Prediction of runoff hydrographs in ungauged basins - Runoff Prediction in Ungauged Basins: Synthesis across Processes, Places and Scales - page 232

Figure 10.5. Dominant processes
of runoff generation mechanisms at
the hillslope scale. From Dunne
( 1978 ).

Direct precipitation
and return flow
dominate hydrograph;
subsurface stormflow
less important

Thin soils: gentle
concave (footslopes);
wide valley bottoms;
soils of high to low
permeability

Horton overland
flow dominates
hydrograph; contributions
from subsurface stormflow
are less important

VARIABLE
SOURCE
CONCEPT

Topography
and soils

Subsurface stormflow
dominates hydrograph
volumetrically: peaks
produced by return flow
and direct precipitation

Steep, straight
hillslopes: deep, very
permeable soils;
narrow valley bottoms

Arid - to - subhumid
climate; thin vegetation:
or disturbed by Man

Humid climate:
dense vegetation

Climate, vegetation and land - use

to relatively little erosion and a less efficient drainage
system in the landscape than in other areas (Gaál et al.,
2012 ). Because of this, a comparative approach across
different climates and landscape regions may reveal inter-
esting patterns of catchment behaviour (Falkenmark and
Chapman, 1989 ). The schematic in Figure 10.5 suggests
that catchments in arid climates will more likely experi-
ence higher precipitation intensities and sparser vegetation
cover leading to the runoff hydrograph being dominated by
Horton overland flow. Infiltrated water in arid places has
less opportunity to pass through the root zone and recharge
groundwater due to the high atmospheric demand and
hence high evaporation rates. On the other hand, the dom-
inance of horizontal fluxes increases with increasing
humidity and steepness of catchments. Catchments in
humid regions tend to have hydrographs dominated by
subsurface stormflow, particularly where soils are deep
and permeable. If soils are thinner, subsurface storm flow
may be less important and saturation excess runoff may be
more important. Figure 10.5 , in fact, can be interpreted as
depicting schematically the co-evolution of catchments
and the resulting runoff mechanisms. Catchments located
at higher altitudes typically have significant snow cover
and therefore a strong seasonal storage component. The
seasonality of water inputs tends to be out of phase with
that of the energy inputs in these catchments and snow
accumulation and snowmelt often dominate the runoff
hydrograph.
With increasing catchment size, runoff variability is
increasingly dominated by stream channel flow processes,
including delay and attenuation associated with the
hydraulics of the flow, along with channel losses and
uptake by riparian vegetation (in arid regions), floodplain
inundation, and changing morphology and hydraulic

geometry of river networks. Large catchments are also
affected by increased spatial heterogeneity, including sub-
stantial changes in the nature of climate inputs and domin-
ant processes (e.g., dominance of snow processes in
headwater regions and the influence of porous aquifers in
downstream areas). The runoff hydrograph observed at any
point in the stream network therefore embeds within it all
the hydrological variability associated with the co-
evolution of the catchment.
The runoff dynamics of catchments can change with
time, in particular if human modifications have occurred.
The most common modification is through the construc-
tion of dams and other hydraulic structures, which, through
the effects of associated impoundments and controls, tend
to reduce the variability of runoff. Extraction of river water
for irrigated agriculture has the effect of increasing evap-
oration and reducing river runoff, whereas withdrawals of
water for municipal water supply and other human uses are
eventually returned after human consumption, which may
reduce the runoff variability and increase low flows (Wang
and Cai, 2009 ). In many agricultural landscapes tile drain-
age is a major factor that may modify the dynamics of the
runoff hydrographs (see Figure 10.2 ). Changes in the
vegetation cover may alter the runoff dynamics signifi-
cantly. Typically, the removal of forest leads to reductions
of evaporation, increases in antecedent wetness, and thus
increases in the fraction of precipitation that is converted to
runoff (Bosch and Hewlett, 1982 ).
Figure 10.6 illustrates the effect of forest removal on the
runoff dynamics, the result of a paired catchment study in
the forested south-west of Western Australia, experiencing
a Mediterranean climate. Salmon (0.82 km 2 ) and Wights
(0.94 km 2 ) are neighbouring catchments. Salmon has
remained fully forested, whereas Wights was fully cleared