Figure 9. 100-MHz time section profile across the water well, the depth of which was 70.1 m. The asymptotic slopes
of the hyperbolic diffraction from the well (1) give an nm = 1.59; the slopes of the more shallow hyperbola (2) from an
unknown object give nm = 1.46; and the slopes of the resonant waves (3) generated by surface objects give nm = 1.36.
A refractive index of 1.59 (ε = 2.53) gives an al-
beneath the profile). The slopes of the linear slop-
most perfect hyperbolic fit to this diffraction, and
ing events throughout the record (e.g., event 3 in
corresponds to an overburden average density of
Fig. 9) give a refractive index of 1.36 (density =
about 0.7 g/cm3. Alternatively, given the time
0.45 g/cm3). The accuracy of these measurements
delay and the measured depth to the water of 70.1
is about 4%, based on the error of making a best
m, the effective refractive index for the firn
fit to the diffractions.
overburden is determined to be 1.63 (ε = 2.66).
A migration of the water well profile is shown
Although the estimated bulb diameter is 9 m,
in Figure 10. The time section profile had to be
there is no evidence (e.g., two hyperbolae) of sep-
compressed by a factor of two (subsequently
arate diffractions from each end of the bulb, as the
expanded) so that the migration aperture could
temporal displacement of their asymptotes
encompass most of the water well diffraction.
would be over 80 ns at the ends of the profile.
The migration collapses the hyperbolic diffrac-
Therefore, the diffraction itself, centered exactly
tion from the well to a segment approximately 16
at the center of the water well, is primarily a
m wide. This is wider than the estimated bulb
response to the complete well surface, but
diameter of 9 m and may be partly an artifact of
undoubtedly modified somewhat by the air bulb.
the algorithm and the finite pulse width, as theo-
There are many other events in Figure 9, the
ry (Yilmaz 1987) shows that a perfect diffraction
origins of which are not known. The area was ex-
hyperbola generally collapses to this finite sized
tremely cluttered. Events with a slope corre-
form rather than to a singular point. For estimat-
sponding to a refractive index near 1.3 are either
ing the encroachment of the sumps, we have
surface or near-surface objects. Event 2 in Figure
assumed that the current diameter of the water
9 is a hyperbola that gives a refractive index of
well could be as great as 16 m.
1.46 (density = 0.55 g/cm3 at an approximate
Three sumps in this area, sump 1 and sump 2,
depth of 29.5 m, assuming the object was directly
and one below a utilidor vent south of the dome,
10
Previous Page
