Estimating abundance, age structure and sex ratio of a recently discovered New Zealand tusked weta

Estimates of abundance, age structure and sex ratio are essential for monitoring the status of populations. We report the first attempt to reliably estimate these parameters in a population of the recently discovered Raukumara tusked weta (Motuweta riparia), which is found almost entirely near streams. On two occasions we searched a 211-m section of creek for 4–5 successive nights and individually marked all weta. We estimated abundance of adults and juveniles using closed-population mark-recapture analysis. The choice of mark-recapture model made a substantial difference to the estimated abundance (116–238) and proportion of juveniles (32–72%). However, no single model was clearly better supported than any other. We therefore used model averaging to account for uncertainty in model choice, giving an estimate of 142 (95% CI 105–231) weta including 56 (95% CI 41–234) adults and 77 (95% CI 46–209) juveniles. This corresponds to a density of 0.11 (95% CI 0.08–0.18) weta per m2, assuming 3 m of habitat on either side of the creek. Recapture probability was much lower for adults (n = 2) than juveniles (n = 10), possibly caused by a difference in handling (adults were held overnight whereas juveniles were not). Slightly more juvenile males (n = 22) were caught than juvenile females (n = 21), but more adult females (n = 51) were caught than adult males (n = 31), suggesting a potentially higher mortality in males. An initial assessment of population size is crucial when considering the conservation status and management strategy of rare animals, such as the highly endangered Middle Island tusked weta. The immediate practical benefit of the methods developed here is therefore clear. In the long-term, these methods also have important implications for the continued monitoring of trends in other threatened invertebrate populations.