Past and Future Trajectories of Forest Loss in New Zealand
Total Page：16
File Type：pdf, Size：1020Kb
BIOLOGICAL CONSERVATION 133 (2006) 312– 325 available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/biocon Past and future trajectories of forest loss in New Zealand Robert M. Ewersa,b,c,*, Andrew D. Kliskeyd,1, Susan Walkere, Daniel Rutledgef, Jon S. Hardinga, Raphael K. Didhama,* aSchool of Biological Sciences, University of Canterbury, Private Bag 4800, Christchurch, New Zealand bInstitute of Zoology, Zoological Society of London, Regent’s Park, London NW1 4RY, UK cDepartment of Zoology, Cambridge University, Downing Street, Cambridge CB2 3EJ, UK dDepartment of Geography, University of Canterbury, Private Bag 4800, Christchurch, New Zealand eLandcare Research, Private Bag 1930, Dunedin, New Zealand fLandcare Research, Private Bag 3127, Hamilton, New Zealand ARTICLE INFO ABSTRACT Article history: Historically, New Zealand was dominated by forest below the alpine treeline, but about Received 23 March 2006 1000 years of Polynesian and European colonisation has resulted in the destruction of nearly Received in revised form three-quarters of the indigenous forest cover. In this study, the historical patterns of defor- 12 June 2006 estation and forest fragmentation were assessed in relation to major topographical, climatic Accepted 20 June 2006 and anthropogenic variables that may drive forest loss. Deforestation has occurred almost Available online 14 September 2006 equally on the two main islands, the North and South Islands, although the remaining indigenous forest is more fragmented in the North Island. Most deforestation has occurred Keywords: in regions with a high-density of road networks, although gradients in climatic water avail- Deforestation ability and soil fertility also had weak effects. Deforestation rates over the period 1997–2002 Extinction threshold were very low (nationwide deforestation rate of just À0.01% p.a.), but varied widely among Habitat loss political districts. Expansion of plantation forestry was the single most important driver of Landcover change recent deforestation. Only 10 of 73 political districts are afforded long-term protection of Landscape threshold native forest cover (having more than 30% forest cover that is managed by the Department Restoration targets of Conservation). Forest cover in the majority of New Zealand landscapes has been reduced below the level of an expected ‘extinction threshold’ (circa 30% native habitat cover) in 55 political districts, and long-term trajectories predict that ongoing deforestation threatens to force another five districts below the critical threshold within the next 45 years. Except for the most heavily deforested regions, relatively modest annual rates of habitat restora- tion could bring forest cover back above the extinction threshold by the year 2050. Ó 2006 Elsevier Ltd. All rights reserved. 1. Introduction (McGlone, 1989), but the arrival of the early Maori people about 1000 BP initiated widespread forest destruction. The Three thousand years ago, forest covered virtually the entire Maori burned significant areas of lowland forest to encourage land surface area of New Zealand below the alpine treeline the growth of bracken fern (Pteridium aquilinum) that was used * Corresponding authors: Tel.: +44 1223 336 675; fax: +44 1223 336 676 (R.M. Ewers), Tel.: +64 3 3642059; fax: +64 3 3642590 (R.K. Didham). E-mail addresses: [email protected] (R.M. Ewers), [email protected] (A.D. Kliskey), [email protected] (S. Walker), [email protected] (D. Rutledge), [email protected] (J.S. Harding), [email protected] (R.K. Didham). 1 Present address: Department of Biological Sciences, University of Alaska Anchorage, 3211 Providence Drive, Anchorage, AK 99508, United States. 0006-3207/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocon.2006.06.018 BIOLOGICAL CONSERVATION 133 (2006) 312– 325 313 as a food source, to make cross-country travel easier and also trees the debt may still be current more than a century fol- as a strategy for hunting moa (Stevens et al., 1988). Maori lowing deforestation (Turner et al., 1996; Vellend et al., were, however, not the sole cause of deforestation during this 2006). time, as climatic change, volcanism and naturally ignited fires Second, much research has been focused on landscape have all been implicated as factors driving Holocene vegeta- and extinction thresholds in fragmented landscapes (Ewers tion change in New Zealand (Fleet, 1986; McGlone, 1989). As and Didham, 2006). Landscape thresholds are the result of a result of these combined factors, forest cover had been ‘‘rapid changes in the size and isolation of patches at critical reduced to an estimated 68% of the land surface by the time proportions of habitat in the landscape’’ (Andre´n, 1994), and European settlers arrived in the early 1800s (Salmon, 1975), are predicted to occur at approximately 30% remaining habi- and about half of the lowland forests had been destroyed (Ste- tat cover (Andre´n, 1994). Below this value, there is often an vens et al., 1988; McGlone, 1989). abrupt increase in the gap structure of landscapes, a marked The first European settlers in the early 19th Century ini- discontinuity in dispersal between isolated sub-populations tially cleared forest at a relatively slow rate (Arnold, 1994). and a sharp decline in the probability of metapopulation per- However, with a growing population, improvements to sistence (Hanski, 1998; Hanski and Ovaskainen, 2002), at roads and a new rail system, large-scale clearance of forest which point an extinction threshold has been passed. The on the plains began in earnest in the 1870s (Arnold, 1994). extinction threshold compounds the risks posed by forest Early New Zealand landholders were required by law to destruction, as even a small loss of habitat near the threshold improve their land, and many achieved this via the simple may result in a precipitous decline in the probability of spe- act of burning the forest (Salmon, 1975). Primary forest cies persistence (With and King, 1999). clearance continued into the mid-20th century, and after Third, many invasions do not propagate through undis- the Second World War increasing amounts of forest in the turbed habitats and invasion risk increases disproportion- mountain ranges were converted to farmland (Stevens ately following anthropogenic habitat loss (May and Norton, et al., 1988) or fast-growing exotic plantations (Fleet, 1986). 1996; Lonsdale, 1999; Hobbs, 2001). A growing body of evi- The net result of Maori and European exploitation of New dence now shows that landscape context may be a critical Zealand’s indigenous forest was the destruction of approxi- determinant of local-scale predation intensity exerted by mately three-quarters of the forest, reducing it from 82% to invasive species (Robinson et al., 1995; Hartley and Hunter, 23% of the land surface area (Fleet, 1986; Leathwick et al., 1998; Ohlemu¨ ller et al., 2006). Furthermore, forest edges are 2003b, 2004). focal points for invasions of remnant habitat (Wiser et al., 1998). In small fragments that consist completely of edge- 1.1. Conservation implications of historical deforestation affected habitat, invasive plant species like Tradescantia fluminensis (Commelinaceae) can influence the long-term via- In New Zealand, it is often assumed that clearing of primary bility of forest remnants by dramatically reducing the species native forests is no longer of significant concern and that richness, abundance and survival rates of native seedlings the conservation issues associated with forest loss are no (Standish et al., 2001), as well as reducing invertebrate diver- longer relevant to threatened species management (Craig sity and altering invertebrate community composition (Toft et al., 2000; Clout, 2001). However, the historical effects of for- et al., 2001). est loss are frequently recognised as one of the greatest threats to endangered species (Tilman et al., 1994; Sala 1.2. Goals of this study et al., 2000), and there is an emerging recognition of the role that historical landuse change has in structuring present- There has been no extensive analysis of patterns of defores- day species assemblages (Harding et al., 1998). Moreover, tation in New Zealand, nor an investigation into the natural theoretical developments on the nature of extinction debts and anthropogenic drivers of those patterns. The purpose of (Tilman et al., 1994), extinction thresholds (With and King, this study was to determine historical patterns of deforesta- 1999), and synergistic interactions between forest loss and tion and describe current patterns of forest fragmentation. invasive species (Didham et al., 2005a,b; Ewers and Didham, These variables were assessed in relation to major topo- 2006), advise caution in adopting the overly simplistic inter- graphical and climatic features. Furthermore, because hu- pretation that historical deforestation is not relevant to pres- mans have been the dominant force behind the loss of ent-day conservation. native forests worldwide, correlations between human popu- First, extinction debts create a long-term decline in spe- lation density, recent changes in human population size and cies richness for generations following the isolation of hab- proximity to highways (a spatial indicator of human activity) itat remnants, and illustrate the long-term conservation were assessed as potential drivers of forest loss. Finally, the implications of historical forest loss. The term extinction amount