New Zealand Journal of Ecology (2010) 34(1): 48- 65

Impacts of introduced deer and extinct moa on New Zealand ecosystems

Review Article
David M. Forsyth 1*
Janet M. Wilmshurst 2
Robert B. Allen 2
David A. Coomes 3
  1. Arthur Rylah Institute for Environmental Research, Department of Sustainability and Environment, 123 Brown Street, Heidelberg, Victoria 3084, Australia
  2. Landcare Research, PO Box 40, Lincoln 7640, New Zealand
  3. Department of Plant Sciences, University of Cambridge, Downing Street, Cambridge CB3 2EA, UK
*  Corresponding author

There has been considerable ongoing debate about the extent to which the impacts of introduced deer on native vegetation have replaced those of moa, and since the 1980s there have been major changes in thinking about the impacts of deer and ratites on ecosystems. Although it has long been known that deer caused a predictable sequence of changes in forest understorey composition, recent work has shown that the foliage of species preferred by deer contains lower concentrations of fibre – and decomposes faster – than avoided species. Analyses of long-term permanent plot data suggest that some preferred species are failing to regenerate in forest types where deer are present. As well as likely altering the long-term biogeochemistry of forest ecosystems, deer have a strong negative effect on the abundance of litter-dwelling macrofauna (most likely through trampling). Estimating the impacts of extinct taxa on an ecosystem has much uncertainty, but recent experiments have shown that extant ratites and deer may have more similar feeding preferences than previously believed. It is likely that moa were important seed dispersers, but this has not been studied for deer in New Zealand. Although collectively the various taxa of deer in New Zealand use all of the habitats utilised by moa, and there is partial overlap in the diets of deer and moa, deer can attain densities and biomasses 100- fold greater than reasonably surmised for moa. We believe that the impacts of introduced deer on ecosystems are markedly different from those of moa. One way to compare the impacts of moa and deer is to use pollen to reconstruct the vegetation at a forested site in recent millennia and evaluate vegetation dynamics during the moa period, following the extinction of moa but prior to the arrival of deer (i.e. the moa gap), and following the arrival of deer. We illustrate the potential of this approach with a soil core from Chester Burn, Murchison Mountains in Fiordland. Five other areas that deserve further research are also identified.