How do you develop guidelines?

The ANZECC and ARMCANZ 2000 guidelines recommend two main approaches to developing guidelines: guidelines based on direct impact studies, and guidelines based on a referential approach. A detailed discussion of guideline development can be found in the ANZECC 2000 Guidelines – see in particular pages 3.3-6 to 3.3-9.

Guidelines based on direct impact studies

This approach is best suited to development of guidelines for in-stream pressure indicators. It relies on development of a relationship between the in-stream pressure and the EV in question. Examples of this would include impacts of salinity on crops; impacts of blue green algae on humans or stock; impacts of toxicants or low dissolved oxygen on biota; and impacts of reductions in light penetation on seagrass growth.

Based on known relationships, we can determine what is an acceptably safe level of the in-stream pressure indicator and use this as a guideline value. Developing guidelines using this approach requires a high level of technical expertise and resources, and this approach is therefore usually confined to universities or other research organisations.

Direct impact studies are the preferred approach to developing guidelines. For ‘human’ use EVs, most guidelines are based on this approach. However, such information is often lacking or, in the case of value/use indicators, the direct impact approach is inappropriate. In these situations, we can apply the referential approach.

Guidelines based on a referential approach

This approach is best suited to value/use indicators (e.g. macroinvertebrate diversity) or to in-stream pressure indicators (e.g. nutrients) for which there are no simple direct impact relationships. Under this approach, guideline values for a particular indicator are determined with reference to the condition of that indicator in a system that is relatively undisturbed (i.e. reference condition). For example, dissoved oxygen guidelines would be based on typical dissolved oxygen values in a relatively undisturbed system.

The actual guideline value is calculated on the basis of some acceptable departure from reference condition. What constitutes an acceptable departure can be debated at length, however the default acceptable departure suggested in the ANZECC 2000 Guidelines (Section is that the guideline value should be based on the 20th and/or 80th percentile (whichever is most appropriate for the indicator) of values at the reference site.

This approach is most commonly applied to develping guidelines for ecosystem protection.

This referential approach can be applied not only to traditional water quality indicators but also to biological and habitat indicators. It could also be applied to load indicators, for example, load guidelines could be set in terms of some small departure from loads characteristic of relatively pristine catchments.

The referential approach is amenable to use by regional groups. Provided that reference data is collected according to agreed protocols (number of reference sites, number of data values, time period of collection, adequate quality assurance) then such groups can use their own monitoring data to develop local guidelines.

Additional resources

Water Quality Targets Online
This has been designed to assist regional groups to identify the EVs of water and to set appropriate water quality guidelines for their catchments/region. Water Quality Targets Online uses default guidelines derived from the Australian and New Zealand Guidelines for Fresh and Marine Water Quality (2000) (Department of Environment and Heritage).

Water quality guidelines Hart, B.T. (2001). Water quality guidelines, In ‘Burden, F.R., Forstner, U., Guenther, A. & McKelvie, I.D. (eds.) Handbook of Environmental Monitoring. McGraw Hill, New York.