Assessing proposed designs for a non-invasive DNA collection device suitable for use in possum (Trichosurus vulpecula) population monitoring

Abstract

The Australian bushtail possum (Trichosurus vulpecula) is an introduced mammalian pest species within New Zealand ecosystems, responsible for causing extensive damage to native flora and fauna at the same time as transmitting bovine tuberculosis between cattle herds. Due to their impact on environments and economies, they are intensively controlled using a variety of methods including aerial toxin delivery and trapping, led by the Department of Conservation and TBfree NZ. Along with control, wildlife managers monitor population trends across the country using an array of monitoring tools developed for various scenarios. One major need for those monitoring T. vulpecula populations is the ability to identify individual animals, to harness an idea of total population abundance and disease prevalence. Non-invasive genetic monitoring from saliva DNA samples taken from WaxTags® has proved promising in the past, however environmental exposure degrades DNA quality making individual identification difficult. This report describes laboratory, pen and field trials aimed at designing a device capable of collecting and preserving DNA samples from possum saliva. A DNA trial investigated the effect of environmental exposure on non-invasively collected saliva samples. After 14 days of weathering, individual possum identifications were at 80% and 20% for covered and uncovered DNA samples respectively. It was found protection following collection significantly increased the ability to genotype saliva based DNA (p = 0.033). Pen trials on captive animals developed prototypes of prospective devices, testing for initial encounter behaviours, trigger rates, ability of animals to re-access triggered baiters and durability throughout possum interactions. Multilocation field trials allowed for comments to be made on the DNA devices ability to calculate population abundances and densities when compared with BMI and RTCI estimates. Other field trial results allow for comments to be made on the devices performance in a realistic environment pertaining to ease of use, non-target interactions and device sensitivity. Finally, a Landcare Research field trial collected 122 saliva samples using the devices described throughout this report, leading to the identifcation of 17 individual possums in a regenerating mixed podocarp forest plot, with a 68% genotyping success rate. Ear notch analysis of 22 subsequently detained individuals revealed 7 matches to the DNA device collections, meaning 10 animals were not caught using leghold traps and 15 animals were not surveyed using DNA devices. This leaves the recommendation that future monitoring operators wishing to accurately calculate population abundance should implement a combination of both live trapping and DNA analysis to complement one another. Finally, future research options are discussed including improved collection media and increasing device sensitivity.