Geoscience Reference
In-Depth Information
New water
A-horizon lateral flow has a higher
matrix flow component than B horizon,
possibly as a result of ponding on top
of the B horizon.
New
water
B-horizon lateral flow is dominantly
through preferential flow paths with some
matrix flow during saturated periods.
Flow into tuff appears to be minor
compared with that in the A and B
horizons.
V
e
r
tical
byp
a
ss
ing
Organic litter and grass cover
Sandy loam—A horizon
Clay—B horizon
Restrictive layer (inferred)
Bandelier Tuff
Matrix flow
Preferential flow
Fig. 11.9 Perceptual flow mechanisms in a semiarid forested slope in New Mexico. The lateral matrix flow in
the A horizon is larger than that in the B horizon, possibly as a result of some ponding on top of the
B horizon; in the B horizon the flow takes place mainly through preferential flow paths, with some
matrix flow and leakage into the underlying tuff. (From Newman
et al
., 1998.)
illustrated how such interpretations can evolve over time, as more and better measurement
techniques are brought to bear on the analysis (McGlynn
et al
., 2002). In the early studies
by Mosley (1979) it was concluded from local flow and dye tracer measurements inpits
that macropore flow of mostly new water, in storms of moderate to large intensity, can
bypass the soil matrix, where the pre-event water is normally stored, and is capable
of generating the channel stormflow. On the basis of subsequent investigations with
electrical conductivity and natural tracers, Pearce
et al
. (1986) and Sklash
et al
. (1986)
arrivedatadifferent conclusion; they deduced from the measurements that it was mainly
old water throughflow that was responsible for hydrograph generation, and that the flow
of new water above the ground surface or below it through the soil matrix or through
macropores could not explain the streamflow response. To resolve these discrepancies,
athird set of studies was carried out by McDonnell (1990; McDonnell
et al
., 1991a) in
which a chemical tracer analysis was supplemented with soil water pressure observations
by means of tensiometers installed in near-stream, mid-hollow and upslope positions. It
was observed that the soil water pressure response was dependent on storm magnitude,
intensity and antecedent water content. For storm events producing peak runoff less
than 2 mm h
−
1
, the water appeared to infiltrate downward as a wetting front in the soil
matrixwithout appreciable macropore bypass flow; no water table developed along the
slope and the streamflow consisted of old water issuing mainly from the near stream
valley bottom groundwater. For events with peak storm runoff in excess of 2 mm h
−
1
,
