6 Hypothetical result when evaluating the effectiveness of road mitigation measures at a new road with mitigation. The new road with mitigation is constructed at time zero. In addition to the mitigation site, measurements are carried out—before and after road construction—at a no-mitigation control site and a no-road control site. Generally, there are four possible scenarios 1 the road mitigation measures are 100 % effective and population density remains at the level of the no-road control site, 2 the road mitigation measures are only partly effective #GSK2118436 randurls[1|1|,|CHEM1|]# and population density decreases compared to the no-road control site but does not reach the level of the no-mitigation

ACP-196 control site, 3 the road mitigation measures are not effective and population

density decreases to the level of the no-mitigation control site, 4 the road mitigation measures worsen the situation and population density decreases below the level of the no-mitigation control site All control sites need to be far enough away from the mitigation sites and each other to ensure statistical independence, yet still be as similar as possible. If possible, control sites should be sited along the same road as the mitigation site(s), as road age, design and traffic characteristics of the same section of road are probably similar. Such control sites should never immediately border the mitigation site(s), as possible edge effects of mitigation measures, e.g., an unnaturally high number of road-kill just at the end of the wildlife fencing, may influence the measurements. Select appropriate find more spatial scale of study Two factors need to be considered when determining the spatial scale of a study. First, the spatial scale of the study should match the spatial scale of the effect being mitigated. Stipulating a one-size fits all approach to determine the spatial scale of the study is not possible because the size of the road effect zone (Forman and Deblinger 2000; Forman et al. 2003) varies depending on the

effect, the species of concern, and the local situation (e.g., habitat type, topography). Second, the sampling effort should be apportioned equally across the road effect zone, as the road effect of concern may vary significantly within this zone. The effect-size of the road—and consequently the effect-size of road mitigation measures—will be often at its maximum in close proximity to the road. However, situations occur where the opposite is true, e.g., due to an increase in suitable edge habitat at the roadside (Mumme et al. 2000) or due to home range pile-up adjacent to the road due to barrier effects (Riley et al. 2006). It is often necessary to do a best guess about where the road effect zone ends.