Keystone effects of prairie dogs (Cynomys spp.) on grassland birds: Current knowledge and future directions

Abstract

Keystone species are critical to maintaining ecological relationships and ecosystem functioning around the world. Prairie dogs (Cynomys sp.) are a genus of colonial keystone species endemic to North American grasslands that modify bird community composition via ecological disturbance, but this topic has not been reviewed despite many studies on grassland bird associations with prairie dogs. We conducted a systematic literature review (n = 111 studies) where we assessed three hypotheses and found consistent support for each: (1) prairie dogs increase abundance of early‐successional grassland birds through grazing, (2) prairie dog towns modify bird community composition compared to nearby grassland because they increase heterogeneity within grasslands at large spatial scales, and (3) prairie dogs increase raptor abundance by serving as a prey base. Significant gaps in our understanding of the effects of prairie dogs on birds remain. Most research was conducted on black‐tailed prairie dogs during the bird breeding season and focused on how prairie dogs modify the occurrence and abundance of birds. Research is needed on the effects of other prairie dog species, effects on birds during the non‐breeding or migratory seasons, and effects on demography, particularly for songbirds. Continued work on this topic can increase our understanding of how keystone species affect bird communities and inform conservation of grassland birds globally. Due to rapid declines of grassland birds in North America, we suggest conserving prairie dog towns as vital sources of early successional grassland and habitat heterogeneity.

INTRODUCTION

Keystone species shape the occurrence, composition, life history, and demography of bird communities (Augustine & Baker, 2013; Boyce et al., 2021; Buyandelger & Otgonbayar, 2022; Feldman et al., 2023; Lai & Smith, 2003; Zhao et al., 2020), and restoration of keystone species can benefit bird populations (Boyce et al., 2022; Griebel et al., 1998; Nummi & Holopainen, 2014; Powell, 2006). Social, burrowing, herbivorous mammals (hereafter, keystone rodents) such as ground squirrels (Sciuridae), Patagonian Maras (Dolichotis patagonum), and Springhares (Pedetes capensis) are an important yet globally imperiled functional group of keystone species found in grasslands worldwide, where they affect bird communities by grazing and clipping vegetation, burrowing, and serving as prey (Davidson et al., 2012). Foremost, their grazing, burrowing, and clipping create distinct patches of early successional, short-stature vegetation and create large-scale habitat heterogeneity within grasslands. This ecosystem engineering has important effects on grassland bird communities globally because grassland birds inhabit a spectrum of grazing intensity: for example, many birds prefer taller, infrequently disturbed habitat (e.g., Cock-tailed Tyrant (Alecturus tricolor) and Henslow's Sparrow (Centronyx henslowii)), while others such as the Ground Tit (Pseudopodoces humilis) inhabit sparse, short-stature vegetation maintained by regular disturbance (Augustine & Derner, 2012; Farnsworth et al., 2020; Fitzpatrick & Sharpe, 2020; Gosler et al., 2020; Green et al., 2020; Herkert et al., 2020; Jaramillo, 2020; Knopf & Wunder, 2023; Remsen & Sharpe, 2020; Wiersma et al., 2020). Other species require varying levels of disturbance for different portions of their life histories. For example, Greater Prairie-Chickens (Tympanuchus cupido) use short-stature grassland for lek sites but use more dense vegetation for nesting & brood rearing. Thus, bird species that require either specific or multiple disturbance levels within grasslands can benefit from heterogeneity created by keystone species (Fuhlendorf et al., 2006, 2009; Hovick et al., 2015).

Keystone rodent impacts go beyond simply impacting bird habitat. Certain birds use their burrows as nest sites; their burrows increase the diversity of arthropod prey items (Davidson et al., 2012; Davidson & Lightfoot, 2007), and they themselves serve as an important food source for many raptor species (Davidson et al., 2012; Zhao et al., 2020). At the same time, keystone rodents could have negative effects on or create tradeoffs for grassland birds. For example, they could increase predation risk via apparent competition by attracting generalist predators (Holt, 1977), providing dens for them, and decreasing vegetation for nest concealment within their colonies (Baker et al., 1999, 2000).

Prairie dogs (Cynomys sp.) are a genus of five social, burrowing, herbivorous rodents and are keystone species in the grasslands and steppes of western North America (Davidson et al., 2012; Hoogland, 2013; Kotliar, 2000; Kotliar et al., 1999). Black-tailed Prairie Dogs (Cynomys ludovicianus) are the most well-studied, have the largest distribution, and live in the most dense colonies (Hoogland, 2013). Because prairie dogs are highly social and live in dense colonies (“towns”), their grazing, burrowing, and vegetation clipping create large, spatially discrete patches of open habitat with shorter grass, increased bare ground, and increased forbs that support distinct bird communities (Cid et al., 1991; Davidson et al., 2012; Hoogland, 2013; Winter et al., 2002). Prairie dogs also serve as a prey base for many predators including raptors, provide refuge for herptiles and mammals via their burrows, and change the composition of arthropod and small mammal communities (Cartron et al., 2002; Cully et al., 2010; Davidson & Lightfoot, 2007; Shipley & Reading, 2006).

All prairie dog species have declined drastically due to habitat destruction, introduced disease, and systematic persecution, and the Utah Prairie Dog (Cynomys parvidens) and Mexican Prairie Dog (Cynomys mexicanus) are considered endangered by the IUCN (Álvarez-Castañeda et al., 2019; Davidson et al., 2012; Hoogland, 2013; Miller et al., 2007; Roach, 2016). Estimating the historical abundance of prairie dogs is difficult, but accounts describe vast Black-tailed Prairie Dog colony complexes covering thousands of hectares in the early twentieth century prior to eradication efforts (Bailey, 1905; Knowles et al., 2002). Today Black-tailed Prairie Dogs occupy 2% of their historical range, and where they are found, their colonies are smaller and more fragmented (Hoogland, 2013). Because grassland birds select habitats at large spatial scales (Duchardt, Beck, & Augustine, 2020; Fisher & Davis, 2010), the loss and fragmentation of prairie dog colonies can have negative consequences for birds that benefit from their presence (Augustine et al., 2008; Cully, 1991; Desmond et al., 2000; Dinsmore & Smith, 2010).

Prior research has focused on Burrowing Owls and Mountain Plovers as inhabitants of prairie dog towns, raptor predation of prairie dogs, and surveys of bird occurrence on vs. off prairie dog colonies (Augustine & Baker, 2013; Cook et al., 2003; Desmond & Savidge, 1996; Dinsmore et al., 2005). However, as keystone species, prairie dogs could potentially have both direct and indirect effects on bird communities with profound outcomes on their life histories (Duchardt et al., 2019, 2021).

Despite the numerous studies on how prairie dogs affect birds, this body of work has not been systematically reviewed. Reviewing these studies is required to synthesize a more complete understanding of how and why prairie dogs affect grassland birds and add to the growing literature on how keystone species shape animal communities (Davidson et al., 2012). The steep declines of grassland birds further underscore the need to better understand how different bird species may respond to the multifaceted effects of prairie dogs. Since keystone species structure grassland bird communities, conserving and restoring prairie dogs could benefit grassland bird populations. We set out to review the state of knowledge on how grassland birds are affected by prairie dogs, evaluate hypotheses about those effects, identify any biases or knowledge gaps in the literature, and suggest future directions for research in this system.

We identified three hypotheses from the literature to evaluate: (1) prairie dog towns increase local occurrence and abundance of bird species that prefer shorter vegetation heights because prairie dog grazing creates more suitable habitat for them, (2) prairie dog towns modify bird community composition compared to nearby grassland because they increase heterogeneity within grasslands at large spatial scales, and (3) prairie dog towns increase local occurrence and abundance of raptors because prairie dogs are an abundant food source for them. To identify potential biases or knowledge gaps, we examined the representation of differences in geography, seasonality, taxonomy, town size, and response variables across publications.

METHODS

Data collection

We conducted our systematic review in accordance with recommended best practices (Page et al., 2021). Prior to conducting our systematic search, we finalized the aims, hypotheses, and approach of our study. We were interested in papers on the ecological effects of prairie dogs on birds, indirect effects of prairie dogs on taxa that could affect birds (e.g., arthropods), and indirect effects of prairie dog mortality via sylvatic plague on bird population dynamics. We excluded papers not relevant to community ecology (e.g., prairie dog sociality or medical details of plague).

We conducted a search of the Web of Science database on September 26, 2022 using the following keywords: “prairie dog* AND bird* OR avian OR songbird* OR passerine OR raptor* OR communit* OR ecolog*.” After our initial search, we updated our criteria to exclude papers on selenium accumulation in plants, prairie dog genetics, prairie dog conservation, Black-footed Ferret captivity, and flea biology, but also to include papers on direct effects of plague mitigation on birds. Our final search returned 641 results (Supplement 1). From there, we screened the titles and abstracts for relevance based on our inclusion/exclusion criteria and retrieved 237 relevant papers. Of these, 221 were able to be retrieved, and their full texts were assessed for eligibility, leaving a total of 186 eligible papers (Supplement 1). Although we did not exclude Spanish-language publications, all eligible papers were published in English.

We then used the “backwards snowball” method to review the references of these papers to include any other relevant papers not captured by the search (Page et al., 2021). This resulted in a final total of 298 papers to evaluate. Of these studies, 111 directly tested at least one of our hypotheses, and we compiled them into a synthesis matrix. We extracted the following information from each paper: (1) prairie dog species; (2) bird species; (3) locality information: country, state/province, approximate latitude/longitude, Bird Conservation Region (ecoregions used for conservation planning; United States US NABCI Monitoring Subcommittee, 2007), and whether land was tribally owned/managed; (4) study length by year(s)/month(s) of sampling and seasonality information; (5) management information: habitat type(s), patch size, livestock grazing regime, fire regime, prairie dog town size; (6) response type (e.g., occurrence, nesting, diet, prey consumption, eavesdropping); (7) response variable measured by authors (e.g., abundance, alpha & beta diversity); and (8) response variable data: sample sizes, means, variance, effect sizes. Many studies examined multiple response variables, so we also considered the number of analyses per study. Our resulting matrix consisted of 166 discrete analyses. We defined each analysis as consisting of a unique response variable under evaluation (e.g., a study that tested alpha and beta diversity would consist of two analyses). The lead author (RAD) determined relevance for all papers and extracted data from each of them. Many of the papers did not report effect sizes, and there were a wide variety of response variables used to study a given hypothesis, so conducting a meta-analysis was not possible. Instead, we proceeded with a systematic review.

We categorized habitat types based on author descriptions and assigned each to one of the following categories: shortgrass prairie, mixed-grass prairie, sagebrush steppe, badlands, piñon-juniper woodland, desert grassland, agriculture, and urban. We noted the presence of livestock as co-grazers based on author descriptions, but if the authors did not specify a co-grazer, we assigned one using online records of public and private grazing land (e.g., Bureau of Land Management, business listings). We assigned seasonality based on the biology of the bird taxa under study (e.g., breeding season, post-breeding migration). We sorted the bird species studied into four categories based on the taxa commonly studied in publications: burrowing owls, mountain plovers, raptors, and general. The general category included studies on any bird species not covered in the other three categories, as well as full-community studies that included many bird species. We defined raptors following McClure et al. (2019), but we delineated burrowing owls from other raptors in our categorization because they do not consume prairie dogs and because they were usually studied exclusively (e.g., studies were only on burrowing owls, or studies on other raptors excluded Burrowing Owls). Mountain Plovers were placed in their own category because of the large number of publications focusing solely on this species and its relationship with prairie dogs. For studies that tracked prairie dog town sizes over multiple years, we took the maximum size for the reported summary statistics.

We assessed if prairie dog towns increased local occurrence or abundance of birds that prefer shorter vegetation heights (Hypothesis 1) by evaluating all analyses that examined any effect of prairie dogs on the presence or abundance of grassland bird species and determining whether the effect of prairie dogs was positive, neutral, or negative for each species. We classified effects on presence or abundance based on direction with respect to prairie dogs (e.g., declines in a bird species following declines in prairie dogs translates to a positive effect of prairie dogs on their abundance, while a decrease in a bird species on a prairie dog town compared to nearby grassland would represent a negative effect). We then checked these species against known habitat associations in the literature to confirm their habitat preferences (Askins et al., 2007; Samson & Knopf, 1996; Somershoe, 2018). Analyses that reported significant, positive responses for bird species that are associated with shorter vegetation heights exemplified support for Hypothesis 1.

We determined whether prairie dog towns modify bird community composition compared to nearby grassland (Hypothesis 2) by examining analyses of how prairie dog presence affected community composition and/or beta diversity. We considered Hypothesis 2 supported for analyses that showed significant differences in bird community composition between prairie dog towns and grassland without prairie dogs, or significant increases in bird beta diversity in grasslands that include prairie dog towns compared to those without. We also visually assessed our summary of the net effects of prairie dogs on the presence and abundance of bird species from hypothesis 1 (Figure 1) to determine whether prairie dogs modify bird community composition on prairie dog towns compared to nearby grassland by increasing some species while decreasing others.

[IMAGE OMITTED. SEE PDF]

To understand if prairie dog towns were associated with higher occurrence and abundance of raptors (Hypothesis 3), we included all analyses of how prairie dog presence influenced the presence or abundance of raptors. Studies were considered supportive when raptors consumed prairie dogs as a major (top 3) dietary item and/or when prairie dog presence was associated with positive effects on the presence, abundance, or nesting of raptors. Abundance effects were classified based on effect direction (e.g., declines in raptors following declines in prairie dogs translate to a positive effect of prairie dogs on their abundance). We accepted the alpha levels set by the authors for determining significance for all analyses. We performed all analyses in R (version 4.3.2, R Core Team 2023).

RESULTS

A total of 111 studies containing 166 analyses directly tested our hypotheses about the effects of prairie dogs on birds. Burrowing Owls were the most studied bird species, involved in 43% of publications (48), followed by 37% (41) that included Mountain Plover and 26% (29) that included Ferruginous Hawk. Across all studies, 105 bird species were documented using prairie dog towns in some capacity (Supplementary Table 1). Each of our three hypotheses about how prairie dogs affect birds was supported by a substantial majority of analyses. Prairie dog towns had a positive effect on the occurrence and abundance of grassland birds that inhabit early successional habitat compared to grassland without prairie dogs in 79% of relevant analyses (100 out of 127; Figure 2) for 22 (28%) different species out of 78 total (Figure 1). Compared to surrounding habitat, 20 (25%) species were negatively associated with prairie dog towns in our review, and 36 (46%) had a neutral association (Figure 1). Similarly, 88% of relevant analyses (15) found support for the hypothesis that prairie dog towns change bird community composition compared to nearby grassland because they increase heterogeneity within grasslands at large spatial scales. Last, 88% of relevant analyses (36) supported the hypothesis that prairie dogs as prey are associated with increased occurrence and abundance of nine raptor species (Figure 3).

[IMAGE OMITTED. SEE PDF]

[IMAGE OMITTED. SEE PDF]

Black-tailed Prairie Dogs were the most frequently studied species of Cynomys by far (included in 88% of studies; 98). Studies including the other four Cynomys species were far fewer: 9% (10) on Gunnison's Prairie Dog, 8% (nine) on White-tailed Prairie Dog, 5% (six) on Mexican Prairie Dog, and 1% (one) on Utah Prairie Dog. Black-tailed Prairie Dogs were the only species with enough studies to adequately calculate summary statistics on prairie dog town size. Across all studies, the median total area of Black-tailed Prairie Dog towns was 731 ha, the median average size of a single town was 35 ha, the median minimum town size was 3 ha, and the median maximum town size was 240 ha (studies with multiple response variables were only counted once). For habitat types represented in studies, shortgrass prairie was included most frequently at 73% (81), followed by mixed grass prairie with 48% (53), 29% (32) in sagebrush, 17% (19) in agriculture, 9% (10) in desert grassland, 5% (6) in urban, 5% (5) in badlands, and 5% (5) in Piñon-Juniper woodland (5).

Megaherbivores were common co-grazers in studies. Domestic cattle (Bos taurus) were included in 89% of study systems (99), 6% (7) included Bison (Bison bison), 7% (8) had no other grazer present, 4% (4) included goats, and 2% (2) included sheep. Of these studies, 2% (2) were conducted at the continental scale and thus were not assigned a co-grazer. Due to the limited number of studies on non-cattle co-grazers, we did not consider them further in our analysis. Fire was rarely present as a co-occurring disturbance. Three studies included prescribed dormant season burns, and one included a wildfire. Only 2% (3) compared the effects of prairie dog towns, prescribed fire, and cattle grazing as disturbances on a grassland bird species.

Although prairie dogs are distributed across Canada, Mexico, and the United States, 94% of studies (104) took place exclusively within the United States, while only 9% of studies (10) included Mexico and just one included Canada (Figure 4). Within Bird Conservation Regions, 37% of studies (41) were conducted in the Shortgrass Prairie region, followed by 23% (25) in the Badlands and Prairies region, and 12% (13) in the Southern Rockies/Colorado Plateau region (Supplementary Figure 2).

[IMAGE OMITTED. SEE PDF]

Several sites stood out as hubs for the study of prairie dog effects on birds; 44% (49) of studies included at least one of these localities: Phillips County, MT (13); Thunder Basin National Grassland, WY (10); eastern Colorado (7); the Oklahoma Panhandle (7); Pawnee National Grassland, CO (6), and Denver, CO (6). Prairie dog populations persist over large spatial scales at each of these sites, and three of these sites are predominantly federal lands administered by the Bureau of Land Management or the Forest Service. Just 7% (8) of studies included data collection on federally reserved tribal land (i.e., reservations).

Of the studies, 93% (103) included the breeding season of the focal bird species, while 11% (12) of studies examined the northern hemisphere winter (hereafter winter), 4% (four) of studies studied either post-breeding migration, and an additional 10% (11) of studies sampled across the full annual cycle. The median study duration was 2 years.

The effect of prairie dogs on bird occurrence and/or abundance was the most common topic studied with 48% (79) of discrete analyses, followed by 37% (61) on nesting, 11% (18) on diet, 3% (five) on movement, and 2% (three) on survival (Figure 5). Overall, the effect types studied varied substantially depending on the type of bird under study. Mountain Plover was the only species to have a publication examining each effect type, even though raptors and burrowing owls had the greatest number of studies overall (Figure 5). Of the 18 diet analyses, 67% (12) focused on raptor predation of prairie dogs. Only Mountain Plover, Burrowing Owl, and raptors had studies on how prairie dogs affect nesting, movement, or survival. There were no studies on how prairie dogs affect nesting, movement, or survival for any wild bird in the general category (Figure 5). The two studies on nesting within the general category used artificial nests.

[IMAGE OMITTED. SEE PDF]

DISCUSSION

Keystone species have multifaceted effects on birds that can affect individual fitness, community composition, and demographic trends (Augustine & Baker, 2013; Boyce et al., 2021; Buyandelger & Otgonbayar, 2022; Feldman et al., 2023; Lai & Smith, 2003; Zhao et al., 2020), and keystone rodents are found in grasslands worldwide (Davidson et al., 2012). Given the extensive and rapid decline of grassland birds globally, we require a better understanding of how keystone rodents like prairie dogs affect the occurrence, abundance, and demography of grassland birds to inform conservation efforts. Our goal was to systematically review the literature on how prairie dogs affect grassland bird communities by evaluating hypotheses about their effects, identifying knowledge gaps, and synthesizing new research directions. Overall, our review shows that prairie dogs, especially Black-tailed Prairie Dogs, are an important source of disturbance for bird communities. We found strong support in the literature for all of the hypotheses we tested. Specifically, prairie dogs increase the occurrence and abundance of early-successional bird species through their ecosystem engineering, drive changes in avian community composition via habitat heterogeneity, and increase raptor occurrence and abundance by serving as prey. More research beyond bird occurrence and abundance (e.g., nesting, movement, diet), outside the breeding season, with more prairie dog species, and over a wider geographic range will help inform grassland bird conservation efforts.

Our hypothesis that prairie dog towns provide early successional habitat for birds was supported. In our review, 22 out of 78 bird species (28%) preferentially occurred on prairie dog towns compared to nearby habitat (Figure 1). We infer that prairie dogs creating early successional habitat is the primary mechanism for these associations because of the extensive literature detailing how vegetation structure drives habitat selection for grassland birds, and the species associated with prairie dog towns are known to prefer short-stature vegetation (Fisher & Davis, 2010; Fuhlendorf et al., 2009; Samson & Knopf, 1996). We could not test other mechanisms, such as modifications to arthropod communities, and these may also contribute to species associations with prairie dogs because short-stature vegetation alone is not necessarily sufficient for prairie dog associates (Augustine & Derner, 2012). Given that prairie dogs support grassland birds that prefer early-successional habitat, we can infer that loss of prairie dog colonies has resulted in declines of birds associated with prairie dogs, and this suite of birds extends beyond known associates like Mountain Plover and Burrowing Owl.

We found support for our hypothesis that prairie dogs prairie dog towns modify bird community composition compared to nearby grassland because they increase heterogeneity within grasslands at large spatial scales. (Figure 1). This not only benefits birds associated with early-successional habitat, but it also benefits species that use multiple habitat types for different purposes or life stages (e.g., nesting in taller grass and foraging in shorter grass) like Long-billed Curlew, Upland Sandpiper, and Mourning Dove (Augustine & Baker, 2013). At the same time, we found that 20 species were negatively associated with prairie dog towns, while 36 had no preference for prairie dog towns compared to surrounding habitat (Figure 1). Species that were negatively associated included those that prefer dense ground cover or late-successional habitat (e.g., Grasshopper Sparrow, Baird's Sparrow) and shrubland species (e.g., Brewer's Sparrow, Lark Bunting; Duchardt et al., 2018, 2019; Duchardt, Augustine, & Beck, 2020). However, long-term prairie dog disturbance did not result in declines of two other sagebrush obligates: Greater Sage-Grouse (Centrocercus urophasianus) or Sage Thrasher (Oreoscoptes montanus; Duchardt, Augustine, & Beck, 2020). Sprague's Pipit was also not negatively associated with prairie dogs, which was surprising given their preference for late successional habitat in the breeding season (Figure 1). Overall, this finding parallels how prairie dogs drive community composition changes in studies of other taxonomic groups including arthropods and small mammals (Cully et al., 2010; Davidson & Lightfoot, 2007, 2008; Nimwegen et al., 2008). Habitat heterogeneity is a central concept in grassland management because of how multiple disturbance sources have shaped the evolution of grassland ecosystems (Fuhlendorf et al., 2006). Since prairie dogs play a vital role in providing habitat heterogeneity for grassland birds, the widespread declines in prairie dog colonies have likely resulted in decreased beta diversity for grassland bird communities across the continent.

Our hypothesis that prairie dogs increase raptor occurrence and abundance was also supported because prairie dogs were major prey items for multiple raptor species during the breeding season and winter (Figure 3). Abundance of Ferruginous Hawk and Golden Eagle declined with decreases in prairie dogs (Cully, 1991), and Golden Eagles, Bald Eagles, and Prairie Falcons selected nest sites closer to prairie dog towns than expected at random (Aagaard et al., 2021). Ferruginous Hawks had higher productivity closer to prairie dog towns (Cartron et al., 2002). Prairie dog towns had a negative effect on the presence of Swainson's Hawk and American Kestrel in 50% of studies (5 and 2, respectively), which is expected because Swainson's Hawks rarely eat prairie dogs, and American Kestrels are too small to eat them. While 11 of the 12 studies that examined raptor diet directly found prairie dogs were important prey items, there are other potential mechanisms that may contribute to raptor associations with prairie dogs that we were unable to evaluate. By grazing early-successional habitat, prairie dogs may create open areas where raptors can capture prey more effectively (Duval & Whitford, 2012), and prairie dogs increase the abundance of other common prey for raptors such as mice (Peromyscus sp.) and cottontails (Sylvilagus sp.) (Cully et al., 2010; Lomolino & Smith, 2004; Shipley & Reading, 2006). Overall, these findings suggest that losses of prairie dogs negatively impact some raptor populations, which is concerning because raptors have valuable ecological roles as apex predators (Donázar et al., 2016; Sergio et al., 2014).

As the species most closely associated with prairie dogs, Mountain Plover, Burrowing Owl, Ferruginous Hawk, and Horned Lark may stand to benefit the most from prairie dog conservation efforts, at least during their breeding seasons. Mountain Plover in particular may be both the species at greatest risk with a population size of only 20,000 individuals (Bird Life International, 2024) and the species that would benefit the most from prairie dog conservation because cattle grazing alone does not create sufficient disturbance for them (Augustine & Derner, 2012). Yet, our review has identified a broader suite of species that benefit from prairie dog towns, such as Long-billed Curlews, Chestnut-collared Longspurs, and Prairie Falcons. Therefore, conserving prairie dogs as keystone species on landscapes would likely benefit an entire suite of grassland bird species.

We must note that publication bias against negative or neutral results in ecology and literature reviews overall is an issue (Jennions & Møller, 2002; Lortie et al., 2007). Out of 217 instances where the effect of prairie dogs on the occurrence or abundance of a species was analyzed (Figure 1), 76 (35%) were positive, 85 (39%) were neutral/no effect, and 56 (26%) were negative. Because the proportion of positive responses is not highly skewed, we infer there is a low magnitude of publication bias affecting this review, but we encourage researchers to publish on neutral or negative results in this system, particularly because understanding how effects vary with environmental context is essential for keystone species ecology.

At the same time, our review identifies important knowledge gaps in grassland bird ecology and synthesizes new avenues to learn more about the effects of prairie dogs on birds. Most studies focused on Black-tailed Prairie Dogs, likely because they have the densest colonies and largest range (Hoogland, 2013). Despite the focus on areas with more intact prairie dog populations, the median average size of Black-tailed prairie dog towns studied (35 ha) was well below historical descriptions of common prairie dog town sizes (Knowles et al., 2002), suggesting that much of our knowledge derives from reduced prairie dog colonies. Only 11 studies evaluated how fluctuations in prairie dog populations affect birds, namely through epizootics of introduced sylvatic plague, which cause prairie dog colonies to collapse (Augustine et al., 2008; Cully, 1991; Davidson et al., 2022; Dinsmore & Smith, 2010; Duchardt et al., 2022; Eads & Biggins, 2015). Prairie dog species besides black-tailed prairie dogs may be less tractable or less compelling to study because their colonies are smaller and more fragmented (Giglio et al., 2021; Scott-Morales et al., 2004; Scott-Morales et al., 2005). However, it is important to understand the effects of prairie dogs on birds across species because Utah Prairie Dogs (C. parvidens) and Mexican Prairie Dogs (C. mexicanus) are endangered (Álvarez-Castañeda et al., 2019; Roach, 2016). It is likely that all prairie dog species maintain habitat heterogeneity to some degree, similar to our findings with Black-tailed Prairie Dogs, but other prairie dog species may not be as effective at creating early successional habitat at larger scales because their colonies are smaller and less dense. Research expressly testing hypotheses over a range of prairie dog town sizes and through fluctuations in prairie dog populations is needed to better understand how such effects may vary with the scale and abundance of prairie dogs.

Most studies on bird occurrence compared grassland birds on vs. off prairie dog colonies. Yet grassland birds select habitat at the landscape scale and frequently use multiple habitat types for different purposes (Derner et al., 2009; Fisher & Davis, 2010), and prairie dog towns produce edge effects (Duchardt et al., 2019), so patterns of bird occurrence may not be consistent within a given town or at a given time. The focus on binary, on vs. off colony comparisons may be due to the ease of statistical analysis or logistical constraints (Agnew et al., 1986; Augustine & Baker, 2013; Ray et al., 2015; Winter et al., 2003). To produce a more complete view of how grassland birds use prairie dog towns as early-successional habitat, we suggest studies on the edge effects of prairie dog towns in greater detail, movement biology of individual species with respect to prairie dog towns, and landscapes with and without prairie dogs (Dreitz, 2009; Duchardt et al., 2021; Haun et al., 2024).

Few studies investigated components of avian life history besides occurrence, perhaps because point count surveys are a more cost-effective methodology compared to other study types. There were no studies on the nesting biology of songbirds in the context of prairie dogs. Raptors were the focus of 67% of diet studies (12), and only three studies examined the movement of birds around prairie dog colonies. This may be due to technological limitations, but advances in DNA metabarcoding (McClenaghan et al., 2019; Rytkönen et al., 2019; Souza-Cole et al., 2022) and tracking technology (Costa-Pereira et al., 2022; Jetz et al., 2022) make such studies increasingly feasible. Studies on survival were lacking, fitting with overall knowledge gaps on grassland birds (Bernath-Plaisted et al., 2023; Hostetler et al., 2015). We recommend studies that research how prairie dogs may modify the nesting, movement, diet, and demography of grassland birds (Figure 5). For instance, Burrowing Owls benefit from increased foraging efficiency on prairie dog towns and eavesdrop on prairie dog alarm calls (Bryan & Wunder, 2013; Duval & Whitford, 2012), but no studies have been conducted on whether other grassland birds derive similar benefits.

Following biases in ornithology and ecology (Marra et al., 2015), we found spatio-temporal bias in our review. Studies overwhelmingly focused on short-term effects during the breeding season, with only 12 and four studies focusing on winter and migration, respectively. Since successful bird conservation requires understanding full annual life cycles (Hostetler et al., 2015; Marra et al., 2015), these are important gaps to fill. There was also a bias towards the Northern and Central Great Plains versus the Intermountain West and Southern Plains and a bias towards studies on black-tailed prairie dogs in the western part of their range compared to the eastern part (Figure 4). This spatial bias may be driven by the focus on areas where Black-tailed Prairie Dogs remain in more ecologically relevant numbers and on public land (Figure 4). The exception to this spatio-temporal bias was studies on raptors and prairie dogs, since raptors were the focus of 75% (9) of winter studies (Figure 4). However, we have no data on raptor species that overlap with prairie dogs on the Northern Great Plains during winter, such as Snowy Owl (Bubo scandiacus), Short-eared Owl (Asio otus), and Gyrfalcon (Falco rusticolus), which may consume prairie dogs and/or small mammals and lagomorphs associated with prairie dog towns (Cully et al., 2010; Goguen, 2012; Nimwegen et al., 2008). More research on how prairie dogs affect bird communities outside of the breeding season (M. Desmond, 2004), over longer periods of time (e.g., through cycles of drought and plague, Duchardt et al., 2022), and in less-sampled parts of the Great Plains will greatly increase our understanding of community ecology and inform grassland bird conservation efforts.

Further, the stark lack of studies on tribal land is concerning. Tribal nations steward a disproportionate amount of prairie dog colonies (Hoogland, 2013; Olimb et al., 2022; Sidle et al., 2001) and are often conservation leaders despite limited resources (Pejchar et al., 2021; Sanderson et al., 2008; Shamon et al., 2022). We acknowledge that this review is limited to the Web of Science database, so there may be tribal research that was not accessible within the parameters of our review and from our Western science perspective. Reviewing traditional ecological knowledge on the relationship between birds and prairie dogs would be extremely valuable but is outside our scope and expertise. Regardless, we encourage researchers to redress this bias and follow best practices to collaborate equitably with Indigenous and local communities who steward a large portion of North America's remaining grasslands (Hanna & Vanclay, 2013; Robinson et al., 2021).

The findings of this review have major implications for the conservation of both grassland birds and social, burrowing, herbivorous mammals, which are often subjects of population control due to perceived conflict with agriculture (Davidson et al., 2012; Lai & Smith, 2003). In particular, prairie dog management involves conflict and controversy because prairie dogs are systematically controlled due to perceived competition with beef cattle (Augustine et al., 2024; Augustine & Derner, 2021; Elmore et al., 2007; Fox-Parrish & Jurin, 2008; Miller et al., 2007; Titus & Jachowski, 2021), and prairie dog control efforts involve secondary poisoning of raptors and ground-foraging birds (Ruder et al., 2011; Vyas et al., 2017; Vyas, Hulse, Meteyer, & Rice, 2012; Vyas, Hulse, & Rice, 2012; Witmer et al., 2016). However, the evidence reviewed here demonstrates that eliminating or reducing prairie dogs causes indirect harm to grassland birds since it removes or reduces a valuable source of ecological disturbance, habitat heterogeneity, and prey.

Given the dire state of grassland bird populations and conflict around prairie dog management, there is a need to better understand how prairie dogs benefit grassland birds so rangeland managers can accurately balance biodiversity conservation and agricultural production, particularly under multiple-use mandates on public land. Grazing interactions between prairie dogs and cattle are complex because prairie dogs can have neutral or positive effects on cattle grazing under certain conditions, but they can also have negative effects on cattle weight gains during drought, particularly on productive soils (Augustine et al., 2024, and references therein). Coupled with wide inter-annual variability in precipitation and prairie dog extent, this presents a challenge for managers of rangeland systems to support ecological integrity while remaining economically viable (Davidson et al., 2022; Duchardt et al., 2022).

Despite their benefits to birds, prairie dogs have largely not been considered in conservation plans for bird species other than Mountain Plover and Burrowing Owl (e.g., Bernath-Plaisted et al., 2023; Somershoe, 2018). Similar issues characterize birds and keystone rodent conservation planning in other grasslands, such as Plateau Pikas (Ochotona curzoniae) in China (Arthur et al., 2008; Jiapeng et al., 2015; Lai & Smith, 2003; Lambert et al., 2022).

There is still much to learn about how prairie dogs and keystone rodents affect grassland birds, and there is an urgent need for improved conservation and management practices to reverse pervasive, global declines in grassland birds (Douglas et al., 2023; Rosenberg et al., 2019). As a functional group that often benefits birds yet are often subjects of management conflict with agriculture, keystone rodents need to be considered in conservation planning for grassland birds, especially in programs to incentivize ranching practices that promote coexistence between cattle and prairie dogs. Specifically, prairie dog-derived ecosystem services should be integrated into cost–benefit frameworks for rangeland management, incentives for retaining prairie dogs should be expanded (Gripne, 2005), protected areas should be established where economic pressures do not cause conflict with agricultural production, and tools to mitigate sylvatic plague should be expanded and improved (Eads et al., 2021; Matchett et al., 2023).

We hope that this review provides an informative foundation for researchers to study how keystone species affect bird communities and for land managers, producers, and practitioners working in grasslands to better integrate social, burrowing, herbivorous mammals into conservation planning for grassland birds.

AUTHOR CONTRIBUTIONS

R. A. D., A. J. B., and H. P. J. conceived the idea and formulated the hypotheses. R. A. D. conducted the literature search, compiled the data into the synthesis matrix, analyzed the results, and wrote the manuscript. A. B. J. and H. P. J. supervised the research and provided substantial edits to the manuscript.

ACKNOWLEDGMENTS

The authors would like to thank the many dedicated researchers who contributed studies on birds and prairie dogs that made this synthesis possible, the members of the Evidence-based Restoration Lab and Boyce Lab who gave constructive feedback on the draft manuscript, and the anonymous peer reviewers who further improved the manuscript. This work was supported by Northern Illinois University and Smithsonian's National Zoo and Conservation Biology Institute. It was also made possible by the generous support of John and Adrienne Mars. The study was a systematic review that analyzed publicly available studies and was conducted following best practices for reviews.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

DATA AVAILABILITY STATEMENT

Analyses reported in this article can be reproduced using the data provided by Dreelin et al. (2024).

References

Aagaard, K., Conrey, R. Y., & Gammonley, J. H. (2021). Nest distribution of four priority raptor species in Colorado. Journal of Raptor Research, 55(4), 510–523. [DOI: https://dx.doi.org/10.3356/JRR-20-47]