Prediction of low flows in ungauged basins - Runoff Prediction in Ungauged Basins: Synthesis across Processes, Places and Scales - page 168

Figure 8.4. Seasonality ratio, i.e., the
ratio of summer and winter Q 95 low
flow runoff in Austria. Blue shades
indicate a winter low flow regime,
red shades indicate a summer low
flow regime. White: no data. From
Laaha and Blöschl ( 2006b ).

Seasonality ratio
< 0.5
0.5 - 0.8
0.8 - 0.9
0.9 - 1.1
1.1 - 1.25
1.25 - 2.0
> 2.0

Summer evaporation

Alpine snow pack

50 km

flow seasonality index (Young et al., 2000c ; Laaha and
Blöschl, 2006b ). This index is similar to the seasonality
index used by Burn ( 1997 ) to analyse floods, but estimates
the mean day of occurrence from a partial series of the
Julian date when the flow decreases below some threshold.
The mean day of occurrence and the strength of seasonality
are derived from circular statistics (Mardia, 1972 ). Another
measure is the seasonal histogram (Laaha and Blöschl,
2006b ), which is obtained by plotting the frequency of
low flow days within each calendar month against months.
The seasonal histogram contains more detailed information
on the low flow regime than the seasonality index.
A simple and informative measure is the seasonality ratio
(Laaha and Blöschl, 2006b ), which is the ratio of summer
and winter low flows, based on some low flow index
calculated separately for the summer and winter seasons.
Seasonality ratios greater than one indicate higher runoff in
summer than in winter and indicate a winter low flow
regime. Seasonality ratios less than one indicate a summer
low flow regime. The magnitude of the number represents
the strength of the seasonality. Figure 8.4 shows an
example of the seasonality ratio, which highlights the
prevalence of winter low flows (blue) in the west and
summer low flows (red) in the east of the region. Together
with the other seasonality measures discussed above, the
map is useful for identifying dominant low flow processes
as a basis for regionalisation (Laaha and Blöschl, 2006a ).
The seasonality of low flows is an important link to
another runoff signature, the seasonal runoff ( Chapter 6 ),
and is therefore an important variable to characterise the
hydrological behaviour of catchments.
A stream may receive water from different flow paths:
overland flow, interflow, and shallow and deep ground-
water discharge ( Chapter 4 ). Overland flow and interflow
respond quickly to rainfall or melting snow, whereas
groundwater discharge responds slowly with a time lag of

several days, months or years. Catchments dominated by
overland flow, interflow and/or shallow saturated subsur-
face flow are usually quickly responding or

' '

catch-
ments (e.g., van Lanen et al., 2004a ; van Lanen et al.,
2012 ). Typical examples are clay catchments with shallow
water tables and a dense drainage network, and catchments
with steep relief and shallow impermeable bedrock. Catch-
ments fed primarily by groundwater discharge are usually
slowly responding catchments. They are typically lowland
catchments with a substantial aquifer. The flashiness of a
catchment may be quantified by the recession behaviour: a
flashy catchment with a fast response to precipitation tends
to exhibit steeper recession behaviour. Recession param-
eters can be calculated from the runoff hydrograph during
periods without rain and can then be combined into master
recession curves to characterise the overall catchment
behaviour during recessions (see Chapter 4 ). Flashiness
can also be quantified by the baseflow contribution. Base-
flow is usually estimated by baseflow separation tech-
niques that divide the total runoff into a quick (usually
shallow flow) component and a slow (storage-delayed)
component (see e.g., IH, 1980 ; Hisdal and Tallaksen,
2004 ). This can be achieved either by several variants of
digital filters (Lyne and Hollick, 1979 ; Arnold et al., 1995 )
or with the support of tracer techniques (see Chapter 4 ).
The long-term ratio of the slow runoff component and total
runoff is the baseflow index. Figure 8.5 shows typical
hydrographs of a flashy and a slowly responding catch-
ment. For the Thompson catchment the baseflow index is
0.31, while for the Elkhart it is 0.9 (Sawicz et al., 2011 ).
The Elkhart catchment contains large wetlands and lakes,
and the thick glacial sediments and complex topography
detain water, releasing it slowly over long periods of time,
thus providing for more sustained flow rates long after the
initiating precipitation events (USACE, 2010 ).

flashy

In the

Thompson the storage is much smaller.