In highly human modified landscapes, roads can be a major driver of species decline. They fragment the landscapes, increase mortality rates, and change animal behaviour, not to mention the potential for harm to humans from collisions with wildlife! So, understanding the impact of of roads on wildlife has been a significant preocupation of many ecologists for quite some time. However, far less attention has been paid to identifying which strategies for minimising impacts of roads on wildlife are most cost effective. In two new papers from our group (Rhodes et al. 2014. Plos One & Polak et al. 2014. Journal of Applied Ecology) we address this issue based on two different strategies for limiting impacts on wildlife: (1) by designing the road network to minimise mortality impacts, and (2) by locationing overpasses and fencing on existing road networks to limit wildlife access to road surfaces and minimise mortality impacts. In each case we consider a constraint; in terms of road traffic volumes in the former and in terms of the cost of mitigation measures in the latter.
Rhodes et al. (2014) tackle the question of whether future increases in traffic volume are better acommodated through upgrades on exisiting roads to increase capapcity, or to build new roads when the aim is to minimise impacts on wildlife mortality. They use a realistically parametersied model of koala movement and assess impacts on mortality rates from crossing roads under these two scenarios. Interestingly they show that under the vast majority of different parameterisations of the model, it was preferable to increase capacity on exisiting roads rather than build new roads (i.e., bigger roads are generally better than more roads). Only when the existing road density was very low and traffic volumes on roads very high was it preferable to build new roads. Therefore, that we should preferably increase capacity on existing roads rather than building new roads appears to be a relatively robust rule of thumb for many realsitic situations.
Polak at el. (2014) aslo use the koala as a case study species, but tackle the problem of where to build fencing and road crossing to maximise population abundance. They develop a new method to achieve this using ideas from the decision sciences and show that, in their case, the trade-off curves between costs and population size are almost linear. The implication is that there there are no options to achieve large gains in population size for low cost. Nonetheless, the approach provides a novel framework for identifying priorities for road retrofitting that is explicit about costs and benefits and this has not previously been possible with existing approaches.