| 1.1 Background
1.2 Approach
1.3 Study Area
1. Introduction
1.1 Background
The National Parks and Wildlife Service (NPWS) were contracted by
the Lower Hunter and Central Coast Regional Environment Management
Strategy (LHCCREMS) to carry out vegetation survey and mapping across
the region.
The LHCCREMS is an initiative of seven local councils (Gosford,
Wyong, Lake Macquarie, Newcastle, Port Stephens, Maitland and Cessnock)
to develop improved consistency and responses to natural resource
and environmental management issues at a regional scale. In mid-1996
the group embarked on the development of the Regional Biodiversity
Conservation Strategy (RBCS), in partnership with key State agencies
and community networks.
The strategy and its implementation program will:
* Be guided by comprehensive baseline mapping of the biodiversity
values in the landscape.
* Identify the conservation status and significance of all land
in the study area
* Provide input into future planning and development strategies
for the region.
* Seek to identify and facilitate implementation of an innovative
mix of approaches to conservation of natural areas of significance
outside the formal reserve system
The vegetation survey and mapping study discussed in this report
form the first major part of the data collection and analysis effort
for this project.
The primary aims of the study were to:
1. produce a detailed cross tenure map of the distribution of the
vascular plant communities in the Lower Hunter and Central Coast
region;
2. provide estimates of the distribution of the plant communities
prior to the arrival of Europeans (pre-1750);
3. utilise project methodologies which are consistent with the Comprehensive
Regional Assessment (CRA) work conducted on public land in the region
so that the two surveys can be amalgamated into a single database
to greatly improve the resolution and usefulness of each; and,
4. Provide a context for future studies so that data collection
is compatible with, and contributes to the on-going development
of a comprehensive regional database.
1.2 Approach
Environmental issues consistently transcend administrative
boundaries. The establishment of the LHCCREMS initiative is recognition
that these issues require a regional context to assist the environmental
planning process. Similar conclusions have been reached (NFPS, 1992)
in an effort to address land use conflicts involving the timber
industry.
The management and protection of biodiversity in the Lower Hunter
and Central Coast, as a pre-requisite to future planning and sustainable
development of the region, is one of the key objectives of the Lower
Hunter & Central Coast Regional Environmental Management Strategy
implementation program (LHCCREMS Progress Report, 1998).
The management and protection of biodiversity in the study area
is one of the key environmental objectives of the LHCCREMS program
(LHCCREMS, 1998). Outside of the invertebrate kingdom, one of the
richest reservoirs of biodiversity lies in the plant world. However,
managing individual species of biota is not a realistically achievable
task given the many thousands present in a study area of this size.
Such complexities have led to the development of biodiversity surrogates
(Ferrier and Watson 1997, JANIS, 1996) to help explain and encapsulate
the variation in biota across a landscape. The development and application
of these concepts has underpinned the recent CRA program to guide
conservation management priorities on public lands. Similar needs
exist to meet LHCCREMS objectives, and hence, classification and
mapping of vegetation communities across all land tenures is required.
Assemblages of floristic species or vegetation communities
have been used to derive surrogates for biodiversity in regional
analyses (Keith & Bedward, 1999; NPWS, 1999(a), NPWS, 1999(b)).
Consequently, considerable discussion on approaches to defining
forest ecosystems and vegetation communities preceded new assessment
work for the CRA program (FEWG, 1997). FEWG (1997) presented the
following broad approach to ensure consistency of method, without
specifying consistency of classification.
The preferred approach to vegetation survey and mapping
should be based on field data and involve the following sequence
of steps:
1. Classification of vegetation assemblages using species composition
data from field samples
2. Characterisation of the relationships between these assemblages
and remote data variables (eg. Aerial photo pattern, climatic
variables, terrain variables and geology);
3. Mapping (spatial interpolation) of assemblages using the empirically
established relationships with remote data variables; and,
4. Evaluation of map reliability
Payne (1998) undertook an audit of existing biodiversity related
studies and noted a large variation in vegetation sampling methods
previously employed in the LHCCREMS study area. Classification of
vegetation has generally been carried out using qualitative assessments
of the patterns observable in aerial photos. This may take the form
of canopy (tree) species important for timber management (SFNSW
Forest Typing) through to canopy species in combination with prominent
features of the understorey (Payne, 1998; Biosis, 1998; Benson and
Fallding, 1986; Benson, 1986).
This has been a prohibitive factor to the establishment of a uniform
classification system across the study area. Without field data
it is difficult to objectively compare one vegetation classification
from another. The need for field data that describes the floristic
composition of a site is now well recognised (Sun et al. 1997; NPWS,
1997). These methods are now recommended in numerous local council
biodiversity survey guidelines (Lake Macquarie, 1998; Wyong, 1998).
The survey of sites that provide an inventory of species occurrence
and a measure of abundance using quantitative (actual counts of
species numbers) or semi quantitative measures of species abundance
(eg. Braun-Blanquet scales) presents the opportunity to numerically
analyse field information.
Site data has been collected for this project for two reasons. Firstly
it provides the means to develop an objective and repeatable method
of classifying vegetation. Secondly, it establishes a foundation
for the ongoing development of a biodiversity data resource, where
information can be revisited and used for a multitude of other purposes
(eg. fauna habitat descriptions, finer scale community analysis
etc).
The aim of vegetation classification is to describe homogenous patterns
in the landscape, thereby representing the floristic variation present
in the study area. Faith et al. (1987) suggests that any analysis
of floristic variation between survey sites must be based on some
objective measure of floristic composition. To this end, a numerical
classification system describing compositional dissimilarity between
sites has been used in this project.
Mapping the patterns in vegetation is always achieved using remote
data and is always a model or extrapolation of features in the landscape.
As described above, patterns have most often been detected using
air photos, in combination with field experience and judgements
of the interpreter. However, this process is limited by what is
visible, and so important features describing the understorey are
likely to be overlooked. Further, relationships used to guide the
mapping of a Community are not explicit making the extrapolation
of vegetation communities to cleared environments difficult.
Computer modelling using Geographic Information Systems has been
used in this project to describe patterns in the distribution of
vegetation communities across extant vegetation and cleared land.
A decision tree method was employed to develop statistical relationships
between remote spatial variables and sites that were used to describe
each Community. The mapping of communities drew on relationships
with a wide variety of spatial variables that describe terrain,
climate, substrate and vegetation structure.
1.3 Study Area
The seven participating councils define the LHCCREMS study area.
These are Gosford, Wyong, Lake Macquarie, Newcastle, Port Stephens,
Maitland and Cessnock (Map 1). This encompasses a total land area
of over 563,000 hectares. Over 35% of the original vegetation cover
has been cleared.
The study area lies within the greater Sydney Bioregion (Thackway
and Creswell, 1995), although portions of Port Stephens and Maitland
fall within the North Coast Bioregion. The Sydney Bioregion is characterised
by the sandstone valleys and plateaux, which are found in the southern
and western ranges of the study area. A large coastal plain extends
along the coast from the foothills in Gosford to Port Stephens.
The Hunter Valley marks a transition zone for many plant and animal
species between the sub tropical influences of the north to the
cooler, less fertile conditions of the south.
The terrain of the study area has a large influence on the vegetation
patterns in the study area. The Watagan Range and Escarpment exerts
an orographic influence on rainfall patterns, providing sufficient
moisture to support the major areas of wet sclerophyll forest and
rainforest in the region. The drier plains and valleys contain dry
sclerophyll forests and woodlands. The plains also support large
areas of impeded drainage where wetland ecosystems are sustained.
Coastal aeolian sand features predominate in the Tomago region and
coastal zone supporting a variety of heaths, swamps and woodland
communities
An expanding urban environment along the eastern coastal plains
dominates in areas of Newcastle, Wyong, Lake Macquarie, Gosford
and Port Stephens. Large coal mining activities predominate across
the Lower Hunter Valley. The Eastern Ranges remain available for
forestry activities in the SFNSW Morriset Management Area. Conservation
reserves have been established across large areas of the sandstone
plateaux and in isolated pockets along the coastal plain. The most
recent declarations have arisen from the forestry reform process,
resulting in new reserves on the Hunter Valley Floor, Watagan Range
and the Medowie plains.
Map 1: LHCCREMS Study Area
High resolution image
version [660kb]
Map 2: LHCCREMS Land Tenure
High resolution image
version [700kb]
Top
|
|
