A synthesis framework for runoff prediction in ungauged basins - Runoff Prediction in Ungauged Basins: Synthesis across Processes, Places and Scales - page 11

2 A synthesis framework for
runoff prediction in ungauged
basins

Contributors: T. Wagener,* G. Blöschl, D. C. Goodrich,
H. V. Gupta, M. Sivapalan, Y. Tachikawa, P. A. Troch and
M. Weiler

2.1 Catchments are complex systems
2.1.1 Co-evolution of catchment characteristics
Landscapes present amazing patterns that appear to be ubi-
quitous at any scale one looks. At the pore scale, microbes
colonise soil particles and form biofilms that alter water
flow paths and water

interacting species, driven by natural selection (Thompson,
1994 ). In the case of catchments, co-evolution implies a
process of reciprocal evolutionary change of soils, vegeta-
tion and topography, mediated by material and energy
fluxes, in response to fast climate dynamics and slow
geological processes ( Figure 2.2 ). The patterns that emerge
reflect the legacy of past processes, their interconnections
over a long period of time leading to the complex spatial
patterns that we see in the landscape today (Sivapalan,
2005 ). These spatial patterns are also responsible for the
temporal patterns in runoff response, but the connection
between these spatial and temporal patterns is still poorly

-

sediment contact time, thus affecting
geochemical weathering and the nucleation of secondary
minerals. Biogeochemical alteration of the mineral water

interface results in stable particle aggregates allowing the
fast movement of water in interconnected flow paths. At the
patch scale, rills form in response to rain-splash erosive
action and overland flow redistributes important nutrients
and carbon that affect soil properties, such as infiltration
capacity. Vegetation responds to this spatial variability in
water and nutrient availability to form clusters characteristic
of the dominant flow processes. At the hillslope scale, clear
patterns emerge in soil characteristics as a result of the
interplay of water and carbon movement, erosion, soil
formation and both vegetation and animal action. At the
landscape scale, the interplay of land uplifting and erosion

-

-
deposition processes generates landforms that feed back to
ecological and pedological processes. At the same time,
climate interacts with vegetation, soils and landforms
through hydrological processes to produce large-scale vege-
tation patterns. It stands to reason that the co-evolution of
climate, vegetation and soils at the landscape scale leads to
specific ways of hydrological partitioning that are reflected
in runoff records. The satellite image of the landscape in the
Channel Country in south-western Queensland shown in
Figure 2.1 illustrates the complexity of the landscape pat-
terns where an intricate network of riverbeds has evolved in
the alluvial fans made mostly of clays (Baker, 1986 ). Many
of the challenges highlighted in Chapter 1 could be
addressed if these landscape patterns could be connected
quantitatively to catchment hydrological response.
Taken from biology, the concept of co-evolution refers
to the process of reciprocal evolutionary change between

Figure 2.1. Channel Country in south-western Queensland,
Australia, as a false-colour composite image of Landsat 7 s ETM+

sensor on 10 January 2000. http://earthobservatory.nasa.gov/IOTD/
view.php?id=3346.

'