Quantifying terrestrial mammal responses to roads

 Roads cover a significant portion of the Earth’s land surface, fragmenting landscapes and intersecting many key biodiversity areas. While pivotal for human mobility, roads restrict mammal movement, potentially affecting their fitness and abundance. Given the impact and predicted expansion of roads, it is important to understand how mammals respond to roads.

While some studies have investigated species’ responses to roads, they have primarily focused on a small selection of species or locations, limiting our general understanding of road impacts. To address this gap, we used tracking data of over 6000 individuals across 64 mammal species (ranging from 1.5 kg to 4500 kg) on all continents to quantify road avoidance and crossing probability. Due to the global scope of the study, we defined these metrics in such a way that they are easy to understand, fast to calculate and scalable, while being good indicators of road-related behaviour. 

We explored two ways to use these crossing and avoidance metrics to understand road impacts on mammal behaviour. We first compared them to road-crossing and avoidance rates from randomly generated tracks, to gain insight into whether roads are altering animal movement. Secondly, we examined how these behaviours relate to species traits, road characteristics and environmental factors. Our initial analyses suggest that road crossing probabilities across all species range from 1 % (highways) to 5% (smaller roads). We also found that crossing rates are mainly related to road type, road density, species’ body mass and canopy cover. Besides differences between species, there is also considerable individual variation (18% of explained variance). 

Based on our findings, road crossing and road avoidance probabilities are potential metrics to link movement ecology to biodiversity monitoring. By identifying general patterns and predictors of road impacts across species and landscapes, our research contributes to a more systematic understanding of mammal-road interactions and supports the development of strategies to reduce ecological disruption from infrastructure.