Geography Reference
In-Depth Information
0.4
a)
0.3
0.2
0.1
0.0
J
F
M
A
M
J
J
A
S
O
N
D
b)
60˚N
30˚N
0˚
30˚S
60˚S
180˚
˚
120˚W
60˚W
0˚
60˚E
120˚E
180˚
0
3
6
9
12
Month of the Year
Figure 6.14. Clusters of mean monthly runoff: (a) Pardé coefficients of the clusters; (b) distribution of sites in the clusters, with cluster indicated
by the peak month in (a). From Dettinger and Diaz (
2000
).
the
of Laaha and Blöschl
(
2006a
) is based on this idea. In the most basic case, both a
model structure and model parameters are applicable
across a geographic region; however, contiguous regions
may still be considered homogeneous if parameters are
location specific, but a single model structure can be held
in common (Gottschalk,
1985
).
An alternative to defining coherent regions based on
proximity is to rely on global climatological classifications
to identify areas with similar flow regimes. Existing classi-
fications, e.g., Köppen (
1936
) and Budyko (
1974
), have
indeed been identified as determinants of runoff seasonality
in some studies (Beckinsale,
1969
). The relationship
between climatic regions and seasonal runoff variability is
'
regional regression approach
'
often less direct than the relationship between climatic
region and annual runoff, due to the nested/aggregated
nature of river basins and the importance of geology, catch-
ment morphology and land use for driving runoff generation
at intra-annual time scales (
Section 6.2.1
). These compli-
cations have led some authors to suggest that climatic
regions should not be used as a basis for the extrapolation
of flow regimes between basins (Haines et al.,
1988
).
Clustering techniques can be applied to the delineation
of geographical areas. In an example of 139 catchments in
Sweden (Gottschalk,
1985
), the results of the cluster analy-
sis were graphically summarised in a dendrogram. Groups
from the dendrogram were mapped and used to identify
hydrological regions with similar monthly runoff patterns,
Search WWH ::
Custom Search