Landcare Research - Manaaki Whenua

Landcare-Research -Manaaki Whenua

Are large arboreal invertebrates threatened by possum control?

A common phenomenon in vertebrate pest control is an increase in the number of one pest species when another competing pest species is controlled. A good example of this in New Zealand is the increase in the abundance of rats following the poisoning of possums: in some forests, rat populations recover quickly over several months following poisoning, while possum populations are much slower and take several years to recover. With less competition from possums for seeds and fruit, rats attain densities far in excess of their pre-poisoning levels. What is not well understood is the consequence of this for native biodiversity. Do more rats undo some of the benefits of fewer possums in these forests?

Peter Sweetapple reported negative consequences of this effect for some ground-dwelling invertebrates and robins in issue 11 of Kararehe Kino. Here, Peter and Mandy Barron look at the impacts of increased rat numbers following possum control on selected large-bodied invertebrate species that live in or on forest trees in the Tararua Range. Aerial poisoning of possums was undertaken there for the first time in spring 2010, when the team started monitoring arboreal invertebrate populations by extracting tree wētā, stick insect and slug faecal material (frass), and cockroach egg cases, from litter trays (Photos a–d). Samples were taken both inside and outside the poisoned areas at two widely separated sites (Tōtara Flat and Waitewaewae) over the November–February period for 5 years starting November 2010. These collections spanned a second poison operation in December 2013. Rats were monitored using tracking tunnels over the same 5-year period.

Trends in rat and invertebrate abundance

When first measured after the first poison operation, the abundance of rats (revealed from their faeces) was low throughout both sites and near zero in poisoned areas (Figure a). Tracking tunnel indices showed that rat numbers rose to high levels in the poisoned areas (relative to unpoisoned areas) over the first 2 years, and that this was more pronounced at Tōtara Flat.  Rat numbers were high again, particularly within the poisoned areas, 1 year after the second possum control operation.

Throughout much of the study, cockroaches, stick Insects and tree wētā at Tōtara Flat all exhibited abundance trends that were the opposite of those for rats. Their abundances were initially similar inside and outside the poisoned area, but then they increased outside to peak about three times higher than inside the poisoned area (Figure b, c, d).  However, the initial trends in tree wētā frass fall at Tōtara Flat did not persist beyond 2013, for unknown reasons. Abundances of all three invertebrates was generally much lower at Waitewaewae than at Tōtara Flat (stick insects were very scarce at Waitewaewae), and there were virtually no differences in abundance trends with and without the 1080 poisoning of possums.

Slugs, on the other hand, did not respond to pest control at Tōtara Flat but showed complex abundance trends at Waitewaewae, which may have been unrelated to pest abundance and control (Figure e).

Overall, there were marked differences between the two sites in terms of the responses of rats and arboreal invertebrates to poisoning. Rats and invertebrates were generally less abundant at Waitewaewae, which probably reflects the wetter climate and less diverse, beech-dominated vegetation there. They were generally more abundant at Tōtara Flat, where beech was a minor part of the forest but fruit- and seed-producing species (including hīnau, rimu, supplejack and pigeonwood) were common. As seen elsewhere, the ‘release’ of rat populations after the removal of possums appears largely restricted to food-rich mixed podocarp−hardwood forests (the Tōtara Flat site in this study) that have high rat- and possum-carrying capacities. In these forests, sustained possum-only control does appear to have negative consequences for tree-dwelling invertebrates.

This work was funded by the Ministry of Business, Innovation and Employment (contract C09X0909).

Peter Sweetapple
sweetapplep@landcareresearch.co.nz

Mandy Barron