https://vital.seals.ac.za/vital/access/manager/Index ${session.getAttribute("locale")} 5 https://vital.seals.ac.za/vital/access/manager/Repository/vital:4655 Wed 12 May 2021 23:48:25 SAST ]]> https://vital.seals.ac.za/vital/access/manager/Repository/vital:5606 50 km/h and were also significantly influenced by light conditions (i.e. detection success was greater when the sun was high) and the position of the roadkill on the road (i.e. smaller roadkill on verges were often missed). These results suggest that roadkill sampling was most effective between 1.5 hours after dawn and 1.5 hours before dusk and that driving at slower speeds (<50 km.h⁻ₑ) was required to detect roadkill. This protocol was implemented across three ecological seasons on a 100 km paved road and a 20 km unpaved road in the Greater Mapungubwe Transfrontier Conservation Area, Limpopo Province, South Africa. Driven daily over a 120-day period (three periods consisting each of 40 days), a total of 1,027 roadkill were recorded. These comprised 162 species from all terrestrial vertebrate groups with birds being the most commonly encountered roadkill (50% of all incidents). The high numbers of vertebrates identified as roadkill suggests that road traffic could have potentially unsustainable impacts on wildlife populations and hence the biodiversity of the area. Seventeen variables were identified as possible determinants of roadkill occurrence with season, rainfall, minimum and maximum temperature, habitat type, grass height, grass density, fence type and vehicle type significantly influencing roadkill numbers. Significantly more roadkill were detected on the paved road (9.91/100km) than on the unpaved road (1.8/100km) probably because of greater traffic volumes and the increased speed that vehicles travelled on the paved road. Warmer temperatures and increased rainfall in the preceding 24 hours also increased road mortality numbers as animals tended to become more active during these times. Interestingly, more roadkill was detected in open roadside habitats compared to dense roadside habitats on both the paved and unpaved roads and when grass on the roadside verge was of intermediate height. Open habitat possibly may provide a natural corridor for wildlife which ultimately end up on the road. Roadkill numbers increased when certain other physical barriers, such as cattle fences, were present, probably because these barriers were more penetrable than electric fencing. A series of mitigation measures are proposed to reduce the impacts of roads on wildlife in South Africa. These mitigation measures highlight the need to address the balance between the development of a country’s transport infrastructure and the conservation of its fauna. It is important that research on the impacts of roads becomes standardised to enable robust statistical comparisons which will provide a greater understanding of the potential threats to vertebrate biodiversity]]> Wed 12 May 2021 19:11:07 SAST ]]>