Geography Reference
In-Depth Information
Figure 10.26. Relationships between
catchment characteristics (minimum
and maximum LAI (leaf area index),
mean elevation and mean catchment
slope) and different model parameters
for six non-snow dominated
catchments in the Rocky Mountains,
USA. From Carrillo et al.( 2011 ).
0.3
10
8
0.2
6
4
0.1
2
0
0.0
0.2 0.4 0.6 0.8 1.0
LAI.min (-)
0 1 2 3 4 5
LAI.max (-)
1.0
100
0.8
80
0.6
60
0.4
40
0.2
20
0.0
0
0.2 0.4 0.6 0.8 1.0
LAI.min (-)
0 250 500 750 1000
Mean elevation (m )
0.8
12
8
0.6
4
0.4
0.2
0
0.2 0.4 0.6 0.8 1.0
LAI.min (-)
0.0
0.1
0.2
0.3
Catchment slope (-)
discussed above and in the literature (e.g., Sefton and
Howarth, 1998 ; Peel et al., 2000 ; Fernandez et al., 2000 ),
often no significant relationships between calibrated model
parameters and catchment characteristics can be found.
There are two main issues involved (Blöschl, 2005 ).
First, catchment characteristics may not represent the
hydrological processes of interest very well. This applies
in particular to subsurface characteristics such as soil and
geological properties. This also applies to characteristics
such as percentage land use or percentage soil unit that
may be easily calculated from existing maps, but the
hydrological interpretation may not always be clear. Also,
soil texture may not always be a meaningful characteristic,
depending on the runoff mechanisms operative in a catch-
ment. Catchment characteristics may be needed that better
reflect the hydrological processes expected in a particular
catchment rather than a one size fits all solution. The
HOST soil classification is an example where soils are
indeed quantified from a hydrological perspective. The
interpretation of catchment and climate characteristics
may involve co-evolution processes of
Second, runoff model parameters generally cannot be
identified from observed runoff very well. Parameter
identification is an ill-posed problem related to parameter
inter-dependencies (Beven and Freer, 2001 ), i.e., different
sets of parameter values give similar runoff situation per-
formance (Kokkonen et al., 2003 ; Wagener et al., 2004 ).
There are a number of ways to address the second issue
and improve parameter estimation, either by reducing the
number of free model parameters by fixing some of them,
changing the model structure, calibrating the model at a
number of stream gauges simultaneously, or by using
additional information on the hydrological processes such
as a-priori information and proxy data. In the following,
the last two options are explored in more detail as they are
most relevant to ungauged basins.
Regional calibration and downscaling of parameters
An alternative to the approaches above is to establish
relationships between parameters and catchment character-
istics and to calibrate the coefficients of these relationships
instead of the parameters themselves. Usually, a number of
stream gauges in a region are used for calibration. This
the catchment
system at long time scales.
 
Search WWH ::




Custom Search