Next Article in Journal
The Evolution of Mosasaurid Foraging Behavior Through the Lens of Stable Carbon Isotopes
Previous Article in Journal
High Diversity of Waters, Communities and Stressors—Design of Croatian Fish Index for Rivers (CFIR)
Previous Article in Special Issue
Current Distribution of the Nine-Banded Armadillo (Dasypus novemcinctus) in the United States
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Diversity
Volume 17
Issue 4
10.3390/d17040290
diversity-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Habitat and Predator Influences on the Spatial Ecology of Nine-Banded Armadillos

by
Robert C. Lonsinger
1,*,
Ben P. Murley
2,
Daniel T. McDonald
3,
Christine E. Fallon
3 and
Kara M. White
4
1
U.S. Geological Survey, Oklahoma Cooperative Fish & Wildlife Research Unit, Oklahoma State University, Stillwater, OK 74078, USA
2
Department of Natural Resource Ecology & Management, Oklahoma State University, Stillwater, OK 74078, USA
3
U.S. Fish & Wildlife Service, Wichita Mountains Wildlife Refuge, Indiahoma, OK 73552, USA
4
Institute of Arctic Biology, University of Alaska Fairbanks, Fairbanks, AK 99775, USA
*
Author to whom correspondence should be addressed.
Diversity 2025, 17(4), 290; https://doi.org/10.3390/d17040290
Submission received: 5 March 2025 / Revised: 10 April 2025 / Accepted: 18 April 2025 / Published: 19 April 2025
(This article belongs to the Special Issue Ecology, Behavior, and Conservation of Armadillos)
Browse Figures
Versions Notes

Abstract

:
Mesopredator suppression has implications for community structure, biodiversity, and ecosystem function, but mesopredators with physical defenses may not avoid apex predators. We investigated nine-banded armadillos (Dasypus novemcinctus) in southwestern Oklahoma (USA) to evaluate if a species with physical defenses was influenced by a dominant predator, the coyote (Canis latrans). We sampled nine-banded armadillos and coyotes with motion-activated cameras. We used single-species and conditional two-species occupancy models to assess the influences of environmental factors and coyotes on nine-banded armadillo occurrence and site-use intensity (i.e., detection). We used camera-based detections to characterize the diel activity of each species and their overlap. Nine-banded armadillo occupancy was greater at sites closer to cover, with lower slopes, and further from water, whereas coyote space use was greater at higher elevations; both species were positively associated with recent burns. Nine-banded armadillo occurrence was not influenced by coyotes, but site-use intensity was suppressed by the presence of coyotes. Nine-banded armadillos (strictly nocturnal) and coyotes (predominantly nocturnal) had a high overlap in summer diel activity. Nine-banded armadillos are ecosystem engineers but are often considered a threat to species of concern and/or a nuisance. Thus, understanding the role of interspecific interactions on nine-banded armadillos has important implications for conservation and management.

1. Introduction

Mesopredator suppression by apex predators is widespread and has implications for community structure, biodiversity, and ecosystem function [1,2]. Mesopredators must capture prey while avoiding predation from apex predators, and top-down pressure on mesopredators may release (or reduce constraints on) primary consumers [3,4]. Apex predators may influence mesopredators directly via predation or killing [5,6] or indirectly through fear-induced shifts in resource use, temporal activity, or spatial occurrence of mesopredators [2]. Thus, investigating the mechanism by which apex predators suppress mesopredators is important to disentangling trophic cascades, understanding factors influencing patterns of biodiversity, and informing conservation.
Mesopredators have evolved a suite of strategies to mitigate top-down pressure or evade apex predators. Small carnivores often rely on crypsis (e.g., short-tailed weasels; Mustela erminea) or adaptations such as fossorial (e.g., black-footed ferrets; Mustela nigripes), arboreal (e.g., African linsangs; Poiana richardsonii), or scansorial (e.g., ringtails; Bassariscus astutus) tendencies that limit accessibility by apex predators [7,8]. Aposematic mesopredators such as skunks (e.g., Family Mephitidae) signal antipredator defenses that may deter predators through innate or learned avoidance [9,10]. Insectivorous mesopredators that forage in habitats with limited cover have commonly evolved spines (e.g., hedgehogs; Erinaceus spp.), scales (e.g., pangolins; Manis spp.), or osteoderms (e.g., armadillos; Dasypus spp.) as physical defenses or deterrents [8,11]. If evolved antipredator defenses are sufficient to withstand or deter predation, mesopredators with evolved defenses may not need to avoid apex predators. For example, experimental studies have indicated that predators were hesitant to approach [12] or learned to avoid [9] striped skunk (Mephitis mephitis) models, and studies have suggested that apex predators may have no [13] or limited [1,14] influence on the space-use patterns the free-ranging striped skunks in some systems.
Nine-banded armadillos (Dasypus novemcinctus) are insectivorous mesopredators distributed broadly across much of the western hemisphere from northern Argentina and Uruguay through central North America, and they have evolved both physical and behavioral antipredator defenses [15]. Nine-banded armadillos are armored by osteoderms (i.e., ossified dermal plates or scutes forming a carapace), and they dig burrows that serve as escape refugia [15]. Although the burrowing and foraging behaviors of nine-banded armadillos have led to them being considered a nuisance in some areas (e.g., due to damage to golf courses and lawns) [16], nine-banded armadillos are important ecosystem engineers due to their burrowing activities [17,18]. In the United States, nine-banded armadillos have been expanding in distribution since the 1850s [19,20], and their burrows may replace some of the ecosystem function lost by the extirpation or decline of other burrowing species, e.g., [21,22]. Nine-banded armadillos are poor thermoregulators [15], and their range expansion is expected to be limited by annual precipitation and low winter temperatures (or the number of freezing days) [19], but they have already expanded beyond initial predictions [20]. More recent predictions of the physiological limits of nine-banded armadillos indicate potential continued expansion across much of the eastern United States to a latitude of ~40° N [23].
The influence of apex predators and top-down regulation on nine-banded armadillos has not been investigated but has implications for understanding the effectiveness of antipredator defenses and disentangling factors limiting nine-banded armadillos within their range to inform conservation and management. It has been suggested that humans may be the only predator capable of limiting nine-banded armadillos [15,19]. In portions of the United States recently colonized by nine-banded armadillos, coyotes often serve as the de facto apex predator [24]. Nine-banded armadillos have been detected in the diets of coyotes, but only at trace levels (<3%, e.g., [25,26,27]), and the source of consumption (e.g., predation versus scavenging of vehicle-killed animals) was unknown. Research in Texas found no evidence of adult nine-banded armadillos being killed by predators, but >71% of juvenile mortalities were attributed to predation, suggesting that smaller-bodied individuals with less-developed carapaces may be more susceptible to predation [28]. Although coyotes have been documented pursuing adult nine-banded armadillos (e.g., Figure 1 [29]), they may not avoid coyotes if their physical and behavioral antipredator defenses are sufficient.
We investigated the spatial ecology of nine-banded armadillos in the Wichita Mountains Wildlife Refuge (WMWR) as a model system to assess the influence of coyotes on nine-banded armadillos. We coupled camera-based sampling with single-species and conditional two-species occupancy models [30,31] to investigate the relative influences of environmental factors (e.g., distance to cover) and an apex predator (i.e., coyotes) on the probabilities of occurrence (occupancy) and detection for nine-banded armadillos. Occupancy modeling provides a flexible framework to investigate the probability that a species occurs within a site while accounting for imperfect detection, environmental factors, and interspecific interactions [30,31]. We interpreted variation in nine-banded armadillo detection as reflecting variation in site-use intensity [32,33,34,35]. We hypothesized that the detection of both species would be influenced by temperature and precipitation. We identified five general environmental covariates hypothesized to influence species-specific occupancy or space-use patterns: (i) terrain, (ii) water availability, (iii) land cover, (iv) fire history, and (v) human activity. We predicted that after accounting for the effects of environmental factors, coyotes would not limit the occurrence of nine-banded armadillos. In contrast, we predicted that coyotes would limit the site-use intensity of nine-banded armadillos, presumably due to fear-induced behavioral changes [36] and/or decreases in relative abundance due to predation risk to juveniles [28]. Finally, we used camera-based detections to characterize diel activity patterns of each species and hypothesized that both nine-banded armadillos [36,37] and coyotes [38] would be primarily nocturnal with a high degree of overlap in diel activity patterns.

2. Materials and Methods

2.1. Study Area

Our study was conducted at the WMWR in southwestern Oklahoma, USA (Figure 2), which encompassed ~240 km2 of lands managed by the U.S. Fish and Wildlife Service [39]. The WMWR was within a humid subtropical climate characterized by mean annual precipitation of 87.8 cm (range = 64.9–152.8 cm) and a mean annual temperature of 15.3 °C (range = 6.9–17.7 °C); the wettest months were typically May (mean = 13.1 cm; range = 1.2–54.3 cm) and June (mean = 10.2 cm; range = 2.9–37.5 cm), whereas the hottest months were July (mean max temperature = 34.8 °C) and August (mean max temperature = 34.7 °C) [40]. The WMWR was characterized by elevations from 390 to 756 m, with critically imperiled southern mixed-grass prairie and deep soils common at lower elevations, Crosstimbers deciduous forests and shallower soils at moderate elevations, and rocky outcropping common across the higher-elevation granite mountains [39]. Mixed-grass prairies were dominated by little bluestem (Schizachyrium scoparium), big bluestem (Andropogon gerardii), Indian grass (Sorghastrum nutans), switchgrass (Panicum virgatum), and gramas (Bouteloua spp.), whereas post oak (Quercas stellata), blackjack oak (Q. marilandica), and eastern redcedar (Juniperus virginiana) dominated the Crosstimbers [39]. Natural and prescribed fires occurred regularly within the WMWR, with prescribed fires used to mimic natural fire regimes and promote historical landcover conditions [39].

2.2. Field Surveys and Image Processing

We conducted camera-based sampling from May to August 2023. We gridded the WMWR into 72.25-ha (850 m × 850 m) cells and randomly selected 100 cells (hereafter, sites) for sampling. The size of cells used to select sites was initially established for investigating wild pig (Sus scrofa × domesticus) space use [41]. The spacing between sites was sufficient to ensure independence among sites for nine-banded armadillos (home-range size = 2–20 ha [42]) but was smaller than the expected home-range size of coyotes (550–690 ha [43,44]). Consequently, we assumed closure was met for nine-banded armadillos during our sampling period and interpreted estimates of occupancy as the probability that a nine-banded armadillo occupied the site. In contrast, we may have violated the closure assumption for coyotes and, therefore, interpreted occupancy estimates as the probability of use (i.e., space use [45,46]). We deployed one motion-triggered Bushnell TrophyCam (Bushnell Corporation, Overland Park, KS, USA) camera at the center of each site. We positioned each camera ~0.5 m high and parallel to the ground on a tree or t-post [47]. We programmed cameras to operate for the full (24-h) diel cycle, capture a 3-image burst when triggered, and have a 10-s resting period between triggers. We deployed cameras for ≥28 days at each site. At the time of deployment, we trimmed vegetation within 3 m of the camera to minimize false triggers caused by wind-blown vegetation [48]. We did not use any visual or olfactory attractants (e.g., lure or bait).
Following field-based data collection, we used Timelapse2 version 2.2.3.0 [49] to classify images to species with a single-observer approach (BPM, i.e., single review), which has high accuracy when conducted by an experienced observer [50]. We focused subsequent data processing and analyses on nine-banded armadillos and coyotes. We defined two sequential images of the same species (at the same site) as independent detections if they were separated by ≥30 min [35,51]. To reduce biases that may be introduced by temporal autocorrelation in camera-based detections [52], we defined each survey for occupancy modeling as one week (i.e., a 7-day period). For each species, we then used the start of each independent detection to generate site-specific encounter histories indicating if the species was detected (1) or not (0) during each survey.

2.3. Detection and Occupancy Covariates

We identified covariates expected to influence the detection of each species. Armadillos have limited homeothermic control relative to most mammals, but temperate armadillos maintain higher body temperatures than tropical species [53,54]. Nine-banded armadillo activity has been positively associated with temperature and negatively associated with precipitation [38,55], and we predicted similar patterns in our system. Still, maximum temperatures reported at WMWR (i.e., 41.1 °C in July and August 2009 [40]) exceeded the putative upper lethal limit for nine-banded armadillos (40 °C [53]), so we hypothesized that there might be a maximum temperature above which activity decreases and considered a quadratic form of temperature as a covariate for detection. The effects of temperature and precipitation on coyote detection have varied among study systems. For example, camera-based detections of coyotes were not associated with temperature or precipitation in South Dakota [56], negatively associated with temperature but not associated with precipitation in Illinois [57], not associated with temperature but positively associated with precipitation in South Carolina [38], and negatively associated with temperature and precipitation in Nebraska [58]. Considering the greater temperatures in our study system relative to more northern studies, we predicted that coyote detection would be negatively associated with temperature and positively associated with precipitation. We obtained minimum daily temperature (°C) and total daily precipitation (mm) for our study area from the National Weather Service Advanced Hydrologic Prediction Service [40] and Oikolab [59]; we then used these data to calculate the mean daily minimum temperature and mean daily precipitation for each weekly survey period. Temporal variation in detection that is not accounted for by covariates can introduce bias into occupancy estimates [60]. Consequently, we included a covariate for day-of-year (DOY) to characterize the relative timing of each survey. Finally, we included a covariate for effort that represented the number of days within each weekly survey period that a camera was operational and accounted for decreased effort when a camera was inactive for a portion of a survey (e.g., due to camera malfunction or full memory cards).
We expected elevation and slope (i.e., two terrain variables) to be correlated, as higher elevations in the WMWR were typically characterized by steeper slopes, greater ruggedness, shallower soils, and more rocky outcroppings [39]. Consequently, we predicted that terrain would influence both species, with nine-banded armadillo occupancy being negatively associated with higher elevation and steeper slope due to limitations on foraging and burrowing [15], and coyote space use being negatively associated with elevation and slope due to their tendency to use areas with lower ruggedness [61]. We obtained elevation (decameter; dm) and slope (degrees) layers at a 30-m resolution from the 2023 LANDFIRE program [62] and characterized each site by calculating the mean elevation and mean slope within 425 m of each camera location.
Water availability has been identified as important for nine-banded armadillos [15] and coyotes [63]. Nine-banded armadillo occupancy decreased farther from water in South America [64]. Similarly, coyote occupancy was greater at sites closer to water [65], and activity at water sources increased with decreases in precipitation [66]. We predicted the occupancy of both species would be greater at sites closer to water. We used spatial layers provided by WMWR for natural and anthropogenic water sources and estimated the distance (hectometer; hm) between each camera and the nearest water source (i.e., distance to water).
Nine-banded armadillos have commonly been classified as habitat generalists, but they have tended to be associated with forest cover [64,67] and avoided developing burrows in open grasslands in Oklahoma [68]. Coyotes have also been considered habitat generalists and have used a range of open and closed habitat types across their broad distribution, e.g., [69,70,71]. We predicted that nine-banded armadillos would be positively associated with (or occur near) woodland or forest land cover, but coyotes would not be influenced by land cover within the WMWR. We obtained existing vegetation layers at a 30-m resolution from LANDFIRE [62] and characterized site-level landcover by calculating both (i) the proportion of cover > 1 m tall and (ii) the proportion of cover classified as forest within 425 m of each camera location. We also calculated the distance (hm) from each camera to the nearest cover (i.e., vegetation > 1-m tall). The relative abundance of nine-banded armadillos has been positively associated with fire heterogeneity [72], and coyotes have selected recently burned areas [73]. Consequently, we predicted that nine-banded armadillo occupancy and coyote space use would be greater in areas with recent burns. We used fire perimeter data provided by WMWR—which included both natural and prescribed fires—and categorized sites based on whether they were recently burned (i.e., ≤5 years prior to sampling) or not.
Species-specific responses to human activity are common and context-dependent [74]. If species respond to humans in opposing or different ways, failure to control or account for human activity could influence the reliability of inferences related to interspecific interactions. We indexed spatial variation in relative human activity using two metrics that we derived from a layer provided by the WMWR for linear human features (i.e., paved roads, unpaved roads, and hiking trails; hereafter, roads): road density within 425 m of each camera location (km/km2) and distance to the nearest road (hm). Nine-banded armadillos commonly use areas near humans and may use roads (or roadsides) for travel corridors or foraging areas [67,75,76]. Coyotes use roads for movement, but road density may influence use patterns, and resident animals may avoid roads [71,77,78]. We, therefore, predicted that roads would positively influence nine-banded armadillo occupancy but would negatively influence coyote space use. We processed all Geographic Information Systems layers with ArcGIS (version 10.7.1, ESRI, Redlands, CA, USA).

2.4. Occupancy Analyses

We conducted occupancy analyses using a structured approach [30]. We (i) first tested environmental covariates for correlations and refined global models for the detection and occupancy of each species. We tested pairwise correlations between continuous covariates using a Pearson correlation coefficient test in R v. 4.4.1 [79] and did not include covariates correlated at |r| > 0.5 in the same model [80]. As expected, covariates that represented alternative characterizations of the same general factor (e.g., roads characterized by road density versus distance to roads) were commonly correlated (refer to Results). For each species, we assessed relative support for correlated variables to establish global models for detection and occupancy by comparing models with Akaike’s Information Criterion with small sample size correction (AICc) [81]. We evaluated competing global models for detection (p) and occupancy (ψ) independently while maintaining the model for the other parameter constant at the intercept-only (i.e., null) model. For each species, we retained the most-supported covariates (or characterizations) for subsequent analyses.
We (ii) then conducted single-season, single-species occupancy analyses to evaluate the effects of environmental covariates on species-specific detection and occupancy (or space use) [30]. We developed a candidate model set for each species by considering all additive combinations of covariates within each submodel (i.e., for p and ψ) and all combinations of submodels [82]. We evaluated candidate models with AICc and evaluated the influence of predictors by considering the structure of the most-supported model, beta coefficients and associated 85% confidence intervals (CI), and relative predictor importance (i.e., cumulative Akaike weights across the entire model set) [81,83,84]. We excluded models that failed to converge. For each species-specific global model, we assessed goodness-of-fit with 1000 parametric bootstrap replicates on an χ2 statistic appropriate for binary data with the unmarked package in R [85,86,87].
Finally, we (iii) carried the most-supported model structures from each single-species analysis forward into a conditional two-species occupancy (hereafter, co-occurrence) analysis to assess interspecific interactions. We evaluated support for 12 a priori models of co-occurrence [31]. Co-occurrence models assume that one species is dominant (Species A) and influences the occurrence of a subordinate species (Species B) but that the subordinate species does not influence the occurrence of the dominant species. To this end, co-occurrence models include parameters for the occupancy of Species A (ψA) and the occupancy of Species B in the presence (ψBA) and absence (ψBa) of Species A, as well as parameters for detection of Species A (pA) and Species B (pB) in the absence of the other species, and detection of Species A (rA) and Species B (rB) in the presence of the other species; rB can be decomposed further into detection of Species B when Species A was present and detected (rBA) or not (rBa). We assumed that coyotes were dominant and could influence the occurrence of nine-banded armadillos, but that nine-banded armadillo presence had no influence on coyote space use. Co-occurrence models provide flexibility for investigating nuanced interspecific interactions that may manifest as variation in detection probabilities [31], and variation in detection is often attributed to differences in animal behavior or relative abundance among sites; we interpreted variation in nine-banded armadillo detection as reflecting variation in site-use intensity [32,33,34,35]. We evaluated candidate models with AICc and assessed relative support for competing interspecific patterns (e.g., Species A influencing the occurrence of Species B [ψBA ≠ ψBa] versus not influencing Species B [ψBA = ψBa]) by considering the cumulative weight of evidence. We performed all occupancy analyses using an information-theoretic approach with program MARK version 6.2 [88].

2.5. Diel Activity Analyses

We used camera-based detections to characterize diel activity patterns of each species and hypothesized that both nine-banded armadillos [36,37] and coyotes [38] would be primarily nocturnal with a high degree of overlap in diel activity patterns. Camera-based data provide information on diel variation in detection and, with sufficient sample sizes [89], can be used to reliably estimate diel activity patterns [90,91,92]. For each species, we restricted diel activity analyses to independent detections, generated kernel density estimates of diel activity, estimated the coefficient of overlap (∆4) between activity curves, and generated 95% confidence intervals for ∆4 with 1000 bootstraps with the overlap package in R [79,93]. We subsequently tested whether the diel activity patterns of nine-banded armadillos and coyotes were significantly different using two statistical tests. We implemented a non-parametric Kolmogorov–Smirnov Test to evaluate if diel activity distribution differed between species. Because activity curves represent circular data, we also tested whether diel activity patterns differed between species with a non-parametric Watson-Wheeler Test specifically designed for circular data with the circular package in R [94].

3. Results

3.1. Data Collection

We collected 1,108,262 images from 16 May to 8 August 2023 across 99 sites (one site was not sampled due to a camera malfunction). Sites were sampled for an average of 37.4 days (±2.8 SD).
We observed nine-banded armadillos in 180 independent detections at 36 sites (36.4%), whereas we observed coyotes in 119 independent detections at 47 sites (47.5%, including 19 sites at which nine-banded armadillos were detected).

3.2. Environmental Characteristics and Occupancy Patterns

Mean daily minimum temperatures ranged from 15.3 to 27.6 °C across sites and surveys, and mean daily precipitation ranged from 0.0 to 12.8 mm. Temperature was highly correlated with DOY (i.e., r > 0.9). For nine-banded armadillos, model selection results indicated that temperature was more supported than either DOY or temperature2 (Table S1). Similarly, model selection results indicated that temperature was more supported than DOY for coyotes (Table S2). Consequently, we retained the temperature for subsequent analyses of each species.
Elevation (mean = 533.5 m) and slope (mean = 7.5°) across sites were positively correlated (r = 0.54). Road density (mean = 1.42 km/km2) and distance to the nearest road (mean = 299 m) were negatively correlated (r = −0.60). The proportion of cover >1 m tall (mean = 22.0%) and the proportion of forest (mean = 22.2%) were strongly correlated (r > 0.99), suggesting that cover >1 m tall was predominantly forest cover; the proportion of cover >1-m tall and proportion of forest were both negatively correlated (r = −0.57) with distance to cover (mean = 98 m). Distance to water (mean = 661 m) was not correlated with any other spatial covariate. Among sites, 54 (54.5%) were within an area that had burned during the preceding 5 years. For nine-banded armadillos, model selection results indicated that slope was more supported than elevation, distance to cover was more supported than other landcover characterizations, and distance to the road was more supported than road density (Table S1). In contrast, model selection results for coyotes indicated elevation was more supported than slope, the proportion of forest cover was more supported than other landcover characterizations, and road density was more supported than the distance to the road (Table S2).
We did not find evidence for a lack of fit based on the χ2 statistic for nine-banded armadillos (p = 0.50) or coyotes (p = 0.08). For nine-banded armadillos, estimates from the most-supported model indicated that weekly detection probability was negatively associated with temperature (Table 1) and consequently decreased over the season from 0.52 (SE = 0.04, 95% = 0.44–0.61) to 0.40 (SE = 0.05, 95% CI = 0.30–0.50). The most-supported model indicated that nine-banded armadillo occupancy was positively associated with distance to water, negatively associated with slope and distance to cover, and greater in areas that have burned (Table 1, Figure 3). When considering the full model set, cumulative model weights indicated relatively strong support for the effects of distance to cover and slope, moderate support for the effect of burns, and weaker support for the effect of distance to water (Table 1). For coyotes, we did not find an influence of environmental covariates on detection but did find that detection was influenced by effort; at full effort, estimates of weekly detection probability from the most-supported model was 0.23 (SE = 0.03, 95% = 0.17–0.31). The most-supported model indicated coyote occupancy was positively associated with elevation and greater in areas that had recently burned (Table 1, Figure 3). When considering the full model set, cumulative model weights indicated moderate support for the effect of burns and weaker support for the effect of elevation (Table 1).
The most-supported co-occurrence model and patterns across the model set indicated that coyote presence did not influence nine-banded armadillo occupancy (i.e., ψBA = ψBa; cumulative weight = 0.99) but did influence nine-banded armadillo site-use intensity (i.e., pB ≠ rB; cumulative weight = 0.78; Table 2); nine-banded armadillo site-use intensity was significantly greater when coyotes were absent than when they were present (Figure 4). When all else was the same, models with the rBA = rBa structure (cumulative weight = 0.76) always outperformed those with the rBA ≠ rBa structure (Table 2), indicating that despite coyote presence influencing nine-banded armadillo site-use intensity, there was no evidence that nine-banded armadillo detection was affected by coyote detection during the same survey period. Finally, nine-banded armadillo presence did not influence coyote detection (cumulative weight = 0.91; Table 2).

3.3. Diel Activity Patterns

Diel activity analyses indicated moderate to high levels of overlap in the activity of nine-banded armadillos and coyotes (∆4 = 0.79, 95% CI = 0.71–0.87; Figure 5). Nine-banded armadillos were strictly nocturnal, whereas coyotes were predominantly nocturnal, with some activity during other periods (Figure 5). Despite seemingly moderate to high overlap in diel activity, the Kolmogorov–Smirnov Test indicated that the activity patterns of nine-banded armadillos and coyotes were significantly different (D = 0.194, p < 0.01). Similarly, Watson’s Two-Sample Test of Homogeneity indicated significant differences in the circular distributions of nine-banded armadillo and coyote diel activity (U2 = 0.428, p < 0.001).

4. Discussion

Space use and activity of mesopredators are often influenced by mesopredator suppression [1,4], but the role of antipredator defenses in mitigating top-down effects is poorly understood. Although physical (i.e., morphological) defenses have evolved to reduce predation risk in some insectivorous mesopredators [95,96], fear of predation may still drive spatial or temporal shifts in the activity of defended mesopredators [97]. We investigated patterns of co-occurrence and diel activity for sympatric coyotes and nine-banded armadillos as a model system to investigate the influence of top-down effects on a mesopredator with physical antipredator defenses. Consistent with our predictions, nine-banded armadillo occurrence was not influenced by coyotes, but the site-use intensity was suppressed by the presence of coyotes, suggesting that—despite their physical defenses—nine-banded armadillos experienced mesopredator suppression from coyotes. Nine-banded armadillo diel activity was strictly nocturnal, whereas diel activity of coyotes was predominantly nocturnal, suggesting nine-banded armadillos did not avoid coyotes temporally.
Contrasting natural histories between sympatric species may lead to species-specific selection for divergent habitat conditions, opposing effects of environmental conditions, or both. For example, if high temperatures suppress the activity of one species while facilitating the activity of another, failure to account for the effect of temperature could lead to erroneous conclusions or suggestions that the activity of one species is influenced by the other. Similarly, two species that select different landcover types independent of one another may appear to avoid one another if appropriate landcover attributes are not considered [31]. Thus, disentangling interspecific interactions from the environmental factors influencing a species is complex. Co-occurrence models offer a framework to examine interspecific interactions while explicitly accounting for imperfect detection and environmental covariates predicted to influence species [31].
Coyote space-use patterns supported some, but not all, of our predictions. In contrast to our predictions, coyote space use was positively associated with elevation and we found no effect of landcover. The Wichita Mountains occur at a distinctively higher elevation than surrounding areas [39], and the use of higher elevations by coyotes during summer may relate to selection for areas with greater thermal relief (e.g., cooler temperatures and thermal cover). In our study area, elevation alone would be expected to influence temperatures by up to ~3 °C. Nonetheless, our sites were distributed within a narrower range of elevations (432–678 m) than available at the WMWR—with only two sites located in the highest 25% of elevations—and this may have limited the strength of effect observed for coyotes. Across sites, the distance to water was within the daily movement capacity of coyotes in Oklahoma (6.0–6.3 km [98]), and we found no effect of water availability on their space use. Despite substantial human activity within the WMWR (~1.5 million visitors annually over a 10-year period [39]), we did not find any evidence that road density (a proxy for spatial variation in human activity) influenced coyote space use. Coyote harvest was not permitted within the WMWR during our sampling period, and this may have limited the influence of human activity.
Nine-banded armadillo occupancy patterns also supported some of our predictions. Previous research indicated that nine-banded armadillos selected sloped areas for establishing burrows [17,68], and our results indicated occupancy was greater in areas with more slope. Also, in agreement with our predictions, nine-banded armadillo occupancy was greater in areas close to cover. Nine-banded armadillo occupancy has been associated with forest cover across a wide range of environmental conditions [64,67]. In our system, cover >1 m tall was highly correlated with forest cover, and greater nine-banded armadillo occupancy in areas with tall cover may relate to greater ground litter and availability of prey (i.e., soil invertebrates), more stable thermal conditions, or both [64]. In Oklahoma, nine-banded armadillos did not establish burrows in grasslands [68], but they have been documented moving from areas with dense cover to grasslands to forage [15]. Despite being correlated, distance to nearest cover was more supported than the proportion of cover or forest, suggesting that a mix of forested and open landcover types may have been important to nine-banded armadillo occupancy. The reported effect of distance to water on nine-banded armadillo presence has varied: nine-banded armadillo occupancy was greater near water in Brazil [64], but the distance to water had “little effect” on the presence of nine-banded armadillos in their northern range extent [76]. In contrast to these patterns and our prediction, nine-banded armadillo occupancy in the WMWR was greater in areas farther from water. Although historical work suggests free-standing water may be important to nine-banded armadillos [15], the relatively low metabolic rate of nine-banded armadillos may promote water conservation [99]. High water economy, combined with ephemeral water available through periodic rainfall, may have allowed nine-banded armadillos to select for areas away from permanent water sources, which may have relatively high predator activity [66] and low prey availability (i.e., soil macroinvertebrates) due to soil compaction by large-bodied ungulates in our study system (e.g., American bison [Bison bison] and cattle [Bos taurus taurus]) [100,101].
Fire history was the only variable that had a similar influence on both species, with coyote space use and nine-banded armadillo occupancy being greater at sites that were recently burned. Increased use and occupancy of recently burned areas supported our predictions. In the WMWR, prescribed fire has been employed as a habitat management tool [39] and has been shown to influence vegetation structure, litter, and the abundance of prey (e.g., small mammals and soil invertebrates [102,103,104]) important to mesopredators, and these changes may have driven increased use of recently burned areas by nine-banded armadillos and coyotes.
Mesopredators must secure sufficient resources while avoiding predation, but species with effective physical defenses may be less likely to shift their spatial or temporal distributions to avoid predators. If physical defenses do not eliminate predation risk, they may still sufficiently deter predation. For example, due to the extensive handling time to bypass protective scales, predators often do not attempt to kill pangolins [11]. Similarly, dorsal armor—as adorned by nine-banded armadillos—is heavy and burdensome and may deter predation attempts by larger carnivores [95]. Thus, species with physical defenses may be able to rely on their defenses, rather than crypsis or vigilance, to avoid predation [96]. Consistent with our predictions, occupancy of nine-banded armadillos was not influenced by the presence of coyotes, and diel activity of nine-banded armadillos had a high overlap with peak coyote activity.
Even when mesopredator occupancy is independent of the presence of a dominant predator, suppression of site-use intensity by mesopredators may be driven by an apex predator [4]. In our study, the probability of detection for nine-banded armadillos was consistently reduced at sites where coyotes were present compared to those where coyotes were absent. Variation in detection has commonly been interpreted as reflecting variation in site-use intensity [32,33,34,35]. Differences in site-use intensity can be driven by shifts in the frequency of use by individuals (e.g., fear-induced shifts in behavioral patterns) or divergence in local abundance (i.e., abundance-induced heterogeneity in detection [105]). Our sampling approach does not allow us to disentangle these potential explanations. Predation attempts on nine-banded armadillos (even if unsuccessful) could lead to fear-induced shifts in site-use intensity by individual nine-banded armadillos. Despite the deterrent properties of armor, some individuals may still be susceptible to predation. Juvenile nine-banded armadillos with less-developed carapaces tend to suffer greater predation-related mortality than adults [28]. Thus, site-use intensity could also potentially be suppressed by coyotes, limiting the recruitment of nine-banded armadillos into the adult class.
Understanding the role of interspecific interactions on nine-banded armadillos has important implications for conservation and management. Nine-banded armadillos are commonly considered a nest predator of ground-nesting species of conservation concern (e.g., northern bobwhites [Colinus virginianus]) and a nuisance due to human-wildlife conflicts (e.g., damage to crops and landscaping) [16]. Elucidating the effects of coyotes on nine-banded armadillos may offer insights into predator management strategies. For example, nine-banded armadillos accounted for substantially more northern bobwhite nest depredations than coyotes in the southeastern United States [106]; thus, if the management objective is to decrease nest depredations, our data indicate that coyote control may lead to increased site-use intensity by nine-banded armadillos (i.e., mesopredator release) and be counterproductive to the objective. Nine-banded armadillos are expected to continue their range expansion in North America, but previous efforts to understand their potential range have relied primarily on environmental conditions [19,20,23]. Our study builds on recent efforts to better understand how sympatric species—e.g., invasive wild pigs (Sus scrofa × domesticus) [35] and domestic dogs (Canis familiaris) [36]—influence nine-banded armadillos, and collectively, these studies provide greater insight into the factors that may promote or limit nine-banded armadillo spatial ecology.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/d17040290/s1, Table S1. Model selection results from single-species detection and occupancy analyses evaluating relative support for combinations of correlated variables to establish global models for nine-banded armadillos (Dasypus novemcinctus). Table S2. Model selection results from single-species detection and occupancy analyses evaluating relative support for combinations of correlated variables to establish global models for coyotes (Canis latrans).

Author Contributions

Conceptualization, R.C.L. and B.P.M.; Methodology, R.C.L. and B.P.M.; Formal Analysis, R.C.L. and K.M.W. Investigation, R.C.L., B.P.M., and K.M.W.; Resources, R.C.L., D.T.M., and C.E.F.; Data Curation, R.C.L. and B.P.M.; Writing—Original Draft Preparation, R.C.L.; Writing—Review and Editing, B.P.M., D.T.M., C.E.F., and K.M.W.; Visualization, R.C.L. and K.M.W.; Supervision, R.C.L.; Project Administration, R.C.L.; Funding Acquisition, R.C.L. All authors contributed critically to manuscript drafts and gave final approval for publication. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the U.S. Geological Survey via grant number G21AC10442.

Institutional Review Board Statement

The study was conducted in accordance with guidelines endorsed by the American Society of Mammalogists [107] and approved by the Institutional Animal Care and Use Committee of Oklahoma State University (protocol number: IACUC-22-18; approved: 10 April 2022).

Data Availability Statement

The original data presented in the study is openly available as a U.S. Geological Survey data release [108] https://doi.org/10.5066/P1GCXNVB.

Acknowledgments

Logistical support was provided by the Oklahoma Cooperative Fish and Wildlife Research Unit, Oklahoma State University, and the Wichita Mountains Wildlife Refuge. We thank Ronie Loffelmacher and Cody Conroy for their assistance with data collection. The Oklahoma Cooperative Fish and Wildlife Research Unit is supported by the Oklahoma Department of Wildlife Conservation, Oklahoma State University, U.S. Geological Survey, U.S. Fish and Wildlife Service, and Wildlife Management Institute. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

Conflicts of Interest

The authors declare no conflicts of interest or competing interests.

References

  1. Crooks, K.R.; Soulé, M.E. Mesopredator Release and Avifaunal Extinctions in a Fragmented System. Nature 1999, 400, 563–566. [Google Scholar] [CrossRef]
  2. Ritchie, E.G.; Johnson, C.N. Predator Interactions, Mesopredator Release and Biodiversity Conservation. Ecol. Lett. 2009, 12, 982–998. [Google Scholar] [CrossRef] [PubMed]
  3. Vance-Chalcraft, H.D.; Rosenheim, J.A.; Vonesh, J.R.; Osenberg, C.W.; Sih, A. The Influence of Intraguild Predation on Prey Suppression and Prey Release: A Meta-Analysis. Ecology 2007, 88, 2689–2696. [Google Scholar] [CrossRef] [PubMed]
  4. Gordon, C.E.; Feit, A.; Gru, J.; Letnic, M. Mesopredator Suppression by an Apex Predator Alleviates the Risk of Predation Perceived by Small Prey. Proc. R. Soc. B 2015, 282, 20142870. [Google Scholar] [CrossRef]
  5. Palomares, F.; Caro, T.M. Interspecific Killing among Mammalian Carnivores. Am. Nat. 1999, 153, 492–508. [Google Scholar] [CrossRef]
  6. Lourenço, R.; Penteriani, V.; Rabaça, J.E.; Korpimäki, E. Lethal Interactions among Vertebrate Top Predators: A Review of Concepts, Assumptions and Terminology. Biol. Rev. 2014, 89, 270–283. [Google Scholar] [CrossRef]
  7. Newman, C.; Buesching, C.D.; Wolff, J.O. The Function of Facial Masks in “Midguild” Carnivores. Oikos 2005, 108, 623–633. [Google Scholar] [CrossRef]
  8. Stankowich, T.; Campbell, L.A. Living in the Danger Zone: Exposure to Predators and the Evolution of Spines and Body Armor in Mammals. Evolution 2016, 70, 1501–1511. [Google Scholar] [CrossRef]
  9. Walker, H.; Caro, T.; Bell, D.; Ferguson, A.; Stankowich, T. Predation Risk Drives Aposematic Signal Conformity. Evolution 2023, 77, 2492–2503. [Google Scholar] [CrossRef]
  10. Fay, C.; Young, J.K.; Stankowich, T. Aposematic Learning in a Mammalian Predator–Prey System. Anim. Behav. 2024, 212, 39–48. [Google Scholar] [CrossRef]
  11. Peplinski, J.; Malone, M.A.; Fowler, K.J.; Potratz, E.J.; Pergams, A.G.; Charmoy, K.L.; Rasheed, K.; Avdieiev, S.S.; Whelan, C.J.; Brown, J.S. Ecology of Fear: Spines, Armor and Noxious Chemicals Deter Predators in Cancer and in Nature. Front. Ecol. Evol. 2021, 9, 682504. [Google Scholar] [CrossRef]
  12. Hunter, J.S. Familiarity Breeds Contempt: Effects of Striped Skunk Color, Shape, and Abundance on Wild Carnivore Behavior. Behav. Ecol. 2009, 20, 1315–1322. [Google Scholar] [CrossRef]
  13. Prange, S.; Gehrt, S.D. Response of Skunks to a Simulated Increase in Coyote Activity. J. Mammal. 2007, 88, 1040–1049. [Google Scholar] [CrossRef]
  14. Wang, Y.; Allen, M.L.; Wilmers, C.C. Mesopredator Spatial and Temporal Responses to Large Predators and Human Development in the Santa Cruz Mountains of California. Biol. Conserv. 2015, 190, 23–33. [Google Scholar] [CrossRef]
  15. McBee, K.; Baker, R.J. Dasypus novemcinctus . Mamm. Species 1982, 162, 1–9. [Google Scholar] [CrossRef]
  16. Ober, H.K.; Degroote, L.W.; McDonough, C.M.; Mizell, R.F.; Mankin, R.W. Identification of an Attractant for the Nine-Banded Armadillo, Dasypus novemcinctus. Wildl. Soc. Bull. 2011, 35, 421–429. [Google Scholar] [CrossRef]
  17. Sawyer, C.F.; Brinkman, D.C.; Walker, V.D.; Covington, T.D.; Stienstraw, E.A. The Zoogeomorphic Characteristics of Burrows and Burrowing by Nine-Banded Armadillos (Dasypus novemcinctus). Geomorphology 2012, 157–158, 122–130. [Google Scholar] [CrossRef]
  18. DeGregorio, B.A.; Veon, J.T.; Massey, A. Wildlife Associates of Nine-Banded Armadillo (Dasypus novemcinctus) Burrows in Arkansas. Ecol. Evol. 2022, 12, e8858. [Google Scholar] [CrossRef]
  19. Taulman, J.F.; Robbins, L.W. Recent Range Expansion and Distributional Limits of the Nine-Banded Armadillo (Dasypus novemcinctus) in the United States. J. Biogeogr. 1996, 23, 635–648. [Google Scholar] [CrossRef]
  20. Taulman, J.F.; Robbins, L.W. Range Expansion and Distributional Limits of the Nine-Banded Armadillo in the United States: An Update of Taulman & Robbins (1996). J. Biogeogr. 2014, 41, 1626–1630. [Google Scholar] [CrossRef]
  21. Davidson, A.D.; Detling, J.K.; Brown, J.H. Ecological Roles and Conservation Challenges of Social, Burrowing, Herbivorous Mammals in the World’s Grasslands. Front. Ecol. Environ. 2012, 10, 477–486. [Google Scholar] [CrossRef]
  22. Lamb, B.D.; Anderson, C.D.; McDonough, C.M.; Lockhart, J.M.; Butler, Z.P. A Comparison of Vertebrate Associates of Gopher Tortoise and Nine-Banded Armadillo Burrows in South Georgia. Chelonian Conserv. Biol. 2023, 22, 184–196. [Google Scholar] [CrossRef]
  23. Feng, X.; Papeş, M. Ecological Niche Modelling Confirms Potential North-East Range Expansion of the Nine-Banded Armadillo (Dasypus novemcinctus) in the USA. J. Biogeogr. 2015, 42, 803–807. [Google Scholar] [CrossRef]
  24. Gompper, M.E. Top Carnivores in the Suburbs? Ecological and Conservation Issues Raised by Colonization of North-eastern North America by Coyotes. Bioscience 2002, 52, 185. [Google Scholar] [CrossRef]
  25. Best, T.L.; Hoditschek, B.; Thomas, H.H. Foods of Coyotes (Canis latrans) in Oklahoma. Southwest. Nat. 1981, 26, 67–69. [Google Scholar] [CrossRef]
  26. Cherry, M.J.; Turner, K.L.; Howze, M.B.; Cohen, B.S.; Conner, L.M.; Warren, R.J. Coyote Diets in a Longleaf Pine Ecosystem. Wildl. Biol. 2016, 22, 64–70. [Google Scholar] [CrossRef]
  27. Watine, L.N.; Giuliano, W.M. Factors Determining Coyote (Canis latrans) Diets. Open J. Ecol. 2017, 7, 650–666. [Google Scholar] [CrossRef]
  28. McDonough, C.M.; Loughry, W.J. Patterns of Mortality in a Population of Nine-Banded Armadillos, Dasypus novemcinctus. Am. Midl. Nat. 1997, 138, 299–305. [Google Scholar] [CrossRef]
  29. Rooney, B.; Kays, R.; Cove, M.V.; Jensen, A.; Goldstein, B.R.; Pate, C.; Castiblanco, P.; Abell, M.E.; Adley, J.; Agenbroad, B.; et al. Snapshot USA 2019–2023: The First Five Years of Data from a Coordinated Camera Trap Survey of the United States. Glob. Ecol. Biogeogr. 2025, 34, e13941. [Google Scholar] [CrossRef]
  30. MacKenzie, D.I.; Nichols, J.D.; Lachman, G.B.; Droege, S.; Royle, J.A.; Langtimm, C.A. Estimating Site Occupancy Rates When Detection Probabilities Are Less than One. Ecology 2002, 83, 2248–2255. [Google Scholar] [CrossRef]
  31. Richmond, O.M.W.; Hines, J.E.; Beissinger, S.R. Two-Species Occupancy Models: A New Parameterization Applied to Co-Occurrence of Secretive Rails. Ecol. Appl. 2010, 20, 2036–2046. [Google Scholar] [CrossRef] [PubMed]
  32. MacKenzie, D.I.; Nichols, J.D.; Royle, J.A.; Pollock, K.H.; Bailey, L.L.; Hines, J.E. Occupancy Estimation and Modeling: Inferring Patterns and Dynamics of Species Occurrence, 2nd ed.; Academic Press: Cambridge, MA, USA, 2018. [Google Scholar]
  33. Pasa, J.B.; Arrais, R.C.; Massara, R.L.; Pereira, G.; de Azevedo, F.C.C. Factors Influencing the Habitat Use by Ocelots in One of the Last Large Atlantic Forest Remnants in Southeastern Brazil. Ecol. Evol. 2021, 11, 4631–4643. [Google Scholar] [CrossRef] [PubMed]
  34. Otto, C.R.V.; Bailey, L.L.; Smart, A.H. Patch Utilization and Flower Visitations by Wild Bees in a Honey Bee-Dominated, Grassland Landscape. Ecol. Evol. 2021, 11, 14888–14904. [Google Scholar] [CrossRef] [PubMed]
  35. Broadway, M.S.; Todaro, H.M.; Koeck, M.M.; Dotterweich, C.N.; Cain, S.A.; Buehler, L.; Chitwood, M.C.; Lonsinger, R.C. Interspecific Effects of Invasive Wild Pigs (Sus scrofa) on Native Nine-banded Armadillos (Dasypus novemcinctus). J. Mammal. 2025, in press.
  36. Costa, R.T.; Fornitano, L.; de Cassia Bianchi, R. Nine-banded Armadillos Temporally Avoid Sites Visited by Domestic Dogs and Native Carnivores. Wildl. Res. 2024, 51, WR23047. [Google Scholar] [CrossRef]
  37. DeGregorio, B.A.; Gale, C.; Lassiter, E.V.; Massey, A.; Roberts, C.P.; T. Veon, J. Nine-Banded Armadillo (Dasypus novemcinctus) Activity Patterns Are Influenced by Human Activity. Ecol. Evol. 2021, 11, 15874–15881. [Google Scholar] [CrossRef]
  38. Saldo, E.A.; Jensen, A.J.; Muthersbaugh, M.S.; Butfiloski, J.W.; Cantrell, J.; Kilgo, J.C.; Ruth, C.; Yarrow, G.K.; Jachowski, D.S. Spatiotemporal Overlap with Invasive Wild Pigs (Sus scrofa) Varies by Species and Season in a Temperate Ecosystem. Ecosphere 2023, 14, e4500. [Google Scholar] [CrossRef]
  39. Wichita Mountains Wildlife Refuge [WMWR]. Wichita Mountains Wildlife Refuge Comprehensive Conservation Plan and Environmental Assessment; U.S. Fish and Wildlife Service: Idahome, OK, USA, 2013. [Google Scholar]
  40. National Oceanic and Atmospheric Administration [NOAA]. Climate Data Online. 2023. Available online: https://water.weather.gov/precip/download.php (accessed on 13 November 2023).
  41. Murley, B.P. Feral Swine Space Use and Effects on Ecological Communities in the Wichita Mountains Wildlife Refuge. Master’s Thesis, Oklahoma State University, Stillwater, OK, USA, 2024. [Google Scholar]
  42. Loughry, W.J.; McDonough, C.M. Spatial Patterns in a Population of Nine-Banded Armadillos (Dasypus novemcinctus). Am. Midl. Nat. 1998, 140, 161–169. [Google Scholar] [CrossRef]
  43. Gese, E.M.; Rongstad, O.J.; Mytton, W.R. Home Range and Habitat Use of Coyotes in Southeastern Colorado. J. Wildl. Manage. 1988, 52, 640–646. [Google Scholar] [CrossRef]
  44. Nelson, J.L.; Cypher, B.L.; Bjurlin, C.D.; Creel, S. Effects of Habitat on Competition between Kit Foxes and Coyotes. J. Wildl. Manage. 2007, 71, 1467–1475. [Google Scholar] [CrossRef]
  45. Rota, C.T.; Fletcher, R.J.; Dorazio, R.M.; Betts, M.G. Occupancy Estimation and the Closure Assumption. J. Appl. Ecol. 2009, 46, 1173–1181. [Google Scholar] [CrossRef]
  46. Gould, M.J.; Gould, W.R.; Cain, J.W.; Roemer, G.W. Validating the Performance of Occupancy Models for Estimating Habitat Use and Predicting the Distribution of Highly-Mobile Species: A Case Study Using the American Black Bear. Biol. Conserv. 2019, 234, 28–36. [Google Scholar] [CrossRef]
  47. Cove, M.V.; Kays, R.; Bontrager, H.; Bresnan, C.; Lasky, M.; Frerichs, T.; Klann, R.; Lee, T.E.; Crockett, S.C.; Crupi, A.P.; et al. Snapshot USA 2019: A Coordinated National Camera Trap Survey of the United States. Ecology 2021, 102, e03353. [Google Scholar] [CrossRef]
  48. Werdel, T.J.; Piper, C.W.; Ricketts, A.M.; Peek, M.S.; Ahlers, A.A. Scale-Specific Landscape Effects Impose Range-Limiting Constraints on the Distribution of a Prairie-Obligate Carnivore. Landsc. Ecol. 2022, 37, 2065–2079. [Google Scholar] [CrossRef]
  49. Greenberg, S.; Godin, T.; Whittington, J. Design Patterns for Wildlife-Related Camera Trap Image Analysis. Ecol. Evol. 2019, 9, 13706–13730. [Google Scholar] [CrossRef]
  50. Lonsinger, R.C.; Dart, M.M.; Larsen, R.T.; Knight, R.N. Efficacy of Machine Learning Image Classification for Automated Occupancy-Based Monitoring. Remote Sens. Ecol. Conserv. 2024, 10, 56–71. [Google Scholar] [CrossRef]
  51. Iannarilli, F.; Erb, J.; Arnold, T.W.; Fieberg, J.R. Evaluating Species-Specific Responses to Camera-Trap Survey Designs. Wildl. Biol. 2021, 2021, 1–12. [Google Scholar] [CrossRef]
  52. Goldstein, B.R.; Jensen, A.J.; Kays, R.; Cove, M.V.; McShea, W.J.; Rooney, B.; Kierepka, E.M.; Pacifici, K. Guidelines for Estimating Occupancy from Autocorrelated Camera Trap Detections. Methods Ecol. Evol. 2024, 15, 1177–1191. [Google Scholar] [CrossRef]
  53. Johansen, K. Temperature Regulation in the Nine-Banded Armadillo (Dasypus novemcinctus mexicanus). Physiol. Zool. 1961, 34, 126–144. [Google Scholar] [CrossRef]
  54. McNab, B.K. Energetics and the Limits to a Temperate Distribution in Armadillos. J. Mammal. 1980, 61, 606–627. [Google Scholar] [CrossRef]
  55. McDonough, C.M.; Loughry, W.J. Influences on Activity Patterns in a Population of Nine-Banded Armadillos. J. Mammal. 1997, 78, 932–941. [Google Scholar] [CrossRef]
  56. Dart, M.M.; Perkins, L.B.; Jenks, J.A.; Hatfield, G.; Lonsinger, R.C. The Effect of Scent Lures on Detection Is Not Equitable among Sympatric Species. Wildl. Res. 2023, 50, 190–200. [Google Scholar] [CrossRef]
  57. Lesmeister, D.B.; Nielsen, C.K.; Schauber, E.M.; Hellgren, E.C. Spatial and Temporal Structure of a Mesocarnivore Guild in Midwestern North America. Wildl. Monogr. 2015, 191, 1–61. [Google Scholar] [CrossRef]
  58. Madsen, A.E.; Corral, L.; Fontaine, J.J. Weather and Exposure Period Affect Coyote Detection at Camera Traps. Wildl. Soc. Bull. 2020, 44, 342–350. [Google Scholar] [CrossRef]
  59. Oikolab. Weather and Climate Data for Analysts. 2023. Available online: https://oikolab.com (accessed on 13 November 2023).
  60. Sollmann, R. Mt or Not Mt: Temporal Variation in Detection Probability in Spatial Capture-Recapture and Occupancy Models. Peer Community J. 2024, 4, e1. [Google Scholar] [CrossRef]
  61. Bender, L.C.; Rosas-Rosas, O.C.; Weisenberger, M.E. Seasonal Occupancy of Sympatric Larger Carnivores in the Southern San Andres Mountains, South-Central New Mexico, USA. Mammal Res. 2017, 62, 323–329. [Google Scholar] [CrossRef]
  62. LANDFIRE. LANDFIRE 2.2.0. U.S. Department of the Interior, Geological Survey, and U.S. Department of Agriculture. 2023. Available online: http://www.landfire.gov (accessed on 13 November 2023).
  63. Golightly, R.T.; Ohmart, R.D. Water Economy of Two Desert Canids: Coyote and Kit Fox. J. Mammal. 1984, 65, 51–58. [Google Scholar] [CrossRef]
  64. Rodrigues, T.F.; Chiarello, A.G. Native Forests within and Outside Protected Areas Are Key for Nine-Banded Armadillo (Dasypus novemcinctus) Occupancy in Agricultural Landscapes. Agric. Ecosyst. Environ. 2018, 266, 133–141. [Google Scholar] [CrossRef]
  65. Lombardi, J.V.; Comer, C.E.; Scognamillo, D.G.; Conway, W.C. Coyote, Fox, and Bobcat Response to Anthropogenic and Natural Landscape Features in a Small Urban Area. Urban Ecosyst. 2017, 20, 1239–1248. [Google Scholar] [CrossRef]
  66. Atwood, T.C.; Fry, T.L.; Leland, B.R. Partitioning of Anthropogenic Watering Sites by Desert Carnivores. J. Wildl. Manage. 2011, 75, 1609–1615. [Google Scholar] [CrossRef]
  67. DeGregorio, B.A.; McElroy, M.R.; Johansson, E.P. Occupancy and Activity Patterns of Nine-Banded Armadillos (Dasypus novemcinctus) in a Suburban Environment. Diversity 2023, 15, 907. [Google Scholar] [CrossRef]
  68. Zimmerman, J.W. Burrow Characteristics of the Nine-Banded Armadillo, Dasypus novemcinctus. Southwest. Nat. 1990, 35, 226–227. [Google Scholar] [CrossRef]
  69. Koehler, G.M.; Hornocker, M.G. Seasonal Resource Use among Mountain Lions, Bobcats, and Coyotes. J. Mammal. 1991, 72, 391–396. [Google Scholar] [CrossRef]
  70. Cherry, M.J.; Howell, P.E.; Seagraves, C.D.; Warren, R.J.; Conner, L.M. Effects of Land Cover on Coyote Abundance. Wildl. Res. 2016, 43, 662–670. [Google Scholar] [CrossRef]
  71. Lonsinger, R.C.; Gese, E.M.; Bailey, L.L.; Waits, L.P. The Roles of Habitat and Intraguild Predation by Coyotes on the Spatial Dynamics of Kit Foxes. Ecosphere 2017, 8, e01749. [Google Scholar] [CrossRef]
  72. Jorge, M.H.; Garrison, E.P.; Conner, L.M.; Cherry, M.J. Fire and Land Cover Drive Predator Abundances in a Pyric Landscape. For. Ecol. Manage. 2020, 461, 117939. [Google Scholar] [CrossRef]
  73. Stevenson, E.R.; Lashley, M.A.; Chitwood, M.C.; Garabedian, J.E.; Swingen, M.B.; Deperno, C.S.; Moorman, C.E. Resource Selection by Coyotes (Canis latrans) in a Longleaf Pine (Pinus palustris) Ecosystem: Effects of Anthropogenic Fires and Landscape Features. Can. J. Zool. 2019, 97, 165–171. [Google Scholar] [CrossRef]
  74. Burton, A.C.; Beirne, C.; Gaynor, K.M.; Sun, C.; Granados, A.; Allen, M.L.; Alston, J.M.; Alvarenga, G.C.; Álvarez Calderón, F.S.; Amir, Z.; et al. Mammal Responses to Global Changes in Human Activity Vary by Trophic Group and Landscape. Nat. Ecol. Evol. 2024, 8, 924–935. [Google Scholar] [CrossRef]
  75. Trovati, R.G. Differentiation and Characterization of Burrows of Two Species of Armadillos in the Brazilian Cerrado. Rev. Chil. Hist. Nat. 2015, 88, 19. [Google Scholar] [CrossRef]
  76. Haywood, C.J.; Nielsen, C.K.; Jiménez, F.A. Potential Distribution of Colonizing Nine-Banded Armadillos at Their Northern Range Edge. Diversity 2021, 13, 266. [Google Scholar] [CrossRef]
  77. Hinton, J.W.; Van Manen, F.T.; Chamberlain, M.J. Space Use and Habitat Selection by Resident and Transient Coyotes (Canis latrans). PLoS ONE 2015, 10, e0132203. [Google Scholar] [CrossRef] [PubMed]
  78. Lombardi, J.V.; Mackenzie, D.I.; Tewes, M.E.; Perotto-Baldivieso, H.L.; Mata, J.M.; Campbell, T.A. Co-Occurrence of Bobcats, Coyotes, and Ocelots in Texas. Ecol. Evol. 2020, 10, 4903–4917. [Google Scholar] [CrossRef] [PubMed]
  79. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2024. [Google Scholar]
  80. Dormann, C.F.; Elith, J.; Bacher, S.; Buchmann, C.; Carl, G.; Carré, G.; Marquéz, J.R.G.; Gruber, B.; Lafourcade, B.; Leitão, P.J.; et al. Collinearity: A Review of Methods to Deal with It and a Simulation Study Evaluating Their Performance. Ecography 2013, 36, 27–46. [Google Scholar] [CrossRef]
  81. Burnham, K.P.; Anderson, D.R. Multimodel Inference: Understanding AIC and BIC in Model Selection. Sociol. Methods Res. 2004, 33, 261–304. [Google Scholar] [CrossRef]
  82. Doherty, P.F.; White, G.C.; Burnham, K.P. Comparison of Model Building and Selection Strategies. J. Ornithol. 2012, 152, 317–323. [Google Scholar] [CrossRef]
  83. Arnold, T.W. Uninformative Parameters and Model Selection Using Akaike’s Information Criterion. J. Wildl. Manage. 2010, 74, 1175–1178. [Google Scholar] [CrossRef]
  84. Sutherland, C.; Hare, D.; Johnson, P.J.; Linden, D.W.; Montgomery, R.A.; Droge, E. Practical Advice on Variable Selection and Reporting Using Akaike Information Criterion. Proc. R. Soc. B Biol. Sci. 2023, 290. [Google Scholar] [CrossRef]
  85. Fiske, I.; Chandler, R. Overview of Unmarked: An R Package for the Analysis of Data From Unmarked Animals. 2015. Available online: https://cran.r-project.org/web/packages/unmarked/index.html (accessed on 1 October 2024).
  86. Steenweg, R.; Whittington, J.; Hebblewhite, M.; Forshner, A.; Johnston, B.; Petersen, D.; Shepherd, B.; Lukacs, P.M. Camera-Based Occupancy Monitoring at Large Scales: Power to Detect Trends in Grizzly Bears across the Canadian Rockies. Biol. Conserv. 2016, 201, 192–200. [Google Scholar] [CrossRef]
  87. Lonsinger, R.C.; Knight, R.N.; Waits, L.P. Detection Criteria and Post-Field Sample Processing Influence Results and Cost Efficiency of Occupancy-Based Monitoring. Ecol. Appl. 2021, 31, e02404. [Google Scholar] [CrossRef]
  88. White, G.C.; Burnham, K.P. Program MARK: Survival Estimation from Populations of Marked Animals. Bird Study 1999, 46, S120–S139. [Google Scholar] [CrossRef]
  89. Lashley, M.A.; Cove, M.V.; Chitwood, M.C.; Penido, G.; Gardner, B.; DePerno, C.S.; Moorman, C.E. Estimating Wildlife Activity Curves: Comparison of Methods and Sample Size. Sci. Rep. 2018, 8, 4173. [Google Scholar] [CrossRef] [PubMed]
  90. Ridout, M.S.; Linkie, M. Estimating Overlap of Daily Activity Patterns from Camera Trap Data. J. Agric. Biol. Environ. Stat. 2009, 14, 322–337. [Google Scholar] [CrossRef]
  91. Rowcliffe, J.M.; Kays, R.; Kranstauber, B.; Carbone, C.; Jansen, P.A. Quantifying Levels of Animal Activity Using Camera Trap Data. Methods Ecol. Evol. 2014, 5, 1170–1179. [Google Scholar] [CrossRef]
  92. Niedballa, J.; Sollmann, R.; Courtiol, A.; Wilting, A. camtrapR: An R Package for Efficient Camera Trap Data Management. Methods Ecol. Evol. 2016, 7, 1457–1462. [Google Scholar] [CrossRef]
  93. Meredith, M.; Ridout, M.S. Overview of the Overlap Package. Central R Archive Network Online. 2014. Available online: https://kar.kent.ac.uk/id/eprint/41474 (accessed on 5 December 2024).
  94. Agostinelli, C.; Lund, U. R Package ‘Circular’. Circular Statistics (Version 0.4-7). 2013. Available online: https://cran.r-project.org/web/packages/circular/index.html (accessed on 5 December 2024).
  95. Stankowich, T. Armed and Dangerous: Predicting the Presence and Function of Defensive Weaponry in Mammals. Adapt. Behav. 2012, 20, 32–43. [Google Scholar] [CrossRef]
  96. Stankowich, T.; Romero, A.N. The Correlated Evolution of Antipredator Defences and Brain Size in Mammals. Proc. R. Soc. B Biol. Sci. 2017, 284, 20161857. [Google Scholar] [CrossRef]
  97. Haswell, P.M.; Jones, K.A.; Kusak, J.; Hayward, M.W. Fear, Foraging and Olfaction: How Mesopredators Avoid Costly Interactions with Apex Predators. Oecologia 2018, 187, 573–583. [Google Scholar] [CrossRef]
  98. Litvaitis, J.A.; Shaw, J.H. Coyote Movements, Habitat Use, and Food Habits in Southwestern Oklahoma. J. Wildl. Manage. 1980, 44, 62–68. [Google Scholar] [CrossRef]
  99. McNab, B.K. The Influence of Food Habits on the Energetics of Eutherian Mammals. Ecol. Monogr. 1986, 56, 1–19. [Google Scholar] [CrossRef]
  100. Sulc, R.M.; Tracy, B.F. Integrated Crop-Livestock Systems in the U.S. Corn Belt. Agron. J. 2007, 99, 335–345. [Google Scholar] [CrossRef]
  101. Wehr, N.H.; Litton, C.M.; Lincoln, N.K.; Hess, S.C. Relationships between Soil Macroinvertebrates and Nonnative Feral Pigs (Sus scrofa) in Hawaiian Tropical Montane Wet Forests. Biol. Invasions 2020, 22, 577–586. [Google Scholar] [CrossRef]
  102. Engle, D.M.; Fuhlendorf, S.D.; Roper, A.; Leslie, D.M. Invertebrate Community Response to a Shifting Mosaic of Habitat. Rangel. Ecol. Manag. 2008, 61, 55–62. [Google Scholar] [CrossRef]
  103. Litt, A.R.; Steidl, R.J. Interactive Effects of Fire and Nonnative Plants on Small Mammals in Grasslands. Wildl. Monogr. 2011, 176, 1–31. [Google Scholar] [CrossRef]
  104. Butler, A.; Davis, C.A.; Fuhlendorf, S.D.; Wilder, S.M. Effects of Fire on Ground-Dwelling Arthropods in a Shrub-Dominated Grassland. Ecol. Evol. 2021, 11, 427–442. [Google Scholar] [CrossRef]
  105. Royle, J.A.; Nichols, J.D. Estimating Abundance from Repeated Presence–Absence Data or Point Counts. Ecology 2003, 84, 777–790. [Google Scholar] [CrossRef]
  106. Staller, E.L.; Palmer, W.E.; Carroll, J.P.; Thornton, R.P.; Sisson, D.C. Identifying Predators at Northern Bobwhite Nests. J. Wildl. Manage. 2005, 69, 124–132. [Google Scholar] [CrossRef]
  107. Sikes, R.S.; Animal Care and Use Committee of the American Society of Mammalogists. 2016 Guidelines of the American Society of Mammalogists for the Use of Wild Mammals in Research and Education. J. Mammal. 2016, 97, 663–688. [Google Scholar] [CrossRef]
  108. Lonsinger, R.C.; Murley, B.P. Nine-Banded Armadillo and Coyote Detection Data and Site-Specific Data from the Wichita Mountains Wildlife Refuge During Summer 2023; U.S. Geological Survey: Reston, VA, USA, 2025. [Google Scholar] [CrossRef]
Diversity 17 00290 g001
Figure 1. Image of a coyote (Canis latrans) attempting to capture a nine-banded armadillo (Dasypus novemcinctus) in the Wichita Mountains Wildlife Refuge, OK, USA (image collected during Snapshot USA 2022 sampling [29]).
Figure 1. Image of a coyote (Canis latrans) attempting to capture a nine-banded armadillo (Dasypus novemcinctus) in the Wichita Mountains Wildlife Refuge, OK, USA (image collected during Snapshot USA 2022 sampling [29]).
Diversity 17 00290 g001
Diversity 17 00290 g002
Figure 2. Location of 99 sites (Cameras) surveyed with motion-activated cameras for nine-banded armadillos (Dasypus novemcinctus) and coyotes (Canis latrans) during summer 2023 within the Wichita Mountains Wildlife Refuge, OK, USA. Hillshade depicts relative elevation and slope overlayed with the distribution of vegetation >1-m tall; hillshade and vegetation layers were derived from 2023 LANDFIRE data [http://www.landfire.gov].
Figure 2. Location of 99 sites (Cameras) surveyed with motion-activated cameras for nine-banded armadillos (Dasypus novemcinctus) and coyotes (Canis latrans) during summer 2023 within the Wichita Mountains Wildlife Refuge, OK, USA. Hillshade depicts relative elevation and slope overlayed with the distribution of vegetation >1-m tall; hillshade and vegetation layers were derived from 2023 LANDFIRE data [http://www.landfire.gov].
Diversity 17 00290 g002
Diversity 17 00290 g003
Figure 3. Estimated probability of occurrence (Occupancy) for nine-banded armadillos (Dasypus novemcinctus) as a function of (a) distance to water, (b) distance to cover, and (c) slope, and (d) probability of use (Space Use) for coyotes (Canis latrans) as a function of elevation, during summer 2023 in the Wichita Mountains Wildlife Refuge, OK, USA. Probabilities were plotted based on the most-supported model structure for each species, using the mean values for other numeric covariates and as a function of fire history (with burned indicating sites were burned within the 5 years preceding sampling). Shaded areas indicate 95% confidence intervals.
Figure 3. Estimated probability of occurrence (Occupancy) for nine-banded armadillos (Dasypus novemcinctus) as a function of (a) distance to water, (b) distance to cover, and (c) slope, and (d) probability of use (Space Use) for coyotes (Canis latrans) as a function of elevation, during summer 2023 in the Wichita Mountains Wildlife Refuge, OK, USA. Probabilities were plotted based on the most-supported model structure for each species, using the mean values for other numeric covariates and as a function of fire history (with burned indicating sites were burned within the 5 years preceding sampling). Shaded areas indicate 95% confidence intervals.
Diversity 17 00290 g003
Diversity 17 00290 g004
Figure 4. Estimated probability of detection for nine-banded armadillos (Dasypus novemcinctus) when coyotes (Canis latrans) were absent (pB) or present (rB) at a site during summer 2023 in the Wichita Mountains Wildlife Refuge, OK, USA. Probabilities plotted based on the mean values of continuous predictors in the most-supported detection model for armadillos (i.e., mean daily minimum temperature; °C).
Figure 4. Estimated probability of detection for nine-banded armadillos (Dasypus novemcinctus) when coyotes (Canis latrans) were absent (pB) or present (rB) at a site during summer 2023 in the Wichita Mountains Wildlife Refuge, OK, USA. Probabilities plotted based on the mean values of continuous predictors in the most-supported detection model for armadillos (i.e., mean daily minimum temperature; °C).
Diversity 17 00290 g004
Diversity 17 00290 g005
Figure 5. Kernel density estimates of activity patterns for nine-banded armadillos (Dasypus novemcinctus) and coyotes (Canis latrans) resulting from camera-trapping surveys conducted during summer 2023 in the Wichita Mountains Wildlife Refuge, OK, USA.
Figure 5. Kernel density estimates of activity patterns for nine-banded armadillos (Dasypus novemcinctus) and coyotes (Canis latrans) resulting from camera-trapping surveys conducted during summer 2023 in the Wichita Mountains Wildlife Refuge, OK, USA.
Diversity 17 00290 g005
Table 1. Beta coefficients (β), standard errors (SE), and 85% lower (LCL) and upper (UCL) confidence limits for covariates in the most-supported model structures of detection (p) and occupancy (ψ) for nine-banded armadillos (Dasypus novemcinctus) and coyotes (Canis latrans) sampled during summer 2023 in the Wichita Mountains Wildlife Refuge, OK, USA; cumulative Akaike model weights (Σwi) indicate relative predictor importance when considering the full model set, and bold indicates beta coefficients of predictors with 85% confidence intervals that did not overlap zero and cumulative weights ≥0.60.
Table 1. Beta coefficients (β), standard errors (SE), and 85% lower (LCL) and upper (UCL) confidence limits for covariates in the most-supported model structures of detection (p) and occupancy (ψ) for nine-banded armadillos (Dasypus novemcinctus) and coyotes (Canis latrans) sampled during summer 2023 in the Wichita Mountains Wildlife Refuge, OK, USA; cumulative Akaike model weights (Σwi) indicate relative predictor importance when considering the full model set, and bold indicates beta coefficients of predictors with 85% confidence intervals that did not overlap zero and cumulative weights ≥0.60.
ArmadilloParameterβSELCLUCLΣwiCoyoteParameterβSELCLUCLΣwi
pIntercept2.000.990.573.43-pIntercept−3.551.32−5.45−1.65-
Effort----0.45 Effort0.340.190.060.610.77
Temp−0.100.05−0.17−0.030.61 Temp----0.44
Prec----0.33 Prec----0.27
ψIntercept1.740.890.463.03-ψIntercept−5.613.40−10.50−0.71-
RxB0.930.560.131.740.64 RxB1.320.660.372.270.66
DistW0.090.060.010.170.57 DistW----0.43
DistRd----0.34 DensRd----0.25
Slope−0.310.10−0.45−0.170.99 Elevation0.100.060.010.190.55
DistCov−1.400.39−1.95−0.841.00 %Forest----0.40
Notes: Effort = measure of sampling effort during a survey period; Temp = mean daily minimum temperature (Celsius); Prec = mean daily precipitation (mm); DistW = Distance to water (hm); DistRd = Distance to roads (hm); DensRd = Road density (km/km2); DistCov = Distance to cover > 1-m tall (hm); %Forest = Proportion of site classified as forest; Slope = mean slope (degrees); Elevation = mean elevation (dm); RxB levels included sites burned or unburned within the preceding 5 years (the intercept reflected unburned sites); a dash indicates the covariate was included in the model set but was not present in the most-supported model.
Table 2. Model selection results for conditional two-species occupancy models of coyotes (Canis latrans; Species A) and nine-banded armadillos (Dasypus novemcinctus; Species B) sampled during summer 2023 in the Wichita Mountains Wildlife Refuge, OK, USA, with models ranked by Akaike’s Information Criterion adjusted for small samples (AICc) and differences in AICc (ΔAICc, where Δi = AICci− AICcmin), and reported with number of parameters (K), Akaike weight (wi), and deviance (Dev.).
Table 2. Model selection results for conditional two-species occupancy models of coyotes (Canis latrans; Species A) and nine-banded armadillos (Dasypus novemcinctus; Species B) sampled during summer 2023 in the Wichita Mountains Wildlife Refuge, OK, USA, with models ranked by Akaike’s Information Criterion adjusted for small samples (AICc) and differences in AICc (ΔAICc, where Δi = AICci− AICcmin), and reported with number of parameters (K), Akaike weight (wi), and deviance (Dev.).
Model StructureKAICcΔAICcwiDev.
ψA, ψBA = ψBa, pA = rA, pB, rBA = rBa14828.9240.0000.49795.92
ψA, ψBA = ψBa, pA = rA, pB, rBA, rBa16830.5921.6680.21791.96
ψA, ψBA = ψBa, pA = rA, pB = rBA = rBa12830.7151.7910.20803.09
ψA, ψBA = ψBa, pA, rA, pB, rBA = rBa16833.5104.5860.05794.88
ψA, ψBA = ψBa, pA, rA, pB, rBA, rBa18834.7615.8370.03790.21
ψA, ψBA = ψBa, pA, rA, pB = rBA = rBa14835.4976.5730.02802.50
ψA, ψBA, ψBa, pA = rA, pB, rBA = rBa19837.9469.0220.01790.33
ψA, ψBA, ψBa, pA = rA, pB, rBA, rBa21841.26912.3450.00787.27
ψA, ψBA, ψBa, pA, rA, pB, rBA = rBa21841.43312.5090.00787.43
ψA, ψBA, ψBa, pA = rA, pB = rBA = rBa17841.83512.9110.00800.28
ψA, ψBA, ψBa, pA, rA, pB, rBA, rBa23844.14315.2190.00783.42
ψA, ψBA, ψBa, pA, rA, pB = rBA = rBa19847.02018.0960.00799.40
Notes: ψA = occupancy of coyotes; ψB = occupancy of armadillos in the presence (ψBA) and absence (ψBa) of coyotes; pA = detection of coyotes in the absence of armadillos; rA = detection of coyotes in the presence of armadillos; pB = detection of armadillos in the absence of coyotes; rB = detection of armadillos when coyotes were present and detected (rBA) or not (rBa); models for detection (i.e., p and r parameters) and occupancy (ψ parameters) included covariates for coyotes (i.e., p ~ Effort; ψ ~ Elevation + Burn history) and nine-banded armadillos (i.e., p ~ Temperature; ψ ~ Slope + Distance to water + Distance to cover + Burn history) that were identified as important during single-species analyses.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Lonsinger, R.C.; Murley, B.P.; McDonald, D.T.; Fallon, C.E.; White, K.M. Habitat and Predator Influences on the Spatial Ecology of Nine-Banded Armadillos. Diversity 2025, 17, 290. https://doi.org/10.3390/d17040290

AMA Style

Lonsinger RC, Murley BP, McDonald DT, Fallon CE, White KM. Habitat and Predator Influences on the Spatial Ecology of Nine-Banded Armadillos. Diversity. 2025; 17(4):290. https://doi.org/10.3390/d17040290

Chicago/Turabian Style

Lonsinger, Robert C., Ben P. Murley, Daniel T. McDonald, Christine E. Fallon, and Kara M. White. 2025. "Habitat and Predator Influences on the Spatial Ecology of Nine-Banded Armadillos" Diversity 17, no. 4: 290. https://doi.org/10.3390/d17040290

APA Style

Lonsinger, R. C., Murley, B. P., McDonald, D. T., Fallon, C. E., & White, K. M. (2025). Habitat and Predator Influences on the Spatial Ecology of Nine-Banded Armadillos. Diversity, 17(4), 290. https://doi.org/10.3390/d17040290

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop