There’s no accounting for taste: bait attractants and infrared digital cameras for detecting small to medium ground-dwelling mammals

Context Reliable information about the occurrence and distribution of threatened forest-dwelling mammals is critical for developing effective conservation plans. To optimise limited resources, advances need to be made to the toolkit available for detecting rare and cryptic fauna. Aims We trialled three bait attractants (peanut butter with oats, live mealworms and black truffle oil) in combination with infrared digital cameras to determine whether detection rates of forest-dwelling native mammals in south-eastern Australia were influenced by: (1) bait type; (2) previous visits by conspecifics; (3) previous visits by Rattus; and (4) duration of bait deployment. Methods Bait attractants were set at 40 camera stations in combination with odourless controls. Over two fortnight-long deployments, 1327 images were captured of 22 mammal and bird species. From these data, detailed statistical analyses were conducted of six mammal genera. Key results Peanut butter with oats was found to be a significantly better attractant than empty bait holders for Antechinus, Isoodon, Perameles and Rattus, but not for Potorous or Pseudocheirus. Truffle oil and mealworms were also significantly better attractants than the control for Rattus but not the other five genera. When Antechinus, Isoodon, Potorous or Rattus were detected at a bait station there was a significant likelihood they had been detected there during the previous 24 h. This was not the case for Perameles or Pseudocheirus. A prior visit by Rattus to a station had no significant influence on the detection probabilities of Antechinus, Isoodon, Perameles, Potorous and Pseudocheirus during the subsequent 24 h. Detection probabilities for Isoodon and Rattus declined significantly during the fortnight-long deployments but trends for the other genera were not significant. Conclusions Peanut butter with oats is an excellent general purpose bait for detecting small to medium-sized mammals. However, scope exists for using other baits to target species. For example, truffle oil baits may reduce by-catch of non-target Rattus in labour intensive cage trapping of bandicoots. Regardless of bait type, longer deployments are necessary to detect Perameles, Potorous or Pseudocheirus than Antechinus, Isoodon or Rattus. Implications Targeted detection of predominantly ground-dwelling mammals may be improved by better understanding the attraction of species to baits and required bait deployment times.