Understanding and Managing Climate Change Effects on River Ecosystems - River Conservation and Management - page 104

suggest that these should at least match, as far
as possible, the 30-40 year timescales currently
being used for climate change projection. This, in
turn, means that few such systematic data are
available and some were collected mostly for other
purposes. Interpretation is therefore often founded
on weak inference (i.e. correlation), characterized
by assumptions, affected by confounding effects
and therefore limited in scope for definitive
interpretation. Experimental studies to investigate
processes are difficult at the scales required
to mimic those involved in climate change
effects on catchment-river ecosystems - though
smaller-scale approaches have been attempted and
potentially offer support for understanding larger-
scale observations (Hogg and Williams, 1996).
Cross-sectional analyses of rivers with contrasting
thermal regimes, caused for example by geothermal
heating,

content and extreme hydro-climatic events, such as
the 2003 heatwave (Daufresne et al. , 2007). Shifts
from cooler-water to warmer-water taxa have been
detected, even at family level, in streams on other
continents, most recently Australia (Chessman,
2010; Figure 9.2). The extent to which these
changes in different regions are driven by similar
processes is not yet clear.
The effects of inter-annual variations in
discharge on stream invertebrates are also
emerging, but interpretation of this depends
on a better understanding of the complexity of
prevailing flow effects on organisms and the ways
in which flow patterns are changing as climate
change effects occur. For example, variations in
precipitation in the Mediterranean regions of
California have been accompanied by variation
among invertebrates of contrasting tolerance
to low or high flow, leading in some locations
to sustained compositional changes (Beche and
Resh, 2007). Similarly, in Sonoran desert streams,
long-term shifts from perennial to intermittent
flow conditions have been accompanied by a
changing aquatic macroinvertebrate composition.
Here, progressive change or recovery in species
structure following flood events closely tracks
previous flow conditions, with inter-annual
variations depending on how organisms resisted
channel drying, for example by using hyporheic
refugia (Sponseller et al. , 2010). By complete
contrast, in glacial-fed alpine systems the richness,
abundance and community composition of
stream invertebrates varies with the relative
contributions of melt-water to runoff (Brown et al. ,
2007). Different species persist under different
conditions, with high endemicity typical of the
species-poor communities in peri-glacial channels
- for example, those characterized by elevated
suspended sediments, but low water temperature
and conductivity. Glacial retreat now risks reducing
the occurrence of such locations, so the extinction
of melt-water specialists is a real possibility. Our
own work has revealed further permutations of
climatically induced flow changes, not all adverse.
For example, increasing discharge appeared to
reduce sensitivity to warming among chalk-stream
invertebrates (Durance and Ormerod, 2009). In

also offer potential (Woodward et al. ,

2010a).
Despite these difficulties, available long-term
data are now providing increasingly strong
evidence that warming effects on river ecosystems
are real and possibly widespread - at least as
revealed by commonly studied groups. Patterns
reflect combinations of changes in abundance,
species composition and occurrence of some
scarcer taxa, as well as some evidence of a
reduction in body size (Daufresne et al. , 2009).
In some British upland headwaters, for example,
spring invertebrate abundances have declined with
stream warming by around 20% for every 1 ◦ C rise
- implying potentially large effects (Durance and
Ormerod, 2007; Figure 9.2). In this case, more than
80% of the core invertebrate community persisted
through inter-annual temperature variations of
around 3 ◦ C, but species typical of cooler-water
conditions, (e.g. some cool-water Plecoptera and
triclads), have nevertheless been lost. These loss
effects are not only consistent between locations
(Daufresne et al. , 2004; Durance and Ormerod,
2007, 2010; Figure 9.2), but also matched by long-
term shifts among macroinvertebrates in European
lakes where temperatures have also increased
(Burgmer et al. , 2007). In the upper French Rh one,
species changes have been linked not only to
higher temperatures, but also to decreased oxygen