Ecological Archives M085-008-A15
David M. Forsyth, Deborah J. Wilson, Tomás A. Easdale, Georges Kuntsler, Charles D. Canham, Wendy A. Ruscoe, Elaine F. Wright, Lora Murphy, Andrew M. Gormley, Aurora Gaxiola, and David A. Coomes. 2015. Century-scale effects of invasive deer and rodents on the dynamics of forests growing on soils of contrasting fertility. Ecological Monographs 85:156–179. http://dx.doi.org/10.1890/14-0389.1
Appendix O. Evaluation of model predictions.
Short-term predictions
Our SORTIE/NZ alluvial and marine terrace forest models would be supported if their predictions for the trajectories of seedling and sapling densities in the presence and/or absence of deer and rodents were consistent with those observed in New Zealand forests in the presence and/or absence of deer and/or rodents. The predictions for the alluvial terrace forest in the presence and absence of deer (Figs. 5, 7) are consistent with meta-analyses showing that plant densities in the browse layer are higher inside compared with outside 20 × 20 m ungulate exclosures throughout New Zealand (Wardle et al. 2001, Spear and Chown 2009, Wright et al. 2012). Furthermore, larger-leafed "preferred" species such as Schefflera digitata (Appendix J: Fig. J1 A) were frequently "very abundant inside the exclosure and effectively absent outside it" (Wardle et al. 2001, p. 593; see also Forsyth et al. 2010, Mason et al. 2010), which is consistent with: (1) the results of our field study evaluating seedling growth and mortality rates inside and outside deer exclosures in the alluvial terrace forest (Fig. 4; Appendix I), and (2) the century-level predictions for the alluvial terrace forest in the presence and absence of deer (Fig. 7; Appendix M).
Some sub-canopy species become more abundant in the presence of deer (Allen et al. 1984, Wardle et al. 2001). Pseudowintera colorata is avoided by deer (Forsyth et al. 2002, 2005), is more abundant outside ungulate exclosures (Allen et al. 1984, Wardle et al. 2001), and is regenerating nationally in the presence of deer (Forsyth et al. 2010). Our alluvial terrace forest model predicted that P. colorata seedlings, saplings and adults would become more abundant in the presence of deer than in their absence (Fig. 7; Appendix M).Hence, our model predictions for P. colorata in the presence of deer are consistent with short-term trends observed in field studies in the presence of deer.
All life-history stages of the dominant canopy species in the alluvial terrace forest, Weinmannia racemosa, increased significantly in our model simulations in which deer were excluded. Weinmannia racemosa is New Zealand's most numerous canopy species (Peltzer and Payton 2006) and seedlings and saplings have increased inside compared with outside deer exclosures (Husheer 2007, Mason et al. 2010). In the Murchison Mountains, 120 km from Waitutu Forest, seedling and sapling densities were significantly higher inside compared with outside red deer exclosures erected on fertile terraces (Tanentzap et al. 2009). The seedlings of preferred and avoided species were more abundant inside and outside the deer exclosures, respectively (Tanentzap et al. 2009), consistent with the predictions of the alluvial terrace forest model (Fig. 7; Appendix M).
The increasing dominance of all Schefflera digitata life-history stages in the alluvial terrace forest in the absence of deer is realistic, with Mark and Baylis (1982) reporting this species dominating stands on Secretary Island, 110 km from Waitutu Forest, when deer had only recently invaded. The density of canopy trees there was 1370 per hectare, of which 42% were S. digitata (Mark and Baylis 1982). The median densities and proportions of S. digitata canopy trees predicted at 500 years in the alluvial terrace forest in the absence of deer were similar to those recorded on Secretary Island in the absence of deer: 1183 trees per hectare of which 19% were S. digitata in the absence of deer and presence of rodents, and 1155 trees per hectare of which 16% were S. digitata in the absence of rodents (Fig. 5; Appendix M). Seedling and sapling densities on Secretary Island were not reported in a way that is comparable with the predictions of our model, but we believe that two factors may be responsible for the very high densities of S. digitata seedlings and saplings predicted after ~400 years in the alluvial terrace forest model in the absence of deer. First, flooding is likely to increase the mortality rates and decrease the growth rates of seedlings in the alluvial terrace forest (Gaxiola et al. 2010). We did not include flooding in our alluvial terrace model, but it would be worthwhile exploring the importance of this disturbance in future work. Second, the growth and mortality rates of S. digitata estimated in the absence of deer may have been artificially high and low, respectively, in very low light environments (i.e., <2.5% GLI) because those conditions did not occur in the deer exclosures. Although we attempted to minimize this effect by using a bi-level function to set a higher probability of mortality in deep shade (Appendix B), it may still have contributed to the high abundances of S. digitata seedlings, saplings and adults in the alluvial terrace forest in the absence of deer.
House mice were overwhelmingly the most common rodent in Waitutu Forest, at least during mast years (Ruscoe 2004, Ruscoe et al. 2004, W. A. Ruscoe et al. unpublished data), and there are no data with which to validate the short-term predictions of our model independent of the data used to parameterize it (Wilson et al. 2006). However, the effects of Rattus species on seedling recruitment were investigated on Breaksea Island, ~80 km from Waitutu Forest, where deer have never been present (Allen et al. 1994). Nothofagus solandri var. cliffortioides and N. menziesii seedling densities increased over a 5-year period following the eradication of Norway rats (Rattus norvegicus) in 1988 (Allen et al. 1994). Although absolute densities of seedlings were not reported by Allen et al. (1994), and the Norway rat was not detected during our field studies at Waitutu Forest, their observations are in broad agreement with the predictions of our alluvial terrace forest model: Nothofagus menziesii seedlings, saplings and eventually adults became much more abundant in the absence of rodents (Fig. 7; Appendix M).
Inferences about forest dynamics inside and outside exclosures in New Zealand are hampered by an absence of data on soil variables collected at the appropriate scale (Mason et al. 2010), so it is not possible to explicitly consider soil fertility in our evaluation of model predictions. However, exclosure effects have been shown to be greatest in disturbed forest stands (Mason et al. 2010), which, all other things being equal, will have higher soil fertility and light, two factors that favor the fast-growing angiosperm species that are generally preferred by deer. Our findings highlight the need for exclosure studies to explicitly consider soil fertility, measured at the appropriate scale, as an explanatory variable.
Long-term predictions
Our models would be upported if their long-term predictions for the trajectories of the canopy dominants (i.e., their basal areas) in the absence of deer and rodents were consistent with those observed in palynological studies conducted near our study area. The adult basal areas of Nothofagus menziesii were predicted to become increasingly abundant, usually at the expense of the co-dominant Weinmannia racemosa, during the 500 years, in the alluvial terrace forest scenarios (Fig. 7; Appendix M). The palynological record shows widespread and rapid increase in Nothofagus spp. in southwestern South Island (i.e., including Waitutu Forest) after around 6000 BP, and the genus "may still be spreading" (McGlone et al. 1996, p. 109; see also McGlone and Wilmshurst 1999, Wilmshurst et al. 2002). The increasing dominance of N. menziesii in the alluvial terrace forest is consistent with this trend (i.e., it is still invading).
In contrast to the alluvial terrace forest, the abundances of Nothofagus menziesii and N. solandri var. cliffortioides declined in the marine terrace forest (Fig. 8; Appendix N). Both species were much less abundant than the dominant Dacrydium cupressinum (Coomes et al. 2005; Fig. 8; Appendix N), and their decline is consistent with the hypothesis that these species are competitively excluded by podocarps on less fertile and poorly drained soils (Urlich at al. 2005). In particular, the forest canopy on sites with infertile soil may not be dense enough to shade out podocarp seedlings and saplings (Coomes et al. 2005). Finally, this result is also in agreement with simulation studies using an ecosystem process model showing a dominance of D. cupressinum at some sites in southwestern South Island (Hall and Hollinger 2000, Hall and McGlone 2001, 2006).
Literature Cited
Allen, R. B., W. G. Lee, and B. D. Rance. 1994. Regeneration in indigenous forest after eradication of Norway rats, Breaksea Island, New Zealand. New Zealand Journal of Botany 32:429–439.
Allen, R. B., I. J. Payton, and J. E. Knowlton. 1984. Effects of ungulates on structure and species composition in the Urewera forests as shown by exclosures. New Zealand Journal of Ecology 7:119–130.
Coomes, D. A., et al. 2005. The hare, the tortoise and the crocodile: the ecology of angiosperm dominance, conifer persistence and fern filtering. Journal of Ecology 93:918–935.
Forsyth, D. M., D. A. Coomes, G. Nugent, and G. M. J. Hall. 2002. The diet and diet preferences of ungulates (Order: Artiodactyla) in New Zealand. New Zealand Journal of Zoology 29:323–343.
Forsyth, D. M., S. J. Richardson, and K. Menchenton. 2005. Foliar fibre predicts diet selection by invasive red deer Cervus elaphus scoticus in a temperate New Zealand forest. Functional Ecology 19:495–504.
Forsyth, D. M., J. M. Wilmshurst, R. B. Allen, and D. A. Coomes. 2010. Impacts of introduced deer and extinct moa on New Zealand ecosystems. New Zealand Journal of Ecology 34:48–65.
Gaxiola, A., S. M. McNeill, and D. A. Coomes. 2010. What drives retrogressive succession? Plant strategies to tolerate infertile and poorly drained soils. Functional Ecology 24:714–722.
Hall, G. M. J., and D. Y. Hollinger. 2000. Simulating New Zealand forest dynamics with a generalized temperate forest gap model. Ecological Applications 10:115–130.
Hall, G. M. J., and M. S. McGlone. 2001. Forest reconstruction and past climatic estimates for a deforested region of south-eastern New Zealand. Landscape Ecology 16:501–521.
Hall, G. M. J., and M. S. McGlone. 2006. Potential forest cover of New Zealand as determined by an ecosystem process model. New Zealand Journal of Botany 44:211–232.
Husheer, S. W. 2007. Introduced red deer reduce tree regeneration in Pureora Forest, central North Island, New Zealand. New Zealand Journal of Ecology 31:79–87.
Mark, A. F., and G. T. S. Baylis. 1982. Further studies of the impact of deer on Secretary Island, Fiordland, New Zealand. New Zealand Journal of Ecology 5: 67–75.
Mason, N. W. H., D. A. Peltzer, S. J. Richardson, P. J. Bellingham, and R. B. Allen. 2010. Stand development moderates effects of ungulate exclusion on foliar traits in the forests of New Zealand. Journal of Ecology 98:1422–1433.
McGlone, M. S., D. C. Mildenhall, and M. S. Pole. 1996. History and paleoecology of New Zealand Nothofagus forests. Pp. 83–130 in T. T. Veblen, R. S. Hill, and J. Read, editors. The ecology and biogeography of Nothofagus forests. Yale University Press, New Haven, Connecticut, USA.
McGlone, M. S., and J. M. Wilmshurst. 1999. A Holocene record of climate, vegetation change and peat bog development, east Otago, New Zealand. Journal of Quaternary Science 14:239–254.
Peltzer, D. A., and I. J. Payton. 2006. Analysis of Carbon Monitoring System data for indigenous forests and shrublands collected in 2002/03. Landcare Research Contract Report LC0506/99. Landcare Research, Lincoln, NZ.
Ruscoe, W. A. 2004. A new location record for kiore (Rattus exulans) on New Zealand's South Island. New Zealand Journal of Zoology 31:1–5.
Ruscoe, W. A., D. Wilson, L. McElrea, G. McElrea, and S. J. Richardson. 2004. A house mouse (Mus musculus) population eruption in response to heavy rimu (Dacrydium cupressinum) seedfall in southern New Zealand. New Zealand Journal of Ecology28:259–265.
Spear, D., and S. L. Chown. 2009. Non-indigenous ungulates as a threat to biodiversity. Journal of Zoology, London, 279:1–17.
Tanentzap, A. J., L. E. Burrows, W. G. Lee, G. Nugent, J. M. Maxwell, and D. A. Coomes. 2009. Landscape-level vegetation recovery from herbivory: progress after four decades of invasive red deer control. Journal of Applied Ecology 46:1064–1072.
Urlich, S. C., G. H. Stewart, R. P. Duncan, and P. C. Almond. 2005. Tree regeneration in a New Zealand rain forest influenced by disturbance and drainage interactions. Journal of Vegetation Science 16:423–432.
Wardle, D. A., G. M. Barker, G. W. Yeates, K. I. Bonner, and A. Ghani. 2001. Introduced browsing mammals in New Zealand natural forests: aboveground and belowground consequences. Ecological Monographs 71:587–614.
Wilmshurst, J. M., M. S. McGlone, and D. J. Charman. 2002. Holocene vegetation and climate change in southern New Zealand: linkages between forest composition and quantitative surface moisture reconstructions from an ombrogenous bog. Journal of Quaternary Science 17:653–666.
Wilson, D. J., W. A. Ruscoe, L. E. Burrows, L. M. McElrea, and D. Choquenot. 2006. An experimental study of the impacts of understorey forest vegetation and herbivory by red deer and rodents on seedling establishment and species composition in Waitutu Forest, New Zealand. New Zealand Journal of Ecology 30:191–207.
Wright, D. M., A. J. Tanentzap, O. Flores, S. W. Husheer, R. P. Duncan, S. K. Wiser, and D. A. Coomes. 2012. Impacts of culling and exclusion of browsers on vegetation recovery across New Zealand forests. Biological Conservation 153:64–71.