Abstract
The spatial ecology of Stewart Island weka (Galliralluis australis scotti) was compared between a seabird island (Taukihepa/Big South Cape Island) and a predator-free non-seabird island (Ulva Island). Both islands were situated off Stewart Island in southern New Zealand. Differences in the spatial ecology of weka can be expected to occur between islands due to the recognised increases in productivity of seabird islands and the corresponding differences in ecology of animals high up the food chain, hence weka on the seabird island were expected to have smaller home ranges. Comparisons between weka home range, body size, social structure and 3rd order habitat selection were made between islands, after radio-tracking 11 and 10 weka on a seabird and non-seabird island respectively between January and May 2005. The results were contrary to those expected for a highly productive seabird island. This most likely occurred due to a reduction of male weka on the seabird island.
The mean home range size of weka did not differ between the two islands (Taukihepa: 3.91 ± 0.83 ha, Ulva: 2.54 ± 0.50 ha) and was similar to weka home range sizes on other islands. There was no significant difference in the home range size between male (Taukihepa: 3.55 + 0.83 ha, Ulva: 2.84 ± 0.79 ha) and female (Taukihepa: 4.12 + 1.2 ha, Ulva: 2.09 ± 0.44 ha) weka which differed from past research. Male weka on Taukihepa had a greater amount of their home ranges overlapped by breeding partners than those on Ulva Island. There was a female-biased sex ratio on Taukihepa and a polygamous breeding strategy. All observed nests of polygamous weka failed. Conversely the sex ratio was even on Ulva Island, weka were monogamous and breeding success appeared to be high. In contrast to the hypothesis of resource-rich seabird islands resulting in an increase of body size for top predators, weka on seabird islands did not attain greater body weights than those on non-seabird islands.
Habitat selection of weka on Taukihepa showed a strong preference for areas of low cover. On Ulva Island habitat selection was random with weka utilising mixed-podocarp forest and open coastal areas according to their availability. On Ulva Island coastal resources enabled weka living in the marine interface to have smaller home ranges than those living inland. This research stands as the only record of the home range and habitat use of Stewart Island weka. Due to the declines of weka populations throughout New Zealand, such information is important for their future management.