Abundance and distribution of possums
Background
This measure relates to indicator 1.3.2 Invasive species dominance.
Possums are widespread across Public Conservation Land (PCL) in New Zealand and can have significant impacts on native flora and fauna through competition, predation and browsing on leaves, fruits and flowers (see pests and threats for more information). Therefore, understanding the national distribution and abundance of possums can help to direct management efforts by the Department of Conservation (DOC) and others to protect New Zealand’s biodiversity. These data are also a baseline to compare with future trends and/or management results.
Possum numbers are stable in forest and scrub but slightly increasing in non-woody habitats.
What did we measure?
DOC has developed a national monitoring programme to assess the status and trends of biodiversity using approximately 1,400 sites that are spaced evenly across PCL (Figure 1 - Map). A random sample of approximately 280 of these sites are measured each year, so that every site is measured once on a five-year rotation (Table 1). Possum abundance is measured at each site, and data are also collected about vegetation, birds, and pellets of other introduced mammals.
Abundance of possums at each site is measured on four transect lines, each with ten chew cards set out for one night. Before 2015, measurements used trap (TCI) or wax tag (WTI) indices. To show long term trends, these are calibrated to chew card indices (CCI). Occupancy is the proportion of sites where possums occur, accounting for the likelihood of detecting a possum if it is present and is estimated using these indices and data from four faecal pellet lines.
Data (chew card index and faecal pellet presence/absence) were modelled with a Bayesian zero-inflated negative binomial model to show the effects of habitat type (woody or non-woody), land status (national park or other PCL), and year on possum occupancy (positive indices and/or presence of pellets) and abundance (calibrated CCI), accounting for underlying differences between sites. This model is suitable for data with many zero-observations and allows for incomplete detection (when possums do occupy the site, but this is not observed).
What did we find?
- Overall habitat and conservation land types, occupancy was 60.6% (95% credible interval (CI) = 57.8 - 63.4%) in the 2020/21 season.
- Abundance and occupancy were slightly increasing in non-woody habitats (Figure 2).
- Both occupancy and abundance were lower in national parks than in other PCL (Figure 3).
- Average abundance indices for individual parks are shown in Figure (4).
Figures
Figure 1: Mean relative abundance of possums, measured with CCI, from the most recent measurement at sites on PCL over nine seasons from 2012/13 to 2020/21. ‘Present/Absent’ outlines sites where possums were detected (by faecal pellet counts and/or any of the indices) with a black line around the circles. Plot locations have been randomly jittered.
Figure 2: Model estimates of possum abundance using calibrated chew card indices (CCIs) on public conservation land (PCL) over the last nine seasons from 2012/13 to 2020/21. ‘Occupied CCI’ represents the trend in possum-occupied areas, whereas ‘PCL CCI’ represents the overall trend on PCL. Different habitat types (woody, non-woody) and land status (national park, other PCL) are represented by different line types and colours.
Figure 3: Model estimates of possum occupancy on PCL over nine seasons from 2012/13 to 2020/21. Different habitat types (woody, non-woody) and land status (national park, other PCL) are represented by different line types and colours.
Figure 4: Observed average possum abundance indices at each park. Choose one or more parks in the selection box at the top and hover over an individual point to see the details. Values are mean ± 1 standard error for each index: chew card (CCI), wax tag (WTI), and trap catch (TCI). Some outliers are not visible but can be seen using the tools in the top right of the figure.
Table
| Vegetation class | Conservation Status | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | Total plots |
|---|---|---|---|---|---|---|---|---|---|---|---|
| non-woody | national park | 21 | 18 | 24 | 22 | 17 | 25 | 22 | 22 | 22 | 112 |
| non-woody | other PCL | 43 | 57 | 45 | 56 | 59 | 59 | 59 | 45 | 56 | 286 |
| woody | national park | 8 | 54 | 73 | 72 | 60 | 65 | 53 | 64 | 70 | 323 |
| woody | other PCL | 22 | 149 | 126 | 115 | 120 | 123 | 141 | 126 | 113 | 642 |
| annual total | 94 | 278 | 268 | 265 | 256 | 272 | 275 | 257 | 261 | 1363 |
Data quality
This measure is classified as a partial measure of high accuracy and complies with the data quality guidelines used in the Environmental Reporting framework.
Glossary of terms
Abundance is the number of animals that occupy an area. Here, we use indices of relative abundance, estimated from the number of devices (cards or traps) that show evidence of animals.
Chew cards are small rectangular plastic sheets that have attractant paste pressed into their internal channels. A chew card index (CCI) is calculated by deploying up to 40 chew cards along four evenly spaced 200-m long transect lines (10 chew cards per line at 20-m intervals). An estimate of possum abundance is then generated by calculating the percentage of chew card interference (tooth impressions or scratches).
Credible interval (CI): A 95% CI indicates that the true mean lies inside the interval with 95% probability given the posterior probability distribution.
Occupancy indicates whether a site is being used by a species. A model is used to estimate the proportion of sites at which the species occurs, adjusted for the probability of detection.
Trap catch indices (TCIs) are used to measure possum abundance. A TCI is calculated by deploying up to 40 leg-hold possum traps along four evenly spaced 200-m transect lines (10 traps per line at 20-m intervals). The TCI is defined as the percentage of trap-nights on which a possum was captured. This information can be used to infer the presence of possums and their approximate abundance within an area.
Wax tags are scented, ice-cube sized wax blocks and are used to estimate possum abundance. A wax tag index (WTI) is calculated by deploying up to 40 wax tags along four evenly spaced 200-m transect lines (10 wax tags per line at 20-m intervals). The percentage of wax tag interference (tooth impressions or scratches) is then used to estimate possum abundance.
Additional resources
DOC’s Outcome Monitoring Framework: https://www.doc.govt.nz/our-work/outcome-monitoring-framework
Dorazio, R.M., Royle, A.J., 2005. Estimating size and composition of biological communities by modeling the occurrence of species. Journal of the American Statistical Association 100, 389–398.
Forsyth, D.M., Perry, M., Moloney, P., McKay, M., Gormley, A.M., Warburton, B., Sweetapple, P., Dewhurst, R., 2015. Calibrating brushtail possum (Trichosurus vulpecula) abundance estimates in DOC’s Biodiversity Monitoring and Reporting System: wax tags, chew cards and leg-hold traps (Unpublished client report for the Science and Policy Group, Department of Conservation). Arthur Rylah Institute for Environmental Research Department of Environment, Land, Water; Planning.
Gormley, A.M., Forsyth, D.M., Wright, E.F., Lyall, J., Elliott, M., Martini, M., Kappers, B., Perry, M., McKay, M., 2015. Cost-effective large-scale occupancy–abundance monitoring of invasive brushtail possums (Trichosurus vulpecula) on New Zealand’s Public Conservation Land. PloS one 10, e0127693.
Lee, W., McGlone, M., Wright, E., 2005. Biodiversity inventory and monitoring: A review of national and international systems and a proposed framework for future biodiversity monitoring by the Department of Conservation. Landcare Research Contract Report LC0405/122 (unpublished) for the Department of Conservation, Wellington.
McGlone, M.S., McNutt, K., Richardson, S.J., Bellingham, P.J., Wright, E.F., 2020. Biodiversity monitoring, ecological integrity, and the design of the New Zealand biodiversity assessment framework. New Zealand Journal of Ecology 44, 3411.