No evidence for sampling bias caused by capture method or time in Apteryx mantelli

Sampling bias can have dire consequences for research. One potential source of bias is combining different sampling methods in the same study. However, combining methods can be unavoidable, for instance, when sampling method selection depends upon factors such as population density or terrain. A case at hand is the use of night-time encounter catching by people or daytime catching using certified dogs for studies of Apteryx mantelli, North Island brown kiwi, in Aotearoa New Zealand. Here, we compare these sampling methods to determine whether (1) combining them risks inducing a demographic bias to the sample set, and (2) they differ in regards to blood parameters used for comparing populations (packed cell volume, glucose, plasma protein, haemoglobin). Sixty-five birds were caught during the day from their roosts using a certified dog, and 62 birds were caught at night while foraging. The results suggest that both methods capture a comparable subset of a population, with the potential exception that more very young juveniles were caught using the day method. Furthermore, no physiological effects were evident from comparing haematological parameters. We also found no difference in blood sampling success between night and day, but observed that blood extraction was more difficult at night. Hence, we demonstrate that either method, or a combination of both, can be considered for future studies. Notably, we found that night-time encounter catching had a superior success rate in very high-density populations. Since this method also negates dependency on the limited number of certified dogs, we suggest that benefits may exist through increasing the utilisation of night-time encounter catching in A. mantelli research. We suggest that future studies should consider measuring the stress levels caused by each of the methods, and quantify the effects of habitat type and terrain on sampling success.