0.915 [0.859 0.972]; P ¼ 0.806). However, juveniles had higher autumn survival rates than adults (0.917 [0.838 0.996] vs. 0.671 [0.536 0.807]; P ¼ 0.002). Adult winter survival rate was 0.792 (0.732 0.851). Survival rates did not differ among seasons for juveniles (Table 3). For adults, summer survival rates were higher than all other seasons (Table 3). Annual survival patterns (hazard functions) between adults and juveniles differed, with the main divergence occurring during autumn as adult male survival declined sharply (Fig. 2). Seasonally, spring and summer survival patterns were not different between adults and juveniles (Fig. 3A,B), while autumn survival patterns were different (Fig. 3C). For juveniles, survival patterns did not differ among seasons (Table 3) but adult survival patterns during summer differed from both spring and autumn (Table 3). Movement Analyses We used 3,306 juvenile and 4,715 adult telemetry locations to estimate movement rates. The distance of day locations from nearest known roost estimate appeared unrelated to survival rate (Table 4). Lowest seasonal movement rates occurred during summer for successive estimated roost locations (Table 4). Seasons 908 The Journal of Wildlife Management 70(4)

Table 3. Pooled survival rates by season and age class for Rio Grande wild turkeys in the Texas Panhandle and southwestern Kansas, USA, from Jan 2000 through Aug 2002. Data on the upper right within an age class represent Z-test statistics and associated P values for comparisons of seasonal survival rates, while data on the lower left within an age class represent chi-square test statistics and associated P values for comparisons of seasonal survival patterns. Season Survival rate 95% CI Winter Spring Summer Autumn Stat P Stat P Stat P Stat P Juvenile Spring 0.813 0.736 0.891 1.773 0.083 a 1.837 0.066 Summer 0.904 0.837 0.972 2.339 0.126 0.242 0.809 Autumn 0.917 0.838 0.996 2.391 0.122 0.078 0.780 Adult Winter 0.792 0.732 0.851 1.366 0.172 2.940 0.003 1.590 0.112 Spring 0.725 0.651 0.799 3.707 0.054 3.991,0.001 0.683 0.495 Summer 0.915 0.859 0.972 3.644 0.056 10.897 0.001 3.254 0.001 Autumn 0.671 0.536 0.807 2.654 0.103 0.776 0.378 12.747,0.001 a Differences were considered significant at the Bonferroni adjusted error rate of a ¼ 0.017 for juveniles and a ¼ 0.008 for adults. with greater movement rates corresponded to seasons when lowest adult male survival rates occurred. A similar correlation was not found for juveniles as juvenile male survival did not differ among seasons (Table 4). We documented that 47 juveniles and 52 adults moved to new core-use areas. Most (52.4%) relocations of core-use area occurred during spring, followed by winter (36.3%), summer (8.9%), and autumn (2.4%). Most winter core-use area relocations occurred during late winter and may have been associated with early breeding. The average distance of core-use area relocation was 9.0 km (SD ¼ 6.5 km), with a range of 1.4 43.8 km. All but 2 were.2.6 km. With hunting mortalities censored, survival rates did not differ between turkeys that moved to new core-use areas and those that stayed within the same core-use area (0.469 [0.354 0.584] vs. 0.588 [0.511 0.665]; P ¼ 0.092). However, when hunting mortalities were included, turkeys that moved to new core-use areas had lower survival than those that did not (0.383 [0.282 0.484] vs. 0.535 [0.460 0.609]; P ¼ 0.018). Of the 63 turkeys that died after relocating their core-use area, 25 died within the first 30 days, 35 had died by day 60, and 41 had died by day 90. There was a linear positive relationship (P ¼ 0.001) between survival rate and number of days since a core-use area relocation (Fig. 4). Discussion Survival Analyses Contrary to our predictions, males did not have higher survival during summer, autumn, and winter than during spring. Additionally, survival rates between age classes did differ. These results were generally inconsistent with other studies. Peterson (1998) called for studies examining the effects of mortality resulting from factors other than hunting. Previous studies of wild turkey survival have been conducted on populations where anywhere from 33 83% of the turkeys that died were harvested (McDougal et al. 1990, Godwin et al. 1991, Ielmini et al. 1992, Paisley et al. 1996, Vangilder 1996, Wakeling and Figure 2. Annual survival patterns of juvenile and adult male Rio Grande wild turkeys in the Texas Panhandle and southwestern Kansas, USA, from Jan 2000 through Aug 2002. The chi-square test statistic (and associated P value) for the comparison of survival patterns is given. Holdstock et al. Turkey Survival, Movements 909

Godwin et al. (1991), Lint et al. (1996), and Paisley et al. (1996) found no difference in annual survival between juveniles and adult male eastern wild turkeys. Ielmini et al. (1992) reported that adult male eastern wild turkeys had lower survival than juveniles. Our results indicated that adults had lower survival rates than juveniles, with the main difference occurring during autumn and to a lesser extent during spring. Figure 3. Seasonal survival patterns of juvenile and adult male Rio Grande wild turkeys in the Texas Panhandle and southwestern Kansas, USA, from Jan 2000 through Aug 2002 for spring (A), summer (B), and autumn (C). The chisquare test statistics (and associated P values) for each comparison of survival patterns are given. Rogers 1998, Wright and Vangilder 2001). In these studies, male turkeys typically experienced their lowest survival during spring, much of which was because of hunting (Little et al. 1990, Godwin et al. 1991, Paisley et al. 1996, Vangilder 1996). However, we found that spring survival rates of male turkeys in our populations where hunting deaths were censored were similar to rates reported in the literature where hunting deaths were included. Previous studies have reported low autumn survival for juveniles prior to arriving at winter roosts (Logan 1973, Little et al. 1990). However, in our study adults rather than juveniles had the lowest survival during autumn. This trend was consistent among study sites and years. Movement Analyses Our prediction that higher movement rates would lead to lower survival rates held true for adults when movements were large. This was particularly true when those movements terminated in a different core-use area. Many studies have reported increased turkey movement during spring (McMahon and Johnson 1980, Clark 1985, Kelley et al. 1988, Godwin et al. 1994, Lambert et al. 1990). While all of these support our findings that male turkeys make large movements during spring, none report large autumn movements and none make any claims about reduced survival in relation to longer movements. Previous studies have often estimated turkey movements by measuring the average distance between successive locations within each season (Fig. 1A). We determined that this did not accurately estimate turkey movement in that, at the very least, turkeys had to return to a roost between locations and that there was a bias because of the different sampling frequencies among different seasons. For these reasons, we did not assess movements using this method, but rather we chose to assess movements using 2 other measurements. The average distance of day locations from the nearest known roost was the first of these 2 measurements. Because we felt confident that all turkey roosts were documented during our study, we felt that this method was appropriate for estimating minimum daily movements without the bias of different seasonal sampling frequencies. We found no correlation between survival rates and this movement estimate. The average distance between successive estimated roost locations was the second measurement. Because such movements were uncommon, we feel that few were missed because of the unequal seasonal sampling frequency. Average distance between successive estimated roost locations was lowest for adults during summer when adult survival rates were highest. Although juvenile movement rates differed among seasons, survival did not. Male movements for a variety of species have been linked to attempts at seeking mating opportunities (Gaines and McClenaghan 1980, Shields 1987, Lidicker and Stenseth 1992, Byrom and Krebs 1999). Spring male turkey movements have also been related to breeding behaviors (Watts 1968, Godwin et al. 1990, Hurst et al. 1991). Scott and Boeker (1975) found that Merriam s (M. g. merriami) males were observed with females in 64% of the visual sightings in April, corresponding to their peak gobbling period. This percentage was smaller in May (38%) and June (3%). For Rio Grande turkeys in Oklahoma, USA, courtship was only reported in areas where hens were often seen feeding, indicating that males followed hens to their feeding areas rather than attempting to breed at random locations (Logan 1973). Kelley et al. (1988) suggested that large spring home ranges may 910 The Journal of Wildlife Management 70(4)

Table 4. Survival rates and the average distances (x ) between day locations and the nearest roost and between successive estimated roost locations by age and season for male Rio Grande wild turkeys in the Texas Panhandle and southwestern Kansas, USA, from Jan 2000 through Aug 2002. Distance from roost Successive roosts Age Season Survival rate Distance SD Distance SD Juvenile Spring 0.813A a 673A b 537 1,404A 2,393 Summer 0.904A 469B 363 785B 1,492 Autumn 0.917A 543C 355 1,200A 1,634 Winter N/A c 508CD 334 997A 1,855 Adult Spring 0.725A 710A 513 1,342A 2,218 Summer 0.915B 624B 461 685BC 1,212 Autumn 0.671A 631ABC 496 1,131AD 1,615 Winter 0.792A 511C 349 1,114CD 1,940 a Survival rates with the same letter within a column and age class were not different at the experiment-wise error rate a ¼ 0.05. b Movement estimate means with the same letter within a column and age class were not different at the experiment-wise error rate a ¼ 0.05. c Juvenile survival rates were not calculable for winter because juveniles were reclassified as adults midway through winter and new cohorts of juveniles were not captured until late Jan or Feb. Figure 4. Linear regression of days since movement to a new core-use area on survival rate for male Rio Grande wild turkeys in the Texas Panhandle and southwestern Kansas, USA, from Jan 2000 through Aug 2002. indicate that males move great distances searching for females with which to mate. No other study of male turkeys has reported long-distance movements during autumn. Autumn and winter movements appear to be driven by food availability, particularly mast such as pine (Pinus spp.) seeds, acorns, and other fruits (Ellis and Lewis 1967, Porter 1992). Mast was somewhat limited at our study sites. The only oaks (Quercus spp.) at our study sites were shinnery oak (Q. havardii) and post oak (Q. stellata), much of which did not occur within daily male turkey ranges. Hackberry (Celtis laevigata, C. occidentalis, C. reticulata) fruits also were available as autumn foods, however these were widely scattered. We speculate that the relatively high autumn movements at our study sites may have been because of lower mast availability compared to other populations. Large-scale movements outside of an animal s core-use area may be because of several different types of movements including dispersal, exploratory trips, and nomadic behavior (Lidicker and Stenseth 1992). While no turkeys in our study exhibited nomadic behavior, many made dispersal-like movements (movement to a new core-use area), and all made occasional exploratory trips into previously unused areas. McMahon and Johnson (1980) documented exploratory trips lasting 2 6 days during spring, but none during summer. Core-use area relocations were easy to document with radiotelemetry because they were relatively long-term. However an unknown number of exploratory trips were likely missed because these trips occurred over a span of only a few days. Studies of eastern and Rio Grande turkeys have reported average dispersal from 2.0 3.2 km (Ellis and Lewis 1967, Logan 1973, Speake et al. 1975, Holbrook et al. 1987). Hoffman (1991) measured mean spring movements of 8.7 km and 5.3 km, respectively, for juvenile and adult male Merriam s turkeys in Colorado, USA, while longest movements for eastern turkeys in Missouri, USA, were 11.5 km (Kurzejeski and Lewis 1990). Average distance traversed during a movement to a new core-use area in our study was much greater. We suspect the longer core-use area relocations that we documented resulted in expenditure of energy during the movement as well as the increased energy demands of unfamiliar habitat described by Shields (1987) and Stenseth and Lidicker (1992). Further evidence of this was that turkeys that moved to new coreuse areas during our study had lower survival than those that remained in the same core-use area. This pattern occurred only when hunting mortalities were included. However, it was reasonable to assume that unfamiliarity with locations of hens with which to breed in the new habitat may have made male turkeys more apt to be called in by hunters. Survival of turkeys that relocated their core-use area increased as time elapsed since the movement. Although long-distance movements ultimately benefit turkey populations for reasons including genetic exchange among roosts and reducing the potential for inbreeding, they present individual risks. However, Miller et al. (1995) found that dispersal distance between winter roost location and nest site in Rio Grande wild turkey females was not related to hen survival in Kansas. Movements alone did not explain the difference in annual survival between adults and juveniles. Throughout the year, juveniles at our study sites were often found with groups of other juvenile males and hens. Adults were often found alone, or in small groups of 3 5 adult males, except during winter. Therefore, juvenile males may have benefited from the heightened ability of larger flocks to detect potential predators. As summer proceeded, Holdstock et al. Turkey Survival, Movements 911

grasses cured and were grazed or trampled by cattle, decreasing the amount of horizontal cover for turkeys. Differences in size and iridescence between adult and juvenile males may have made adults more detectable and more vulnerable to predation during autumn. Furthermore, the smaller body mass of juveniles may have required them to spend less time foraging in the open. Management Implications Managers should be careful not to overlook potential natural mortality factors that can occur at high rates during both spring and autumn hunting seasons, as these can be easily masked in heavily hunted populations. Managers interested in increasing survival of male turkeys should concentrate on efforts that will provide needed resources in close proximity to roosts in order to decrease the necessity to move to new core-use areas for reasons other than genetic exchange. Acknowledgments J. Bowman, E. Sobek, M. Gray, and D. Wester provided statistical assistance and logistical support. B. Bedford, M. Butler, G. Hall, R. Houchin, and R. Huffman provided comments on the Literature Cited Ballard, W. B., M. C. Wallace, J. H. Brunjes, T. Barnett, D. Holdstock, R. Phillips, B. Spears, M. Miller, B. Simpson, S. Sudkamp, R. D. Applegate, and P. Gipson. 2001. Changes in land use patterns and their effects on Rio Grande turkeys in the Rolling Plains of Texas (annual report 2001). Texas Tech University, Lubbock, USA. Byrom, A. E., and C. J. Krebs. 1999. Natal dispersal of juvenile arctic ground squirrels in the boreal forest. Canadian Journal of Zoology 77:1048 1059. Clark, L. G. 1985. Adjustment by transplanted wild turkeys to an Ohio farmland area. Proceedings of the National Wild Turkey Symposium 5:33 47. Colorado Division of Wildlife. 2002. 2002 turkey: Colorado hunting spring and fall season information. Colorado Department of Natural Resources, Denver, USA. Davis, B. D. 1994. A funnel trap for Rio Grande turkey. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 48:109 116. Ecological Software Solutions. 1999. LOAS location of a signal. Ecological Software Solutions, Sacramento, California, USA. Ellis, J. E., and J. B. Lewis. 1967. Mobility and annual range of wild turkeys in Missouri. Journal of Wildlife Management 31:568 581. Gaines, M. S., and L. R. Mcclenaghan, Jr. 1980. Dispersal in small mammals. Annual Review of Ecology and Systematics 11:163 196. Glazener, W. C., A. S. Jackson, and M. L. Cox. 1964. The Texas drop-net turkey trap. Journal of Wildlife Management 28:280 287. Godwin, K. D., G. A. Hurst, and R. L. Kelley. 1991. Survival rates of radioequipped wild turkey gobblers in east-central Mississippi. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 45:218 226. Godwin, K. D., G. A. Hurst, and B. D. Leopold. 1994. Movements of wild turkey gobblers in central Mississippi. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 48:117 122. Godwin, K. D., W. E. Palmer, G. A. Hurst, and R. L. Kelley. 1990. Relationship of wild turkey gobbler movements and harvest rates to management area boundaries. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 44:260 267. Hoffman, R. W. 1991. Spring movements, roosting activities, and home-range characteristics of male Merriam s wild turkeys. Southwestern Naturalist 36: 332 337. Holbrook, H. T., M. R. Vaughan, and P. T. Bromley. 1987. Wild turkey habitat preferences and recruitment in intensively managed Piedmont forests. Journal of Wildlife Management 51:182 187. Hurst, G. A., D. R. Smith, J. D. Burk, and B. D. Leopold. 1991. Wild turkey gobbler habitat use and home range in loblolly pine plantations. Proceedings manuscript. We thank T. Barnett, B. Bedford, A. Braden, B. Buckley, J. Bullock, D. Butler, J. Davis, S. Dempsey, A. Denton, K. Derzapf, J. Doty, D. Drummond, J. Duke, C. Evans, D. Earl, D. Ferris, C. Frosch, R. Hanson, W. Hough, D. King, J. McJunkin, J. Milliken, B. Mills, J. Moon, G. Moreno, S. Pedersen, L. Robison, N. Sears, V. Spearman, K. Spears, A. Teaschner, A. Thomas, R. Ward, and N. Wilson for assistance with data collection. D. Cook, R. Cranford, D. Dvorak, T. Hinkle, G. Ibarra, D. Lucia, G. Miller, M. Miller, B. Rogers, B. Simpson, S. Sudkamp, and D. Wright, all of Texas Parks and Wildlife Department, and C. Swank and M. Mitchener of Kansas Department of Wildlife and Parks, assisted with turkey captures and other aspects of this study. We also thank the many private landowners who allowed us access to their properties. Texas Parks and Wildlife Department, Kansas Department of Wildlife and Parks (Federal Aid Grant W-R-54), the National Wild Turkey Federation, and the Texas State Chapter of the National Wild Turkey Federation funded this study. This is College of Agricultural Sciences and Natural Resources, Texas Tech University publication number T-9 994. of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 45:115 123. Ielmini, M. R., A. S. Johnson, and P. E. Hale. 1992. Habitat and mortality relationships of wild turkey gobblers in the Georgia Piedmont. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 46:128 137. Jefferson, J. 2002. Texas Parks and Wildlife outdoor annual: hunting and fishing regulations. Texas Parks and Wildlife Department, Austin, USA. Kansas Department of Wildlife and Parks. 2002. 2002 Kansas hunting and furharvesting regulations. Kansas Department of Wildlife and Parks, Topeka, USA. Kelley, R. L., G. A. Hurst, and D. E. Steffen. 1988. Home ranges of wild turkey gobblers in central Mississippi. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 42:470 475. Kurzejeski, E. W., and J. B. Lewis. 1990. Home ranges, movements, and habitat use of wild turkey hens in northern Missouri. Proceedings of the National Wild Turkey Symposium 6:67 71. Kurzejeski, E. W., L. D. Vangilder, and J. B. Lewis. 1987. Survival of wild turkey hens in north Missouri. Journal of Wildlife Management 51:188 193. Lambert, E. P., W. P. Smith, and R. D. Teitelbaum. 1990. Wild turkey use of dairy farm timberland habitats in southeastern Louisiana. Proceedings of the National Wild Turkey Symposium 6:51 60. Lenth, R. V. 1981. On finding the source of a signal. Technometrics 23:149 154. Lidicker, W. Z., and N. Z. Stenseth. 1992. To disperse or not to disperse: who does it and why? Pages 21 36 in N. C. Stenseth and W. Z. Lidicker, Jr., editors. Animal dispersal. Chapman and Hall, London, United Kingdom. Lint, J. R., B. D. Leopold, G. A. Hurst, and K. J. Gribben. 1996. Population size and survival rates of wild turkey gobblers in central Mississippi. Proceedings of the National Wild Turkey Symposium 7:33 38. Little, T. W., J. M. Kienzler, and G. A. Hanson. 1990. Effects of fall either-sex hunting on survival in an Iowa wild turkey population. Proceedings of the National Wild Turkey Symposium 6:119 125. Logan, T. H. 1973. Seasonal behavior of Rio Grande wild turkeys in western Oklahoma. Proceedings of the Annual Conference of the Southeastern Game and Fish Commissioners 27:74 91. McDougal, L. A., M. R. Vaughan, and P. T. Bromley. 1990. Wild turkey and road relationships on a Virginia national forest. Proceedings of the National Wild Turkey Symposium 6:96 106. McMahon, G. L., and R. N. Johnson. 1980. Introduction of the wild turkey into the Carlos Avery Wildlife Management Area. Proceedings of the National Wild Turkey Symposium 4:32 44. Miller, M. S., D. J. Buford, and R. S. Lutz. 1995. Survival of female Rio Grande 912 The Journal of Wildlife Management 70(4)

turkeys during the reproductive season. Journal of Wildlife Management 59: 766 771. Nams, V. O. 1990. Locate II user s guide. Pacer Computer Software, Truro, Nova Scotia, Canada. Neu, C. W., C. R. Byers, and J. M. Peek. 1974. A technique for analysis of utilization availability data. Journal of Wildlife Management 38:541 545. Nicholson, D. S., R. L. Lochmiller, M. D. Stewart, R. E. Masters, and D. M. Leslie, Jr. 2000. Risk factors associated with capture-related death in eastern wild turkey hens. Journal of Wildlife Diseases 36:308 315. Otis, D. L., and G. C. White. 1999. Autocorrelation of location estimates and the analysis of radiotracking data. Journal of Wildlife Management 63:1039 1044. Paisley, R. N., R. G. Wright, and J. F. Kubisiak. 1996. Survival of wild turkey gobblers in southwestern Wisconsin. Proceedings of the National Wild Turkey Symposium 7:39 44. Pelham, P. H., and J. G. Dickson. 1992. Physical characteristics. Pages 32 45 in J. G. Dickson, editor. The wild turkey: biology and management. Stackpole, Mechanicsburg, Pennsylvania, USA. Peterson, M. J. 1998. Book review: proceedings of the seventh National Wild Turkey Symposium. Journal of Wildlife Management 62:816 818. Pollock, K. H., S. R. Winterstein, C. M. Bunck, and P. D. Curtis. 1989. Survival analysis in telemetry studies: the staggered entry design. Journal of Wildlife Management 53:7 15. Porter, W. F. 1992. Habitat requirements. Pages 202 213 in J. G. Dickson, editor. The wild turkey: biology and management. Stackpole, Mechanicsburg, Pennsylvania, USA. Samuel, M. D., and M. R. Fuller. 1994. Wildlife radiotelemetry. Pages 370 418 in T. A. Bookhout, editor. Research and management techniques for wildlife and habitats. Fifth edition. The Wildlife Society, Bethesda, Maryland, USA. Schauster, E. R., E. M. Gese, and A. M. Kitchen. 2002. Population ecology of swift foxes (Vulpes velox) in southeastern Colorado. Canadian Journal of Zoology 80:307 319. Schemnitz, S. D. 1994. Capturing and handling wild animals. Pages 106 124 in T. A. Bookhout, editor. Research and management techniques for wildlife and habitats. Fifth edition. The Wildlife Society, Bethesda, Maryland, USA. Schwartz, C. C., and A. W. Franzmann. 1992. Dispersal and survival of subadult black bears from the Kenai Peninsula, Alaska. Journal of Wildlife Management 56:426 431. Scott, V. E., and E. L. Boeker. 1975. Ecology of Merriam s wild turkey on the Fort Apache Indian Reservation. Proceedings of the National Wild Turkey Symposium 3:141 158. Shields, W. M. 1987. Dispersal and mating systems: investigating their causal connections. Pages 3 24 in B. D. Chepko-Sade and Z. T. Halpin, editors. Mammalian dispersal patterns. University of Chicago, Illinois, USA. Speake, D. W., T. E. Lynch, J. E. Fleming, G. A. Wright, and W. J. Hamrick. 1975. Habitat use and seasonal movements of wild turkeys in the southeast. Proceedings of the National Wild Turkey Symposium 3:122 130. Spears, B. L., W. B. Ballard, M. C. Wallace, R. S. Phillips, D. P. Holdstock, J. H. Brunjes, R. Applegate, P. S. Gipson, M. S. Miller, and T. Barnett. 2002. Retention times of miniature radiotransmitters glued to wild turkey poults. Wildlife Society Bulletin 30:861 867. Spraker, T. R., W. J. Adrian, and W. R. Lance. 1987. Capture myopathy in wild turkeys (Meleagris gallopavo) following trapping, handling and transportation in Colorado. Journal of Wildlife Diseases 23:447 453. Stenseth, N. C., and W. Z. Lidicker, Jr. 1992. The study of dispersal: a conceptual guide. Pages 5 20 in N. C. Stenseth and W. Z. Lidicker, Jr., editors. Animal dispersal. Chapman and Hall, London, United Kingdom. Vangilder, L. D. 1996. Survival and cause-specific mortality of wild turkeys in the Missouri Ozarks. Proceedings of the National Wild Turkey Symposium 7: 21 31. Wakeling, B. F., and T. D. Rogers. 1998. Summer resource selection and yearlong survival of male Merriam s turkeys in north-central Arizona, with associated implications from demographic modeling. Arizona Game and Fish Department Technical Report Number 28, Phoenix, USA. Watts, R. C. 1968. Rio Grande turkeys in the mating season. Proceedings of the North American Wildlife and Natural Resources Conference 33:205 210. Worton, B. J. 1989. Kernel methods for estimating the utilization distribution in home-range studies. Ecology 70:164 168. Wright, G. A., and L. D. Vangilder. 2001. Survival of eastern wild turkey males in western Kentucky. Proceedings of the National Wild Turkey Symposium 8: 187 194. Associate Editor: Brennan. Holdstock et al. Turkey Survival, Movements 913