Landscape disturbance by roads may increase abundance of prey in verges 20 (i.e., strips of terrain adjacent to roadways) or create other features that can attract carnivores 21 and expose them to a higher risk of mortality by vehicle collision. We studied a system that 36 linear transects within the study area, 18 along motorways and 18 in control zones ( Fig. 1). 119We defined control zones as those with the same habitat structure and ≥4 km from habitat effect. We double-checked transects for habitat similarity; first we inspected habitat . We calculated the index value as the sum of all pellets counted in each transect. 142We estimated a carnivore abundance index as the number of scats detected in each 143 transect (Long et al. 2008 of the raw data in each transect type followed the same pattern for almost all the taxa, with 240higher mean values in motorways than in control sites (Fig. 2). Global carnivore abundance 241 increased with rabbit abundance (β = 0.103 ± 0.044, P = 0.019) and carnivore abundance was 242 higher in motorway transects than control sites (β = 0.400 ± 0.175, P = 0.022). 243The red fox showed higher abundance in motorways (β = 0.469 ± 0.197, P = 0.017) 244 but no significant response to rabbit abundance (P = 0.129; Fig. 3). The non-synanthropic 245 species showed the opposite response, with abundances positively related to rabbit abundance 246(β = 0.119 ± 0.044, P = 0.007) but no differences in abundance due to the proximity of the 247 motorway (P = 0.084; Fig. 3). 248Cascading Effects: Rabbit Abundance, Carnivore Abundance, and Roadkills 255In the analyses for the global carnivore community, roadkills were positively and numbers. For carnivores related closely to prey abundance, high prey abundance in verges 275 creates a local cascading effect that leads to carnivore roadkills. 276The increased abundance of the global carnivore community near motorways is 277 probably due to higher fox abundance in these areas, as foxes comprised almost half of the 295On the other hand, our group of non-synanthropic carnivores followed a different carnivores that prey on them will follow rabbit distribution. we controlled for by conducting DNA analysis to assess the accuracy of scat identification. 332We controlled for other potential sources of bias by choosing areas that were comparable in 333 habitat characteristics, which can affect fecal persistence and scat detectability (Long et al. are necessary to make informed management decisions that favor conservation objectives. 364Understanding wildlife responses to human-modified landscapes provides us with better 365 knowledge of the communities and identifies which processes should be the focus of our 366 conservation efforts. 367 MANAGEMENT IMPLICATIONS 368When prey are abundant near motorways, carnivores increase their abundance in those areas. 369Higher abundances of carnivores were strongly related to higher mortality by vehicles. 370Therefore, the design and management of verges could be a key aspect for conservation of 371 wil...

No abstract