Last term, the decision to take up the ASEAN Environments elective was a no brainer to me. Being someone from the Southeast Asian region and most probably returning to Malaysia for a colourful career (fingers crossed) in conservation, it was the most natural choice. Choosing the Hilary term elective however was more of a challenge. I was torn between going for a technical course in the form Analytical Geographic Information System (GIS) techniques or more electives that would be similar to the other modules – involving the reading of a few journals followed by weekly classes.
In the end, along with a few other brave BCMers I chose the former. On the first day of class, our elective leader Rob made it very clear to us that this was a self-taught course of which he will pop in on a weekly basis to help out. We are to decide on an individual topic and focus our energy in the coming weeks to cough up a respectable 4,000 word elective essay complete with ‘pretty maps’. Yes, he said pretty maps. That left us all with a problem – how could we with barely any knowledge of what GIS could do, settle on come up with a pertinent research question let alone a mini research topic?
The answer was to learn on the job. And today, 6 weeks into the course and 2 binned ideas (I’ll be more than willing to share about my failures) later I would like to think I have pinned down a potential topic!
Since we are only doing a short term project, it is almost impossible to do manual data collecting so most of us are relying on bits and pieces of data we can muster from the miracle known as the internet. We literally spend hours scouring the far reaches of the world wide web (just to add a little drama) for usable data. Among the few popular sites include WorldClim (climate data), GBIF.org (biodiversity data) and Magic (most environmental data in the UK). Trust me when I say this – data is your best friend when it comes to GIS.
Coming back to my topic, I initially wanted to map out changes of seahorse distribution in the Johor Straits and relate it to seagrass coverage obtained from satellite imagery. The hopes of doing so were crushed by the low resolution and data limitation of satellites since:
- Data is in 30m resolutions when my area of interest was only about 150m long and
- The seagrass bed I was looking at is fully submerged at low tide.
It was a long shot to begin with.
Moving on, at that time, the report of a wild tigress being run over by a Multi Purpose Vehicle (MPV) made the headlines in Malaysia. It gained a huge social media following as well. This was since the tigress was found to pregnant with two cubs at the time of death. The incident was a huge loss to the already dwindling tiger population in Malaysia. It was then that I mulled the possibility of doing a project on roadkill.
Image taken from online news portal FreeMalaysia Today (http://www.freemalaysiatoday.com/category/nation/2016/02/07/tigress-killed-in-road-accident-was-a-mother-to-be/)
After some soul searching followed by some actual internet searching, I was surprised that there was an existing citizen science database for roadkills known as Project Splatter run by Dr Sarah Perkins of Cardiff University. The project allows people from all around UK to record basic details of observed roadkill including. According to the Project Splatter wordpress blog, the data collected will be used to estimate the impact of roads on UK wildlife. A few emails later, Dr Sarah agreed to provide me with the actual database to work on. Bingo! A first breakthrough in weeks.
Getting the data is one thing, knowing what to do with it is another. As of now, I’m trying to map out roadkill prone roads in the UK and try to ground truth it with Project Splatter data. This is in hopes to identify areas in the UK where no observations were recorded there from the lack of citizen scientists. Easier said than done of course.
Mapped out distribution of roadkill by Project Splatter
From my readings, different taxa are affected by different conditions. For example, Rosa and Bager (2012) identified that bird roadkills in summer can be attributed to increase traffic from transport of grain and vacationers in Brazil. Aside from this, Farmer and Brooks (2012) identified 8 risk factors for vertebrate roadkills in Southern Ontario of which the main 4 were distance from wetlands, habitat diversity, maximum daily temperature and posted road limit.
Admittedly, the data collected data might have some potential biases. Ratton et al., 2014 cautioned about the carcass permanency and removal rates. The idea is that roadkill only persists for a short amount of time and this could affect data quality. Another obvious factor to consider is reporting bias. Some areas might have more enthusiastic or diligent citizen scientists than others resulting in a huge data skew. In the Project Splatter map above, you can observe that England has significantly more data points than Scotland.
Once my project is completed, potential mitigation measures could be put in place to reduce the occurrence of roadkills in the UK. Magnus et al. (2014) suggested mitigation via animal behaviour change from the use of ultrasonic whistles as deterrents. The effectiveness of this technique was proven to be low. A more realistic measure that could be put in place was the placement of road signs to stimulate human behavioural change (Coulson, 1982, Dique et al., 2003). Another measure suggested by Seiler (2005) was to increase roadside mowing leading to an increased visibility of which would potentially reduce roadkills.
In the weeks leading to submission, it is highly likely that I would be spending inordinate amounts of time in the computer lab at the department struggling with the beast that is GIS. Wish me luck!
Do also contribute to the amazing citizen science database that is Project Splatter by reporting roadkill or follow them on twitter @ProjectSplatter for more information.
(Flickr creative commons, Username: Aaron)
Coulson, GM (1982) . Road-Kills of Macropds on a Section of Highway in Central Victoria. Australian Wildlife Research 9 , 21–26.
Dique, D. S., Thompson, J., Preece, H. J., Penfold, G. C., De Villiers, D. L., & Leslie, R. S. (2003). Koala mortality on roads in south-east Queensland: The koala speed-zone trial. Wildlife Research, 30(4), 419–426. doi:10.1071/WR02029
Farmer, R. G., & Brooks, R. J. (2012). Integrated risk factors for vertebrate roadkill in southern Ontario. Journal of Wildlife Management, 76(6), 1215–1224. doi:10.1002/jwmg.358
Magnus, Z., Kriwoken, L. K., Mooney, N. J., & Jones, M. E. (2004). Reducing The Incidence Of Wildlife Roadkill : Improving the visitor experience in Tasmania, 42. Retrieved from http://www.crctourism.com.au/wms/upload/resources/bookshop/Wildlife_RoadKillFINAL.pdf
Ratton, P., Secco, H., & da Rosa, C. A. (2014). Carcass permanency time and its implications to the roadkill data. European Journal of Wildlife Research, 60(3), 543–546. doi:10.1007/s10344-014-0798-z
Rosa, C. A. da, & Bager, A. (2012). Seasonality and habitat types affect roadkill of neotropical birds. Journal of Environmental Management, 97(1), 1–5. doi:10.1016/j.jenvman.2011.11.004
Seiler, A. (2005). Predicting locations of moose-vehicle collisions in Sweden. Journal of Applied Ecology, 42(2), 371–382. doi:10.1111/j.1365-2664.2005.01013.x