Ecological communities are complex to classify. Each State/Territory jurisdiction applies its own system to classify ecological communities which can cause challenges when cross-referring amongst systems. They may also vary in accuracy to the on-the-ground situation, particularly if based on maps and modelling. Any reference to vegetation and mapping units as equivalent to a national ecological community, at the time of listing, should be taken as indicative rather than definitive. A unit that is generally equivalent may include elements that do not meet the description. Conversely, areas mapped or described as units other than those referred to may sometimes meet the description. Judgement of whether an EPBC-protected ecological community is present at a particular site should focus on how an area meets the description, particularly the key diagnostic characteristics for the national ecological community.
B3.2. Relationships to national and state vegetation systems
A consistent vegetation classification system for Western Australia was developed by John Beard and associated ecologists from the mid 1960s. A brief history of this project along with an updated broad-scale vegetation map are provided in Beard et al. (2013). The available state-wide vegetation datasets are hierarchical and encompass three scales (WA DPaW, 2013):
vegetation systems – a broad landscape-level classification that combines elements of vegetation and landscape types;
vegetation associations – purely vegetation units that share a consistent structure and with the same dominant species in the main vegetation layer; and
vegetation system–associations – a finer-scale classification that provides more detail about vegetation associations in the context of particular systems.
The National Vegetation Information System (NVIS) is an hierarchical and consistent system used to classify vegetation across the Australian continent. NVIS has very broadly defined Major Vegetation Groups that identify continental-scale vegetation patterns. Below this, there are up to six levels that provide progressively more detailed information, from broad class and structural formation (levels 1-2) to floristic associations and sub-associations (levels 5-6) that detail the dominant taxa within each vegetation layer. Equivalent NVIS level 5/6 descriptions have been applied to the Western Australian vegetation association and system–association data.
The most relevant datasets for this assessment are the Beard vegetation associations, coupled with equivalent NVIS level 6 descriptions, as this provides a consistent and detailed framework that can be applied over large areas and include detailed information on extent and decline. Table 4 in the main text of the Conservation Advice identifies 45 vegetation associations that correspond to the WA Wheatbelt Woodland ecological community, either entirely or in part. These corresponding woodland vegetation associations can be grouped into three categories based on their pre-European representation and extent in the WA wheatbelt (definition of categories is explained further in footnotes to Table 4).
Dominant vegetation associations that collectively account for the most common and extensive vegetation types in the wheatbelt bioregions. Twelve vegetation associations fall into this category, most of which also are unique to the wheatbelt.
Unique vegetation associations that are mostly restricted to the wheatbelt but were not dominant components of the original wheatbelt vegetation. There are 25 associations in this category.
Other minor vegetation associations that are neither dominant or unique to the wheatbelt but had more than 20 percent of their total pre-European extent present in the wheatbelt. Eight associations fall into this category.
Woodlands were extensive across southwestern WA, ranging from the higher rainfall zone of the coast and ranges into the semi-arid interior. The nature and composition of the woodlands changes over this vast range, not only due to differences in landscape and climate but also due to human impacts. The natural patterns of vegetation change, however, are not always clearly distinct and some intergradation is evident, particularly across the eastern wheatbelt into the Great Western Woodland.
The wheatbelt region falls mostly within the Transitional Rainfall botanical province of the Southwest Australian Floristic Region (Hopper and Gioia, 2004). It covers a range of vegetation types, and the most extensive vegetation associations of the wheatbelt are detailed in Table B3. Eucalypt woodlands account for a substantial proportion of the native vegetation of the WA wheatbelt, overall about 44.6 percent, but is especially high in the Katanning subregion (AVW02) where over 80 percent of the vegetation originally was eucalypt woodland. The most extensive vegetation types in the wheatbelt, other than eucalypt woodlands, are shrubland and mallee scrub communities dominated either by mallee eucalypts or species of Acacia, Casuarina or Melaleuca. These are distinguished from the WA Wheatbelt Woodlands by an absent, or at best, very sparse presence of eucalypt trees. Where any tree canopy is present, it is dominated by a range of mallee eucalypt species or by non-eucalypt tree species. Many of the most extensive associations are shrublands in which a distinct tree canopy is absent.
The vegetation to the west and south of the wheatbelt, in the Jarrah Forest and Esperance Plains bioregions mostly lies within the High Rainfall Province of the Southwest Australian Floristic Region. It comprises extensive forest, woodland and shrubland associations (Table B4). However, most of these forests and woodlands are distinguished from the WA Wheatbelt Woodland by the dominance of eucalypt species not typical of wheatbelt woodlands, notably Eucalyptus marginata (jarrah), Corymbia calophylla (marri), E. pleurocarpa (tallerack, a mallee) and E. redunca (black marlock). There also were extensive tracts of kwongkan shrublands, sometimes associated with mallee eucalypts in mallee-scrub complexes in the Esperance Plains bioregions. These are listed as a separate nationally endangered ecological community known as the “Proteaceae Dominated Kwongkan Shrublands of the Southeast Coastal Floristic Province of Western Australia”. This ecological community is distinct from the WA wheatbelt woodlands by its location further south, in the Esperance Plains bioregion, and the absence of a distinct tree canopy dominated by single-stemmed eucalypt species.
A low proportion of the vegetation, between one to six percent, comprises vegetation associations corresponding to the WA Wheatbelt Woodlands that extend into these adjacent western bioregions. In the Jarrah Forest bioregion, these are mainly woodlands with wandoo, York gum and/or yate and are included as part of the ecological community where they occur in the drier parts of the bioregion receiving less than 600 mm annual rainfall. Some yate and York gum woodlands extend into the Esperance Plains bioregion, but are not considered part of the ecological community. These occurrences appear to be scattered through the bioregion and include areas not associated with the Yilgarn craton.
Table B3. The dominant Beard vegetation associations across the wheatbelt, by IBRA subregion. Numbers indicate percentage of each subregion occupied by an association. These associations collectively account for 85 percent of total native vegetation extent, as determined by rankings of pre-European extent [most to least extensive] for all associations in each subregion. Associations are presented in order of their dominance across the wheatbelt. Boldedassociations are part of the WA Wheatbelt Woodland.