New Zealand Journal of Ecology (2014) 38(1): 26- 38

Simulating long-term vegetation dynamics using a forest landscape model: the post-Taupo succession on Mt Hauhungatahi, North Island, New Zealand

Research Article
Timothy Thrippleton 1*
Klara Dolos 1
George L. W. Perry 2,3
Jürgen Groeneveld 2,4
Björn Reineking 1
  1. Biogeographical Modelling, BayCEER, University of Bayreuth, Universitätsstr. 30, 95447 Bayreuth, Germany
  2. Department of Ecological Modelling, Helmholtz Centre for Environmental Research – UFZ, Permoserstr. 15, 04318 Leipzig, Germany
  3. School of Biological Sciences, The University of Auckland, Private Bag 92019, Auckland 1142, New Zealand
  4. School of Environment, The University of Auckland, Private Bag 92019, Auckland 1142, New Zealand
*  Corresponding author

Forest dynamics in New Zealand are shaped by catastrophic, landscape-level disturbances (e.g. volcanic eruptions, windstorms and fires). The long return-intervals of these disturbances, combined with the longevity of many of New Zealand’s tree species, restrict empirical investigations of forest dynamics. In combination with empirical data (e.g. palaeoecological reconstructions), simulation modelling provides a way to address these limitations and to unravel complex ecological interactions. Here we adapt the forest landscape model LandClim to simulate dynamics across the large spatio-temporal scales relevant for New Zealand’s forests. Using the western slope of Mt Hauhungatahi in the central North Island as a case study, we examine forest succession following the Taupo eruption (c. 1700 cal. years BP), and the subsequent emergence of elevational species zonation. Focusing on maximum growth rate and shade tolerance we used a pattern-oriented parameterisation approach to derive a set of life-history parameters that agree with those described in the ecological literature. With this parameter set, LandClim was able to reproduce similar spatio-temporal patterns in vegetation structure to those seen in pollen reconstructions and contemporary vegetation studies. The modelled successional sequence displayed a major shift in forest composition between simulation years 400 to 700, when the dense initial stands of conifers (dominated mainly by Libocedrus bidwillii) were progressively replaced by the angiosperm Weinmannia racemosa in the montane forest. From around year 1000, the contemporary elevational species zonation was attained. Light-competition controlled the major successional trends and, together with temperature-limitation, explained the observed elevational species zonation. Although designed for European temperate forests, LandClim can simulate New Zealand landscape dynamics and forest response to catastrophic disturbances such as the Taupo eruption. We suggest that LandClim provides a suitable framework for investigating the role of spatial processes, in particular disturbance, in New Zealand’s forest landscapes.