Abstract
SummaryThe concept of using a single index to encapsulate the ‘ecological condition’ or ‘quality’ of native vegetation is reviewed and critiqued, along with the history of the adoption of such indices by governments at the State and Federal levels in Australia. Despite the value judgements inherent in such a concept, these indices provide a currency that is useful for comparing sites for purposes such as to provide a nexus between vegetation proposed to be removed under permit and other vegetation that will be protected or improved as an ‘offset’. The oldest of the reviewed indices – Habitat Hectares, from the Victorian Government – has been the subject of more scrutiny than others but there has been negligible change to it in response to recommendations made. A recommendation of particular significance is for values of the index to respond smoothly to changes in continuously variable observables like foliage cover, rather than discretising the data into broad class intervals. The step‐wise jumps in index values that result from discretisation can have the effects of creating false indications of difference between sites or over time and masking actual differences. The index's value for monitoring change is thereby impaired. The Biodiversity Assessment Method of New South Wales is the only index reviewed here that avoids discretisation. However, many of the other recommendations that have been made for changes to Habitat Hectares were adopted in 2003 for a simplified variant of that method – the ‘Land manager self‐assessment method’. Therefore, this variant has been used as the basis for a new index – HH2.0 – that retains the simplifications but does not discretise continuous variables or allocate scores in a step‐wise manner. The process of translating field observations into scores is automated by a Microsoft Excel workbook, saving time and reducing human error. HH2.0 will suit landowners, land managers, local government and regional authorities, for non‐regulatory purposes such as determining what planning protection to apply to an area or prioritising vegetation management among different areas. It requires significantly less gathering of field data than Habitat Hectares, but like that method, it places weight on vegetation's similarity to a long‐undisturbed state. No restriction is made on whether it is applied to plots of specified size or shape. A pre‐existing set of observational data has been used to retrospectively determine scores under HH2.0 and compare them with Habitat Hectares scores. The data came from 37 sites spanning a range of vegetation types and ecological condition. The root‐mean‐square difference between the two methods' scores is 5.1 points (out of a theoretical maximum of 100). For reference, Tolsma and Newell (2003) concluded that a difference of 20 points between two Habitat Hectares assessments cannot be relied upon to indicate a genuine difference in ecological condition. A larger data set would be desirable but even if there were no comparison at all between indices (as normal for a new index), that would not invalidate the use of HH2.0 for its intended non‐regulatory purposes. HH2.0 is designed for use in Victoria but the changes that have been made compared with Habitat Hectares could be applied to create faster, simpler variants of indices used in other jurisdictions.