Introduction To CSAMT
Introduction To CSAMT
Introduction To CSAMT
By Kenneth L. Zonge
Zonge Engineering and Research Organization, Inc.
S ilver B e ll C S AM T Line 28
S m oo th-m od el R esistiv ity (o hm -m )
1000
1200
1400
1600
1800
2000
2200
2400
2600
2800
3000
3200
3400
3600
3800
4000
4200
4400
4600
4800
5000
5200
5400
5600
5800
6000
6200
6400
6600
6800
7000
7200
7400
-4 00
-2 00
200
400
600
800
0
3 0 00
2 8 00
2 6 00
2 4 00
E le va tion (fe e t)
2 2 00
2 0 00
1 8 00
1 6 00
1 4 00
1 2 00
1 0 00
S ta tic-co rre cte d C S A M T da ta fro m L 2 8.A V G
C o rrecte d w ith 4 09 6 H z T M A filte r
Figure 1: Field setup for a scalar reconnaissance CSAMT survey using multiple E-fields and one H-field
measurements.
Figure 2: Field configurations for: (a) a
scalar CSAMT survey, (b) a multiple E-
field reconnaissance CSAMT survey, or
controlled source audio frequency
(CSAET)survey, (c) a vector CSAMT
survey, (d) a tensor CSAMT survey. An
alternative set-up for the tensor survey
would be to have one transmitter at its
present location to the west of the
receiver station (oriented N-S) and the
other to the north (oriented E-W).
If you are going to make measurements on a sparse grid, for example making individual soundings every
kilometer or every 500 meters instead of running continuous stations along a line, you may want to
consider vector or tensor measurements. (See Figures 2c and 2d). These measurements can provide
information on anisotropy (current channeling) and 2-D or 3-D behavior of the subsurface on an individual
station basis. Vector or tensor arrays are not necessary for normal continuous line profiling since
variations from station-to-station provide information on geologic dimensionality.
Of the four arrays shown in Figure 2, the most commonly used systems are scalar CSAMT (Figure 2a) and
the reconnaissance CSAET measurement technique (Figure 2b).
Grounded dipoles detect the electric field and magnetic coil antennas sense the magnetic field. The ratio
of orthogonal, horizontal electric and magnetic field magnitudes (e.g. Ex and Hy) yields the apparent
resistivity. This is usually referred to as the apparent or Cagniard resistivity after the French geophysicist
who was instrumental in the development of the magnetotelluric (MT) method in the early 1950's
(Cagniard, 1953). The difference between the phase of the electric and magnetic fields yields the
impedance phase, which we will often just call the phase or phase difference.
A parameter used extensively in EM work is skin depth for frequency domain systems (diffusion depth in
time domain systems). Skin depth, , is equal to 503 a/f meters; where a = apparent (measured)
resistivity, and f = signal frequency. This is the depth at which the amplitude of a plane wave signal has
dropped to 37 percent of its original value.
AMT depth of exploration or depth of investigation, D, is equal to /2 or 356 a/f meters. This equation
holds for CSAMT when the separation between the transmitter dipole and the receiver station is greater
than three skin depths or 3. Although this equation predicts unlimited depths of investigation under the
right circumstances, we have found that in actual practice the maximum depth of investigation is limited to
about 3 km.
Lateral resolution is controlled by the electric field dipole length, which normally is between l0 and 200
meters. Vertical resolution is 5 percent to 20 percent of the depth of exploration, depending upon
resistivity contrasts, geologic complexity and electrical noise.
Table I lists the basic equations most commonly used for calculating apparent or Cagniard resistivity, skin
depth and depth of investigation. These equations hold for a layered earth with the transmitter placed an
infinite distance from the receiver. The limitations brought on by a finite distance between transmitter and
receiver will be discussed later. Figure 3 shows how to determine exploration depth from resistivity and
frequency.
Cagniard Resistivity, ρa:
2
E
ρ a
=
1
5f
(ohm-meters)
H ⊥
Phase difference,Φ :
φ = φ E − φH (milliradians)
2 ρ
δ= = 503
a
(meters)
µ σ ω f
ρ
D = 356 a
(D in meters)
f
Wavelength, λ:
λ = 2πδ
where:
E = electric field in mV/km
H = magnetic field in gamma (γ) or nanoteslas (nT)
Φ = phase in milliradians
µ = magnetic permeability of air = 4 p x 10-7
σ = conductivity in siemem/meter = 1/r
ω = 2 p f, f=frequency in Hz
ρ = resistivity in ohm-metrs = 1/ σ
If it is impractical to set up a large dipole for a transmitting source, you can also use a large loop of wire
for the transmitting antenna. The main drawback of the loop source is its inefficiency in coupling energy
into the ground. Given a dipole and square loop of the same size (say a one kilometer dipole and a one
kilometer square loop) it takes eight to ten times more loop current than dipole current to get the same
signal level at the receiving site. For this reason, loops are seldom used as transmitting sources. However,
the Cagniard resistivity and phase measurements are the same in the far-field for both sources, but differ in
the transition and near-field zones.
Figure 3: Effective investigation depth for CSAMT as a function of frequency and ground resistivity.
Far-field: The electric and magnetic fields, that are generated by driving current into the ground with the
large transmitting dipole, propagate along the surface of the ground and penetrate nearly vertically into the
ground at distances beyond half a wavelength (approximately 3) from the transmitter. This zone of
vertical penetration is called the far-field, and in this area the electric and magnetic fields behave as plane
waves, similar to natural source MT and AMT fields. Therefore we can use the simplified MT and AMT
equations for modeling structure in the far-field.
Near-field: When the electrical distance between the transmitter and receiver becomes less than 3 (about
1/2 wavelength - see Table 1) at the frequency being measured, the electric and magnetic fields change
gradually from plane-wave to curved, and the Cagniard resistivity formula no longer provides realistic
apparent resistivity values. When operating in the far-field or plane-wave zone over a layered earth, both
E and H-fields drop off as 1/r3, where r is the separation between the transmitter and receiver, and both
fields vary as a function of frequency and earth resistivity. In the transition zone, the H-field drop off
begins to change to 1/r2 and the dependency on frequency and earth resistivity begins to change as well.
The E-field continues to drop off as 1/r3 and retains its function of earth resistivity, but its dependency on
frequency also begins to change.
In the near-field the H-field decays at 1/r2, becomes saturated and no longer varies as a function of
frequency or resistivity. The E-field still remains a function of resistivity, decays at 1/r3, but is
independent of frequency. It is in this near-field zone that depth of investigation becomes independent of
frequency and dependent upon array geometry. This is the condition that actually puts a practical limit on
the depth of exploration for CSAMT soundings. For consistency, we continue to use the Cagniard
resistivity calculation through the transition zone and into the near-field, although the values calculated in
these areas do not reflect the actual resistivity values of the earth.
Following are some general rules for phase interpretation:
Depth of exploration
As observed in the graph in Figure 1 and the equations in Table 1. The depth of exploration or
investigation is related to the square root of ground resistivity and the inverse square root of signal
frequency. Although the equations do not provide any limit to the depth of exploration obtainable, the
maximum usable depth achieved in practice is usually between 2 and 3 kilometers. As a general rule,
when sounding over a relatively homogeneous half-space, the separation between transmitter and receiver
should be about 5 times the depth of exploration. Therefore, if you want to see down to 1 km you should
have receiver-transmitter separation of about 5 km. If the background resistivity is 1000 m, you will only
have to sound down to about 100 Hz to penetrate 1 km; if the background resistivity is 100 m, you will
have to sound down to about 10 Hz, etc. Refer to Figure 1 for more information.
The limiting factor on depth of exploration with all of the data in the far-field is usually signal level. The
electric and magnetic fields drop off as the inverse cube of the separation distance (1/r3) between the
receiver and transmitter, so the signal level drops off very quickly. It is not unusual to work in
environments where the background noise is more than 10 times the signal level. Most surveys are run
with receiver-transmitter separations between 5 and 15 kilometers.
Lateral resolution
Lateral resolution is mainly determined by the size of the E-field dipole. Theoretically you could make the
dipole as small as you wish to get the desired lateral resolution, but again signal strength and noise inter
the picture. Received signal strength is proportional to the length of the dipole, so if you cut the dipole
size in half, you cut your signal strength in half. I think the smallest dipole we have used in normal
production is about 25 feet (8 m).