- Title
- Defining and mapping rare vegetation communities: improving techniques to assist land-use planning and conservation
- Creator
- Bell, Stephen A. J.
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2013
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- Although steeped in history, the identification and classification of vegetation communities is rarely capable of addressing the full diversity of vegetation in an area, specifically those communities that are rare or of restricted distribution. This has proven particularly problematic for land managers who are charged with the responsibility of balancing human progress with meaningful conservation. The Zurich-Montpellier school of vegetation classification, established in the early 20th Century in Europe, outlined the basic building blocks of the current-day discipline of vegetation science. Within this school, the two central themes of sampling unit (‘quadrat’) and vegetation community (‘association’) are pivotal, and provide a mechanism for establishing order in a seemingly chaotic environment. In recent decades, particularly in Australia, there has been a shift away from the ideals of the Zurich-Montpellier school, specifically that involving sampling design. Detailed, ground-based sampling, classification and mapping has transitioned into one where environmental surrogates, remotely accessed data (aerial photography, satellite imagery), and computer modelling takes precedence. At the same time, there has been an increasing demand for accurate, local-scale map products to inform land-use and conservation planning, and to satisfy legislative requirements for the protection of biodiversity. This juxtaposition of broad, regional-scale classification and mapping products onto local-scale landscapes and land conflicts is an unproductive dichotomy that requires resolution. A new paradigm for the classification and mapping of vegetation at local and regional scales is outlined in this study, incorporating new methods based on old principles to facilitate inclusion of rare and restricted communities in land-use planning. Data-informed Sampling and Mapping (D-iSM), is illustrated through three common scenarios in natural resource management: assessment of vegetation for the development industry, defining and classifying vegetation within conservation reserves, and identifying significant vegetation within a sub-regional context. All seven steps in the new paradigm are detailed for each scenario, and the results are compared to previous classification and mapping for the three study areas, highlighting considerably improved accuracies. Central to the D-iSM method is the old adage, from the Zurich-Montpellier school, know your study area well, combined with preferential (non-random) sampling to ensure a thorough and representative dataset. For larger regional, State or National contexts, there is provision within D-iSM to incorporate ‘cutting-edge’ 3-dimensional interpretation of high resolution aerial images to overcome perceived shortfalls (financial, time, access constraints) in using the technique across extensive or rugged regions. Benefits of the D-iSM method include more efficient and more representative sampling, more realistic and repeatable classifications, considerably higher user accuracy in vegetation mapping, increased ability to detect and map rare vegetation communities and ready application to a range of classification and mapping projects. These techniques are then used to define new communities (Scribbly Gum-dominated vegetation on the Central Coast of New South Wales) and to refine existing communities (Lower Hunter Spotted Gum-Ironbark Forest threatened ecological community), highlighting the extent of community biodiversity overlooked using existing methods. Implementing a more ground-based approach to classification and mapping, particularly for local- and regional-scale classification products, will significantly improve the detection and recognition of rare and restricted vegetation communities. The successful detection and mapping of rare communities is conditional upon a number of key themes in vegetation science. One of the most influential of these is that of sample selection: how sample locations are chosen within the wider environment. As is shown in this thesis, adoption of a simple change in the way that sampling is undertaken (preferential rather than random) can dramatically improve the detection and definition of rare communities. The D-iSM approach to classification and mapping encapsulates the core principles of direct sampling of observed variations, rather than relying on chance that such variations will be captured by an environmental stratification. The collection of ground-data points prior to establishing a sampling framework enables more efficient and representative sampling and results in a more reliable vegetation map with lasting relevance. Existing standards and guidelines for vegetation classification vary throughout Australia and the world. In Australia, it is suggested that a re-setting of the focus is required so that classification in general, and detection of rare communities in particular, can be more reliably documented to bring them onto an equal footing with knowledge on rare and threatened plant taxa. This shift in focus, away from environmentally stratified random sampling and towards preferential sampling, can be facilitated through the D-iSM process. Seven critical steps in thinking are outlined for this shift to occur: (1) raising the perception and value of rare vegetation communities; (2) improved use of public money in regional classifications; (3) improved reporting in the development industry; (4) improved assessment of conservation reserves; (5) review of existing threatened communities; (6) review of listing structure in threatened species legislation; and (7) continued establishment of a hierarchical classification of Australian vegetation. In conclusion, the simple conceptual framework of Top-Down, Bottom-Up information processing has been used to illustrate why this re-focusing is necessary, and to facilitate the change in thinking required for vegetation scientists in Australia. Classifications established on Bottom-Up theory will provide more reliable and accurate information than Top-Down classifications.
- Subject
- vegetation community; classification; rare; mapping; conservation planning
- Identifier
- http://hdl.handle.net/1959.13/937476
- Identifier
- uon:12567
- Rights
- Copyright 2013 Stephen A. J. Bell
- Language
- eng
- Full Text
- Hits: 990
- Visitors: 3456
- Downloads: 1900
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | ATTACHMENT01 | Abstract | 711 KB | Adobe Acrobat PDF | View Details Download | ||
| View Details Download | ATTACHMENT02 | Thesis | 7 MB | Adobe Acrobat PDF | View Details Download |