• amphibian movement;
  • animal behaviour;
  • dispersal;
  • road ecology;
  • road experiment;
  • wildlife underpass


Many amphibians in Europe and North America regularly use under-road tunnels during dispersal. However, such structures have not been shown to mitigate the impacts of road mortality on frog populations in Australia. We tested the behavioural response of three Australian frog species to a 12-m amphibian under-road tunnel in controlled ex situ conditions. In April and October 2012, we tested 34 striped marsh frogs Limnodynastes peronii, 54 green and golden bell frogs Litoria aurea, and 15 broad-palmed frogs Lit. latopalmata. The overall proportion of frogs entering the tunnel (tunnel usage), and entering and exiting the tunnel at the opposite end (tunnel efficiency), among the three species was 0.13 and 0.05, respectively. The proportion of usage for Lim. peronii, Lit. aurea and Lit. latopalmata was 0.21, 0.07 and 0.13, respectively. The mean time taken for an individual to enter the tunnel was 14 min 22 s. There was some evidence of directionality in the movement of Lim. peronii and Lit. aurea within the experimental arena. Tunnel usage was not likely related to air temperature, humidity or light levels inside the tunnel, but requires further investigation. Our study showed that there was low usage of the under-road tunnel compared with the relatively high usage rates of tunnels by amphibians in the Northern Hemisphere. Our results demonstrate that the effectiveness of wildlife tunnels in mitigating the impact of roads on frog species in Australia and tropical regions requires further investigation. We recommend testing these tunnels once installed under several roads near wetlands to assess how frogs respond to them during the breeding period.