To disturb or not disturb : radio-telemetry based territory mapping and camera traps for monitoring cryptic species using Rakiura tokoeka (Apteryx australis australis) : a thesis presented in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Conservation Biology at Massey University, Albany, New Zealand

Abstract

Successful wildlife conservation is dependent on effective population monitoring. The objectives of monitoring can be multi-faceted, but population density is often coveted. However, estimating the density of species in which individuals cannot be identified is challenging without a complete census, which is rarely possible. In New Zealand, kiwi (Apteryx spp.) are morphologically and behaviourally cryptic species that are difficult to monitor, yet monitoring results are essential to their conservation. The gold standard of kiwi monitoring is territory mapping using radio-telemetry, yet this method had never been formally evaluated, and the alternative method is call counts, which are difficult to relate to population variables.

Using Rakiura tokoeka (Apteryx australis australis) as my case study species, I implemented territory mapping with radio-telemetry at two novel locations on Rakiura/Stewart Island, one a pest-free island. My aim was to establish ‘known’ populations (minimum densities), and evaluate the traditional territory mapping method (Robertson, 2018) by comparing different ways of establishing territory boundaries (field-workers-estimates, convex 100% and concave polygons) and incorporating increasing numbers of locations by extending the length of the surveys.

In these two study locations, and an additional two locations where historical territory mapping surveys of Rakiura tokoeka had provided minimum density estimates, I then trialled a novel method for monitoring kiwi populations, camera trap grids. I ran camera trap surveys for Rakiura tokoeka seasonally between 2018 and 2020 in the four locations, using the same study areas that were used for territory mapping. I used the data from the camera surveys in two ways, firstly for a relative abundance index (calculating camera trap rate), and secondly in a statistical abundance model that estimated point abundance and detection probability (Royle-Nichols, 2003). I also evaluated the use of radio-telemetry and camera traps for monitoring Rakiura tokoeka through the breeding season and providing information on the survival, growth, and dispersal of young Rakiura tokoeka.

I found that territory mapping using radio-telemetry is an effective monitoring method for establishing minimum densities of Rakiura tokoeka, but that current methods could be improved. Convex polygon territory boundaries were similar to field-workers-estimate boundaries but could be applied more objectively and increasing the number of location fixes decreased density estimates. Camera trapping was an effective monitoring method for Rakiura tokoeka populations. Camera trap rates were an easy to apply index that is widely applicable to the many community and conservation groups working with kiwi that do not have access to radio-telemetry for monitoring. Camera trap rates indicated that the pest free island (Ulva Island) had lower detections of Rakiura tokoeka than the other three locations. This could have been due to lower densities of Rakiura tokoeka over some parts of the island, or lower detection probabilities. From the abundance model, estimates of density of Rakiura tokoeka at the four locations were not significantly different to those from territory mapping. This indicated that camera trap surveys could be used as a non-invasive alternative to territory mapping with radio-telemetry for Rakiura tokoeka, and potentially other kiwi species that has lower cost and requires less effort.

I found the use of radio-telemetry and camera traps effective for monitoring different aspects of breeding behaviour, nesting, and chick variables, with the best use being a combination of the two methods. From the data collected on age structure, breeding and survival, I found no reason for immediate concern or management intervention for the Rakiura tokoeka population.

This project shows the importance of questioning established methods, trialling new methods, combining methods, and considering whether the invasiveness of a method is warranted to meet the project objectives.