Table 3. Field saturated hydraulic conductivity,
than straight ruts (P < 0.01). In other words, turn-
Kfs, measured 13 May 1996. Locations shown in
ing ruts with the greatest amount of initial distur-
Figure 4.
bance (highest average standard deviations in
December 1995) had the highest decrease in stan-
Outofrut Kfs
Inrut Kfs
dard deviation over time. Much of this initial
Plot*
(cm/sec)
(cm/sec)
smoothing appeared to originate from rapid ero-
1.52104
Plot C, M1, 4, moist, straight
4.14 104
sion of thin edges of asymmetric rut lips and sub-
Plot C, M1, 4, dry, straight
4.68104
4.04104
sequent infilling of the compacted channels near
Plot C, M1, 1, moist, turn
4.29104
2.22106
the center of the ruts.
Plot E, M1, 2, moist, straight
1.86104
2.09104
As suggested above, changes in profile did not
Plot E, M1, 1, moist, turn
1.91103
3.79104
occur uniformly within the same rut. In general,
* Plot nomenclature syntax is in the form of plot, vehicle
the greatest changes in rut surface microrelief,
type, number of passes, antecedent soil moisture at time of
during 1996, occurred at the highest or lowest
tracking, and track path. Location notes refer to the map
elevations of the rut profile (e.g., App. A, turn C
shown in Figure 4.
2-4). A net loss of profile height was most often
measured at the rut lip. In contrast, the base of the
sidewalls of the ruts were the zones of deposition
or infilling. Little change was detected along the
steep sidewalls. However, the profile meter
Saturated hydraulic conductivity, Kfs
records only profile changes that lie in an unob-
Table 3 shows that soil compacted by the tank
structed vertical pin path. Careful field inspection
can have a reduced Kfs relative to the adjacent
showed that soil slumping sometimes resulted in
untrafficked soil. However, how much Kfs is
concave or undercut rut sidewall geometry not
reduced appears to be influenced by the amount
detectable with this instrument.
of soil moisture at the time of tracking. For a loca-
Our profile measurements revealed inter- and
tion where tracks had been formed on moist soil
intra-plot variability in rut shape and depth; this
at site C, the Kfs inside a rut was less than half that
variability was not clearly correlated with the
measured in adjacent uncompacted soil. Con-
number of vehicle passes. Such variability sug-
versely, at a location where tracks had been
gests that rut formation is strongly influenced by
formed on dry soil, the in-rut and out-of-rut Kfs
soil variables and antecedent soil moisture.
was nearly identical. The Kfs rate measured out-
Another important source of rut surface vari-
side a turning rut was comparable to values out-
ability is related to soil surface conditions and
side the two straight ruts, but Kfs was much
soil moisture at the time of measurement. We
lower inside the turning rut than in straight ruts.
collected initial readings on 8 December 1995,
This suggests that the shearing and vertical
when soil was locally frozen and partially snow
forces generated during tank turning decrease
covered. Soil was near field capacity during our
the potential for subsequent water movement in
next readings on 27 March 1996. The third set
the soil more than when a tank is moving straight.
of readings was collected on 16 July 1996 when
Our measurements suggest Kfs is more spa-
the soil surface contained 05% water and
tially variable at site E than site C. The highest rate
of Kfs (1.91 103 cm/sec) was recorded in uncom-
shrink-swell cracks were evident. Accurate mea-
surements require that the profile-meter support
pacted soil on a small ridge less than 100 m from
bars remain horizontally and vertically stable and
a location, where the uncompacted value was an
that the reading pins rest exactly on the soil sur-
face. Thus apparent changes in profile-meter mea-
However, like site C, the Kfs observed at site E was
surements at a specific pin location may result
lower inside a turning rut than out of rut. Unlike
from actual changes of the rut profile but will also
site C, little difference in Kfs was observed between
reflect other mechanisms, such as frost heave or
a straight rut and adjacent uncompacted soil, sug-
shrinking and swelling due to wetting-drying
gesting tank compaction did not affect potential
cycles, that shift the reference position (upright
for water movement at this location.
rebars). Also, the profile-meter pins may penetrate
extremely dry, loose or wet soil and introduce a
Soil penetration resistance, SPR
We observed similar in-rut and out-of-rut pat-
error into the profile readings. We observed this
terns of average penetrometer readings at both
phenomenon in July 1996 for measurements in
sites. Average SPR was low near the surface,
extremely dry soil at site E (see App. B).
9
Previous Page
