Thin Lithosphere Beneath the Ethiopian Plateau Revealed by a Joint Inversion of Rayleigh Wave Group Velocities and Receiver Functions

Thin Lithosphere Beneath the Ethiopian Plateau Revealed by a

Joint Inversion of Rayleigh Wave Group Velocities

and Receiver Functions

Mulugeta T. Dugda,1 Andrew A. Nyblade,1 and Jordi Julia2

Received 22 December 2006; revised 4 April 2007; accepted 3 May 2007; published 11 August 2007.

[1] The seismic velocity structure of the crust and upper mantle beneath Ethiopia andDjibouti has been investigated by jointly inverting receiver functions and Rayleigh wavegroup velocities to obtain new constraints on the thermal structure of the lithosphere. Mostof the data for this study come from the Ethiopia broadband seismic experiment,conducted between 2000 and 2002. Shear velocity models obtained from the jointinversion show crustal structure that is similar to previously published models,with crustal thicknesses of 35 to 44 km beneath the Ethiopian Plateau, and 25 to 35 kmbeneath the Main Ethiopian Rift (MER) and the Afar. The lithospheric mantle beneaththe Ethiopian Plateau has a maximum shear wave velocity of about 4.3 km/s and extendsto a depth of �70–80 km. Beneath the MER and Afar, the lithospheric mantle has amaximum shear wave velocity of 4.1–4.2 km/s and extends to a depth of at most 50 km.In comparison to the lithosphere away from the East African Rift System in Tanzania,where the lid extends to depths of �100–125 km and has a maximum shear velocity of4.6 km/s, the mantle lithosphere under the Ethiopian Plateau appears to have been thinnedby �30–50 km and the maximum shear wave velocity reduced by �0.3 km/s. Resultsfrom a 1D conductive thermal model suggest that the shear velocity structure of theEthiopian Plateau lithosphere can be explained by a plume model, if a plume rapidlythinned the lithosphere by �30–50 km at the time of the flood basalt volcanism(c. 30 Ma), and if warm plume material has remained beneath the lithosphere since then.About 45–65% of the 1–1.5 km of plateau uplift in Ethiopia can be attributed to thethermally perturbed lithospheric structure.

Citation: Dugda, M. T., A. A. Nyblade, and J. Julia (2007), Thin Lithosphere Beneath the Ethiopian Plateau Revealed by a Joint

Inversion of Rayleigh Wave Group Velocities and Receiver Functions, J. Geophys. Res., 112, B08305, doi:10.1029/2006JB004918.

1. Introduction

[2] In this paper, we investigate the seismic velocitystructure of the crust and upper mantle beneath Ethiopiaand Djibouti by jointly inverting receiver functions andRayleigh wave group velocities to obtain new insights intothe thermal structure of the lithosphere. Much of Ethiopiaand Djibouti has experienced Cenozoic hot spot tectonism,including flood basalt volcanism, plateau uplift, and rifting,and the impact of these processes on the lithosphericstructure of the region is poorly understood, particularlyin regard to the thickness of the lithosphere beneath theEthiopian Plateau.[3] The hot spot tectonism began with basaltic volcanism

in southwestern Ethiopia at about 45–40 Ma [Davidson andRex, 1980; Zanettin et al., 1980; Berhe et al., 1987;

WoldeGabriel et al., 1990; Ebinger et al., 1993; Georgeet al., 1998]. The main episode of volcanism, however,initiated during the Oligocene (c. 29–31 Ma) with theemplacement of thick (500–2000 m) flood basalts andrhyolites in the future sites of the Red Sea, the easternmostGulf of Aden, and the central Ethiopian Plateau [Hofmannet al., 1997; Mohr and Zanettin, 1988; Baker et al., 1996;Ayalew et al., 2002; Coulie et al., 2003; Kieffer et al., 2004].Less voluminous syn-rift shield volcanoes formed between30 and 10 Ma, locally creating an additional 1000 to 2000 mof relief on top of the flood basalts [Berhe et al., 1987;Coulie et al., 2003]. Uplift of the Ethiopian Plateau com-menced between 20 and 30 Ma [Pik et al., 2003].[4] The formation of the Red Sea and Gulf of Aden rifts

began in the Oligocene when Africa started separating fromArabia, and can be linked to the complex geometry ofcollision along the Alpine-Himalayan chain [Wolfendenet al., 2005]. The opening of the Eastern Branch of theEast African Rift System to form the Afar triple junctionoccurred long after Arabia separated from Africa. Extensioncommenced around 11 Ma in the northern sector of theMain Ethiopian Rift (MER) [Wolfenden et al., 2004;Chernet et al., 1998; WoldeGabriel et al., 1999] and at

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 112, B08305, doi:10.1029/2006JB004918, 2007

1Department of Geosciences, Pennsylvania State University, UniversityPark, Pennsylvania, USA.

2Department of Geological Sciences, University of South Carolina,Columbia, South Carolina, USA.

Copyright 2007 by the American Geophysical Union.0148-0227/07/2006JB004918$09.00

B08305 1 of 14

c. 18 Ma in southwestern Ethiopia. Wolfenden et al. [2004]suggest that there was a hiatus in volcanism between 6.5and 3.2 Ma within the MER, after which time deformationmigrated toward a narrow zone in the rift center, and that by1.8 Ma volcanism and faulting had localized to magmaticsegments within the rift.[5] Much of the Quaternary volcanism in the MER has

occurred within the magmatic segments, but some has alsooccurred along the rift shoulders [Ayalew et al., 2006;Furman et al., 2006; Wolfenden et al., 2004]. Historicalflows [Gibson, 1967] and elevated temperatures at shallowcrustal depths in geothermal fields in the MER suggest thatmagmatic processes within the MER have been activerecently. Ebinger and Casey [2001] and Casey et al.[2006] proposed that the magmatic segments act now asthe locus of extension within this transitional rift settingrather than the rift border faults.[6] Rooney et al. [2005] have used xenoliths from loca-

tions in and along the sides of the MER to investigatelithospheric and sublithospheric processes under Ethiopia,and their findings indicate that the lithospheric mantlebeneath Ethiopia has been modified significantly by silicatemelts, forming pervasive dikes and veins. The resultsfrom Rooney et al. indicating thermal modification of thelithosphere away from the rift axis are consistent withinterpretations of shear wave splitting observations andmagnetotelluric data suggesting thermal modification ofthe lithosphere under the northwestern part of the EthiopianPlateau [Kendall et al., 2005, 2006; Whaler and Hautot,2006; Keir et al., 2005].[7] Because the Cenozoic volcanism, plateau uplift and

rifting in the Horn of Africa cannot be explained easily bysimple passive rifting related to the development of the Afartriple junction, many authors have invoked one or moremantle plumes to account for the hot spot tectonism. Forexample, two plumes (i.e., one at c. 45 Ma in southernEthiopia and one at c. 30 Ma in the Afar region) have beeninvoked by George et al. [1998] and Rogers [2006], a singleplume has been argued for by Manighetti et al. [1997],Ebinger and Sleep [1998], and Courtillot et al. [1999], anda superplume has been suggested by Ritsema et al. [1999]and Furman et al. [2004, 2006].[8] In an attempt to evaluate the different plume models,

Benoit et al. [2006a, 2006b] imaged upper mantle structureunder Ethiopia using data from the 2000 to 2002 Ethiopiabroadband seismic experiment [Nyblade and Langston,2002]. By tomographically imaging the P and S velocitystructure of the upper mantle and by examining topographyon the transition-zone discontinuities, Benoit et al. conclud-ed that the hot spot activity could be the surface manifes-tation of a broad mantle upwelling that is part of the AfricanSuperplume located in the lower mantle beneath southernAfrica. However, because of limited model resolution,Benoit et al. were not able to image structure at depthsabove 100–150 km and therefore could not comment onthe structure of the lithosphere. In a related tomographicstudy using teleseismic P and S traveltimes recorded bythe Ethiopia Afar Geoscientific Lithospheric Experiment(EAGLE), Bastow et al. [2005] were able to image velocityvariations starting at depths as shallow as 75 km beneath thenorthern part of the MER. However, the region imaged by

Bastow et al. did not extend very far across the EthiopianPlateau or include Djibouti.[9] Building on the results of these previous studies, in

this study we investigate further the nature of the litho-sphere beneath Ethiopia and Djibouti, particularly beneaththe Ethiopian Plateau where little is known about litho-spheric mantle structure, by simultaneously invertingreceiver functions and Rayleigh wave group velocities.The shear wave velocity models obtained from the jointinversion can be used to estimate the amount of heatingand thinning of the mantle lithosphere that has occurredbeneath the Ethiopian Plateau related to the Cenozoic hotspot activity, and provide new constraints on how muchof the plateau uplift may have been caused by thermalalteration of the mantle lithosphere.

2. Data and Methodology

[10] Seismic data collected between 2000 and 2002 bythe Ethiopia broadband seismic experiment (EBSE) havebeen used for this study together with data from permanentstations in the region (the IRIS/GSN station FURI and theGEOSCOPE station ATD; Figure 1). 22 of the 27 seismicstations in the EBSE were located either on the eastern orwestern side of the Ethiopian Plateau (Figure 1), and the restof the stations were situated in the MER or Afar Depression.Additional details of the station configuration and recordingparameters used in the EBSE can be found in Nyblade andLangston [2002].[11] In the joint inversion, we used receiver functions and

fundamental mode Rayleigh wave group velocities. Lovewave group velocities were not included because there arefew high quality measurements available for Ethiopia[Benoit, 2005]. The two kinds of data used are complemen-tary. Receiver functions are a time series that represent theradial impulse response (or, in the frequency domain, radialcomponent transfer function) of the shallow structure of theEarth in the neighborhood of the seismic station [Langston,1979], and can be used to resolve velocity contrasts atdiscontinuities and relative traveltimes [Ammon et al., 1990;Julia et al., 2000, 2005]. Rayleigh wave group velocities,on the other hand, can be used to constrain the averageshear wave velocity between the discontinuities [Julia et al.,2000]. Joint inversions of receiver functions and surfacewave dispersion measurements have been performed bymany authors using data from a variety of tectonic settings[e.g., Ozalaybey et al., 1997; Du and Fougler, 1999; Juliaet al., 2000; Tkalcic et al., 2006].

2.1. Joint Inversion of Receiver Functions andRayleigh Wave Group Velocities

[12] We have used the method developed by Julia et al.[2000] for the joint inversion. This method makes use of alinearized inversion procedure that minimizes a weightedcombination of least squares norms for each data set, amodel roughness norm, and a vector-difference normbetween inverted and preset model parameters. The velocitymodels obtained are, consequently, a compromise betweenfitting the observations, model simplicity and a prioriconstraints. For the joint inversion, the two data sets mustbe consistent (i.e., they should sample the same study area).To make the contribution of each data set to the joint least

B08305 DUGDA ET AL.: THIN LITHOSPHERE UNDER ETHIOPIAN PLATEAU

2 of 14

B08305

squares misfit comparable, a normalization of the data set isnecessary, and this is done using the number of data pointsand variance for each of the data sets. An influence factorcan then be used to control the trade-off between fitting thereceiver functions and the group velocity curves.

2.2. Rayleigh Wave Group Velocities

[13] Rayleigh wave group velocities between 10 to 85 speriod from the model of Pasyanos [2005] and between 90and 175 s period from the Harvard model [Larson andEkstrom, 2001] were used for the joint inversion. The modelof Pasyanos includes group velocity measurements fromBenoit et al. [2006c], who conducted a surface wave tomo-graphy study of eastern Africa by adding to the dispersionmeasurements of Pasyanos et al. [2001] new measurementsmade with data from the EBSE. Typical uncertainties ingroup velocities range between 0.01 and 0.02 km/s but

increase to 0.02 and 0.03 km/s for the shortest periods inthe Pasyanos model. Uncertainties in the Harvard groupvelocities are not specified [Larson and Ekstrom, 2001].We did not observe any systematic differences in the twogroup velocity models where they overlap. In order to create asmooth dispersion curve for each station, we extracteddispersion curves from the two models, joined them, andthen applied a 3-point moving (running) average to obtain asmooth composite curve.[14] Figure 2 shows Rayleigh wave group velocity

curves for three different regions in Ethiopia and for theMozambique Belt in northeastern Tanzania. The threeregions in Ethiopia include the Main Ethiopian Rift(MER), and the eastern and western sides of the EthiopianPlateau. At periods greater than �70 s, lower group veloc-ities can be seen for the MER and the two Ethiopian Plateauregions (western and eastern sides), indicating a difference

Figure 1. Map showing the topography of the study area, seismic station locations from the 2000 to2002 Ethiopia Broadband Seismic Experiment, and permanent seismic stations in the region (FURI,ATD). The two lines, AA0 and BB0, show the location of the profiles in Figure 5.


3 of 14

B08305

in lithospheric structure between Ethiopia and the Mozam-bique Belt in northeastern Tanzania. We use the groupvelocities for the Mozambique Belt in Tanzania as areference since the lithosphere under most of Ethiopia priorto rifting was unperturbed Mozambique Belt lithosphere[Burke and Sengor, 1986; Shackelton, 1986; Berhe, 1990;Vail, 1988, 1985; Kroner et al., 1987]. Although theMozambique Belt in northeastern Tanzania sits on theeastern edge of the East African Plateau and may thus besomewhat perturbed by the Cenozoic tectonism affectingmost of eastern Africa, this region provides the best possibleestimate from within eastern Africa for the structure ofunperturbed Mozambique Belt lithosphere [Weeraratneet al., 2003; Julia et al., 2005].

2.3. Receiver Functions

[15] Receiver functions were computed using seismo-grams from teleseismic events between distances of 30�and 95� with magnitudes greater than 5.5. A list of eventsused in this study can be found in Dugda et al. [2005]. Mostof the events are from the east (the Indonesian and WesternPacific subduction zones) or the northeast (Hindu Kush–Pamir region).[16] The time domain iterative deconvolution method of

Ligorria and Ammon [1999] was employed to compute thereceiver functions, and the quality of the receiver functionswas evaluated using a least squares misfit criterion. Thismisfit criterion provides a measure of the closeness of areceiver function to an ideal case, and it is calculated byusing the difference between the radial component seismo-gram and the convolution of the vertical component seis-mogram with the already determined radial receiverfunction. Usually, receiver functions with a fit of 90% andabove were used in the inversion. However, in a few cases,when it was difficult to get a reasonable number of receiverfunctions for a station, receiver functions with fits of 70–90% were included. The receiver functions were filtered

with Gaussian pulse widths of 1.0 and 2.5 to obtain low andhigh frequency band receiver functions, respectively. Radialand tangential receiver functions were examined for evi-dence of lateral heterogeneity and for dipping structure.Events with large amplitude tangential receiver functionswere not used.[17] In the joint inversion, we used three groups of

receiver functions each corresponding to a range of rayparameters from 0.04 to 0.049, from 0.05 to 0.059, and from0.060 to 0.069 (Figure 3b). In addition, for each grouping ofreceiver functions, we computed and stacked two sets ofreceiver functions that have overlapping frequency bands; alower frequency band of f � 0.5 Hz (Gaussian of 1.0), and ahigher frequency band of f � 1.25 Hz (Gaussian of 2.5). Byinverting receiver function stacks over a range of rayparameter and frequency, details of lithospheric structurecan be imaged, such as sharp versus gradational disconti-nuities [Julia et al., 2005; Cassidy, 1992].[18] The initial model used for the inversion consisted of

constant velocity layers that increase in thickness withdepth. Layer thicknesses were 1 and 2 km at the top ofthe model, 2.5 km between 3 and 60.5 km depth, 5 kmbetween 60.5 and 260.5 km depth, and 10 km below a depthof 260.5 km.[19] Our initial inversions showed that the model results

do not depend on the starting model but that velocities atlithospheric depths (��150–100 km) trade-off with veloc-ities below about 190 km depth. To minimize this trade-off,we forward modeled the structure below 190 km depth by apriori fixing the S-velocities in this depth range. Thevelocity structure below 190 km depth was determinedthrough a trial-and-error process by finding models thatbest fit the 140–175 s group velocities for several stationsin each tectonic region. This was done by fixing velocitiesbelow 190 km depth between a range of �15 to +10% ofPREM velocities [Dziewonski and Anderson, 1981], whileat the same time inverting for the velocity structure above190 km depth.[20] Figure 3 shows models with velocities of 5, 7, 10 and

15% less than PREM below 190 km depth for one station.For each station tested, the model with velocities of 10%less than PREM (Figure 3c) fit the long period surfacewave observations the best, while all the models fit therest of the dispersion measurements equally well. Thus, forour final inversion we fixed velocities below 190 km depthto values 10% less than PREM for all stations and invertedfor velocity structure above 190 km depth. The Poisson’sratios in the velocity models were held constant during theinversion. Poisson’s ratios for the crust were taken fromDugda et al. [2005], and for the mantle we used Poisson’sratio from the PREM model. The Poisson’s ratios used forthe crust are average values obtained for the crust beneatheach station. Further testing of our models indicated thatour results are not sensitive to reasonable variations inPoisson’s ratio.[21] To estimate the uncertainties in our model results, we

followed the approach of Julia et al. [2005] and repeatedlyperformed inversions using a range of weighting parame-ters, constraints, and Poisson’s ratio. Similar to the results ofJulia et al., by repeating the inversions for many combina-tions of model parameters and data, we found the uncer-tainties in the shear wave velocities to be about 0.1 km/s in

Figure 2. Rayleigh wave group velocity curves fordifferent parts of the study area; the Main Ethiopian Rift(MER), the western side of the Ethiopian Plateau, and theeastern side of the Ethiopian Plateau. The curves are fromthe following stations: NAZA (MER), JIMA (western sideof the Ethiopian Plateau), and GOBA (eastern side of theEthiopian Plateau). The group velocity curve for theMozambique Belt in Tanzania is taken from Pasyanos[2005] for station KIBE (see location in Nyblade et al.,1996), and is shown for comparison.


4 of 14

B08305

the crust and uppermost mantle, and 0.2 km/s in the lowerpart of the upper mantle, and uncertainties in the depth ofdiscontinuities to be about 2–3 km in the crust anduppermost mantle.

3. Results

[22] Results from the joint inversion for all stations areshown in Figure 4, and in Figure 5 we show results forselected stations along two E-W profiles highlighting thestructure above 100 km depth. A summary of our resultsplus a comparison with results from other studies follows.

3.1. Crust

[23] The crustal thickness beneath both the eastern andwestern sides of the Ethiopian Plateau ranges between 35and 40 km, and for the MER, the crustal thickness rangesbetween about 30 and 35 km. For the two stations in Afar

(ATD and TEND), the crustal thickness is about 25 km.These crustal thickness estimates agree closely with thestructure obtained by Dugda et al. [2005], Stuart et al.[2006], and Dugda and Nyblade [2006] using the H-kreceiver function stacking technique of Zhu and Kanamori[2000] (Figure 5), where H = Moho depth and k = Vp/Vs.The H-k stacking technique provides robust estimates ofcrustal thickness and Poisson’s ratio by incorporating theP-to-S converted phase from the Moho and two laterarriving crustal reverberations in a stacking procedure.[24] The average crustal structure (e.g., Moho depth and

average crustal velocity) obtained from the joint inversionalso agrees closely with the structure obtained from seismicrefraction profiles in places where seismic stations arewithin about 50 km of the refraction profiles [Makris andGinzburg, 1987; Mackenzie et al., 2005; Maguire et al.,2006] (Figure 5). Only a few stations lie closer than about

Figure 3. Figure illustrating, for station ARBA, the procedure used to determine structure below 190km depth. The four columns show different models tested for structure below 190 km depth usingvelocities 15%, 10%, 7%, and 5% less than PREM [Dziewonski and Anderson, 1981]. (a) Observed(black line) and predicted (gray line) group velocity dispersion curves. The highlighted part in thedispersion curves shows the fitting of the longest period group velocities for the four different modelstested. (b) Observed (black line) and predicted (gray line) receiver functions. See text for explanation ofthe different receiver functions. (c) The shear wave velocity models obtained from the joint inversion(black line) and the PREM shear wave velocity model (gray line) for reference. The 10% less than PREMmodel for shear velocities below 190 km depth gives the best fit to the longest period group velocities.The difference in the fit of the receiver functions is not significant between the models.


5 of 14

B08305

Figure 4. Results of the joint inversion. For each station, three panels are shown: receiver functions(top), dispersion curves (middle), and shear wave velocity models (bottom).


6 of 14

B08305

Figure 4. (continued)


7 of 14

B08305

Figure 4. (continued)

Figure 5. Graph showing the correlation between the Moho depth estimates from the joint inversiontechnique and previous studies. A comparison with Moho depths from refraction profiles is only madewhere seismic stations are within about 50 km of a refraction profile. See text for further explanation.


8 of 14

B08305

20 km from the refraction profiles and thus a more detailedcomparison of crustal structure, such as intercrustal layer-ing, between the ensemble of our results and the refractionprofiles is not warranted.

3.2. Upper Mantle

[25] For the uppermost mantle, if we assume a Poisson’sratio between 0.28 and 0.30 and compare our S-wavevelocity structure to the P-wave velocity structure fromseveral refraction profiles [Makris and Ginzburg, 1987;Mackenzie et al., 2005; Maguire et al., 2006], we also findgood agreement. The average Sn velocity beneath theEthiopian Plateau is about 4.3 km/s, while for the MER itis about 4.0 to 4.1 km/s, corresponding to Pn velocitiesfound by Mackenzie et al. [2005] of �7.9–8.0 km/s for theEthiopian Plateau and �7.6–7.7 km/s for the MER. Theassumed Poisson’s ratios are not unreasonable consideringthat there may be partial melts in some areas of the WesternPlateau [Whaler and Hautot, 2006] and that very highPoisson’s ratios are characteristic of the Main EthiopianRift and Afar.[26] Figure 6 shows two profiles, one crossing the west-

ern side of the Ethiopian Plateau, the MER, and the easternside of the Ethiopian Plateau (Profile A-A0), and the othercrossing the western side of the Ethiopian Plateau, theMER, and the Afar (Profile B-B0). It can be seen from thisfigure that the lithosphere beneath the plateau has a thin‘‘lid’’ structure that extends from the Moho to about 60 or80 km depth, while the lid beneath the MER and Afar iseither much thinner or nonexistent. The maximum shearwave velocity in the upper mantle lid is about 4.3 km/sfor both sides of the Ethiopian Plateau, and is underlainby a low velocity zone characterized by velocities of about

4.1 km/s. When there is a lid beneath the MER/Afar,the maximum velocity in the lid reaches a value of about4.1–4.2 km/s, with velocities of about 4.0 km/s under thelid.[27] In the tomographic image of Bastow et al. [2005], at

a depth of 75 km there is a maximum difference of about5% in shear velocities beneath the Ethiopian Plateau and theMER. The difference in shear wave velocities beneath thetwo sides of the Ethiopian Plateau and the MER at 75 kmdepth in our models is �0.2 km/s, or �5%, consistent withthe results of Bastow et al. [2005].

4. Discussion

[28] The main finding of this study is a thin seismic lid(between the Moho and �60 to �80 km depth) under theEthiopian Plateau. We also find essentially little or no lidunder the MER and Afar, in accord with the results of theEAGLE project [Bastow et al., 2005; Mackenzie et al.,2005; Maguire et al., 2006]. As is common, we interpret theseismic lid to represent the lithospheric mantle. To examinethe extent to which the Ethiopian lithosphere has beenperturbed, we use the velocity structure from relativelyunperturbed Mozambique Belt in northeastern Tanzaniafor comparison, as discussed previously.[29] In Figure 7, we compare the upper mantle structure

obtained in this study for each station with the resultobtained by Julia et al. [2005] for Mozambique Belt uppermantle structure beneath northeastern Tanzania, alsoobtained using a joint inversion of receiver functions andsurface wave dispersion measurements. By comparingthe profiles, it can be seen that the lithosphere beneathEthiopia and Afar, including Djibouti, relative to northeastern

Figure 6. Two profiles of shear wave velocity along lines A-A0 and B-B0 (shown in Figure 1) used tohighlight the structure of the seismic lid and its variation in the MER/Afar and the Ethiopian Plateau.Only the top 100 km of the inversion results are displayed. From the profiles, it can be seen that theseismic lid (shaded) beneath the Ethiopian Plateau is thin and that virtually no lid exists beneath the Afar/MER. Station names are indicated at the top of each velocity model and the lid is taken to be any regionwith velocities > 4.0–4.1 km/s.


9 of 14

B08305

Tanzania, has been pervasively modified. The maximumvelocities beneath the Ethiopian Plateau are 0.3 km/s to0.4 km/s less than under the Mozambique Belt in north-eastern Tanzania to depths of �100–120 km. The differencein the lid structure is also illustrated in Figure 8, whichshows the average shear wave velocity structure beneath theEthiopian Plateau and the Mozambique Belt. The lid underthe Ethiopian Plateau has an average maximum velocity ofabout 4.3 km/s, whereas the lid under the MozambiqueBelt in Tanzania has an average maximum velocity ofabout 4.6 km/s [Julia et al., 2005; Weeraratne et al.,2003]. The lid structure under the MER and Afar is evenmore perturbed.[30] What is the cause of this large perturbation in

lithospheric mantle structure beneath Ethiopia? Possiblecauses of seismic velocity variations in the mantle includechanges in temperature, grain size, fluid content (melt),

dislocations and composition, but temperature variations aremost often invoked [Faul and Jackson, 2005]. The deriv-ative of the shear wave velocity with respect to temperaturein the upper mantle depends mainly on grain size and theparticular temperature at which this derivative is considered.The derivative also varies with seismic frequency. Jacksonet al. [2002] and Faul and Jackson [2005] report a deriv-ative for shear wave velocity with respect to temperature of�1.2 m/s/K for an upper mantle grain size of 10 mm at anaverage temperature of 1250�C. Thus, to reduce the shearwave velocity in the mantle lithosphere by �0.3 km/s(Figure 8), rock temperatures must increase by �250 K.[31] Because of the strong indications from previous

geochemical and geophysical studies for thermally per-turbed upper mantle structure beneath Ethiopia, as reviewedin section 1.0, to explain the modified shear velocitystructure of the Ethiopian mantle lithosphere we considerthe effect of a thermal plume instantaneously eroding the

Figure 7. Shear wave velocity models for Ethiopia and Djibouti stations (black lines) compared to ashear wave velocity model for the Mozambique Belt lithosphere in Tanzania (gray line). Station namesare indicated at the top of each velocity model. The model for the Mozambique Belt in Tanzania is forstation KIBE [Julia et al., 2005].


10 of 14

B08305

bottom of the lithosphere at the time of the flood basaltvolcanism (c. 30 Ma) and then remaining under the thinnedlithosphere to the present day. While instantaneous thinningof the mantle lithosphere does not capture the complexity ofgeodynamic (hot spot) processes in the region, it nonethe-less provides a reasonable approach for assessing first-orderchanges in the thermal structure of the mantle lithosphereresulting from a plume impinging on the lithosphere.[32] We model the thermal effect of the plume on the

instantaneously-thinned lithosphere by considering the tran-sient thermal response within a slab whose lower boundarytemperature is increased by DTL, according to the equation

DT z; tð Þ ¼ DTLz=L þ2 =p

X1n¼1

1

ncos npð Þ

(

: sinnpzL

h iexp

�n2p2atL2

� ��ð1Þ

where DT (z, t) is the anomalous temperature at depth zafter time t, L is slab thickness, and a is thermal diffusivity[Carslaw and Jaeger, 1959]. Thus, a specific type of plume(i.e., single plume, multiple plumes, superplume) is notmodeled, but rather the consequence of simply thinning andheating the lithosphere by any kind of plume. For a thermaldiffusivity of 32 km2/m.y. [Turcotte and Schubert, 2002],and different thicknesses of the instantaneously thinnedlithosphere, temperatures within the lithosphere at depths of30 and 60 km are shown in Figure 9. If the Ethiopianlithosphere was eroded by the plume from some initialthickness to a thickness of 70 or 80 km, and if the plumematerial sitting beneath the lithosphere remained some 300to 400 K hotter than the surrounding mantle [Schilling,1991; Wyllie, 1988; Mckenzie and Bickle, 1988], then thetemperature at �60 km depth in the lithosphere would

increase by �250 K or more after 30 Ma (Figure 9),sufficient to account for the reduction of the maximumshear wave velocity in the lid beneath the EthiopianPlateau of about 0.3 km/s. From this analysis, we concludethat thermal alteration of the Ethiopian lithospheric mantlecan, to a first order, account for the low shear wavevelocities in the lithosphere, provided that the lithospherewas thinned to a thickness of �70–80 km at c. 30 Ma andthat the warm mantle material remained under the thinnedlithosphere since that time.[33] The amount of uplift of the Ethiopian Plateau due to

the thermally perturbed mantle lithosphere can be estimatedfrom alDT, where a is the coefficient of thermal expansion,l is the thickness of rock being heated, and DT is thetemperature perturbation. If we consider the altered mantlelithosphere to be 40 km thick with a temperature perturba-tion of 250 K, then �0.30 km of uplift is created isostat-ically, assuming a = 3 10�5 K�1 [Turcotte and Schubert,2002]. And if we assume that the lithosphere was 120 kmthick prior to 30 Ma, an additional 0.36 km of uplift wouldresult from the warm plume material with a DT = 300 Kreplacing the �40 km thick bottom section of the litho-sphere. Thus, a significant portion (0.66 km) of theobserved plateau uplift in Ethiopia of 1–1.5 km (Figure 1)can be attributed to the thermal alteration of the lithosphere.[34] To explain the full 1–1.5 km of uplift across the

Ethiopian Plateau, other sources of buoyancy may berequired, which could come from thermally perturbedstructure deeper in the upper mantle and/or modificationsto crustal structure. Regional and global tomographicmodels [e.g., Benoit et al., 2006a, 2006b; Bastow et al.,2005; Debayle et al., 2001; Ritsema et al., 1999; Grand,2002] show low velocity anomalies beneath Ethiopia andAfar in the upper mantle extending to depths of �400 kmand these models are consistent with the lower-than-PREM

Figure 8. Average lithospheric structure of the EthiopianPlateau (black line) and the Mozambique Belt underTanzania (grey line). A pronounced reduction in the shearwave velocity of the seismic lid under Ethiopia is seencompared to the Mozambique Belt under Tanzania. Theaverage lithospheric structure for Ethiopia is obtained byaveraging shear velocities from both the eastern and westernsides of the Ethiopian Plateau, as they are similar. Thevelocity profile for the Mozambique Belt in Tanzania is forstation KIBE [Julia et al., 2005].

Figure 9. Graph showing the thermal response of thelithosphere to a plume sitting beneath the lithosphere atdepths between 70 km and 120 km since 30 Ma. The solidcurves show temperature responses at 60 and 30 km depthsfor plume material that is 300 K above ambient mantletemperatures. The dashed lines show temperature responsesat 30 and 60 km depths for plume material that is 400 Kabove ambient mantle temperatures.


11 of 14

B08305

velocities in our models (Figures 4a, 4b, and 4c) extendingto �400 km depth. All of the models suggest that the uppermantle beneath Ethiopia is thermally perturbed to greatdepths, providing possible sources of thermal buoyancythat could play a role in creating the Ethiopian Plateau,both isostatically and dynamically [e.g., Daradich et al.,2003]. In addition, crustal structure across the EthiopianPlateau is variable, with underplating and thermal perturba-tions providing yet other possible sources of buoyancy[Mackenzie et al., 2005; Maguire et al., 2006; Stuart et al.,2006; Whaler and Hautot, 2006].[35] Although the focus of this discussion has been on the

lithosphere beneath the Ethiopian Plateau, the thin or non-existent mantle lithosphere under the MER and Afar is alsonoteworthy. The seismic velocity structure of the uppermostmantle under the MER and Afar is similar to that foundbeneath the Kenya Rift where it has been commonlyattributed to extension of the lithosphere together withupwelling warm plume material that has thermally modifiedthe thinned lithosphere [Fuchs et al., 1997; Prodehl et al.,1994, and references therein]. Such a combined plume-riftexplanation can also account for the nature of the uppermostmantle structure under the MER and Afar, as suggested bymany authors [e.g., Maguire et al., 2006; Stuart et al., 2006;Mackenzie et al., 2005; Bastow et al., 2005; Makris andGinzburg, 1987].[36] Finally, the thinned and thermally perturbed mantle

lithosphere under the MER, Afar, and Ethiopian Plateau hasimportant consequences for interpreting seismic estimatesof mantle anisotropy derived from shear wave splitting.Shear wave splitting studies for Ethiopia using teleseismicdata, showed delay times of �0.5 to 1.7 s for the MER,Afar, and the Ethiopian Plateau, with fast-polarizationdirections generally parallel to the orientation of the riftsystem [Gashawbeza et al., 2004; Ayele et al., 2004;Kendall et al., 2005, 2006]. Different interpretations for thesource of the anisotropy have been published. Gashabezaet al. [2004] suggest that the anisotropy could be controlledby fossil anisotropy in the Mozambique Belt lithosphere,while the studies by Ayele et al. [2004] and Kendall et al.[2005, 2006] attribute the anisotropy to melt-filled micro-cracks and dikes in the lithosphere. And Keir et al. [2005]provide for a similar interpretation using estimates ofseismic anisotropy obtained from regional seismicity. Theaverage SKS splitting time is about �1.0 s for the region,comparable to the splitting times observed in and aroundthe Kenya Rift [Walker et al., 2004]. Walker et al. [2004]show that for the hot, thin mantle beneath and surroundingthe Kenya Rift, fossil anisotropy in the lithosphere is notlikely a dominant source of anisotropy. A similar argumentcan be used for Ethiopia, thus calling into question inter-pretations that attribute the shear wave splitting observa-tions to fossil anisotropy in the lithosphere.

5. Summary

[37] The seismic velocity structure of the crust and uppermantle beneath Ethiopia and Djibouti has been investigatedin this study using a joint inversion of receiver functionsand Rayleigh wave group velocities. Crustal structureobtained from the joint inversion is similar to crustalstructure reported in previous studies, with crustal thick-

nesses of 35 to 44 km beneath the Ethiopian Plateau, and 25to 35 km beneath the Main Ethiopian Rift and the Afar. Thelithospheric mantle beneath the Ethiopian Plateau has amaximum shear wave velocity of about 4.3 km/s andextends to a depth of �70–80 km. Beneath the MER andAfar, the lithospheric mantle has a maximum shear wavevelocity of 4.1–4.2 km/s and extends to a depth of at most50 km.[38] Prior to the Cenozoic hot spot tectonism, Ethiopia

and Djibouti were underlain with unperturbed MozambiqueBelt lithosphere. In comparison to Mozambique Belt litho-sphere in Tanzania along the edge of the East AfricanPlateau where the lid extends to depths of �100–125 kmand has a maximum shear velocity of 4.6 km/s, the mantlelithosphere under the Ethiopian Plateau appears to havebeen thinned by �30–50 km and the maximum shear wavevelocity reduced by �0.3 km/s. The results of a 1Dconductive thermal model suggest that the shear velocitystructure of the Ethiopian lithosphere can be explained by amantle plume, if the plume instantaneously thinned thelithosphere by at least 30–50 km at the time of the Afarflood basalt volcanism (c. 30 Ma), and if the warm plumematerial has remained beneath the lithosphere since then.About 45–65% of the 1–1.5 km of plateau uplift inEthiopia can be attributed to the thermally perturbed litho-spheric mantle structure.

[39] Acknowledgments. We would like to extend our thanks to DrLaike-Mariam Asfaw and Dr Atalay Ayele, as well as the technical staff ofthe Geophysical Observatory of Addis Ababa University, for their help withthe Ethiopian broadband seismic experiment. We also thank MichaelPasyanos and Cindy Ebinger for constructive reviews. This research hasbeen funded by the National Science Foundation (grants EAR 993093,0003424, and 0505812).

ReferencesAmmon, C. J., G. E. Randall, and G. Zandt (1990), On the nonuniquenessof receiver function inversions, J. Geophys. Res., 95, 15,303–15,318.

Ayalew, D., P. Barbey, B. Marty, L. Reisberg, G. Yirgu, and R. Pik (2002),Source, genesis, and timing of giant ignimbrite deposits associated withEthiopian continental flood basalts, Geochimica et Cosmochimica Acta,66, 8, 1429–1448.

Ayalew, D., C. J. Ebinger, E. Bourdon, E. Wolfenden, G. Yirgu, andN. Grassineau (2006), Temporal compositional variation of syn-rift rhyo-lites along the western margin of the southern Red Sea and northern mainEthiopian Rift, Geological Society Special Publications, 259, 121–130.

Ayele, A., G. Stuart, and J. M. Kendall (2004), Insights into rifting fromshear splitting and receiver functions; an example from Ethiopia, Geo-physical Journal International, 157, 1, 354–362.

Baker, J. A., L. W. Snee, and M. A. Menzies (1996), A brief Oligoceneperiod of flood volcanism in Yemen; implications for the duration andrate of continental flood volcanism at the Afro-Arabian triple junction,Earth and Planetary Science Letters, 138, 1–4, 39–55.

Bastow, I. D., G. W. Stuart, J.-M. Kendall, and C. J. Ebinger (2005), Upper-mantle seismic structure in a region of incipient continental breakup;northern Ethiopian Rift, Geophysical Journal International, 162(2),479–493.

Bellahsen, N., C. Faccenna, F. Funiciello, J. M. Daniel, and L. Jolivet(2003), Why did Arabia separate from Africa?, Earth and PlanetaryScience Letters, 216, 365–381.

Benoit, M. H. (2005), The upper mantle seismic velocity structure beneaththe Arabian Shield and East Africa, PhD Thesis, Pennsylvania StateUniversity at University Park, PA, USA.

Benoit, M. B., A. A. Nyblade, and M. E. Pasyanos (2006a), Crustal thinningbetween the Ethiopian and East African Plateaus from modeling Rayleighwave dispersion, Geophys. Res. Lett., 33, L13301, doi:10.1029/2006GL025687.

Benoit, M. H., A. A. Nyblade, T. J. Owens, and G. Stuart (2006b), Mantletransition zone structure and upper mantle S velocity variations beneathEthiopia: Evidence for a broad, deep-seated thermal anomaly, Geochem.Geophys. Geosyst., 7, Q11013, doi:10.1029/2006GC001398.


12 of 14

B08305

Benoit, M. H., A. A. Nyblade, and J. C. VanDecar (2006c), Upper mantleP-wave speed variations beneath Ethiopia and the origin of the AfarHotspot, Geology (Boulder), 34(5), 329–332.

Berhe, S. (1990), Ophiolites in northeast and East Africa: Implications forProterozoic crustal growth, J. Geol. Soc. Lon., 147, 41–57.

Berhe, S. M., B. Desta, M. Nicoletti, and M. Teferra (1987), Geology,geochronology and geodynamic implications of the Cenozoic magmaticprovince in W and SE Ethiopia, J. Geological Society, London, 144,213–226.

Burke, K., and A. M. C. Sengor (1986), Tectonic escape in the evolution ofthe continental crust, Geodyn. Ser., 14, 41–53, AGU.

Carslaw, H. S., and J. C. Jaeger (1959), Conduction of heat in solids,Clarendon Press, 510 p.

Casey, M., C. J. Ebinger, D. Keir, R. Gloaguen, and F. Mohamed (2006),Strain accommodation in transitional rifts; extension by magma intrusionand faulting, in Ethiopian Rift magmatic segments, edited by G. Yirgu,C. J. Ebinger, and P. K. H. Maguire, Geological Society Special Publica-tions, vol. 259, pp. 143-163.

Cassidy, J. F. (1992), Numerical experiments in broadband receiver func-tion analysis, Bulletin of the Seismological Society of America, 82(3),1453–1474.

Chernet, T., W. K. Hart, J. L. Aronson, and R. C. Walter (1998), New ageconstraints on the timing of volcanism and tectonism in the northernMain Ethiopian Rift-southern Afar transition zone (Ethiopia), Journalof Volcanology and Geothermal Research, 80, 3–4, 267–280.

Coulie, E., X. Quidelleur, P. Y. Gillot, Courtillot Vincent, J. C. Lefevre, andS. Chiesa (2003), Comparative K-Ar and Ar/Ar dating of Ethiopian andYemenite Oligocene volcanism; implications for timing and duration ofthe Ethiopian traps, Earth and Planetary Science Letters, 206, 3 –4,477–492.

Courtillot, V., C. Jaupart, I. Manighetti, P. Tapponnier, and J. Besse (1999),On Causal links between flood basalts and continental breakup, EarthPlanet. Sci. Lett., 166, 177–195.

Daradich, A., J. X. Mitrovica, R. N. Pysklywec, S. D. Willett, and A. M.Forte (2003), Alessandro M, Mantle flow, dynamic topography, and rift-flank uplift of Arabia, Geology (Boulder), 31(10), 901–904.

Davidson, A., and D. C. Rex (1980), Age of volcanism and rifting insouthwestern Ethiopia, Nature, 283, 657–658.

Debayle, E., J.-J. Leveque, and M. Cara (2001), Seismic evidence for adeeply rooted low-velocity anomaly in the upper mantle beneath thenortheastern Afro/Arabian continent, Earth and Planetary Science Let-ters, 193(3–4), 423–436.

Du, Z. J., and G. R. Fougler (1999), The crustal structure beneath thenorthwest fjords, Iceland, from receiver functions and surface waves,Geophys. J. Int., 139, 2, 419–432.

Dugda, M. T., and A. A. Nyblade (2006), New constraints on crustalstructure in eastern Afar from the analysis of receiver functions and sur-face wave dispersion in Djibouti, Geological Society Special Publica-tions, 259, 239–251.

Dugda, M. T., A. A. Nyblade, J. Julia, C. A. Langston, C. J. Ammon,and S. Simiyu (2005), Crustal Structure in Ethiopia and Kenya fromReceiver Function Analysis: Implications for Rift Development ineastern Africa, J. Geophys. Res., 110, B01303, doi:10.1029/2004JB003065.

Dziewonski, A. M., and D. L. Anderson (1981), Preliminary ReferenceEarth Model (PREM), Physics of the Earth and Planetary Interiors,25, 4, 297–356.

Ebinger, C. J., and M. Casey (2001), Continental breakup in magmaticprovinces; an Ethiopian example, Geology (Boulder), 29, 6, 527–530.

Ebinger, C. J., and N. H. Sleep (1998), Cenozoic magmatism throughoutEast Africa resulting from impact of a single plume, Nature (London),395, 6704, 788–791.

Ebinger, C. J., T. Yemane, G. Woldegabriel, J. L. Aronson, and R. C. Walter(1993), Late Eocene-Recent volcanism and faulting in the southern mainEthiopian rift, J. Geological Society, London, 150, 99–108.

Faul, U. H., and I. Jackson (2005), The seismological signature of tempera-ture and grain size variations in the upper mantle, Earth and PlanetaryScience Letters, 234, 1–2, 119–134.

Fuchs, K., R. Altherr, B. Muller, and C. Prodehl (Eds.) (1997), Structureand Dynamic Processes in the Lithosphere of the Afro-Arabian RiftSystem, Special Issue, Tectonophysics.

Furman, T., J. G. Bryce, J. Karson, and A. Iotti (2004), East African RiftSystem (EARS) plume structure; insights from Quaternary mafic lavas ofTurkana, Kenya, Journal of Petrology, 45, 5, 1069–1088.

Furman, T., J. Bryce, T. Rooney, B. Hanan, G. Yirgu, and D. Ayalew(2006), Heads and tails; 30 million years of the Afar plume, GeologicalSociety Special Publications, 259, 95–119.

Gashawbeza, E. M., S. L. Klemperer, A. A. Nyblade, K. T. Walker, andK. M. Keranen (2004), Shear-wave splitting in Ethiopia; Precambrianmantle anisotropy locally modified by Neogene rifting, GeophysicalResearch Letters, 31, 18, 4.

George, R., N. Rogers, and S. Kelley (1998), Earliest magmatism inEthiopia: Evidence for two mantle plumes in one flood basalt province,Geology, 26, 923–926.

Gibson, I. L. (1967), Preliminary account of the volcanic geology ofFantale, Shoa, Bulletin of the Geophysical Observatory, 10, 59–67.

Grand, S. P. (2002), Mantle shear-wave tomography and the fate of sub-ducted slabs, Philosophical Transactions - Royal Society. Mathematical,Physical and Engineering Sciences, 360(1800), 2475–2491.

Hempton, M. R. (1987), Constraints on Arabian plate motion and exten-sional history of the Red sea, Tectonics, 6, 687–705.

Hofmann, C., V. Courtillot, G. Feraud, P. Rochette, G. Yirgu, E. Ketefo, andR. Pik (1997), Timing of the Ethiopian flood basalt event and implica-tions for plume birth and global change, Nature, 389, 838–841.

Jackson, I., J. D. Fitz, J. D. Fitz Gerald, U. H. Faul, and B. H. Tan (2002),Grain-size-sensitive seismic wave attenuation in polycrystalline olivine,J. Geophys. Res., 107(B12), 2360, doi:10.1029/2001JB001225.

Julia, J., C. J. Ammon, R. B. Herrmann, and A. M. Correig (2000), Jointinversion of receiver functions and surface wave dispersion observations,Geophys. J. Int., 143, 99–112.

Julia, J., C. J. Ammon, and R. B. Herrmann (2003), Lithospheric structureof the Arabian Shield from the joint inversion of receiver functions andsurface-wave group velocities, Tectonophysics, 371(1–4), 1–21.

Julia, J., C. J. Ammon, and A. A. Nyblade (2005), Evidence for mafic lowercrust in Tanzania, East Africa, from joint inversion of receiver functionsand Rayleigh wave dispersion velocities, Geophys. J. Int., 162, 2, 555–569.

Kendall, J.-M., G. W. Stuart, C. J. Ebinger, I. D. Bastow, and D. Keir(2005), Magma-assisted rifting in Ethiopia, Nature, 433, 146–148.

Kendall, J. M., S. Pilidou, D. Keir, I. D. Bastow, G. W. Stuart, and A. Ayele(2006), Mantle upwellings, melt migration and the rifting of Africa;insights from seismic anisotropy, edited by G. Yirgu, C. J. Ebinger,and P. K. H. Maguire, Geological Society Special Publications, 259,55-72.

Keir, D., J. M. Kendall, C. J. Ebinger, and G. W. Stuart (2005), Variations inlate syn-rift melt alignment inferred from shear-wave splitting in crustalearthquakes beneath the Ethiopian Rift, Geophysical Research Letters,32, 23, 4.

Kieffer, B., et al. (2004), Flood and shield basalts from Ethiopia; magmasfrom the Africa Superswell, Journal of Petrology, 45, 4, 793–834.

Kroner, A., R. Greiling, and T. Reischman (1987), Pan-African crustalevolution in the Nubian segment of Northeast Africa, in ProterozoicLithospheic Evolution, edited by A. Kroner, Geodyn. Ser. 17, AGU,235-257.

Langston, C. A. (1979), Structure under Mount Rainier, Washington,inferred from teleseismic body waves, J. Geophys. Res., 84, 4749–4762.

Larson, E. W. F., and G. Ekstrom (2001), Global Models of Surface WaveGroup Velocity, Pure Appl. Geophys., 158(8), 1377–1400.

Last, R. J., A. A. Nyblade, C. A. Langston, and T. J. Owens (1997), Crustalstructure of the East African Plateau from receiver functions and Ray-leigh wave phase velocities, J. Geophys. Res., 102, 24,469–24,483.

Ligorria, J. P., and C. Ammon (1999), Iterative deconvolution and receiver-function estimation, Bull. Seism. Soc. Am., 89, 1395–1400.

Mackenzie, G. D., H. Thybo, and P. K. H. Maguire (2005), Crustal velocitystructure across the main Ethiopian Rift; results from two-dimensionalwide-angle seismic modelling, Geophysical Journal International, 162,3, 994–1006.

Maguire, P. K. H., et al. (2006), Crustal structure of the northern MainEthiopian Rift from the EAGLE controlled-source survey; a snapshotof incipient lithospheric break-up, edited by G. Yirgu, C. J. Ebinger,and P. K. H. Maguire, Geological Society Special Publications, 259,269–291.

Makris, J., and A. Ginzburg (1987), The Afar Depression: transition bet-ween continental rifting and sea floor spreading, Tectonophysics, 141,199–214.

Manighetti, I., P. Tapponnier, V. Courtillot, S. Gruszow, and P.-Y. Gillot(1997), Propagation of rifting along the Arabia-Somalia plate boundary;the Gulf of Aden and Tadjoura, J. of Geophys. Res., 102(B2), 2681–2710.

McKenzie, D., and M. J. Bickle (1988), The volume and composition ofmelt generated by extension of the lithosphere, Journal of Petrology, 29,3, 625–679.

Mohr, P. (1983), Ethiopian flood basalt province, Nature, 303, 577–584.Mohr, P., and B. Zanettin (1988), The Ethiopian flood basalt province, inContinental Flood Basalts, edited by J. D. MacDougall, Kluwer Aca-demic Pub., 63-110.

Nyblade, A. A., C. Birt, C. A. Langston, T. J. Owens, and R. J. Last (1996),Seismic experiment reveals rifting of craton in Tanzania, Eos, Transac-tions, American Geophysical Union, 77, 51, 517, 521.

Nyblade, A. A., and C. A. Langston (2002), Broadband seismic experi-ments probe the East African rift, EOS Trans. AGU, 83, 405–408.


13 of 14

B08305

Ozalaybey, S., M. K. Savage, A. F. Sheehan, J. N. Louie, and J. N. Brune(1997), Shear-wave velocity structure in the northern Basin and RangeProvince from the combined analysis of receiver functions and surfacewaves, Bulletin of the Seismological Society of America, 87, 1, 183–199.

Pasyanos, M. E., W. R. Walter, and S. E. Hazler (2001), A surface wavedispersion study of the Middle East and North Africa for monitoring theComprehensive Nuclear-Test-Band Treaty, Pure and Applied Geophysics,158, 8, 1445–1474.

Pasyanos, M. E. (2005), A variable resolution surface wave dispersionstudy of Eurasia, North Africa, and surrounding regions, J. of Geophys.Res., 110(B12), 22.

Pallister, J. S. (1987), Magmatic history of the Red Sea rifting: Perspectivefrom the Central Saudi Arabian Coastal Plane, Geol. Soc. Am. Bull., 98,400–417.

Pik, R., B. Marthy, J. Carignan, and J. Lave (2003), Stability of the UpperNile Drainage Network (Ethiopia) deduced from (U-Th)/He thermochro-nometry: implications for uplift and erosion of the Afar plume dome,Earth and Planetary Science Letters, 215, 73–88.

Prodehl, C., G. R. Keller, and M. A. Khan (1994), Crustal and upper mantlestructure of the Kenya Rift, Tectonophysics, 236, 1–483.

Ritsema, J., H.-J. van Heijst, and J. H. Woodhouse (1999), Complex shearwave velocity structure imaged beneath Africa and Iceland, Science, 286,5446, pp. 1925–1928.

Rogers, N. W. (2006), Basaltic magmatism and the geodynamics of the EastAfrican Rift System, in The Afar Volcanic Province within the EastAfrican Rift System, SP259, edited by G. Yirgu, C. J. Ebinger, andP. K. H. Maguire, Geological Society Special Publications, pp. 336.

Rooney, T. O., T. Furman, G. Yirgu, and D. Ayalew (2005), Structure of theEthiopian lithosphere; xenolith evidence in the main Ethiopian Rift, Geo-chimica et Cosmochimica Acta, 69, 15, pp. 3889–3910.

Schilling, J.-G. (1991), Fluxes and excess temperatures of mantle plumesinferred from their interaction with migrating mid-ocean ridges, Nature(London), 352, 6334, pp. 397–403.

Shackelton, R. M. (1986), Precambrian collision tectonics in Africa, inCollision tectonics, edited by M. P. Coward and A. C. Reis, Geol. Soc.Spec. Pup., 19, 329-349.

Stuart, G. W., I. D. Bastow, and C. J. Ebinger (2006), Crustal Structure ofthe northern Main Ethiopian Rift from Receiver Function Studies, in TheAfar Volcanic Province within the East African Rift System, SP259, editedby G. Yirgu, C. J. Ebinger, and P. K. H. Maguire, Geological SocietySpecial Publications, pp. 336.

Tkalcic, H., M. E. Pasyanos, A. J. Rodgers, R. Gok, W. R. Walter, andA. Al-Amri (2006), A multistep approach for joint modeling of surfacewave dispersion and teleseismic receiver functions: Implications forlithospheric structure of the Arabian Peninsula, J. Geophys. Res., 111,B11311, doi:10.1029/2005JB004130.

Vail, J. R. (1985), Pan-African (late Precambrian) tectonic terrains and thereconstruction of the Arabian-Nubian Shield, Geology, 13, 839–849.

Vail, J. R. (1988), Tectonics and evolution of the Proterozoic basement ofNortheastern Africa, in The Pan-African Belt of Northeast Africa andadjacent areas, edited by S. El-Gaby and R. Grieling, Friedr. Viewegand Sohn, 195-226.

VanDecar, J. (1991), Upper-mantle structure of the Cascadia subductionzone from nonlinear teleseismic traveltime inversion, PhD Thesis, Uni-versity of Washington, Seattle, WA, USA.

Weeraratne, D. S., D. W. Forsyth, K. M. Fischer, and A. A. Nyblade (2003),Evidence for an upper mantle plume beneath the Tanzanian Craton fromRayleigh wave tomography, J. of Geophys. Res., B, Solid Earth andPlanets, 108, 9,17 pp.

Whaler, K. A., and S. Hautot (2006), The electrical resistivity structure ofthe crust beneath the northern Main Ethiopian Rift, edited by G. Yirgu,C. J. Ebinger, and P. K. H. Maguire, Geological Society SpecialPublications, 259, pp. 293–305.

WoldeGabriel, G., J. L. Aronson, and R. C. Walter (1990), Geology andgeochronology, and rift basin development in the central sector of themain Ethiopian rift, Geol. Soc. of Am. Bull., 102, 439–458.

WoldeGabriel, G., R. C. Walter, W. K. Hart, S. A. Mertzman, and J. L.Aronson (1999), Temporal relations and geochemical features of felsicvolcanism in the central sector of the Main Ethiopian Rift, From: Bocca-letti, Mario; Peccerillo, Angelo, Acta Vulcanologica, 11, 1, pp. 53–67.

Wolfenden, E., C. Ebinger, G. Yirgu, A. Deino, and D. Ayalew (2004),Evolution of the northern main Ethiopian Rift; birth of a triple junction,Earth and Planetary Science Letters, 224, 1–2, pp 213–228.

Wolfenden, E., C. Ebinger, G. Yirgu, P. R. Renne, and S. P. Kelley (2005),Evolution of a volcanic rifted margin; southern Red Sea, Ethiopia, Geo-logical Society of America Bulletin, 117, 7–8, 846–864.

Wyllie, P. J. (1988), Solidus Curves, mantle plumes and magma generationbeneath Hawaii, J. Geophys. Res., 93, 4171–4181.

Yirgu, G., C. J. Ebinger, and P. K. H. Maguire (Eds.) (2006), The AfarVolcanic Province within the East African Rift System, SP259, Geologi-cal Society Special Publications, pp. 336.

Zanettin, B., E. Justin-Visentin, M. Nicoletti, and E. M. Piccirllo (1980),Correlations among Ethiopian volcanic formations with special refer-ences to the chronological and stratigraphical problems of the ‘‘TrapSeries’’, Atti Convegni Lincei, 47, 231–252.

��M. T. Dugda and A. A. Nyblade, Department of Geosciences,

Pennsylvania State University, University Park, PA, USA. ([email protected])J. Julia, Department of Geological Sciences, University of South

Carolina, Columbia, SC, USA.


14 of 14

B08305

Thin Lithosphere Beneath the Ethiopian Plateau Revealed by a Joint Inversion of Rayleigh Wave Group Velocities and Receiver Functions

Documents