The ability to move is crucial for the survival of nearly all wildlife species. Yet, human development has fragmented natural habitats and created barriers to wildlife movement across much of the Earth’s surface (Barnosky et al., 2012; Cozzi et al., 2013; Davis et al., 2024). The loss of habitat connectivity and landscape permeability can reduce survival of individuals (Coppola et al., 2021), decrease the viability of populations (Fahrig and Merriam, 1994; Olson et al., 2009), suppress gene flow (Aspi et al., 2009; Sawaya et al., 2019), and increase extinction risk (Benson et al., 2016; Clobert, 2012; Gilpin, 1986). Hence, maintaining and increasing habitat connectivity and understanding barriers to movement are vital to conservation efforts (Abrahms et al., 2017; Ford et al., 2020; Heller and Zavaleta, 2009).

Conservation efforts across geopolitical boundaries can be hindered by varying environmental regulations and customs (Rick et al., 2004; Thornton and Branch, 2019). Geopolitical boundaries may also act as physical barriers to conservation and wildlife movement because of the growing presence of infrastructure designed to control the movements of humans and goods. Indeed, construction of border barriers has accelerated globally in recent decades with barriers now present in 74 countries (Vallet, 2022). The ecological impacts of such border barriers can be severe as barriers are often long, continuous, and built with minimal accommodations for wildlife due to security concerns (Aspi et al., 2009; Linnell et al., 2016).

The USA-Mexico border exemplifies the recent proliferation of border barriers (Lasky et al., 2011; McCallum et al., 2014). After the passage of the Secure Fence Act in 2006, 1,053 km of border barriers were built, including 570 km of pedestrian border walls and 483 km of vehicle barriers (Isacson, 2023; US Government Accountability Office, 2017, 2023). Pedestrian border walls (hereafter referred to as, “border walls”) vary but are typically steel bollard walls, constructed with square carbon steel tubing 15 cm wide and 5.5-9.1 m tall. Interstitial spaces between bollards measure ≤10 cm to stop all human crossings. Vehicle barriers are a variable system of steel posts arranged to stop vehicles but not pedestrians (US Government Accountability Office, 2021). Between 2017 and 2020, the USA replaced most vehicle barriers with border walls and built new walls in previously open areas (all border walls built between 2017 and 2020 were 9.1 m tall). As a result, 1,023 km of border walls and 169 km of vehicle barriers exist along the USA-Mexico border in 2024 (US Government Accountability Office, 2023).

Generally, wildlife movements have not been considered a priority when border barriers have been constructed along the USA-Mexico border. On some federal conservation lands (e.g., National Wildlife Refuges, etc.), small wildlife passages measuring roughly 21.5 x 27.8 cm were added to border walls to facilitate movements of endangered species like ocelots (Leopardus pardalis) in south Texas, USA (US Fish and Wildlife Service, 2016). Additionally, floodgates were installed where the border wall crosses arroyos and riverbeds. These gates are opened to accommodate seasonal floods but remain closed most of the year. A settlement in July of 2023 with the Department of Homeland Security will allow select floodgates to remain open longer (Sierra Club v. Biden, 2023), however, most gates remain closed except during high water events. As such, crossing opportunities are limited for wildlife along most of the border wall to the interstitial spaces between bollards, scarce small wildlife passages, and, rarely, open floodgates.

The USA-Mexico border crosses some of the most biologically diverse regions in North America. For example, the Madrean Sky Islands span the USA-Mexico border in southeastern Arizona, USA, and northeastern Sonora, MX. These mountains are recognized as a global biodiversity hotspot and are home to the highest density of bee species in North America (Minckley and Radke, 2021), the highest diversity of mammals, reptiles, and ants in the USA (Warshall, 1994), and one of the highest rates of plant endemism in the USA (McLaughlin, 1994). In its totality, the USA-Mexico border intersects the geographic ranges of 1,077 native animals and bisects important habitat for numerous federally listed species (Peters et al., 2018). Border barriers, especially border walls, may severely limit movements of many of these native animals, including American black bear (Ursus americanus), mountain lion (Puma concolor), desert bighorn sheep (Ovis canadensis nelsoni), mule deer (Odocoileus hemionus), white-tailed deer (O. virginianus), and federally listed species such as jaguar (Panthera onca), Sonoran pronghorn (Antilocapra americana sonoriensis), and Mexican gray wolf (Canis lupus baileyi). Border barriers may hinder recovery efforts for these listed species and threaten other species not currently listed by restricting movement, isolating populations, severing corridors, and limiting gene flow (Chambers et al., 2022; Lasky et al., 2011; Peters et al., 2018).

Presently, land managers and conservation practitioners have insufficient information to manage or mitigate the effects of the USA-Mexico border wall on wildlife populations. Further, the efficacy of small wildlife passages to facilitate movement of wildlife is untested in this region. To address such knowledge gaps, we deployed motion-activated cameras near border barriers along 163.5 km of the USA-Mexico border to: 1) evaluate crossing rates for 20 terrestrial species through border walls and vehicle barriers, and 2) quantify use of the small wildlife passages.

We collected 12,780 videos of the 20 focal species, resulting in 9,240 events. Most events (5,225 of 9,240; 56.6%) involved no interaction with border barriers, whereas 22.7% (2,095 of 9,240) of events featured an animal interacting but not crossing, and 20.8% (1,920 of 9,240) of events featured a successful crossing. We documented one fatality of an adult collared peccary that became stuck between bollards in the border wall and subsequently died from exposure. Coyote was the most common species with 1,759 events, followed by collared peccary (n = 1,402), and white-tailed deer (n = 1,170; Table 1 ). We documented all 20 focal species at vehicle barriers, 90% (18 of 20) at small wildlife passages, and 85% (17 of 20) at border walls. When considering all focal species together, average daily activity (as number of barrier interactions per camera trap-day) varied significantly by barrier type (KW₂ = 20.7, p = <0.001) as did the average number of crosses per camera trap-day (KW₂ = 50.2, p = <0.001). Specifically, average daily activity was 3.2 times higher near small wildlife passages (µ = 0.60, SE = 0.08) than at border walls (µ = 0.19, SE = 0.03) and the average number of crosses per camera trap-day was 16.7 times greater at small wildlife passages (µ = 0.25, SE = 0.05) than at border walls (µ = 0.015, SE = 0.004; Supplementary Table S1 ).

Of the focal species we documented at each barrier, 55.6% (10 of 18) crossed through small wildlife passages, 58.8% (10 of 17) crossed through the interstitial spaces in the border wall, and 100% (20 of 20) crossed through the vehicle barrier ( Table 1 ). Within species, crossing rates were consistently highest at vehicle barriers and lowest at border walls ( Figure 3 ). This was evident even for smaller species (e.g., skunk species, bobcat, etc.) that could easily cross through interstitial spaces in the border wall. For example, bobcats showed no physical limitations when crossing through interstitial spaces of border walls but were still 5.1 times more likely to cross through vehicle barriers than border walls (0.82, SE = 0.06 vs 0.16, SE = 0.05). Although many relatively small species were able to cross through all border barriers, we did not observe larger species crossing through border walls. Indeed, white-tailed deer, mule deer, American black bear, American badger, wild turkey, and mountain lion never successfully crossed (crossing rate = 0.0) through interstitial spaces in the border wall but crossed during >50% of events at vehicle barriers (crossing rates of 0.5-0.79).

Small wildlife passages benefited larger-bodied species than we expected. Indeed, the crossing rate of collared peccary (average mass = 28.5 kg) was 24 times greater through small wildlife passages than through interstitial spaces in the border wall (0.38, SE = 0.06 vs 0.016, SE = 0.016; Figure 3 ). Furthermore, mountain lions (average mass = 74.5 kg) never crossed through interstitial spaces in border walls but had a crossing rate of 0.42, SE = 0.12 at small wildlife passages (Figure 3). However, small wildlife passages did not offer any crossing opportunities for white-tailed deer, mule deer, American black bear, or wild turkey, despite many interactions from these species near the passages. Smaller species such as gray fox, northern raccoon, and skunks had similar crossing rates through interstitial spaces in the border wall and small wildlife passages. Finally, we did not observe a single crossing event by a human through any border barrier.

We present a novel assessment of crossing rates for 20 terrestrial wildlife species through barriers across 163.5 km of the USA-Mexico border. We demonstrate that several large terrestrial species are unable to cross through border walls that now bisect some of the most biodiverse regions of North America and bound approximately 1,023 km of the USA-Mexico border. Moreover, we evaluate for the first time the efficacy of the small wildlife passages and show that they are used frequently by several species. Nevertheless, these small openings are inadequate for large carnivores and ungulates that may have the greatest need to move to find mates, secure resources, or avoid competition. Additional research efforts such as this study are needed to elucidate the ecological consequences of border barriers and potential mitigation strategies.

We observed 10 focal species crossing through the small wildlife passages including two species apparently unable or unwilling to cross through interstitial spaces in the border wall (i.e., mountain lion and American badger) and two species unlikely to cross through interstitial spaces (i.e., collared peccary and coyote). However, only 13 small wildlife passages exist in our study area, with 8 distributed across 6.3 km near the San Bernardino National Wildlife Refuge and 5 distributed across 0.5 km at the San Pedro Riparian National Conservation Area. These clusters of openings are separated by 95 km of continuous border wall, thereby limiting their utility to individuals that do not reside near one of these conservation areas. Moreover, current small wildlife passages measure only 21.5 x 27.8 cm (roughly the size of a piece of A4 printer paper) and provided no crossing opportunities for American black bear, white-tailed deer, mule deer or wild turkey in our study. It is unlikely existing passages would provide crossing opportunities for similarly sized mammals of conservation concern in the area including pronghorn, desert bighorn sheep, jaguar, and Mexican gray wolves. The latter two species are larger than mountain lions and coyotes we observed crossing through the small wildlife passages (De La Torre and Rivero, 2017), and probably would not fit through existing passages. Gaps between constructed sections of border walls remain in some regions of the border (Traphagen, 2021), but the border wall section we monitored is uninterrupted for 130.2 km ( Figure 1 ). Beyond the 13 small wildlife passages, no permanent openings exist for wildlife along this entire stretch of wall. More small wildlife passages should be installed (e.g., every 5-10 m) and different sized passages should be considered to more meaningfully mitigate the impacts of the border wall on wildlife connectivity.

The impacts of the border wall may manifest at multiple levels. At the demographic level, border walls may affect survival and reproductive rates by impeding access to resources and mating opportunities (Cozzi et al., 2013; Hering et al., 2022; Safner et al., 2021). At the population level, border walls may limit dispersal, thereby increasing isolation and decreasing viability of subpopulations (Flesch et al., 2010). For example, American black bears disperse regularly between mountain ranges in the USA and Mexico, and genetic analyses show populations in southeastern Arizona are closely connected with those of northern Sonora, Mexico (Atwood et al., 2011; Gould et al., 2022; Lara-Díaz et al., 2013; Varas et al., 2010). However, our data indicate that the border wall is mostly impermeable for American black bears and will thus hinder their dispersal between the USA and Mexico, where they are listed as federally endangered (SEMARNAT, 2010). Border walls could thus threaten the population viability of American black bears in Mexico. Border walls may also influence community composition and structure by excluding some species but not others from adjacent habitats. For example, many mesocarnivores (e.g., bobcat, gray fox, etc.) had relatively high crossing rates through the border walls, whereas large carnivores were effectively excluded. Finally, climate change is driving the distribution of many species to higher latitudes and elevations (Chen et al., 2011; Thornton and Branch, 2019). The presence of largely impermeable barriers along 1,023 km of the USA-Mexico border may hinder the ability of many species to reach climate refuges, thereby increasing their vulnerability to such changes (Chambers et al., 2022; Davis et al., 2024; Egan et al., 2018).

Other factors associated with the border barriers may influence crossing rates and warrant further research. For example, we considered only the barriers themselves when evaluating crossing rates and use of the small wildlife passages. Border barriers are often accompanied by parallel roads (10-18 m wide in our study area) and a network of access roads with varying levels of law enforcement and maintenance traffic (Sayre and Knight, 2010; US Government Accountability Office, 2021). Moreover, some areas of the USA-Mexico border have high-intensity security lighting that illuminates the barriers (US Government Accountability Office, 2021). We did not include lighting in our analyses because lighting infrastructure has been installed but not activated in our study area. Furthermore, landscape characteristics (e.g., terrain) and vegetation conditions (e.g., vegetation cover adjacent to border barriers) probably affect wildlife occurrence and crossing rates. Future work should evaluate the influence of human activity, additional security infrastructure such as lighting (including for taxonomic groups like birds, bats, and insects that may not be as directly affected by border barriers), and environmental covariates on wildlife crossing rates through border barriers. Additionally, we considered all non-crossing interactions as a single category in our analyses. However, further insights into the impacts of the border barriers could be gained by evaluating non-crossing interactions in greater detail. Finally, we evaluated crossing rates of 20 wildlife species, but many other species may be impacted by border barriers that we could not evaluate because they did not occur in our study area (e.g., peninsular bighorn sheep, Mexican gray wolf, etc.) or because we did not detect them on our cameras, such as desert tortoise (Gopherus agassizii), pronghorn, ocelot, jaguar, and ferruginous pygmy owl (Glaucidium brasilianum; Flesch et al., 2010). The impacts of border barriers on these species should not be overlooked.

Approximately 1,023 km of the USA-Mexico border (including 63% of the southern border of Arizona) is now bounded by border walls 5.5-9.1 m tall with interstitial spaces between bollards measuring ≤10 cm. Our study is among the first to explicitly evaluate the permeability of these barriers for terrestrial species along the USA-Mexico border. Our results indicate border walls are largely impermeable to many terrestrial mammals, including several federally listed species in the USA and MX. We further demonstrate that existing wildlife passages have the potential to improve crossing rates for some species, but more passages (including larger openings) are needed. Our work may help conservation practitioners and land managers to understand the impacts of current border barriers on wildlife movement and improve design and distribution of wildlife passages. As construction of border barriers accelerates globally, research into the full extent of ecological consequences of these barriers and effective mitigation strategies will be paramount for transboundary conservation efforts throughout the world.