Abstract
New Zealand’s endemic fauna face a significant risk of population decline due to predation from introduced mammalian predators. This is partially due to the lack of shared evolutionary history with these terrestrial mammalian predators, which may prevent endemic fauna from developing appropriate antipredator behaviours to cope with the predation strategies of these mammals. Significant population declines have been observed in many of New Zealand’s endemic species, primarily due to predation from small ground-dwelling rodents and mustelids, with rats and stoats posing the highest risk. Insects have been less well studied than birds and reptiles, but many are at risk from introduced predators, as many are large-bodied, slow moving, odorous, and naïve to the foraging strategies of mammals. Of particular interest are endemic wētā, a group of over 70 species unique to New Zealand, 16 of which are threatened or declining as a result of unsustainable predation from introduced mammals. While general antipredator behaviours in this group are well documented, behavioural responses/antipredator strategies to cope with mammalian predator exposure have not been examined in detail.
Firstly, this thesis examines the potential differences in activity, refuge seeking, level of aggression, and response to looming stimuli between two groups of Wellington tree wētā (Hemideina crassidens), those living in an ecosanctuary, free of most invasive mammalian predator presence, and those living in urban environments where there is little to no predator control. Sampling in a predator free area allows a comparison of wētā that are naïve vs. predator experienced, and tests whether protected/naïve wētā display antipredatory behaviours. A ‘protected’ group of 30 wētā was collected within the bounds of Zealandia, a predator-free sanctuary in Wellington subject to rigorous predator control. A second, ‘non-protected’ group of 27 wētā were gathered from urban sites in the Wellington area where predator control was low or non-existent. Both groups were subject to identical behavioural phenotyping assays, and repeated measures of their behavioural responses were video recorded and analysed using EthoVision XT behavioural quantification software. The results of this field study demonstrate that wētā living in areas where mammalian predators are present are less active and less aggressive than wētā in predator-free habitats. Lower activity levels in the non-protected wētā may be a strategy to avoid predator encounters/detection, though may also have negative consequences for their foraging and mating opportunities, whereas the higher activity observed in the protected wētā suggests that these wētā may be vulnerable to attack if they were to encounter unfamiliar predators. The lower rates of aggression observed in the non-protected group may also be the result of wētā modifying their behaviour to avoid encounters with mammalian predators, but may be attributed to a reduction of male-male competition in the fragmented population outside of Zealandia’s bounds. Aggressive behaviour is a key factor in this species’ mating system, and a reduction in this aggression may negatively influence mate selection in terms of quality. If the loss of aggressive behaviour is an adaptation for predator avoidance, highly aggressive protected wētā may be further at risk when encountering mammalian predators.
Secondly, this thesis presents a laboratory experiment examining the response of H. crassidens to predator simulations over the course of six months while in captivity. This experiment also examined the effectiveness of behavioural conditioning for the development of antipredator behaviours, and capacity for olfactory memory of predator cues in H. crassidens. Wētā collected in the field were split into three treatment groups and subject to a regime intended to condition treatment wētā to associate the scent of rat droppings with a negative experience. One group (n= 13) received simulated predator encounters while in the presence of rat droppings, another (n= 13) received simulated predator encounters with no scent cue, and the final control group (n= 7) received normal husbandry practices with no intervention. Similar to the field study, activity, refuge seeking, level of aggression, and response to looming stimuli were recorded, in addition to a trial which measured their activity in the presence of a dish containing rat droppings and their average distance from the dish. Wētā subject to simulated predator encounters increased their rate of movement while in a novel arena, and aggression levels declined across all groups. Higher activity levels in the treatment group wētā may be a response to the regular attacks experienced over the course of this experiment, which likely raised the stress levels of treatment wētā, a factor which increases locomotion in other species. The increased rate of movement may also indicate a modification of antipredator behaviour, as typical aggressive responses were not effective in deterring the simulated predator attacks. A decline in aggressive expression may be the result of habituation to a captive environment, though results gathered in the field suggest that aggressive behaviour is downregulated in areas with high levels of predator pressure similar to the pressures experienced by the treatment groups. Due to a diminished control sample, neither hypothesis could be concluded in a concrete manner.
The results of this study have implications for the continued health and reproductive quality of unprotected wētā living in environments with invasive predator pressure, and raise concerns about the vulnerability of protected, predator-naïve populations in ecosanctuaries and offshore islands. It is important for conservations efforts to consider the ways in which behavioural differences between populations may impact the efficacy of existing management strategies, especially translocation efforts, and protected areas vulnerable to predator incursions. The implication that captivity may cause a reduction in aggressive expression is concerning, especially when considering the vital role that aggressive displays play in the mating system of this species, as well as their role in predator deterrence. Captive rearing programs should consider the ways in which captivity can influence the behavioural expression of wētā, and the ways in which inappropriate antipredator expression, and a reduced capacity for reproductive competition may leave wētā vulnerable in an unfamiliar environment.