See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/240793547 Effects of Artificial Escape Dens on Swift Fox Populations in Northwest Texas Article in Wildlife Society Bulletin · October 2006 DOI: 10.2193/0091-7648(2006)34[821:EOAEDO]2.0.CO;2 CITATIONS READS 9 61 6 authors, including: Kerry Nicholson Brian Cypher 33 PUBLICATIONS 92 CITATIONS 67 PUBLICATIONS 716 CITATIONS Alaska Department of Fish and Game SEE PROFILE California State University, Stanislaus SEE PROFILE Jan Kamler Panthera 109 PUBLICATIONS 1,103 CITATIONS SEE PROFILE All content following this page was uploaded by Jan Kamler on 27 December 2016. The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately. Peer Reviewed Effects of Artificial Escape Dens on Swift Fox Populations in Northwest Texas BRADY K. MCGEE, Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, TX 79409, USA WARREN B. BALLARD,1 Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, TX 79409, USA KERRY L. NICHOLSON, Department of Range, Wildlife, and Fisheries Management, Texas Tech University, Lubbock, TX 79409, USA BRIAN L. CYPHER, Endangered Species Recovery Program, Bakersfield, CA 93389, USA PATRICK R. LEMONS II, Department of Environmental and Resource Sciences, University of Nevada, Reno, NV 89512, USA JAN F. KAMLER, Wildlife Conservation Research Unit, Oxford OX1 3PS, United Kingdom Abstract Throughout the range of swift fox (Vulpes velox), coyotes (Canis latrans) are the primary source of swift fox mortality. Coyotes may suppress swift fox populations where densities are high. Because coyote numbers have increased since wolves (Canis lupus) have been exterminated, we hypothesized that escape habitats may limit swift foxes. To test our hypothesis, we installed artificial escape dens in 3 spatially separated (treated) areas on the Rita Blanca National Grasslands in Dallam County, Texas, USA. From January 2002 to August 2004, we captured, ^ ¼ 0.81) was higher than in radiocollared, and monitored 55 swift foxes. Annual swift fox survival in artificial escape-den–treated areas ( S ^ ¼ 0.52, P ¼ 0.07). Relative swift fox abundance was higher in treated than untreated areas in 2002 ( Yates’ v2 ¼ 4.61, P ¼ 0.03) untreated areas ( S and in 2003 (Yates’ v2 ¼ 4.70, P ¼ 0.03) but not in 2004 (Yates’ v2 ¼ 2.67, P ¼ 0.10). However, recruitment rates were no different between treated and untreated areas in 2002 (Yates’ v2 ¼ 0.21, P ¼ 0.65) or 2003 (Yates’ v2 ¼ 0.41, P ¼ 0.52). Ninety-five percent fixed-kernel estimates of home-range sizes revealed no difference (P ¼ 0.91) between treated and untreated areas, but swift foxes increased their distribution by moving into an area that had been unoccupied for at least 3 years before this study. Our results suggest that artificial escape dens contributed to increasing swift fox distributions in our study area. (WILDLIFE SOCIETY BULLETIN 34(3):821–827; 2006) Key words abundance, artificial escape dens, Canis latrans, coyotes, dens, mortality, recruitment, survival, swift fox, Vulpes velox. Swift fox (Vulpes velox) once were abundant throughout the shortand mid-grass prairies of North America but declined with expansion of human settlement (Egoscue 1979). Populations were reduced by habitat destruction and the indiscriminate use of traps and poison baits to control large carnivores, principally wolves (Canis lupus) and coyotes (Canis latrans; Bekoff 1977, Hines 1980, Scott-Brown et al. 1987). Swift fox populations may have begun to increase by the mid-twentieth century due to the elimination of poisoning campaigns, but they still remain below historic levels (Egoscue 1979, Samuel and Nelson 1982). The swift fox was a candidate for endangered species listing by the United States Fish and Wildlife Service (USFWS) from 1992 to 2001 (USFWS 1995, 2001, Allardyce and Sovada 2003). Throughout the range of swift fox, coyotes are the primary source of swift fox mortality when the cause of death is identifiable (Kitchen et al. 1999, Matlack et al. 2000, Schauster et al. 2002). Kamler et al. (2003a) reported a 46–63% annual survival rate for swift fox at one study site in northwest Texas, USA, where coyotes were abundant. Consequently, that population was considered a sink (Pulliam 1988) for swift fox due to heavy predation from coyotes (Kamler et al. 2003a). Other studies have reported similar annual survival rates, ranging from 43–53%, where coyotes were prevalent (Sovada et al. 1998, Allardyce and Sovada 2003, Anderson et al. 2003). Coyote-related mortality appears to be the result of interference competition rather than predation (Kamler et al. 2003b). Although it occasionally occurs, coyotes rarely consume the swift foxes they kill (Allardyce and Sovada 2003). During a study 1 E-mail: Warren.Ballard@ttu.edu McGee et al.  Effects of Artificial Escape Dens on Swift Fox conducted by Sovada et al. (1998) in western Kansas, USA, only 1 of 13 swift foxes killed by coyotes appeared to be eaten. However, it was not uncommon for swift foxes to be cached or buried (Sovada et al. 1998, Kitchen et al. 1999). Coyote control can be an effective way to relieve depredation and increase swift fox densities (Kamler et al. 2003a). However, implementing effective control programs can be extremely difficult for a number of reasons. In general, predator control programs can be costly because they usually must be administered over large areas for multiple years to achieve effective control (Connelly and Longhurst 1975), and in some cases, coyote control has not been effective in improving fox survival (Cypher and Scrivner 1992). Also, such programs are not always popular with some members of the public, even when conducted for the conservation of a rare, native species (Goodrich and Buskirk 1995). During the Kamler et al. (2003a) study, there was also risk to human safety because the United States Department of Agriculture’s Animal and Plant Health Inspection Service (APHIS) personnel shot coyotes from a fixed-wing aircraft. Other less-costly, more-effective and acceptable, and less-hazardous methods are needed to increase swift fox populations. We installed artificial escape dens as a new method for reducing coyote-related mortalities of swift foxes. Swift fox are one of the most burrow-dependent canids in North America (Egoscue 1979). Dens may constitute crucial escape cover. White et al. (1994) suggested that kit foxes (Vulpes macrotis) were able to survive in coyote home ranges by establishing a large number of dens (
20) to facilitate escape. Also, Kitchen et al. (1999) indicated that coyotes were more likely to kill swift foxes when foxes were a substantial distance from a den. Increasing den density may reduce 821 vulnerability to attack by predators, thereby, increasing survival rates of swift fox. Because coyotes have increased in number, we hypothesized that lack of escape-den sites may be limiting swift fox populations in northwest Texas, USA. Our study is the first assessment of using artificial dens to increase swift fox populations. Our objectives were to determine home ranges, survival rates, relative abundance, and recruitment of swift foxes in treated (artificial escape-dens installed) and untreated areas (no dens installed). We predicted that artificial escape dens would result in increased home-range size and increased survival, abundance, and recruitment of swift foxes. We also installed artificial escape dens in an area previously unoccupied by swift foxes to determine if they would immigrate into an area saturated with dens. By providing greater opportunity for escape, we predicted that swift fox would suffer fewer coyoterelated mortalities, and swift fox populations would increase. Study Area We monitored swift foxes in a contiguous 100-km2 area on the Rita Blanca National Grassland (NG) in Dallam County, approximately 43 km northwest of Dalhart, Texas, USA (Fig. 1). The NG consisted of native rangelands with short-grass prairie dominated by blue grama (Bouteloua gracilis), side-oats grama (Bouteloua curtipendula), burrograss (Scleropogon brevifolius), and buffalograss (Buchloe dactyloides) that were moderately to intensively grazed by cattle (Bos taurus). Neither coyotes nor swift foxes were heavily hunted on the NG, although the area was open to hunting by the general public (McGee 2005). Methods From January 2002 to August 2004, we captured, radiocollared, and monitored 55 swift foxes on the NG. We captured swift foxes using Havahartt cage traps (Woodstream Corp., Lititz, Pennsylvania.; 25.4 3 30.5 3 81.3 cm) baited with carcasses of prey species, including black-tailed prairie dogs (Cynomys ludovicianus), black-tailed jackrabbits (Lepus californicus), and desert cottontails (Sylvilagus audubonii). We operated traps for 2 or 3 consecutive nights. We placed traps individually or in pairs every 0.4–0.8 km along fences, washes, or drainages, which may have been travel routes for swift foxes. We checked traps once daily at dawn. Trapping efforts occurred throughout the NG study site but also opportunistically near active dens or where unmarked foxes were sighted. We suspended trapping during pup-rearing season (Apr– Jun) to avoid late-pregnancy and early pup-rearing periods. Research protocols were approved by the Animal Care and Use Committee (Protocol No. 01105-04) at Texas Tech University. This is Texas Tech University, College of Agricultural Sciences and Natural Resources Technical Publication T-9-1049. We collected the following data from each study animal: sex, estimated age class, and capture location. We ear-tagged each animal with one unique identification number. We classified swift foxes as adults (.6 months), juveniles (3–6 months), or pups (,3 months) based on size, weight, and tooth-wear at the time of capture. We placed 40-g radiotransmitter collars (Advanced Telemetry Systems, Inc., Isanti, Minnesota) on each fox. We released all foxes at their capture sites. We recorded independent telemetry locations (Erickson et al. 822 Figure 1. Map of the 100-km2 study area located on the Rita Blanca National Grassland in northwest Dallam County, Texas, USA. Artificial escape dens (black dots; 108 total) were installed in 3 separate grid locations. 2001) for each swift fox 2–4 times per week from 1900 to 0100 hours throughout the study period to obtain locations of study animals when they were most active (Kitchen et al. 1999, Allardyce and Sovada 2003). We considered locations independent when recorded .3 hours apart. We performed all radiotracking remotely using a vehicle-mounted, null-peak system with dual, 4-element Yagi antennas (Advanced Telemetry Systems). We calculated location estimates using the maximum-likelihood option in the LOASe 2.6 (Ecological Software Solutions, Sacramento, California) radiotelemetry software. Based on readings of test collars placed in 30 different locations (White and Garrott 1990), we determined that the mean error was 47.4 m. During April 2002, we constructed 72 artificial escape dens in 2 spatially separated areas (10 km apart) occupied by swift foxes. We considered swift foxes as belonging to the treated area if their home range overlapped an artificial escape-den area by
50%. We considered foxes in untreated groups as those that were not captured nor radiotracked within an artificial escape-den–treated area during our designated biological year (Sep–Aug). We Wildlife Society Bulletin  34(3) Figure 2. Annual home ranges of swift foxes (dark gray polygons, n ¼ 19) from 1999–2001 before installation of artificial escape dens on the Rita Blanca National Grassland (solid line) in northwest Texas, USA. Thirty-six artificial escape dens were installed in a grid (dotted rectangle) in Apr 2002 to determine if foxes would immigrate into an area saturated with dens. assumed that coyote abundance was equally represented between treated and untreated areas. Also, we considered foxes to belong to the same family group if they used the same area and dens concurrently (Kitchen et al. 1999, Kamler et al. 2003a). Escape dens consisted of 4.04-m-long, 20.32-cm-diameter, corrugated-plastic sewer pipes with 20.32-cm holes cut in the middle to allow foxes to modify and expand subterranean dens (US$6.41/m; Amarillo Plumbing Supply, Inc., Amarillo, Texas). The diameter size of our artificial escape dens were based on previous studies that reported a mean den opening height of 20.0 cm for swift fox dens (Cutter 1958, Hillman and Sharps 1978, Pruss 1999, Jackson and Choate 2000). Coyote dens were reported to be 30–37 cm in diameter (Bekoff 1977, 1982, Althoff 1980, Harrison and Gilbert 1985). We assumed that artificial escapeden entrances, being the same diameter as natural swift fox dens, were too narrow for coyotes. We used a John Deeree 260 skid loader (Deere and Company, Moline, Illinois) to install and cover the sewer pipe with only the 2 open ends exposed. Escape dens were randomly oriented and spaced approximately 322 m apart in a 2.59-km2 grid pattern for a density of 36/2.59 km2. To determine whether foxes would immigrate into an area with an abundance of escape dens, we placed an additional 36 artificial escape dens in an area unoccupied by foxes for the 3 years before the study (Fig. 2; Kamler et al. 2003b). This area consisted of overgrazed native rangeland, which has been reported as primary swift fox habitat (Allardyce and Sovada 2003), but coyotes may have previously prevented swift foxes from establishing territories within the area (Kamler et al. 2003b). When this grid site was installed, it was .5 km from the nearest known existing swift fox home range. To determine whether foxes were in the vicinity of the proposed grid site, we placed traps in and near the area over a 3-month period before den installation. We captured no swift fox McGee et al.  Effects of Artificial Escape Dens on Swift Fox in the 240 trap-nights before artificial den installation. Foxes that eventually established home ranges overlapping this grid area by
50% were also considered treated foxes. We estimated annual home-range sizes and core areas for swift foxes using 95% and 50% fixed kernel (FK) methods with leastsquares cross-validation as the smoothing parameter (Seaman et al. 1999) as calculated by Home Range extension (Rodgers and Carr 1998) for ArcViewe 3.2 (Environmental Systems Research Institute, Redlands, California). To allow for comparisons with previously published studies of swift foxes, we calculated 95% and 50% minimum convex polygon estimates (Mohr 1947) of home ranges and core areas. We calculated home ranges for foxes with .30 locations and .9 months of radiotracking (Seaman et al. 1999). We calculated differences between mean home-range sizes using 2-way ANOVAs in SPSSe 12.0 (Chicago, Illinois; SPSS 2003) and deemed them significant when P , 0.05. We evaluated annual survival rates by monitoring radiocollared animals. To facilitate the detection of dead foxes, radiocollars produced a ‘‘mortality signal’’ (pulse rate approximately doubled) if an animal remained motionless for
2 hours. If we detected mortality signals, we recovered dead foxes as soon as possible and performed necropsies to determine cause of death. We assessed causes of mortality with methods similar to those described by Kamler et al. (2003a). We classified mortalities as coyote, raptor, natural causes, or unknown. We excluded a single trap-related mortality that occurred during the project from analyses. We calculated annual survival rates beginning in August 2002, approximately 4 months after installation of artificial escape dens. This was to allow swift foxes an adjustment period to the artificial dens and to calculate 2 full years of annual survival. We determined annual survival rates for swift foxes using the program MICROMORTe (Heisey and Fuller 1985). We assumed that constant survival occurred during all seasons. We calculated radiodays to the midpoint between last-known live signal and the initial mortality signal. We compared swift fox mortality rates using Z-tests (Heisey and Fuller 1985). Given small sample sizes for comparisons, differences in survival rates were deemed significant when P , 0.10. Preliminary analyses indicated no statistical differences between years, so we pooled data to increase sample sizes. We calculated abundance estimates of swift foxes using a catchper-unit-effort index (Schauster et al. 2002). We calculated the total number of captures/100 trap-nights as an index of relative abundance. We compared relative swift fox abundance between treated and untreated areas using Yates’ corrected chi-square tests (Zar 1999, Kamler et al. 2003a). We determined swift fox recruitment rates from the minimum number of juveniles per reproducing adult that survived until dispersal or 1 year of age (Kamler et al. 2003a). We did not calculate recruitment rates for year 3 because the study ended in August and juvenile dispersal generally did not occur until October (Sovada et al. 2003, Nicholson 2004). We compared recruitment rates between treated and untreated areas using Yates’ corrected chi-square tests, and P , 0.05 was deemed significant (Zar 1999). For statistical analysis, we considered individual foxes as the sample unit even though foxes within areas were not independent. 823 Table 1. Average 50% and 95% fixed kernel (FK) and minimum convex polygon (MCP) estimates of annual home-range sizes (km2) and standard errors for males, females, and all foxes combined on Rita Blanca National Grassland (NG) in northwest Texas, USA, 2002–2004. 50% FK Study area Treated All foxes Males Females Untreated All foxes Males Females n _ x 13 6 7 5 2 3 95% FK SE _ x 4.5 5.4 3.8 0.6 0.8 0.9 3.5 6.6 1.5 1.8 4.1 0.5 SE 19.9 23.5 16.8 2.5 3.3 3.5 17.1 31.5 7.5 8.6 19.3 2.3 Randomly assigning foxes to treatment areas was logistically impossible because swift foxes tended to live in male–female pairs and were often with their pups during the summer. Thus, the test statistics based on an area were not independent, so there is potential for error. Results From January 2002 to August 2004, we captured 55 swift foxes (31 M, 24 F). Based on foxes with .30 locations and .9 months of radiotracking, we calculated annual home ranges for 18 adult swift foxes (8 M, 10 F; Table 1). Ninety-five percent FK estimation of home-range sizes (mean 6 SE) revealed no difference (F ¼ 0.01, P ¼ 0.91, 1  b ¼ 0.05) between treated (19.9 6 2.5 km2, n ¼ 13) and untreated swift fox groups (17.1 6 8.6 km2, n ¼ 5), but there was a difference (F ¼ 7.07, P ¼ 0.02, 1  b ¼ 0.70) between males (27.5 6 5.5 km2, n ¼ 8) and females (12.2 6 2.9 km2, n ¼ 10). There was no interaction between area and sex (F ¼ 2.3, P ¼ 0.16, 1  b ¼ 0.29). We found that, for all foxes combined, core areas (mean 6 SE) were not different (F ¼ 0.18, P ¼ 0.68, 1  b ¼ 0.07) in treated (4.5 6 0.6 km2, n ¼ 13) and untreated areas (3.5 6 1.8 km2, n ¼ 5). However, there were sex differences (F ¼ 6.27, P ¼ 0.25, 1  b ¼ 0.64), with males having larger core areas (5.4 6 0.8 km2, n ¼ 6) than females (3.8 6 0.9 km2, n ¼ 7). We documented that, 11 months after installing artificial escape dens, a male–female pair of swift foxes moved into and established home ranges completely overlapping the artificial escape-den area, where no fox home ranges previously existed (Fig. 3). This pair of swift foxes remained in this treated area through the duration of our study. We documented 11 swift fox mortalities from January 2002 to August 2004. Of known causes of death, 70% were due to coyotes. Other mortalities included 3 unknown, 1 natural cause, and 1 raptor. The swift fox that died from natural cause was found dead inside its den and was necropsied after being dug out. Field necropsy revealed no hemorrhaging or puncture wounds. No swift fox mortalities occurred within an artificial escape-den grid, but 5 foxes belonging to the treated group were found dead outside the artificial den grid site. Of those 5 foxes, 2 were killed by coyotes, 2 were unknown, and 1 was due to natural causes. We calculated survival on 41 swift foxes (28 treated, 13 untreated). Fourteen juvenile swift foxes were captured during the summer of 2004 and were not used in survival analysis. We 824 50% MCP _ x 95% MCP SE _ x SE 1.9 2.1 1.8 0.4 0.4 0.6 11.3 12.5 10.3 1.7 2.1 2.7 1.1 2.1 0.5 0.6 1.2 0.2 7.1 9.4 5.6 1.9 4.3 1.8 determined that annual swift fox survival was greater (Z ¼ 1.47, P ¼ 0.07) on treated sites (S^ ¼ 0.81) than untreated sites (S^ ¼ 0.52; Table 2). Relative swift fox abundance (Table 3) was higher in treated than untreated areas in 2002 (Yates’ v2 ¼ 4.61, P ¼ 0.03) and 2003 (Yates’ v2 ¼ 4.70, P ¼ 0.03) but not in 2004 (Yates’ v2 ¼ 2.67, P ¼ 0.10). Swift fox recruitment rates were not different between treated and untreated areas in 2002 (Yates’ v2 ¼ 0.21, P ¼ 0.65) or 2003 (Yates’ v2 ¼ 0.41, P ¼ 0.52) but were higher on treated areas in both years (Table 3). The nonsignificance in recruitment rates may be attributed to low sample sizes. Discussion Our study demonstrated that the addition of artificial dens sites could improve swift fox abundance and survival. On 4 separate occasions, we radiotracked and observed swift foxes within artificial escape dens during the day. As part of a larger study, we tracked swift foxes to their dens once a week (McGee 2005). Figure 3. Annual home ranges of swift foxes (light gray polygons, n ¼ 11) from 2003–2004 after installation of artificial escape dens (dotted rectangle) to determine if foxes would immigrate into an area saturated with dens on the Rita Blanca National Grassland (solid line) in northwest Texas, USA. Dark gray polygons represent annual home ranges of swift foxes (n ¼ 19) from 1999– 2001 before artificial den installation. Wildlife Society Bulletin  34(3) Table 2. Annual survival rate, number of swift foxes monitored, and radiodays for males, females, and all swift foxes combined on Rita Blanca National Grassland in northwest Texas, USA, 2002–2004. Treated areas Study period Aug 2002–Aug 2003 All foxes Males Females Aug 2003–Aug 2004 All foxes Males Females Pooled years All foxes Survival (95% CI) n Radiodays Survival (95% CI) n 0.71 (0.45–1.00) 1.00 (1.00–1.00) 0.52 (0.21–1.00) 15 6 9 2,158 1,033 1,125 0.39 (0.10–1.00) 0.26 (0.02–1.00) 0.48 (0.12–1.00) 6 2 4 772 271 501 0.89 (0.70–1.00) 0.78 (0.48–1.00) 1.00 (1.00–1.00) 13 7 6 3,031 1,480 1,551 0.60 (0.30–1.00) 0.45 (0.15–1.00) 1.00 (1.00–1.00) 7 4 3 1,441 906 535 0.81 (0.64–1.00) 28 5,189 0.52 (0.27–0.97) 13 2,213 Theoretical design and placement of artificial escape dens in our study was to provide swift foxes with temporary escape cover. We assumed that swift foxes would use the artificial dens as a place of escape from predators while away from their natural dens during their normal activities. Further evidence of use was indicated by higher swift fox survival (S^ ¼ 0.71–0.89) and abundance (11.65– 17.91 foxes/100 trap-nights) in treated areas. Swift fox survival in untreated areas (S^ ¼ 0.52) was similar to estimates reported in Texas (S^ ¼ 0.47; Kamler et al. 2003a), Colorado (S^ ¼ 0.52–0.53; Covell 1992), Kansas (S^ ¼ 0.45; Sovada et al. 1998), and New Mexico (S^ ¼ 0.53; Harrison 2003). We demonstrated that swift foxes would immigrate into a previously unoccupied area where artificial escape dens had been installed (Fig. 3). We chose an experimental area on NG where no swift fox home ranges had occurred for 3 years before treatment (Fig. 2). Less than 1 year after artificial escape-den installation, a swift fox pair immigrated into the area and remained for the duration of our study. Similar to other research (Covell 1992, Sovada et al. 1998, Harrison 2003, Kamler et al. 2003a), we found that coyotes were the primary cause of mortality when the cause of death was identifiable. Kamler et al. (2003b) suggested that interference competition with coyotes could suppress and reduce swift fox populations. Historically, wolves may have preyed upon and reduced coyote populations with probably little competition between wolves and swift foxes because they occupied less-related niches (Ballard et al. 2003, Kamler et al. 2003b). Thus, swift fox populations likely were much larger than what they are today. High annual mortality rates have been documented in many studies conducted throughout the range of swift fox (Allardyce and Sovada 2003). High mortality rates are indicative of poor Table 3. Estimates of relative abundance and recruitment for swift foxes on Rita Blanca National Grassland in northwest Texas, USA, 2002–2004. Relative abundancea Study period 2002 2003 2004 a b Recruitmentb Treated Untreated Treated Untreated 15.38 11.65 17.91 7.04 5.86 11.00 2.33 1.25 1.00 0.00 Relative abundance is number captured/100 trap-nights. Recruitment is number of young/reproducing adult. McGee et al.  Untreated areas Effects of Artificial Escape Dens on Swift Fox Radiodays habitat quality. We have shown that swift fox survival was greater in areas with artificial escape dens. Thus, we believe that artificial escape dens were an improvement in habitat quality on our study site. We predicted that artificial escape dens would increase swift foxes’ home ranges by allowing swift foxes to travel farther from their natural dens. However, our results indicated that artificial escape dens had little effect on swift fox home-range sizes. We suspect that artificial escape-den grid areas were not large enough to have an influence on swift fox home ranges. Our grid areas were 2.59 km2. Generally, swift fox home-range sizes ranged from 7.6– 32.3 km2 (Allardyce and Sovada 2003). During our study, average swift fox home ranges were 3.9–19.9 km2 (95% FK; Table 1). Swift fox home ranges in treated areas may have been larger if treated areas had been larger than swift fox home-range sizes. Swift foxes may have ranged over a larger area to use the artificial dens. Our results supported the hypothesis that lack of escape-den sites limited swift fox populations in northwest Texas, USA. We have shown that swift fox survival, abundance, and distribution were greater in areas with artificial escape dens. We believe that artificial escape dens can enhance swift fox populations by relieving coyote suppression in areas where den sites are limited. Management Implications Much of the habitat within historical swift fox range has been fragmented into native rangeland, conservation reserve program lands, and agricultural fields (Allardyce and Sovada 2003). However, many researchers have shown that swift foxes primarily inhabit areas of ‘‘overgrazed’’ native rangeland (Allardyce and Sovada 2003). Because of habitat loss and high depredation rates from coyotes, swift fox populations remain restricted in distribution throughout the Great Plains (Allardyce and Sovada 2003). Herein, our objectives were to examine the use of artificial escape dens for the enhancement of swift fox survival and conservation. We have demonstrated that artificial escape dens can benefit swift foxes in the short term, but further research on artificial dens is necessary to determine whether, as a result of den construction, swift fox abundance, survival, recruitment, and spatial expansion continues over the long term. Future research should include larger numbers of artificial escape dens over larger grid areas that 825 encompass or exceed swift fox home ranges or, perhaps, over areas of occupied habitat where escape cover is known to be low. We have shown that escape dens are an important factor in swift fox survival. Swift foxes use dens year-round for reproduction, resting, protection from predators, and avoidance of extreme climatic conditions (Egoscue 1979). Thus, den availability and the escape habitat they provide can constitute a limiting factor for swift fox populations. Installation of artificial escape dens for increasing swift fox populations in areas of high coyote abundance or for reintroduction efforts may be a viable alternative to coyotecontrol programs. However, long-term studies are needed to determine the extent and effectiveness of artificial escape dens on swift fox conservation. In April 2002, we spent a total of US$2,796.68 on 436.3 m of 20.32-cm-diameter corrugated plastic sewer pipes (US$6.41/m). With help from United States Forest Service personnel, which included use of their skid loader, 4 people were able to install all dens in 24 hours. Time and personnel costs were not considered in estimates. Acknowledgments Funding was provided by the National Fish and Wildlife Foundation and Texas Tech University. We thank the United States Forest Service personnel for allowing us conduct research on the Rita Blanca National Grasslands and for helping to install artificial escape dens. We thank those who assisted with the project, including M. Butler, A. McGee, M. Wallace, C. Tyler, R. Gilliland, J. Stith, H. Whitlaw, C. Boal, and P. Zwank. We also thank E. and B. Hampton for providing us with a place to stay while conducting field research. Literature Cited Allardyce, D., and M. A. Sovada. 2003. A review of the ecology, distribution, and status of swift fox in the United States. Pages 3–18 in M. A. Sovada and L. Carbyn, editors. The swift fox: ecology and conservation of swift foxes in a changing world. Canadian Plains Research Center, University of Regina, Saskatchewan, Canada. Althoff, D. P. 1980. Den and den-site characteristics of coyotes in southeastern Nebraska. Transcripts of Nebraska Academy of Science 8:9–14. Anderson, D. E., T. R. Laurion, J. R. Cary, R. S. Sikes, M. A. McLeod, and E. M. Gese. 2003. Aspects of swift fox ecology in southeastern Colorado. Pages 129–148 in M. A. Sovada and L. Carbyn, editors. The swift fox: ecology and conservation of swift foxes in a changing world. Canadian Plains Research Center, University of Regina, Saskatchewan, Canada. Ballard, W. B., L. N. Carbyn, and D. W. Smith. 2003. Wolf interactions with nonprey. Pages 259–271 in L. D. Mech and L. Boitani, editors. Ecology and conservation of the wolf. University of Chicago, Illinois, USA. Bekoff, M. 1977. Canis latrans. Mammalian Species 79:1–9. Bekoff, M. 1982. Coyote Canis latrans. Pages 447–459 in J. A. Chapman and G. A. Feldhamer, editors. Wild mammals of North America: biology, management, and economics. The Johns Hopkins University, Baltimore, Maryland, USA. Connelly, G. E., and W. M. Longhurst. 1975. The effects of control on coyote populations: a simulation model. University of California, Division of Agriculture Science Bulletin 1872:1–37. Covell, D. F. 1992. Ecology of the swift fox (Vulpes velox) in southeastern Colorado. Thesis, University of Wisconsin, Madison, USA. Cutter, W. L. 1958. Denning of the swift fox in northern Texas. Journal of Mammalogy 39:70–74. Cypher, B. L., and J. H. Scrivner. 1992. Coyote control to protect endangered San Joaquin Kit foxes at the Naval Petroleum Reserves in California. Pages 42–47 in J. E. Borrecco and R. E. Marsh, editors. Proceedings of the fifteenth vertebrate pest conference. University of California, Davis, USA. Egoscue, H. F. 1979. Vulpes velox. Mammalian Species 122:1–5. Erickson, W. P., T. L. McDonald, K. G. Gerow, S. Howlin, and J. W. Kern. 2001. Statistical issues in resource selection studies with radio-marked animal. Pages 209–242 in J. J. Millspaugh and J. M. Marzluff, editors. Radiotracking and animal populations. Academic, San Diego, California, USA. Goodrich, J. M., and S. W. Buskirk. 1995. Control of abundant native vertebrates for conservation of endangered species. Conservation Biology 9: 1357–1364. Harrison, D. J., and J. R. Gilbert. 1985. Denning ecology and movements of coyotes in Maine during pup rearing season. Journal of Mammalogy 66:712– 719. Harrison, R. L. 2003. Swift fox demography, movements, denning, and diet in New Mexico. The Southwestern Naturalist 48:261–273. Heisey, D. M., and T. K. Fuller. 1985. Evaluation of survival and cause-specific mortality rates using telemetry data. Journal of Wildlife Management 49:668– 674. Hillman, C. N., and J. C. Sharps. 1978. Return of swift fox to northern Great Plains. Proceedings of the South Dakota Academy of Science 57:154–162. 826 Hines, T. D. 1980. An ecological study of Vulpes velox in Nebraska. Thesis, University of Nebraska, Lincoln, USA. Jackson, V. L., and J. R. Choate. 2000. Dens and den sites of the swift fox, vulpes velox. The Southwestern Naturalist 45:212–220. Kamler, J. F., W. B. Ballard, R. L. Gilliland, P. R. Lemons, and K. Mote. 2003a. Impacts of coyotes on swift foxes in northwestern Texas. Journal of Wildlife Management 67:317–323. Kamler, J. F., W. B. Ballard, R. L. Gilliland, and K. Mote. 2003b. Spatial relationships between swift foxes and coyotes in northwestern Texas. Canadian Journal of Zoology 81:168–172. Kitchen, A. M., E. M. Gese, and E. R. Schauster. 1999. Resource partitioning between coyotes and swift foxes: space, time, and diet. Canadian Journal of Zoology 77:1645–1656. Matlack, R. S., P. S. Gipson, and D. W. Kaufman. 2000. The swift fox in rangeland and cropland in western Kansas: relative abundance, mortality, and body size. The Southwestern Naturalist 45:221–225. McGee, B. K. 2005. Swift fox ecology in northwest Texas. Dissertation, Texas Tech University, Lubbock, USA. Mohr, C. O. 1947. Table of equivalent populations of North American small mammals. American Midland Naturalist 37:223–249. Nicholson, K. L. 2004. Swift fox occurrence in prairie dog towns in northwest Texas; with notes on movement. Thesis, Texas Tech University, Lubbock, USA. Pruss, S. D. 1999. Selection of natal dens by the swift fox (Vulpes velox) on the Canadian prairies. Canadian Journal of Zoology 77:646–652. Pulliam, H. R. 1988. Sources, sinks, and population regulation. American Naturalist 132:652–661. Rodgers, A. R., and A. P. Carr. 1998. HRE: the home range extension for ArcViewe: user’s manual. Centre for Northern Forest Ecosystem Research, Ontario Ministry of Natural Resources, Thunder Bay, Canada. Samuel, D. E., and B. B. Nelson. 1982. Foxes. Pages 475–490 in J. A. Chapman and G. A. Feldhamer, editors. Wild mammals of North America: biology, management, and economics. John Hopkins University, Baltimore, Maryland, USA. Schauster, E. R., E. M. Gese, and A. M. Kitchen. 2002. An evaluation of survey methods for monitoring swift fox abundance. Wildlife Society Bulletin 30: 464–477. Scott-Brown, J. M., S. Herrero, and J. Reynolds. 1987. Swift fox. Pages 433– 441 in M. Novak, J. A. Baker, M. E. Obbard, and B. Malloch, editors. Wild furbearer management and conservation in North America. Ontario Ministry of Natural Resources, Toronto, Canada. Seaman, D. E., J. J. Millspaugh, B. J. Kernohan, G. C. Brundige, K. J. Raedeke, and R. A. Gitzen. 1999. Effects of sample size on kernel home range estimates. Journal of Wildlife Management 63:739–747. Sovada, M. A., C. C. Roy, J. B. Bright, and J. R. Gillis. 1998. Causes and rates of mortality of swift foxes in western Kansas. Journal of Wildlife Management 62:1300–1306. Sovada, M. A., C. C. Slivinski, R. O. Woodward, and M. L. Phillips. 2003. Home range, habitat use, litter size, and pup dispersal of swift foxes in two distinct landscapes of western Kansas. Pages 149–1160 in M. A. Sovada Wildlife Society Bulletin  34(3) and L. Carbyn, editors. The swift fox: ecology and conservation of swift foxes in a changing world. Canadian Plains Research Center, University of Regina, Saskatchewan, Canada. SPSS. 2003. SPSS 12.0 user’s guide. SPSS, Chicago, Illinois, USA. U.S. Fish and Wildlife Service. 1995. Endangered and threatened wildlife and plants: 12-month finding for a petition to list the swift fox as endangered. Federal Register 60:31663–31666. U.S. Fish and Wildlife Service. 2001. Endangered and threatened wildlife and plants: annual notice of findings on recycled petitions. Federal Register 66: 1295–1300. White, G. C., and R. A. Garrott. 1990. Analysis of radiotracking data. Academic, San Diego, California, USA. White, P. J., K. Ralls, and R. A. Garrott. 1994. Coyote–kit fox interactions as revealed by telemetry. Canadian Journal of Zoology 72:1831–1836. Zar, J. H. 1999. Biostatistical analysis. Fourth edition. Prentice-Hall, Upper Saddle River, New Jersey, USA. Brady K. McGee received his Ph.D. in wildlife science in May 2005 at Texas Tech University, where he studied swift fox ecology. He received his M.S. in wildlife biology from Texas State University in San Marcos and a B.A. in biology and a B.S. in zoology from the University of Texas at Austin. He currently is working as a wildlife biologist for U.S. Fish and Wildlife Service in Alamo, Texas. Warren B. Ballard is professor and associate chair in the Department of Range, Wildlife, and Fisheries Management at Texas Tech University. His research interests include predator–prey relationships and population dynamics of carnivores and ungulates. Kerry L. Nicholson is currently a Ph.D. candidate in wildlife at the University of Arizona. Her McGee et al. View publication stats  Effects of Artificial Escape Dens on Swift Fox dissertation research is focused on urban mountain lion issues. She earned her B.S. degrees from the University of Alaska Fairbanks in Wildlife Biology and Biological Sciences and her M.S. degree from Texas Tech University. Her M.S. degree focused on swift fox and black-tailed prairie dog interactions. Brian L. Cypher is a research ecologist with the California State University–Stanislaus, Endangered Species Recovery Program. His primary research interest is the ecology and conservation of wild canids. His research experience includes work on wolves, coyotes, gray foxes, red foxes, kit foxes, and island foxes. Since 1990, he has been involved in research and conservation efforts for endangered San Joaquin kit foxes and other sensitive species in the San Joaquin Valley of California. He serves on recovery teams for San Joaquin kit foxes and island foxes and also is a member of the International Union for Conservation of Nature and Natural Resources Canid Specialists Group. Patrick R. Lemons II currently is a Ph.D. candidate in Ecology, Evolution, and Conservation Biology at the University of Nevada, Reno. His dissertation research is focused on secondary reproductive strategies of arctic nesting geese in Alaska. He received his M.S. from Texas Tech University in 2001 and his B.S. in Wildlife Biology and Natural History from Kansas State University in 1999. Jan F. Kamler received his B.S. in biology from the University of Kansas, his M.S. in wildlife biology from Kansas State University, and his Ph.D. in wildlife science from Texas Tech University. He currently is a Marie Curie Fellow at Oxford University conducting postdoctoral research on canid interactions in South Africa. His research interests include conservation biology, predator–prey relationships, and the ecology and interactions of carnivores. Associate Editor: Pitt. 827