Geography Reference
In-Depth Information
distribution of lakes and wetlands in riparian landscapes
(e.g., billabongs in Australia).
8.6 Summary of key points
Most of the methods used for low flow predictions are
statistical, partly because there is not enough informa-
tion on hydrogeology (which is an important control)
and this information, when available, is hard to quantify.
There is considerable potential to use more dynamic
predictors that include time (e.g., sequencing of rainfall
events, spring runoff, runoff recession curves), link
these to regression coefficients obtained during the
application of statistical methods, and to interpret them
from hydrological process perspectives.
This chapter dealt with low flows, the part of the spec-
trum of runoff variability when there is very little flow in
the river. Low flows can be defined in several ways, the
most common being the annual runoff minima, or the
magnitude of runoff that is exceeded 95% of the time.
Sometimes, runoff variability during the period of time in
the year over which flows remain low or are close to the
minimum is used to characterise the low flow regime.
The low flow distribution represents a composite
signature that reflects the interplay of several features:
climate during the dry period of the year, storage in sub-
surface (including deep aquifers) and associated long flow
paths, evaporation (especially from the riparian zone vege-
tation) and, in cold climates, the effects of snow storage.
Comparative assessment of several prediction methods
for low flows indicates that predictive performance gets
worse with increasing aridity (both Level 1 and Level 2
assessments). The performance improves with increas-
ing catchment area (Level 2 assessment), ostensibly
because of the presence of longer water flow pathways
that accompany increasing catchment size. The avail-
ability of short records is particularly useful to improve
performance of low flow predictions (both Levels 1 and 2),
especially in humid regions, and is perhaps not as useful in
arid regions because of strong inter-annual variability
(Level 2). Of the various methods, regional regressions
have been shown to be better than global regressions
(from Level 1 and Level 2 assessment).
Winter low flows (in cold, snow-affected regions) are
controlled by temperature and antecedent precipitation.
Summer low flows (as a result of prolonged dry spells)
are controlled by aridity of the catchment, the sequence
of rain events during the normal dry part of the year, by
storage
properties
of
the
underground,
and
by
vegetation.
Characteristic similarity measures include the time of
the year that low flows occur (e.g., winter or summer),
local geology (which determines the long flow pathways
that bring water to the river from deep water stores) and
large-scale climate patterns (which determine larger
scale patterns of low flows). Co-evolutionary indices
that determine similarity include patterns of riparian
vegetation (e.g., an oasis present in the middle of a
desert
There is considerable scope to use process information
on the hydrogeological catchment architecture and cli-
mate drivers (summer/winter low flows) to identify low
flow regimes at regional and global scales and interpret
the differences and similarities through comparative
hydrology.
is an extreme example), connectivity and
Search WWH ::
Custom Search