Skip to content


May 23, 2011

There is much data on human road casualties, but very little on animal road deaths and particularly not in Southern Africa. Despite road traffic being a known cause of wildlife deaths, studies are poorly represented, and many existing studies are either localised and not representative of a national population. Consequently, the Endangered Wildlife Trust’s Roadkill Research and Mitigation Project (EWT-RRAMP) has developed a Rapid Assessment Protocol for the detection of roadkill on our country’s roads. The project is being conducted across two phases. During Phase 1, which will be completed by July 2011, a Rapid Assessment Protocol for the detection of roadkill on roads was developed. This is the first time a method for roadkill detection has been standardized. During Phase 2, the Rapid Assessment Protocol will be tested in order to obtain baseline Wildlife Road Traffic Accident (WRTA) rates in one important conservation area in South Africa, namely the Greater Mapungubwe Transfrontier Conservation Area (GMTFCA), which incorporates the surrounds of the De Beers Venetia Limpopo Nature Reserve and Mapungubwe National Park.  This focal survey will identify the factors affecting WRTA rates, and provide the basis for a series of mitigation measures that can reduce the impact of road infrastructure and use on biodiversity, as well as contribute towards national norms and standards for future road design, maintenance and improvement.  It will also provide a better understand of the factors contributing to WRTAs and consequently to assess threats to biodiversity from WRTAs across the country.

These objectives will be achieved by using the Rapid Assessment Protocol to examine the seasonal incidence of WRTAs across a one-year period. Selected sections of road on the R572 and R521 will be driven to incorporate a variety of road usage categories, including  tourism traffic, mine traffic and local traffic. The Rapid Assessment Protocol outlines the most economical and time-saving approach to assessing WRTA rates, including the best speed at which to drive, direction in which to travel, time of day to sample, and how frequently to drive as well as how far to drive.

                                   Figure 1: Greater Mapungubwe Trans Frontier Conservation Area. (Peace Parks Foundation, 2009)


 Figure 2: Proposed Transect showing detail of the Mapungubwe National Park, De Beers Venetia Limpopo Nature Reserve and surrounds.

The characteristics of each road kill will be recorded to evaluate their influence on the incidence of WRTAs. These have been broadly classed into four categories: physical road characteristics such as bridges, and culverts, road width, road condition and road surface type; environmental characteristics such as surrounding vegetation, verge habitat, grass height, type of fencing, climatic conditions, moon phase and season; human road usage characteristics including traffic volume, time of day, day of the week; and aspects specific to the species killed such as animal behavior patterns, the incidence of secondary roadkill, and presence of live animals on road either through dung or direct observation. These will all be recorded to determine how they influence roadkill rates.

Traffic volumes will be recorded using a PicoCount 2500 counter and pneumatic road tube. A rubber hose will be placed across the width of the road and secured in place with nails, hose grips and tape.  Attached to this will be a traffic counter which will record traffic volume and speeds. Using a download cable, data can be collected at the end of each roadkill survey and roadkill rates can then be compared to traffic volumes. The road tube will be randomly rotated on the different roads sampled.

Through the implementation of the WRTA Rapid Assessment Protocol the findings from this study will:

  • Contribute to considerations for improved future road design and maintenance.
  • Provide a platform to lobby government departments (and in particular the Department of Transport) in terms of mitigation measures for WRTAs.
  • Supplement existing assessments of species range distributions in southern Africa.

 Furthermore, the WRTA Rapid Assessment Protocol can be employed in areas already identified as ‘hotspots’ such as Hluhluwe-iMfolozi Park (HiP) in KwaZulu-Natal. The area has been highlighted as a hotspot because of the high frequency of WRTAs involving African Wild Dogs Lycaon pictus. Implementing the WRTA Rapid Assessment Protocol to sample the roads in HiP, will enable a stronger case to be built to lobby government departments to make road improvements. This is just one species at threat from WRTAs in one known hotspot area.

Future Environmental Impact Assessments (EIAs) will also be able to consider temporal assessments of the impacts of WRTAs elsewhere and make use of the Rapid Assessment Protocol for detecting roadkill.

Wendy Collinson, the project co-ordinator, has provided some interesting, if rather shocking data in the recent roadkill surveys whilst developing the Rapid Assessment Protocol. 374 wildlife roadkills were detected over a 30-day period on a 67km transect. Birds were by far the most commonly impacted species (figure 3) with 2 Giant Eagle Owls and 2 Spotted Eagle Owls forming part of the statistics. A simple breakdown of species found is provided, although there is still a long way to go with analysing the overall data (Figures 4, 5, 6, 7).

 Figure 3: Number of roadkill detected (Preliminary roadkill study, March 2011, Collinson. W)

Map 1: Arcview map showing all roadkill detected over a 30-day period. (Preliminary roadkill study, March 2011, Collinson. W)


Figure 4: Number of Aves roadkill detected (Preliminary roadkill study, March 2011, Collinson. W)


Figure 5: Number of Ambhibia roadkill detected (Preliminary roadkill study, March 2011, Collinson. W)


Figure 6: Number of Reptilia roadkill detected (Preliminary roadkill study, March 2011, Collinson. W)


Figure 7: Number of Mammalia roadkill detected (Preliminary roadkill study, March 2011, Collinson. W)

Figure 8 shows the total number of roadkill detected over a 30-day period, and then what had been removed by the next day. All roadkill has a ‘shelf-life’ on the road through natural decomposition or scavenging. It is the latter that is of interest, and ‘what’ exactly is removing the fresh roadkill by the next day. On average, 50% of roadkill has disappeared by the next day. This suggests that there may be more roadkill that is going undetected as it is being removed for various reasons, such as muti-use, meat, trophy, or naturally scavenged. A proposed questionnaire is being developed to asess human intervention, whilst a sniffer dog (supplied by Rox Brummer of Green Dogs) will be employed to assist with determining how much roadkill is undetected on the roadside verge. Many roadkill figures do not account for animals hit on the road and then crawl off the road to die.


Figure 8: Total number of roadkill detected over a 30-day period with number of fresh roadkill removed by the next day (February – March 2011)




A huge thank you to Andre Botha, Ian Little, Duncan MacFadyen, Sean Thomas, Dan Parker and Ric Bernard for assisting with the roadkill identifications. Your time and patience was much appreciated ….. And I hope you enjoyed the challenge!


Thank you too, to De Beers Consolidated Mines and E Oppenheimer & Son for their continued support of the project.


Project supported by the Diamond Route.


The Diamond Route is a massive national project which focuses on linking the conservation properties of the Oppenheimer family and De Beers. These properties conserve vast conservation areas and provide a safe haven for a wide variety of unique, rare and ecologically important plants and wildlife and provide endless photograph opportunities. These properties are open to the public whom are encouraged to explore this wealth of tourism opportunities.


Please visit the website or contact Duncan MacFadyen, Manager of Research and Conservation: E Oppenheimer & Son on

3 Comments leave one →
  1. owlmanrsa permalink
    May 23, 2011 8:48 am

    Hey Wendy!

    Our paper on the roadkill between the Kgalagadi and Upington has finally been published. Send me an email so that I have your address and can forward a copy to you…

    Keep up the good work!

  2. CurlyKim permalink
    May 23, 2011 9:08 am

    Interesting arcview map. Someone needs to tell those amphibia to abandon bottom left and take up residence on the rhs of the map!
    You’ve collected a lot of data!

  3. yolandi permalink
    May 24, 2011 9:52 am

    Hi Wendy,

    Thanks for the interesting read!

    The amount of road kill in only a 30 day period really is shocking…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: