National Wilderness Inventory
Australia: Our national stories
Australian Heritage Commission, 2003
ISBN 0 642 23561 9
4. Creating The Wilderness Database
Although NWI analysis is standardised, there can be variation in procedures at certain stages in the creation of the wilderness database. This variation is related to regional differences in the composition and availability of primary data. This section of the handbook describes the way in which these principles and the primary data were used in order to implement the NWI baseline survey. Descriptions of the primary data used in analysis are mainly dealt with in Section 3 (The Primary Database).
4.1 Wilderness Database Framework
Establishment and maintenance of the wilderness database is a key aspect of the NWI. Survey procedures allow wilderness quality data to be compiled for any area at any level of resolution. However, for operational convenience and to facilitate access and use of the database a standard spatial framework has been used.
This framework consists of individual survey units tied to the national 1:250 000 topographic mapping grid. The structuring of wilderness survey units to conform with standard national topographic mapping allows direct comparison between general mapped information and wilderness quality. It also permits the scope of wilderness survey work to be widened at some future time within the same survey framework.
The process for establishing the analysis framework is shown in Figure 4.1. The survey unit framework is shown in Figure 3.1.
The wilderness quality database of the baseline NWI survey is created by establishing sampling point lattices over those areas selected for analysis, on a survey unit by survey unit basis. It is to these lattices that the entire wilderness database is permanently assigned. Wilderness quality may be surveyed across a full spectrum of natural land conditions as shown in Figure 1.1. However, the baseline NWI survey focuses on natural areas.
As these are all potentially subject to survey the first step in undertaking wilderness quality analysis is therefore determining which natural areas should be examined. For the NWI baseline survey, survey area selection has generally been pragmatic. All areas mapped as natural cover in the primary data compilation process formed the basis for survey area selection. Sampling lattices were established across all these areas.
As is evident from descriptions in Section 3.3.2, which deals with the compilation of the land_cov primary data, criteria used in mapping these areas varied considerably from study area to study area in accordance with local practicalities and data availability.
Consistent with the criteria established for mapping natural cover, areas selected for survey in NWI baseline survey work vary from a 5 000 hectare minimum size requirement for the pilot survey work conducted in Tasmania (now being updated), through to a 20 hectare requirement for parts of agricultural South Australia. NWI techniques allow the point spacing of wilderness database lattices to be varied according to primary data suitability and the scale of investigations.
However, for the purposes of baseline survey work, a dual sampling standard has been adopted. For more complex environments and more developed regions a sampling lattice of 500-metres easting by 500-metres northing is used. The remaining parts of the continent are being sampled at 1000-metres. Lattice point resolution is assigned on a 1:250 000 survey unit basis. The sampling resolution of survey units is illustrated in Figure 4.2
Although the wilderness database comprises lattices of sampling points which are referenced to the standard 1:250 000 survey unit tiles, these lattices are stored as discrete ARC/INFO coverages, not as layers within the library manager. The reason for this is that sampling lattices extend for a nominal distance beyond equivalent library tile limits, thus ensuring reasonable presentation of results across tile boundaries.
The lattices are named according to their respective 1:250 000 survey tile (e.g. SI5409), however they are referenced in this handbook using the generic name 'lattice'. Primary data are processed using a series of ARC/INFO AML programs to produce a wilderness quality database in the standard form of a lattice of points, each point with an associated set of attributes describing remoteness and naturalness at that site (AML is described in Section 2.4). The content and structure of these lattices has, over the course of the baseline survey, varied to a minor extent as adjustments have been made to the methodology and allowances made for the special characteristics of particular study areas. Following the compilation of the baseline survey, the structure of these lattices has been standardised.