Response monitoring and cost–benefit analysis drives adaptive management of critically endangered skinks
Grand skink g03009 after an encounter with a stoat. Direct evidence of predation by cats has also been observed. Image – Nathan Whitmore.
The Department of Conservation (DOC) has over the last six years expanded its protection of critically endangered grand and Otago skinks from an experimental scale to an area providing medium-term security for both species. This has been made possible by accurate monitoring of populations of skinks to detect trends under different management regimes. Once effective management options were identified, cost–benefit analysis and metapopulation modelling were used to plan the extended control area.
Grand and Otago skinks (GAOS) are two of New Zealand’s largest and most threatened lizards. Both are endemic to Otago and closely tied to deeply-incised schist rock tors, which they use for thermoregulation and refuge from threats. Both are now restricted to a small fraction of their historical range. In 2005, DOC’s GAOS Recovery Programme (involving Andy Hutcheon and colleagues) embarked on a 5-year science-by-management experiment at Macraes Flat to test the hypothesis that introduced mammalian predators were a cause of the decline of both species. All mammals were removed from a site enclosed by mammal-proof fencing, while other sites were located at the core, periphery and outside an intensive trapping operation covering 2100 ha and targeting weasels, stoats, ferrets, cats and hedgehogs.
 |
 |
 |
| Otago skinks born within DOC’s mammal-proof fence. Image – James Reardon | A grand skink basking on rock habitat. Image – James Reardon | Typical GAOS habitat: deeply incised schist rock in dry tussock country. Image – James Reardon |
By the end of the third monitoring season of the 5-year multi-treatment trial, the combination of the rapid response by both species of skinks and tight confidence intervals on their abundance estimates made it clear that both treatments worked: skink numbers increased both inside mammal-proof fences and when protected by a 1.5-km buffer of trapping (Fig. 1). With two tools demonstrating effectiveness, the next adaptive management step was to protect a larger expanse of skink habitat.
Cost–benefit analysis showed that, for areas greater than 10 ha, trapping gave the best return on dollars invested (Fig. 2). This analysis took account of the capital charge and depreciation costs associated with fencing when compared with the otherwise higher annual running costs of extensive trapping. The non-homogeneous nature of skink habitat at Macraes Flat was captured in a metapopulation model that both guided the selection of areas for testing the two protection approaches and identified opportunities to use local translocations to kick-start grand skink populations in newly protected habitat.
Between 2008 and 2010, trapping at Macraes Flat more than doubled in extent, and now covers 4600 ha and uses a thousand traps in a mix of six trap types and three different baits. Drop in catch rate of the various predator species from the edge towards the centre of the trapping operation is presumed to correlate with a reduction in predator densities and explains why the populations of skinks are able to recover under this control regime.
Good design and robust monitoring has demonstrated the effectiveness of both management systems. This, in conjunction with modelling, supported the decision to extend protection across populations at Macraes Flat. To complete the good-news story for these two iconic species, Andy and his team need new tools to effectively control invasive predators over a landscape scale and allow service frequencies of traps in months rather than days or weeks. Such innovations would reduce costs and allow protection of skinks in places that are less accessible.
This work was funded by the Department of Conservation and the Ministry of Science and Innovation (Programme C09X0505).
Andy Hutcheon
Contact Andy: ahutcheon@doc.govt.nz
Grant Norbury
Nathan Whitmore (DOC)
Fig. 1. Monitoring results of Otago skinks. Grand skink monitoring tells a similar story, but with a catastrophic decline at the non-treatment site. Unmanaged populations appear to exhibit stability punctuated by episodic decline from which they do not recover. Error terms are +-95%CL.
Fig. 2. Cost per unit area protected for the two methods shown to be effective at Macraes Flat (open points indicate trapping, orange points indicate fencing) for areas of <45 ha (left) and up to 1200 ha (right). For small areas (left), the buffering needed with trapping makes fencing more appropriate. Note different scales on the y axis.
