Climate Change: Defining Reference Conditions and Restoring Freshwater Ecosystems - Climate Change Impacts on Freshwater Ecosystems - page 221

35

30

25

20

Fish

15

Invertebrates

Macrophytes

10

Benthic diatoms

5

0

Figure 9.8 Among-stream variance (%) in fish (blue bars), invertebrate (red),
macrophyte (green) and benthic diatom (purple) assemblages explained by geographic
(GEO), regional (REG) and local (LOC) factors. LL, lowland; MN, mountain streams.
(Results are taken from Johnson et al . 2007.)

Ecological relationships between communities and spatial scale
Recognition of how aquatic systems are linked with their surrounding landscapes
and how linkages affect resistance and resilience to change is important to underpin
management decisions. Johnson et al . (2007), analysing relations between benthic
diatoms and fish and different aspects of spatial scale, postulated that organism-
response would vary according to life histories. They hypothesized that fish, being
long-lived, mobile organisms, would be more related to large-scale, regional
variability, while small organisms, like benthic diatoms and invertebrates, would be
more related to local factors. They found that local scale variables such as water
chemistry and substratum explained, on average, more than four times the amount
of variance than was accounted for by geographic position (latitude and longitude)
and more than twice the variance that was accounted for by regional-scale variables
such as ecoregion and catchment land use/cover (Fig. 9.8). However, contrary to
expectations, no support was found for the conjecture that large organisms (fish)
are responding more to large-scale (regional) factors, while small organisms (e.g.
benthic diatoms) are responding more to small scale (habitat) factors. The finding
that local scale variables were important predictors of stream assemblages agrees
with several previous studies (e.g. Hawkins et al . 2000; Paavola et al . 2006; Hering
et al . 2006), and has direct implications for designing programmes for stream
restoration (see the use of woody debris in stream restoration below).
Furthermore, even though in-stream variables were better predictors of taxonomic
composition, this does not imply that large-scale factors such as catchment land use
should be ignored in the planning and implementation of conservation
and management programmes. Indeed, since many disturbances are the result of