Publication details

Prevention of wildlife-vehicle collisions: An evaluation of odour repellent effectiveness

Authors

BÍL Michal ANDRÁŠIK Richard BARTONIČKA Tomáš KŘIVÁNKOVÁ Z SEDONÍK Jiří

Year of publication 2018
Type Conference abstract
MU Faculty or unit

Faculty of Science

Citation
Description Wildlife-vehicle collisions (WVC) pose a safety issue in many nations? Collisions with large mammals result in numerous deaths and severe injuries to animals, property damage and injuries to car passen-gers? Therefore, several safety measures have been recommended to decrease the number of WVC? The best results (up to 83% of WVC decrease) were accomplished by the use of fencing combined with over- or underpasses? On the other hand, fencing is only recom- mended for highways and primary roads due to their overall costs? Affordable measures, such as odour repellents, can be considered as an appropriate alter- native for secondary roads? We examined the effectiveness of odour repellents concerning the reduction of WVC? The Before-After-Control-Impact study design was applied to con-trol both the effect of odour repellents and the expected natural variation in wildlife populations at the monitored sites over time? Eighteen places were selected on the Czech road network where WVC were frequent in previous years on the basis of traffic crash data? Control places were delimited on the same road segments to keep the exposition to the traffic the same? We compared the WVC fre-quencies before and after odour repellent installation at treated sites to the WVC frequencies at con-trol sites? Wildlife carcass gathering was carried out once per week during spring and autumn from 2014 to 2016? We also used the police crash database to supplement carcass data? In total, 201 killed mammals (roe deer, red deer and wild boar) were identified over 47 months (January 2013 – Novem-ber 2016)? First, we analysed the data using a standard technique based on the comparison group method? However, this conventional approach is suitable only for sufficiently large datasets? Low frequencies usually lead to insignificant results? Therefore, we also applied an approach based on the Bayesian inference? We calculated the probability that the odour repellents have a positive effect on the number of WVC? We found out that a WVC decrease accounting for 26-43% can be expected at treated sites? The odour repellents can be used as affordable measures to reduce the number of WVC? Some specific requirements have to be met, however, to ensure that the repellents work properly (installation, rejuvenation, etc