an electrical resistivity model of the southeast

An Electrical Resistivity Model of the Southeast

Australian Lithosphere and Asthenosphere

Thesis submitted in accordance with the requirements of the University ofAdelaide for an Honours Degree in Geophysics.

Kate Robertson

October 2012

Electrical Resistivity in Southeast Australia 3

ABSTRACT

A combination of magnetotelluric and geomagnetic depth sounding data were used

to attempt to image the electrical resistivity structure of southeast Australia, to

investigate the physical state of the crust and upper mantle. A 3D forward model of

southeast Australia comprised of regional sets of broadband and long-period mag-

netotelluric and geomagnetic depth sounding data, over an area of 440 x 300 km2,

was used to map broad-scale lithospheric properties. Model results show an order of

magnitude decrease in resistivity from the depleted continental mantle lithosphere

of the Delamerian Orogen in the west, to the more conducting oceanic mantle of

the Lachlan Orogen in the east. The decrease in resistivity in conjunction with a

0.1 km/s decrease in P-wave velocity at depths of 50-250 km, suggest a change in

temperature (∆T∼200C) due to lithospheric thinning toward the east as the likely

cause, in conjuction with a change in geochemistry and/or hydration. A high reso-

lution two-dimensional inversion using data from 37 new and 39 existing broadband

magnetotelluric stations mapped crustal heterogeneity beneath the Delamerian Oro-

gen in much greater detail. Lateral changes in resistivity from 10-10 000 Ωm occur

over the space of a few kilometres. Low resistivity (∼10 Ωm) regions occur at depths

of 10-40 km. Narrow paths of low resistivity extend to the surface, coinciding with

locations of crustal faults from seismic interpretations. Movement of mantle fluids

up these faults, during periods of extension prior to the Delamerian Orogen, may

have produced a carbon-rich, low resistivity lower crust, leaving a resistive upper

mantle, depleted of volatiles.

KEYWORDS

electrical conductivity, resistivity, magnetotellurics, lithosphere, crust, upper man-

tle, Delamerian Orogen, Lachlan Orogen, southeast Australia

4 Electrical Resistivity in Southeast Australia

Table of Contents

Introduction 7

Lithospheric Resistivity 9

Previous Results From Seismic Studies 10

MT Theory 11

MT Distortion and Phase Tensor Theory . . . . . . . . . . . . . . . . . . . 13

Geomagnetic Depth Sounding . . . . . . . . . . . . . . . . . . . . . . . . . 15

Geological Background 16

Observations and Results 18

3D Regional Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

3D Forward Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

Crustal scale MT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

Data Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

Data Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Responses, Strike and Dimensionality . . . . . . . . . . . . . . . . . . 23

Phase Tensor Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 25

2D Inversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Occam 2D Model with Seismic Interpretation . . . . . . . . . 27

Sensitivity analysis . . . . . . . . . . . . . . . . . . . . . . . . 29

Occam 2D Response Curves . . . . . . . . . . . . . . . . . . . 29

Discussion 30

3D Forward Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

2D Inversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Comparison of MT with Seismic Interpretation . . . . . . . . . . . . . 31

Possible Causes of Mid-Lower Crustal Low Resistivities . . . . . . . . 31


Geological Implications . . . . . . . . . . . . . . . . . . . . . . . . . . 32

Crustal and Mantle Heterogeneity . . . . . . . . . . . . . . . . . . . . . . . 33

Conclusions 34

Acknowledgments 35

References 35

Appendix A 39

Appendix B 40


List of Figures1. Elliptical representation of the phase tensor . . . . . . . . . . . . . . 152. Geological map of southeast Australia with MT and GDS station

locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163. Cross-section of P-wave velocity structure beneath the Delamerian

and Lachlan Orogens, with electrical resistivities for forward modelincluded . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

4. Induction arrows of two forward models at 0.01 Hz; One with a west-east decrease in resistivity across velocity boundary between Delame-rian and Lachlan Orogens, one unchanging across boundary . . . . . 21

5. Pseudosection of TE and TM modes, indicating data quality andstatic shift effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

6. Geoelectric strike . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257. Phase tensor pseudosection . . . . . . . . . . . . . . . . . . . . . . . . 268. Phase tensor ellipses over a total magnetic intensity map . . . . . . . 279. Occam smooth 2D inversion of all data points of the Southern De-

lamerian transect, SD01’ . . . . . . . . . . . . . . . . . . . . . . . . . 2810. Phases and apparent resistivity values for stations SD01, DB24C,

DB30C and SD38 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3011. Appendix A-Station information spreadsheet . . . . . . . . . . . . . . 3912. Appendix B-Processing information spreadsheet . . . . . . . . . . . . 40


INTRODUCTION

A better understanding of the electrical resistivity structure underlying southeast

Australia would improve the current knowledge regarding the physical state of the

crust and upper mantle. Uncertainties involving this physical state include: hy-

dration of the upper mantle, rheological controls defining the depth and nature of

the lithosphere-asthenosphere boundary, formation and deformation history of the

sub-continental lithospheric mantle, and broad-scale heterogeneity of the continen-

tal lithosphere and asthenosphere to the transition zone.

With the exception of xenolith/xenocryst data (O’Reilly & Griffin 1985), geo-

physical surveys are the only means of investigating depths associated with the

mid to lower crust and upper mantle. Gravity, magnetic, electromagnetic (EM)

and reflection seismic (Brewer & Oliver 1980; Korsch et al. 2002) surveys can all

image crustal depths. Reflection seismics show major faults and structural bound-

aries within the crust, which preserves the brittle and ductile footprints of events

(Brewer & Oliver 1980). Magnetotellurics (MT) (Dennis et al. 2012; Hamilton

et al. 2006; Heinson & White 2005; Jones 1992) and seismic tomography (Graeber

et al. 2002; Rawlinson & Kennett 2008; Rawlinson & Fishwick 2011) are the only

methods in common use capable of investigating sub-Moho depths, with the latter

commonly used for estimating elasticity (a function of temperature and composition)

in the upper mantle to asthenosphere. Seismic tomography is limited in its ability to

resolve crustal and even some upper mantle structures (Rawlinson & Fishwick 2011).

The signature of the geological processes that determine the physical state of the

lithosphere are substantially different at mantle and crustal scales. MT has played

an increasing role at both mantle and crustal scales (Bedrosian 2007), with different

approaches often used for deep lithosphere (grid array of MT sites) and crustal struc-

tures (transects of MT sites), however these approaches are interchangeable. MT is


a passive EM technique that records the electrical resistivity response of the Earth

as a function of different inducing frequencies (Cagniard 1953; Tikhonov 1950). Low

resistivity (i.e. high conductivity) could be due to changes in one or more of the

following: temperature, melt fractions, hydration and trace mineralogy, making it a

diverse method for determining the physical state of the Earth (Selway 2012). For

maximum benefit, combined interpretation of MT in conjuction with other geophys-

ical data sets is recommended (Jones 1987).

In this study, a series of 3D forward models were used to determine the regional-

scale variability of the southeast Australian electrical resistivity structure. Two

forward models were constructed to investigate if a change in electrical resistivity

occured at the location of a change in velocity of 0.1 km/s where the Delamerian and

Lachlan Orogens meet (Graeber et al. 2002; Rawlinson & Fishwick 2011). New and

existing broadband (0.0064s-81.92s) MT data of the Southern Delamerian transect

were then used to determine the electrical resistivity structure of the Delamerian

Orogen subsurface in more detail.

Both of these new models of electrical resistivity, combined with existing passive

seismic velocity models, geometric boundaries from reflection seismology and addi-

tional geochronological restraints were used to better constrain the physical state of

the lithosphere in southeastern Australia. Particular items addressed in this study

include; the length and resistivity-scale over which changes in the lithosphere occur,

the temperature and hydration state of the lithosphere from electrical resistivity

and seismic tomography, comparisons of mantle and crustal heterogeneity, and how

crustal heterogeneity relates to crustal deformation imaged by reflection seismics

and mapped by geology and geochemistry.


LITHOSPHERIC RESISTIVITY

Low resistivity regions in the upper mantle are caused by higher temperatures, in-

terconnected conducting phases, hydrated minerals (Karato 1990) and iron content

(Jones 1999). Lilley et al. (1981) have found that at depths greater than 150 km,

electrical resistivity from central Australia to southeast Australia decreases by an

order of magnitude. The suggested explanation for this, at the time of the study,

was the presence of a small fraction (approximately 5%) of basaltic melt. Studies

by O’Reilly & Griffin (1985) also suggested elevated temperatures in southeast Aus-

tralia when compared with normal continental geotherms.

The possible existence of a partial-melt zone under southeast Australia is sup-

ported by EM measurements, seismic studies and thermal models which involve

crustal intrusion from a sub-lithospheric magma source. Gaillard et al. (2008) pro-

posed that carbonatite melt in the top of the asthenosphere is capable of producing

conductive anomalies present in the upper asthenosphere. A study by Yoshino

(2010) suggests at pressures of 3 GPa, carbonatite melt bearing olivine aggregates

show an order of magnitude lower resistivity than silicate melt bearing olivine aggre-

gates. Yaxley et al. (1998) provide evidence for the presence of carbonatite in melts

in southeast Australia, offering an alternate explanation to basaltic melt (Lilley

et al. 1981) causing the lower resistivity to the east.

Low electrical resistivity (i.e. high conductivity) regions within the crust are

commonly explained by highly interconnected, conductive phases in grain boundary

areas (e.g. Wei 2001). Yang (2011) states that this mechanism for low resistivity

is not always supported by geochemical and geophysical data. He suggests that in

parts of the lower continental crust (particularly beneath stable areas), low resistiv-

ity can often be accounted for by solid-state conduction of the main constituents,

due to chemical composition (including some major elements and water content),


temperature, and mineral fabrics.

The South African MT Experiment (Hamilton et al. 2006), found that the crustal

MT results obtained in South Africa are strongly related to large-scale geological

structures within the Kapvaal Craton and the surrounding terranes. Resistivity

shows much greater variability at crustal depths than for upper mantle depths, with

only some crustal structures still distinct at mantle depth.

The use of 3D forward modelling and 2D inversions of MT and geomagnetic depth

sounding (GDS) data will determine if crustal resistivity is heterogeneous at smaller

scale-lengths than upper mantle depths in southeast Australia.

PREVIOUS RESULTS FROM SEISMIC STUDIES

Rawlinson and Fishwick (2011) constructed a high-resolution, 3D P-wave velocity

model of the southeast Australian upper mantle, to a depth of 300 km. A distinct

feature that the model resolved was a higher velocity region that they associated

with the Delamerian Orogen, extending beneath the lower velocity Lachlan Oro-

gen separated by the presence of an easterly dipping velocity transition zone, most

prominent at depths of 150-200 km. The transition zone is marked by an increase

in velocity at these dpeths of ∼0.1

The sudden variation in velocity of 0.1 km/s across the boundary could be a

result of elevated temperatures. Higher temperature to the west could arise from

lithospheric thinning towards the east as a result of the break-up of Australia and

Antarctica and the opening of the Tasman Sea. The velocity model supports a

change from Proterozoic mantle lithosphere of continental origin to Phanerozoic

mantle lithosphere of oceanic origin, and is consistent with the changes in both


composition (Cammarano et al. 2003; Griffin et al. 1998) and temperature (Faul &

Jackson 2005) required for these two mantle types.

MT THEORY

The MT method was first suggested by Tikhonov (1950) and later expanded upon

by Cagniard (1953). A summary of the theory of MT will be presented here. For a

comprehensive theory of MT see Simpson (2005) and Chave (2012). Unless stated

otherwise these books (and references therein) are the source of information for this

MT Theory section.

MT is a passive EM technique capable of determining the electrical resistivity

response of the Earth down to depths of more than 600 km. MT uses naturally

occurring geomagnetic variations external to the Earth as a power source for EM in-

duction. The fluctuations in the natural magnetic fields and induced electric fields

are measured in orthogonal directions at the Earth’s surface. The distance over

which EM fields diffuse is dependent on the frequency of the variations and the

resistivity of the Earth, allowing MT to yield both lateral and vertical constraints.

The periods of these variations that are naturally occurring are of the order of 10−3

to 105 s (frequencies 103 to 10−5 Hz). At frequencies of 0.5-5 Hz, the natural source

signal has low amplitude, known as the dead-band, resulting in a low signal-to-noise

ratio in the measured data.

In the frequency domain, complex ratios known as impedances can be derived,

that describe the natural variations of the electric (E) and magnetic fields (H). The


impedance tensor Z is defined as follows:

Ex

Ey

=

Zxx Zxy

Zyx Zyy

Hx

Hy

(1)

For a 1D resistivity structure, the diagonal components Zxy = −Zyx = Z,

Zxx = Zyy = 0. For a purely 2D structure, the impedance tensor can be rotated

to eliminate Zxx and Zyy. The TE mode has the electric field parallel to the strike,

and the TM mode has the magnetic field parallel to the strike.

Z is complex and can be separated into real (X) and imaginary (Y) parts:

Z = X + iY (2)

The components of the matrix Z have both magnitude and phase;

The apparent resistivity (ρaij(ω) - the magnitude of Z, where ω = 2πf is the an-

gular frequency), is the average resistivity for the volume of Earth penetrated by a

particular MT sounding period, T , where T = 1f, and is given by the equation:

ρaij(ω) =1

µ0ω|Zij(ω)|2 (3)

. If the Earth were homogeneous, the apparent resistivity would be the actual re-

sistivity. However for the true, multi-dimensional Earth, the apparent resistivity

is actually the average resistivity represented by a uniform half-space (a volume

bounded only by an infinite plane).

The phase (component of Z, Φij) represents the phase difference between the

electric and magnetic fields that are used to calculate the MT impedance, and is

equal to:

Φij = tan−1=Zij<Zij

(4)


For a uniform half-space, impedance phases are 45 . Impedance phases >45 are

indicative of resistivity decreasing with depth, and impedance phases less than

45 means resistivity increases with depth.

Penetration depths depend on the EM sounding period and the Earth’s resistivity

structure. The depth of penetration of which resolvable signal is obtained approxi-

mately equals one skin depth (the depth at which EM fields are attenuated to e−1

of their surface amplitude), where a skin depth ρ(T) in simplified form is given by;

ρ(T ) = 500√Tρ (5)

From this it can be seen that at longer periods, waves are attenuated more slowly

than shorter periods, and therefore penetrate deeper into the Earth.

MT Distortion and Phase Tensor Theory

Heterogeneities in the near-surface resistivity cause galvanic effects, which distort

the regional MT response, and static shift, which distorts the depth and resolution

of underlying structures (Jones 1988). The phase tensor is extremely useful for

strike and dimensionality interpretation, because it preserves the regional phase

information, and is the component of the impedance tensor that is unaffected by

galvanic distortion or static shift (Caldwell et al. 2004; Bibby et al. 2005). Use of

the phase tensor is suitable even if both heterogeneity and resistivity structures are

3D, as no assumptions about the resistivity dimensionality are required to calculate

it. The phase tensor, Φ is represented by the equation

Φ = X−1Y (6)


where X and Y were defined in Equation 2, as the real and imaginary components

of Z. The coordinate invariants of the phase tensor are the minimum and maximum

phases and the skew angle. For 1D regional electrical resistivity structures, the

phase tensor is represented by a single coordinate invariant phase equal to the 1D

impedance tensor phase. For 2D regional electrical resistivity, the phase tensor is

symmetric with one of its principal axes aligned parallel to the regional strike. The

geoelectric strike is the preferred direction of current flow and is given by α or α+90

(as there is ambiguity between the two principal axes (Caldwell et al. 2004)), where:

α =1

2tan−1Φxy + Φyx

Φxx − Φyy

(7)

. In the 3D case the phase tensor is non-symmetric and a third coordinate invariant,

the skew (β), is introduced. The skew is a measure of the asymmetry of the regional

MT response, and is given by:

β =1

2tan−1Φxy − Φyx

Φxx + Φyy

(8)

.

The phase tensor can be represented graphically by an ellipse (Caldwell et al. 2004)

(Figure 1), which gives a visual indication of resistivity-dimensionality and strike di-

rection. The major and minor axes of the ellipse represent the principal axes of the

tensor. The orientations of the phase tensor principal axes reflect lateral variations

(gradients) in the regional resistivity structure. Phase tensor pseudosections (see

Figure 7) are a convenient way of visualising these gradients.


Figure 1: Elliptical representation of the phase tensor. The ellipse axes representthe principal axes of the phase tensor and α represents the geoelectric strike. Ifthe phase tensor is non-symmetric (i.e. the resistivity structure is 3D), a thirdcoordinate invariant known as the skew is needed (represented by β).

Geomagnetic Depth Sounding

Geomagnetic depth sounding was developed when Schuster (1889) showed the exis-

tence of magnetovariational fields arising from induction. Like MT, GDS determines

the electrical resistivity structure of the subsurface. However, in MT, where either

4 or 5 components are measured (Ex, Ey, Hx, Hy, ± Hz), only the three magnetic

components are measured (Hx, Hy, Hz) in GDS. The advantage of not measuring

the electric fields in GDS, is immunity from the effects of local distortions of the

electric field, manifest as static shift (Jones 1988) and galvanic distortion (Bibby

et al. 2005). A reference station outside of the survey area is used to improve signal

to noise ratios (Gamble et al. 1979). GDS can probe over scale-lengths both lat-

erally and vertically of a few hundred kilometres, as it detects lateral variations in

resistivity.

In most studies, induction arrows are plotted for graphical representations of

these lateral heterogeneities in the resistivity structure. These induction arrows


are vector representations of the complex ratios of vertical to horizontal magnetic

field components. The vertical magnetic fields are generated by lateral resistivity

gradients, so induction arrows can be used to infer the presence or absence of lateral

variation in resistivity (Whellams 1989). The Parkinson convention for induction

arrows is most commonly used (Parkinson 1959), where the arrows point towards

conductive bodies. The length of the arrow is determined by a combination of the

magnitude of the anomaly and its distance from the MT site, with range increasing

for longer periods.

GEOLOGICAL BACKGROUND

Figure 2: Simplified geological map with approximate locations of younger coversequences. The locations of all MT and GDS stations used in this study have beenincluded. The east-west (SD01’) and north-south (SD02’) Southern Delamerian MTtransects are highlighted with a blue box around it. The inset shows the region inthe blue box zoomed in, with stations used in Figure 10, labelled. Note the transectcrosses the Glenelg Zone, the Yarramyljup Fault and the Grampian Stavely Zone.Adapted from Rawlinson and Fishwick (2011).


The southeast Australian lithosphere formed during a Palaeozoic Orogeny that

commenced in the Early Cambrian and ceased in the Middle Triassic (Glen 2005).

The new and reworked lithosphere from this orogeny, covering the eastern third of

Australia is collectively known as the Tasmanides (Glen 2005). Orogens comprising

the Tasmanides include the Delamerian Orogen, the Lachlan Orogen to the east and

the New England Orogen along the east coast of Australia (Foster 2000).

The Delamerian Orogeny began 514 ± 3 Ma (Foden et al. 2006), lasting for

approximately 24 m.y., ending 490 Ma with a period of rapid uplift, cooling and

extension. The Delamerian Orogen consists of Precambrian and Early Cambrian

rocks that experienced deformation and metamorphism in the Cambrian (Foden

et al. 2006). The Australian Precambrian cratons are separated from the younger

Palaeozoic to Mesozoic orogenic belts of eastern Australia by the Delamerian Oro-

gen. Westward-verging folds and thrust faults are abundant in the region (Jenkins

& Keene 1992). The orogen comprises the Adelaide Fold belt in South Australia,

the Glenelg Complex in western Victoria and parts of western New South Wales and

Tasmania. The MT transect, SD01’, of which the second phase of data collection

was part of this study, runs east-west through the Adelaide Fold Belt and across the

Glenelg and Grampian-Stavely Zones (see Figure 2).

The Ordovican-Devonian Lachlan Fold Belt in western Victoria developed from

about 485 to 340 Ma (Kemp et al. 2005). The geodynamic plate tectonic setting

for the Lachlan fold belt is still unresolved but may include an accretionary oro-

gen system (Foster 2000), with numerous complications such as the inclusion of

rifted Proterozoic continental fragments into the region during deformation (Cayley

et al. 2002; Cayley & Taylor 1997).


The location of the boundary between the Delamerian Orogen and the Lachlan

Orogen has been a topic of debate (Vandenberg 1978; Glen 1992; Baille 1985). A

more recent study by (Miller et al. 2005) deduced that the boundary is likely to be

complex and transitional across the vicinity of the Stawell Zone in western Victoria

(Miller et al. 2005).

In addition to this Palaeozoic history, this region lay just to the north of the

breakup of Australia and Antarctica. This resulted in the formation the Otway Basin

and the opening of the Tasman Sea, between 80-90 Ma, causing lithospheric thinning

near the present passive margin of southeast Australia. Mafic underplating may be

associated with the formation of the passive margin (Rawlinson & Fishwick 2011).

The Newer Volcanic Provinces, located in western and central Victoria, consist of

substantial Quaternary basalt cover that originated from hot-spot volcanism (Price

et al. 1997).

OBSERVATIONS AND RESULTS

3D Regional Modelling

3D FORWARD MODEL

WinGLink R© (Rodi & Mackie 2001), a non-linear conjugate algorithm, was used to

create a database consisting of; two broadband MT transects comprising 14 sta-

tions, a grid of 39 long-period MT stations (Aivazpourporgou et al. 2012), a grid

of 25 GDS stations (Whellams 1989), in addition to the 39 existing and 37 newly

attained (in this study- see subsections ’Data Acquisition’ and ’Data Processing’ for

more information) broadband MT stations of the Southern Delamerian transects

(see Figure 2 for all station locations).


Figure 3: A cross section of the P-wave velocity structure (Vp) from seismic tomog-raphy in southeast Australia, at 36.7 S, 138-150 E. Red indicates regions whereP-waves travel slower, purple indicates faster regions. Depths greater than 300 kmwere not included in the seismic tomography model and are coloured grey. Thelocation of the transitional velocity region at the proposed boundary of the De-lamerian/Lachlan Orogens is the east-dipping dashed line at 142 E. The resistivityvalues marked on the figure indicate the values that were used in the second 3Dforward model of MT and GDS data, Model B, to test if the change in velocityacross the boundary could be modelled by a change in resistivity. Model A had nochange in resistivity across the boundary, with all values set to those of the westernhalf of Model B. The MT transect (SD01’) is at approximately 140-142 E (notshown on this map). Adapted from Rawlinson and Fishwick (2011).

Construction of two 3D forward models aimed to put the newly acquired Southern

Delamerian MT data into a regional framework. Resistivity values were set using

findings from Heinson and White (2005). For depths down to 40 km in Model A,

the first model, a 1D resistivity structure was assumed (Figure 3). Bathymetry and

the Otway Basin sediments (location and depth as indicated by the cross sections

from Petkovic (2004), were incorporated into the model. At depths greater than

40 km, resistivities to the west of the apparent seismic boundary between the De-

lamerian and Lachlan orogens were set one order of magnitude greater than those

to the east. For the Newer Volcanic Provinces, which is a low velocity zone, resis-

tivities were set anomalously low (∼45 Ωm at depths of 120-180 km) (Rawlinson &

Kennett 2008; Rawlinson & Fishwick 2011). An alternative model, Model B, was


also constructed, with resistivites unchanging across the velocity boundary (all val-

ues were set to that of the resistive Delamerian Orogen in Figure 3). Both models

extended to depths exceeding 800 km.

Forward modelling results are shown in Figure 4. The induction arrows for each

model were plotted for frequencies of 0.1 Hz, 0.01 Hz, 0.001 Hz and 0.0002 Hz to

investigate the fit of the two models with the actual data. The closer in magni-

tude and orientation the induction arrows are, the closer that the model created in

WinGLink R© is to the actual structure of the Earth. The modelled induction ar-

rows reproduce the observed induction arrows better when the model incorporates

a change in resistivity across the Moyston Fault, to correspond with the 0.1 km/s

change in P-wave velocity (Figure 3). Figure 4 shows the best fit induction arrows,

occurring at a frequency of 0.01 Hz. To the middle of the figure and further east is

where the arrows of Model B show particular improvement over those of Model A.

The lower resistivity in the east has rotated the modelled data arrows so that they

have a more similar orientation to the observed data arrows. Induction arrows at

frequencies higher than the 0.01 Hz (at shallower depths) plotted in Figure 4, do

not show any trends and point in many different directions, due to local effects of

low resistivity crustal regions.

The MT stations of SD01’ and SD02’, do not have induction arrows as only the

x and y (i.e. horizontal) components of the magnetic field were recorded; the z-

component is also required to create induction arrows. However, three stations from

Aivazpourporgou et al. (2012) long-period grid overlap the eastern end of SD01’,

with induction arrows for periods greater than 100 s (frequency <0.01 Hz) fitting

well in this region, indicating that below the transect, the mantle is predominantly

homegeneous.


Figure 4: The real induction arrows of the two different forward models. Thearrows point towards the anomalous internals concentrations of current. The redarrows are the induction arrows from the data (note that the Southern Delamerianstations do not have induction arrows). The blue arrows indicate the inductionarrows created by the 3D WinGLink R© forward model. How well the data fits themodel is given by how closely the magnitude and orientations align.a) The induction arrows for Model A, with no change in resistivity across the bound-ary.b) The bottom figure shows the induction arrows for Model B, the model with theorder of magnitude decrease in resistivity, west to east, across the boundary.


Crustal scale MT

DATA ACQUISITION

In June 2012, broadband MT data were collected at 37 sites along a 75 km east-

west transect in west Victoria, Australia. Locations of the sites can be found in

Appendix A. Data collection comprised the second phase of a two-part survey. The

first 39 stations were collected in 2009 for Phase 1, along the main transect running

east-west labelled SD01’ (stations SD01-SD10, SD14, SD21-39), and a much shorter

transect, SD02’, (stations SD11-SD13, SD15-SD20) running north-south. The sur-

vey was run to coincide approximately with the 2009 Southern Delamerian AuScope

seismic survey line.

The stations of Phase 2 were placed along the eastern half of the 140 km SD01’

transect. The second survey was conducted because preliminary inversions of the

original 39 stations (unpublished) suggested a much more heterogeneous resistivity

structure to the east than the west. This corresponds with more complex and

steeply dipping geology within this region. An extra 1-3 sites were infilled for each

existing site SD24-SD39 (named DB24a, DB24b etc.see Figure 2). Combining the

two surveys resulted in a 67 station, 140 km transect, SD01’, with 5 km spacing for

stations 1-24, and approximately 1.2 km spacing for sites 24-39. Data were recorded

at a frequency of 1000 Hz for a duration of 20-40 hours using AuScope equipment.

Each station recorded two horizontal components of the electric field using 40-50 m

dipoles (Ex, Ey with x north-south and y east-west) and two components of the

magnetic field (Hx, Hy). At least 3 stations recorded simultaneously allowing for

remote referencing (Gamble et al. 1979).


DATA PROCESSING

The time series data for each site were processed and converted from the time-domain

to the frequency-domain, using a robust remote referencing code, BIRRP (Chave &

Thomson 2004). EDI files were output from this code, which contain information

regarding the impedance tensor in the frequency domain. 50, 150 and 250 Hz noise

occurred when stations were situated near powerlines, and were removed using a

notch filter. A regional geomagnetic declination of approximately 10 has been cor-

rected for, ensuring that the x and y directions represent geographic north and east.

Apparent resistivity and phases were calculated from MT impedance components

and plots of these values were used to analyse noise across the processed frequency

range. If plots appeared to show noisy data, a different time window was chosen

to process. Coherence values comparing the electric response with the magnetic

response provided a further indication of the quality of the data. Most stations

recorded good quality data.

RESPONSES, STRIKE AND DIMENSIONALITY

The pseudosection in Figure 5 was plotted using the processed data. The modes

are quite smooth, with minor evidence of static shift effects, apparent in the vertical

shift between stations of the apparent resistivities. The effect of this small shift

would be minimal, however, so it was not corrected for prior to inversion (Figure 5).

The regional geological strike is north-west, as evident from TMI images and

geological maps. Strike analysis returned geoelectric strike directions of 37 east of

North. Prior to inversion, all data were rotated by an angle of 45 east of North.

Dimensionality of the resistivity structure is given by the shape of the ellipse: a

circular phase tensor indicates 1D resistivity structure and an elliptical phase tensor

could indicate either 2D or 3D resistivity structure. Analysis of the skew and ellip-

ticity for each station, at every period, was undertaken to find 3D effects. Periods

where the ellipticity was greater than 0.1 and the skew was outside the range of


Figure 5:a) The apparent resistivity ( Ωm) for the TE mode. Small vertical shifts station-to-station are visible.b) The phase () of the TE mode. The phase is unaffected by static shift anddistortion.c)The apparent resistivity for the TM mode; note that it is much smoother thanthe TE mode, and static shift effects are less.d)The phase of the TE mode. There is some evidence of static shift effects,

-5 to 5 were said to be 3D. Very few data points had 3D effects and the strike is

reasonably invariant with depth; it was determined that 2D modelling of the data

were appropriate.


Figure 6: The angle of geoelectric strike plotted against the sampling period, forevery station. At short periods, a trend is not obvious, however at longer periods aconstant strike of about 40 east of North is evident, indicated by the aqua dashedline

PHASE TENSOR ANALYSIS

In Figure 7, the minimum phase angle is less than 45 for periods up to 1.28 s,

which corresponds to an increase in resistivity with depth. At slightly longer peri-

ods (greater than 1.28 s) the minimum phase angle is greater than 45 indicating

that resistivity decreases with depth.

The orientation angle of the ellipses across the stations is approximately 45 east

of North at longer periods, with little variance. At short periods, ellipses can be

grouped into 4 broad regions of similar orientation and shape, which are outlined in

black in Figure 7. The western-most group, group 1, has very thin ellipses, indicating

possible current channeling with low resistivity in the direction of elongation. For

group 2, at periods less than 1.28 s, ellipses are almost circular. This is indicative of

a 1D environment which is typical for sediments. Group 3 has very uniform ellipses

that have an angle of orientation of approximately 20 east of North. This group

corresponds to a very resistive section in the MT inversion (Figure 9. Group 4 is

similar to the west-most group and corresponds to a low resistivity region, C3 (see


Figure 7: Elliptical representation of the phase tensors displayed in a pseudosec-tion. The distance from station DB24A to DB38C is approximately 75 km. Thedirection of the ellipse axes align with the direction of maximum conductivity. Thecolour is a representation of the invariant minimum phase value relating to resistiv-ity changes with depth (phase >45 implies resistivity decreases with depth, phase<45 implies resistivity increases with depth). The black boxes group trends ofellipses, and are numbered 1-4 in red.

Figure 9).

Figure 8 shows phase tensors at a frequency of 0.1 Hz on top of a total magnetic

intensity (TMI) image. The northwest geological strike can be seen. The phase

tensors are oriented approximately northeast, with the phase tensors to the east

rotating slightly counter-clockwise.


Figure 8: This figure provides another way of viewing the phase tensor ellipses.Ellipses of phase tensors at 0.1 Hz are superimposed on a TMI image. Geologicalstrike of approximately northwest is visible. Phase tensors are all quite similar, withthe principal axis aligned approximately northeast.

2D INVERSION

Occam 2D Model with Seismic Interpretation

The data points for all stations of the SD01’ transect were inverted using the smooth

2D Occam inversion code (Constable et al. 1987; de Groot-Hedlin & Constable 1990).

A smooth modelling technique was used to reduce overinterpretation, reflecting the

actual resolving power of the MT method. Error floors are set to 10% and 20% for

the apparent resistivity and phase, respectively (Rodi & Mackie 2001; Siripunvara-

porn 2012).

The Occam 2D smooth inversion in Figure 9 shows crustal heterogeneity over

small distances, with four orders of magnitude variation in resistivity. The most

striking feature is C1, the large (approximately 30 km wide) crustal structure at a

depth of 10-40 km, of very low resistivity (<10 Ωm) immediately adjacent to the

high resistivity (>10 000 Ωm) region (R1) to the west. Two low resistivity vertical

features, C2 and C3, extend up to the surface and appear to be approximately

5 km in width. A combination of the smooth modelling technique and the diffusive


Figure 9: A smooth Occam 2D inversion of broadband (0.0064 s-81.92 s) MT datacollected in 2009 and 2012. The data has been rotated 45 east of North to alignthe TE mode with the geoelectric strike. The background resistivity was set to100 Ωm. The root-mean squared (RMS) value is 4.9 with a roughness of 229. Redindicates highly conductive (i.e. low resistivity) regions and blue, resistive. Theseismic interpretation from the 2009 Southern Delamerian AuScope line has beensuperimposed on the MT, with faults of significance labelled with an F and theMoho with an M. Regions of low or high resistivity anomalies have been labelledwith a C or an R respectively for reference purposes. The location of the GlenelgZone (GZ) and the Grampian-Stavely Zone (GSZ) have also been marked. Twovertical faults (interpreted from MT results) have been added to the interpretation.Note that the western-most of the vertical faults separates the GZ from the GSZ.The star indicates a region where the mantle was particularly close to the surfacein the past, during a period of extension.

nature of MT means that resolution at depth is poor and for this reason these

low resistivity zones are very likely to have a smaller width and be much more

concentrated along fractures. The mantle beneath the Delamerian Orogen has little

variation in resistivity, with an average value of approximately 1000 Ωm.


Sensitivity analysis

To investigate whether conductive anomalies were real or perhaps artefacts of the

inversion, the starting background resistivity was set to 10, 100, 1000 and 10 000

Ωm in consecutive inversions. Each inversion imaged almost identical features and

the upper mantle resistivity was invariant, indicating confidence in the ability of MT

to resolve resistivities at mantle depths and the robustness of the resistivity struc-

ture. An inversion using only the TM mode (rather than the common simultaneous

inversion of the TE and TM modes) resulted in the same resistivity anomalies being

imaged. These sensitivity analysis tests indicate that the features of the model in

Figure 9 are very robust and are unlikely to be artefacts of the inversion.

Occam 2D Response Curves

Figure 10 shows the apparent resistivities and phases for various stations along the

transect, and how the inversion result fit the data. The target root-mean squared

value (RMS) was set to 1 for each inversion. The closer the final RMS is to this

target, the better the model fit the data. After the lowest RMS was reached (in

this case 4.77), the inversion was repeated, this time with the target RMS set just

above this lowest RMS (4.9 for this inversion). This caused Occam to generate

a much smoother (i.e. minimal structure) and hence more geologically plausible

model (de Groot-Hedlin & Constable 1990), with the roughness decreasing from 565

to 229. Note that in each station there is a general increase in apparent resistivity

with period, implying that resistivity is increasing with depth. An exception to this

is at a period of approximately 10 s where there is a dip in resistivity. It is likely this

dip in resistivity is a result in loss of signal strength associated with the dead-band,

but it is possible that this could be a real feature.


Figure 10: Plots showing the phases and apparent resistivities for periods rangingfrom 0.0064 s to 81.92 s for stations SD01, DB24C, DB30C and SD38 along the 2DMT transect. Values for periods greater than 81.92 s were removed due to calibrationissues. The TE mode is in blue and the TM mode is in red. The green and purplelines indicate the data fit of the Occam 2D inversion results (see Figure 9, for the TEand TM modes respectively. The RMS values are listed in the title for each station.The percentage errors floors allowed for the resistivities were 5%, and phases 10%.

DISCUSSION

3D Forward Model

To account for the change in resistivity and velocity (Rawlinson & Fishwick 2011) be-

tween the Delamerian and Lachlan Orogens, changes in one or more of the following


must occur: temperature, composition or hydration. Large contrasts in temperature

occur between the orogens, due to varying lithospheric thickness (Rawlinson & Fish-

wick 2011). A model of dry olivine electrical resistivity as a function of temperature

indicates that for a change in electrical resistivity from 103 to 102 Ωm (occurring

at depths of 80-110 km in the model), a temperature increase of approximately

200Cwould be required (Constable 2006).

2D Inversion

COMPARISON OF MT WITH SEISMIC INTERPRETATION

Fluid pathways often form along faults and can be associated with low resistivity

regions, either due to fluids present, or because of alteration in mineralogy due

to fluid interaction in the past. A major northwest-southeast trending fault (F2)

has been highlighted by a white dashed line in Figure 9. This fault is a possible

pathway for fluids from the mantle, which would provide an explanation of the

highly anomalous, low resistivity C1 region. Although the fault continues past the

vertical black dashed line, the subsurface changes drastically from very conductive

to very resistive. A possible explanation for this is the timing of geological events-

either the fluid averted to the northeast-southwest fault at this point, or the fault

extended beyond this point after the fluids had ceased their upward movement.

POSSIBLE CAUSES OF MID-LOWER CRUSTAL LOW RESISTIVITIES

A variety of mechanisms could be responsible for low resistivities observed in the

mid-lower crust, as investigated by Haak & Hutton (1986), Jones (1987) and Jones

(1992). The presence of carbon causes low resistivity either as: a product of CO2

mantle degassing of carbonate phases, mobilised during low degrees of partial melt-

ing in extensional periods, biogenic in the form of graphite (downthrust during

collision) (Glover 1996) or slab devolatilisation (Luque et al. 1998). Graphitisation

is promoted by shear stress and therefore its presence may be indicative of shear


deformation (Ross & Bustin 1990). Free fluids is another common cause of low re-

sistivity (Wei 2001), but the availability of fluids is very dependent on the region.

For instance, a young and active tectonic region might have a ready supply of flu-

ids, however in stable crust the presence of fluids is less likely and other causes of

low resistivity may be favoured. The presence of minerals other than carbon in the

Earth, that can cause observable low resistivities are magnetite, haematite, pyrite,

pyrrhotite and other sulphides. In certain settings, sulphides are of particular im-

portance and are in sufficient abundance and distribution to effect the electrical

resistivities significantly (Chouteau et al. 1997; Livelybrooks 1996).

GEOLOGICAL IMPLICATIONS

The star on Figure 9 indicates where the upper mantle was close to the surface due

to rifting before the Delamerian Orogeny compressed the region (Foden et al. 2006).

It is possible that during this period of extension, when the crust was at its thinnest,

fluids were able to travel up through the Proterozoic crust. However, the low re-

sistivity region extends up into the oceanic material overlying the crust, which was

thrust up after the extensional period, complicating the matter. An alternative

explanation is that the fluids originated from the compression of the Delamerian

Orogeny.

The MT transect is intersected by the Yarramyljup Fault at station SD33, the

west-most of the two vertical faults. This fault is thought to be a late Delamerian

structure (Morand et al. 2003). The dip of the fault is unknown, but the dominant

foliation in the rocks either side of it is close to vertical, suggesting vertical dip for the

fault (e.g. Gibson 1992). This fault forms the boundary between the Glenelg Zone

to the east, and the Grampians-Stavely Zone to the west (Cayley & Taylor 1998).

The conductive region, C2, coincides with the location of the fault. This further


suggests the cause of low resistivity to be fluids. The fault has been interpreted as

vertical on Figure 9, however it could be slightly dipping to the east.

Crustal and Mantle Heterogeneity

Dennis et al. (2012) MT survey in the southeast Australia region also show a very

heterogeneous crust, with low resistivity regions occuring at depths of about 10-

70 km, extending slightly deeper than low resistivity regions in this study (see Fig-

ure 9). Crustal features as interpreted from the deep seismic section commonly are

expressed in the form of low resistivity (eg the Yarramyljup Fault), but the sub-

Moho lithosphere has no resistivity variations in the location of this major fault.

Features that are evident in the crust are either confined to the crust, or do not

have a significant enough signature in the upper mantle to resolve using MT. An

exception to this is a possible continuation of the Moyston Fault, expressed at depth

as a change in resistivity and velocity.


CONCLUSIONS

2D inversions resulted in a heterogeneous crust with lateral variations of the order

of 104 Ωm over tens of km’s. The lithospheric mantle changes over a much broader

scale, order of magnitude decrease in resistivity from the Delamerian to the Lachlan

Orogen than no variation at all, as indicated by the fit of induction arrows.

A 0.1 km/s variation in velocity across the apparent boundary between the De-

lamerian and Lachlan Orogens implies a temperature and/or composition change.

A resistivity decrease from the Delamerian to the Lachlan Orogen is a result of in-

creased temperature (∼200C) from west to east and a change in composition from

continental to oceanic mantle. A change to a more hydrated mantle may also occur.

Despite the distinct crustal heterogeneity, the sub-Moho lithosphere contains al-

most no expression of the events that shaped the crust with regards to resistivity.

The electrical resistivity and velocity change between the Delamerian and Lachlan

Orogen appears to be a deeper expression of the crustal structure, the Moyston

Fault, however resolution is not high enough to be sure if this is the case.

Anomalously low resistivity regions in the crust align well with fault structures

from seismic interpretations interpreted as fluid pathways. There are low resistivities

(< 10 Ωm) at a depth of 10-40 km, with smaller paths of lower resistivity leading to

the surface. It is possible that fluid movement from the mantle may have produced

a carbon rich lower crust and very low resistivities, leaving the upper mantle in this

region to be depleted of volatiles.


ACKNOWLEDGMENTS

I would like to thank David Taylor for his assistance in the field and for helpful

discussions regarding the geology of Victoria; Jared Peacock for his assistance in the

field, with data processing, and explaining MT; Sebastian Schnaidt for assistance

in the field and reviewing my work; Goran Boren for preparing the field equipment;

Dr Lars Krieger for technical assistance; my peer-reviewers Simon Carter and Paige

Honour; the geophysics honours students, particularly Millicent Crowe, discussions

as we learnt the MT method together; Sahereh Aivazpourporgou for providing her

MT data for my usage and AuScope for the use of their equipment. I would also

like to thank my supervisor Professor Graham Heinson for his wonderful guidance

throughout the year and sharing his extensive knowledge of the theory of MT, and

also my secondary supervisor, Dr Stephan Thiel, for always being willing to help

with problems and providing a different perspective on my project.

REFERENCESAivazpourporgou, S., Thiel, S., Hayman, P., Moresi, L. & Heinson, G. (2012). The upper mantle

thermal structure of the newer volcanic province, western victoria, australia from a long periodmagnetotelluric (mt) array, 21st Electromagnetic Induction Workshop, Darwin, Australia.

Baille, P. (1985). A palaeozoic suture in eastern gondwanaland, Tectonics 4: 653–660.

Bedrosian, P. (2007). Mt+, integrating magnetotellurics to determine earth structure, physicalstate, and processes, Surveys in Geophysics 28: 121–167.

Bibby, H. M., Caldwell, T. G. & Brown, C. (2005). Determinable and non-determinable parametersof galvanic distortion in magnetotellurics, Geophysical Journal International 163(3): 915–930.

Brewer, J. A. & Oliver, J. E. (1980). Seismic reflection studies of deep crustal structure, AnnualReview of Earth and Planetary Sciences 8: 205–230.

Cagniard, L. (1953). Basic theory of the magnetotelluric method of geophysical prospecting,Geophysics 18: 605–635.

Caldwell, T. G., Bibby, H. M. & Brown, C. (2004). The magnetotelluric phase tensor, GeophysicalJournal International 158(2): 457–469.

Cammarano, F., Goes, S., Vacher, P. & Giardini, D. (2003). Inferring upper-mantle temperaturesfrom seismic velocities, Physics of the Earth and Planetary Interiors 138: 197 – 222.

Cayley, R. A. & Taylor, D. H. (1997). Grampians special map area geological report.


Cayley, R. A., Taylor, D. H., VandenBerg, A. H. M. & Moore, D. H. (2002). Proterozoic - earlypalaeozoic rocks and the tyennan orogeny in central victoria: the selwyn block and its tectonicimplications, Australian Journal of Earth Sciences 49(2): 225–254.

Cayley, R. & Taylor, D. (1998). The lachlan margin victoria: the moyston fault, a newly recognizedterrane boundary, Geological Society of Australia Abstracts 49: 73.

Chave, A. D. & Thomson, D. J. (2004). Bounded influence magnetotelluric response functionestimation, Geophysical Journal International 157(3): 988–1006.

Chave, AD, J. A. (2012). The Magnetotelluric Method: Theory and Practice, Cambridge UniversityPress.

Chouteau, M., Zhang, P., Dion, D. J., Giroux, B., Morin, R. & Krivochieva, S. (1997). Delineat-ing mineralization and imaging the regional structure with magnetotellurics in the region ofchibougamau (canada), Geophysics 62(3): 730–748.

Constable, S. (2006). Seo3: A new model of olivine electrical conductivity, Geophys. J. Int.166: 435437.

Constable, S. C., Parker, R. L. & Constable, C. G. (1987). Occam’s inversion; a practical algorithmfor generating smooth models from electromagnetic sounding data, Geophysics 52(3): 289–300.

de Groot-Hedlin, C. D. & Constable, S. C. (1990). Occam’s inversion to generate smooth, two-dimensional models from magnetotelluric data, Geophysics 55(12): 1613–1624.

Dennis, Z. R., Thiel, S. & Cull, J. P. (2012). Lower crust and upper mantle electrical anisotropyin southeastern australia, Explor. Geophys. .

Faul, U. H. & Jackson, I. (2005). The seismological signature of temperature and grain sizevariations in the upper mantle [rapid communication], Earth and Planetary Science Letters234: 119–134.

Foden, J., Elburg, M. b., Dougherty-Page, J. & Burtt, A. c. (2006). The timing and durationof the delamerian orogeny: Correlation with the ross orogen and implications for gondwanaassembly, Journal of Geology 114(2): 189–210. cited By (since 1996) 79.

Foster, D.A. & Gray, D. (2000). Evolution and structure of the lachlan fold belt of eastern australia,Annual Review of Earth and Planetary Sciences 28: 47 80.

Gaillard, F., Malki, M., Iacono-Marziano, G., Pichavant, M. & Scaillet, B. (2008). Carbonatitemelts and electrical conductivity in the asthenosphere, Science 322(5906): 1363–1365.

Gamble, T. D., Goubau, W. M. & Clarke, J. (1979). Error analysis for remote reference magne-totellurics, Geophysics 44(5): 959–968.

Gibson, G. M. (1992). Yarramyljup fault zone: eastern boundary of glenelg river complex andpossible crustal suture in western victoria, Geol. Soc. Aust. Abstracts 32: 232–233.

Glen, R. (1992). Thrust, extensional and strike-slip tectonics in an evolving palaeozoic orogen, astructural synthesis of the lachlan orogen of southeastern australia, Tectonophysics 214: 341–380.

Glen, R. A. (2005). The tasmanides of eastern australia, Geological Society, London, SpecialPublications 246(1): 23–96.

Glover, P. (1996). Graphite and electrical conductivity in the lower continental crust; a review,Physics and Chemistry of the Earth 21: 279–287.

Graeber, F. M., Houseman, G. A. & Greenhalgh, S. A. (2002). Regional teleseismic tomography ofthe western lachlan orogen and the newer volcanic province, southeast australia, GeophysicalJournal International 149(2): 249–266.


Griffin, W. L., O’Reilly, S. Y., Ryan, C. G., Gaul, O. & Ionov, D. A. (1998). Secular variation in thecomposition of subcontinental lithospheric mantle: Geophysical and geodynamic implications,Geodyn. Ser., Vol. 26, AGU, Washington, DC, pp. 1–26.

Haak, V. & Hutton, R. (1986). Electrical resistivity in continental lower crust, Geological Society,London, Special Publications 24(1): 35–49.

Hamilton, M. P., Jones, A. G., Evans, R. L., Evans, S., Fourie, C., Garcia, X., Mountford, A. &Spratt, J. E. (2006). Electrical anisotropy of south african lithosphere compared with seismicanisotropy from shear-wave splitting analyses, Physics of the Earth and Planetary Interiors158(24): 226 – 239. ¡ce:title¿Continental Anisotropy¡/ce:title¿.

Heinson, G. & White, A. (2005). Electrical resistivity of the northern australian litho-sphere: Crustal anisotropy or mantle heterogeneity?, Earth and Planetary Science Letters232(12): 157 – 170.

Jenkins, C. & Keene, J. (1992). Submarine slope failures of the southeast australian continentalslope: a thinly sedimented margin, Deep Sea Research Part A. Oceanographic Research Papers39(2): 121 – 136.

Jones, A. (1988). Static shift of magnetotelluric data and its removal in a sedimentary basinenvironment, Geophysics 53(7): 967–978.

Jones, A. (1992). Electrical conductivity of the continental lower crust, in Fountain, D.M., et al.,eds., Continental lower crust: New York, Elsevier p. 81143.

Jones, A. G. (1987). Mt and reflection: an essential combination, Geophysical Journal of the RoyalAstronomical Society 89(1): 7–18.

Jones, A. G. (1999). Imaging the continental upper mantle using electromagnetic methods, Lithos48(14): 57 – 80.

Karato, S.-I. (1990). The role of hydrogen in the electrical conductivity of the upper mantle, Nature347: 272273.

Kemp, A., Whitehouse, M., Hawkesworth, C. & Alarcon, M. (2005). A zircon u-pb study ofmetaluminous (i-type) granites of the lachlan fold belt, southeastern australia: implicationsfor the high/low temperature classification and magma differentiation processes, Contributionsto Mineralogy and Petrology 150(2): 230–249.

Korsch, R. J., Barton, T. J., Gray, D. R., Owen, A. J. & Foster, D. A. (2002). Geological inter-pretation of a deep seismic-reflection transect across the boundary between the delamerianand lachlan orogens, in the vicinity of the grampians, western victoria, Australian Journal ofEarth Sciences 49(6): 1057–1075.

Lilley, F. E. M., Woods, D. V. & Sloane, M. N. (1981). Electrical conductivity from Australianmagnetometer arrays using spatial gradient data, Physics of the Earth and Planetary Interiors25: 202–209.

Livelybrooks, D. (1996). Magnetotelluric delineation of the Trillabelle massive sulfide body inSudbury, Ontario, Geophysics 61: 971.

Luque, F. J., Pasteris, J. D., Wopenka, B., Rodas, M. & Barrenechea, J. F. (1998). Natural fluid-deposited graphite; mineralogical characteristics and mechanisms of formation, AmericanJournal of Science 298(6): 471–498.

Miller, J., Phillips, D., Wilson, C. & Dugdale, L. (2005). Evolution of a reworked orogenic zone: theboundary between the delamerian and lachlan fold belts. southeastern australia, AustralianJournal of Earth Sciences 52: 921–940.

Morand, V., Wohlt, K., Cayley, R., Taylor, D., Kemp, A., Simons, B. & Magart, A. (2003).Glenelg special map area - geological report, Technical Report Report 123, Geological Surveyof Victoria.


O’Reilly, S. Y. & Griffin, W. (1985). A xenolith-derived geotherm for southeastern australia andits geophysical implications, Tectonophysics 111: 41 – 63.

Parkinson, W. D. (1959). Direction of rapid geomagnetic fluctuations, Geophys. J. R. Astron. Soc2: 1–14.

Petkovic, P. (2004). Time-depth functions for the otway basin, Geoscience Australia Record2004/02 2.

Price, R., Gray, C. & Frey, F. (1997). Strontium isotopic and trace element heterogeneity in theplains basalts of the newer volcanic province, victoria, australia, Geochimica et CosmochimicaActa 61(1): 171 – 192.

Rawlinson, N. & Fishwick, S. (2011). Seismic structure of the southeast australian lithospherefrom surface and body wave tomography, Tectonophysics (0): –.

Rawlinson, N. & Kennett, B. (2008). Teleseismic tomography of the upper mantle beneath thesouthern lachlan orogen, australia, Physics of the Earth and Planetary Interiors 167(12): 84– 97.

Rodi, W. & Mackie, R. L. (2001). Nonlinear conjugate gradients algorithm for 2d magnetotelluricinversion, Geophysics 66(1): 174–187.

Ross, J. V. & Bustin, R. (1990). The role of strain energy in creep graphitization of anthracite,Nature 343(6253): 58–60.

Schuster, A. (1889). The diurnal variation of terrestrial magnetism, Phil trans. Lond. A, 180: 467–518.

Sealing, C. (2007). Honours thesis.

Selway, K. (2012). Interpreting magnetotelluric data in tectonically stable lithosphere, 21st Elec-tromagnetic Induction Workshop, Darwin, Australia.

Simpson, F, B. K. (2005). Practical Magnetotellurics, Cambridge University Press.

Siripunvaraporn, W. (2012). Three-Dimensional Magnetotelluric Inversion: An Introductory Guidefor Developers and Users, Surveys in Geophysics 33: 5–27.

Tikhonov, A. (1950). On determining electrical characteristics of the deep layers of the earth’scrust, Doklady 73: 295–297.

Vandenberg, A. H. M. (1978). The tasman fold belt system in victoria, Tectonophysics 48: 267297.

Wei, W. (2001). Detection of widespread fluids in the tibetan crust by magnetotelluric studies,Science 292: 716–719.

Whellams, J. (1989). The Effect of Geomagnetic Induction Anomalies on Aeromagnetic Surveys,School of Earth Sciences, Flinders University of South Australia.

Yang, X. (2011). Origin of high electrical conductivity in the lower continental crust: A review,Surveys in Geophysics 32: 875–903.

Yaxley, G. M., Green, D. H. & Kamenetsky, V. (1998). Carbonatite metasomatism in the south-eastern australian lithosphere, Journal of Petrology 39(11): 1917–1930.

Yoshino, T. (2010). Laboratory Electrical Conductivity Measurement of Mantle Minerals, Surveysin Geophysics 31: 163–206.


APPENDIX A

Figure 11: Spreadsheet showing information regarding the date of deployment andlocation of MT stations. The length of electric dipoles at deployment is listed in thecolumns, length of Ex, and length of Ey. The UTM zone for all stations was 54H.

Station Information

Station Date Longitude Latitude Elevation Length Ex Length Ey

DB24A 15/06/2012 141.32399 -36.87218 156 46 42

DB24B 15/06/2012 141.34357 -36.87256 165 47 48

DB24C 15/06/2012 141.35291 -36.87309 165 45 44

DB25A 15/06/2012 141.39014 -36.87864 169 49 50

DB26A 15/06/2012 141.41534 -36.88198 196 43 45

DB26B 15/06/2012 141.43466 -36.89242 194 45 46

DB26C 15/06/2012 141.44086 -36.87981 173 45 46

DB27A 16/06/2012 141.47414 -36.88734 169 46 46

DB27B 16/06/2012 141.48697 -36.88790 160 46 45

DB28A 16/06/2012 141.51950 -36.89550 174 46 50

DB28B 16/06/2012 141.53470 -36.89159 177 50 46

DB28C 16/06/2012 141.54878 -36.88997 198 45 44

DB29A 16/06/2012 141.57323 -36.89105 195 46 45

DB29B 16/06/2012 141.58533 -36.88955 176 46 46

DB29C 17/06/2012 141.59918 -36.88540 175 46 46

DB30A 17/06/2012 141.62903 -36.88461 184 46 45

DB30B 17/06/2012 141.64012 -36.88411 182 45 45

DB30C 18/06/2012 141.65433 -36.88347 182 44 45

DB31A 17/06/2012 141.68387 -36.88064 188 45 42

DB31C 17/06/2012 141.71442 -36.88051 177 49 47

DB32A 17/06/2012 141.73845 -36.87222 188 48 44

DB32B 17/06/2012 141.74998 -36.86918 194 47 35

DB32C 18/06/2012 141.76314 -36.86594 187 47 48

DB33B 18/06/2012 141.79994 -36.87624 186 43 50

DB33C 18/06/2012 141.82000 -36.88000 165 46 46

DB34A 18/06/2012 141.83642 -36.88514 181 47 45

DB34B 18/06/2012 141.85664 -36.88742 162 48 48

DB35A 19/06/2012 141.88564 -36.89779 160 48 50

DB35B 19/06/2012 141.90083 -36.89919 173 48 43.5

DB35C 19/06/2012 141.91573 -36.90064 165 47 47

DB36A 19/06/2012 141.94290 -36.90334 172 47 44.5

DB36B 19/06/2012 141.95755 -36.90483 166 45 46

DB36C 19/06/2012 141.97548 -36.90582 164 48 45

DB37II 19/06/2012 141.99450 -36.90700 159 46 45

DB37A 19/06/2012 142.00635 -36.90890 160 48 45

DB37B 20/06/2012 142.02155 -36.91085 162 45 44

DB38A 20/06/2012 142.05013 -36.90384 168 46 45

DB38B 20/06/2012 142.06378 -36.89836 171 46 48

DB38C 20/06/2012 142.07709 -36.88776 158 45 47


APPENDIX B

Figure 12: Spreadsheet listing parameters for data processing, for 1000 Hz dataand 100 Hz decimated data. The start and stop times indicate the window of datathat were processed. The columns RR indicate the station that were used for remotereferencing purposes.

Stat

ion

Freq

(H

z)Ex

, Ey

and

to

tal r

ota

tio

nD

aySt

art

tim

e (

hh

mm

ss)

Sto

p T

ime

(hh

mm

ss)

RR

Day

Star

t ti

me

St

op

tim

e R

R

DB

24

A10

000

,90,

-9.8

5916

711

0000

1300

00D

B2

4A16

700

0300

2158

00D

B2

4A

DB

24B

1000

0,9

0,17

0.14

116

708

0000

1000

00D

B2

6A16

702

1000

2240

00D

B2

4C

DB

24C

1000

0,9

0,-9

.859

167

1200

0014

0000

DB

26A

167

0210

0022

4000

DB

24C

DB

25

A10

000

,90,

170.

141

167

0800

0010

0000

DB

24B

167

0800

0023

0000

DB

26C

DB

26

A10

000

,90,

170.

141

167

1100

0013

0000

DB

24B

167

0555

0023

5000

DB

26C

DB

26B

1000

0,9

0,17

1.14

116

708

0000

1000

00D

B2

4B16

706

5000

2355

00D

B2

6A

DB

26C

1000

0,9

0,17

0.14

116

708

0000

1000

00D

B2

4B16

7,16

807

4000

,000

000

2400

00,0

1300

0D

B2

6B

DB

27

A10

0018

0,9

0,17

0.1

4116

808

0000

1000

00D

B2

9A16

8,16

905

4000

,000

000

2400

00,0

3550

0D

B2

7B

DB

27B

1000

180

,270

,17

0.14

168

1200

0014

0000

DB

28B

168

0200

0012

0000

DB

27B

DB

28

A10

000

,90,

170.

141

168

1100

0013

0000

DB

29A

168

0235

0022

1000

DB

28B

DB

28B

1000

0,9

0,17

0.14

116

811

0000

1300

00D

B2

8B16

802

3500

2200

00D

B2

8A

DB

28C

1000

0, 2

70,

-9.

859

168

1100

0013

0000

DB

28A

168

0800

0018

0000

DB

29B

DB

29

A10

000

,90,

170.

141

168

1100

0013

0000

DB

29A

168

0410

0023

3000

DB

27B

DB

29B

1000

0,27

0,17

0.14

116

808

0000

1000

00D

B2

8A16

808

0000

1800

00D

B2

8C

DB

29C

1000

0,9

0,17

0.14

116

908

0000

1000

00D

B2

9C16

901

2500

2335

00D

B3

1A

DB

30

A10

0018

0,9

0,17

0.1

4116

911

0000

1300

00D

B3

2A16

9,17

006

2500

,000

000

2400

00,0

1200

0D

B3

2B

DB

30B

1000

0,9

0,17

0.14

116

908

0000

1000

00D

B3

2A16

907

0000

1700

00D

B3

2B

DB

30C

1000

0,27

0,17

0.14

117

008

0000

1000

00D

B3

3C17

002

4500

2220

00D

B3

2C

DB

31

A40

180

,90,

170

.141

169

1000

0012

0000

DB

30A

169

0510

0021

0000

DB

31C

DB

31C

1000

0,9

0,17

0.14

116

908

0000

1000

00D

B3

1C16

910

0000

2000

00D

B3

1C

DB

32

A10

000

,90,

170.

141

169

2200

0024

0000

DB

30B

169,

170

0350

00,0

0000

024

0000

,012

000

DB

30A

DB

32B

1000

180

,270

,17

0.14

169

0800

0010

0000

DB

30B

169

0400

0014

0000

DB

30A

DB

32C

1000

0,9

0,17

0.14

117

008

0000

1000

00D

B3

3C17

003

0000

1300

00D

B3

0C

DB

33B

1000

180

,90,

170

.141

170

1100

0013

0000

DB

33C

170

0630

00,0

0000

024

0000

,223

500

DB

33B

DB

33C

1000

180

,90,

170

.141

170

0800

0010

0000

DB

32C

170

0515

0022

2000

DB

30C

DB

34

A10

000

,270

,170

.14

170

0800

0010

0000

DB

32C

170

0515

0022

2000

DB

30C

DB

34B

400

,90,

170.

141

170

0800

0010

0000

DB

34B

170

0600

0021

5500

DB

34B

DB

35

A10

000

,90,

170.

141

171

0800

0010

0000

DB

37I

IA17

1,17

224

5000

,000

000

2400

00,2

2000

0D

B3

6B

DB

35B

1000

180

,270

,17

0.14

171

0800

0010

0000

DB

37I

IA17

1,17

241

5000

,000

000

2400

00,2

3000

0D

B3

6C

DB

35C

1000

0,9

0,17

0.14

117

108

0000

1000

00D

B3

7IIA

171,

172

4500

00,0

0000

024

0000

,220

500

DB

35A

DB

36

A10

000

,90,

170.

141

171

0800

0010

0000

DB

36A

171,

172

2450

00,0

0000

024

0000

,220

500

DB

35A

DB

36B

1000

0,9

0,17

0.14

117

108

0000

1000

00D

B3

7IIA

171,

172

4150

00,0

0000

024

0000

,220

500

DB

35A

DB

36C

1000

0,9

0,17

0.14

117

108

0000

1000

00D

B3

7IIA

171,

172

5550

00,0

0000

024

0000

,220

500

DB

35A

DB

37I

I10

0018

0,9

0,17

0.1

4117

108

0000

1000

00D

B3

5B17

1,17

270

0000

,000

000

2400

00,2

3300

0D

B3

6A

DB

37I

IA10

000

,90,

170.

141

171

0800

0010

0000

DB

35B

172

0200

0012

0000

DB

35A

DB

37I

IB10

000,

270,

170.

141

172

0800

0010

0000

DB

38B

172

0200

0012

0000

DB

35A

DB

38

A10

000,

270,

170.

141

172

0800

0010

0000

DB

38B

172

0130

0023

5000

DB

38C

DB

38B

1000

180

,270

,17

0.14

172

0800

0010

0000

DB

38A

172

0130

0023

5000

DB

38B

DB

38C

1000

180,

270

, -9

.859

172

0800

0010

0000

DB

38B

172

0130

0023

5000

DB

37I

IB

Dec

imat

ed D

ata-

10

Hz

Un

dec

imat

ed D

ata-

100

0 H

z

Pro

cess

ing

Info

rmat

ion

an electrical resistivity model of the southeast

Documents