Geoscience Reference
In-Depth Information
Fig. 11.10 Perceptual runoff production mechanisms in a midslope hollow of a humid catchment in New Zealand.
As shown, the precipitation rate (
P
) exceeds the hydraulic conductivity (
k
0
) of the mineral soil, and
moves down through vertical cracks. The invading new water perches at the soil-bedrock interface,
and backs up into the newly saturated soil matrix, where it mixes with the much larger volume of
stored old water. Once free water (with positive pore water pressures) exists, the larger pipes in the
lower soil zones quickly dissipate transient water tables laterally downslope, producing a rapid
throughflow response of well-mixed, albeit mainly pre-event water. (From McDonnell, 1990.)
the lower soil horizons along the slope responded almost instantaneously, indicating a
rapid macropore flow, as Mosley (1979) had already surmised. The predominance of
old water in the streamflow runoff was explained by McDonnell (1990) by the fact that
the rapid flow of new rainwater through downward crack macropores backs up into the
soil matrix at the soil-bedrock interface, which isstill dry; this rapidly causes saturated
conditions, and results in the emergence of well-mixed old water from the matrix into
lateral pipe macropores and rapid downslope transport (see Figure 11.10). In a fourth set
of experiments, Woods and Rowe (1996) dug a trench 60 m long and 1.5 m deep along
the toe of a hillslope hollow, with 30 subsurface flow collection points along its length.
The outflow from the hillslope was found to be very variable; this led to the conclusion
that outflow data from single hillslope throughflow pits should not be extrapolated to an
entire hillslope and further (Woods
et al
., 1997) that thisvariability depends on wetness
and surface topography. The latter conclusion was refuted by McDonnell and associates
