magnetotelluric and transient electromagnetic ... - Orkustofnun

Report 21 411 Wameyowhere ρ = Bulk resistivity (Ωm);ρ w = Resistivity of the pore fluid (Ωm);φ t = Porosity in proportion of total volume;α = Empirical parameter for describing type of porosity varying from less than 1 forinter-granular porosity to more that 1 for joint porosity, but is usually around 1;n = Cementation factor, usually around 2.Archie’s law is sometimes written as:ρ = ρwF; F =ρ(4)ρwwhere F= Formation factor.The permeability of a rock is the ability of fluids to move within its matrix. Permeability depends onthe interconnectivity of the pore spaces within the rock matrix rather than the porosity of the rock.The amount of fluid flowing through a rock can also largely be dictated by fractures (secondaryporosity), common in geothermal areas. The wider the fracture, the higher fracture porosity, hence,high permeability is expressed by the following equation (ISL, Michigan State University, 1999):where K = Permeability (m 2 );Q = Fluid flow rate (m 3 /s );η = Fluid viscosity (kg/ms);L = Length of the rock (m);A = Cross-sectional area available for flow (m 2 ); andP = Pressure drop (Pa).QηLK = (5)APDue to the negligible electrical conduction in most minerals, the majority of charge transport in rocksand sediments occurs in the electrolytes. High mobility (due to high permeability) of the chargecarrier within a rock matrix has the effect oflowering the electrical resistivity of the rock.Geological processes such as faulting, shearing,columnar jointing and weathering usuallyincrease permeability and porosity, thereforeincreasing electrical conductivity, whereasprecipitation of calcium carbonate or silicareduces porosity and hence increases resistivity.2.2 SalinityAn increase in the amount of dissolved ions inthe pore fluid can increase the conductivitygreatly (Figure 1). Conduction in solutions islargely a function of salinity and the mobility ofthe ions present in the solution. This isexpressed in the equation (Hersir andBjörnsson,1991):FIGURE 1: Pore fluid conductivity vs. salinity(concentration) for a variety of electrolytes(modif. from Keller and Frischknecht, 1966)

Wameyo 412Report 21( c q m + c q ...)σ = Fm(6)1 1 1 2 2 2 +where σFcqm= Conductivity (S/m);= Faraday’s number (9.65×10 4 C);= Concentration of ions;= Valence of ions;= Mobility of ions.Both an increase in the water content and the total amount of dissolved ions are sometimes associatedwith geothermal activity.2.3 TemperatureAt temperatures well below the critical temperature (374°C), increased temperature enhances theconductivity in the pore fluid (given that the pressure is sufficiently high to prevent boiling). Increasein temperature reduces water viscosity, resulting in increased mobility of ions (Hersir and Björnsson,1991). This relationship is described by Dakhnov (1962) as:ρρw0w = (7)1 + α( T - T0)where ρ w0 = Resistivity (Ωm) of the fluid at temperature T 0 ;α = Temperature coefficient of resistivity, α ≈ 0.023°C -1 for T 0 = 23°C and 0.025°C -1for T 0 = 0°CConductivity enhancement, with an increase intemperature, is quite great between 20 and350°C for most electrolytes (Figure 2). Athigher temperatures, there is a decrease in thedielectric permittivity of water, resulting in adecrease in the number of dissociated ions insolution. This effectively increases fluidresistivity.2.4 PressureConfining pressure has the net effect ofincreasing the bulk resistivity of a rock bydecreasing pore volume as the rock iscompressed. The pressure effect can be dramaticin fractured rock where the fractures normal tothe principle stress close while others remainopen. This can cause a significant anisotropy ofthe rock (Morris and Becker, 2001).FIGURE 2: Resistivity of an electrolyte as afunction of temperature at different pressures(mod. from Quist and Marshall, 1968)2.5 Water-rock interaction and interface conditionsApart from the reduction in resistivity by the pore fluid, the bulk resistivity of the rock is also reducedby the presence of hydrous secondary minerals (such as clays) as a result of fluid-rock interaction.This interface conductivity (alteration) is expressed by the equation:

Report 21 413 Wameyoσ = 1 σ w + σ s(8)Fwhere σ = Bulk conductivity (S/m);σ w = Conductivity of water (S/m);σ s = Interface conductivity (S/m);F = Formation factor of the rock, given by ρ/ρ w (from Equation 3).The interface conductivity, σ s , is caused by fluid-matrix interaction and depends mostly on themagnitude of the internal surfaces (porosity) and of their nature (surface conditions). The main tworeasons for surface conductivity are the presence of clay minerals (alteration) and surface double-layerconduction.The alteration process and the resulting type of alteration minerals are dependent on the type ofprimary minerals, chemical composition of the geothermal fluid and temperature. The intensity of thealteration is furthermore dependent on the temperature, time and the texture of the host rocks.Alteration intensity is normally low for temperatures below 50-100°C. At temperatures lower than220°C, low-temperature zeolites and the clay mineral smectite are formed. Smectite has hydrated andloosely bound cations between the silica plates, making the mineral conductive and with a high cationexchange capacity (Árnason et al., 2000).In the temperature range from 220 to about 240-250°C, the low-temperature zeolites disappear and thesmectite is transformed into chlorite in a transition zone, the so-called mixed-layered clay zone, wheresmectite and chlorite coexist in a mixture. At about 250°C the smectite disappears and chlorite is thedominant mineral, marking the beginning of the chlorite zone, hence, highly resistivity, since chloriteminerals have cations that are fixed in a crystal lattice, making the mineral resistive. At still highertemperatures, about 260-270°C, epidote becomes abundant in the so-called chlorite-epidote zone andthe resistivity becomes even higher.In a typical geothermal system, one would expect to find a deep anomaly in electrical conductivityassociated with thermal excitation of conduction in the massive crystalline rock comprising thebasement. At shallower depths in the section, one would expect to find an anomaly in electricalconductivity associated with a reservoir filled with hot geothermal fluids.3. CENTRAL-LOOP TEM METHOD3.1 TheoryReceiverIn the Central-Loop TransientElectromagnetic method, a steadyReceiver coilcurrent is transmitted in a wire loop,laid on the ground at the area to beexamined (Figure 3). This currentis allowed to flow for a sufficientlylong time to allow turn-on transientsTransmitterin the ground to dissipate. Thissteady current is then abruptlyterminated in a controlled fashion(Figure 4). At the instant ofTransmitter loopFIGURE 3: Central-loop time domain TEM layouttransmitter turn-off, eddy currents reproduce the static magnetic field due to the transmitter loop butthen decay rapidly. The decaying primary magnetic field, in accordance with Faraday’s law, induceseddy currents immediately below the transmitter loop. As the initial near-surface eddy current decays,

Wameyo 414Report 21Transmitter current (and primary magnetic field)e.m.f. induced in the earth bytransmitter current turnttits distribution in theground in turn inducesa secondary magneticfield which also decayswith time. Thisprocess continues overtime with an everweakening secondarymagnetic fieldinducing currents atincreasing depth.Secondary (target) current andmagnetic fieldFIGURE 4: Transmitted and received waveforms in a typicaltime-domain system (after Ward, 1983)The magnitude and rateof decay of thesecondary magneticfield is monitored bymeasuring the voltageinduced in a receivercoil, placed at thecentre of thetransmitter loop, as afunction of time after the transmitter current is turned off. This is then interpreted in terms of asubsurface resistivity structure (Árnason, 1989).t3.2 Confined target responsesFor a simple confined conducting target in a resistive half-space, the receiver-coil output voltage isproportional at all times to the time rate-of-change of the secondary magnetic field in the form(Geonics Ltd, 1980):1 t- τe0 ∝ e(9)τwhere τLR= L/R, i.e. time-constant of the target;= Inductance of the target (H);= Target resistance (Ω).From the above expression, we see that conductive targets (i.e. those having small resistance andtherefore a large value of τ) yield signals with small amplitude but decay relatively slowly. Resistivetargets, however, have high initial amplitudes that decay rapidly.As an example take a conducting sphere (to represent a confined target, i.e. a conductor surrounded byan insulator) (Figure 5a), with a radius r and a conductivity σ in a uniform magnetic field H p which issuddenly terminated at time t = 0. Currents will immediately flow on the surface of the sphere (earlytimes) with a distribution that tries to maintain the original uniform magnetic field within the sphere.The current distribution at the early times (Figure 5b) is independent of the conductivity of the sphere.We can say that at this stage we are in the high-frequency limit since the current distribution is similarto that which would flow if the sphere was located in a very high-frequency alternating magnetic field.The electrical skin-depth, δ, of the sphere material defined by the following relationship:12⎛ 2 ⎞δ = ⎜ ⎟(10)⎝ μσω ⎠

Report 21 415 Wameyowhere μωf= 4π × 10 -7 (H/m);= 2πf;= Frequency (Hz),The high frequency in this case is such that the skin-depth is much less than the radius of the sphere.Figure 5c shows a crosssectioncutting an equatorialplane of the conductivesphere; at time t = 0 thecurrent is concentrated at thesurface layer of the sphere.However the amplitude of thecurrent commences todecrease due to ohmic lossesin the conductive spherematerial. The local magneticfield due to the current alsodecreases inducing electromagneticfield (throughFaraday’s Law) which causesnew current to flow as shownin Figure 5d. This processcontinues with the result thatthe current flow movesinward as time elapses. Thecurrents appear to bediffusing radially inwards asa result of the interaction withtheir magnetic fields in theconductive body. Thisperiod, when the currentdistribution is in motion, istermed as the ‘intermediatetimes’.During this period themagnetic field associatedwith the decreasing currentsdecays rapidly with time.ACEBDA stage is reached, however,where the relative spatialcurrent distribution becomesinvariant with time butdecreases at an equal rate,FIGURE 5: Target conductive sphere and current distributionwith the form indicated inFigure 5e. Near the sphere centre, the current density increases linearly with radial distance becomingrelatively uniformly distributed at one-half the radius and slightly decreasing towards the edge. Thisperiod is termed as the ‘late-times’. The inductance and resistance associated with each current ringhave stabilized and from this point onwards, both currents and their associated external magneticfields commence to decay exponentially with a time-constant given by (Geonics Ltd, 1980) as:2σμrτ = (11)2π

Wameyo 416Report 21For any confined conductor, the current distribution at the early times is a function of time and theoverall decay is not exponential. At the late times however, the current distribution becomes invariantwith time and the decay becomes exponential at a rate determined by the shape, size and conductivityof the body.3.3 Layered earth responseWhen dipole moment of a vertical magnetic dipole transmitter located on a homogenous half-space,shown in Figure 6, is abruptly reduced to zero at time t 0,, a short-duration voltage pulse is induced inthe ground which causes a loop of current to flow in the immediate vicinity of the transmitter wire (inaccord with Faraday's law) distributed in such a manner as to maintain the magnetic field everywhereas it existed before turn-off.TransmitterReceiverInducea Induced currentFIGURE 6: Transient current flow in the ground(mod. from Hersir and Björnsson, 1991)The finite ground resistivity causesthe amplitude of this current to startdecaying immediately. This surfacecurrent flow decreases as r -4 fromthe transmitter, where r is distancefrom the transmitter (Geonics Ltd,1980). Close to the transmitter theinduced surface current starts todiffuse into the homogeneous halfspace,whereas the current at a greatdistance maintains the valuedictated by r -4 fall-off. As timeelapses, the current effectivelyappears to move outwards anddownwards as a diffusing ring dueto the deficiency of current near thetransmitter (Figure 6). Themagnitude of the current induced attime t = 0 at the surface isindependent of the conductivity butits apparent velocity at which itsring expands is inverselyproportional to the square root ofthe conductivity of the medium it isdiffusing through. This decaying current and magnetic flux induces a voltage that causes more currentto flow, but now at a larger distance from the transmitter loop, and also at greater depth. This deepercurrent flow also decays due to finite resistivity of the ground, inducing even deeper current flow andso on.The amplitude of the current flowing as a function of time is measured by monitoring its decayingmagnetic field using a small multi-turn receiver coil located at the centre of the transmitter loop. Bytaking measurements of the voltage out of the receiver coil at successively later times, measurement ismade of the current flow and thus also of the electrical resistivity of the earth at successively greaterdepths.Induced voltage in the receiver coil (in the frequency domain), due to alternating current I 0 e iwttransmitted into a circular loop of radius r on the top of an N-Layered halfspace, is given by (Árnason,1989):

Report 21 417 WameyoV∞ 2eωt− iωμ0λ S0c ( ω,r)= ArnrAsnsI01 )πr∫m S0T0 0 − 0J ( λrdλ(12)where A r = Cross-sectional area of the receiver coil (m 2 );n r = Number of windings in the receiver coil;μ 0 = Magnetic permeability in vacuum (H/m);A s = Cross-sectional area of the transmitter loop (m 2 );n s = Number of windings in the transmitter loop;r = Radius of the source loop (m).S 0 and T 0 are given by the recursion relations;TSi−1 = Sicosh( midi) − Tisinh( midi)mi= Sisinh( midi ) − Ticosh( midm[ ) ]i−1 ii−1STN − 1 = 1;N −1=mmNN −1where d im i= The thickness of the i’th layer (m); and= The impedance of the i’th layer.The quantities S 0 and T 0 which determine the voltage in Equation 12 depend on ω and the2 2conductivities, σ i through m = λ − k where k 2 = ω 2 με – iωμσ (ε is the dielectric permittivity). Inthe quasi-stationary aproximation ε ~ 0, hence k 2 = -iωμσ.Mutual impedance between the source and the receiver coil (by analogy with Ohm’s law) is defined bythe ratio between the measured voltage and the transmitted current. From Equation 12, the mutualimpedance for central-loop configuration is:Z∞c ( 1 )0 0 0 0 − 02VL( ω,r)− iωμ0λ S0ω,r)= = A n A nJ λrdλiωtr r s s(I eπr∫m S T(13)Equation 12 can be expressed in the time domain by a Fourier expansion of the function describing thetransmitted current (Árnason, 1989). If the transmitted current is described by the function I(t), aFourier expansion of the current function will be:∞1~iωtI(t)= I(ω)e1 ∫(14)(2π)2 −∞where∞∫−∞~1I(ω ) = I(t)e1(2π)2−iωtdt(15)From Equation 13, the induced voltage in the receiver coil in terms of mutual impedance (as a functionof frequency) and the Fourier transform of the transmitted current are expressed as:

Wameyo 418Report 21∞1 ∫1~iV t rZ ω r I ω eω tc ( , ) = L ( , ) ( ) dω(16)(2π)2 −∞The measured voltage as a function of time after the steady current is abruptly turned off at t = 0 isthen expressed by:V− It)=2π0−( ∞∫−∞∞∫−∞Z(ω)ie ω t I0idω= ω e ω tΦ ( ) dωiω2π(17)If for simplicity, we define:Z(ω)Φ( ω)=(18)− iωФ(ω) only depends on ω through ω 2 and iω, hence:*Φ ( −ω)= Φ(ω)(19)where * denotes the complex conjugation and thereforeRe Φ ( −ω)=Re Φ (ω); Im Φ ( −ω)= -Im Φ (ω)(20)Equation 17 can therefore be simplified to:∞∫−∞2I0V−( t)= Re Φ ( ω) cos( ωt)dω(21)πTransient voltage generated in the receiver coil due to a linearly ramped step function is given by(Árnason, 1989):00+TOFFI 0I0 V ( t)= V t τ dτTOFF∫ −(− ) = ∫V−( τ ) dτ(22)TOFF−TOFFIn practice the current is not abruptly turned off, but turned off linearly in a time interval of lengthTOFF.For a homogeneous half-space of conductivity σ, the induced voltage in the receiving coil is givenapproximately by (Árnason 1989):tV ( t,r)22C(μ30σr)μ0= I 0, where C= A1 5rnrAsns32 210πt2πr(23)As was discussed in Section 3.2, the time-behaviour of the diffusing current is in three phases; earlytimes, intermediate and late times (Figure 7). At late times the measured voltage V(t) decays in time−52as t and varies as σ 3/2 (Árnason 1989), thus making TEM intrinsically more sensitive to smallvariations in the conductivity than conventional resistivity methods.Apparent resistivity ρ a , of a homogeneous half-space in terms of induced voltage at late times after thesource current is turned off is given by (Árnason 1989):

Report 21 419 Wameyo23μ ⎡0 2μ⎤⎢0ArnrAsnsI0ρ a =(24)5 2( t,r)4π⎢ 5 ⎥ ⎥ ⎣ t V ⎦where t = Time elapsed after the transmitter current is turned off (s);A r = Cross-sectional area of the receiver coil (m 2 );n r = Number of windings in the receiver coil;μ 0 = Magnetic permeability in vacuum (H/m);A s = Cross-sectional area of the transmitter loop (m 2 );n s = Number of windings in the transmitter loop;I 0 = Transmitter current (A);V (t,r ) = Measured voltage (V). 1E-005t -5/2In the case of a homogeneous half-spaceof resistivity ρ, the apparent resistivitygives the true resistivity at late times.The definition of apparent resistivity isbased on the time behaviour of thereceiver coil output voltage at late timeswhen it decays as t -5/2 as seen from Figure7. At earlier and intermediate times,receiver voltage is too low (relative to thedashed line indicating the approximationof late-time voltage back in time) and thus,from Equation 24, the apparent resistivityρ α (t) will be too high. This results in a"descending branch" at early time wherethe apparent resistivity is higher than thehalf-space resistivity as shown in Figure 8.However, at late time, the apparentresistivity ρ α (t) is equal to half-spaceresistivity ρ 1 .Log Receiver output voltage (t)1E-0061E-0071E-008Early timeIntermediate timeLate time0.0001 0.001 0.01 0.1Log TimeFIGURE 7: Log plot of receiver output voltageversus timeIf we let the earth be two-layered with theupper layer having resistivity ρ 1 , andthickness h, and basement resistivity ρ 2(>ρ 1 ), at early times when the currents areentirely in the upper layer of resistivity ρ 1the decay curve will look like that of ahomogeneous half-space of resistivity ρ 1 ifthe layer is sufficiently thick (Figure 7),and the apparent resistivity curve will looklike Figure 8 for a while. Later on thecurrents will lie in both layers, and atmuch later times, they will be locatedentirely in the basement of resistivity ρ 2 .Log ρ a (t)1000ρDescending branchSince ρ 2 > ρ 1 , Equation 23 shows that (asindicated in curve a in Figure 9) themeasured voltage will now be less than itwould have been for the homogeneoushalf-space of resistivity ρ 1 . The effect on1000.0001 0.001 0.01 0.1Log TimeFIGURE 8: Apparent resistivity of a half-spaceversus time

Report 21 421 WameyoNatural electric and magnetic field strengths are simultaneous recorded in two orthogonal, horizontaldirections as a function of time. The resulting time-series data is then Fourier transformed to thefrequency domain and processed to derive the impedance tensors of the apparent resistivities andphase.For geothermal exploration, the MT method targets deep brine reservoirs and hot (or partially molten)rocks that act as the heat source for a geothermal system under survey (Ward and Wannamaker, 1983).4.2 Homogeneous earthWhen electromagnetic monochromatic plane waves propagate downwards (along the z-axis) in ahomogeneous and isotropic medium, the electric and magnetic field vectors are orthogonal and theratio of the electric to magnetic field intensity is a characteristic measure of the electromagneticproperties of the medium, often called characteristic impendence (Cagniard, 1953; Keller andFrischknecht, 1966). This characteristic impendence is described by the following equation:Zωμ=k=0EHxyE= −Hyx(25)where Zωμ 0E x,yH x,y= Characteristic impendence;= Angular frequency (2πf) where f is frequency (Hz);= Magnetic permeability (H/m);= Electric field intensity (V/m) in x,y direction;= Magnetic field intensity (T or Tesla) in x,y;i ωμ iωε÷ σ stands for the wave propagation constant;k = ( )ε= Dielectric permittivity (C/Vm);If displacement currents are neglected, k = iωμσand Equation 25 can be rewritten as:ωμiπ0Z = = i ωμ40ρ= ωμ0ρe(26)ρiμ σω0and there is a π/4 = 45° phase differencebetween E x and H y, with E x leading H y.(Figure 11).The resistivity of the homogeneous halfspaceis given as (Ward and Wannamaker,1983):1 2ρ = Z ;ωμ0EHxZ = (27)yElectric field intensity (mV/km)Magnetic field intensity in (nT)This can be reformulated by using theFIGURE 11: Homogeneus half-space response ofmagnetic field intensity, H, and theelectric and magnetric field intensityr rmagnetic field as related by B = μH(μ = 4π ×10 -7 ). In practice the electric field is expressed inmV/km and the magnetic field in nT (or 10 -9 Tesla = gamma). Therefore, Equation 27 can beexpressed as:

Wameyo 422Report 2122−6T Ex10μ TμEx6 Ex= = 10 = 0.2T−92πμB102πByByρ (28)where Tρ= Period in seconds;= Resistivity (Ωm).For a non-homogeneous earth, apparent resistivity, ρ a , and phase, θ a , are defined as:20ρ a = 0.2TZand θ a = arg(Z 0 ) ≠ 45°where Z 0= Impedance at the surface4.3 Layered earthFor horizontally N-layered earth the plane wave impedance is given by the recursive formula (Wardand Wannamaker, 1983):whereZˆωμN= ZN=kNZμZn 0 ω = (intrinsic impedance of the n th layer); withk nh nZ ∧nk n= √(iωμσ n ); and= Thickness of the n th layer;∧∧Z n + Zn−1tanhZ n−1= Zn−1(29)∧n−1+ Zntanh( ikn−1hn−1)( ik h )n−1n−1= Impedance at the top of the n th layer and Z 1 = Z 0 is that on the surface.For two-layered earth (Figure 12), Equation 28 becomes:∧Z1∧1( k1h1)( k h )Z 2 + Z1tanh= Z1(30)∧Z + Z 2 tanhwhere Z 1 = Z 0 = Impedance at the surface; andFIGURE 12: Two-layered earthi i2πμk1=ωμ=ρ1ρ1TFor a large period T, k 1 h > ρ 2 , i.e. we have a good conductor at depth h 1 and Z 1>> Z ∧ 2 , Equation 31 becomes:

Report 21 423 WameyoandZZ + iZ k hiωμ= (32)2 1 1 10 Z1≈ iZ1k1h1= * k1h1= iωμh1kZ11+ Z 2 k1h1ρ 1 2 12 2πμh1a = Z0= ( ωμh1)ωμ ωμ=(33)T2The depth to the good conductor, h 1 , can be calculated from ρ a at large T according to:ρ aTh1= (34)2πμEquation 33 shows that when plotting log ρ vs. log T the slope is -1 at large TWhen ρ 1

Wameyo 424Report 215. TEM AND MT RESISTIVITY SURVEY IN MENENGAI5.1 Geological setting6The Kenya Rift Valley is aprominent feature that is ofgreat geographic andgeological interest. It is a partof the East Africa Rift system,an intra-continental divergentzone where rift tectonics areaccompanied by activevolcanism.SeveralQuaternary volcanoes arefound within the Kenyasegment of the rift floor.Some of these volcanic centresare dotted with hydrothermalactivities and are pictured tohost geothermal systemsdriven by magmatic heatsources (Mungania, 2004).Menengai geothermal prospectis located within an areacharacterized by complextectonic activity associatedwith a rift triple junction. Thisis at the zone where theNyanza rift joins the mainKenya rift (Figure 13).Menengai is located about 20km south of the equator. Themajor structural systems in thearea are the Menengai caldera,the Molo tectonic axis and theLongitudes543210-1-2-3-4-5UGANDAL. VictoriaTANZANIACentral IslandEburuOlkariaSuswaSilaliPakaKorosiL. BaringoMenengaiLongonotL. MagadiShomboleKENYAMt. KenyaVolcanic centreRift faults-632 33 34 35 36 37 38 39 40 41 42LatitudesSolai graben (Figure 14). Menengai caldera is elliptical with the major and minor axis measuring about12 and 8 km, respectively. The ring structure has been disturbed by the Solai graben faults on thenortheast end and one fracture at the south-southwest end of the caldera wall extending southwards(Figure 14). The Molo TVA / Ol rongai fracture system intersects the Menengai caldera on the northnorthwestpart. The caldera floor covers an area of about 88 km 2 , and is partially covered by youngrugged lava flows that are post caldera in age. Most of the lavas are from fissure eruptions that flowedout of fracture openings. An extensive area around the caldera is to a large extent covered bypyroclastics erupted from centres that are associated with Menengai volcano.The Molo TVA is quite a prominent structural feature represented on the surface by a zone with aconcentration of faults and fractures along which volcanic eruptions have taken place (Figure 14). Theother structure that is of significance in this area is the Solai tectonic axis. It comprise numerousfaults/fracture systems trending N-S. It cuts through the caldera from the northern end and extendssouthwards beneath the Menengai volcanic pile. This fracture system is important as a hydrologicalcontrol and possibly contributes to permeability enhancement of the brittle lava formations underlyingthe Menengai eruptives.NyanzaNorth IslandBarrierNamarunuEmuruangogolakBogoriaSouth IslandNairobiOlorgesailleETHIOPIAMt. KilimanjaroFIGURE 13: Map of Kenya showing location of Menengaigeothermal prospect and other Quaternary volcanic systemsalong the Kenyan rift axisSOMALIA

Report 21 425 Wameyo100000009995000Solai9990000Olbanita swampKampi ya motoNorthings (m)99850009980000KabarakBahati997500099700009965000LEGENDMajor fault(rift scarp)Rift floor faultsInferred faultsLava eruptioncentre160000 165000 170000 175000 180000 185000Eastings (m)Nakuru townFIGURE 14: Map of Menengai geothermal prospect showingthe caldera outline, volcanic centres and other main structures5.2 TEM surveyA total of 62 TEM soundings, covering an area of about 50 km 2 , were carried out in the Menengaiprospect area using a central loop TEM array. About 38 soundings were done within the caldera floorand its surroundings, while the remaining soundings were scattered in other parts of the prospect field(Figure 15).The equipment that was used for this survey (purchased from Zonge Engineering Co. except for thepower generator) comprised the following units:• Transmitter - GGT-3 (maximum loop current of 12 amps)• Transmitter controller - XMT-16• Data logger - GDP-16 (three channel unit)• Receiver coil - TEM/3 (receiver moment of 10000 m 2 )• Voltage regulator - VR-1• Power source - 5HP Honda generator (120 Vac at 400Hz)In all the soundings, a 300 m x 300 m transmitter wire loop was used. A half-duty square wavecurrent was transmitted at frequencies 16, 8, 4 and 2 Hz. Logarithmically spaced sampling gates wereused with 16 Hz having 25 gates starting at 36.14 μs to 12.18 ms; 4 Hz had 31 sampling gates startingat 36.14 μs to 48.42 ms and 2 Hz had 34 gates starting at 84.18 μs to 96.85 ms. At each repetition rate,several repeated transients were stacked and stored in a memory cache inside the data logger and werelater transferred to a personal computer (PC) for processing.

Wameyo 426Report 21100000009995000NS-4SolaiTEM15 TEM17TE M 16 NS-5TEM19Olbanita swampNorthings (m)999000099850009980000Kampi ya motoTEM02TEM06NS-1NW-SEEW-1 TE M 61EW-2TEM58KabarakEW-3NS-2TEM28TEM12TEM43TEM42NS-3TEM45TEM44TEM5 9TEM41TEM48TE N 62TEM60TE M 18TE M 47TE M 50TEM57TEM36TE M 09TE M 13TEM22TEM23BahatiTEM49TEM21TEM20EW-4TEM11TE M 30TE M 31 TEM32TEM33TEM51TEM35TE M 34TEM54TEM52TE M 379975000TE M 03EW-5EW-6TEM01 TEM08TE M 24TEM25TEM56TEM10TEM53TEM26TEM27TEM46TEM40TEM07TEM55LEGENDTEM14TEM05TE M 29TE M 39TE M 389970000Major fault(rift scarp)EW-7NE-SWTEM04Rift floor faultsInferred faultsNakuru townTEM Sounding9965000160000 165000 170000 175000 180000 185000Eastings (m)FIGURE 15: Map showing location of TEM soundings and resistivity cross-sectionsThe transmitter and the receiver timing were controlled by synchronized high-precision crystal clocks.This was achieved by synchronizing the receiver and the transmitter controller prior to the dataacquisition procedure to ensure that induced voltage was measured by the receiver only after thetransmitter was turned off.5.2.1 Data processing and interpretationTEMX program was used to read raw data files downloaded from the GDP-16 receiver. This programcalculates averages and standard deviations of repeated transient voltage measurements and calculateslate time apparent resistivity as a function of time. The program has a graphical-user interface (GUI)and offers the possibility of editing raw data by rejecting some of the noisy readings.The CLTINV interpretation program was used to perform 1D inversion on the data. The program wasdeveloped by Knútur Árnason at ÍSOR. The program assumes that the source loop is a square loop

Report 21 427 Wameyoand that the receiver coil/loop is at the centre of the source loop. The current wave form is assumed tobe a half-duty bipolar semi-square wave with exponential current turn-on and linear current turn-off.The inversion algorithm used in the program is the nonlinear least-squares inversion of the Levenberg-Marquardet type (Árnason, 1989). The misfit function is the root-mean-square difference betweenmeasured and calculated values (chisq), weighted by the standard deviation of the measured values.The user is offered the option of choosing whether the program fits the measured voltage or the latetime resistivity values. Smooth models were also achieved by using Occam’s (minimum structure)inversion. In this case many layers (20) were used with fixed equal thicknesses on a log-scale betweenthe surface and the basement depth and variable resistivity values. The variation in resistivity waskept smooth with little contrast between layers (Appendix I).5.3 MT surveyIn all, 67 MT soundings were carried out in Menengai prospect. The sounding distribution over theprospect area is as shown in Figure 16. For the MT soundings 3 sets of 5-channel MT data acquisitionsystems were used (MUT-5A) from Phoenix Geophysics-Canada, as shown in Figure 17. MT-5Asystems acquire MT data in frequencies ranging from about 400 Hz to 0.0000129 Hz (a period of100000009995000Solai9990000MT17Olbanita swampKampi ya motoMT02MT05NW -SEMT66Northings (m)99850009980000KabarakMT52MT11MT42MT64MT62MT54MT47MT57MT49MT59MT53MT55MT51MT50Bahati9975000MT04MT10MT58MT14MT13MT60MT29MT01MT65MT61LEGENDMT339970000Major fault(rift scarp)Rift floor faultsInferred faultsNE-SWNakuru townMT Sounding9965000160000 165000 170000 175000 180000 185000Eastings (m)FIGURE 16: Map showing location of MT soundings and resistivity cross-sections

Wameyo 428Report 21FIGURE 17: Field array for a 5-channel MT dataacquisition system (Phoenix Geophysics Ltd, 2003)about 21.5 hrs), and issuitable for shallow anddeeper investigations. One ofthe 5-channel units wasinstalled at a fixed location,about 30 km from Menengaicaldera (which was the mainarea of focus), and was usedas a remote-reference stationfor the other two dataacquisition systems. Thiskind of array takes advantageof the fact that electromagneticnoise from thepower line (50 or 60 Hz) andhuman activities tends to varyconsiderably over distance,whereas the natural magnetic signal tends to be the same over large distances; the lower the frequency,the less variation.Since the instruments are synchronized to Co-ordinated Universal Time (UTC) via signals from theGlobal Positioning System (GPS) satellites, low-noise time-series data acquired from the remotereference station were processed in combination with data from the other stations (acquiredsimultaneously) to reduce the effects of local noise and improve the quality and reliability of thesurvey results.5.3.1 Data processing and interpretationTime-series data downloaded from the the MTU-5A units were viewed using the program SyncTSV.This program allows viewing and printing of graphical representations of the raw time-series data,power spectra derived from the time-series data and coherence between pairs of orthogonal channels.Using the program SSMT2000, provided by Phoenix Geophysics-Canada (Phoenix Geophysics,2003), Fourier transforms were produced from the raw time-series data. Fourier coefficients were thenreprocessed using data from the reference site to filter out noise-affected data. The cross-powers werestored in files and could be displayed graphically using the program, MTU-Editor. Those files werethen converted to industry-standard EDI format for use with WinGLink, geophysical interpretationsoftware (from Geosystem).TEM data (averaged stacks from GDP-16, TEM raw data) collected at the same locations as the MTsite were also exported to the WinGLink program, where their 1-D models were used for static shiftcorrections on the MT data. The WinGLink program was used for 1-D inversion of the MT data.Appendix II shows an example of interpretateion of MT data.6. RESULTS OF THE TEM AND MT SURVEY6.1 General discussionThe resistivity structure of high-temperature geothermal systems is generally characterized by a lowresistivitycap at the outer margins of the reservoir, underlain by a more resistive core towards theinner part. This structure has been found in both freshwater systems as well as brine systems, with thelatter having lower resistivity values. Comparison of this resistivity structure with data from wells hasbeen carried out in high-temperature geothermal fields, in Iceland (Árnason et al., 2000) and in the

Report 21 429 WameyoAsal Rift (Djibouti, East Africa) (Árnason et al., 1988). The results have shown a good correlationwith alteration mineralogy.As discussed in Section 2.5, the resistivity is relatively high in cold unaltered rocks outside thereservoir and rather low at temperatures of 50-100°C as alteration intensity is normally low in thistemperature range. At temperatures of 100-220°C, low-temperature zeolites and the clay mineralsmectite are formed that have high conductivity. In the temperature range from 220 to about 240-250°C, the zeolites disappear and smectite is transformed into chlorite, forming the transition from thelow-resistivity cap to the more resistive core. At about 240-250°C the smectite disappears and theresistive chlorite becomes the dominant mineral. This observation is of great importance, because thetemperature dependence of the alteration mineralogy makes it possible to interpret the resistivitylayering in terms of temperature, provided that the temperature is in equilibrium with the dominantalteration mineral.The purpose of this research is to analyse TEM and MT data that had been collected in the Menegaigeothermal prospect to study its resistivity structure in order to delineate the reservoir area, its heatsource and possible up-flow zones. All the TEM soundings and used MT soundings together with theinversion models are published as special appendices to this report (Wameyo, 2005). The TEMresistivity structure of Menengai prospect is presented in 12 resistivity cross-sections crossing majorgeological structures, as well as by 2 iso-resistivity maps at different elevations with respect to sealevel. The location of the resistivity cross-sections is shown in Figure 15.6.2 TEM cross-sectionsResistivity cross-section EW-2 (Figure 18) at the northern border of Menengai caldera shows a lowresistivitycap, an anomaly defined with resistivities

Wameyo 430Report 21Elevation (m)W2000180016001400Menengai CratermE184159140TEM43122110TEM28TEM13TEM59 TEM4186TEM4871TEM36645953494541363127232018151232000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 15000Horizontal distance (m)High resistivity coreFIGURE 19: TEM resistivity cross-section EW-3Resistivity cross-section EW-3 (Figure 19) cuts across the caldera 2 -3 km to the south of EW-2. Itreveals a low-resistivity cap over a high-resistivity core. The low-resistivity cap is observed at anelevation of 1800 m a.s.l. in the westernmost sounding (TEM28). It slopes to the east but there is noindication of sloping to the west in this section. A high resistivity core is seen in the western part ofthe cross-section, reaching an elevation of almost 1600 m a.s.l. in sounding TEM28 sloping down tothe east.Elevation (m)2000180016001400WTEM11TEM30 TEM31 TEM32TEM33Menengai CraterTEM511000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 15000 16000TEM35Horizontal distance (m)TEM37FIGURE 20: TEM resistivity cross-section EW-4TEM34High resistivity coreEm184159140122110867164595349454136312723201815123Resistivity cross-section EW-4 passes right across the caldera (Figure 20). It clearly defines a lowresistivitycap over a high-resistivity anomaly in the northwest part of the caldera. The low-resistivitycap reaches an elevation of approximately 1800 m a.s.l. in the westernmost sounding (TEM11), andthen shows gradual sloping to the east along the cross-section. There is an indication of the lowresistivitycap doming up in the easternmost sounding (TEM34) but that cannot be confirmed as moresoundings are needed for that. The high-resistivity core anomaly is sloping well on both sides,defining boundaries within this profile. An interesting feature is the upward doming of the lowresistivitycap almost to the surface in soundings in the middle of the caldera (TEM54 and 33) wheresurface manifestations are observed.Resistivity cross-section EW-5, across the caldera (Figure 21), shows the low-resistivity cap across thesection. It is close to the surface to the west and east of the caldera, but at greater depth within it. Thelow resistivity is underlain by higher resistivity in the western part and west of the caldera.

Report 21 431 WameyoElevation (m)2000180016001400WTEM03TEM01TEM08TEM542000 4000 6000 8000 10000 12000 14000 16000 18000 20000Horizontal distance (m)FIGURE 21: TEM resistivity cross-section EW-5Resistivity cross-section EW-6 crosses the southern part of caldera (Figure 22) showing low resistivitysloping down to the east as the cross-sections reach the southern part of the caldera. Higher resistivityis seen at depth under the caldera indicating that the cross-section is close to the southern margin ofthe reservoir.Menengai CraterTEM10Menengai CraterTEM56TEM40TEM37TEM07High resistivity coreEm184159140122110867164595349454136312723201815123Elevation (m)W2000180016001400TEM24TEM27 TEM46TEM25TEM26TEM554000 6000 8000 10000 12000 14000 16000 18000 20000 22000Horizontal distance (m)High resistivity coreFIGURE 22: TEM resistivity cross-section EW-6Em184159140122110867164595349454136312723201815123Resistivity cross-section EW-7, Figure 23, shows a clear thickening of the low resistivity in thewestern part, indicating that the cross-section cuts the cap as it slopes down to the south. A highresistivity body is seen at depth. The cross-section confirms, as seen in EW6, that it is close to thesouthern margin of the reservoir.Elevation (m)2000180016001400WTEM14TEM050 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 15000TEM29Horizontal distance (m)TEM39FIGURE 23: TEM resistivity cross-section EW-7TEM38High resistivity coreEm184159140122110867164595349454136312723201815123

Wameyo 432Report 21Elevation (m)2000180016001400NTEM61TEM58TEM284000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 15000 16000 17000 18000Horizontal distance (m)Resistivity cross-section NS-1 is a north-south bound resistivity cross-section, passing west of thecaldera (Figure 24). It shows a low-resistivity layer sloping down to the north. It is underlain byhigher resistivity except in the far south.Resistivity cross-section NS-2, (Figure 25) crosses the western edge of Menengai caldera and Olrongai hill to the north. The cross-section shows high resistivity at a shallow depth within the calderaand a continuous low-resistivity layer at depth. This low-resistivity cap is shallower to the north, atabout 1800 m a.s.l., and the northern limit is not seen as there is no data farther north of this profile.In the northern part the low resistivity is underlain by higher resistivity, dipping to the south. Wherethe low resistivity is closest to the surface in the northern part, there is fumarole activity along Olrongai hill.Resistivity cross-section NS-3, (Figure 26) passes right through the middle of the caldera and has alow-resistivity cap that opens to the north but slopes down to the south in the northern part of thecaldera (between TEM44 and TEM59). The high-resistive core is deeper (at 1550 m a.s.l.) in thisprofile compared to NS-2 (1650 m a.s.l.), immediately to the west. This shows that a boundary isapproached as we move further east. It is interesting to note the extension of the low-resistivity cap upto the surface in the middle of the caldera, just underneath surface manifestations.Resistivity cross-section NS-4 that also cuts through the caldera on the eastern side (Figure 27) shows athick low-resistivity layer under and north of the caldera (below TEM60) and south (TEM40). Signsof the approaching of a boundary to the east are confirmed by the high-resistivity core below this capthat is at deeper levels (1400 m a.s.l.) when compared to NS-1 and NS-2.TEM11FIGURE 24: TEM resistivity cross-section NS-1Menengai CraterTEM01TEM24High resistivity coreSm184159140122110867164595349454136312723201815123Elevation (m)N2000TEM12TEM05TEM25TEM08TEM28TEM301800160014002000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 15000Horizontal distance (m)High resistivity coreFIGURE 25: TEM resistivity cross-section NS-2Sm184159140122110867164595349454136312723201815123

Report 21 433 WameyoMenengai CraterElevation (m)N2000180016001400TEM45TEM44TEM59 TEM412000 4000 6000 8000 10000 12000 14000 16000TEM51Horizontal distance (m)TEM52TEM56TEM53FIGURE 26: TEM resistivity cross-section NS-3TEM27TEM39SHigh resistivity corem184159140122110867164595349454136312723201815123NMenengai CraterSElevation (m)22002000180016001400TEM17TEM62TEM60TEM18TEM470 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 22000 24000TEM35Horizontal distance (m)TEM46TEM40FIGURE 27: TEM resistivity cross-section NS-4TEM38TEM04High resistivity corem184159140122110867164595349454136312723201815123Resistivity cross-section NW-SE passes through the caldera and Ol ronga to the northwest (Figure 28)and has a thick low-resistivity layer or cap (TEM38 – TEM51) over a deep high-resistivity core that isseen in the northwest (below TEM61) marking the southeastern margin of the system. The low-Menengai Crater mElevation (m)NW2000180016001400TEM38TEM56TEM33 TEM51TEM46TEM12TEM04TEM43TEM614000 6000 8000 10000 12000 14000 16000 18000 20000 22000 24000 26000Horizontal distance (m)High resistivity coreFIGURE 28: TEM resistivity cross-section NW-SESE184159140122110867164595349454136312723201815125

Wameyo 434Report 21SWMenengai CraterTEM21NEmElevation (m)22002000180016001400TEM05TEM26TEM19TEM53TEM56TEM52TEM51TEM35TEM362000 4000 6000 8000 10000 12000 14000 16000 18000 20000 22000 24000 26000Horizontal distance (m)High resistivity coreresistivity cap in this profile is at shallow levels below TEM12, at about 1850 m a.s.l., but deepens tothe northwest. Surface manifestations are observed around TEM12, where the low-resistivity cap isclosest to the surface.Resistivity cross-section NE-SW also passes through the caldera in the Bahati area (Figure 29). Athick low-resistivity cap over a deep resistive core, below 1400 m a.s.l., at the southwest end of thissection marks the southern margin of the anomaly observed in the western area of the caldera fromprevious resistivity sections. To the northeast, the low-resistivity cap deepens until we get to TEM36where emergence of yet another system is observed below TEM09 with the emergence of a new highresistivitybody. Further to the northeast the low-resistivity layer reaches elevation of approximately2000 m a.s.l. For clarification, this system will here be called the Solai-system.TEM09TEM23FIGURE 29: TEM resistivity cross-section NE-SWTEM491841591401221108671645953494541363127232018151036.3 TEM resistivity mapsThe results of the surveyare also presented as isoresistivitymaps showingresistivity at a certaindepth.Iso-resistivity map at 1800m a.s.l. (Figure 30), cutsthrough the top of the lowresistivitycap where itreaches the highestelevation just west of theMenengai caldera. Due toinsufficient data in the areawest of the caldera the areais “open” to the west. Thelow resistivity in thenortheast shows the top ofthe low-resistivity capcovering the Solai system.The indication of the Solaigeothermal system, is onlyNorthings (m)10000000999500099900009985000998000099750009970000So laiTEM19TEM15TE M 17TEM16Olbanita swampTEN62Kampi ya motoTE M 21TEM22TEM06TEM 60TEM02TEM 61TE M 49TE M 50TEM45TEM23TEM20TEM18TEM12TEM44TE M 58TEM 42TEM47TEM57TEM09KabarakTEM28TEM43TEM 59TE M 13 BahatiTEM41TEM 48TEM36TEM 11TEM35TEM30 TEM31 TE M 32 TEM33TEM51TE M 34TE M 54TE M 52TEM37TEM56TEM03TEM40TEM01 TEM08TE M 10TEM07TEM53TEM46TEM55TE M 24TEM25 TE M 26TEM27TEM14TEM05 TE M 29TEM 39TEM38TEM04Nakuru town9965000160000 165000 170000 175000 180000 185000Eastings (m)FIGURE 30: TEM resistivity distribution at 1800 m a.s.l.m1276551403327231816141210851

Report 21 435 Wameyofounded by a few soundings in the northeastern part of the survey area. High resistivity (> 200 Ωm) isobserved within most parts of the Menengai caldera related to the unaltered young lava flows thatcover most parts of the caldera and its environs at shallow levels.Northings (m)100000009995000999000099850009980000997500099700009965000Kampi ya motoTEM026.4 MT resultsTEM03TEM06KabarakTEM14TEM 61TEM58TEM 11TEM01TEM24TEM28TEM 08TEM12TE M 25TEM 05TE M 43TEM15TEM 26TEM 29TE M 42TEM10TEM44TEM27160000 165000 170000 175000 180000 185000TEM45Eastings (m)TE M 17TEM18TE M 35TEM04TEM36TEM09TEM13TEM07TEM16TEM30 TEM31 TEM 32 TEM33TEM51TEM 34TEM54Estimated Menegaigeothermal boundaryOlbanita swampTEM59TE M 53TEM39TEM41TE M 48TEM52TEM56TEN6 2TE M 60TEM47TEM46TEM40TEM38Nakuru townTE M 50TEM57TE M 37SolaiTEM22TEM23BahatiTEM49TEM55TEM21TEM20TEM19?High resistivity coreFIGURE 31: TEM resistivity distribution at 1600 m a.s.l.m1276551403327231816141210851Iso-resistivity map at 1600m a.s.l., at an average 400m depth from surface(Figure 31), cuts the lowresistivitycap of theMenengai system as itslopes down towards northand south. The map cutsthe top of the highresistivitycore within thecaldera and northwest ofthe caldera. Here it isclear that the Menengaigeothermal system extendsto the northwestoutside the survey area.The high resistivity in theeastern part of the calderareflects rocks outside thegeothermal reservoir, notaffected by thermalalteration. Northeast isnot well defined due toinsufficient data coveragebut soundings east ofSolai show a clearevidence of the anomalythat has been referred to asthe Solai system.MT resistivity cross-section NE-SW, shown in Figure 32, passes right through the middle of Menegaicaldera along its main axis. A low-resistivity anonamly is seen along the whole profile at 0.5-1 kmMenengai Crater2000MT01 MT60MT13 MT58 MT57MT55 MT53 MT59 MT51Elevation (m)0-2000-4000-60004000 6000 8000 10000 12000 14000 16000 18000Horizontal distance (m)FIGURE 32: MT resistivity cross-section NE-SWm18415914012211086716459534945413631272320181051

Wameyo 436Report 21Menengai Crater2000MT66MT42 MT64 MT62 MT57 MT58 MT65mElevation (m)0-2000-4000-6000184159140122110867164595349454036312723201812512000 4000 6000 8000 10000 12000 14000 16000 18000 20000 22000Horizontal distance (m)FIGURE 33: MT resistivity cross-section NW-SEdepth, with the lowest resisitivity to the southwest (< 10 Ωm) and the northeast (

Report 21 437 Wameyo7. CONCLUSIONS AND RECOMMENDATIONS• The Menengai geothermal system exceeds 150 km 2 at 1300 m a.s.l. as defined by resistivitystructures.• A low-resistivity cap delineates the geothermal system, showing boundaries in all directionsexcept northwest.• The low-resistivity cap is underlain by a high-resistivity core that reaches an elevation ofapproximately 1750 m a.s.l. close to the northeastern caldera rim.• The existence of a high-resistivity core indicates reservoir temperatures exceeding 240-250°C,provided there is an equilibrium between the temperature in the reservoir and the alteration ofthe rocks, as expressed by the resistivity.• Soundings in the northeast part of the survey area indicate the existence of another geothermalsystem, the Solai system. A more extensive survey is needed to verify that and to delineate it.• MT resistivity cross-sections show low resistivity at deeper levels that could be related to heatsources.• Major geological structures appear to have a strong influence on the resistivity distribution inthis prospect area, as observed in the MT results.The Menengai prospect appears to host a large geothermal system that can be developed to boostKenya´s power demand. This field could be defined better if more TEM and MT soundings were donewest and north of the prospect. Additional soundings inside the caldera will also fine-tune the pictureinside the caldera, especially in areas where the low-resistivity cap is not well defined due to large dataspacings.The experience during prospecting for geothermal resource in Menegai area proved that using bothTEM and MT has proven to be very quick and effective in exploring for geothermal resources. Thesensitivity of electromagnetic methods to low resistivity makes it easier for TEM to define thicknessand depth of low-resistivity caps that are common in high-temperature geothermal systems. MT, onthe other hand, is handy in locating and defining deep resistivity anomalies such as low resistivityassociated with thermal excitation of conduction in the massive crystalline rock comprising thebasement.ACKNOWLEDGEMENTSI wish to express my sincere gratitude to KenGen for the chance and support to participate in thistraining. Also to Martin Mwangi, Silas Simiyu, Zack Muna, Nicholas Mariita, Stephen Onacha andthe entire KenGen geophysics staff for support during my preparation for the course. I wish to thankour human resources staff Mrs. Beatrice Soy, Mary Gachuhi, Suzy Ogwaye and Matthew Chebogelfor their assistance and support in granting me study leave and getting necessary travelling documents.To the UNU, I am very grateful for the financial support and the rare opportunity that you havegenerously given me to gain wonderful knowledge from this course. I wish to express my unreservedappreciation to the UNU-GTP staff, Dr. Ingvar B. Fridleifsson (Director), Mr. Lúdvík S. Georgsson(Assistant director) and Mrs. Gudrún Bjarnadóttir for making our stay in Iceland comfortable and forthe superb coordination of all our training activities and excursions. I am greatly indebted to mysupervisors, Knútur Árnason assisted by Dr. Hjálmar Eysteinsson and Ragna Karlsdóttir, who are partof ISOR’s geophysics team, for their guidance, sharing knowledge and experience with me, not tomention the encouragement they constantly gave me during my specialized course and project writeup.I feel honoured to have been your student.Finally, I wish to thank my wife, Phoebe Agufa and my sons, Phillip Wameyo and Pascal Mudave forenduring my absence, for their encouragement and support. May God bless you and grant you yourwishes and happiness in life.

Wameyo 438Report 21REFERENCESArchie, G.E., 1942: The electrical resistivity log as an aid in determining some reservoircharacteristics. Tran. AIME, 146, 54-67.Árnason, K., 1989: Central loop transient electromagnetic sounding over a horizontally layeredearth. Orkustofnun, Reykjavík, report OS-89032/JHD-06, 129 pp.Árnason, K., Björnsson, G., Flóvenz, Ó.G., and Haraldsson, E.H., 1988: Geothermal resistivity surveyin the Asal Rift in Djibouti. Orkustofnun, report OS-88301/JHD-05 prepared for UND-OPS andISERT, vol. I, 48 pp.Árnason, K., Karlsdóttir, R., Eysteinsson, H., Flóvenz, Ó.G., and Gudlaugsson, S.Th., 2000: Theresistivity structure of high-temperature geothermal systems in Iceland. Proceedings of the WorldGeothermal 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.Geonics Ltd, 1980: Application of Transient Electromagnetic techniques. Geonics Ltd., Technicalnote NT-17.Hersir, G.P., and Björnsson, A., 1991: Geophysical exploration for geothermal resources. Principlesand applications. UNU-GTP, Iceland, report 15, 94 pp.ISL, Michigan State University, 1999: Liquid molding. ISL, Michigan State University, webpageislnotes.cps.msu.edu/trp/rtm/modl_mes.html .Jones, A.G., 1988: Static shift of magnetotelluric data and its removal in a sedimentary basinenvironment. Geophysics, 53-7, 967-978Keller, G.V., and Frischknecht, F.C., 1966: Electrical methods in geophysical prospecting. PergamonPress Ltd., Oxford, 527 pp.Morris, F.H., and Becker A., 2001: Berkeley course in applied geophysics. University of California,Berkeley, Ca, unpublished.Mungania, J., 2004: Geological studies of Menengai geothermal prospect. KenGen Ltd., internalreport, 18 pp.Phoenix Geophysics, 2003: V5 System 2000MTU/MTU-A User Guide. Phoenix Geophysics, Ltd.,Toronto.Quist, A.S., and Marshall, W.L., 1968: Electrical conductances of aqueous sodium chloride solutionsfrom 0 to 800°C and at pressures to 4000 bars. J. Phys. Chem., 72, 684-703.Wameyo, P.M., 2005: Appendices to the report: Magnetotelluric and Transient Electromagneticmethods in geothermal prospecting, with examples from Menengai, Kenya. UNU-GTP, Iceland,Report 21 appendices, 26 pp.Ward, S.H, and Wannamaker, P.E., 1983: The MT/AMT Electromagnetic method in geothermalexploration, UNU-GTP, Iceland, report 5, 107 pp.

Report 21 439 WameyoAPPENDIX I: TEM 1-D modelling using CLTINVAPPENDIX II: MT 1-D modelling