Where do roads impact Gopher Tortoise conservation in Georgia?


Male Gopher Tortoise basking on a cool winter day – Houston Chandler

            In Georgia, roads are a pervasive landscape feature. The ever expanding road network connects cities and winds through almost every corner of the state.  This dense road network is spurred on by Georgia’s increasing population, which has grown by over a million residents in the last 10 years to a total of over 10.6 million.  Many of the roads in Georgia have now existed on the landscape for many decades, and they will likely remain for many more decades if not centuries.  I have written previously about how roads can negatively impact wildlife populations through direct mortality, isolation, and a myriad of other effects (https://www.oriannesociety.org/perils-crossing-road/).  Even though the negative effects of roads are generally well-known, it can be challenging to identify locations where mitigation efforts can reduce these effects, especially over large geographic areas with many thousands of kilometers of roadways.

            The Gopher Tortoise range in Georgia covers the majority of the southern portion of the state, approximately 92,000 km2 of total land area.  How many kilometers of public road pass through this portion of Georgia?  We recently completed a project for the Georgia Department of Transportation (DOT) designed to answer this question and identify locations where roads are interacting with viable tortoise populations and tortoise habitat.  Potential impacts of roads on Gopher Tortoises are particularly concerning because turtle populations generally respond poorly to the presence of roads (Howell and Seigel 2019; Steen and Gibbs 2003).  Gopher Tortoises can also be attracted to road shoulders where open canopy, grassy habitats are artificially maintained.

Gopher Tortoise traveling adjacent to one of Georgia’s highways – Houston Chandler

            To answer my earlier question, there are over 101,000 km (62,700 miles) of public road mapped by GA DOT in the Gopher Tortoise’s Georgia range.  In fact, there are few large areas in the southern portion of the state where roads do not exist.  Most of the large areas without public roads are the floodplains of major rivers, the Okefenokee Swamp, and large military installations.  Floodplains and swamps make poor tortoise habitat and most military installations actually have dense road networks for military use.  The types of roads within the tortoise range run the gamut from small, rarely used dirt roads to major interstates and highways.  Just over 60% of the roads are paved, allowing vehicles to travel at higher speeds and increasing the chances of wildlife-vehicle collisions.  The Gopher Tortoise’s preferred upland habitats are often the ones that make ideal locations for road placement, leaving the vast majority of upland habitats in Georgia divided into relatively small sections by some type of public road. 

The Gopher Tortoise range and public road network in southern Georgia.

            To identify where roads are most likely to impact tortoise populations, we first looked for documented observations of tortoises crossing roads in Georgia.  However, despite a large road network and the conservation concern for tortoises, there are relatively few such cases.  The ones that have been recorded are spread across the state, with few occurring on the same road.  Most studies attempting to identify hotspots for vehicle-wildlife collisions use large databases with hundreds or thousands of records or examine crossings at much smaller spatial scales (i.e., one road passing through a protected area).  Thus, we decided to focus on tortoise habitat and the already delineated viable tortoise populations for our analyses.

            There are 122 known viable Gopher Tortoise populations in Georgia that have been identified as part of the Gopher Tortoise Conservation Initiative.  These populations occur on both public and private lands across southern Georgia, and much of the conservation effort for tortoises in Georgia is now focused on these remaining viable populations.  All but three of these viable populations have at least 1 km of road bordering or bisecting the tortoise habitat (range: 0–75.6 km).  Unsurprisingly, the majority of these roads (60%) are small, generally unpaved roads with low traffic volumes.  These small roads likely have minimal impacts on tortoise populations through direct mortality but do increase access and degrade surrounding habitat.  More concerning are the 79 instances where paved roads completely bisect one viable population or serve as the border between two adjacent populations.  In these cases, tortoises and high quality habitat are present on both sides of the road, likely increasing the chances that individuals will attempt to cross to the other side.  These road segments represent ideal locations to better understand how tortoises respond to roads and to explore potential mitigation strategies.

An example of how roads interact with viable tortoise populations. Functional classes range from generally unpaved, local roads (7) to interstates and major highways (1).  Red dots are observations of tortoises on or near a road.

            Examining viable tortoise populations only tells part of the story because tortoise habitat and smaller populations are spread across the state, and there could even be currently unknown viable populations in some areas.  At this broader spatial scale, we used a habitat model for Gopher Tortoises created by researchers at the University of Georgia to assign a ‘Gopher Tortoise Score’ to every road segment in the tortoise’s range.  The habitat model assigns a suitability score across a 30 x 30 m grid using various environmental variables that effect habitat quality for tortoises.  Suitability scores can then be classified as unsuitable, low, medium, or high quality habitat.  For each road segment, we calculated an average suitability value for the habitat that the road passes through, and then scored that segment with a 0–3 (unsuitable to high quality) based on the average habitat suitability.  We then gave each segment an additional point if it passed through a viable population or if it passed through a high quality habitat patch bigger than 100 ha.  The resulting data layer scored each road (all 101,000 km) segment with a score ranging from 0–5, indicating the relative quality of the surrounding tortoise habitat.  Higher scores indicate better habitat with known tortoise populations and/or large habitat patches.  Areas of the state with a high density of upland sandhill were, unsurprisingly, the most likely to receive high scores.

            These two approaches give GA DOT and others involved in the Gopher Tortoise Conservation Initiative an objective method to assess roads based on the surrounding tortoise habitat and provide a short-list of roads that directly impact viable tortoise populations.  The extent of the road network in Georgia makes it nearly impossible to eliminate the negative effects of roads on wildlife.  However, mitigation efforts (e.g., under passes or fencing) can be effectively employed to reduce these impacts.  Finally, there is surprisingly little known about how Gopher Tortoises interact with roads, including how and when they attempt to cross them.  Even if tortoises do not often have direct conflict with vehicles, populations could be impacted within a large area surrounding a road through avoidance behavior or other road effects (Boarman and Sazaki 2006).  Roads are just one of the many conservation challenges for tortoises and other herpetofauna, but they are a near omnipresent factor throughout the southeastern U.S.  This is unlikely to change anytime soon and so creative solutions must be identified to minimize the negative effects that they have on wildlife populations.

Literature Cited

Boarman, W. I., and M. Sazaki. 2006. A highway’s road-effect zone for desert tortoises (Gopherus agassizii). Journal of Arid Environments 65:94–101.

Howell, H. J., and R. A. Seigel. 2019. The effects of road mortality on small, isolated turtle populations. Journal of Herpetology 53:39–46.

Steen, D. A., and J. P. Gibbs. 2003. Effects of roads on the structure of freshwater turtle populations. Conservation Biology 18:1143–1148.