Abstract:
Methods for improving the protection of kakapo nests from kiore predation were investigated during February-May 1996 on Codfish Island. Nest protection protocols are critical to help boost the current kakapo population, which at present is comprised of less than fifty birds.
The protection of kakapo nests involves only short-term, localised management of kiore. The critical state of the kakapo population limits methods of rodent removal to techniques which pose no threat to nesting females (or chicks). Elliot live-capture traps are therefore considered to be the only safe option for kiore management in the vicinity of kakapo nests. The success of a trapping protocol is dependent on trap spacing and location. In the present study high density trapping significantly increased the number of rats captured over a small area. Similarly, the location of traps and associated vegetation, significantly influenced the probability of a rat encountering a trap. However the design of the standard Elliot trap is not condusive to kiore entry and methods of increasing Elliot trap captures were consequently investigated.
The use of Elliot traps which had previously been successful at catching kiore did not significantly influence the probability of future captures, although there was a slight trend for dirty traps to capture more rats. Modifying the standard Elliot trap to a mesh-ended design, significantly increased the number of kiore captured, further emphasising the inadequacy of Elliot traps at capturing kiore for the purposes of localised rodent control. Observation of behavioural interactions between kiore at a localised feeding source revealed conflict is increased between rats as rat density increases. Conflict as a result of high rat densities may potentially deter entry into traps and reduce capture rate in a local area.
A comparison of standard Elliot trap and Ezeset kill-trap captures, revealed Elliot traps capture rats of a lighter weight than Ezesets and may therefore be more accurate at revealing the presence of juveniles within the population. Conversely, Ezeset traps captured more adults and may thus be more effective at removing groups within the population which pose the greatest threat as potential predators (i.e adults).
Food abundance in an area also influences the characteristics of kiore captured and may therefore influence the capture rate of a particular trapping regime. Dampened population oscillations were observed in areas in which food was abundant year round, and rats were typically heavier in weight than those from areas in which food was seasonally abundant. In contrast, the populations of rats captured in areas in which food was seasonally abundant, were slightly lighter in weight, and experienced breeding activity which appeared to be concentrated over a shorter period.
Based on the findings of this study, the importance of tailoring a protocol for rodent control to the aims of the programme, to the characteristics of the species under protection, and to the characteristics of the target species, is emphasised. Further research into the behavioural characteristics of kiore will further facilitate more effective means of removal. However further research is also required into the effect of kiore on native flora and fauna particularly when providing justification for kiore control and removal.
