Prediction of flow duration curves in ungauged basins - Runoff Prediction in Ungauged Basins: Synthesis across Processes, Places and Scales - page 144

Figure 7.10. Concavity index (IC) observed at gauged catchments in France plotted on the location of their centre of gravity. From Sauquet and
Catalogne ( 2011 ).

snowmelt-fed processes display SR 1, whereas this

variable is above 1 for typical rainfall-fed catchments with
low flow in summer and high flow in winter. The vari-
ation in SR is governed by geology and air temperature,
and is consequently subject to topographic influences in
France. Sauquet and Catalogne ( 2011 ) use IC and SR
together to identify homogeneous regions in terms of the
shape of FDC in France through different classification
methods (see Section 7.2.3 ).
In contrast to these studies, Ganora et al.( 2009 ) defined
a metric to quantify the differences as distances between
FDCs and relate these distances to differences among
catchments in terms of catchment and climatic characteris-
tics. The main advantage of using distances between
curves, rather than specific characteristics such as the
slopes or model parameters, is that the entire curve is
considered and not just their statistics/parameters. The
method also allows one to compare the difference between
FDCs with the differences between curves relating to cli-
matic and catchment characteristics (e.g., information on
the hypsometric curve can be used instead of the mean
catchment elevation of the elevation range).

<

Climate similarity
Regionalisation of FDCs works best if the daily runoff is
normalised by the mean daily runoff. The mean daily run-
off can be predicted, to first order, by the catchment s

aridity index, which then becomes the primary indicator
of catchment similarity. For example, Cheng et al.( 2012 )
found a strong relationship between mean daily runoff and
the aridity index, consistent with results presented in Chap-
ter 5 . Castellarin et al. ( 2007a ) show that the parameter
representing the position of the FDC, which is linked to the
mean annual runoff, is related to the mean annual net
precipitation, along with the catchment area. Similar find-
ings have been found by other studies (e.g., Viola et al.,
2011 ;Liet al., 2010 ). Once the runoff is scaled by the
mean daily runoff, the resulting annual FDCs are governed
by several climatic and landscape characteristics that
impact the transformation of the within-year variability of
precipitation into the corresponding variability of runoff.
The scale parameter used by Castellarin et al.( 2007a ) ,
which represents the within-year variability of runoff, is
correlated to the variability of the annual net precipitation.
As per the framework provided by Yokoo and Sivapalan

'