Geoscience Reference
In-Depth Information
monitoring data from near-natural sites and
modelling simulations (Sandin and Verdonschot,
2006; Hawkins
et al
., 2010; Nestler
et al
., 2010).
These data typically provide short- to medium-
term records (typically
20 years) of instream
community and habitat composition (Monk
et al
.,
2007). In most instances, quantitative records of
instream ecological dynamics are only available for
the time period after the most significant human
impact to river systems. As a result many 'reference
sites' are typically located in headwater regions
(upstream of the most severe human impacts),
while lowland and somewhat larger sites, which
have experienced the most severe degradation, are
poorly represented. In some lowland river types,
even headwater reference sites may be unavailable
owing to large-scale land-use intensification across
whole catchments.
There are several methods available for defining
reference conditions for macroinvertebrate and
diatom communities in river systems, such as
RIVPACS (Wright, 2000) and DARES (Kelly
et al
.,
2008). RIVPACS (River InVertebrate Prediction
and Classification System) assesses river quality by
predicting the macroinvertebrate fauna expected to
be found in the absence of major environmental
stress and comparing this with the observed
fauna at each site (Wright, 2000). DARES
(Diatoms for Assessing River Ecological Status)
uses modern diatom assemblages to determine
reference conditions based on comparisons across
a number of sites. However, while diatoms can
be used to determine the level of impact, they
cannot at present be used in rivers for assessing
ecological status, as required for the WFD (Kelly
et al
., 2008). In addition, neither RIVPACS nor
DARES can define a reference condition for a
riverine site in the absence of high quality/low
impact reference sites of a 'similar' type. As a result
there is a need for alternative approaches which
draw on palaeoecological and palaeoenvironmental
techniques. For many lowland rivers, with a long
history of management and human impacts, this
may be the only potentially viable method available
for the identification of historic instream 'reference
conditions'.
Palaeoecological techniques are well established
and applied in the study of lakes across the globe,
although their application to river ecosystems has
been limited in comparison (notable exceptions
being: Ogden, 2000; Ogden
et al
., 2001; Brown,
2002; Smith and Howard, 2004; Greenwood
et al
., 2006; Davis
et al
., 2007; Kelly
et al
.,
2008; Howard
et al
., 2009; Reavie and Edlund,
2010). It has been demonstrated that it is
possible to identify lentic and lotic sediment
deposits before the first evidence of human
activity and use these to define 'pristine' reference
conditions (Bennion
et al
., 2011). However, this
and other research has also clearly recognized
that pristine systems are not static and that
natural variability occurs over short (inter-annual
to decadal) and long (centennial to millennial)
time-scales. It is also increasingly acknowledged
that a 'pristine' reference condition is not always
the most useful, appropriate or achievable in
the study of lakes or rivers and that a more
pragmatic approach may be required. Bennion
et al.
(2011) proposed the use of a 'reference' concept
that incorporates: (1) the extent of degradation
recorded contemporaneously compared with a
historic state/condition; and (2) the potential for
recovery (natural or associated with restoration)
in the absence of human impacts. Based on these
criteria a range of reference conditions can be
identified across many catchments at a regional
scale, prior to significant industrial or agricultural
events throughout the period of human impact.
These conditions may vary from specific periods
(e.g. industrial revolution - 1860-1850; or the
periods of agricultural intensification - e.g. post-
1945 in the UK), through to dates when disruption
of longitudinal or lateral connectivity within the
fluvial environment occurred (e.g. construction of a
dam or mill, or land drainage activities). As a result
it may be more appropriate to consider multiple
reference conditions as a series of benchmarks
rather than a target for restoration activities.
Palaeoecology and habitat
restoration
Advances in the field of palaeoenvironmental and
palaeoecological research have centred primarily
on
lakes
and
the
development
of
quantitative
