Geoscience Reference
In-Depth Information
Resistivity tomogram prior the water pulse injection
Well
Iteration 3 RMS error 4.7%
0
2.5
0.4
2.4
2.3
0.8
2.2
1.2
2.1
2
1.6
1.9
2
0
2
4
6
8
10
12
14
16
Distance (m)
(a)
Resistivity tomogram after the water pulse injection
Well
Iteration 3 RMS error 4.7%
0
2.5
0.4
2.4
2.3
0.8
2.2
1.2
2.1
2
1.6
1.9
2
0
2
4
6
8
10
12
14
16
Distance (m)
(b)
Figure 5.44
Inverted DC resistivity section using the Gauss
Newton method. The data are collected using 32 electrodes with
50 cm spacing.
a)
Tomogram prior the pulse injection.
b)
Tomogram after the pulse injection. (
See insert for color representation
of the figure
.)
-
anomaly, which amount 6
8 mV, and that it is centered
on the injection well. This anomaly is inverted in the next
section to localize the source current density responsible
for this anomaly.
-
density, especially at the tip of the injection well, at
a depth of 65 cm. We used Tikhonov regularization
to perform the inverse computation. The value of the
regularization parameter,
, was determined using the
L-curve approach with a regularization parameter in
the range of [10
−
3
,10
3
]. We obtain
α
5.5.3 Localization of the causative source
of the self-potential anomaly
The injection of water into a porous material produces
a source current density,
J
S
. The pulse injection exper-
iment is therefore responsible for such a source current
= 0.121 as the opti-
mized regularization parameter. In this case, we have
M
= 8241 and
N
= 28 (4 noisy channels were removed)
from the data set. The kernel is computed in 3D
by extending the resistivity distribution in the strike
α
