Climate Change: Defining Reference Conditions and Restoring Freshwater Ecosystems - Climate Change Impacts on Freshwater Ecosystems - page 205

disturbance; (ii) historical condition, to refer to a condition at some previous
time; (iii) least disturbed condition is reserved for sites in the landscape having
the best available physico-chemical and biological conditions; and (iv) best
attainable condition is equivalent to the expected ecological or the least disturbed
condition if the best possible management practice were in use for some time.
Many approaches are used to establish reference conditions (e.g. Stevenson et al .
2004). The most common can be grouped into four categories: (i) spatial approaches
such as survey; (ii) temporal approaches such as contemporary time series, historical
data and palaeo-reconstruction; (iii) modelling approaches such as hindcasting; and
(iv) expert judgement. In areas where land use has not drastically altered the
landscape, the identification of reference conditions is rather straightforward and
spatial methods are frequently used. In such areas, use of survey data is common
since the approach either explicitly (sites are sampled to include among-year
variability) or implicitly (space-for-time substitution) includes natural variability.
Another reason for the popularity of using survey data is transparency - the
definition of what constitutes a reference condition is established a priori .
In many areas of Europe and elsewhere, humans have, however, extensively
altered the landscape over long periods and hence reference conditions cannot
adequately be determined using space-for-time approaches. When contemporary
reference sites are lacking, models and hindcasting are commonly used to establish
a reference condition. For instance, relationships between response and predictor
variables can be used to predict the expected reference condition (e.g. community
composition or palaeo-reconstruction of water chemistry and biota) in the
absence of stress (e.g. Wright 1995).
Recognition that even relatively pristine systems may change a great deal,
periodically cycling through different successional states and exhibiting erratic or
unexpected changes in system behaviour (Holling 1992), has been valuable in
promoting new approaches to resource management and for establishing
reference conditions and ecological targets. The use of palaeo-reconstruction and
contemporary time series data are two temporal methods that may capture the
dynamic nature of aquatic ecosystems. For lakes, reconstructing past conditions
is often done either directly using the remains of taxa stored in the sediment to
reconstruct an assemblage, or indirectly using taxon information to infer past
water chemistry. For example, weighted-averaging transfer functions have
allowed for quantitative inferences of nutrient enrichment, hypolimnetic oxygen,
pH and temperature (Bennion & Battarbee 2007). By contrast, the use of
sediment remains to reconstruct community assemblages directly is less frequent
due to high costs and spatial limitations (reconstructions are site-specific). In
contrast to the use of time-series data, use of historical data is often a static
measure of the reference condition; another weakness is that data availability is
often limited (e.g. only qualitative data are available), although some interesting,
novel approaches, such as the use of journals of pirates and century old cookbooks
are being used in marine reconstruction (Schrope 2006).
A problem with many of these approaches is that they are essentially reductionist,
often providing information on a single population or assemblage of organisms,
and are seldom used to reconstruct the elements of whole communities or
ecosystems. Expert judgement is one way of amalgamating many different types
of information, such as empirical data, opinion and present-day concepts,