Estimating the proportion of Archey’s frogs in the Wharekirauponga mine vibration footprint 13 within a 100 ha area in the Wharekirauponga catchment. The closest recorded Archey’s frog sightings to plots at the other three sites are: two 1986 sightings just over 180 m from the Golden Cross plots; several sightings between 1981 and 2007 over 90 m from the Tapu Ridge plots; and two 1981 sightings 1.3 km from the Crosbies plots. The expected frog counts in 100 m2 plot at each of the three areas was estimated using quasipoissonv regression models[26]. Estimates of the numbers of frogs present in plots were obtained from parameter estimates of the expected frog count per plot by dividing them by the mean detection probability for Archey’s frogs during nocturnal searches (i.e. 0.322, SE ±0.055). Eighty-three plots were surveyed in the Wharekirauponga catchment (WKP) during the period 2018–2019 (Table 8). Fifty-nine of the Wharekirauponga plots were in manukakanuka scrub with general hardwood forest (M1, n5) between 100 m and 300 m a.s.l.; five in kauri forest with mixed native scrub (N2, m6) between 200 m and 400 m a.s.l; and nineteen in kauri forest with podocarp-hardwood forest (N2, n3a) between 200 m and 400 m a.s.l.. Sixty-eight of the plots in the Wharekirauponga catchment were in the predicted vibration footprint for the proposed Wharekirauponga mine. Plots at Golden Cross, Crosbies and Tapu Ridge, were surveyed during early 2022 (Table 8). There were: eight plots at Golden Cross, all in podocarp-hardwood forest (N3a), between 300 and 400 m a.s.l.; six plots at Crosbies, all in manuka-kanuka scrub with low producing grassland (M1, p2), between 600–700 m a.sl; and two plots at Tapu Ridge, both in podocarp-hardwood forest with some kauri (N3a, n2), between 450–550 m a.s.l.. Most plots were surveyed on more than one night to obtain replicate counts. For analyses, data from plots in the same region and within the same vegetation type were grouped together to provide separate sets of estimates (Table 8). Estimates include: the expected count of frogs found during a single survey of a 100 m2 plot, the number of frogs present in a 100 m2 plot; and the density of frogs ha-1. v The quasi-poisson regression is a generalization of the Poisson regression and is used when modelling an overdispersed count variable.
RkJQdWJsaXNoZXIy MjE2NDg3