RUNOFF - Fundamentals of Hydrology - page 83

One of the first theories put forward concerning
the contribution of throughflow to a storm hydro-
graph was by Horton and Hawkins (1965) (this
Horton was a different person from the proposer of
Hortonian overland flow). They proposed the
mechanism of translatory or piston flow to explain the
rapid movement of water from the subsurface to the
stream. They suggested that as water enters the top
of a soil column it displaces the water at the bottom
of the column (i.e. old water), and the displaced
water enters the stream. The analogy is drawn to a
piston where pressure at the top of the piston
chamber leads to a release of pressure at the bottom.
The release of water to the stream can be modelled
as a pressure wave rather than tracking individual
particles of water. Piston flow has been observed in
laboratory experiments with soil columns (e.g.
Germann and Beven, 1981).
At first glance the simple piston analogy seems
unlike a real-life situation since a hillslope is not
bounded by impermeable sides in the same way as
a piston chamber. However the theory is not as far-
fetched as it may seem, as the addition of rainfall
infiltrating across a complete hillslope is analogous
to pressure being applied from above and in this case
the boundaries are upslope (i.e. gravity) and the
bedrock below. Brammer and McDonnell (1996)
suggest that this may be a mechanism for the rapid
movement of water along the bedrock and soil inter-
face on the steep catchment of Maimai in New
Zealand. In this case it is the hydraulic gradient
created by an addition of water to the bottom of the
soil column, already close to saturated, that forces
water along the base where hydraulic conductivities
are higher.
Ward (1984) draws the analogy of a thatched roof
to describe the contribution of subsurface flow to a
stream (based on the ideas of Zaslavsky and Sinai,
1981). When straw is placed on a sloping roof it is
very efficient at moving water to the bottom of the
roof (the guttering being analogous to a stream)
without visible overland flow. This is due to the pre-
ferential flow direction along, rather than between,
sloping straws. Measurements of hillslope soil
properties do show a higher hydraulic conductivity

in the downslope rather than vertical direction. This
would account for a movement of water downslope
as throughflow, but it is still bound up in the soil
matrix and reasonably slow.
There is considerable debate on the role of macro-
pores in the rapid movement of water through the
soil matrix. Macropores are larger pores within a
soil matrix, typically with a diameter greater than
3 mm. They may be caused by soils cracking, worms
burrowing or other biotic activities. The main
interest in them from a hydrologic point of view is
that they provide a rapid conduit for the movement
of water through a soil. The main area of contention
concerning macropores is whether they form con-
tinuous networks allowing rapid movement of
water down a slope or not. There have been studies
suggesting macropores as a major mechanism con-
tributing water to stormflow (e.g. Mosley, 1979,
1982; Wilson et al ., 1990), but it is difficult to detect
whether these are from small areas on a hillslope or
continuous throughout. Jones (1981) and Tanaka
(1992) summarise the role of pipe networks (a form
of continuous macropores) in hillslope hydrology.
Where found, pipe networks have considerable
effect on the subsurface hydrology but they are not
a common occurrence in the field situation.
The role of macropores in runoff generation is
unclear. Although they are capable of allowing rapid
movement of water towards a stream channel there
is little evidence of networks of macropores moving
large quantities of water in a continuous fashion.
Where macropores are known to have a significant
role is in the rapid movement of water to the
saturated layer (e.g. Heppell et al ., 1999) which may
in turn lead to piston flow (McGlynn et al ., 2002).

Groundwater contribution to
stormflow

Another possible explanation for the presence of
old water in a storm hydrograph is that it comes
from the saturated zone (groundwater) rather
than from throughflow. This is contrary to conven-
tional hydrological wisdom which suggests that
groundwater contributes to baseflow but not to the