GEOTHERMAL TRAINING PROGRAMME Reports 2010
Orkustofnun, Grensásvegur 9, Number 12
IS-108 Reykjavík, Iceland
151
MAGNETOTELLURIC AND TRANSIENT ELECTROMAGNETIC
METHODS IN GEOTHERMAL EXPLORATION WITH EXAMPLES
FROM THE KRÝSUVÍK AREA, SW-ICELAND
Tamrat Fantaye
Geological Survey of Ethiopia (GSE)
Geothermal Resource Exploration and Assessment Core Process
P.O. Box 2302, Addis Ababa
ETHIOPIA
tamuka04@yahoo.com
ABSTRACT
Geophysics plays a great role in subsurface exploration for geothermal resources.
Several geophysical exploration methods are applicable, divided into direct and
indirect (structural) methods. One of the direct geophysical methods for
geothermal exploration is the resistivity method. Electrical resistivity methods like
magnetotellurics (MT) and transient electromagnetics (TEM) are powerful tools in
mapping subsurface conductivity variations. In this report the main geophysical
methods are discussed in general while electromagnetic methods are discussed in
detail. MT and TEM data collected along a profile from Krýsuvík hightemperature
geothermal area were processed and interpreted and results are
presented as resistivity cross-sections. The MT soundings on the profile were
aligned in a N-S direction and the corresponding nearby TEM data were analysed
for each MT station. 1-D inversion for the TEM data and 1-D joint inversion of
MT and TEM data were performed along the profile. The results are presented as
resistivity cross-sections. The uppermost layer was found to be resistive, unaltered
volcanic rocks while the second corresponds to the conductive smectite-zeolite and
mixed layer clay zone. Below these, a relatively resistive layer, corresponding to
the chlorite-epidote zone was clearly mapped. The correlation between resistivity
values and alteration mineralogy and temperature in wells is also discussed.
1. INTRODUCTION
Geophysics is a science that applies different kind of physical principles to study Earth´s subsurface.
Some physical parameters studied are temperature, resistivity, density, magnetization and seismicity.
Among these, resistivity and thermal methods are the most successful methods for studying
geothermal sites and they are referred to as direct methods. Indirect or structural methods like
magnetics, gravity and seismics help to examine the physical parameters of the host rock.
Electromagnetic (EM) methods are frequently used in geothermal exploration. This includes
magnetotellurics (MT) and time domain electromagnetics (TEM) which are sensitive to resistivity
structures beneath. In this report, the main geophysical methods in geothermal exploration, especially
EM methods, are discussed and 1-D modelling of MT and TEM soundings from Krýsuvík geothermal
field in SW-Iceland is carried out and presented as resistivity cross-sections.
Tamrat Fantaye 152 Report 12
2. GEOPHYSICAL METHODS IN GEOTHERMAL EXPLORATION
Geophysical methods play a key role in geothermal exploration. Different types of geophysical
methods are used for geothermal resource exploration. These methods are divided into direct and
indirect or structural methods (Hersir and Björnsson, 1991). The former include thermal and electrical
methods while the latter include magnetic, gravity and seismic methods.
The most direct method for studying geothermal systems is that of subsurface temperature
measurements. Measurements can be made in drillholes as shallow as a few metres or more in the
soil, but the preference presently is to conduct temperature surveys in wells that are at least 100 m
deep (Manzella, 2007). Thermal methods include direct measurements of temperature or heat and
correlate better with the properties of the geothermal system than other methods.
The electrical methods are the most important geophysical methods in the exploration of geothermal
systems. This includes direct current (DC) methods and electromagnetic (EM) methods.
DC methods: a constant current is injected into the ground that creates a potential field that can be
measured to infer the subsurface resistivity. Schlumberger soundings, head-on profiling and dipole
soundings are all DC geophysical methods.
Electromagnetic (EM) methods: Electromagnetic (EM) sounding methods include natural-field
methods (magnetotelluric and audiomagnetotellurics) and controlled-source induction methods, as
well as high-frequency radiation techniques such as radar-probing. Because of the depth of the main
targets in geothermal exploration, only natural-source and controlled-source methods (time domain
electromagnetic) are used (Manzella, 2007). This method is described in detail in Section 4.
The gravity method is a structural method through which subsurface geology is investigated on the
basis of variations in the earth’s gravitational field due to density contrasts between subsurface rocks.
It is possible to locate local masses of greater or lesser density than the surrounding formations and get
information from the irregularities in the earth´s field. Gravity data are generally affected by several
factors and must be corrected for variations such as instrument drift, latitude, elevation and so on.
Magnetic method, also a structural method, is useful in mapping near surface volcanic rocks that are
often of interest in geothermal exploration revealing faults or intrusions, but the power of this method
lies in its ability to detect the depth at which the Curie temperature is reached (Manzella, 2007). Curie
temperature is the critical temperature at which ferromagnetic materials lose their magnetic
susceptibility.
Seismic methods are divided into passive and active. Passive seismic methods are natural microearthquakes
caused by fracturing, sometimes related to geothermal activities; active seismic methods
are based on the timing of artificially-generated pulses of elastic energy propagating through the
ground and picked up by electromechanical transducers (geophones) as detectors.
3. RESISTIVITY METHODS IN GEOTHERMAL EXPLORATION
Measuring subsurface electrical resistivity is the most powerful method in geothermal exploration
(Hersir and Björnsson, 1991). This is because resistivity is directly related to the properties of interest,
such as temperature, alteration, salinity and porosity (permeability) (Hersir and Árnason, 2009).
Report 12 153 Tamrat Fantaye
3.1 Resistivity of rocks
The specific resistivity (ρ) of a material is defined as the electrical resistance (R) between the opposite
faces of a material (Figure 1), as given by the following equation:
(1)
where ρ = Specific resistivity of the material (Ωm);
R = Resistance (Ω);
l = Length (m);
A = Cross-sectional area of the conducting material (m2).
Electric currents propagate in rocks and minerals in three ways, through
electronic (ohmic), electrolytic and dielectric conduction. The first is the
normal type of current flow in materials containing free electrons such as
metals, while in electrolytic conduction the current is carried by ions at a
slower rate. Dielectric conduction takes place in insulators or poor
conductors that have no or very few free carriers and conduction takes
place by displacement current.
3.2 Factors affecting electrical resistivity of water-bearing rocks
The resistivity of rocks is dependent on and affected by various factors. The rock matrix itself is an
insulator and electric conduction occurs through an aqueous solution of common salts distributed
throughout the pores of rocks, and through the alteration minerals. The main factors that control the
resistivity of rocks are:
• Temperature
• Porosity and permeability
• Salinity
• Water-rock interaction and alteration
3.2.1 Temperature
As shown in Figure 2, at moderate temperatures, 0-200°C, the resistivity of an aqueous solution
decreases with increasing temperature. This is because of increasing mobility of the ions caused by a
decrease in the viscosity of the electrolytic solution. But at higher temperatures, a decrease in
dielectric permittivity of water results in a decrease in the number of dissociated ions in the solution.
Above 300°C, this starts to increase the fluid resistivity (Quist and Marshall, 1968).
The relationship between resistivity and temperature of the rock saturated with an electrolyte has been
described by Dakhnov (1962) as:
1 (2)
where ρw = Resistivity (Ωm) of the fluid at temperature T;
ρwo = Resistivity (Ωm) of the fluid at temperature To;
α = Temperature coefficient of resistivity, around 0.023/°C for To = 23°C,
and 0.025/°C for To = 0°C.
FIGURE 1: A piece
of resistive material
with electrical
contacts on both ends
Tamrat Fantaye 154 Report 12
The effect of temperature variations is greatest at low
temperatures (less than 150°C), but decreases at higher
temperature when other factors, such as porosity, salinity and
alteration mineralogy become dominant (Flóvenz et al.,
1985).
3.2.2 Porosity and permeability
The fractional porosity, ϕ
t of a material is defined as the ratio
of the pore volume to the total volume of the material and is
given by the formula:
(3)
where ϕ
t = Fractional porosity;
Vφ = Volume of voids;
V = Total volume of the material.
Fluid is important for electrical conduction of a rock,
therefore the degree of saturation (governed by porosity) is of
importance to the bulk resistivity of the rock. The following equation, usually referred to as Archie’s
law, describes how resistivity depends on porosity if ionic conduction dominates other conduction
mechanisms in a rock (Archie, 1942; Hersir and Björnsson, 1991):
(4)
where ρ = Bulk (measured) resistivity (Ωm);
ρw = Resistivity of the pore fluid (Ωm);
ϕt
= Fractional porosity;
α = Empirical parameter that describes the type of porosity varying from <1 for
intergranular porosity to >1 for joint porosity, but usually around 1;
n = Cementing factor, an empirical parameter, varies from 1.2 for unconsolidated
sediments to 3.5 for crystalline rocks, but usually around 2.
Archie’s law is valid if the resistivity of the pore fluid is of the order of 2 Ωm or less, but doubts are
raised if the resistivity is much higher (Flóvenz et al., 1985). However, Archie’s law seems to be a
fairly good approximation when the conductivity is dominated by the saturating fluid (Árnason et al.,
2000).
As Figure 3 shows, for pore fluid with resistivity less than 2 Ωm the dominant conductivity is pore
fluid conductivity and, hence, Archie’s law applies. For rocks saturated with fluids, having resistivity
higher than 2 Ωm at room temperature, the bulk resistivity is practically independent of the resistivity
of the fluid, but rather dependent on porosity and temperature. The dominant conductivity is mineral
and/or surface conductivity.
The permeability of a rock is the ability of fluids to move within its matrix. Permeability depends on
the interconnectivity of the pore spaces within the rock matrix. The amount of fluid flowing through a
rock can also largely be dictated by fractures (secondary porosity), common in geothermal areas. The
wider the fracture, the higher the fracture porosity, hence, high permeability is expressed by the
following equation (ISL, Michigan State University, 1999):
FIGURE 2: Electrical resistivity
of aqueous solution as a function
of temperature for
different pressures
(Hersir and Björnsson, 1991)
Report 12 155 Tamrat Fantaye
(5)
where K = Permeability (m2);
Q = Fluid flowrate (m3/s);
η = Fluid viscosity (kg/ms);
L = Length of the rock (m);
A = Cross-sectional area available for flow (m2);
P = Pressure drop (Pa)
Geological processes such as faulting, shearing, columnar jointing and weathering usually increase
permeability and porosity, thereby increasing electrical conductivity; however, the precipitation of
calcium carbonate or silica reduces porosity and, hence, increases resistivity.
3.2.3 Salinity
The conductivity (σ) of a solution depends on the mobility and concentration of the ions present in the
solution. This relationship is described by the following equation (Keller and Frischknecht, 1966):
… (6)
where σ = Conductivity (S/m);
F = Faraday’s number (9.65×104 C);
ci = Concentration of ions;
qi = Valence of ions;
mi = Mobility of ions.
As shown in Figure 4, when the amount of dissolved ions in the pore fluid increases, conductivity
increases. Conduction in a solution is greatly affected by salinity and the mobility of ions present
in the solution.
FIGURE 3: Bulk resistivity as a function of pore fluid resistivity
for different temperatures and porosities (Flóvenz et al., 1985)
Tamrat Fantaye 156 Report 12
3.2.4 Water-rock interaction and
alteration
The alteration process and the resulting
type of alteration minerals are dependent
on the type of primary minerals, the
chemical composition of the geothermal
fluid and temperature. The intensity of the
alteration is furthermore dependent on
temperature, time and the texture of the
host rocks. Alteration intensity is
normally low for temperatures below 50°C
(Figure 5). At temperatures lower than
220°C, low-temperature zeolites and the
clay mineral smectite are formed.
Smectite has hydrated and loosely bound
cations between the silica plates, making
the mineral conductive and with a high
cation exchange capacity (Árnason et al.,
2000).
In the temperature range from 220 to
about 240°C, the zeolites disappear and
the smectite is transformed into chlorite in
a transition zone, the so-called mixedlayered
clay zone, where smectite and
chlorite coexist in a mixture (Figure 6).
At about 240°C smectite disappears and
chlorite is the dominant mineral, marking
the beginning of the chlorite zone with
increased resistivity, since chlorite
minerals have cations that are fixed in a
crystal lattice, making the mineral
resistive. At still higher temperatures,
above 240°C, epidote becomes abundant
in the so-called chlorite-epidote zone
(Árnason et al., 2000).
4. ELECTROMAGNETIC METHODS
Electromagnetic methods have become
powerful geophysical tools in mapping
subsurface conductivity variations. The
method is widely used in the exploration
of geothermal resources (Árnason et al.,
2010). It mainly involves the propagation
of continuous wave or transient
electromagnetic fields in the earth. The
source may be natural or artificial. In a
natural source method like
magnetotellurics (MT), the fluctuation in
the earth’s natural magnetic field induces
FIGURE 4: Pore fluid conductivity vs. salinity for
various electrolytes (modified from
Keller and Frischknecht, 1966)
FIGURE 5: A summary of general resistivity
structures of high-temperature areas in Iceland
(Flóvenz et al., 2005)
FIGURE 6: Alteration mineralogy and temperature
Report 12 157 Tamrat Fantaye
an electric field (Figure 7). By measuring the
electrical and magnetic field at the surface of the
earth, inferences are made about the conductivity
distribution in the subsurface.
In artificial or controlled source methods like the
Transient Electromagnetic Method (TEM), a
magnetic field is created by transmitting a
current of known magnitude through a loop of
wire on the earth’s surface and when the current
is abruptly turned off, the magnetic field starts to
decay with time. This is used to determine
subsurface resistivity structures.
In magnetotellurics the presence of near-surface
resistivity inhomogeneities can distort the
electrical field, since the field is not continuous
across a resistivity boundary. This galvanic distortion effect is known as static shift. It shifts the MT
apparent resistivity sounding curve (i.e. log ρa vs. log Τ) by some constant scale factor downward or
upward. The static shift is a non-inductive change of the MT apparent resistivity response that
severely impairs the interpretation of the data. Electromagnetic methods which only measure
magnetic fields such as TEM do not have the static shift problems that affect MT soundings (Simpson
and Bahr, 2005). Therefore, TEM data can be used in conjunction with MT data from the same or
nearby site in order to correct for static shifts (Sternberg et al., 1988). The correction is in such a way
that the MT sounding curve is shifted vertically so that the high-frequency part of the MT curve agrees
with the TEM curve and consequently, the low frequency MT curve provides an undistorted picture of
the deep resistivity section (Jones, 1988).
4.1 Magnetotelluric (MT) method
The MT method is a passive-surface electromagnetic geophysical technique that measures variations
in the Earth's natural electromagnetic field to investigate the electrical resistivity structure of the
subsurface from depths of tens of metres to tens of kilometres (Vozoff, 1991). The method is passive
in the sense that it utilizes naturally occurring geomagnetic variations as the power source. Worldwide
lightning activity of frequencies from 10,000 to 1 Hertz (Hz) and geomagnetic micro-pulsations of
frequencies from 1 to 0.001 Hz provide the majority of natural signals used by the MT method.
The low frequencies which are less than 1 Hz and originate from solar wind interacting with the
earth’s magnetic field and ionosphere and are used for deep crustal investigations, while the high
frequencies, which originate from thunderstorm activities close to the equator, are used to map
resistivity variations of the upper crust. Data acquisition for
a single MT station (Figure 8) is done by measuring the
input fields, two horizontal magnetic components Hx and Hy
and the resulting earth response, two horizontal electrical
fields, Ex and Ey and the vertical magnetic field, Hz. The
resulting time-series data are recorded and Fourier
transformed to the frequency domain for further processing
to get the impedance tensors of the apparent resistivity and
phase.
For homogeneous earth, when monochromatic
electromagnetic plane waves propagate vertically
downward, the ratio of the electric to magnetic field
FIGURE 7: Interaction of solar wind with the
magnetosphere (SOHO, 2010)
FIGURE 8: Field layout for a 5-
channel MT data acquisition system
(Phoenix Geophysics, 2009)
Tamrat Fantaye 158 Report 12
intensity is a characteristic measurement of the electromagnetic properties of the medium, often called
characteristic impendence (Cagniard, 1953; Keller and Frischknecht, 1966):
μ
(7)
where Z = Characteristic impendence;
ω = Angular frequency (2πf), where f is frequency (Hz);
μo = Magnetic permeability in vacuum (H/m);
Ex,y = Electric field intensity (V/m) in x, y direction;
Hx,y = Magnetic field intensity (A/m) in x, y direction;
k = μ is the wave propagation number;
ε = Dielectric permittivity (F/m);
σ = Electrical conductivity (S/m).
For the quasi-stationary approximation, σ >> ωε, the wave propagation number is approximated to
and Equation 7 can be rewritten as:
μ
μ
√ μ μ л/ (8)
The phase difference between Ex and
Hy is π/4 =45° (Figure 9).
For homogeneous earth, the
resistivity is given as:
1
|| ;
(9)
E (V/m) is the electric field and the
practical units are mV/km and for the
magnetic field B and magnetic field
intensity or magnetization field H we
have B = μoH, where μo=4π×10-7 and
the practical unit is gamma (1 gamma
= 10-9 T). Imposing these practical
units of E and B to Equation 9 gives:
2
μ10
10
2
10 0.2|
| (10)
where ρ = Resistivity (Ωm);
T = Period (s)
And for a non-homogeneous earth, we define the apparent resistivity (ρa) and the phase (θa) as:
ρ
2
μ10
10
μ
2
10 0.2|
|
(11)
0.2|| arg 45°
FIGURE 9: Homogeneous half-space response of electric
and magnetic field intensity
Report 12 159 Tamrat Fantaye
where Zo = Impedance at the surface.
For horizontally N-layered earth (Figure 10), the plane wave impedance is given by the recursive
formula (Ward and Wannamaker, 1983) as:
;
(12)
where Zn =
0
(intrinsic impedance of the nth layer);
kn = );
hn = Thickness of the nth layer;
= Impedance in the nth layer;
Z1 = Zo is impedance at the surface.
For a two layered earth
(Figure 11) where layer one
has resistivity ρ1 and layer
two has resistivity ρ2,
Equation 12 becomes:
(13)
where Z1 = Zo = The impedance at the surface; and
For large periods, we have k1h1 << 1 and this implies tanh(ik1h1) ≈ ik1h1 and then Equation 13
becomes:
(14)
When the second layer at depth h1 is a good conductor, ρ1 >> ρ2 and Z1 >>
2, Equation 14 becomes:
(15)
and
1
||
1
2
(16)
Therefore, the depth to the good conductor, h1, can be calculated as:
FIGURE 10: N-layered earth
FIGURE 11:
Two-layered earth
Tamrat Fantaye 160 Report 12
2
(17)
When the second layer at depth h1 is an insulator, we have ρ1 << ρ2 and Z1 <<
2, and Equation 14
becomes:
1
1
1
1
(18)
where S = Conductance of layer 1.
The apparent resistivity can be rewritten as:
1
||
1
1
2
1
(19)
Therefore, the conductance of the uppermost layer can be calculated from ρa for long periods as:
2
(20)
The depth of investigation in MT is a function of subsurface resistivity and frequency (or the inverse
of the period) of the electromagnetic signals. The penetration depth can be roughly related to the
period by the use of the skin depth. The skin depth δ (m) at which the electromagnetic field amplitude
is reduced to e-1 of its original value at the surface is given as:
1
1
2
2
2 · 4 · 10
10
. 20/8 .
(21)
0.5
4.2 The central-loop TEM method
The transient electromagnetic method (TEM) is a
time domain method in which a current pulse is
transmitted and the decay of the induced
magnetic field is measured as a function of time.
A transient event is a short-lived burst of energy
in a system caused by a sudden change of state.
Unlike MT, in TEM a current is artificially
generated in the ground by a time varying field
or a step current transmitted into a loop on the
surface. A loop of wire is placed on the ground
and a constant magnetic field of known strength
is built up by transmitting a constant current into
the loop (Figure 12). When the current is turned
off abruptly, the magnetic field starts to decay
and induces electrical current in the ground. The
current distribution in the ground induces a
secondary magnetic field decaying with time. The rate of this decay is monitored by measuring the
FIGURE 12: Transient current flow in the
ground (modified from
Hersir and Björnsson, 1991)
Report 12 161 Tamrat Fantaye
voltage induced in a receiver coil at the centre of the transmitter loop in time gates (Figure 13). The
current distribution and the decay rate of the secondary magnetic field depend on the resistivity
structure of the earth. The decay rate, recorded as a function of time after the current in the transmitter
loop is turned off, can therefore be interpreted in terms of the subsurface resistivity structures
(Árnason, 1989).
The primary field impulse (transient) creates eddy currents below the transmitter loop and as the initial
near-surface eddy currents decay, they in turn induce eddy currents at greater depths. The rate of
change of the secondary field, due to induced eddy currents, is measured using an induction coil. The
depth of exploration that will be mapped in a vertical sounding configuration can vary from tens of
metres to a thousand metres, depending on the transmitter loop size.
The basic equation is:
,
(22)
where Ar = Cross-sectional area of the receiver coil (m2);
nr = Number of windings in the receiver coil;
μ = Magnetic permeability (H/m);
As = Cross-sectional area of the transmitter loop (m2)
ns = Number of windings in the transmitter loop;
r = Radius of transmitter loop (m).
So and To , which depend on layer resistivity and thickness, are given by the recursion relationships:
FIGURE 13: Waveforms of the primary and secondary magnetic field
(modified from Rowland, 2002)
Tamrat Fantaye 162 Report 12
cosh sinh (23)
sinh cosh (24)
with 1 ;
; and
where di = The thickness of the ith layer;
mi = The impedance of the ith layer.
So and To which determine the voltage (Equation 22) depend on angular frequency, ω and the
conductivities σi, through
where
and ε is the dielectric
permittivity. In the quasi-stationary approximation, σ >> ωε and hence k2 = -iωμσ.
From Equation 22, the mutual impedance for a central-loop configuration between the source and the
receiver coil is defined by the ratio of the measured voltage to the transmitted current as:
,
,
(25)
Equation 22 can be expressed in the time domain by a Fourier expansion of the function describing the
transmitted current (Árnason, 1989). If the transmitted current is described by function I (t), a Fourier
expansion of the current function will be:
1
2
(26)
or
1
2
(27)
The induced voltage in the receiver coil in terms of mutual impedance and the Fourier transform of the
transmitted current can now be written as:
,
1
2
, (28)
When the steady current is turned off, the voltage measured as a function of time is given by:
2
2
(29)
where Φ(ω) = Z(ω) /- iω is a function that depends on ω, its complex conjugate, Φ* is given as
Φ*(-ω) = Φ (ω).
Therefore, we have:
Re Re and Im Im (30)
Therefore, Equation 29 after some calculations becomes:
Report 12 163 Tamrat Fantaye
2
cos
2
sin
(31)
Transient voltage generated in the receiver coil due to a linearly ramped step function is given by
(Árnason, 1989):
(32)
If the current is turned off instantaneously, it would induce infinite voltage in the source loop.
Practically, the current is not abruptly turned off but rather in a linear manner in an interval of time
called turn-off time (TOFF). The turn-off time is measured by the transmitter and fed by the operator
into the receiver (Árnason, 2006b).
For a homogeneous half space of conductivity σ, the induced voltage in the receiving coil is
approximately given by (Árnason, 1989):
,
10
,
2 (33)
The time behaviour of the diffusing current has three phases: early times, intermediate times and late
times. The induced voltage is constant at the early stage and starts to decrease with time in the
intermediate stage. At late times, the measured voltage V(t) decays in time as t-5/2 and varies as σ3/2
(Árnason, 1989).
Apparent resistivity ρa, of a homogeneous half-space in terms of induced voltage at late times after the
source current is turned off is given by (Árnason, 1989):
4
2
5
,
(34)
where t = Time elapsed after the transmitter current is turned off (s);
Ar = Cross-sectional area of the receiver coil (m2);
nr = Number of windings in the receiver coil;
μo = Magnetic permeability in vacuum (H/m);
As = Cross-sectional area of the transmitter loop (m2);
ns = Number of windings in the transmitter loop;
Io = Transmitter current (A);
V(t,r) = Measured voltage (V).
5. MT AND TEM SURVEY AT KRÝSUVÍK HIGH-TEMPERATURE GEOTHERMAL
FIELD
5.1 Introduction
Krýsuvík is one of the high-temperature geothermal areas in Iceland, located at the centre of the
Reykjanes peninsula in SW-Iceland, about 25 km from the capital, Reykjavík (Figure 14). The
Krýsuvík high-temperature geothermal field is divided into five sub-fields: Trölladyngja, Hveradalir-
Seltún, Austurengjar, Köldunámur and Sandfell.
A total of 96 MT and more than 200 TEM sites have been acquired from the Krýsuvík hightemperature
geothermal field and the surrounding area since 1989 and 1-D inverted (Eysteinsson, 1999
Tamrat Fantaye 164 Report 12
and 2001; Hersir et al.,
2010). These data were
used in this report with the
permission of HS Orka to
study the subsurface
resistivity distribution of
the Krýsuvík hightemperature
geothermal
field. Multidimensional
inversion of the data was
done recently by Lemma
(2010) as a part of his MSc
work. In this study, 11 MT
and 11 TEM soundings on
a profile were processed
and 1-D joint inversion of
TEM and MT data was
done to correct the static
shift. The results are
presented as resistivity
cross-sections.
5.2 Geologic and tectonic settings of Krýsuvík and surroundings
Krýsuvík is characterised by extensive post-glacial lava fields, steep-sided mountains and ridges of
pillow lavas, pillow breccias, and
hyaloclastites (Figure 15). It is
located at the boundary between an
area of predominantly Interglacial
eruptions to the east and an area of
predominantly sub-glacial eruptions
to the west, with formations such as
hyaloclastite ridges. The Inter- and
Postglacial volcanism, i.e. volcanic
activity during the ice free periods, is
represented by sub-aerial volcanic
products and morphological
landscape like explosion craters and
lava flows. The common products
are lava flows, pyroclastic scoria,
welded lava and scoria and explosion
breccias.
Krýsuvík lies in a NE-SW elongated
valley within the active volcanic zone
characterised by fissure swarms
striking NE-SW. The Krýsuvík
fissure swarm is one of the large enechelon
structural units (fissure
swarms) of the Reykjanes Peninsula.
Different kinds of surface
manifestations observed in the
Krýsuvík geothermal system include:
FIGURE 15: Regional geological map of Krýsuvík and
surrounding area (taken from Abdelghafoor, 2007)
FIGURE 14: Krýsuvík and other geothermal areas in Iceland (map
from Ármannsson et al., 2000)
Report 12 165 Tamrat Fantaye
a clay alteration zone, boiling springs, warm springs, mud pools, warm soil, hydrothermal explosion
craters and mineralized water in Graenavatn Lake, deposits of silica sinters, oxidation, sulphate
deposits, steam vents and mineral veins.
5.3 Instrumentation and data processing
MT instrumentation and field layout: A 5-channel MT data acquisition system (MTU-5) and a 2-
channel system (MTU-2EP) from Phoenix Geophysics Ltd were used to record the MT raw data.
MTU-5 can acquire two channels of electric field data and three channels of magnetic data from coil
sensors. MTU-2EP can acquire two channels of electrical field data. The MTU controls the data
acquisition process and converts the signal into a digital format through 24 bit ADU. The instrument
synchronizes to Co-ordinate Universal Time (UTC) via signals from the Global Positioning System
(GPS) satellites.
A typical layout of a five component
MT unit consists of 5 electrodes, 4 of
which measure 2-perpendicular
horizontal components of the electric
field (Ex and Ey), and the fifth
electrode is used to ground the MTU
data logger at the centre. Out of the
3 induction coils used for magnetic
field measurement, 2 of them
measure the 2 perpendicular
horizontal components (Hx and Hy),
and 1 coil measures the vertical
component (Hz) of the magnetic field
(Figure 16).
Processing and interpretation of MT data: Time-series data downloaded from the MTU-5 unit were
viewed using the time series viewer in SSMT2000 program supplied by Phoenix Geophysics Ltd.,
Canada. The time series viewer is important in analysing the different components of electric and
magnetic fields recorded and used to investigate the noisy data (Figure 17). The correlation between
Ey with Hx and Ex with Hy is clearly seen on the viewer. The SSMT2000 program takes as input raw
time series files, calibration files, and site parameter files. In an intermediate step, it produces Fourier
coefficients, which are then reprocessed with data from the reference site, using robust routine
processing. The output is MT plot files (.MTH and .MTL) containing multiple cross powers and auto
FIGURE 16: Phoenix V5 system 2000 hardware
components (Phoenix Geophysics, 2009)
FIGURE 17: Time-series view of a 5-channel MT site
Tamrat Fantaye 166 Report 12
powers for each of the frequencies analysed. These MT plot files are used as input in the MT-Editor.
The MT-Editor program takes as input the MT plot files created by SSMT2000 and displays the
resistivity and phase curves as well as the individual cross powers that are used to calculate each point
on the curves. Cross powers that were affected by noise can be automatically or manually masked
from the calculations.
The main objective of editing using the MT-Editor is to get a smooth apparent resistivity curve (Figure
18) by eliminating from the calculation of each data point cross powers that were greatly affected by
noise. Auto or manual editing can be used according to the data quality; for a normal quality data (low
to moderate noise), it is best to start with auto editing and refine the result manually. For very noisy
data, it is good to start by deleting all the cross powers for a given frequency and then selectively
restoring the best.
The program is also used to display a variety of parameters of the plot files such as tipper magnitude,
coherency between
channels, and strike
direction. After editing
was finished, the
results were saved as
an MPK file and then
exported as industrystandard
edi (Electrical
Data Interchange) files
suitable for use with
geophysical interpretation
programs.
The output edi files
from MT-Editor were
then converted to a
UNIX EDI format and
a PostScript (ps) file
was created to view
parameters like
apparent resistivity and
phase curves (Figure
19). The TEMTD
FIGURE 18: MT-Editor output file for MT station 86,
showing apparent resistivity and phase curves
FIGURE 19: Measured apparent resistivity and phase curves from
sites 85 and 100 on the profile
Report 12 167 Tamrat Fantaye
program (Árnason, 2006a) was then used to invert the MT soundings (EDI files) jointly with the TEM
soundings collected from the same or a nearby site. This joint inversion is important for static shift
correction, as discussed in Section 4.
TEM instrumentation and field layout: A time domain electromagnetic equipment, PROTEM-67 from
Geonics, Ltd., Canada was used to collect the data. It included a digital receiver, a receiver coil and a
square receiver loop, a motor generator, current transmitter and current transmitting loop. The digital
receiver and the current transmitter timing are controlled by synchronized high-precision crystal
clocks. At each time when the current is turned off, the decay of the secondary magnetic field is
monitored by recording the voltage induced in the receiver coil using the digital receiver.
A small receiver coil with an effective area of 100 m2 and a flexible loop with an effective area of
5613 m2 were used along with a transmitter square loop of side 300 m. The transmitted current is
usually in the range of 20-24 A and data were recorded for both high and low frequencies. The
measured induced voltage data were stacked over many cycles and stored in the memory of the data
logger together with the corresponding setting information.
Processing and interpretation of TEM data: the raw data downloaded from the PROTEM receiver
were viewed and outliers masked using the TemX program from ÍSOR (Árnason, 2006b). The
program reads raw data from a PROTEM receiver and calculates the average and standard deviations
of the repeated transient voltage and computes late time apparent resistivity. It offers three types of
averaging, mean, median value and robust mean that reduce or eliminate outliers. Individual voltage
datasets and apparent resistivity for averaged voltage were viewed and noisy readings were masked
using the program. After site information was written in the header editing menu, the TemX program
produced an output file (.INV), ready for interpretation. An interpretation program from ÍSOR,
TEMTD, was used to invert either for voltage or apparent resistivity or both.
TEMTD can be used to invert not only TEM or MT data separately, but also for joint inversion of
TEM and MT data where the best static shift multiplier is determined. It uses gnuplot graphics
program for graphical display during the inversion process (Figure 20). For TEM data, the program
assumes that the source loop is a square loop and that the receiver coil or loop is at the centre of the
source loop. The
current wave form is
assumed to be a halfduty
bipolar semisquare
wave (equal
current-on and currentoff
segments), with
exponential current
turn-on and linear
current turn-off
(Árnason, 2006b). A
typical TEM inversion
model is shown in
Figure 21.
The apparent resistivity
and phases derived
from the determinant
of the MT tensor were inverted jointly with the nearby TEM data by this inversion program. As
shown in Figure 22, the measured TEM apparent resistivity (red/dark diamonds) curve that overlaps
the MT apparent resistivity curve (blue/grey) was used to correct the static shift; the shift multiplier is
shown in the upper right corner (0.714). In general, for the 11 stations, the shift multiplier ranged
from 0.4-1.3. The response of the resistivity model is shown to the right.
FIGURE 20: TEM data plot during inversion using TEMTD program
Tamrat Fantaye 168 Report 12
5.4 1-D inversion of resistivity data at Krýsuvík
A total of 11 TEM and 11 MT sites on a profile that crosses the main geological structures of the
Krýsuvík high-temperature geothermal field were analysed. The TEM data for the 11 sites was first
inverted and interpreted and then jointly inverted with the MT data. The results were used to create
resistivity cross-sections by using the TEMCROSS program, written by Dr. Hjálmar Eysteinsson
(Eysteinsson, 2010), and the GMT (generic mapping tool) program package. The location of the
profile along with main structures and features is shown in Figure 23.
FIGURE 22: Typical result of 1-D joint inversion of TEM and MT soundings
FIGURE 21: TEM inversion model
Report 12 169 Tamrat Fantaye
5.4.1 TEM and MT resistivity cross-sections
The TEM resistivity cross-section from 1-D inversion of each sounding on the profile is shown in
Figure 24. The resistivity structure, down to a depth of 1000 m b.s.l., was clearly mapped in the TEM
cross-section. The inversion model for each TEM station is shown in Appendix I. From the resistivity
cross-section, that runs from north to south, it can be seen, that high-resistivity layers with resistivity
> 70 Ωm are dominant in the uppermost part of the subsurface. This is due to unaltered volcanic
rocks, dry lavas, basalts and hyaloclastites. The presence of water in the subsequent layers decreases
the resistivity to a range of 10-100 Ωm. Below these resistive layers, conductive layers or cap with
resistivity values of < 10 Ωm were observed. This conductive cap was interpreted as the smectitezeolite
zone that characterises the high-temperature geothermal systems in Iceland (Árnason et al.,
2000). The third sequence of layers in this cross-section was a relatively resistive structure that could
be related to the chlorite-epidote zone. Correlation with temperature and alteration mineralogy is
discussed in Section 5.4.2.
FIGURE 23: Location of the resistivity profile showing: TEM stations
(diamonds), MT stations (dots), geothermal wells (inverted triangles),
surface manifestations (stars), faults and fissures (scattered lines)
Tamrat Fantaye 170 Report 12
The results of the joint inversion of the TEM and MT data are presented on the three resistivity crosssections
down to a depth of 2.5, 5 and 20 km as shown in Figures 25, 26 and 27, respectively. The
processed MT data are shown in Appendix II, while the joint TEM and MT inversion model for each
MT station is shown in Appendix III. It is clearly seen also in these resistivity cross-sections that
highly resistive fresh rocks dominate the uppermost layers followed by less resistive layers and then
by a conductive cap. This conductive cap formation is characteristic of high-temperature volcanic
geothermal field and is associated with the smectite-zeolite or mixed layered clay zone. Below these
layers, high-resistivity layers are observed which could be the indication of the high-temperature
alteration zone (chlorite-epidote zone). In the resistivity cross-section, which reaches down to a depth
of 20 km (Figure 27), a deep laying conductor was observed which can be seen under hightemperature
geothermal areas associated with volcanic systems in Iceland (Árnason et al., 2010).
FIGURE 25: Resistivity cross-section from 1-D joint inversion of TEM and
MT data down to a depth of 2.5 km b.s.l.
FIGURE 24: TEM resistivity cross-section along the profile down to 1000 m b.s.l.
Report 12 171 Tamrat Fantaye
5.4.2 Alteration mineralogy and temperature in drillholes
The resistivity cross-section along the profile down to 1000 m b.s.l. was compared to the nearby
drillhole data from wells KR-06 and KR-08 (Arnórsson et al., 1975), see Figure 25. The minerals
observed from the drillhole data include: calcite, laumontite, zeolite, gypsum, smectite, pyrite, mixed
clay minerals, epidote and chlorite. These secondary minerals of the basaltic rock are the result of
hydrothermal alteration under different pressures and temperatures.
FIGURE 26: Resistivity cross-section from 1-D joint inversion of TEM and
MT data down to a depth of 5 km b.s.l.
FIGURE 27: Resistivity cross-section from 1-D joint inversion of TEM and MT
data down to a depth of 20 km b.s.l.
Tamrat Fantaye 172 Report 12
KR-06: This drillhole is 843 m deep. The conductive structure which is correlated with the smectitezeolite
and mixed layered clay zone extends from 50 m to a depth of 450 m. The highest measured
drillhole temperature is at 500 m depth and is equal to 262°C (Figure 28). This measured value is
rather high for the smectite-zeolite and mixed layered clay zone (Figure 25), as it is typical for the
resistive chlorite-epidote zone which is indicative of temperature above 240°C. The measured bottom
hole temperatures are found to be less than 220°C, which also does not conform to the chlorite-epidote
seen there. These deviations show in the first place that in the uppermost part of the drillhole (at least
down to 500 m) the alteration is lagging behind the rock temperature and in the second place that the
geothermal system below 500 m is most likely cooling down and is not in equilibrium. In a
geothermal system, when cooling happens, alteration and resistivity remain stable.
KR-08: This drillhole is 933 m deep with the smectite-zeolite zone extending from 200 to 400 m and
the mixed-layered clay zone from 300 down to 700 m; the measured temperature down to 400 m is
190°C, which is in agreement with the alteration. However, the measured temperature at the bottom
of the drillhole was only found to be 170°C, in the chlorite-epidote zone. This is also an indication
that the geothermal system is presumably cooling down.
6. CONCLUSIONS
The resistivity cross-sections from the TEM and from 1-D joint inversion of MT and TEM data show
three major resistivity structures. The high-resistivity uppermost layers of resistivity >70 Ωm are
interpreted as unaltered basaltic dry lavas and hyaloclastites. The conductive cap (<10 Ωm) is
associated with a smectite-zeolite or mixed layered clay zone followed by resistive layers,
corresponding to the chlorite-epidote zone. Generally, a good correlation is found between the
subsurface resistivity structure and alteration mineralogy of the Krýsuvík high-temperature area.
The measured temperatures for the two drillholes, KR-06 and KR-08, were analysed and compared
with alteration mineralogy. In the chlorite-epidote zone, expected temperature values are above
240°C; however, here the measured temperature was less than 170°C for KR-08 and 218°C for KR-06.
The discrepancy between the measured and the expected values show that parts of the geothermal
system are probably cooling down. The resistivity values in this part of the geothermal reservoir
cannot reflect temperature because the hydrothermal alteration and the rock temperature are not in
equilibrium.
The 1-D joint inversion of MT and TEM data is a powerful method in mapping shallow to deep
resistivity structures. Joint inversion is helpful in correcting static shift caused by near surface
inhomogeneities.
FIGURE 28: TEM resistivity cross-section along the profile down to 1000 m b.s.l.,
also showing temperature in drillholes KR-06 and KR-08
Report 12 173 Tamrat Fantaye
ACKNOWLEDGEMENTS
It is my pleasure to thank the United Nations University Geothermal Training Programme and the
Government of Iceland for sponsoring me to participate in this specialized training. Special gratitude
goes to the director of UNU-GTP, Dr. Ingvar B. Fridleifsson and the deputy director, Mr. Lúdvík S.
Georgsson for offering me the opportunity to participate in the Geothermal Training Program. Thanks
to all staff members of UNU-GTP for their guidance and support.
Special thanks to my advisors Mr. Gylfi Páll Hersir and Mr. Knútur Árnason for their valuable time,
special lectures and guidance. I would also like to thank Dr. Hjálmar Eysteinsson for his great
support. All lecturers, Orkustofnun and ISOR staff members are greatly acknowledged for sharing
their knowledge and experience. I am also grateful to the Geological Survey of Ethiopia for
supporting my studies in Iceland.
Finally, I would like to thank the Almighty God for all his help.
REFERENCES
Abdelghafoor, M., 2007: Geological and geothermal mapping in Sveifluháls area, SW-Iceland.
Report 3 in: Geothermal Training in Iceland 2007. UNU-GTP, Iceland, 1-23.
Archie, G.E., 1942: The electrical resistivity log as an aid in determining some reservoir
characteristics. Tran. AIME, 146, 54-67.
Ármannsson, H., Kristmannsdóttir, H., Torfason, H., and Ólafsson, M., 2000: Natural changes in
unexploited high-temperature geothermal areas in Iceland. Proceedings of the World Geothermal
Congress 2000, Kyushu-Tohuku, Japan, 521-526.
Árnason, K., 1989: Central loop transient electromagnetic sounding over a horizontally layered
earth. Orkustofnun, Reykjavík, report OS-89032/JHD-06, 129 pp.
Árnason, K., 2006a: TEMTD. A program for 1D inversion of central-loop TEM and MT data. A
short Manual. ÍSOR, Reykjavík, 16 pp.
Árnason, K., 2006b: TemX. A graphically interactive program for processing central-loop TEM data.
A short manual. ÍSOR, Reykjavík, 10 pp.
Árnason K., Eysteinsson, H., Hersir G.P., 2010: Joint 1D inversion of TEM and MT data and 3D
inversion of MT data in the Hengill area, SW Iceland. Geothermics, 39, 13–34.
Arnórsson, S., Gudmundsson, G., Sigurmundsson, S.G., Björnsson, A., Gunnlaugsson, E., Gíslason,
G., Jónsson, J., Einarsson, P., and Björnsson, S., 1975: Krýsuvík. A general report on the geothermal
exploration. Orkustofnun, Reykjavík, report OS-JHD-7554 (in Icelandic), 71 pp.
Árnason, K., Karlsdóttir, R., Eysteinsson, H., Flóvenz, Ó.G., and Gudlaugsson, S.Th., 2000: The
resistivity structure of high-temperature geothermal systems in Iceland. Proceedings of the World
Geothermal Congress 2000, Kyushu-Tohoku, Japan, 923-928.
Cagniard, L., 1953: Basic theory of the magneto-telluric method of geophysical prospecting.
Geophysics, 18, 605-635.
Dakhnov, V.N., 1962: Geophysical well logging. Q. Colorado Sch. Mines, 57-2, 445 pp.
Eysteinsson, H., 2010: The resistivity plot programs, TEMMAP, TEMRESD and TEMCROSS. ÍSOR,
Reykjavík, unpublished manuals.
Eysteinsson, H., 1999: Resistivity soundings around Sandfell, Reykjanes Peninsula. Orkustofnun,
report OS-99002 (in Icelandic), 71 pp.
Tamrat Fantaye 174 Report 12
Eysteinsson, H., 2001: Resistivity soundings around Trölladyngja and Núpshlídarháls, Reykjanes
Peninsula. Orkustofnun, report OS-2001/038 (in Icelandic), 110 pp.
Flóvenz, Ó.G., Georgsson, L.S., and Árnason, K., 1985: Resistivity structure of the upper crust in
Iceland. J. Geophys. Res., 90-B12, 10,136-10,150.
Flóvenz, Ó.G., Spangenberg, E., Kulenkampff, J., Árnason, K., Karlsdóttir, R., and Huenges E., 2005:
The role of electrical conduction in geothermal exploration. Proceedings of the World Geothermal
Congress 2005, Antalya, Turkey, CD, 9 pp.
Hersir, G.P., and Björnsson, A., 1991: Geophysical exploration for geothermal resources. Principles
and application. UNU-GTP, Iceland, report 15, 94 pp.
Hersir, G.P., and Árnason, K., 2009: Resistivity of rocks. Paper presented at “Short Course on
Surface Exploration for Geothermal Resources”, organized by UNU-GTP and LaGeo, Santa Tecla, El
Salvador, 8 pp.
Hersir, G.P., Vilhjálmsson, A.M., Rosenkjær, G.K., Eysteinsson, H., and Karlsdóttir, R., 2010: The
Krýsuvík geothermal field. Resistivity soundings 2007 and 2008. ÍSOR - Iceland GeoSurvey,
Reykjavík, report ÍSOR-2010/025 (in Icelandic), 263 pp.
ISL, Michigan State University, 1999: Liquid molding. ISL, Michigan State University, webpage
islnotes.cps.msu.edu/trp/rtm/modl_mes.html.
Jones, A.G., 1988: Static shift of magnetotelluric data and its removal in a sedimentary basin
environment. Geophysics, 53-7, 967-978.
Keller, G.V., and Frischknecht, F.C., 1966: Electrical methods in geophysical prospecting. Pergamon
Press Ltd., Oxford, 527 pp.
Lemma D., Y., 2010: Multidimensional inversion of MT data from Krýsuvík high-temperature
geothermal field, SW Iceland, and study of how 1D and 2D inversion can reproduce a given 2D/3D
resistivity structure using synthetic MT data. University of Iceland, M.Sc. thesis, UNU-GTP, Iceland,
report 4, 94 pp.
Manzella, A., 2007: Geophysical methods in geothermal exploration. Italian National Research
Council, International Institute for Geothermal Research, Pisa, webpage:
http://cabierta.uchile.cl/revista/12/articulos/pdf/A_Manzella.pdf, 40 pp.
Phoenix Geophysics, 2009: Data processing. User’s guide. Phoenix Geophysics, Ltd., Toronto.
Quist, A.S., and Marshall, W.L., 1968: Electrical conductances of aqueous sodium chloride solutions
from 0 to 800°C and at pressures to 4000 bars. J. Phys. Chem., 72, 684-703.
Rowland, B.F., 2002: Time-domain electromagnetic exploration. Northwest Geophysical Associates,
Inc., 6 pp.
Simpson, F., and Bahr, K., 2005: Practical magnetotellurics. Cambridge University Press,
Cambridge, UK, 254 pp.
SOHO, 2010: Solar wind. Solar and Heliospheric Observatory, internet website:
http://sohowww.nascom.nasa.gov/gallery/images/magfield.html.
Sternberg, B.K., Washburn, J.C., and Pellerin, L., 1988: Correction for the static shift in
magnetotellurics using transient electromagnetic soundings. Geophysics, 53-11, 1459-1468.
Vozoff, K., 1991: The magnetotelluric method. In: Nabighian, M.N (ed), Electromagnetic methods
in applied geophysics, 2, 641-711.
Ward, S.H, and Wannamaker, P.E., 1983: The MT/AMT electromagnetic method in geothermal
exploration. UNU-GTP, Iceland, report 5, 107 pp.
Report 12 175 Tamrat Fantaye
APPENDIX I: TEM 1-D inversion models
The measured TEM data curve is shown with red (dark) dots; the calculated TEM data curve is a black
line connecting the red (dark) dots, and the 1-D layered modelling is in green (gray).
Tamrat Fantaye 176 Report 12
APPENDIX II: MT data (EDI)
Report 12 177 Tamrat Fantaye
Tamrat Fantaye 178 Report 12
Report 12 179 Tamrat Fantaye
Tamrat Fantaye 180 Report 12
APPENDIX III: TEMTD 1-D TEM and MT joint inversion models
The red (dark) line represents the TEM curve and blue (gray) represents the MT curve. The green
(gray) line to the right represents 1-D layered resistivity modelling.
Report 12 181 Tamrat Fantaye
Tamrat Fantaye 182 Report 12