**1. Introduction**

200 Biodiversity Loss in a Changing Planet

Vicentini, K. R. F. (1999). History of Fire in the Cerrado: A Pollen Analysis. 235p. *Thesis* (PhD

Whelan, R. J. (2001). *The Ecology of Fire*. Cambridge University Press, ISBN 9780521328722,

in Ecology), Federal University of Brasília (UNB), Brasília/Brazil.

Cambridge, England.

#### **1.1 Biodiversity and habitat provision in New Zealand**

The Millennium Ecosystem Assessment (MEA) found that over the past 50 years, natural ecosystems have changed more rapidly and extensively than in any other period of human history (Millennium Ecosystem Assessment, 2005). In the 30 years after 1950, more land was converted to cropland than in the 150 years between 1700 and 1850, and now one quarter of the earth's surface is under cultivation. In the last decades of the twentieth century, approximately 20% of the world's coral reefs have disappeared and an additional 20% show serious degradation. Of the fourteen major biomes in the world, two have lost two thirds of their area to agriculture and four have lost one half of their area to agriculture. The distribution of species has become more homogeneous, primarily as a result of species introduction associated with increased travel and shipping. Over the past few hundred years, the species extinction rate has increased by a thousand times, with some 10–30% of mammal, bird, and amphibian species threatened with extinction. Genetic diversity has declined globally, particularly among cultivated species.

A framework of ecosystem services was developed to examine how these changes influence human well-being, including supporting, regulating, provisioning, and cultural services (Millennium Ecosystem Assessment, 2003). While overall there has been a net gain in human well-being and economic development, it has come at the cost of degradation to many ecosystem services and consequent diminished ecosystem benefits for future generations. Many ecosystem services are degrading because they are simply not considered in natural resource management decisions. Biodiversity plays a major role in human wellbeing and the provision of ecosystem services (Diaz et al., 2006). For example, natural ecosystems provide humans with clean air and water, play a major role in the decomposition of wastes and recycling of nutrients, maintain soil quality, aid pollination, regulate local climate and reduce flooding.

New Zealand has been identified as a biodiversity hotspot (Conservation International, 2010). Located in the Pacific Ocean, south east of Australia, New Zealand covers 270 thousand square kilometres on three main islands (North, South and Stewart Island). It has a wide variety of landscapes, with rugged mountains, rolling hills, and wide alluvial plains. Over 75 percent of New Zealand is above 200 meters in altitude, reaching a maximum of

Provision of Natural Habitat for Biodiversity: Quantifying Recent Trends in New Zealand 203

combined with field work (Landcare Research, 2011). The survey scale was approximately 1:50,000 and had a nominal date of mid-1970. The legend included 42 vegetation classes, of which six were indigenous forests (coastal, kauri, podocarp-hardwood (lowland or midaltitude), *nothofagus* (lowland or highland), and hardwood) and three were indigenous grass classes (snow tussock, red tussock, and short tussock). The Vegetative Cover of New Zealand was produced at the scale of 1:1,000,000 primarily from the NZLRI (Newsome, 1987). The small scale required mixed vegetation classes to be used, such as "grassland-

The Land Cover Database (LCDB) was derived by photo-interpretation of satellite imagery and has nominal dates of 1995–96 for LCDB1 and 2001–2002 for LCDB2 (Ministry for the Environment 2009). Indigenous classes included tussock grassland, manuka/kanuka, matagouri, broadleaved hardwoods, sub-alpine shrubland, and mangroves; however, different indigenous forest classes were not delineated and were lumped into one class of indigenous forest. Walker et al. (2006) used the LCDB to look at changes to natural habitat between 1995–96 and 2001–2002. They concluded that much of the highland natural habitats had been preserved since pre-Maori times, but also that much of the natural habitat of lowland ecosystems had been lost and continues to be lost. Limitations in the LCDB prevented reliable analysis of the changes in indigenous grassland, wetlands, and

The recently completed Land Use Map (LUM) has extended the date range for indigenous forest to between 1990 and 2008 (Ministry for the Environment, 2010). LUM is primarily helping New Zealand meet its international reporting requirements under the Kyoto Protocol. It tracks and quantifies changes in New Zealand land use, particularly since 1990. For this purpose, it produced national coverages for 1990 and 2008 of five basic land cover classes (indigenous forest, exotic forest, woody-grassland, grassland, and other), from

More recent work by Weeks et al. (in prep) has improved the accuracy and extended the analysis to between 1990 and 2008 on tussock grasslands. Ausseil et al. (2011) have improved the accuracy of wetland mapping and identified changes since pre-European time. These recent analyses, together with the LUM, permit a synthesis of information for assessing recent trends of natural habitat provision in New Zealand. This chapter presents this synthesis and describes a national measure of habitat provision for biodiversity. We look at New Zealand's natural habitat changes from pre-Maori to the present, and also at recent trends. We will focus this chapter on three natural ecosystems: indigenous forest, indigenous grasslands, and freshwater wetlands. The measure of habitat provision will combine information on current and historical extents with a condition index to quantify

Indigenous forests in New Zealand are generally divided into two main types. The first is dominated by beech trees (*Nothofagus*), and the second generally comprises an upper coniferous tier of trees with a sub-canopy of flowering trees and shrubs (the broadleaved species) (Wardle, 1991). However, these two types are not mutually exclusive and mixtures are common. Lowland podocarp-broadleaved forests are structured like forests of the

forest" or "forest-scrub".

satellite imagery.

stress and disturbance.

**2. Indigenous forests** 

regeneration of shrublands to indigenous forest.

**1.3 Proposed assessment of natural habitat provision** 

3,700 meters on Mount Cook. Climate is highly variable and has played a key role in biodiversity distribution (Leathwick et al., 2003).

As New Zealand has been an isolated land for more than 80 million years, the level of endemism is very high, with more than 90% of insects, 85% of vascular plants, and a quarter of birds found only in New Zealand (Ministry for the Environment, 2007). One of the most notable characteristics of New Zealand's biodiversity is the absence of terrestrial mammals, apart from two bat species, and the dominance of slow-growing evergreen forest. New Zealand's indigenous biodiversity is not only unique within a global context – it is also of major cultural importance to the indigenous Maori people. Maori have traditionally relied on, and used, a range of ecosystem services including native flora and fauna for food, weaving, housing, and medicines.

The isolation of New Zealand has preserved its unique biodiversity, but also rendered the biodiversity vulnerable to later invasion. When Maori migrated from the Pacific Islands, circa 700 years ago, predation upon birds began and much lowland indigenous forest was cleared, especially in the South Island. Rats and dogs were also introduced. The birds, having evolved in an environment free of predators, were susceptible to disturbance and many began to decline to the edge of extinction. When Europeans arrived in the early 19th century, they extensively modified the landscape and natural habitats. Large tracts of land were cleared and converted into productive land for pastoral agriculture, cropping, horticulture, roads, and settlements. Only the steepest mountain land and hill country was left in indigenous forest and shrubland. Swamps were drained and tussock grasslands were burned. Not only was the natural habitat significantly altered, but a large range of exotic species were introduced, including deer, possums, stoats, ferrets, and weasels, causing a rapid decline in native birds and degrading native forest. Other introduced plants and animals have had significant effects in the tussock grasslands and alpine shrublands, most notably rabbits, deer, and pigs, and the spread of wilding pines, gorse, broom, and hieracium. Despite significant efforts to control weeds and pests and halt the loss of natural habitat, around 3,000 species are now considered threatened, including about 300 animals, and 900 vascular plants (Hitchmough et al., 2005).

The Economics of Ecosystems and Biodiversity study (TEEB) suggested that it is difficult to manage what is not measured (TEEB, 2010). To prevent further biodiversity loss, decisionmakers need accurate information to assess and monitor biodiversity. However, biodiversity assessment is not a trivial task. As defined by the Convention on Biological Diversity (CBD), biodiversity encompasses "the variability among living organisms from all sources including, inter alia, terrestrial, marine and other aquatic ecosystems and the ecological complexes of which they are part; this includes diversity within species, between species and of ecosystems" (CBD, 1992). Conceptually, biodiversity is a nested hierarchy comprising genes, species, populations, and ecosystems. In order to assess status and trend, these multiple levels need to be assessed simultaneously. Noss (1990) suggested a conceptual framework with indicators providing measurable surrogates for the different levels of organisation. Loss of extent is one of the many indicators in this framework, and it has been widely used internationally in reporting to the CBD (Lee et al., 2005). It is relatively easy to report, and has been recognised as one of the main drivers for biodiversity loss (Department of Conservation [DOC] and Ministry for the Environment [MFE], 2000).

#### **1.2 Previous assessments in natural habitat**

Several national surveys of vegetation cover have been completed. The New Zealand Land Resource Inventory was derived by stereo photo-interpretation of aerial photographs combined with field work (Landcare Research, 2011). The survey scale was approximately 1:50,000 and had a nominal date of mid-1970. The legend included 42 vegetation classes, of which six were indigenous forests (coastal, kauri, podocarp-hardwood (lowland or midaltitude), *nothofagus* (lowland or highland), and hardwood) and three were indigenous grass classes (snow tussock, red tussock, and short tussock). The Vegetative Cover of New Zealand was produced at the scale of 1:1,000,000 primarily from the NZLRI (Newsome, 1987). The small scale required mixed vegetation classes to be used, such as "grasslandforest" or "forest-scrub".

The Land Cover Database (LCDB) was derived by photo-interpretation of satellite imagery and has nominal dates of 1995–96 for LCDB1 and 2001–2002 for LCDB2 (Ministry for the Environment 2009). Indigenous classes included tussock grassland, manuka/kanuka, matagouri, broadleaved hardwoods, sub-alpine shrubland, and mangroves; however, different indigenous forest classes were not delineated and were lumped into one class of indigenous forest. Walker et al. (2006) used the LCDB to look at changes to natural habitat between 1995–96 and 2001–2002. They concluded that much of the highland natural habitats had been preserved since pre-Maori times, but also that much of the natural habitat of lowland ecosystems had been lost and continues to be lost. Limitations in the LCDB prevented reliable analysis of the changes in indigenous grassland, wetlands, and regeneration of shrublands to indigenous forest.

The recently completed Land Use Map (LUM) has extended the date range for indigenous forest to between 1990 and 2008 (Ministry for the Environment, 2010). LUM is primarily helping New Zealand meet its international reporting requirements under the Kyoto Protocol. It tracks and quantifies changes in New Zealand land use, particularly since 1990. For this purpose, it produced national coverages for 1990 and 2008 of five basic land cover classes (indigenous forest, exotic forest, woody-grassland, grassland, and other), from satellite imagery.

### **1.3 Proposed assessment of natural habitat provision**

More recent work by Weeks et al. (in prep) has improved the accuracy and extended the analysis to between 1990 and 2008 on tussock grasslands. Ausseil et al. (2011) have improved the accuracy of wetland mapping and identified changes since pre-European time. These recent analyses, together with the LUM, permit a synthesis of information for assessing recent trends of natural habitat provision in New Zealand. This chapter presents this synthesis and describes a national measure of habitat provision for biodiversity. We look at New Zealand's natural habitat changes from pre-Maori to the present, and also at recent trends. We will focus this chapter on three natural ecosystems: indigenous forest, indigenous grasslands, and freshwater wetlands. The measure of habitat provision will combine information on current and historical extents with a condition index to quantify stress and disturbance.
