IENE 2016

Expanding road networks pose one of the largest threats to the Guiana Shield biodiversity hotspot, where wildlife has mostly remained protected through inaccessibility. Responsibility for smart road building and management should lie with investors and developers; as scientists we can provide data-driven recommendations for wildlife-friendlier roads. We conducted a study on wildlife crossings in Guyana for government and the International Development Bank as part of a larger pre-investment study on the Linden-Lethem Road (LLR) in the framework of the Integration of Regional Infrastructure in South America. This 438km section is the only unpaved part of the road connecting metropolis Manaus (Brazil) to the Atlantic Ocean. LLR bisects Guyana and provides the main access to the country’s hinterland, traversing logging, mining, agriculture, settlements, biodiversity conservation areas and tourism regions. For our short-term study in May-June 2014, we prioritized road sections based on vegetation and land-use, selecting the Rupununi - a unique habitat of seasonally flooded savannah - and Iwokrama – the adjacent protected forested site. Previous research here had demonstrated vast mammal richness and abundance. A road upgrade with augmented traffic flow could have large negative effects on wildlife. Generally, for the Neotropics, but especially for savannah, limited data on road impacts on wildlife are available. LLR offered a unique opportunity to gather data prior to pavement, and present recommendations to the developers. Based on surveys of scat, road kill, live sightings, and camera traps under bridges in the Rupununi, and existing life sighting and road kill data from 6 years in Iwokrama, we identified two major crossing locations in the savannahs, while for the protected forest animals crossed randomly. Under-passage use was negligible. Road kill rate - 0.0048 individual mammals killed per kilometre of savannah road surveyed (0.64 kills/24hrs) - was very low compared to other records from unpaved and paved roads in the Neotropics, but followed similar patterns in species vulnerability. While the most common mammal, the savannah fox (Cerdocyon thous), also had the highest kill rate, this direct relationship was not true for other species. Road kill species diversity was also very low, perhaps due to a short survey time in relation to road kill rates. Applying the lowest reported rates for similar environments, pavement of this section can be expected to result in a minimum 10-fold increase in road kill, depending on traffic volume, with the biggest concern for the threatened giant anteaters (Myrmecophaga tridactyla).

We recommend protection of habitat along the road (wildlife friendly segments), speed limit reduction through signage and speed bumps in the identified locations, and continuation of the no-night-driving policy within Iwokrama. We also recommend implementation of hunting management to help mitigate indirect impacts from the road upgrade on fauna. Finally, very little is known about road impacts on Neotropical mammals, and we advocate for mitigation strategies of road development projects to include research on barrier effects and effectiveness of underpasses and rope bridges in maintaining genetic connectivity to assist with adaptive management of the LLR and inform future road projects in Central-and South America.