If one is aiming for ecosystem-based management, then one needs an idea of what is intended by an ecosystem.  From a management perspective, the definition needs to be politically, legally (jurisdictionally), socially as well as ecologically sensible.  This is likely to mean scientific compromise, presuming that science could give one a perfect geographical definition of an ecosystem in any one location.  Of course, in as far as science is unable to offer the "perfect" definition of ecosystem, the degree of compromise will be unknown.

Given this umbrella, the political, jurisdictional and social context is important.  Depending on these factors, one can treat an estuary and all its components as an ecosystem for the purposes of ecosystem-based management - or a bay or a section of a continental shelf.  If one's role is fisheries management, then the "ecosystem" might be defined by the area (including habitats and communities) used by the fish being managed or by the fishers pursuing the fish.  On the Great Barrier Reef, for the purposes of rezoning the entire GBRMP through the Representative Areas Program, the ecosystem was defined as the composite of all parts of the Marine Park and the World Heritage Area.  This included estuaries and intertidal areas beyond the boundary of the Great Barrier Marine Protected Area (GBRMPA) but within jurisdiction of GBRMPA's management partner, the Queensland government.

The definition of habitat and/or some kind of lower-scale "bioregion" can help managers distinguish areas within their jurisdiction or within their definition of ecosystem.  Again, from a management perspective, it is not necessarily useful to rely on a purely scientific definition of habitat or bioregion - assuming this were even available.  One might first wish to consider what management objectives one aims to achieve.  Water-quality management objectives, fisheries-management objectives and biodiversity-management objectives may require different scientific, social, political and jurisdictional factors to be considered in defining habitats or bioregions.

In my opinion, however, for management purposes, habitats are almost always at a lower geographical scale than ecosystems; and ecosystems are composed of many habitats.  The Convention on Biological Diversity (CBD) definition of habitat is consistent with this thinking:  "the place or type of site where an organism or population naturally occurs."  For example, the Great Barrier Reef "ecosystem" may be said to comprise habitats such as coral reefs, seagrass meadows, mangrove forests, Halimeda spp. beds, soft seabed, inter-reef channels, intertidal flats, algal "gardens", continental slope, estuaries, deepwater shoals and more.

Depending on the management objectives, one may use the defined habitats, for example, to determine appropriate levels of water quality depending on the impact on communities living in various habitats.  Or one may regulate fishing in various habitats depending on fishing impacts on the different fish populations in different habitats or the impacts of fishing on the habitats themselves.  If one is aiming to protect biodiversity more generally, then defining habitats or bioregions as the units within which high-level protection is required can be useful.  In the case of the Great Barrier Reef, we found it useful to describe smaller-scale, largely contiguous "bioregions" versus habitats as a basis for rezoning the GBRMP.  This was in part due to inadequate information to map all the habitats of the GBRMP and in part due to the recognition that the same habitat - for example, coral reefs - was likely to be very different in the north and south of the ecosystem and in the inshore and offshore parts of the ecosystem.

Seventy bioregions were described for the GBRMP, and each is an area where the known animal and plant assemblages and physical features are relatively distinct from the surroundings and the rest of the Great Barrier Reef World Heritage Area.  Examples of habitats (e.g., seagrass meadows or coral reefs) were considered to be more similar to each other within a bioregion than examples of those habitats occurring outside the bioregion.  Bioregions for the Great Barrier Reef ecosystem were defined at the scale of 10s to 100s of kilometers due to the level of information available, the scale over which patterns of distribution and abundance were known to change and the requirements of the management program for which bioregions were defined.  The term "bioregions", although not strictly scientifically correct, was adopted as a compromise between a biologically appropriate description and a socially meaningful one.

In summary, I consider that defining and understanding what habitats or ecosystems (or biodiversity or species or bioregions, etc.) means in the context of ecosystem-based management matters with regard to the management objectives one is aiming to achieve.  And, from a management perspective, if the definitions are robust enough to help you achieve your management objectives and are appropriate and mutually understood in the social, political, scientific and jurisdictional environment in which you have to work - then that is adequate.  No manager can, or even should, aim to define terminology for the world.

Do you have a question on ecosystem-based management?  E-mail it to us at editor@meam.net.  We will pick the best submissions and find experts to answer them.