Post‐translocation dynamics of black‐tailed prairie dogs (Cynomys ludovicianus): A successful conservation and human–wildlife conflict mitigation tool

Abstract Prairie dogs have declined by 98% throughout their range in the grasslands of North America. Translocations have been used as a conservation tool to reestablish colonies of this keystone species and to mitigate human–wildlife conflict. Understanding the behavioral responses of prairie dogs to translocation is of utmost importance to enhance the persistence of the species and for species that depend on them, including the critically endangered black‐footed ferret. In 2017 and 2018, we translocated 658 black‐tailed prairie dogs on the Lower Brule Indian Reservation in central South Dakota, USA, a black‐footed ferret recovery site. Here, we describe and evaluate the effectiveness of translocating prairie dogs into augered burrows and soft‐released within presumed coteries to reestablish colonies in previously occupied habitat. We released prairie dogs implanted with passive integrated transponders (PIT tags) and conducted recapture events approximately 1‐month and 1‐year post‐release. We hypothesized that these methods would result in a successful translocation and that prairie dogs released as coteries would remain close to where they were released because of their highly social structure. In support of these methods leading to a successful translocation, 69% of marked individuals was captured 1‐month post‐release, and 39% was captured 1‐year post‐release. Furthermore, considerable recruitment was observed with 495 unmarked juveniles captured during the 1‐year post‐release trapping event, and the reestablished colony had more than doubled in the area by 2021. Contrary to our hypothesis, yet to our knowledge a novel finding, there was greater initial movement within the colony 1‐month post‐release than expected based on recapture locations compared with the published average territory size; however, 1 year after release, most recaptured individuals were captured within the expected territory size when compared to capture locations 1‐month post‐release. This research demonstrates that while translocating prairie dogs may be socially disruptive initially, it is an important conservation tool.

effectiveness of translocating prairie dogs into augered burrows and soft-released within presumed coteries to reestablish colonies in previously occupied habitat. We released prairie dogs implanted with passive integrated transponders (PIT tags) and conducted recapture events approximately 1-month and 1-year post-release. We hypothesized that these methods would result in a successful translocation and that prairie dogs released as coteries would remain close to where they were released because of their highly social structure. In support of these methods leading to a successful translocation, 69% of marked individuals was captured 1-month post-release, and 39% was captured 1-year post-release. Furthermore, considerable recruitment was observed with 495 unmarked juveniles captured during the 1-year post-release trapping event, and the reestablished colony had more than doubled in the area by 2021. Contrary to our hypothesis, yet to our knowledge a novel finding, there was greater initial movement within the colony 1-month post-release than expected based on recapture locations compared with the published average territory size; however, 1 year after release, most recaptured individuals were captured within the expected territory size when compared to capture locations 1-month post-release. This research demonstrates that while translocating prairie dogs may be socially disruptive initially, it is an important conservation tool.

| INTRODUC TI ON
Prairie dogs (Cynomys spp.) are colonial ground squirrels that have an important role as a keystone species across the North American Great Plains though their distribution has been reduced by 98% of the historical range (Hoogland, 2006a). Through their herbivory and burrowing behaviors, prairie dogs create unique habitat within the western grasslands. Clipping and grazing results in short vegetation patches often consisting of forbs and grasses, combined with the creation and maintenance of burrow networks and mounds offer habitat heterogeneity relative to grasslands uninhabited by these ecosystem engineers (Davidson et al., 2012;Whicker & Detling, 1988). Numerous animal species are associated with prairie dog colonies, including those that use the habitat created, such as burrowing owls (Athene cunicularia) and mountain plovers (Chadrius montanus); and those that prey heavily on prairie dogs, such as ferruginous hawks (Buteo regalis), American badgers (Taxidea taxus), coyotes (Canis latrans), and blackfooted ferrets (Mustela nigripes) (Clark et al., 1982;Kotliar et al., 1999).
Several species that are prairie dog associates are of conservation concern, most notably of which is the endangered black-footed ferret-an obligate predator of prairie dogs and requires extensive and relatively intact prairie dog colonies (Antolin et al., 2002).
There are five species of prairie dogs in North America, two of which are considered threatened (Utah prairie dog (C. parvidens)) or endangered (Mexican prairie dog (C. mexicanus)) and the remaining three species of prairie dogs (black-tailed prairie dog (C. ludovicianus), white-tailed prairie dog (C. leucurus), and Gunnison's prairie dog (C. gunnisoni)) are rare compared with historical populations (Ceballos et al., 1993;USFWS, 2000USFWS, , 2012USFWS, , 2013USFWS, , 2017. The black-tailed prairie dog is listed as threatened in Canada under the Species at Risk (Parks Canada Agency, 2021). Threats to prairie dogs include rangeland conversion, inadequate regulatory mechanisms resulting in direct persecution, and disease (Luce et al., 2006). Prairie dogs are highly susceptible to sylvatic plague, which is caused by a non-native bacterium (Yersinia pestis) and is likely the most significant threat to the persistence of prairie dogs and many of their associated species (Antolin et al., 2002;Cully et al., 2006;Eads & Biggins, 2015;USFWS, 2000). Plague epizootics can lead to extensive landscape level impacts because colonies often can experience >90% mortality and recolonization can be slow (Cully et al., 2010;Eads & Biggins, 2015) or never reach pre-epizootic levels (Antolin et al., 2002;Biggins & Eads, 2019, R. Matchett, personal communication).
Translocation is the human-mediated movement of living organisms from one area, with release in another (IUCN/SSC, 2013).
Developing successful translocation programs for ecologically important species experiencing defaunation has many potential benefits for conservation (Watson & Watson, 2015). Translocation offers a restoration method that can repopulate suitable habitat much more quickly than would occur naturally (Dullum et al., 2005). The reestablishment of self-sustaining wildlife populations is challenging and improvement of approaches, including the inclusion of the social behavior of the species, have been added to the field of conservation translocation science to guide on-the-ground implementation (Goldenberg et al., 2019). Active restoration of prairie dog colonies through translocation can serve to contribute to conservation goals for both prairie dogs and their associated species (Dullum et al., 2005), while simultaneously offering a wildlife-human conflict mitigation opportunity (Nelson & Theimer, 2012).
Prairie dog translocations have been used as a tool for many years, with different methods that include releasing into artificial burrow systems made from corrugated, plastic drainpipe and artificial nest boxes (Bly-Honness et al., 2004;Curtis et al., 2014;Roe & Roe, 2004), intact natural burrows (Truett et al., 2001), and burrows created with gas-powered augers (Dullum et al., 2005).
The biotic and abiotic factors of previously occupied habitat or suitable novel habitat have been important for release site selection; furthermore, site preparations that consider the behavioral ecology of prairie dogs such as group compositions, mowing vegetation to suitable height and predator deterrents have been principal considerations for recipient sites (Long et al., 2006). During translocations, prairie dogs have been hard-released and soft-released, which can include the use of acclimation cages and supplemental feeding (Long et al., 2006;Truett et al., 2001). Some researchers have stressed the importance of releasing prairie dogs with other members of their coteries of the source colonies due to their highly social population structure (Bly-Honness et al., 2004;Shier, 2006). Coteries are polygynous units that typically contain one breeding male, related females, and their young that maintain a spatially discrete territory (Hoogland, 2006b). Across methods, there has been a wide range of success documented ranging from low survival rates and the failure to establish colonies to the establishment of colonies that have expanded in size and persisted (Curtis et al., 2014;Dullum et al., 2005;Roe & Roe, 2004;Shier & Owings, 2007). Choosing the ideal suite of translocation methods

K E Y W O R D S
black-footed ferret, black-tailed prairie dog, coterie, Cynomys ludovicianus, social behavior, translocation

T A X O N O M Y C L A S S I F I C A T I O N
Applied ecology, Behavioural ecology, Biodiversity ecology, Community ecology, Conservation ecology, Demography, Ecosystem ecology, Landscape ecology, Life history ecology, Movement ecology, Population ecology, Restoration ecology, Spatial ecology needs to be based on a variety of factors including but not limited to release site condition (e.g., soil type, pre-existing burrows, etc.), prevalence of predators, history of previous extirpation and sylvatic plague mitigation tools, cultural or ecological significance of the site, and resource availability.
We evaluated the effectiveness of the translocation of blacktailed prairie dogs to reestablish colonies in previously occupied habitat on the Lower Brule Indian Reservation in central South Dakota, USA, where prairie dogs had been absent following sylvatic plague epizootic events in 2011 and 2013. Prior to plague epizootics, the site supported a population of federally endangered black-footed ferrets. Because of adjacent topographical features and the lack of nearby colonies of prairie dogs, natural recolonization of the site had been especially slow with annual site visits confirming the areas remained unoccupied since plague epizootic events occurred (Grassel, personal communication). To meet wildlife conservation objectives for the site, a past translocation was conducted in 2015 when 520 prairie dogs were captured elsewhere and released. The release sites were prepared by mowing vegetation and opening burrows that had partially collapsed with gas-powered augers. Release burrows were treated with deltamethrin to control fleas and prevent mortalities from plague (Seery et al., 2003). Prairie dogs were hard-released without the use of acclimation cages or supplemental food, coteries were not maintained, and prairie dogs failed to establish from this effort (Grassel, personal communication). While this effort implemented some of the guidelines for conservation translocations, there were aspects in the release strategy that offered the opportunity for improvements (IUCN/SSC, 2013). In the next iteration of the translocation effort, further modifications were made to incorporate potential behavioral considerations including social structure and predator avoidance.
Our primary objective was to reestablish prairie dog colonies using soft-release methods while maintaining coterie structure and evaluate restoration success. The site had a history of being occupied and the translocation methods we employed had been successful elsewhere, therefore, we anticipated that translocated prairie dogs would successfully recolonize the site. We considered animal retention one-month post-release as our short-term measure of translocation success, with success further defined as maintaining a sex ratio of adults that would allow for pup recruitment the following spring and greater than 50% of animals released recaptured. Assuming short-term success was realized, we quantified mid-term success 1-year post-release as continued animal retention, juvenile recruitment, and colony expansion, which would all signify a successful translocation. Through continued monitoring and mapping of the site, we evaluated continued success defined as site occupancy and expansion three or more years post-release.
A secondary objective was to enhance understanding of coterie dynamics after release. As previously noted, coterie dynamics are highly complex and vary depending on the family group, sex (male or female), and age (juvenile or adult) composition, all of which influence the group size of the coterie. In turn, these complex interactions may influence the cohesion of the coterie and resiliency to disturbances. Because of the strong social structure of colonial species, we hypothesized prairie dogs would remain in the immediate area where they were released as a coterie. To complement this spatial component, we also examined whether the number of individuals in a coterie and age and sex composition influenced site retention. A better understanding of these objectives is necessary for conservation practitioners working to maintain the ecological function of North America's most imperiled biome (Comer et al., 2018), and ultimately the conservation of the prairie dog ecosystem, including contributions to the recovery of black-footed ferrets.

| Study site
We translocated prairie dogs to two inactive prairie dog colonies

| Site preparation
Prior to capturing prairie dogs at the source colony, the release sites were mowed with a rotary mower powered by a small tractor to ensure that vegetation was of suitable height (i.e., <25 cm).
In 2017, 2.61 ha was mowed at the Charlie site, and in 2018, 2.98 ha was mowed at the Charlie's South site. Starter burrows were created using gas-powered augers with 10-cm-diameter drill bits, ranged from 0.9 to 1.2 m in depth, and were angled at approximately 45°. The release sites mimicked the spatial orientation of the source sites and had areas designated for coteries.
Each area designated for a coterie was approximately 600 m 2 , with 16 m buffers between coterie extents, and starter burrows within coteries were 8 m apart. In 2017, the Charlie release site accommodated 39 coteries and in 2018, the Charlie's South site accommodated 50 coteries. Within each coterie, there were four starter burrows with single-opening and four starter burrows with double openings. Previous translocation efforts observed that prairie dogs released into augered burrows with a single opening sometimes covered themselves with soil, whereas prairie dogs released into pre-existing natural burrows displayed more natural behavior (Dullum et al., 2005), which typically have more than one opening (Cincotta, 1989). Starter burrows with double openings were constructed by using an auger to dig two burrows that intersected. Both single-opening and double-opening burrows had chambers at the deepest point of the burrow, which were created by using hand tools to remove soil and dig out a cavity. The chambers in burrows with single openings were approximately 0.3 m 2 and were large enough to accommodate a minimum of two adult prairie dogs, and the chambers in burrows with double openings were approximately 0.5 m 2 and were large enough to accommodate a minimum of three adult prairie dogs. Deltamethrin was applied to each burrow and annually thereafter as a plague mitigation strategy Seery et al., 2003).

| Prairie dog capture
We captured prairie dogs at source colonies that were near Lower Brule, South Dakota, USA. Source colonies were located approximately 48 km away and were undesired because of land use conflicts (i.e., livestock grazing, within highway right-of-way). Between related or neighboring individuals to be captured, held, and released together. Prior to trap placement, we noted patterns in vegetation created by the foraging and clipping behavior of prairie dogs and conducted observations of prairie dogs, which guided the delineation of coterie boundaries and the location of trap clusters. We prebaited traps with a commercially available oat/molasses feed for 3-5 days prior to trapping for all sites to increase capture success (Long et al., 2006).
Flushing provided a viable alternative when leaving traps for extended periods was not an option. The flushing technique is the flooding of active burrows with approximately 200-800 L of water and a small amount of dishwashing detergent (~ 0.25 L per 100 L of water) from a portable water tank and 5.1 cm hose. A small amount of dishwashing detergent is added to create foam that limits the ability of prairie dogs to see and allows them to exit safely. Prairie dogs are captured by hand as they exit the burrow and placed into portable kennels. All prairie dogs captured from one burrow were placed into one kennel. Kennels were marked by coterie to ensure related or neighboring animals were released together.

| Processing
Upon capture prairie dogs were kept overnight to be processed the following day. Prairie dogs were kept in kennels in a secure large shelter and provided with grain, lettuce, and carrots. Prairie dogs were anesthetized using isoflurane mixed with oxygen and administered with a vaporizer. Prairie dogs were weighed to a tenth of a gram and implanted with a passive integrated transponder (PIT: AVID® Identification Systems, Norco, CA) tag for individual identification. Sex and age classes (juvenile and adult) were recorded as determined by weight and appearance (Hoogland, 1995). Groups that were trapped and flushed in presumed coterie units based on spatial orientation, were processed sequentially, and placed together into labeled plastic kennels lined with grass hay, carrots, and oats for food. Following processing, animals were taken to the release site later that same day. All methods used were approved by the Department of Wildlife, Fish and Recreation of the Lower Brule Sioux Tribe.

| Release
In 2017 and 2018, prairie dogs were released into burrows to maintain the coteries of the source colonies. The spatial orientation of neighboring coteries from the source colonies was also maintained. The locations of all burrows and coteries were recorded with a GPS receiver (Garmin Ltd., Olathe, Kansas). For coteries that had multiple individuals to release, we ensured that individuals were released into both single-opening burrows and into double-opening burrows. Up to three individuals were released into single-opening burrows and up to five individuals were released into double-opening burrows. Groups typically contained one adult male and the remaining individuals were adult females and juveniles of both sexes. We placed a wire mesh cage (0.8 m × 0.9 m × 0.4 m) over each burrow used for release to provide temporary protection from aerial and terrestrial predators and to discourage immediate dispersal. The acclimation cages were kept over the burrow openings for 3 days, during which grain, lettuce, and carrots were provided. Prairie dogs could easily burrow under the acclimation cages to get out or back in.

| Recapture
To sample the population at the release site; animals were recap- Captured animals were weighed, scanned for PIT tag identification number, marked with permanent marker, and released into the nearest burrow. All individuals that did not have PIT tag identifiers were marked with permanent markers, and sex and age recorded.
During subsequent trapping days, traps that contained marked individuals were released immediately.

| Colony expansion
To evaluate colony retention and expansion, we mapped the perimeter of both prairie dog colonies at the time of release and each year until 2021 ( Figure 1). The perimeter was the outermost burrows that had clear visual signs of use (e.g., prairie dog present, fresh scat or diggings), which was the augered burrows at the time of release. We input each boundary in ArcGIS v10.7.1 (ESRI, Redlands, CA, USA) and calculated the area of the colony for each year. If the occupied extent of the colony encompassed the same area, or increased, at the end of the study period, we considered this to be an indication of continued success, i.e., that the site had been successfully recolonized or established 3-4 years post-release.

| Distance moved post-release
To understand whether prairie dogs remained within their coterie as released or whether they moved within the colony and established new territories, we measured the distance from where each prairie dog was initially released to their 1-month location and from their 1-month location to their 1-year location in ArcGIS v10.7.1. Furthermore, we compared the average distance between recapture locations to the diameter of small (0.05 ha), average (0.32 ha), and large (1.01 ha) territory sizes (Hoogland, 2006b) and identified the percent of prairie dogs that were recaptured within the given territory size across each time step. We used chi-square goodness-of-fit tests to determine whether the proportion of prairie dogs that moved in each territory size differed between the Charlie's and Charlie's South colonies. We assumed that the distance between release and 1-month locations and 1-month and 1-year locations could provide a representation of the territory size of their coteries.

| Effect of coterie dynamics on retention
To determine whether the number of individuals in a coterie and the age structure and sex composition of the coterie influenced retention, a binary response variable, we used generalized linear mixed models (GLMM; Jamil et al., 2013) in Program R statistical base package (R Development Core Team, Version 3.6.32020). Individuals were released with the coterie they were captured with, therefore, we included coterie as a random effect to account for potential effects of social structure that might differ by coterie and not be explained by our predictor variables. In addition to sex (male or female), age (juvenile or adult), and the group size of the coterie, we also included weight (g), burrow release type (single or double openings), and release site (Charlie's or Charlie's South) as fixed effect predictor variables to account for additional variation. We also created a combined variable of sex and age (adult male or female and juvenile male or female). We did not allow these combined age-sex variables to enter the same model as age or sex because of the inherent correlation between them. No other variables were correlated at >0.6.
Rather than run all possible model combinations, which can result in overly complicated models with reduced biological meaning, we used expert opinion to determine biologically relevant models based on the biology of the system. We also included a global and a null model with just the random effect and intercept. We calculated AICc because of the small sample size, ΔAICc, AICc weight, and log-likelihood to rank models (Burnham & Anderson, 2002). Models within 2 AICc of the top model were considered indistinguishable from each other (Burnham & Anderson, 2002). Once top models were identified, we calculated model-averaged parameter estimates with 95% confidence intervals (Arnold, 2010). Parameters whose confidence intervals did not overlap zero were determined to be important predictors or prairie dog recapture success. We repeated this process for 1-month and 1-year post-release.

| Release and recapture
We marked and translocated a total of 658 prairie dogs, 293 in 2017 at Charlie's, and 365 in 2018 in Charlie's South (Table 1) (Table 1). Of the 237 unmarked individuals, 14 were adults and 223 were juveniles. When both colonies were examined as a complex, 69% of the prairie dogs were recaptured 1-month post-release and 39% were recaptured 1year post-release (Table 1).

| Colony expansion
Throughout the duration of the study, and beyond, the full extent of the original release site remained occupied, and the overall footprint of the colony expanded (Grassel, personal communication; Figure 1).
At the time of release, the Charlie's release site encompassed 2.61 ha. One year later, the colony had expanded to 5.19 ha, grew slightly larger by year two to 5.22 ha, encompassed 5.74 ha in year three, and was 5.75 ha in year four. At Charlie's South, prairie dogs were released across 2.98 ha. One year later, the Charlie's South colony had expanded to 4.2 ha, by year two the site encompassed 5.44 ha, and grew to 6.35 ha in year three. Combined, the two sites expanded from 5.59 ha to 12.10 ha more than twice the size (2.16 times greater) when the final year of mapping occurred.

| Distance moved post-release
There was a marked difference in the distance prairie dogs moved from the time of release to 1 month and from 1 month to 1 year (Table 2, Figure 2). Within territory sizes, prairie dogs did not move significantly different distances between the two colonies at 1 month (X 2 3 = 1.81, p = .61) or 1 year (X 2 3 = 1.29, p = .73); therefore, we combined the data from both colonies to represent the prairie dog complex. Between the 1-month recapture and the 1-year recapture, the percent of individuals that remained within the small-sized TA B L E 1 Count and retention rate, as a percent, of marked black-tailed prairie dogs released with passive integrated transponders (PIT) and recaptured, with and without PIT tags, at approximately 1-month and 1- territory (≤24 m) increased from an average of 28% to 67%, decreased within the medium-sized territory (25-64 m) from 34% to 24%, decreased within the large-sized territory (65-112 m) from 17% to 3%, and decreased beyond the large territory size (≥ 113 m) from 21% to 6% (Table 2, Figure 2).

| Distance moved by age sex class
In a post-hoc analysis that was deemed valuable after observing the overall distances moved, we also compared the distance moved across age and sex classes using analysis of variance with Tukey's pairwise comparisons to assess whether there were differences between age or sex classes by recapture period. At 1-month post-

| Effect of coterie dynamics on retention
Eight of the 13 models examined were considered top models (<2 AICc) and were similar in describing the retention of PIT-tagged individuals approximately 1-month post-release. Model-averaged TA B L E 2 Percent of marked black-tailed paired dogs that were recaptured within defined territory sizes from release to 1-month post-release, and from 1-month to 1-year for the Charlie's and Charlie's South colonies. Distances are binned based on territory size as defined in Hoogland (2006b (Table 3). Males had greater retention than females, and juveniles were more likely to be recaught than adults (Tables 1   and 3). When age and sex were combined as a grouped variable, juvenile males had the greatest retention rates (Table 3). There was no difference between sites, the number of individuals released as a coterie, or whether the animals were released into burrows with one to two openings.
At 1-year post-release, four models were among the top-ranked models (<2 AICc) in describing retention at the release site. Modelaveraged parameter estimates identified sex and weight as the only F I G U R E 3 Average distance (m) moved (error bars are standard errors) of prairie dogs by age and sex class for adult females (AF), adult male (AM), juvenile female (JF), and juvenile male (JM) between initial release and 1-month postrelease locations, and distance between 1-month locations to 1-year postrelease locations. Letters and ** indicate significant differences in average distance moved within recapture periods based on the Tukey's pairwise test (p < .05).

TA B L E 3 Model averaged parameter estimates for top models (<AIC c)
estimating retention at 1-month postrelease and 1-year post-release informative parameters with confidence intervals not overlapping zero (Table 3). Unlike at 1 month, males displayed lower retention than females, and recapture increased with weight at the time of release (Tables 1 and 3). Similar to the 1-month post-release, there was no site-level effect, the number of individuals released as a coterie, or whether the animals were released into burrows with one or two openings. The combined age and sex variable did not enter the top models during recapture 2 (Table 3).

| DISCUSS ION
The efforts to reestablish prairie dog colonies at this site were predicated on the opportunity to reduce human-wildlife conflict where source prairie dogs were present in an undesired location, while simultaneously restoring the species in the desired location. We translocated prairie dogs from source sites and successfully reestablished two prairie dog colonies that had not recovered from losses sustained from the epizootic plague. Mid-term success was demonstrated by successful recruitment 1-year post-release resulting in more prairie dogs being trapped than were released. Continued success was realized through colony occupancy and expansion for at least three and 4 years beyond the time of the translocation.
Recognizing that earlier efforts of reestablishing prairie dogs via translocation at the same release site were unsuccessful, we posit that the modifications made in our methodology were beneficial to the success of the translocation. Furthermore, we documented that while the translocation initially disrupted the complex social structure of prairie dog coteries, by 1 year later, it appeared that the colony structure had stabilized.
One year after the translocations, 39% of translocated individuals were recaptured plus an additional 495 unmarked juveniles (272 unmarked juveniles at Charlie's, and 223 unmarked juveniles at Charlie's South) indicating that, at minimum, the population had increased by at least 18%. The numbers of juveniles also illustrate that there was an adequate level of adult survival to result in recruitment and the potential for population persistence into the future. Because prairie dogs have been demonstrated to have larger litters in newer colonies than older ones (Garrett et al., 1982), the site likely offered ample resources that allowed the newly translocated individuals to successfully reproduce. Annual mapping of the site post-release confirmed that the entirety of the release site remained occupied, and that the colony extent had more than doubled three-and four-year post-release. We used colony extent and visual confirmation that the entirety of the release site had active prairie dog presence because colony area is the metric commonly used in management; however, we recognize that colony expansion does not quantify prairie dog densities or population size (Bruggeman & Licht, 2020).
While there was only one unmarked adult that was captured at 1-month post-release, there were nearly two dozen unmarked adults that were captured at 1-year recapture event. The unmarked adults captured may indicate immigration into the colony, or that individuals lost pit tags, which is rare but possible. If these unmarked individuals were immigrants, this could be an early indication that the reestablishment of these two colonies may contribute to the restoration and connectivity of the Fort Hale Bottom prairie dog complex. While beyond the scope of this study, further research evaluating how newly translocated colonies may influence the dispersal dynamics of conspecifics from surrounding colonies would add understanding to the composition and functioning of prairie dog complexes.
We followed translocation recommendations to utilize areas that were previously occupied by prairie dogs and had visible evidence of past occupation (Long et al., 2006;Truett et al., 2001). Additionally, the site was located within proximity to other colonies meeting dispersal and complex criteria; and a plan was in place to mitigate the initial reason the site was extirpated, by managing the site for sylvatic plague following the translocation efforts. However, because the previous translocation effort was considered unsuccessful and our selected release site had natural burrows with collapsed or filled-in openings; we made three modifications to our methodology: adding chambers in augered burrows, utilization of acclimation cages, and maintaining coteries at release burrows. Release site preparation often requires the installation of artificial burrows and nest boxes with the use of heavy equipment (Bly-Honness et al., 2004;Curtis et al., 2014;Davidson et al., 2014) or the usage of abandoned preexisting burrows (Long et al., 2006, Truett et al., 2001. Neither of these options were available to us due to the preclusion of using heavy equipment necessary to install artificial burrows and nest boxes and there were no pre-existing burrows that were in good condition. Soils have been identified as one of the most important variables for the translocation of other colonial, burrowing mammals (Swaisgood et al., 2019), and the soils at our site were highly conducive for effectively making temporary release burrows with the use of augers, which is not applicable to all translocation sites. The silty loam soil type at the release site allowed for using augers to create starter burrows and for using hand tools to chisel out chambers and remove excess soil at the deepest part of each augered hole. While the past failed translocation utilized augered starter burrows, this current effort included the creation of chambers that may have been more hospitable to animals upon initial release. Past translocation efforts at other sites have observed prairie dogs rarely using augered starter burrows for longer than a few days (Truett et al., 2001, Sterling Krank, personal observation (Long et al., 2006;Roe & Roe, 2004 Shier (2006) found that prairie dogs translocated with family units increased survival by a factor of five and females had higher reproductive success, but Bly-Honness et al. (2004) suggested the maintenance of coterie unity during translocation did not enhance post-release survival. While we did not examine survival in this study, we did examine what factors influenced site retention. We did not observe an influence of coterie size on retention, though we acknowledge this was not a paired design study with a control group that was not released as a coterie to compare to as that was not a primary goal of this translocation. We did, however, observe that while juvenile males had the highest retention rates 1-month post-release they had the greatest decline in retention by 1-year post-release, whereas adult retention remained relatively stable throughout. Furthermore, adult females and heavier individuals at the time of release also had higher retention rates. Juvenile prairie dogs are less likely than adults to survive until the following spring (Hoogland, 2006c), which may account for the observed reduction in retention. Additionally, adult females tend to have high site philopatry and heavier prairie dogs, regardless of sex, are more likely to survive overwinter (Hoogland, 2006c), which aligns with the observations in this study as well.
To our knowledge, this is the first study that has measured the distance moved by prairie dogs after a translocation, further enhancing our understanding of prairie dog social dynamics posttranslocation. Interestingly, and not as hypothesized, our spatial analysis provided mixed support that prairie dogs remained at the release area designated for their respective coteries. While only 28% of the translocated prairie dogs were recaptured within the small territory size 1-month post-release, 67% of the prairie dogs that were recaptured 1 year later were still within the small territory size of where they were at 1-month post-release. Conversely, 21% of the prairie dogs recaptured at 1 month moved further than the large territory size and just 6% of the recaptured prairie dogs at 1 year had made such considerable movements. This indicates that prairie dogs had largely established new territories 1-month post-release from where they were initially released, and still maintained those new territories 1 year later. Even with greater initial movement than expected, these movements were still within the footprint of the release site providing further support for our overarching objective of establishing prairie dog colonies through translocation.
Translocation is an initially disruptive event, and in this case, the greatest movements were by adults during the first-month post-release. Both male and female adult prairie dogs moved large distances immediately after release, more than twice the average distance moved by juveniles. The different distances moved by age and sex and the directional movements observed radiating from a coterie rather than in a uniform direction (Figure 2), suggests that entire coteries did not shift as a cohesive unit. We can only further hypothesize as to why this is, but one possibility is that adults are more territorial than juveniles and are more likely to venture further to find familiar territory. California ground squirrels have been observed to make large movements in search of familiar territory post-translocation, with 46% of surviving individuals returning to their original territory over the course of 1-16 days (Van Vuren et al., 1997). It was not possible for the prairie dogs observed during our study to return to their original territory because of the extensive distance between capture and release sites, but it is possible that individuals may have exhibited homing movements when acclimation cages were released. Nevertheless, this disruption appears to be short-lived-in that 91% of all prairie dogs that were recaptured 1-year post-release were within the small-and medium-sized territories of where they were 1-month post-release-and importantly, the release site remained occupied and expanded.
Territory sizes maintained by coteries vary and are likely dependent on a suite of variables that can be traced to ecological site conditions such as soil type, temperature, precipitation and ultimately plant productivity (Geaumont et al., 2016) and demographic features (Kusch & Lane, 2021). We used previously identified small, medium, and large territory size ranges (Hoogland, 2006b) (Shier & Swaisgood, 2011) and is particularly salient for highly social species as well (Goldenberg et al., 2019;Shier, 2006). Because our translocation effort met our definition of success, we recommend keeping coteries intact during capture and release, as this was one key modification to our methodology following the earlier failed translocation.

| CON CLUS ION
Sylvatic plague can result in widespread and dramatic population losses of prairie dogs; therefore, wildlife managers and conservation practitioners currently rely on plague mitigation tools and translocations to effectively maintain the prairie dog ecosystem.
Translocations of other mammals with the intention of restoring ecological processes, including many identified with the role that keystone species play, are increasingly being identified as a specific goal for translocation programs (Palmer et al., 2020;Swaisgood et al., 2019;Watson & Watson, 2015). An intended outcome of reestablishing prairie dogs at this site was to facilitate the ecological role that the species plays in North American grasslands. Nesting burrowing owls, ferruginous hawks, badgers, lekking sharp-tailed grouse (Tympanuchus phasianellus), and species of shortgrass prairie songbirds including horned lark (Eremophila alpestris) and killdeer (Charadrius vociferus) were all observed utilizing the new colonies (Guernsey and Grassel, personal observation). Anecdotally, this indicates that the reestablishment of prairie dogs can provide important habitat and prey base for many grassland species. We anticipate with further growth and recovery of these colonies-as part of a larger prairie dog complex-the site will likely support the critically endangered black-footed ferret as it once did.  Patrick Family Trust's support of this work. We also thank two anonymous reviewers for their valuable feedback, which improved the manuscript.

CO N FLI C T O F I NTE R E S T
There are no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
Upon acceptance, data will be made available in a publicly accessible repository such as Dryad.

R E FE R E N C E S
Antolin, M. F., Gober, P., Luce, B., Biggins, D. E., Van Pelt, W. E., Seery, D. B., Lockhart, M., & Ball, M. (2002). The influence of sylvatic plague on north American wildlife at the landscape level, with special emphasis on black-footed ferret and prairie dog conservation.
Transactions of the North American Wildlife and Natural Resources Conference, 67, 104-127.