Upload
dangtuyen
View
218
Download
0
Embed Size (px)
Citation preview
Geological Survey of Ethiopia
Hydrogeology, Engineering geology and Geothermal Department
The Resistivity Structure of the Abaya Geothermal field,
Southern Main Ethiopian Rift, Revealed by Magneto-telluric
Imaging.
Mohammednur Desissa and Yohanes Lema
December, 2006
Addis Ababa
ii
Abstract To understand the crustal electric structure of the geothermal field around the Lake Abaya,
15 MT soundings have been carried out. Site spacing varies between 2.5 and 8 km based on
access condition. Measurements were taken along four lines. The 1D inversion of the data
confirms the presence of low resistivity zones related to structurally weak zone,
hydrothermal alteration, different clayey units; extremely to highly altered ignimbrites or to
magmatic brine (saline water). Very high resistivity zones observed below the surface are
associated to volcanic intrusion while those continue from the surface are due to volcanic
rocks. The foot of volcanic centers (magmatic centers) observed in the surface are not
deeper than 2km. The dominant controlling structure is associated with NE-SW oriented
structure which could be activated by Wonji fault.
iii
Abstract ..................................................................................................................................ii Table of contents....................................................................................................................iii List of figures.........................................................................................................................iv 1. INTRODUCTION ............................................................................................................iv
1.1. Back ground and location ..........................................................................................1 1.2. Basic concepts .............................................................................................................1
2. Review of previous work ...................................................................................................3 3. Geological and tectonic settings .......................................................................................4 4. Objectives of the MT survey.............................................................................................6 5. Methodology and data acquisition ...................................................................................7 6. Data processing and presentation ....................................................................................8 7. Modeling and interpretation ...........................................................................................9 8. Conclusions and Recommendations ..............................................................................16 Acknowledgement................................................................................................................18 References ............................................................................................................................19
iv
List of Figures
Figure1. Location ma of survey area....................................................................2
Figure2. Geological map of the northern Lake Abaya area...............................5
Figure3. Distribution of MT survey lines in relation to geological map...........7
Figure4. Maximum Electric Direction (MED), strike of 2D structure.............9
Figure5. Examples of fitted model...................................................................10
Figure6. Geoelectric section along line 1........................................................11
Figure7. Geolelectric section along line 2.......................................................13
Figure8. Geoelectric section along line 3.......................................................14
Figure9. Geoelectric section along line 4.......................................................15
1
1. INTRODUCTION 1.1. Back ground and location Prominent geothermal systems of Africa are associated with Great Rift Valley, which is one
of the major tectonic structures in the world. Geothermal energy is renewable, indigenous
and environmentally friendly. It is a base load energy which is independent of weather
conditions. Since Ethiopia is affected by prolonged draught, the reliability of hydropower
during dry season is a basic power problem related hydro. Therefore, successful exploration
and exploitation of geothermal energy is of paramount importance.
The Ethiopian Rift (ER), which is part of the East African Rift System (EARS), covers
about 150,000 sq.km (i.e, nearly 15% of the country). In the Ethiopian rift, geothermal
explorations have been carried out since 1969 involving national and international survey
organizations. Previous studies revealed more than 16 geothermal potential areas with an
estimated energy potential of about 1000MWe.
The present study area is located at the southern part of the Main Ethiopian Rift (MER),
bounded between 370 53’-38015’ longitudes and 60 35’- 70 00’ latitudes (about 500km
south of Addis Ababa). The field study was conducted from 4th to 25th April 2006.
1.2. Basic concepts
The MT method is a way of determining the electrical conductivity distribution of
subsurface from measurements of natural transient electric and magnetic fields on the
surface. Results interpreted from measurements at a single site are sometimes compared to
an induction log. Results from a line of measuring stations are interpreted to give underlying
conductivity distribution and structure. That picture of subsurface could in turn be related to
porosity and salinity, temperature and hot magma, etc.
The time variation of earth's electric and magnetic fields at a site are recorded
simultaneously over a wide range of frequencies. The variations are analyzed to obtain their
spectra and apparent resistivities as a function of frequency are computed from spectra. The
MT method depends on the penetration of electromagnetic energy in to the earth. Depth
2
control comes as a natural consequence of the greater penetration of lower frequencies. The
measurements are absolute. Their interpretation gives true resistivity values and depth, not
just anomalies. Depth interpretation based on MT data is therefore much more definitive
than that based on gravity or magnetic data.
b) Within geothermal prospect area
Figure1. Location ma of survey area
3
Table1. Coordinates of MT sites.
2. Review of previous work Previously geological, geochemical and infrared surveys were carried out in the area and
revealed occurrence of several hot springs and fumaroles while geothermal gas studies have
indicated that the temperature of springs' water is around 1770 C. It was concluded that
distribution of hydrothermal active zones is strongly controlled by tectonics and geological
structures. Most of fumaroles and hot springs are related to volcanic complexes.
Geophysical studies including gravity, magnetics and DC resistivity surveys were
conducted in the northern part of the present MT survey. Strong thermal features were seen
to associate with low values of gravity, magnetic and DC resistivity. Moreover, structures
trending in NNE-SSW, N-S and NW-SE were delineated by geophysical methods.
Site No. Easting Northing elevation Remark 004 375389 732990 1257
001 379425 733560 1190
002 382923 729973 1182
003 385682 728580 1207
Line 1
005 376782 742857 1440
007 379494 742768 1438
006 386480 741402 1294
Line 2
015 387597 759153 1508
008 392470 757836 1385
009 399829 754779 1360
Line 3
014 401443 766621 1588
011 404692 766418 1501
012 408611 767351 1638
013 414129 768146 1802
Line 4
4
3. Geological and tectonic settings The collision between east and west Gondwanaland has defined the major geological and
structural features of East Africa. The late Protrozoic Mozambique belt, representing
Himalayan-scale orogeny, abuts the Archaen complexes of western Ethiopia. In Ethiopia,
the belt is transected by numerous predominantly N-S to NNE-SSW suture zones and
associated ophiolitic slivers and NW-SE trending shear zones.
In the area, pre-rift basaltic rocks unconformably overlie the Precambrian basement. The
prominent units in the area are:
• Volcanic rocks: these include the pre-rift basaltic rocks, ignimbrites, trachytes,
pumiceous pyroclastics, rhyolites, and rhyo-obsidian flows.
• Sedimentary succession is represented by Pleistocene volcano-sedimentary rocks,
Holocene lake sediments and recent alluvium and fluvial deposits • Structural observation was revealed presence of series of intense normal faulting
and quaternary volcanism in the area. Duguna-Fango volcanic complex is believed
to represent the southern sector of Wonji Fault (WFB), which is the active
spreading axis of extension in the Main Ethiopian Rift (MER).
• The centers of this belt are considered to be the heat sources for wide spread
hydrothermal activity fed by deep-seated magma chambers.
6
4. Objectives of the MT survey Previously DC resistivity survey was conducted over limited area that covers only the
northern part of the current area (i.e around lines 3 & 4). Deep subsurface information
related to thermal sources could not be acquired due to the following reasons.
• To achieve great depth penetration it requires a long electrode separation, which is
difficult under such terrain conditions.
• It needs equipment capable of generating high power.
• Depth penetration of the DC method is highly affected by the upper resistive and/or
conductive layers.
• Using potential field methods (gravity and magnetic) only qualitative interpretation
was made.
On the other hand, geothermal resources are genetically related by water and heat. Since the
electrical resistivity is affected by changes in temperature and permeability, electromagnetic
method can provide a model of subsurface resistivity variation. Of all EM methods MT
seems to be appropriate due to several kilometers of penetration and the upper resistive
volcanic rocks (as Ethiopian case) couldn’t mask the EM field propagation.
Since the measurements of MT are absolute and the interpretation gives true resistivity
values and depth, not just anomalies, depth interpretation based on MT data is therefore
much more definitive than that based gravity or magnetic data. Therefore, the MT survey
was intended to:
• Map the geo-electrical structure of the area;
• Determine lateral and vertical extent of active hydrothermal zones and possible
heat sources with respective depths
• Map prominent geologic structures associated with WFB and volcanic centers in the
area.
7
5. Methodology and data acquisition Fifteen (15) MT soundings were carried out along four profiles (fig. 3) with the Short
Period Automatic Magnetotelluric (SPAM mark III) data acquisition system (Ritter et al.,
1998). The horizontal electric and magnetic field time series were recorded in 3 and 7
period bands for high frequency and low frequency modes respectively in the magnetic
north (x) and east (y) directions. The electric potentials were measured between two non-
polarizable Cl2–PbCl2 electrodes, with separations of 50 m using cross configuration.
Magnetic field was recorded using metronix MFS05 induction coils. Four geo-electric
sections along four lines oriented nearly perpendiculars to the rift axis are presented. The
station spacing varies between 2.5 and 8 km while line spacing varies between 12 and 25km
based on access. The recordings typically lasted 4 to 10 hours. Position of each station was
obtained with handheld GPS and the one installed within the MT unit.
Figure3. Distribution of MT survey lines in relation to geological map.
8
6. Data processing and presentation The data from high frequency band and in three shortest period bands of low frequency
mode are of excellent quality, leading to precise MT impedance. The data in longer period
bands (3 to 6) did not provide such good transfer functions because of limited recording
time.
The electric and magnetic field time series were transformed into the frequency domain.
The 2 * 2 MT impedance tensor Z relating the horizontal electric (Ex, Ey) to the horizontal
magnetic (Bx, By) fields was determined using the robust remote reference method of
Chave and Thompson (1989).
In a 2-D Earth, the electromagnetic induction equations separate into two modes, the
transverse electric (TE) mode (currents parallel to structure and in the direction of constant
conductivity, the geo-electrical strike) and the transverse magnetic (TM) mode (currents
perpendicular to geo-electrical strike). Currents parallel to structure only induce magnetic
fields perpendicular to it, and vice versa for the TM mode. Thus the diagonal elements of
the impedance tensor vanish.
The MT impedance tensor at each site was rotated to the direction that best-fits an off-
diagonal tensor (xx & yy). The geo-electrical strike direction was determined by rotating the
impedance tensor into the complex direction minimizing its diagonal elements. This was
done choosing the rotation angle of the real axis of the electric or magnetic field that either
maximizes or minimizes the impedance tensor. This angle defines, for each period, a
maximum electric field direction (MED) or a minimum magnetic field direction. In 2-D
geometry the MED corresponds to the strike of the 2-D structure.
At most sites, the rotation angle was well defined and constant with period and indicates a
structural direction. This is illustrated in Figures 4 where the MED are shown for periods of
0.5 and 8 seconds respectively. At 0.5 second period, representative of approximately the
first kilometer below the surface, there are four directions, 120-170N, 90–120N, 40-80_N
and 10–40_N of which the two main directions, 90–120_N and 10–40_N are dominant. The
9
90–120_N direction corresponds to the NW–SE that well agrees with the transverse tectonic
structures while the 10–40_N direction corresponds to the NNE–SSW active WFB.
At a longer period, (8 s) or at deeper depth the NE–SW direction dominates, indicating that
deeper structures are more influenced by the active Wonji fault direction. A correlation
made between resistivity and geo-electric strike (MED) at each site revealed that low
resistivity is associated with NE-SW direction, which indicates that thermal activity is
controlled by this structure.
a) at 0.5seconds b) at 8 seconds
Number of sites
Figureigure 4. Maximum Electric Direction (MED), strike of 2D structure
7. Modeling and interpretation The 1-D inversion method used is based on an iterative technique that minimizes a misfit
function between the data and the 1-D model calculation. Modeling was considered up to a
depth of about 3km, where best fit between the data and model parameters were observed
for all sites. The geo-electrical models are presented at two different depth levels 500m and
2000m for each line. Examples of resistivity and fitted model for four sites from four lines
are shown in figure 5.
D
egre
es c
lock
wis
e
10
10-4 10-3 10-2 10-1 100 101 102 103
100
101
102
YEva
luatio
n C
10-4 10-3 10-2 10-1 100 101 102 103100
101
102
103
eXY E
valua
tion
Site 001 Site 014
10-4 10-3 10-2 10-1 100 101 102
100
101
102
103
Evalua
tion C
opy
Horizontal axis is period
Site 005
10-4 10-3 10-2 10-1 100 101 102 103
C l 1
100
101
102
103
XY Eva
luatio
n
Observed resistivity
Site 008 model resistivity
Figure5. Examples of fitted model
Ρ app
(Ωm
) Ρ a
pp (Ω
m)
Ρ app
(Ωm
)
Ρ
app (Ω
m)
11
a). 500m depth
b). 2000m depth
Figure 6. Geoelectric section along line 1
Site001 is characterized by low resistivity starting from surface and getting wider at around
1200m depth. The site was observed between fumaroles and hot springs which are favored
by structurally weak zone (low resistivity). Low resistivity could also be resulted from
hydrothermal alteration, different clayey units; or it may be related to highly altered
ignimbrites or to magmatic brine or, saline water. This very low vertical resistivity could
12
indicate the upflow zone through which hydrothermal fluids migrate to the surface. Site 002
shows relatively low resistivity (10-15Ωm) up to a depth of around 500-600m and then it
gets very low up to 1500m. Resistivity signature below 1500m is relatively high, which
could be due to volcanic intrusion. Sites 004 and 003 are characterized by very high
resistivity starting from surface up to 500m depth, which could be due to extrusive volcanic
rocks. Below this depth, resistivity under site 004 becomes relatively low while under site
003 it is very low, which might be related to extremely altered rocks or related to magmatic
brine (saline water).
Resistivity signature along line 2 shows different characteristics compared to line1. The
shallower part is mapped by relatively high resistivity except at site 005, where low
resistivity is observed up to 100 m. The conductive anomalous zone is mapped at central
part of the line below 1 km depth under sites 005 and 007, which could be related to the
deep seated heat source. It is noted that site 007 was located at a foot of mount Selwa ; one
of the volcanic centers in the area with hot caldera lake on top. Site 006 is mapped by
relatively low value between 500 and 1000m depth.
15
Line 3 is characterized by very low resistivity below 200 m and 800 m depth under sites 015
and 009 respectively while high resistivity is observed below site 008 up to nearly 2 km
depth, which could be attributed to the presence of thick extrusive volcanic rocks under site
008. Site 009 is close to Bilate River where massive ignimbrite is observed. The MT survey
result revealed that the thickness of ignimbrite is about 1km at site 009 and 2km at site 008.
The same explanation discussed above applies to very low resistivity values under sites 015
and 009.
a). 500m
b). 2000m
Figure9. Geoelectric section along line4
16
Line 4 is characterized by low resistivity zone under site 011 (similar to site 001) while
resistive in the eastern side, which may indicate the upflow zone. The high resistivity under
site 012 continues up to the depth of about 1 km and replaced by very low resistivity below
this depth. The conductive zone observed at site 011 may serve as conduit through which
geothermal heat flows to the surface. The hot spring observed lose to this site (Dimtu
village) could be a manifestation of the presence of deep seated heat source.
8. Conclusions and Recommendations Experiments made by different researchers have proved that resistivity and temperature are
inversely related ( T1
∝ρ ). Temperature parameter influences the resistivity of rocks.
During geodynamic activity, the crustal materials melt at high temperature and pressure
conditions and thereby increasing the conductivity of crustal materials. In such conditions,
high temperature values are associated with low resistivity (high conductivity) regions. In
volcanic and geothermal areas, the causes of high conductivity (low resistivity) at depth
may indicate the presence of hot magma chamber. In the present survey area, the presence
of high temperature was revealed by revious studies. There are a number of hot springs and
fumaroles smoking all the time. Moreover, direct rrespondence was seen between very low
resistivity and NNE-SSW structure, which could be due to WFB. The result of MT survey
in Abaya geothermal prospect area allowed haracterizing the electrical structure of the area.
The result shows lateral and vertical variations of electrical resistivity. It can be concluded
that:
• Surface manifestations of thermal features are strongly related to low resistivity
(high conductivity) zones
• Low resistivity zones are wide and open towards depth, which could indicate that
the area
is potentially promising for geothermal energy resource
• Low resistivity zones are caused by a combined effect of hydrothermal alteration,
saline fluids and structures
17
• The low resistivity observed t deeper depth could be associated with conductive
hydrothermal fluids injected from hot magma or highly fractured and weathered
volcanic rocks or conductive sedimentary unit.
• The foot of volcanic centers (magmatic centers) observed on the surface are no
more deeper than 2km
• The dominant controlling structure is associated with NE-SW oriented structure
which could be activated by Wonji fault .Though the current MT survey was
bserved for short period of time, the results allowed characterizing the electrical
structure of the area and determine lateral and vertical variations of electrical
resistivity. Continues observation and use of MT method will enable us to explore
and map geothermal potential areas, which could secure sustainable energy supply
and utilizing alternative energy sources (petroleum, geothermal, coal etc). Therefore,
systematically girded MT survey observing for long period of time is highly
recommended.
18
Acknowledgement We would like to thank the National Environment Research Council (NERC) geophysical
Equipment facility of the University of Edinburgh (UK), for permission to use the
MT equipment. Our thank also goes to Prof. Kathy A. Whaler and Dr. Sophie
Hautot for their help all the way through.
At last but not least, our heart felt thanks goes to H.E. Ato Alemayehu Tegenu, minister
of Mines and Energy and Dr. Meseret Teklemariam for facilitating the field work
19
References
1. Cagniard L. 1953. Basic Theory of Magneto-telluric Method of Geophysical
Prospecting: Geophysics ,18 605-635
2. Kunetz G. 1972. Processing and Interpretation of Magnetotelluric Soundings
Geophysics, 37 1005-1021
3. Vozoff K. 1972. The magnetotelluric method in the exploration of sedimentary
basins: Geophysics 37 98-141
4. Hautot S. ,P. Tarits and Kathy Whaler, 2000. Deep strcture of Baringo Bogoria Rift
Basin(central Kenya) from Three-Dimensional magnetotelluric imaging:
Implications for rift evolution. J.Geophy.R. 105 23,493-23,518
5. T. Harinarayana et al, 2006. Exploration of exothermal structure in Puga geothermal
field, Ladakh Himalayas, India by magnetotelluric studies. Journal of Applied
Geophysics, 58 280-295.
6. A. Hisafumi et al, 2006.Magnetoteluric resistivity modeling for 3D characterization
of geothermal reservoir in western side of Mt. Aso, SW Japan. J.Geophy. Res. 58
296-312
7.