|dc.description.abstract||The exposed gravel of the braided rivers of the eastern South Island of New Zealand provide nesting habitat for rare endemic bird species. These endemic bird species evolved in the absence of mammalian predators and it is suspected that predation by introduced predators is a major cause of species decline in braided rivers. Large scale monitoring of predation in braided rivers has mainly been confined to the Waitaki Basin, using video cameras. Further study of nest predation in braided rivers may require more economical monitoring techniques in order to cover more rivers and reduce the risk of financial loss through damaged or stolen equipment. Artificial eggs and nests have been widely used over the past two decades and have provided significant advances in nesting biology theory. However an increasing volume of research has shown that the results from artificial nests should be interpreted with caution as they may not sufficiently replicate the real nest situation.
This study provided an initial assessment of the utility of artificial wax eggs as a device to measure interference and artificial nests as a presentation technique for the wax eggs. There are several areas that were tested. Initially aspects of design were considered, did the wax egg need tying down, how should the wax eggs be tied, what strength of monofilament is appropriate, is a lure (natural egg) required to entice interference with the wax egg? Following this cryptically painted wax eggs were tested to replace the dyed wax eggs that had been used previously. Once the wax egg was designed it was trialled in both real Banded Dotterel and artificial nests. Overall rates of predation were compared between real and artificial nests along with the patterns of predation with respect to wax egg colour and location. The difference in predator suite recorded at real and artificial nests is also discussed.
This study found wax eggs were less likely to be lost if they were tied down. Failures of the tethers during Trial 2 were mitigated by increasing the breaking strain of the monofilament from 6.8 kg to 22.7 kg. Well-rooted vegetation (where available) was found to provide a more suitable anchor than rocks, although either were satisfactory. High rates of interference were observed on dyed wax, real quail and chicken eggs. It was decided that these rates of interference were too high to consider these egg types for use during the breeding season. Cryptically painted wax eggs were then trialled and found to not induce a strong response from the predators like that observed with the other egg types.
During the 2005 breeding season wax eggs were accepted by the majority of incubating adults and were seldom left unmarked following a predation event. However, in most cases predation could only be assigned to a predator group (avian or mammalian) rather than species, as the marks often lacked any clear distinguishing patterns for individual species. Impressions were identified to species level on seven occasions, two Black-backed Gulls and five Hedgehogs.
There was no significant difference between the overall predation rates observed at real and artificial nests. Artificial nests also showed similar effects of wax egg colour and location as real nests. However, there was a difference in the predator suite between real and artificial nests. The avian:mammal split for real nests was avian 17 (65%) and mammal 9 (35%) against the artificial nests 24 avian (96%) and mammal 1 (4%). These results highlight a problem with artificial nest experiments which is that although predation rates may be the same between real and artificial nests there may be a bias in predator type and that these biases may be counterbalancing. These results show further evidence that where possible artificial nest studies should be verified or calibrated with real nest data to assess there accuracy. The use of artificial nest should not be discounted because it has been shown to be inaccurate for some situations. For instance artificial nests may be appropriate where a species is highly endangered and therefore real nests simply are not available, or in situations where measuring of predator presence rather than effect on a target prey species is of primary interest.||en