Geophysics 424 September 2016

D4: Applications of magnetotelluric exploration

.
D4.1. Geothermal exploration

Background

Geothermal energy can be utilized over a wide range of temperatures.

If the rocks are in the range 50 -150°C, then the heat is generally used directly (district
heating, greenhouses, alligator farms etc.).

At temperatures above 150° C, electricity can be generated efficiently.

Most existing geothermal energy production is from hydrothermal systems. These are
found in areas with recent volcanic and / or tectonic activity. A typical geothermal
reservoir is a region of high porosity /high permeability, containing high temperature
fluids.

Requirements

(1) Heat
(2) Subsurface fluids to transport heat to the surface.
(3) Natural permeability to allow fluid flow

© Martyn Unsworth, University of Alberta, 2016 1

 Geophysics 424 September 2016

Exploration for hydrothermal systems

 In many geothermal fields, the reservoir is overlain by a zone of clay alteration,

with a low resistivity. Figure on left is from Pellerin et al., (1996).

 In Geophysics 325, we saw how DC resistivity can be used to locate the clay cap.

 To image geothermal reservoirs below the clay cap, combined MT and controlled
source EM methods are widely used. Reservoirs are generally lower in resistivity
than the host rock, but more resistive than the clay cap (B2). As described in
424B2, the base of the low resistivity zone corresponds to the smectite-illite
transition at around 200° C

 Figure on right is courtesy of Geosystem SRL and shows a resistivity model of a

volcano hosted geothermal field obtained with MT. Note the 3 layers.

(1) High resistivity at the surface is extrusive volcanics (red).

(2) Low resistivity layer due to clay (blue)
(3) Underlying low resistivity reservoir (red). The dashed line shows
top of reservoir, based on drilling.

 In D1.4 we saw that MT cannot image a weak conductor below a stronger

conductor. Thus the overlying clay cap, can make it difficult to image the
reservoir.

 Also note that when a geothermal reservoir is exhausted, the clay cap remains and
can confuse exploration. For a general overview of EM exploration with EM, see
Pellerin et al, (1996).

© Martyn Unsworth, University of Alberta, 2016 2

 Geophysics 424 September 2016

 Once a geothermal field has been developed, subsequent MT exploration is more

difficult because pipes and pumps generate a lot of EM noise that will
contaminate the natural EM signals.

Engineered geothermal systems

In many locations heat is available, but fluids / permeability do not occur naturally.
Solution is to create an Engineered Geothermal System by:

 drilling wells to depths where high temperatures are found,

 creating permeability through hydro-fracturing and injecting fluid
 circulating water through a well doublet to extract heat

A number of EGS pilot plants have been built to develop technology. Has potential to
supply significant percentage of electricity needs of North America.

According to “The Future of Geothermal Energy”, a report issued by the

Massachusetts Institute of Technology (MIT) in 2007, the thermal energy
available in America in rocks 3-10km beneath the Earth’s surface is nearly
140,000 times greater than its annual energy consumption. Conservative
estimates suggest just 2% of that energy could be tapped by EGS in
practice, but even that would be far more than is needed to supply all of
America’s electricity. Tapping it will, however, require both technical and
economic hurdles to be overcome.
http://www.economist.com/node/16909897?story_id=16909897

© Martyn Unsworth, University of Alberta, 2016 3

 Geophysics 424 September 2016

D4.2. Hydrocarbon exploration

In hydrocarbon exploration, studies can be divided into two broad classes.

(1) Most geophysical exploration seeks structures that may contain

hydrocarbons e.g. anticlines, saltdomes etc. This approach is based on
previous studies in the same area that has provided information about
which formations are potential source and reservoir rocks.
(2) A limited number of approaches look for direct evidence of
hydrocarbons e.g. oil water interface or oil gas interface can
sometimes be imaged with seismic reflection data.

EM exploration can contribute in both (1) and (2)

General overview

 MT cannot give the same vertical detail as a seismic reflection survey. This is
because EM methods use diffusive signals with a wavelength of the same order of
magnitude as the skin depth. Sharp interfaces cannot be imaged. In contrast
seismic exploration uses wave propagation and features λ/4 in thickness can be
detected.
 However, EM methods can provide valuable information on bulk material
properties. This is similar way to the original application of Archie’s Law, which
was developed for the interpretation of well logs. For general reviews on this
topic, see the articles by Unsworth (2005) and Christophersen (2002).
 EM exploration is generally used in situations where seismic exploration methods
have difficulties, or seismic data quality is low.
 EM exploration was widely used in the former Soviet Union for hydrocarbon
exploration, as reviewed by Spies (1983).

D4.2.1 Imaging potential reservoirs and source rocks

Sub-salt imaging

 Subsalt imaging can challenge seismic exploration, because of the high velocity
contrast between the salt sheet and underlying sedimentary rocks (why?). The top
of a salt sheet can be detected with confidence in a seismic reflection survey, but
the base is more difficult to locate.

 Salt sheets generally exhibit a higher electrical resistivity than the surrounding
sedimentary rocks. Thus the base of salt represents a decrease in electrical
resistivity (increase in conductivity). This geometry is favourable for detection
with MT.

© Martyn Unsworth, University of Alberta, 2016 4

 Geophysics 424 September 2016

 MT has been used onshore to assist in sub-salt imaging. For a case study that
studied salt domes, see den Boer et al, 2000).

Offshore sub-salt exploration

 MT instruments are placed dropped to the seafloor with a heavy anchor attached.
They are recovered when an acoustic command releases the anchor and flotation
spheres pull the instrument to the surface.

 The presence of overlying seawater will attenuate the EM signals, so high

frequency data will not be recorded at the seafloor. Simple skin depth arguments
can be used to determine the frequency band that will work in a given water
depth.

 Electric fields are measured by

placing the electrodes at the end
of flexible plastic pipes. Seafloor
currents can cause the instrument
to move and generates magnetic
noise as the magnetic field
sensors move in the Earth’s
magnetic field.

 Details of measurements can be found at:

http://www.aoageophysics.com/Electrical/MarineMT/WhatsNew.htm

 Several case studies in the Gulf of Mexico have shown that MT can be applied in
this context. Details : http://marineemlab.ucsd.edu/resources/concepts/mtsalt.html

 Note the small range of resistivity values in the MT derived resistivity model.
This model used a smooth inversion to overcome the non-uniqueness that is
inherent in the MT inverse problem.

© Martyn Unsworth, University of Alberta, 2016 5

 Geophysics 424 September 2016

 When generating models, the use of seismic and gravity data can give a
much clearer model. If the location of some boundaries is known
independently, then the MT inversion does not need to smear out structure
in this region. Resistivity model derived by a Sharp Boundary Inversion is
shown by Hoversten et al, (2000).

Onshore sub-basalt exploration

 Basalt layers can result in noisy seismic data. The basalt has a high
resistivity, so MT signals will travel through and can be used to image
structure of underlying sedimentary rocks.
 MT can determine the thickness of basalt flows that overly sedimentary
rocks. This works well because MT is sensitive to the depth to the top of a
conducting layer. Applications in Columbia Plateau (Washington) and
Deccan Traps in India.
 MT can also image the structure of the underlying sedimentary rocks. See
case study from Saudi Arabia described by Colombo et al., (2012). Note
that the resistivity model below uses a reversed colour scale (red = high
resistivity).

© Martyn Unsworth, University of Alberta, 2016 6

 Geophysics 424 September 2016

 MT derived resistivity model was used to provide constraints that

improved the migration

© Martyn Unsworth, University of Alberta, 2016 7

 Geophysics 424 September 2016

Overthrusts
 Overthrusts are often associated with effective hydrocarbon traps. As with
sub-salt exploration, this geometry can place high velocity thrust sheets
above lower velocity rocks, and resolution at depth with seismic reflection
exploration is compromised.
 In addition, weathering and static problems can seriously degrade the
quality of seismic data in this context.
 Can MT contribute in this situation? Again this geometry usually
corresponds to a low resistivity layer (a conductor) below a high resistivity
layer, which is favourable for detection with MT.

 The above figure shows an MT model for a profile in the Foothills

extending from Rocky Mountain House to the Front Ranges of the
Rockies (Xiao and Unsworth, 2006).
 The Alberta Basin is characterized by a low resistivity layer some 4 km
thick. The Cretaceous sedimentary rocks have been thrust beneath a
sequence of older rocks to form the Brazeau Thrust Fault.
 The agreement between the MT model and a resistivity log at W3 is
shown below.

 Why is the MT derived resistivity model smoother than that measured in

the well log?

© Martyn Unsworth, University of Alberta, 2016 8

 Geophysics 424 September 2016

Another overthrust example

 Watts and Pince (1998) reported a similar example from Southeastern
Turkey. Here karst terrain produces very poor seismic reflection data.

D4.2.2 Direct imaging of hydrocarbons with MT (onshore)

 Shallow gas pockets can be readily located with DC resistivity, as they are
associated with a high resistivity anomaly (e.g. Northern Alberta).
 However, at greater depth the signal from a reservoir becomes smaller,
and will be masked by resistivity variations within a particular formation.
 Example from Uzbekistan by Ingerov et al, (2003)

 Example of time changes during production using controlled source EM

see the paper by Ziolkowski et al, (2002). Controlled source methods are
better in detecting these changes, since the transmitted signal is known,
and constant from one day to the next.

© Martyn Unsworth, University of Alberta, 2016 9

 Geophysics 424 September 2016

D4.3. Hydrogeology and environmental geophysics

General overview
See review by Christophersen (2002).

Application to environmental studies at the Amchitka Island nuclear test site

 In 2004 MT was used to study subsurface porosity and salinity on

Amchitka Island in the Aleutian arc. This was the location of 3
underground nuclear tests conducted by the United States from 1965-
1971.

 The goal was to provide data that would constrain hydrogeological

modeling that is being used to determine how long it will take for
radiation to leak into the marine environment.

Cannikin Lake, Amchitka Island, Alaska, 2004

 The results are summarized in a report that can be found at:

http://www.cresp.org
 A detailed description of MT results are listed at:
http://www.instrm.org/Amchitka_Final_Report/finalreport/chapters/06_Chapter6_
F7_26_05.pdf

The following figures are taken from this report

© Martyn Unsworth, University of Alberta, 2016 10

 Geophysics 424 September 2016

Figure 1 : Theoretical study of the effect of subsurface porosity and salinity on the overall
resistivity of a rock (a) Variation of salinity as a function of depth (TDS = total dissolved solids)
(b) resistivity of the ground water assuming the empirical relationship of Block (2001). (c)+(d)
The porosity is constant with depth, resulting in a uniformly decreasing bulk resistivity with
increasing depth. (e)+(f) Porosity decreases with depth, resulting in a more complex variation of
bulk resistivity with depth. TZ = transition zone from fresh to salt water. Note that in (f) the
resistivity decreases through the transition zone, and increases in the saltwater layer.

© Martyn Unsworth, University of Alberta, 2016 11

 Geophysics 424 September 2016

Figure 26: Hydrogeology for Cannikin Ground Zero. (a) Shows the salinity (TDS) at the nearby UAe-1
well (red circles) and the blue line denotes a simplified form. The maximum value permitted is 35 g/l
equivalent to seawater. (b) resistivity of the pore fluid derived from (a) using the empirical equation of
Block (2001). (c) Effective porosity required to give agreement between bulk resistivity and that
determined by the MT data. Computation uses Archies’ Law with exponents m=1, 1.5 and 2 (d) Resistivity
from MT data (red circles) compared to that predicted by data (blue line) in panels (a)-(c). The asterisk (*)
denotes the depth of the shot cavity.

Figure 29: Hydrogeology at the Cannikin Ground Zero. An exponential porosity depth variation was
assumed and the salinity required to reproduce the variation of resistivity with depth was computed. Note
again that a significant increase in salinity is predicted below the depth of the shot cavity.

© Martyn Unsworth, University of Alberta, 2016 12

 Geophysics 424 September 2016

D4.4. Mineral exploration

 Controlled source EM methods are widely used in mineral exploration (as

we will see in the next section of this class).

 AMT and MT is generally used for deeper exploration (>1 km). A good
review is provided by Sheard et al, (2002).

Massive sulphide deposits

 Generally located with other types of EM survey (Airborne EM, VLF etc).

 However ground based AMT data has been used to try and define the
geometry in more detail. See the case study by Jones and Garcia (2003).

Uranium exploration

 Uranium deposits are not usually large enough to be located from the
surface through their low resistivity.

 However, they are often co-deposited with graphite where fluid bearing
basement faults reach an unconformity.

 These graphitic conductors can easily be located with airborne EM if they

are close to the surface.

 However at greater depths (1 km), AMT data are needed to detect them.
The figure below shows sample resistivity model from a test case at the
McArthur River Mine in the Athabasca Basin (NW Saskatchewan). AMT
data were collected on a grid at the surface to allow 3-D imaging (Tuncer
et al, 2006).

© Martyn Unsworth, University of Alberta, 2016 13

 Geophysics 424 September 2016

 Induction vectors were useful in giving additional constraints on the location of

basement conductors.

 Can you see the trend of induction vector reversals in the data? Details in Tuncer
et al., (2006)

© Martyn Unsworth, University of Alberta, 2016 14

 Geophysics 424 September 2016

Kimberlite exploration

 Induction vectors from 500 Hz

AMT data collected over a
kimberlite pipe.

 Data courtesy of NordWest

 Top of pipe is characterized by

low resistivity due to weathering
that causes formation of clay
minerals.

 Induction vectors point at the

kimberlite pipe in the Parkinson
convention.

© Martyn Unsworth, University of Alberta, 2016 15

 Geophysics 424 September 2016

D4.5. Tectonic and lithospheric studies with MT

 Before considering tectonically active regions, it is important to understand what

is considered a normal resistivity structure in stable regions. In the figure, all
resistivity values are in ohm-m.

 Sedimentary basins are generally

low in resistivity. Connected
pore fluids conduct electricity
quite well. Compaction reduces
pore space and thus resistivity
increases with depth. Changes in
salinity have a major effect on
bulk resistivity. e.g. saline
aquifers in WCSB

 Freshly intruded igneous rocks

can be highly resistive, as the
melting process effectively
expels all volatiles that can act as
charge carriers.

 Beneath the sedimentary cover,

upper crustal rocks are generally
resistive.

 The lower continental crust is generally much lower in resistivity than the upper
crust, with a conductance of 200-500 S. The decrease in resistivity is roughly
coincident with the brittle-ductile transition. Some researchers suggest this is due
to interconnected, saline fluids (Marquis and Hyndman, 1992). Others argue that
this is petrologically impossible, and that carbon films are responsible for the low
resistivity (Yardley and Valley, 1997).

 The upper mantle is generally more resistive, but the resistivity is difficult to
image. This is because it usually below a conductor (the lower crust). The Moho
is a seismically defined discontinuity. There is not always a distinct change in
resistivity across the Moho (Jones and Ferguson, 2001).

 The asthenosphere is typically seen at 150-250 km depth as a low resistivity

zone (not shown in the figure above). This is due to the presence of small amounts
of interconnected partial melt.

© Martyn Unsworth, University of Alberta, 2016 16

 Geophysics 424 September 2016

With the above model as background, a number of studies have shown that anomalous
resistivity structures are often associated with present day (and past) tectonic activity.

San Andreas Fault, California:

 Shallow zone of fractured rock in

upper few kilometers of the fault
(Unsworth et al, 1997; Unsworth
et al, 2004).

 May be typical of most large-

scale strike-slip faults worldwide.

 MT data provided guidance for

the SAFOD drilling project.

Tibetan Plateau
 Type example of a continent-continent collision. Collision began 40- 50 million
years ago and has generated the Himalaya and Tibetan Plateau. Crust is double
the global average and locally exceeds 80 km. Mechanism for crustal thickening
is not understood and since 1992 the INDEPTH project has been using integrated
geological and geophysical studies to understand the tectonic evolution of the
Tibetan Plateau. MT has made a major contribution to this study.

 MT has detected a widespread fluid zone at 20 km depth with resistivity around 5-

10 Ωm. Likely due to partial melting, with contribution from aqueous fluids (Wei
et al, 2001; Unsworth et al, 2005). The MT data has also shown that this layer
may be weak enough to flow over geological time.

 Note that in the figure below, two independent geophysical studies indicate the
presence of fluids (1) low resistivity (2) strong, negative polarity seismic
reflections.

© Martyn Unsworth, University of Alberta, 2016 17

 Geophysics 424 September 2016

 Similar features are seen in other plate boundaries. For example, a region of
thickened crust has developed in the Central Andes where the Nazca Plate is
being subducted beneath the South American plate. The tectonic setting here is
quite different to Tibet, yet the end result is quite similar. The Altiplano is 5 km
high, 200 km wide and 1000 km long. Deep sounding MT surveys have revealed
a zone of low resistivity in the mid-crust that has been interpreted as a zone of
extensive partial melting (Brasse et al, 2002).

Pre-Cambrian basement tectonics

 Geological sutures represent the locations where blocks of continental crust were
joined together as the continents grew. In pre-Cambrian basement these regions
are often characterized by zones of low resistivity extending for 100’s or 1000’s
of kilometers.

 Several examples are found in Canada (Red Deer Conductor, Trans-Hudson

Orogen) and they may be due to graphite. For details see Boerner et al (1999).

 Now how did the carbon get down there?

© Martyn Unsworth, University of Alberta, 2016 18

 Geophysics 424 September 2016

References
Block D, 2001, Water Resistivity Atlas of Western Canada Abstract, presented at Rock the
Foundation Convention of Canadian Society of Petroleum Geologists, June 18-22, Calgary.
Boerner DE, RD Kurtz, JA Craven, GM Ross, FW Jones and WJ Davis, Electrical conductivity in
the pre-Cambrian lithosphere of Western Canada, Science, 283, 668, 1999.
Brasse H, P Lezaeta, V Roth, K Schwalenberg, W Soyer and V Haak, The Bolivian Altiplano
Conductivity Anomaly, J. Geophys. Res, 107, doi 10.1029/2001JB000391, 2002.
Christophersen K, EM in the 21st century – looking for oil, gas and water, review presented at
2002 EM workshop, Santa Fe, New Mexico. 2002.
Colombo D, T Keho and G McNeice, Integrated seismic-electromagnetic workflow for sub-
basalt exploration in northwest Saudi Arabia, The Leading Edge, 42-52, January 2012
den Boer, E, J Eikelboom, P van Driel and MD Watts, 2000, Resistivity imaging of shallow salt
with magnetotellurics as a new aid to prestack depth migration: First Break, 18, no. 1, 19-26.
Hoversten GM, SC Constable and HF Morrison, 2000, Marine magnetotellurics for base of-salt
mapping: Gulf of Mexico field test at the Gemini structure: Geophysics, Soc. of Expl.
Geophys., 65, 1476-1488.
Ingerov A et al; Resistivity signature of hydrocarbon deposits: MT surveys in Western
Uzbekistan Poster No. P054, presented at the EAGE Annual Meeting, June 01-04, 2003,
Stavanger, Norway (available on www.phoenix-geophysics.com)
Jones AG, IJ Ferguson, The electric Moho, Nature, 409, 331-333, 2001.
Jones AG, X Garcia, Okak Bay (Labrador) audiomagnetotellurics (AMT) dataset case study:
Lessons in dimensionality and scale, Geophysics, 68, 70-91, 2003.
Marquis G and RD Hyndman, Geophysical support for aqueous fluids in the deep crust: seismic
and electrical relationships, GJI, 110, 91, 1992.
Pellerin L, J Johnston and G Hohmann, Geophysics, 61, 121, 1996.
Sheard SN, E Brand, T Ritchie, Industry applications of EM – base metal and environment, ,
review presented at 2002 EM workshop, Santa Fe, New Mexico.
Spies B, Recent development in the use of surface electrical methods for oil and gas exploration
in the Soviet Union, Geophysics, 48, 1102-1112, 1983.
Tuncer V, MJ Unsworth, W Siripunvaraporn and J Craven, Exploration for unconformity type
uranium deposits with audio-magnetotelluric data: A case study from the McArthur River
Mine, Saskatchewan (Canada), Geophysics, 71 (6), B201-B209, 2006.
Unsworth MJ, PE Malin, GD Egbert and JR Booker, Internal Structure of the San Andreas Fault
Zone at Parkfield, California, Geology, 25, 359-362, 1997.
Unsworth MJ, AG Jones, W Wei, G Marquis, S Gokarn, J Spratt, Crustal rheology of the
Himalaya and Southern Tibet inferred from magnetotelluric data, Nature, 438, 78-81,
doi:10.1038/nature04154, 2005.
Unsworth MJ, and PA Bedrosian, Electrical resistivity at the SAFOD site from MT exploration,
Geophys. Res. Lett., 31, L12S05, doiL10.1029/2003GL019405, 2004.
Unsworth MJ, New developments in conventional hydrocarbon exploration with electromagnetic
methods, CSEG Recorder, pp 34-38, April 2005.
Watts MD and A Pince, Petroleum exploration in overthrust areas using MT and seismic data,
SEG Expanded Abstracts, 1998.
Wei W, MJ Unsworth, AG Jones, JR Booker, H Tan, KD Nelson, L Chen, S Li, K Solon, PA
Bedrosian, S Jin, M Deng, J Ledo, D Kay, B Roberts, Detection of widespread fluids in the
Tibetan crust by magnetotelluric studies, Science, 292, 716-718, 2001.
Xiao W and MJ Unsworth, Structural imaging in the Rocky Mountain Foothills (Alberta) using
magnetotelluric exploration, AAPG Bulletin, 90, 321-333, 2006.
Yardley B and J Valley, The petrologic case for a dry lower crust, JGR, 102, 12,173, 1997.
Ziolkowski A, B Hobbs and D Wright, 2002, First direct hydrocarbon detection and reservoir
monitoring using transient electromagnetics: First Break, 20 (4) pp 224-225.

© Martyn Unsworth, University of Alberta, 2016 19