Keir_thesis Seismicity of the Northern Ethiopian Rift

i

Strain Accommodation by Magmatism and Faulting as Rifting Proceeds to Breakup: Seismicity of the Northern Ethiopian Rift

Derek Keir

Submitted in accordance with the requirements for the degree of Ph.D.

Royal Holloway University of London

Department of Geology

March 2006

The candidate confirms that the work submitted is his own and that appropriate credit

has been given where reference has been made to the work of others

ii

Acknowledgements The help and support of many people made this research possible. I first thank my supervisors

Cindy Ebinger, Graham Stuart and advisor Dave Waltham for giving me the opportunity to

participate in the unique EAGLE project. Cindy’s energy, enthusiasm and interest in my

research encouraged me throughout the Ph.D, and Graham’s sound, practical advice helped

extract more from the EAGLE dataset than I could ever have imagined. Thanks also to Mike

Kendall and Andy Jackson who facilitated the S-wave splitting and local earthquake magnitude

scale studies. Peter Maguire and all participants of the EAGLE project provided useful

suggestions and ideas for my research.

I am much indebted to Alex Brisbourne at SEIS-UK in Leicester for his clear and concise

training in field deployment and data pre-processing procedures.

Six months of my Ph.D. was spent in Ethiopia, where I was always made to feel welcome during

my travels. Special thanks to members of the Addis Ababa University - Drs Atalay, Laike,

Tesfaye, Dereje, Bekele, and Gezahegn, with whom it has been a pleasure to work with. The

technical staff at the Geophysical Observatory were also a great help during my stay in Addis

Ababa. The assistance offered by Ashenafi and Ewenet is also gratefully acknowledged.

Fortunately, tending to all 50 EAGLE II seismic stations by Toyota Landcruiser gave me the

great opportunity to enjoy the spectacular countryside in Ethiopia. Drivers from both Ethio-Der

and Addis Ababa University made the long journeys great fun. Ian Bastow Bastow, Dave C,

Christel Tiberi, Andy Page and Julie Rowland were fantastic to work with, both in the field, and

back home in England.

All friends and colleagues here at Royal Holloway have offered amazing support at various

stages of my research. Ellen Wolfenden and Richard Gloaguen helped get me on my feet right

at the beginning. The help from Mark and Frank with computing is also much appreciated. On a

more personal note; Nick, Jonas, Blair, Claire, Mike C, Liz, Paul, Simon and Helen, you guys

have been great. Thanks also to Eve for being a great friend for the last three years.

Finally, to my family - you have all been a source of comfort and vital support throughout.

Special thanks to Mom, you have been, and will be in my heart every day.

This research was supported by NERC grant NER/A/S/2000/01004 and NERC studentship

NER/S/A/2002/10547

iii

Abstract The seismically and volcanically active Main Ethiopian rift (MER) marks the transition

from continental rifting in the East African rift to incipient seafloor spreading in Afar.

New seismicity data is used to investigate the distribution of strain and its relationship

with magmatism immediately prior to continental breakup. From October 2001 to

January 2003, seismicity was recorded by up to 179 broadband instruments that

covered a 250 km x 350 km area. 1957 earthquakes were located within the network, a

selection of which was used for accurate location with a 3-D velocity model, focal

mechanism determination and shear-wave splitting analyses. Border faults are inactive

except for a cluster of seismicity at the structurally complex intersection of the MER

and the older Red Sea rift, where the Red Sea rift flank is downwarped into the

younger MER. Earthquakes are localized to ~20 km-wide, right-stepping en echelon

zones of Quaternary magmatism and faulting, which are underlain by mafic intrusions

that rise to 8-10 km subsurface. Seismicity in these ‘magmatic segments’ is

characterised by low magnitude swarms coincident with Quaternary faults, fissures and

chains of eruptive centres. Focal mechanisms predominantly show normal dip-slip

motion; the minimum compressive stress is N103oE, perpendicular to Quaternary faults

and aligned volcanic cones. The seismogenic zone lies above the 20 km-wide intrusion

zones; intrusion of magma may induce seismicity and faulting in the upper crust. S-

wave splitting from local earthquakes shows the largest amounts of upper crustal

anisotropy are in Quaternary magmatic segments; anisotropy is most likely caused by

melt-filled micro-cracks and dikes aligned perpendicular to the minimum compressive

stress. New and existing data indicate that during continental breakup, magma

intrusion beneath ~20 km-wide magmatic segments accommodates the majority of

strain and controls the locus of seismicity and faulting in the upper crust. The

observations from the MER do not support detachment fault models of lithospheric

extension but instead support a model of magma assisted rifting whereby the combined

effects of lithospheric stretching and heating by magma injection localises strain and

facilitates continental breakup.

iv

Contents

List of figures vii

List of tables viii

1 Introduction 1

1.1 Overview 1

1.2 Models of rifting 2

1.3 Along-axis segmentation of continental rifts and mid-ocean ridges 6

1.3.1 Segmentation of discrete continental rifts 7

1.3.2 Segmentation of slow-spreading mid-ocean ridges 9

1.4 Thesis aims and summary 11

2 Tectonic setting 13 2.1 Overview 13

2.2 Crustal and mantle structure 16

2.3 Constraints on the direction of extension 22

2.5 Previous seismicity studies 23

3 Seismic network and earthquake data 27 3.1 Introduction 27

3.2 EAGLE seismic stations 27

3.2.1 EAGLE I broadband network 27

3.2.2 EAGLE II broadband network 30

3.2.3 EAGLE III broadband and short-period profiles 30

3.2.4 Permanent broadband network in Ethiopia 31

3.3 Instrumentation 31

3.4 Deployment procedure and station setup 33

3.5 Network management and data collection 35

3.6 Data quality control 36

3.7 Earthquake detection in continuous seismic data 37

3.8 Arrival time measurements 39

3.9 Hypocentre determination methods and errors in earthquake locations 41

3.10 Amplitude measurements and initial magnitude estimation 43

3.11 Summary 45

v

4 Seismicity in the northern Main Ethiopian rift 46 4.1 Introduction 46

4.2 Distribution of seismicity results 46

4.3 Discussion 52 4.3.1 Pattern of seismicity on rift border faults 52

4.3.2 Seismicity in magmatic segments 53 4.3.3 Pattern of along-axis segmentation and episodic rifting 54 4.4 Summary 55

5 Local earthquake magnitude scale and seismicity rate 56 5.1 Introduction 56

5.2 Importance of a calibrated magnitude scale for Ethiopia 56

5.3 Amplitude data 57

5.4 Methodology 60

5.5 Results 61

5.5.1 Magnitude Scale for the MER 61

5.5.2 Local magnitude values and station corrections 65

5.6 Discussion 67

5.6.1 Seismic attenuation 67

5.6.2 Magnitude statistics and annual-cumulative seismicity rate 67

5.6.3 Seismic and volcanic hazards in Ethiopia 69

5.7 Summary 69

6 Style of faulting and stress field orientation determined from earthquake 71 focal mechanisms 6.1 Introduction 71

6.2 Determination of focal mechanisms 71

6.3 Method of inverting focal mechanisms for the regional stress tensor 74

6.4 Focal mechanism results 74

6.5 Stress tensor results 79

6.6 Quaternary volcanoes and faults as strain indicators 79

6.7 Discussion 82

6.7.1 Style of faulting 82

6.7.2 Direction of extension across the MER 83

6.8 Summary 83

vi

7 Shear-wave splitting in crustal earthquakes 85

7.1 Introduction 85 7.2 Mechanisms for seismic anisotropy in the crust 85

7.3 Determination of shear-wave splitting parameters 86

7.4 Results of shear-wave splitting analysis 91 7.5 Discussion 91 7.5.1 Crustal anisotropy beneath the rift-axis 91 7.5.2 Crustal anisotropy beneath the Ethiopian plateau 92 7.5.3 Model of crustal anisotropy beneath the MER 93 7.6 Summary 93

8 Discussion 95 8.1 Evidence for magma-fed along-axis segmentation of the MER 95 8.2 Temporal variations of magma supply and episodic rift opening 99 8.3 Comparison with slow-spreading mid-ocean ridges 101

8.4 Implications for models of continental breakup 103

9 Conclusions 104 Appendices A EAGLE broadband seismic stations 106 B Catalogue of earthquakes located with 1-D velocity model 110 C Publications in peer reviewed journals resulting from Ph. D. research 134

References 136

vii

List of Figures 1.1 Location of the Ethiopian rift with respect to the East African rift system. 3

1.2 Locations of EAGLE permanent broadband seismic stations in Ethiopia. 4

1.3 Conceptual models for extension of continental lithosphere. 6

2.1 Zones of Quaternary faulting and magmatism in the Ethiopian rift. 14

2.2 Structural map of the northern Ethiopian rift. 15

2.3 Schematic cross section of the MER. 16

2.4 Horizontal slice of controlled source tomographic model of the MER. 17

2.5 Profile of topography, Bouguer gravity anomaly, and crustal thickness. 18

2.6 2-D model of resistivity structure across the MER. 19

2.7 P-wave velocity models of the EAGLE controlled source experiment. 20

2.8 Depth slice at 75km of a mantle tomographic model beneath Ethiopia. 21

2.9 SKS splitting results in the MER. 22

2.10 3.2 My - present and current plate motions with respect to the Nubian plate. 24

2.11 Seismic activity of the Horn of Africa since 1960. 25

3.1 Locations of EAGLE broadband stations used for earthquake locations. 28 3.2 Locations of EAGLE III short period, single component “texan” instruments. 29

3.3 EAGLE II CMG-6TD station equipment and construction. 34

3.4 Example of GPS data from station E69 plotted for data quality control. 36

3.5 Example of seismometer Z, E and N mass positions. 37

3.6 Distribution of EAGLE I stations used for earthquake detection. 39

3.7 Example of vertical component recordings of an earthquake in the MER. 40

3.8 Examples of P-wave recordings for various arrival time quality factors. 41

3.9 Minimum 1-D P wave velocity model and Wadati digrams. 43

3.10 Example of processing required for measurement of earthquake amplitude. 44

4.1 Seismicity of the MER from October 2001 to January 2003. 47

4.2 Seismicity located near Quaternary eruptive volcanic centres near Fentale. 48

4.3 Earthquake locations determined using the 3-D P-wave velocity model. 49

4.4 Histograms of earthquakes depths. 50

4.5 Variations of earthquake frequency in sub-areas of the MER. 51

5.1 Distance / magnitude distribution of the earthquake data from the MER. 58

5.2 Distribution of earthquakes and seismic stations (Oct 01 - Jan 03). 59

5.3 Attenuation curves from the literature and new data from the MER. 62

viii

5.4 Magnitude estimated at varying hypocentral distances in the MER. 63

5.5 Variation in magnitude residuals with hypocentral distance. 64 5.6 Spatial variation of station factors in the MER. 65

5.7 Magnitude residuals / hypocentral distance distributions. 66

5.8 Magnitude-frequency, and Gutenburg-Richter distributions for the MER. 68

6.1 Example of a well constrained strike-slip focal mechanism from the MER. 72

6.2 Example of a well constrained normal-slip focal mechanism from the MER. 73

6.3 Selection of focal mechanisms from the MER. 76

6.4 Structural map of the MER with earthquake focal mechanisms. 77

6.5 Orientation of T-axes, slip-planes, and stress axes. 78

6.6 Structural map of the MER showing areas enclosed within Figs. 6.7; 6.8. 80

6.7 Landsat TM image and structural interpretation from Boset-Kone rift segment. 80

6.8 Aster 15m resolution imagery showing the Dofen volcanic edifice. 81

6.9 Orientations of Quaternary faults in two rift segment on the MER. 82

7.1 Examples of shear-wave splitting at EAGLE stations. 87

7.2 Crustal anisotropy measurements at 18 broadband stations in Ethiopia. 89

7.3 Along- and across-axis variations of anisotropy in the MER. 90

8.1 Cartoon sketch of the crustal structure of the MER. 97

8.2 Cartoon sketch of lithospheric structure beneath the MER. 99

List of Tables 3.1 The author’s fieldwork timetable. 35

6.1 Earthquake source parameters determined from EAGLE data. 75

6.2 Earthquake source parameters determined in other studies. 76

7.1 Shear-wave splitting measurements. 88

Chapter 1 - Introduction

1

Chapter 1 Introduction 1.1 Overview Some continental rifts undergo sufficient stretching and strain localisation to rupture the

rigid 150-250 km thick continental lithosphere, leading to production of new oceanic

lithosphere as a mid-ocean ridge (e.g. Ebinger 2005). Models of continental breakup

that assume purely mechanical stretching of the lithosphere predict strain localization

along pre-existing or new shear zones that may accommodate large displacements

(e.g., Lister et al., 1986; Dunbar and Sawyer, 1989). Alternative models of continental

breakup include the influence of magmatism (e.g. Buck 2004). Such models predict

extensional strain accommodation by injection of magma, with small offset faults above

the zone of dyking. Despite the fundamental importance of the processes that control

how continents split apart to form an ocean basin, there is no consensus on how strain

localises to achieve rupture, nor on what proportion of the strain is accommodated by

magmatism and faulting.

The structure of both continental and oceanic rifts provides fundamental constraints on

models of continental breakup. Studies of continental rifts reveal that a series of

discrete kinematically linked basins bounded by large border faults defines a regular

along-axis segmentation of the rift. The dimensions of the basins, uplifted rift flanks,

and the maximum length and depth extent of faults, depends upon the mechanical

properties of the lithosphere, with longer faults correlating with high elastic and/or

seismogenic layer thickness (e.g. Ebinger and Hayward, 1996; Ebinger et al., 1999). In

oceanic rifts the segmentation is dominated by along-axis bathymetric variations

thought to represent variations in the supply of basaltic melts from the underlying

asthenosphere (e.g. Phipps-Morgan and Chen, 1993). The timing and mechanism by

which the along-axis segmentation dominated by mechanical processes in continental

rifts is replaced by the along-axis segmentation controlled by asthenospheric and

magmatic processes at oceanic rifts offers key insights into how continental breakup

occurs. Detailed observations of the relationship between mechanical failure of the

crust shown by seismicity and faulting, and the distribution of intruded magma and


2

extrusive lava in a volcanically active rift setting that is near breakup provides a means

to study the pattern of strain localisation and to assess how strain is partitioned

between faulting and dike injection. It is the aim of this research to analyse the

distribution of seismicity, Quaternary faults and magmatism in the northern Main

Ethiopian rift, a region of incipient continental breakup, to evaluate how and when the

magma-fed along-axis segmentation observed in oceanic rifts replaces the mechanical

segmentation of continental rifts.

The Ethiopian rift forms the third arm of the Red Sea, Gulf of Aden rift-rift-rift triple

junction where the Arabian, Nubian and Somalian plates join in Afar (Fig. 1.1).

Embryonic rifting of the continental lithosphere is observed to the south in the East

African rift system in Kenya and Tanzania (e.g., Nyblade et al., 1996). Northeast of the

Ethiopian rift, incipient sea-floor spreading is evident in the Afar rift and in the Asal-

Ghoubbett rift which is the westward onshore extension of the Gulf of Aden spreading

ridge (e.g., De Chabalier and Avouac, 1994; Ruegg and Kasser, 1987; Stein et al.,

1991).

The seismically and volcanically active northern Main Ethiopian rift (MER) and Afar rifts

are virtually the only places worldwide where the transition between continental and

oceanic rifting is exposed onland. The MER is thus an ideal natural laboratory to study

continental breakup processes. The research that forms the core of this thesis utilised

the wealth of local seismicity data collected by up to 179 broadband seismic stations

deployed as part of project EAGLE (Ethiopia Afar Geoscientific Lithospheric

Experiment) (Fig. 1.2). In addition, the results from this multi-disciplinary project provide

fundamental constraints on crust and upper-mantle structure beneath the MER, set

within a strong regional tectonic framework (e.g. Maguire et al., 2003; WoldeGabriel et

al., 1990; Wolfenden et al., 2004).

1.2 Models of rifting The simplest kinematic models of extension can be represented using purely

mechanical stretching of the lithosphere (e.g. McKenzie et al., 1978; Dunbar and

Sawyer, 1989). Mechanical stretching models accommodate strain by ductile

deformation in the lower crust and lithospheric mantle and by large offset faults in the

brittle layers (Fig. 1.3). Depending on assumed lithspheric rheology, strain may be


3

distributed across a broad zone, or localised along one large displacement fault, or

detachment faults and shear zones that cross cut the lithosphere (e.g., Buck, 1991;

Ebinger, 2005).

Tanzaniacraton INDIAN

OCE AN

Turkana

Depression

LakeVictoria

L. Nyasa

L. Rukwa

L. Turkana

1010°N

5°N

0°

5°S

Malawi rift

3030°E 3434°E 3838°E

1010°S

L. Tanganyika

GulfAden

SeaRed

L. Tana

EthiopianPlateau

W esternRift Eastern

Rift

Plateau

EastAfrican

Som alianPlate

NubianPlate

ArabianPlate

Afar

Figure 1.1: Location of the Ethiopian rift with respect to the East African rift system. Brown lines

enclose regions with an elevation of >1000m. Purple designates flood basalts and rhyolite

deposits. Dark red shows sea floor spreading centres. The red box marks the EAGLE study

area. Modified after Ebinger (2005).


4

37˚

37˚

38˚

38˚

39˚

39˚

40˚

40˚

41˚

41˚

7˚ 7˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

11˚ 11˚

AAEAAE

FURIFURI

WNDEWNDE

Aluto-Aluto-Gedemsa MSGedemsa MS

Boset-Boset-KoneKoneMSMS

Fentale-Fentale-Dofen MSDofen MS

Angelele MSAngelele MS

Guraghe BFGuraghe BF

Arboye BFArboye BF

Asella-Sire BFAsella-Sire BF

AnkoberAnkoberBFBF

37˚

37˚

38˚

38˚

39˚

39˚

40˚

40˚

41˚

41˚

7˚ 7˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

11˚ 11˚

LEMELEME

SHEESHEE

GEWEGEWE

ANKEANKE

KOTEKOTE

MELEMELE MIEEMIEE

BEDEBEDE

MECEMECE

CHAECHAE

SENESENE

AREEAREEGTFEGTFE

BOREBORE

DONEDONE

INEEINEE

AMMEAMME

MEKEMEKE

DIKEDIKE

ADEEADEE

ADUEADUE

DZEEDZEE

WOLEWOLE

ASEEASEE

BUTEBUTE

KAREKARE

HIREHIRE

37˚

37˚

38˚

38˚

39˚

39˚

40˚

40˚

41˚

41˚

7˚ 7˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

11˚ 11˚

35˚ 40˚ 45˚

5˚

10˚

15˚

S

N

ADD

35˚ 40˚ 45˚

5˚

10˚

15˚

GAGA

RSRS

35˚ 40˚ 45˚

5˚

10˚

15˚

1000 2000 3000

Elevation

m

Addo-do MS

Figure 1.2: EAGLE permanent broadband seismic stations used for earthquake location with

respect to major border faults and magmatic segments of the Main Ethiopian rift (MER). Grey

triangles are Phase 1 stations (Oct 2001 - Jan 2003), white triangles are Phase 2 stations (Oct

2002 - Jan 2003, white circles are Phase 3 stations (Nov 2002 - Jan 2003) and white squares

are the IRIS GSN permanent stations FURI, AAE and WNDE. The top left inset shows the

topographic relief, plates and rift zones: A = Arabia; D = Danakil; N = Nubian Plate; S =

Somalian Plate; RS = Red Sea rift; GA = Gulf of Aden rift.

Most passive margins have been affected by magmatic intrusion and volcanic outflows,

even before the onset of faulting and subsidence that mark stretching (e.g. Sengor and

Burke, 1978). For example, the North American east coast was once regarded as a

classic example of a non-volcanic passive margin (e.g. Steckler and Watts, 1981).

However, offshore seismic data and onshore geologic mapping indicate that similar


5

volumes of lava were produced along the east coast as at volcanic margins (Holbrook

and Kelemen, 1993). Models of the opening of the South Atlantic emphasise the effect

of lithospheric stretching and detachment faulting (e.g. Etheridge et al., 1989; Lister et

al., 1991). However, piles of volcanic flows characterise the South Atlantic margin

(Hinz, 1981; White and McKenzie, 1989) and the 2000 km-long Greenland margin

(e.g., Mutter et al., 1988). Despite evidence for copious magmatism at the onset of

rifting in the Red Sea (e.g., Menzies et al., 1997; Pallister, 1987), thermo-mechanical

rifting models for the region ignore the effects of magmatism (e.g., Buck et al., 1988;

Chery et al., 1992; Martinez and Cochran, 1988; Steckler, 1985; Wernicke, 1985).

Models of rifting that consider purely mechanical stretching of the lithosphere assume

that the average stress or tectonic force required to initiate rifting is available. Several

authors have estimated that the tectonic forces likely to be available for rifting is in the

range of 3-5 TeraNt/m (Forsyth and Uyeda; Solomon et al., 1980). Areas of initially thin

lithosphere should rift at relatively low levels of tectonic force. However, the tectonic

force required for amagmatic rifting of thick continental lithosphere has been estimated

at up to an order of magnitude larger than that available (Kusznir and Park, 1987;

Hopper and Buck; 1993). This raises the question: “Is there a missing force ?”.

If magma is available during rifting, deformation of the lithosphere resulting from

extensional stresses not only occurs by fault slip and ductile flow but also by dyke

intrusion (Fig. 1.3). Buck (2004) shows that by considering the intrusion of magma into

the lithosphere, the yield stress for the lithosphere is significantly reduced. In a magma

assisted rifting model, magma intrudes the more ductile lower lithosphere, feeding

dykes that intrude higher in the lithosphere. As stretching leads to increased amounts

of lithospheric thinning, the heat transfer from magmatism further reduces plate

strength, allowing magma to intrude to shallower levels. Melt is buoyant to at least mid-

crustal levels and the buoyancy forces effectively reinforce the plate-driving forces.

Friction along fault surfaces competes with plate driving forces. Magma injection

therefore accommodates strain at lower plate-driving forces than faulting, facilitating

continental breakup at relatively low stresses. Therefore, in addition to the variability in

lithospheric properties considered by models involving purely mechanical stretching,

coupled lithosphere-asthenosphere models that consider injection of magma are

required to explain rifting and breakup of thick continental lithosphere and the onset of

sea-floor spreading (Buck, 2004).


6

Figure 1.3: Two general classes of conceptual models for extension of rheologically layered

continental lithosphere. The top panel shows a mechanical stretching model in which strain is

accommodated by large offset faults (e.g. detachment) in brittle layers, and by ductile

deformation in weaker layers. The lower panel shows a magmatic extension model that includes

the effects of magma intrusion and accompanying heating. The strain localisation and strength

reduction of the lithosphere is enhanced by melt intrusion. Modified after Buck (2004).

1.3 Along-axis segmentation of continental rifts and mid-ocean ridges Systematic along-axis segmentation is a feature of all continental and oceanic rift

systems. The timing and mechanism by which the along-axis segmentation dominated

by mechanical processes in continental rifts is replaced by the along-axis segmentation

controlled by asthenospheric and magmatic processes at oceanic rifts offers key

insights into how continental breakup occurs. This review focuses on the along-axis


7

segmentation observed at discrete continental rifts such as the East African rift system

and at slow and ultra-slow mid-ocean ridges, such as the northern Mid-Atlantic and

Gakkel ridges where extensional velocities are most similar to those observed across

the MER.

1.3.1 Segmentation of discrete continental rifts Discrete rift systems are localised zones of lithospheric extension less than 100 km

wide, such as the East African system (EAR), the Baikal rift zone, the Rhine graben

and the Rio Grande rift (Ruppel, 1995). The archetypal discrete rift, the EAR, consists

of rift depression bounded by large normal faults with uplifted footwall flanks (e.g.

Karner and Weissel, 1991). Discrete continental rifts comprise a relatively narrow zone

of deformation where basins are deep and broad and are segmented along their length

into individual, characteristically asymmetric basins or half grabens (e.g. Bosworth,

1985; Ebinger et al., 1999). The rift basins are typically 30-120 km-long, 40-70 km-wide

and bounded on one or both sides by steeply dipping (>50o) normal faults systems or

border faults (Ebinger et al., 1987; Dunkelman et al., 1989). Tectonic activity on basin

bounding fault systems causes subsidence of the rift floor and uplift of the rift flanks,

generating basins typically 1-7 km-deep (Morley, 1988) with maximum subsidence

adjacent to the centres of the border faults (Ebinger 1989). The majority of strain in the

brittle crust is accommodated by high magnitude earthquakes located along basin

bounding border faults (e.g., Jackson and Blenkinsop, 1997; Langston et al., 1998),

and the seismogenic zone can extend to 35 km depth in regions of strong lithosphere

such as the East African rift in Tanzania (e.g., Nyblade and Langston, 1995; Zhao et

al., 1997; Foster and Jackson, 1998). Extension within individual basin segments is

transferred to adjacent segments by accommodation zones with geometry dependent

on spatial relationship between border faults of adjacent rift segments (e.g. Morley et

al., 1990; Gawthorpe and Hurst 1993).

Although rift segment lengths are variable throughout the EAR, segments in a given rift

province are generally of similar length, with border fault lengths in cratonic parts of the

East African rift 80-120 km, and border fault lengths in younger lithosphere 30-60 km

(e.g. Jackson and Blenkinsop, 1997; Ebinger et al., 1999), indicating that the

dimensions of rift basins are controlled by the mechanical properties of the lithosphere

(Ebinger et al., 1999). Short, narrow basins with narrow uplifted flanks form in young,


8

hot, weak lithosphere, and are characterised by <15 km seismogenic layer thickness

and / or effective elastic thickness. Long, wide, deep basins with broad uplifted flanks

develop in old, cold, strong continental lithosphere with seismogenic layer thickness /

Te > 30 km (Ebinger et al., 1999).

Similar patterns are observed in other rifts. For example, the Central Baikal basin is

underlain by strong lithosphere (Te 45 - 60 km) with little evidence of lithospheric

thinning, heating and weakening (Burov et al., 1994; Grand et al., 1997). The broad

central basin of the rift is bounded by ~120 km-long border faults (Agar and Klitgord,

1995), on which earthquakes occur down to 30 km in the crust and possibly in the

upper mantle down to depths of 40 km (Déverchère et al. 1991; Déverchère et al.,

1993). In contrast, the Rio Grande rift is underlain by relatively hot, weak lithosphere

(Wilson et al., 2005); gravity data suggests thinning of the lithosphere near the rift axis,

seismic tomography models image a broad low velocity anomaly beneath the rift

(Cordell et al., 1991; Slack et al., 1996), and seismogenic layer thickness is less than

20 km (Jaksha and Sanford, 1986). The thermal and mechanical properties of the

lithosphere are broadly similar beneath the Rio Grande rift and the East Afican rift in

Kenya, and reflected near the surface by the similar rift basin width, fault length and

seismogenic layer thickness (Keller et al., 1991), with the higher volumes of rift related

volcanics in the Kenya rift likely due to the presence of a more concentrated vertical

mantle upwelling (Wilson et al., 1994). Similarly, discrete rift basins in the Corinth rift,

which developed over the past 20 Ma in relatively hot lithosphere of a collapsing

orogenic belt, have high angle faults of 25-30 km length and seismogenic layer

thickness and Te of 10-15 km (Roberts and Jackson, 1991; McNeill et al., 2005;

Hatzfeld et al., 2000).

The comparison of the variability between border fault dimensions in the East African,

Baikal and Corinth rifts show that long (> 80 km) border fault segments bound deep

broad rift basins that develop in cold thick lithosphere whereas shorter border faults

bound narrow basins within initially weak lithosphere. Border fault length increases with

seismogenic layer thickness and with effective elastic thickness. These patterns

indicate that the fault and flexural response of the continental lithosphere to rifting

processes is largely controlled by the mechanical properties of the lithosphere (Ebinger

et al., 1999).


9

An along-axis variation in rift segmentation is observed along the Ethiopian rift, which

marks the transition between continental rifting in East Africa rift and sea-floor

spreading in Afar (Hayward and Ebinger, 1996; Ebinger and Casey, 2001). The

Ethiopian rift is characterised by a northward decrease in rift basin length and width,

consistent with the northward decrease in crustal thickness and effective elastic

thickness. Hayward and Ebinger (1996) suggest that localisation of strain away from

border faults to intra-rift grabens is initially controlled by the progressive decrease in

lithospheric strength within the rift from a combination of lithospheric thinning and

resultant increased heating during continued extension. As rifting proceeds to breakup

the progressive decrease of lithospheric strength remains the fundamental control on

rift architecture, until the onset of sea-floor spreading when dyke injection

accommodates the majority of strain (Hayward and Ebinger, 1996). However, coupled

lithosphere-asthenosphere models of continental rifting illustrate the importance of

magma injection well before the onset of sea-floor spreading in localising strain to

achieve rupture of continental lithosphere (e.g. Buck 2004). The controls on the along-

axis segmentation of the MER are re-evaluated in light of new seismicity data

presented in this thesis and interpreted with additional constraints on lithospheric and

asthenospheric structure from new geological, geophysical and geochemical data.

1.3.2 Segmentation of slow-spreading mid-ocean ridges The median valley topography of a slow spreading mid-ocean ridge, such as the Mid-

Atlantic Ridge, results from the creation of new lithosphere through dyke injection,

volcanism and extensional faulting (e.g. Smith and Cann, 1999; Dunn et al., 2005).

Eruption of lava at the surface in a ~10 km-wide inner valley produces magmatic

topography in the form of 200-300 m-high, 10’s of km-long axial volcanic ridges aligned

orthogonal with the plate spreading vector, 100-500 m-high volcanoes or seamounts

and large expanses of hummocky pillow lava flows (Smith and Cann, 1990; Smith and

Cann, 1993). The generation of new lithosphere at mid-ocean ridges makes them

fundamentally different from continental rift systems. However, like continental rifts,

mid-ocean ridges exhibit a regular along-axis segmentation that offers clues to controls

on extensional processes.

Segmentation of slow spreading mid-ocean ridges occurs at a series of length-scales

with a complex range of offset morphologies. Transform faults produce first order ridge


10

segments that are 200-800 km-long. The juxtaposition of cold lithosphere against a

spreading centre at an offset results in a decrease in upwelling melt volumes (Phipps

Morgan and Forsyth, 1988), evident as thinner crust at ridge transform discontinuities

(e.g. Blackman and Forsyth, 1991).

Slow-spreading mid-ocean ridges are also segmented by non-transform offsets with an

along-axis spacing of 10-80 km and lateral offsets of up to 30 km (Sempéré et al.,

1990). The non transform offsets that define this second order segmentation are

characterised by a range of accommodation structures including zones of en echelon

faulting and interbasinal volcanic ridges (Sempéré et al., 1993). Ridge segmentation is

expressed in along-axis variations in bathymetry, with the non-transform offsets

forming local depth maxima whereas ridge segment midsections have the shallowest

bathymetry. The ridge segment midsections are underlain by the thickest crust while

the segment ends exhibit deep and wider axial valleys and thinnest crust, as inferred

from gravity data (Lin et al., 1990; Detrick et al., 1995) and determined seismically

(Tolstoy et al., 1993; Hooft et al., 2000; Hosford et al., 2001; Dunn et al., 2005). Ridge

segment midsections are also underlain by low velocity anomalies in the mid-lower

crust likely indicates the presence of anomalously high temperatures and partial melt

(Dunn et al., 2005). The pattern of seismic anisotropy of the lower crust beneath ridge

segment midsections is also indicative of the presence of partially molten dykes (Dunn

et al., 2005).

Along-axis variations in the amount of seismicity are also observed at slow-spreading

mid-ocean ridges. Monitoring of earthquakes along the northern Mid-Atlantic Ridge

shows that earthquakes are concentrated within ~20 km of the ridge axis and that each

discrete ridge segment can experience characteristic low or high level of seismicity

over decade time-scales (Barclay et al., 2001; Smith et al., 2003). Discrete mid-ocean

ridge segments also experience short-lived temporal variations in seismicity as shown

by the 147 ML > 3.5 earthquakes and continuous broadband tremor, that occurred

during 16-17 March 2001 along the 50 km-long Lucky Strike segment of the northern

MAR in March 2001 (Dziak et al., 2004). The earthquakes were interpreted as caused

by stress perturbation resulting from emplacement of a dyke beneath the ridge

segment. Similarly, a volcanic eruption and sequence of 252 mb > 4.5 earthquakes

occurred over three months along the ultra-slow spreading (full spreading rate of ~1.1

cm/yr) Gakkel ridge in the Arctic Ocean (Tolstoy et al., 2001). Seismicity was located


11

predominantly on faults that bound a ~10 km-wide axial graben, a pattern consistent

with reduction in normal stress at the base of the seismogenic zone from injection of

magma directly from the asthenosphere (Tolstoy et al., 2001).

In oceanic rifts the segmentation observed in along-axis variations in bathymetry are

thought to represent spatial and temporal variations in the supply of basaltic melts from

the underlying asthenosphere (e.g. Lin and Phipps-Morgan, 1992; Phipps-Morgan and

Chen, 1993). Spatial variations in melt flux, with melt preferentially delivered to

segment centres via dikes in the lower crust, are also believed to control crustal

thickness, lithospheric strength, and the partitioning of plate spreading between faulting

and magmatism (e.g. Lin et al., 1990; Sparks et al., 1993; Tucholke and Lin, 1994;

Magde et al., 1997; Parsons et al., 2000; Dunn et al., 2005). Temporal variations in

melt supply result in the great diversity of ridge segment morphologies and along-axis

variations in amount of seismicity (e.g. Cannat, 1993; Grácia et al., 1999; Tolstoy et al.,

2001; Dziak et al., 2004).

1.4 Thesis aims and summary Observations of rift systems shows that mechanical strength of the lithosphere controls

the along-axis segmentation of continental rifts during the initial stages of formation

whereas the along-axis segmentation at mid-ocean ridges is controlled by

asthenospheric and magmatic processes. Therefore, this thesis aims to constrain

where and how strain is partitioned between faulting and dyking in a rift system which

is transitional between continental and oceanic in style in order to distinguish between

models of strain accommodation prior to continental breakup. A high quality seismicity

dataset from the seismically and volcanically active northern Main Ethiopian rift is

analysed and interpreted within a strong regional framework in order to constrain the

relationship between brittle failure in the crust and injection of magma into the

lithosphere.

In the thesis that follows, Chapter 2 outlines the tectonic setting of the MER,

constrained by previous geological, geophysical and geochemical studies. Chapter 3

describes the data acquisition and pre-processing stages required for the analysis of

seismicity dataset recorded by the EAGLE network of broadband seismic stations.

Chapter 4 aims to constrain the locus of brittle deformation and relationship between


12

faulting and dyking by accurately locating earthquakes in the MER and comparing the

distribution of seismicity to the pattern of Quaternary faulting and volcanism observed

at the surface, and distribution of intrusive magmatism constrained by independent

studies. Chapter 5 aims to quantify the size of local earthquakes in our dataset by

accurately estimating local earthquake magnitude using a calibrated magnitude scale

for the MER. Chapter 6 aims to constrain the style of active faulting and orientation of

the stress field in the MER using earthquake focal mechanisms. Results are compared

to structural data, and to regional and global plate kinematic models. Chapter 7 aims to

use shear-wave splitting from local earthquakes to reveal patterns of seismic

anisotropy of the crust. The results are related to structural and geophysical data,

including SKS-splitting studies. In Chapter 8, the results are integrated with

independent geophysical and geological data. Evidence is presented for magma-fed

along-axis segmentation of the MER and compared to processes at slow-spreading

mid-ocean ridges. The implications that these new observations from the MER have for

models of continental breakup are also discussed in Chapter 8. Conclusions are

presented in Chapter 9.

The “tectonic setting” (Chapter 2) and “discussion” (Chapter 8) contain references to

original contributions by the author of this thesis in the form of co-authored publications

in peer reviewed journals; Kendall et al. (2005), Ayele et al. (2006), Casey et al.

(2006), Kendall et al. (2006), Wright et al. (2006). The results chapters 4-7 includes

research that forms three co-authored publications in peer reviewed journals; Keir et al.

(2005); Keir et al. (2006) and Keir et al. (in review). Reprints or the submitted

manuscripts of these papers are bound at the back of the thesis.

Chapter 2 - Tectonic setting

13

Chapter 2 Tectonic setting 2.1 Overview The Ethiopian rift system is on the Ethiopia-Yemen plateau that is thought to have

developed above a mantle plume (e.g., Schilling, 1973; Ebinger and Sleep, 1998;

George et al., 1998). A ~2 km thick sequence of flood basalts and rhyolites erupted

across the Ethiopian-Yemen plateau region between 45 and 22 Ma (e.g., George et

al., 1998; Kieffer et al., 2004) (Fig. 1.1). The majority erupted at ~30 Ma along the Red

Sea margins (e.g., Hofmann et al., 1997; Ukstins et al., 2002) coincident with the

opening of the Red Sea and Gulf of Aden (Wolfenden et al., 2005). Anomalously low

P-wave velocities exist in the mantle beneath Afar to depths of at least 410 km, but

their connection with the profound low velocity zone in the lower mantle beneath

Southern Africa is debated (e.g., Debayle et al., 2001; Benoit et al., 2003; Montelli et

al., 2004).

The MER forms one arm of the complex Afar triple junction zone (Fig. 2.1). Rifting

initiated in the southern and central Main Ethiopian rift between 18 Ma and 15 Ma, but

the northern Main Ethiopian rift (MER) only developed after ~11 Ma (WoldeGabriel et

al., 1990; Wolfenden et al., 2004). Between 12 and 10 Ma, the southern Red Sea

margin propagated southward as the MER propagated NE, effectively linking the

southern Red Sea and Ethiopian rifts, and forming a triple junction for the first time

(Wolfenden et al., 2004).

The MER formed within Precambrian metamorphic basement of the Pan-African shield

(Kazmin et al., 1978). Within the study area, however, Precambrian basement is

exposed in only one locality at the base of the footwall to the Guraghe border fault (Fig.

2.2). Hence, the structural grain of metamorphic basement is based on extrapolation

from exposures > 100 km to the southwest and southeast of the study area (e.g.,

Kazmin et al., 1978). Over 1 km of Mesozoic to early Tertiary marine passive margin

sequences overlie basement, and are in turn covered by 1-2 km of Oligocene to early

Miocene basalt-ignimbrite sequences (Abebe et al., 2005). The Mid-Miocene to


14

Recent infill within the MER comprises interbedded basalt and ignimbrite flows, with

isolated pockets of lacustrine and volcaniclastic strata (e.g., WoldeGabriel et al., 1990;

Wolfenden et al., 2004) (Fig. 2.3).

The NE-trending northern Main Ethiopian rift is a series of linked half grabens bounded

by steep NE-striking Miocene (11 - 5.3 Ma) border faults (WoldeGabriel et al., 1990;

Wolfenden et al., 2004) (Fig. 2.1). Structural patterns suggest a change from N130oE to

N105oE-directed extension sometime in the interval 6.6 to 3 Ma (Boccaletti et al., 1998;

Wolfenden et al., 2004). During this time period extensional strain migrated from border

faults to smaller offset ~N10oE-striking faults and aligned eruptive centers in the central

rift valley (Wolfenden et al., 2004) (Fig 2.1). These Quaternary (<1.8Ma) faults and

volcanic centres define a number of ~20 km-wide, ~60 km-long, right-stepping en

echelon “magmatic segments” (Ebinger and Casey, 2001) (Fig 2.2). Quaternary faults

within magmatic segments show predominantly normal slip and ~50% of faults have

eruptive centres or extrusive lavas along their length (Casey et al., 2006). GPS

measurements show that approximately 80 % of present day extension across the

MER is localized within these magmatic segments (Bilham et al., 1999).

42°E40°E

N

8°N TGD

Afar

Arabia

Ankoberborderfault

10°N

Arboyeborderfault

S. RedSearift

Tendaho Goba’adDiscontinuity

Quaternarymagmaticsegment

border fault

incipient plateboundary zone

Addo-do MS

few data

F -D MS

oceanic crust

Angelele MS

SRS

AdenRift

Nubia

Somalia

100 km

Boset MS

DHA-G MS

Figure 2.1: Zones of Quaternary faulting and magmatism (magmatic segments) with respect to

border faults of the Main Ethiopian rift (MER) and the southern Red Sea rift. Inset shows the

plate kinematic relation of the MER to the Southern Red Sea and Gulf of Aden rifts, and

opening directions between the Nubian, Somalian and Arabian plates. A-G: Aluto-Gedemsa, F-

D: Fentale-Dofen, SRS: Southern Red Sea, DH: Danakil Horst, TGD: Tendaho Goba’ad

Discontinuity. Black dots indicate locations of major shield complexes. After Wolfenden et al.

(2004).


15

38˚

38˚

39˚

39˚

40˚

40˚

41˚

41˚

7˚ 7˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

Quaternary faultsand eruptive centresand mid-Mioceneborder faults

RedSearift

MER

AnkoberBF

ArboyeBF

GuragheBF

AddisAbaba

Nazret

Z

LakeKoka

Fentale

Fentale -Dofenmagmaticsegment

Aluto-Gedemsamagmaticsegment AselaA -

Sire BF

Bosetmagmatic

Dofen

Kone

Boset

Gademsa

Debre Zeit chainD b Z it h i

Butajirachain

GadimottG

Ambo lineament

AtayeBF

Ayelu- Abida

YardiLake

AwashR.

Caldera Lakes

<2 My faultseruptive centers

mid-Miocene-Plioceneborder faults

aAlutoAluto

Figure 2.2: Structural map of the northern MER (after Casey et al., 2006). Quaternary faults

are shown black, and Quaternary eruptive centres are displayed red. Miocene border faults

bounding Main Ethiopian rift basins are grey with dip ticks. Fault plane solutions are lower

hemisphere projections, determined by Harvard CMT, Ayele (2000) and Hofstetter and Beyth

(2003) (Table 6.2). The Quaternary faults commonly have aligned eruptive centres and

extrusive lavas along their length. Miocene border faults likely formed under N130oE directed

extension. The border faults are ~60 km-long and define a number of half graben rift segments.

These Quaternary faults and volcanic centres define a number of ~20 km-wide, ~60 km-long,

right-stepping en echelon “magmatic segments” (Figs. 2.1).

Project EAGLE aimed to probe the structure of the magmatic segments, and relate

them to dynamic processes within the upper mantle in a rift at the transition between

continent and oceanic rifting. The following sections summarize EAGLE results within

the context of earlier work.


16

0

2000

6000

s.l.

4000

0 50 100 150 km

NNW SSE

MegezezVolcano

0

1000

2000

3000

4000

-1000

-2000

-3000

-4000

0 50 100 150 km

m

s.l.

??

?

Boset magmaticsegment

3.5 Ma- 2.5 Ma

10.6 -6.6 Ma~10.5 Ma

Keradifault

3.5 Ma - 2.5 Ma

6.6 Ma -

AdamaBasin

MegezezComplex

pre-rift floodvolcanics

“Balchi Fm”

“Kessem Fm”

ArboyeBorder Faultzone of closely-spaced

small-offset faults(flexure)

AdamaBasin

≤ 28 Ma

≤ 1.8 Ma“Wonji Fm”

7.8 Ma -10.6 Ma - 11?

Figure 2.3: Topography (top) and cross section of the MER through the Boset-Kone magmatic

segment illustrating the original half-graben structure of the basin and onlapping patterns along

the NW side of the basin. Note narrow zone of dyke injection and faulting near the centre of the

basin. The stratigraphic column on the right shows the typical Tertiary sequence overlying

Mesozoic strata and metamorphic basement. From Wolfenden et al. (2004).

2.2 Crustal and mantle structure The <20 km-wide, right-stepping, en echelon magmatic segments in the centre of the

rift are underlain by ~20 km-wide, high velocity (Vp > 6.5 km/s) elongate bodies that

are interpreted as cooled mafic intrusions (Keranen et al., 2004; Mackenzie et al.,

2005; Daly et al., 2006) (Fig. 2.4). These magmatic segments are characterized by

relative positive Bouguer anomalies (Mahatsente et al., 1999; Tiberi et al., 2005) (Fig

2.5). Enhanced conductive anomalies in the upper crust and at ~20 km depth beneath

the Boset magmatic segment shown in magnetotelluric data most likely indicate the

presence of zones of partial melt (Whaler and Hautot, 2006) (Fig. 2.6). Historic fissural

basalt flows at Fentale and Kone volcanoes as recently as 1810 (Harris et al., 1844)

and elevated temperatures at shallow crustal depths in the geothermal fields near

Aluto (Tadesse et al., 2003) indicate ongoing volcanic activity in magmatic segments.

There is also evidence for partial melt outside the fault-bounded rift valley, as

interpreted by Whaler and Hautot (2006) from high conductivity anomalies at 25-30 km

depth beneath the Ethiopian plateau (Fig 2.6).


17

The northern Main Ethiopian rift shows a northward increase in crustal extension and

magmatic modification (Tiberi et al., 2005; Maguire et al., 2006; Stuart et al., 2006).

Crustal thickness estimated from receiver function and wide angle seismic data

beneath the MER decreases from 38 km in the south beneath the caldera lakes to 24

km beneath Fentale volcano in the southern Afar depression (Dugda et al., 2005;

Maguire et al., 2006) (Figs. 2.7). The along-axis thinning is consistent with a northward

along-axis decrease in effective elastic thickness and seismogenic layer thickness

(Ebinger and Hayward, 1996). Seismic refraction / wide angle reflection data show

~40 km-thick crust beneath the southeastern plateau, whereas the western side of the

rift is underlain by 45-50 km-thick crust with a ~10-15 km high velocity (> 7.4 km/s)

lower crust believed to be magmatic underplate (Mackenzie et al., 2005) (Fig. 2.7).

Figure 2.4: Horizontal slice of controlled source tomography model at 10 km subsurface. Thick

contours mark 6 km/s with minor contours at 0.2 km/s intervals. High velocity bodies (red)

interpreted as solidified magma intrusions. From Keranen et al. (2004).


18

- 2 0 0 0

- 1 0 0 0

0

1 0 0 0

2 0 0 0

3 0 0 0

4 0 0 0

5 0 0 0

6 0 0 0

0 5 0 1 0 0 1 5 0 2 0 0km

met

ers

2 5 0 3 0 0 3 5 0

Northern MER

M S

- 4 0 0

- 3 0 0

- 2 0 0

- 1 0 0

0

1 0 0

0 5 0 1 0 0 1 5 0 2 0 0 2 5 0 3 0 0 3 5 0

Bouguer Anomaly

mG

als

- 5 0

- 4 0

- 3 0

- 2 0

- 1 0

0

0

km s

ubsu

rface

5 0 1 0 0 1 5 0 2 0 0 2 5 0km 3 0 0 3 5 0

Crustal thickness

CHAE

AREE

BOREE46

Topography

Figure 2.5: Profile of topography, Bouguer gravity anomaly, and crustal thickness estimates

obtained in a 3D inversion of filtered Bouguer anomaly data across the MER 38oE, 10oN to

41oE, 8o. White circles indicate depths estimated from receiver function studies (Stuart et al.,

2006). A relative high in the Bouguer anomaly is observed across the Boset-Kone magmatic

segment and the crust is thinnest beneath the magmatic segment (ms). From Tiberi et al.

(2005)


19

-40-35-30-25-20-15-10

-50

z (k

m -

scal

e x

2)

-120 -100 -80 -60 -40 -20 0 20 40 60 80 100 120Distance (km)

NW SE

Rift valley

Boset magmatic segmentDebre Zeit Chain

0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0Log Resistivity (Ω.m)

Figure 2.6: 2-D model of the resistivity structure across the MER. The profile follows the same

line as the across rift controlled source experiment. Marked above the profile are the positions

of the rift valley bounded by the Arboye border fault on the eastern margin and by a monoclinal

flexure. The position of the Boset-Kone magmatic segment and the position of the Debre Zeit

volcanic chain are marked. Conductive anomalies are clear at 20-25 km depth, as well as at

shallow levels beneath the Boset-Kone magmatic segment. A deeper anomaly at 25-30 km

depth is observed beneath the Ethiopian plateau. From Whaler and Hautot (2006).

Mantle tomography studies provide information on the geometry of lithospheric

stretching, as well as the role of melt in rift evolution. Bastow et al. (2005) imaged a

low-velocity zone as a narrow sheet rising to 65-75 km beneath the centre of the rift,

and extending to at least 250 km subsurface. Its linkage with the deeply rooted low

velocity zone termed the African superplume remains unclear. This relatively narrow

low velocity zone broadens into the Afar depression towards the Afar triple junction

zone (Bastow et al., 2005) (Fig. 2.8).


20

0

10

20

30

40

50

60

Dep

th (k

m)

0 50 100 150 200 250 300 350 400

3.05.0 5.0

3.2

5.15.90 6.29 6.01 6.20 6.40

6.50 6.40

6.32 6.11 6.42

6.71

6.756.60

6.70 6.70

7.00 7.20

7.10 7.00

7.50 7.50

~

(b) Line 2SW NESP21 SP22 SP23 SP24 SP15/25 SP26 SP27 SP28

0

10

20

30

40

50

60

Depth (km

)

0 50 100 150 200 250 300 350Distance (km)

5.0 4.8 5.26.10

6.166.336.40

6.646.82 7.38 7.70

8.05

6.83

6.63

6.38

6.316.156.07

6.73

6.65

6.546.526.24 6.23

6.516.086.13

3.3

~ ~

(a) Line 1NW SESP11 SP12 SP13 SP14 SP15/25 SP16 SP17 SP18

2 3 4 5 6 7 8 9Vp (km/s)

u.c.

l.c.

HVLCM

L

u.c.

l.c.

M

L

Figure 2.7: Final ray-trace P-wave velocity models for (a) Line 1 across-axis profile after (after

Mackenzie et al., 2005; Maguire et al., 2006) and (b) Line 2 along-axis profile of the EAGLE

controlled source experiment (after Keller et al. 2003; Maguire et al., 2006). The locations of the

refraction experiments are indicated on Figure 2.5. The crust beneath the southeastern plateau

is ~40 km-thick, whereas the western side of the rift is underlain by 45-50 km-thick crust with a

~10-15 km high velocity lower crust believed to be underplate (labeled HVLC). Crustal thickness

beneath the MER decreases from 38 km in the south beneath the caldera lakes to 24 km

beneath Fentale volcano in the southern Afar depression.

Geochemical and seismic data provide constraints on melting and melt emplacement

beneath the MER. The major element compositions of Quaternary mafic lavas from the

MER show the onset of melting occurs in the lower crust and upper sub-continental

lithospheric mantle (Rooney et al., 2005). This is consistent with P- and S-wave

tomographic models that show anomalous low velocity zones in the upper mantle

beneath the rift, attributed to a combination of higher temperatures and the presence of

partial melt (Bastow et al., 2005).


21

Figure 2.8: Depth slice at 75 km subsurface from the tomographic model of Bastow et al.

(2005). The low velocity is narrow and follows the rift axis, broadening into the Afar depression

where station coverage becomes sparse. Note the limb of low velocity material that extends

towards the Ambo lineament and zone of Quaternary eruptive centres (Fig. 2.2). From Bastow

et al. (2005).

SKS splitting shows the polarization direction of the fast shear-wave rotates from ~NE

at station outside magmatic segments to ~NNE along the axis of the rift (Fig. 2.9). The

splitting direction is along the length of the rift axis, perpendicular to the direction

expected for a strain-related fabric. The correlation between SKS polarisation and

Quaternary volcanic cones, as well as increased splitting in zones of more magmatism,

led Gashawbeza et al. (2004) and Kendall et al. (2005) to propose that partial melt

beneath the MER rises through dikes that penetrate through the thinned lithosphere.

Sv and Sh velocity models derived from surface wave dispersion curves are consistent

with a model of anisotropy due to aligned melt-filled pockets from 20-75 km depth

beneath the rift (Kendall et al., 2006)


22

Figure 2.9: SKS splitting results in the MER. The orientation of arrows shows the alignment of

fast shear waves and the length of the arrow is proportional to the magnitude of splitting. Yellow

arrows mark results at seismic stations deployed for 16 months, white arrows are for stations

deployed for three months, and red arrows are for the IRIS permanent stations FURI and AAE.

Heavy black lines show major border faults, dashed lines shows monclines and magmatic

segments are marked red. Top left inset shows topography (A, Red Sea; B, Gulf of Aden; C,

Arabian plate; D, Nubian plate; E, Somalian plate). The lower right inset shows the locations of

events used for analysis. From Kendall et al. (2005).

2.3 Constraints on the direction of extension

The orientation of present-day extension across the Ethiopian rift remains

controversial. Laser ranging and GPS data show that the northern Ethiopian rift over

the period 1969-1997 extended in a direction of N108oE± 10o at 4.5± 0.1 mm/year

(Bilham et al., 1999). The velocity field calculated from permanent GPS stations on

Africa since 1996 shows opening of ~6-7 mm/year at an azimuth of ~N95oE

(Fernandes et al., 2004; Calais et al., 2006) (Fig. 2.10). Global and regional plate


23

tectonic models by Jestin et al. (1994) and Chu and Gordon (1999) average plate

kinematic indicators from the past 3.2 My and find similar extension directions and

extensional velocities of N102oE at 5± 1 mm/year and N96oE± 9o at 6.0± 1.5

mm/year, respectively. Campaign GPS studies indicate an extension direction of

108oE, with an extensional velocity of ~ 6 mm/year.

Source parameters of teleseismically recorded earthquakes show normal, normal left-

oblique and sinistral strike-slip motions with the horizontal component of T-axes

between N135oE and N90oE in orientation (e.g., Ayele and Arvidsson, 1998; Foster

and Jackson, 1998; Ayele, 2000; Hofstetter and Beyth, 2003) (Figs. 2.10, 2.11).

Kinematic indicators on Quaternary faults that dip 70-75o and strike N10-35oE indicate

a principal dip-slip normal movement with a mean direction of ~N95oE (Pizzi et al.,

2006). However, Acocella and Korme (2002) matched pairs of asperities along the

sides of Quaternary extension fractures to show a mean extension direction of

N128oE± 20o. Korme et al. (1997) used the orientation of extension fractures to

determine an extension direction of NW-SE, similar to Wolfenden et al.’s (2004)

N130oE estimate of Miocene-Pliocene extension direction.

2.4 Previous seismicity studies Seismicity data is lacking from the Ethiopian rift due to previous sparse station

coverage (Ayele and Kulhánek, 1997). However, written records in Ghe’ez and Arabic

document seismic activity in the Horn of Africa for the last six centuries (Gouin et al.,

1979). The earliest documented seismic event in the Ethiopian rift is a swarm of

earthquakes in 1841-1842 near Debre Birhan which caused the destruction of the town

of Ankober by landslides (Gouin, 1979) (Fig. 2.11). Historical records spanning the

past 150 years show that large magnitude earthquakes are rare in the MER.

The record of seismicity from 1960-2000 compiled from teleseismic and regional

catalogues complete down to ML ~ 4 shows that the majority of earthquakes are

located along the highly eroded southern Red Sea escarpment north of 9.5oN, 38.7oE

(e.g., Kebede et al., 1989; Ayele, 1995; Ayele and Kulhánek, 1997; Hofstetter and

Beyth, 2003; Ayele et al., in press) (Fig. 2.11). Seismicity located in the MER north of

the Aluto-Gedemsa magmatic segment is concentrated along the axis of the rift

whereas seismicity south of Gedemsa is distributed across a wider (~40 km) zone.


24

Figure 2.10: 3.2 My - present and current plate motions with respect to the Nubian plate; vector

direction points in direction of plate motion of the Somalian plate with respect to the Nubian

plate and vector length scaled to extensional velocity. Ellipses at vector tips show error

estimates. Vectors from four studies are shown; black - Chu and Gordon (1999), red - Sella et

al. (2002) blue - Fernandes et al. (2004), light orange - Calais et al. (2004). Earthquake focal

mechanisms are from the Harvard CMT catalogue and earthquake epicentres and magnitudes

from the NEIC. Earthquakes are sized and coloured by magnitude. The position of the EAGLE

study area in the northern MER is outlined in red. Top left inset shows major plate boundaries

(solid red) and incipient plate boundaries (speckled line) of Africa, Arabia and the Indian Ocean.

From E. Calais (pers. comm.., 2005).


25

36˚

36˚

38˚

38˚

40˚

40˚

42˚

42˚

44˚

44˚

6˚ 6˚

8˚ 8˚

10˚ 10˚

12˚ 12˚

14˚ 14˚

Ethiopian Ethiopian PlateauPlateau

Southeastern Southeastern PlateauPlateau

AfarAfarDepressionDepression

GulfGulfofofAdenAden

RedRedSeaSea

AnkoberAnkoberDebre Debre BirhanBirhan

AddisAddisAbabaAbaba DofenDofen

FentaleFentaleKoneKone

BosetBoset

GedemsaGedemsa

AlutoAluto

AyeluAyelu

AbidaAbida

36˚

36˚

38˚

38˚

40˚

40˚

42˚

42˚

44˚

44˚

6˚ 6˚

8˚ 8˚

10˚ 10˚

12˚ 12˚

14˚ 14˚

Earthquake Magnitude

2 3 4 5 6

0˚ 30˚

-30˚

0˚

30˚

0 1000 2000 3000 4000 5000

Elevation

m

Figure 2.11: Seismic activity of the Horn of Africa since 1960. Earthquake locations and

magnitudes are from Ayele (1995) for the time period 1960 - 1997 and the NEIC catalogue

(1997 - 2005). Earthquake epicentres are scaled by magnitude. Earthquake focal mechanisms

are lower hemisphere projection and obtained from Harvard CMT catalogue, Foster and

Jackson (1998), Ayele and Arvidsson (1998), Ayele (2000) and Hofstetter and Beyth (2003).

Quaternary volcanoes along the axis of the MER are shown by triangles with names labelled.

Seismicity in the MER north of Gedemsa is concentrated along the axis of the rift whereas

seismicity south of Gedemsa is distributed across a wider (~40 km) zone in the rift.


26

An estimate of seismic moment release since 1960 shows that more than 50% of

extension across the MER is accommodated aseismically (Hofstetter and Beyth,

2003). During this period swarms of low magnitude events have been located near

Debre Birhan (Gouin, 1979), and near Fentale volcano in 1981 and 1989 where NNE-

striking surface fissures developed following earthquake swarms of ML<4 (Asfaw,

1982). Similar fissures oriented N20oE and N45oE are observed elsewhere along the

axis of the MER and attributed to tectonic processes (Asfaw, 1982; 1998). Tension

fractures cut welded tuffs at Fentale and Kone volcanoes and suggest a fissuring

episode within the past 7000 years (Williams et al., 2004). In the year preceding the

EAGLE study, the seismicity was concentrated in the Fentale-Dofen and Angelele

magmatic segments (Ayele et al., 2006). From mid-October 2003, after removal of the

EAGLE seismic network, a ~1 month-long earthquake swarm with a mainshock of

ML~5 was recorded by the Geophysical Observatory and reported by local inhabitants

near Dofen volcano (Geophysical Observatory, Addis Ababa University, personal

communication). The epicenter of the mainshock is estimated to be ~9.2oN 40.1oE

from the locations of damaged buildings and trees, reported scree slides in the area,

and personal accounts of ground-shaking (Fig. 2.11).

Hypocentre depths of 5-10 km have been reported for seismic swarms in the MER and

southern Afar (Asfaw, 1982; Ayele et al., 2006). Teleseismically recorded earthquakes

on the eastern side of the MER have been located between 8-12 km depth (Ayele,

2000).

Chapter 3 - Seismic network and earthquake data

27

Chapter 3

Seismic network and earthquake data 3.1 Introduction

Detailed monitoring of seismicity in the Ethiopian and Afar rifts has previously been

hampered by the sparse distribution of seismic stations in Ethiopia. The network of

seismic stations operational during EAGLE provided a dense distribution of three-

component broadband seismic stations that covered the MER and its uplifted flanks for

a period of 16 months from October 2001 - January 2003. The wealth of local

earthquake data recorded by the EAGLE network provides a unique opportunity to

analyse micro-seismicity at a higher resolution than has previously been possible in

Ethiopia and elsewhere in the East African rift system. This chapter outlines how the

EAGLE seismic data were collected, prepared and processed for earthquake locations

and magnitude. It includes accounts of fieldwork, data conversion, data quality control,

local earthquake detection, and arrival-time and trace amplitude measurements.

3.2 EAGLE seismic stations EAGLE comprised a number of discrete passive and controlled source studies that

required different distributions of seismic stations that operated over time periods of

between 10 days and 16 months. All EAGLE seismic stations were available for the

microseismicity study.

3.2.1 EAGLE I broadband network The EAGLE I broadband network was primarily designed to record sufficient

teleseismic earthquakes to map the upper-mantle seismic velocity structure beneath

the MER using body-wave tomography (Bastow et al., 2005). Teleseismic arrivals were

also used for receiver function (Stuart et al., 2006) and SKS-splitting studies (Kendall et

al., 2005). The network consisted of 30 stations (10 Güralp CMG-40TD and 20 Güralp

CMG-3TD instruments) with a nominal spacing of 40 km covering a region 250 x 350

km of the rift and its uplifted flanks (Figure 3.1) (Bastow et al., 2005). The network was

centred on the Boset magmatic segment in the centre of the rift, ~75 km SE of Addis


28

Ababa, and recorded for 16 months between October 2001 and January 2003. The

broad and even distribution of stations was also ideal for monitoring micro-seismicity in

the MER. The EAGLE I network was deployed and managed by Leeds University.

37˚

37˚

38˚

38˚

39˚

39˚

40˚

40˚

41˚

41˚

7˚ 7˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

11˚ 11˚

AAEAAE

FURIFURI

WNDEWNDE

Caldera Caldera LakesLakes






Arboye BFArboye BF


AnkoberAnkoberBFBF

37˚

37˚

38˚

38˚

39˚

39˚

40˚

40˚

41˚

41˚

7˚ 7˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

11˚ 11˚

LEMELEME

SHEESHEE

GEWEGEWE

ANKEANKE

KOTEKOTE

MELEMELE MIEEMIEE

BEDEBEDE

MECEMECE

CHAECHAE

SENESENE

AREEAREEGTFEGTFE

BOREBORE

DONEDONE

INEEINEE

AMMEAMME

MEKEMEKE

DIKEDIKE

ADEEADEE

ADUEADUE

DZEEDZEE

WOLEWOLE

ASEEASEE

BUTEBUTE

KAREKARE

HIREHIRE

37˚

37˚

38˚

38˚

39˚

39˚

40˚

40˚

41˚

41˚

7˚ 7˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

11˚ 11˚

35˚ 40˚ 45˚

5˚

10˚

15˚

S

N

ADD

35˚ 40˚ 45˚

5˚

10˚

15˚

GAGA

RSRS

35˚ 40˚ 45˚

5˚

10˚

15˚

1000 2000 3000

Elevation

m

EAGLE I

EAGLE II Permanent

EAGLE III

Figure 3.1: EAGLE broadband seismic stations used for earthquake location with respect to

major border faults and magmatic segments of the Main Ethiopian rift (MER). Grey triangles are

Phase 1 stations (Oct 2001 - Jan 2003), white triangles are Phase 2 stations (Oct 2002 - Jan

2003, white circles are Phase 3 stations (Nov 2002 - Jan 2003) and white squares are the IRIS

GSN permanent stations FURI, AAE and WNDE. The top left inset shows topographic relief,

plates and rift zones: A = Arabia; D = Danakil; N = Nubian Plate; S = Somalian Plate; RS = Red

Sea rift; GA = Gulf of Aden rift.


29

EAGLE III

stations

Earthquake

Epicentres

37˚

37˚

38˚

38˚

39˚

39˚

40˚

40˚

41˚

41˚

7˚ 7˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

11˚ 11˚

AAEAAE

FURIFURI

WNDEWNDE

Caldera Caldera LakesLakes






Arboye BFArboye BF


AnkoberAnkoberBFBF

37˚

37˚

38˚

38˚

39˚

39˚

40˚

40˚

41˚

41˚

7˚ 7˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

11˚ 11˚

35˚ 40˚ 45˚

5˚

10˚

15˚

S

N

ADD

35˚ 40˚ 45˚

5˚

10˚

15˚

GAGA

RSRS

35˚ 40˚ 45˚

5˚

10˚

15˚

1000 2000 3000

Elevation

m

Figure 3.2: EAGLE III short period, single component “texan” instruments with respect to major

border faults and magmatic segments of the Main Ethiopian rift (MER). White diamonds are

Phase III short period, single component “texan” instruments deployed for 8 days from 9 Jan

2002 - 16 January 2002. White stars are epicentres of earthquakes that occurred during the two

hour operating time windows during detonation of controlled sources. The top left inset shows

the topographic relief, plates and rift zones: A = Arabia; D = Danakil; N = Nubian Plate; S =

Somalian Plate; RS = Red Sea rift; GA = Gulf of Aden rift.


30

3.2.2 EAGLE II broadband network

The EAGLE II network was specifically designed to improve station density in the MER

for detailed analysis of micro-seismicity and to improve ray-coverage for local

earthquake tomography. Stations were thus deployed in areas that experience most

seismic activity and where crustal heterogeneity was expected to be greatest. The

EAGLE II network consisted of 50 stations (Güralp CMG-6TD sensors) that were

deployed with nominal spacing of ~15 km mainly within the rift valley (Fig. 3.1). The

network operated between October 2002 and January 2003 and was deployed and

managed by Royal Holloway University of London.

EAGLE I and II stations were located in relatively secure compounds such as schools

and clinics to safeguard the equipment from wild animals and vandalism. Sites were

selected with an unobstructed sky view to allow the GPS system to receive a signal

from as many satellites as possible and provide solar panels with maximum direct

sunlight. Seismometers were commonly close to sources of high levels of cultural noise

during the day.

3.2.3 EAGLE III broadband and short-period profiles The EAGLE III broadband cross-rift profile was deployed to record teleseismic arrivals

for both receiver function and SKS-splitting studies to assess variations in crustal and

mantle structure across the MER. From November 2002 - January 2003, 91 stations

(Güralp CMG-6TD sensors) were deployed at 5 km intervals along a ~450 km long

profile that traversed the MER between the Blue Nile gorge in the Ethiopian Highlands

and the Bale Mountains on the eastern part of the plateau (Fig. 3.1). The EAGLE III

broadband profile significantly improved the local earthquake tomography ray-coverage

on the Ethiopian and Southeastern plateau.

The EAGLE III controlled source project aimed to image variations in crustal structure

of the MER using wide-angle reflection/refraction and controlled source tomographic

methods. For 8 days in January 2003, the 91 EAGLE III cross-rift broadband stations

were supplemented by short period, single component Reftek ‘texan’ instruments at 1

km intervals to form the cross-rift wide-angle reflection/refraction profile (Mackenzie et

al., 2005) (Fig. 3.2). A profile of ‘texans’ were also positioned at 1 km intervals along


31

the axis of the MER between Lake Shala in the central Ethiopian rift to Gewane in the

northern Ethiopian rift (Keller et al., 2004). ‘Texans’ were also arranged at 2.5 km

nominal spacing as a ~100 km diameter ring centered at the intersection of the cross-

rift and along-rift profiles (Keranen et al., 2004). The texans were powered by an

internal battery and only operated for a two hour window during pre-planned controlled

source detonations. Three earthquakes occurred in the MER during such time windows

(Fig. 3.2).

3.2.4 Permanent broadband network in Ethiopia In addition to the temporary EAGLE network, local earthquake data was available from

three permanent stations (AAE, FURI and WNDE) that are managed by the

Geophysical Observatory, Addis Ababa University (Fig. 3.1). Three-component

broadband waveform data are available from the permanent IRIS GSN station FURI via

the IRIS WILBER II website (http://www.iris.washington.edu/cgi-bin/wilberII_page1.pl).

In addition, P- and S-waves for earthquakes recorded at AAE and WNDE were

analyzed by technicians at the Geophysical Observatory and incorporated into the

arrival-time data from EAGLE stations.

The names, locations, instrumentation and periods of operation of all EAGLE

broadband stations and the permanent network in Ethiopia are summarized in

Appendix A.

3.3 Instrumentation The EAGLE I network consisted of 10 Güralp CMG-40TD instruments and 20 CMG-

3TD instruments. These have a near flat velocity response from 0.008 to 50 Hz (120 s -

0.02 s) and from 0.03 to 50 Hz (30 s - 0.02 s) respectively. Each instrument has an in-

built 24-bit digitizer which was pre-configured to record vertical, north-south and east-

west data streams at 50 s.p.s.. These instruments were powered using a 100 Ahr

battery that was continuously recharged, via a regulator box, by 2x36 W solar panels.

Data were recorded by a SAM (Seismic Acquisition Module) onto 9 Gb removable

SCSI disks capable of storing more than 6 months of continuous data.


32

The EAGLE II and EAGLE III broadband networks consisted of Güralp CMG-6TD

seismometers that have a flat velocity response from 0.03 to 50 Hz and recorded at

100 s.p.s.. These instruments were powered using a 12V/10Ah Dryfit500 battery that

was continuously recharged, via a regulator box, by a ‘BP Solar’ solar panel (Fig. 3.3).

Data were stored on an internal 3 Gbyte FLASH memory card capable of storing 6-8

weeks (depending on noise levels) of continuous data at 100 s.p.s..

The sensors in Güralp seismometers are based on a design of a leaf spring suspended

boom supporting a transducer coil. The boom and the coil form an inertial mass.

Capacitance position sesnors provide a voltage proportional to the displacement of the

masses from their original position. This voltage, after amplification, generates a

current In the force transducer coil which tends to force the mass back to its original

position. The motion of the mass is effectively cancelled and thus provides a force

balance. The feedback voltage is a measure of the force, and thus also of the

acceleration of the mass. A complete account of the operation of all Güralp instruments

used in EAGLE can be found at their company website (http://www.guralp.net).

The digitizer at each station continuously digitized seismometer mass positions and

thus allows sensor stability to be monitored. Tilting of the sensor due to post-

deployment settlement can cause mass positions to depart from the recommended

operational range of ±2.5 million counts.

When a seismometer tilts, the vertical component reduced by a factor cos θ (where θ

is the angle of tilt from the vertical) whilst the horizontal components pick up a fraction

of the vertical component - given by a factor sin θ (e.g., Neuberg et al., 2002). For

small tilt angles this is not problematic for the vertical component since cos θ ≈ 1. For

larger tilt angles, however, sin θ << 1 and the horizontal components are perturbed.

The Güralp CMG-3TD sensors have an autocentre facility that can counter the problem

of tilt up to ±2.5o of tilt. The CMG-40TD and CMG-6TD sensors do not have this facility

but CMG-6TD sensors suffer no observable variation in performance up to ±3o of tilt.

Monitoring of sensor mass positions was therefore an important aspect of data quality

control (section 3.6).

Timing on all broadband stations was determined using an internal clock that was

synchronized using a GPS signal (Fig. 3.3). The GPS provided nominal timing


33

precision of 0.5 µs. Details of the GPS synchronization were recorded as a log file

which was used to check for data timing errors during data quality control (section 3.6).

3.4 Deployment procedure and station setup

The author participated in EAGLE II station deployment in October 2002 as well as

three equipment service and data retrieval visits from November 2002 - January 2003

(Table 3.1). Therefore, details of deployment procedure, station setup, network

management and data retrieval / archiving related to EAGLE II seismic stations are

discussed here. Details of the EAGLE I network are available in Bastow (2005).

EAGLE phase II Güralp CMG-6TD seismometers were installed in a ~0.5 m deep, ~0.5

m wide pit dug in superficial deposits, with the top of the sensor typically 20-50 cm

beneath the surface. Sensors were typically wrapped in two plastic bags to prevent

damage to the instruments by water, soil and insects. Sensors were oriented to true

north using a hand held compass which has an estimated orientation error of ±2

degrees. An air-bubble was used to initially level the sensor within the pit. Prior to

burial, mass-positions, data recording streams and GPS synchronization were all

checked with Güralp ‘shout’ software run from a handheld palm-pilot. Once the health

of the deployed station was ensured, the sensor pit was filled with sand.

A single ‘BP Solar’ solar panel was fastened to a wooden stake typically ~1-1.5 m

away from the sensor pit (Fig. 3.3). The solar panel surface was oriented south and

angled ~20o from the horizontal. The solar panel charged a 12V/10Ah Dryfit500 battery

wrapped in two plastic bags and buried directly beneath the solar panel. The battery

and solar panel were connected in parallel through a regulator to ensure a steady

supply of power to the station and allow the battery to be charged during periods of

sunshine.

The GPS antenna used to synchronize the internal clock of the sensor was fixed to the

top of the wooden stake used to hold the solar panel in position (Fig. 3.3). The GPS

antenna was typically ~1.5 m above the surface. A 1-1.5 m high chicken or barbed wire

fence, supported by four wooden posts, was constructed around the station to prevent

disturbance by animals and people.


34

The location of the station was determined from the average of three measurements

made with a Garmin handheld GPS. Station locations are estimated to be accurate to

±200 m in the horizontal plane and ±100 m in the vertical plane.

Figure 3.3: EAGLE II CMG-6TD station equipment and construction. Top left panel shows

equipment made by Güralp A, GPS module; B, breakout box; C, CMG 6TD seismometer. Top

right panel shows a typical scene during data download with the portable SCSI disk, D,

contained within the rucksack. E is a solar panel with attached GPS module. The breakout box,

battery and firewire cable used to transfer data between the sensor and disk are stored in a

plastic bag, F, buried near the solar panel. In this scene the firewire cable is attached to the

SCSI disk, D. Lower panel shows a completed EAGLE II 6TD station at Kiyensho School, E75.

E, solar panel and attached GPS module; F, position of buried breakupout box, battery and

firewire cable; G, position of buried sensor; H, barbed wire fence and wood fence posts.


35

# Dates Details of Fieldwork

1 2002/09/27-

2002/12/09 • Deployment of EAGLE II stations

• First EAGLE II service and data retrieval

2 2002/12/27-

2003/02/21 • Second EAGLE II service and data retrieval

• Participated in deployment and retrieval of EAGLE II single

component controlled source geophones

• Participated in retrieval of EAGLE III broadband stations

• Assisted in retrieval of EAGLE I network

• Retrieval of EAGLE II stations and shipment of equipment to U.K.

3 2004/07/12-

2004/07/15 • Collaborated with Atalay Ayele (Geophysical Observatory, Addis

Ababa University) for field investigation of earthquake damage from

November 2003 swarm of earthquakes near Dofen volcano.

4 2004/11/15-

2004/12/12 • Collaborated with Julie Rowland (University of Auckland, NZ) for

field mapping of faults and fissures in the Ethiopian rift

Table 3.1: The author’s fieldwork timetable. Two visits to Ethiopia related to EAGLE comprised

5 months from September 2002 - February 2003. Two visits related to field mapping and

earthquake damage investigation comprised 1 month of 2004.

3.5 Network management and data collection The 3 Gbyte memory of the CMG-6TD sensors is set as a circular buffer and thus the

oldest data are overwritten when the memory is full. To avoid loss of data, the EAGLE

phase II stations were visited twice between deployment in October and retrieval in late

January (Table. 3.1). During each visit, all 3 Gbyte of data were copied from the sensor

to a portable SCSI disk (DFD unit) via a firewire cable. The general health of the

seismic station was also monitored at this time. Data streams, sensor leveling, and

GPS signal were checked with a palm-pilot. All solar panels were cleaned, cable

connections checked, and wire fences surrounding the station re-enforced.

The raw data, in Güralp Compressed Format (GCF), were copied from the portable

SCSI disks to the SEIS-UK EAGLE II field Sun Workstation in the Department of

Geology and Geophysics, Addis Ababa University. The data were copied from the

workstation to DLT and DAT tapes and transported to the U.K. Tapes were

downloaded onto the SEIS-UK data processing workstation at the University of


36

Leicester and the raw continuous GCF data converted to miniSEED format. A dataless

SEED file containing both network and station specific information not included in the

miniSEED data was also created. This full data volume was subsequently archived at

the IRIS Data Management Center, Seattle. Full details of the data conversion

procedures are available at SEIS-UK’s online documentation

(http://www.le.ac.uk/geology/seis-uk).

3.6 Data quality control

An internal clock offset of 0.02 s (the sampling rate of EAGLE I instruments) was used

as the quality control benchmark to minimize the timing error in earthquake arrival time

measurements. The offset and drift of the internal clock at the moment of

synchronization was recorded on a log file, which was used to plot a time series of

GPS health (Fig. 3.4). Periods of time that lacked regular clock synchronization were

noted and seismic data during these time periods was not used for measurements of

arrival times. The one hour clock synchronization interval was generally sufficient to

keep the internal clock drift below 0.02 s. Stations that experienced problems acquiring

a GPS fix were set to continuous clock synchronization (Fig. 3.4).

Figure 3.4: Example of GPS data from station E69 plotted for data QC. The GPS did not

function from mid-Dec02 until 6Jan03.


37

Sensor mass positions were recorded in the data stream which was used to plot a time

series of sensor mass positions for each station (Fig. 3.5). Mass positions for CMG-

6TD seismometers should ideally be less than 2.5 million counts and all EAGLE II

instruments were deployed with mass positions below this quality control benchmark.

EAGLE II stations did not experience significant changes in sensor mass positions after

deployment. A summary of the quality control assessment performed on the EAGLE II

data is in Appendix B.

Mass Z

Mass N

Mass E

Date and Time

349593

406713

867543

993015

414985

470167

10/24/02 10/31/02 11/07/02 11/14/02 11/21/02

Figure 3.5: Example of seismometer Z, E and N mass positions from station E39 plotted for QC

purposes. Mass positions should be below 2.5 million. The small fluctuations in mass positions

are due to daily variations in temperature.

3.7 Earthquake detection in continuous seismic data

Earthquakes were detected in the continuous miniSEED seismic data with the

automatic event detection algorithm datascan. The algorithm used a comparison of

average short term amplitude (STA) and average long term amplitude (LTA) to detect


38

events. The STA was calculated for a sliding 1 sec. time window while the LTA was

calculated for a sliding 60 sec. time window. The data were bandpass filtered at 2-15

Hz and a positive detection was made if the STA/LTA ratio exceeded 20:1 on two or

more stations within a 2 minute time window. The STA/LTA ration was chosen by a trial

and error process. Several test runs were made with different STA/LTA ratios for one

week of data containing a known number of earthquakes found by manual inspection.

The lower the STA/LTA ratio used in the trigger algorithm, the larger the number of

events that are detected. A ratio of 20:1 provided the best balance between detecting

too many events (earthquakes and non-seismogenic signals, e.g. rockfalls, gusts of

wind, cultural noise, etc.) and detecting too few genuine earthquakes.

The trigger algorithm could be applied to four stations at a time. Due to the large spatial

coverage of the EAGLE network, applying the algorithm to just four widely spaced

stations did not detect an acceptable number of the lower magnitude earthquakes. The

network was thus divided into six sub-sections and the algorithm applied to four

relatively quiet stations within each section (Fig. 3.6). The subdivision of the network

increased the time dedicated to pre-processing but was very successful in detected low

magnitude earthquakes throughout the MER.

The trigger algorithm was only applied to stations of the EAGLE I network. Data noise

levels from the EAGLE I CMG-3T and CMG-40TD are much lower than that from the

EAGLE II and III CMG-6TDs. Applying the trigger algorithm to the 6TDs returned fewer

earthquakes and more non-seismogenic events. Secondly, the EAGLE I network was

operational for the full duration of the 16 month experiment, with Phase II and III

stations only deployed for the final 4 months. The trigger algorithm was applied to the

same set of stations for the full duration of the experiment to ensure a homogeneous

dataset that would not be biased from temporal variations in the number of operational

stations.

For each event detected by the algorithm the corresponding seismograms were output

as a postscript plot and genuine seismic events were discriminated by visual

inspection. Many of the earthquakes recorded by the EAGLE project were of regional

or teleseismic origin. Events with S-P travel time differentials of larger than 60 seconds

were immediately discarded. This selection process resulted in a dataset of 2139

earthquakes.


39

41˚

8˚

9˚

10˚WEE

TETE

MIEEEMIEE

EDEEDEAREEAREE

GTFEEEGTFE

RERE

INEEINEE

AAA

KEKEKEEKE

ADEEADEE

DZEEDZEE

WOLEWOLE

BUBU

HIREHIRE

3838˚ 3939˚ 4040˚ 41˚

8˚

9˚

10˚

35˚ 40˚ 45˚

5˚

10˚

15˚

S

N

AD

35˚ 40˚ 45˚

5˚

10˚

15˚

GAGA

RSRS

35˚ 40˚ 45˚

5˚

10˚

15˚

1000 2000 3000

SHEESHEE

GEWGEW

ANKEANKE

KAREKARE

WEE

KK

EEEEGTFGTF

NUNU

BOREEEBORE

DONEDODODONE

RF

UR

REEEEGTGTF

UU

MELEMELE

BEBE

MECEMECE

AWAEAWAEEE

LEMELEME

A

MEKEEMEK

DZDZ

UTEUTEAAA

KEKKK

ZEE

A

KEK

DZDZ

UU

Z

KOKOCHAECHAE

SENESENE

A

ININ

OTT

AA

NEEININE

MMEEMME

DIKEDIKEAA

ASEEAASEE

A

A

AMM

AA

Figure 3.6: Distribution of EAGLE I stations used for earthquake detection. The network was

divided into five sections and the datascan algorithm applied to four stations within each

section.

3.8 Arrival time measurements

First arrival times of P- and S-waves on earthquake seismograms were measured

using Seismic Handling Management (SHM) software (http://www.franken-

online.de/seismosite/) (Fig. 3.7). Arrival time of P- and S- phases were initially

measured on Phase I data bandpass filtered between 2-15Hz. Arrival times from Phase

II and Phase III seismic stations were added to the data for earthquakes that occurred

during the respective operation periods of these arrays. Arrival times for stations AAE,

FURI and WNDE measured by the Geophysical Observatory, Addis Ababa University

were also added to the dataset. The work of measuring phase arrival times was

predominantly by the author but data was processed by three other trained analysts (C.

Ebinger, A. Intawong and J. Pollatos).


40

38˚ 39˚ 40˚ 41˚

7˚

8˚

9˚

10˚

11˚

2003/01/10 12:13:56.132003/01/10 12:13:56.13N 8.6112 E 39.447N 8.6112 E 39.447ML 3.4ML 3.4

38˚ 39˚ 40˚ 41˚

7˚

8˚

9˚

10˚

11˚

MIEEMIEE

CHAECHAE

BOREBORE E82E82

12191219

10011001

38˚ 39˚ 40˚ 41˚

7˚

8˚

9˚

10˚

11˚

Figure 3.7: Example of unfiltered vertical component recordings of an earthquake located near

Boset volcano. Stations BORE, MIEE and CHAE are Phase I CMG-40T instruments, E82 is a

Phase II CMG-6TD, and 1219 and 1001 are Phase III CMG-6TD.

Uncertainty in the measured arrival times of both P- and S- waves caused by noise in

the data was estimated using a scheme of arrival time quality factors. Arrival times of

P-waves were made on vertical components only, and assigned a quality factor of 0, 1,

2 or 3 according to estimated measurement errors of 0.05 s, 0.1 s, 0.15 s and 0.2 s,

respectively (Fig. 3.8). Arrival times of the S-waves were made on the clearest

horizontal or vertical seismogram. Quality factors of 0, 1, 2 and 3 were assigned to

arrivals with estimated measurement errors of 0.1 s, 0.175 s, 0.25 s and 0.3 s,

respectively. A total of ~13388 P-wave and ~12725 S-wave arrivals were measured

from the 2139 earthquakes.


41

P0 P1

P2 P3

A B

C D

Figure 3.8: Examples of P-wave recordings for the various quality factors. A 120 s trace length

is displayed above each 3 s window at the P-wave onset. The bars show the time uncertainty of

the observation.

3.9 Hypocentre determination methods and errors in earthquake locations

The travel of a seismic wave is a non-linear function of both hypocentral parameters

and seismic velocities sampled along the ray path between station and hypocenter. In

standard earthquake location the seismic velocity model remains fixed and the

observed travel times are minimized by perturbing the hypocentre to minimize the root-


42

mean-square residual between observed travel-time and expected travel-time (RMS

travel-time residual) for the given velocity model. This procedure neglects the coupling

between hypocentre location, origin time and seismic velocity model in determining the

travel time of a seismic wave and error estimates will be dependant on choice of

velocity model. Precise hypocenter locations require the simultaneous solution of both

velocity and hypocentral parameters. The velocity model achieved with this method

leads to a minimum average of RMS values for all earthquakes and should reflect a-

priori information obtained by refraction and controlled source tomographic studies.

The 1-D P-wave velocity model and station corrections used to locate earthquakes in

the MER were determined by simultaneously relocating earthquakes and inverting for

velocity structure with VELEST (Kissling et al., 1995) (Fig. 3.9). Only earthquakes with

8 or more P-arrivals, an azimuthal gap of less than 180o, and an epicentral distance to

the nearest station of less than twice the focal depth were used to invert for the 1-D P-

wave velocity model. 280 earthquakes satisfied the selection criteria and can be

considered as ‘well-located’ earthquakes. Additional constraints on the 1-D P-wave

model were provided by the controlled source experiment (Mackenzie et al., 2005).

The 280 well-located earthquakes were subsequently relocated using a 3-D P-wave

velocity model determined with SIMULP (e.g., Eberhart-Philips and Michael, 1998;

Haslinger et al., 1999). Hypocenter accuracy of the earthquakes was tested by

relocating shots and randomly adjusting horizontal and vertical positions of

hypocenters. From these tests we estimate hypocenter accuracy for earthquakes of

about ± 600 m in horizontal directions and ± 2000 m in depth. Details of the local

earthquake tomography study are provided in Daly et al. (in review).

The Hypo2000 algorithm, which solves for earthquake location by iteratively converging

on a hypocentre that minimizes the RMS travel-time residual, was used to obtain

locations of all 2139 local earthquakes recorded at four or more stations. The minimum

1-D P-wave velocity model and Vp/Vs ratio of 1.75, calculated from P- and S-wave

travel-times, were employed. Arrival times were weighted according to the quality factor

assigned to the phase, with P-wave quality factors of 0, 1, 2 and 3 given full (1), 0.75,

0.5 and 0.25 weights respectively. S-waves were given half weighting relative to P-

waves of the same quality factor.


43

0

4

8

12

16

20

24

28

Dep

th (

km)

0 2 4 6 8P-wave velocity (km/s)

0

10

20

30

40

S-P

trav

el-t

ime

(s)

0 10 20 30 40 50P-wave travel-time (s)

(a) (b)

Figure 3.9: a) Minimum 1-D P wave velocity model determined by simultaneously inverting for

velocity model and hypocentres of 280 well located earthquakes. The minimum 1-D velocity

model was subsequently used to locate all 2139 earthquakes in the EAGLE dataset. b) Wadati

diagram of S-P travel times versus P-wave travel times for earthquakes in the MER. The

straight line shows the best fit to the data and represents a Vp/Vs ratio of 1.75.

3.10 Amplitude measurements and initial magnitude estimation

A summary of initial processing steps used for magnitude estimation is provided here.

However, chapter 5 describes full account of earthquake magnitude estimation and

magnitude statistics results. Local magnitude was initially estimated using the

maximum body wave displacement amplitudes (zero-to-peak) in mm, measured on a

simulated Wood-Anderson seismograph and distance correction terms of Hutton and

Boore (1987).

Waveforms were thus first corrected for the instrument response of the CMG-3T, CMG-

40TD and CMG-6TD seismometers and convolved with the displacement ground

motions with the nominal Wood-Anderson response using Seismic Analysis Code

(SAC) (http://www.llnl.gov/sac) (Fig. 3.10). The maximum peak-to-peak amplitude in

mm was measured on both horizontal traces. Peak-to-peak amplitudes were halved to

obtain a close approximation to maximum zero-to-peak amplitude. Stations with

malfunctioning horizontal components were removed from the dataset. The dataset of

2139 earthquakes provided 15456 amplitude measurements on each horizontal

component, a total of 30908 measurements.


44

Figure 3.10: Example of processing required for measurement of earthquake amplitudes. Top

panel: unfiltered velocity response on east-west and north-south components. Middle panel:

simulated Wood-Anderson displacement response on E-W and N-S components. Lower panel:

3 sec. window around maximum peak-to-peak amplitude on simulated Wood-Anderson

displacement response. The vertical scale is in mm.


45

3.11 Summary The EAGLE network provided a dense distribution of broadband seismic stations that

recorded earthquakes for 16 months from October 2001 - January 2003. The large

data set has been subjected to rigorous quality control procedures and is of high

quality. A trigger algorithm was used to identify local earthquakes in continuous seismic

data. The arrival times of P- and S-waves were measured and assigned quality factors

dependent on the estimated measurement error caused by noise. High quality arrivals

were given more weight during earthquake location and relocation. 2139 earthquakes

were recorded at four or more stations and located using a minimum 1-D velocity

model, determined using local earthquake tomography. Hypocentres of a subset of 280

well located earthquakes were determined using a 3-D velocity model and have

estimated location errors of ± 600 m in horizontal directions and ± 2000 m in depth.

Maximum zero-peak amplitudes of the earthquakes were measured on simulated

Wood-Anderson seismograms and then used to estimate earthquake magnitude.

Chapter 4 - Seismicity in the northern MER

46

Chapter 4

Seismicity in the northern Main Ethiopian rift

4.1 Introduction

The distribution of seismicity in an active rift that is transitional between continental and

oceanic in style constrains the pattern of strain localisation in the crust just prior to

continental breakup. Analysis of the broadband seismic data recorded by the dense

network of EAGLE stations in the MER provides a unique opportunity to obtain

accurate earthquake hypocentres in a volcanic rift setting that is near breakup. This

chapter presents the distribution of seismicity in the MER from October 2001 to

January 2003. The results are compared to patterns of Quaternary - Recent

deformation shown by other strain indicators at the surface such as Quaternary faults,

calderas, aligned cones and lava flows. The new seismicity data are also interpreted in

light of structural, seismic refraction / wide-angle reflection, gravity, anisotropy, and

crustal and mantle tomographic studies to propose that extension via magma injection

and minor faulting characterises the late stages of continental rifting prior to breakup.

4.2 Distribution of seismicity results

From October 2001 to January 2003, 2139 local earthquakes were recorded by the

EAGLE network (App. B). Of these, 1957 earthquakes were located within the network

of seismic stations (Fig. 4.1). Concentrated seismic activity occurs in the Fentale-Dofen

magmatic segment, which is a 20 km-wide, 70 km-long zone that extends from Fentale

caldera to Dofen volcano (Fig. 4.1). Earthquakes are located in a 10 km-wide, NNE-

trending zone that extends 40 km north of Fentale volcano where the pattern of

seismicity is mirrored by the surface expression of the closely spaced, small offset

Quaternary faults and fractures (Fig. 4.2). Three distinct earthquake clusters are

located near the Pliocene - Recent Dofen volcano (Figs. 4.1, 4.3). The distribution of

earthquakes located with the 3-D P-wave velocity model show that these clusters are

elongate ~N to ~NNE, parallel to the surface expression of major Quaternary fault

systems that cut lavas erupted from fissures (Fig. 4.3).


47

38˚

38˚

39˚

39˚

40˚

40˚

41˚

41˚

7˚ 7˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

11˚ 11˚

Aluto-Gedemsa MSAluto-Gedemsa MS

Boset-Boset-Kone Kone MSMS

Fentale-Fentale-Dofen Dofen MSMS



Arboye BFArboye BF


AnkoberAnkoberBFBF

Ambo FaultAmbo Fault

Addis Addis AbabaAbaba

38˚

38˚

39˚

39˚

40˚

40˚

41˚

41˚

7˚ 7˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

11˚ 11˚


1 2 3 4

38˚

38˚

39˚

39˚

40˚

40˚

41˚

41˚

7˚ 7˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

11˚ 11˚

1000 2000 3000 4000

Elevation

m

Figure 4.1: Seismicity of the MER from October 2001 to January 2003. Earthquakes were

located with the minimum 1-D P-wave velocity model determined from local earthquake

tomography. Only events recorded by at least 4 stations and located within the array of seismic

stations are displayed. Heavy black lines show major border faults; ellipses mark Quaternary

magmatic segments. The star shows the location of the October 2003 earthquake swarm near

Dofen volcano.


48

Figure 4.2: Example of seismicity located near Quaternary eruptive volcanic centres and faults

of the Fentale-Dofen magmatic segment plotted on a grayscale Landsat 741 image. The top left

inset shows the position of the image with respect to border faults and magmatic segments in

the MER.

The frequency-depth distribution of earthquakes within the Fentale-Dofen magmatic

segment located with the 3-D P-wave velocity model shows most earthquakes are 8-14

km depth (Fig. 4.3, 4.4). Hypocentre depths are 8-10 km deep near Fentale and Dofen

volcanoes but are up to 16 km deep in between these major eruptive centres (Fig. 4.3).

The temporal distribution of seismicity in the Fentale-Dofen magmatic segment is

characterised by earthquake swarms that punctuate largely aseismic periods (Fig. 4.5).

Minor seismicity is located within the Boset and Aluto-Gedemsa magmatic segments

(Fig. 4.1). Regions between the right stepping en-echelon magmatic segments are

largely aseismic.


49

Figure 4.3: Earthquake locations determined using the 3-D P-wave velocity model in the

Ankober region and Fentale-Dofen magmatic segment, plotted on 90 m resolution SRTM

topographic data. The earthquakes were recorded with 8 or more P-wave arrivals, have an

azimuthal gap of less than 180o, and an epicentral distance to the nearest station of less than

twice the focal depth. Profiles A-A’ & B-B’ project earthquakes within 30 km of the line of section

onto the profile. The thickened portions of the profiles show where the profile crosses the

Fentale-Dofen magmatic segment.


50

0

2

4

6

8

10

12

14

16

18

20

Dep

th (

km)

0 5 10 15 20

Number of Events

0

2

4

6

8

10

12

14

16

18

20

Dep

th (

km)

0 10 20 30 40 50 60

Number of Eventsa) b) Figure 4.4: Histograms of number of earthquakes per 1 km depth bin interval for the a) Fentale

- Dofen magmatic segment, and b) Ankober region. The hypocentres were located with the 3-D

P-wave velocity model and are displayed on Fig. 4.3.

Seismic activity south of the Aluto-Gedemsa magmatic segment is more diffuse than to

the north (Fig. 4.1). This rift sector lacks the narrow zone of localised faults and

eruptive centres characterizing the magmatic segments. Epicentres are located within

a 30-40 km-wide zone of Quaternary faults along the eastern side of the rift valley. The

amount of seismicity in this rift sector is relatively low and lacks the periods of swarm

activity observed further north in the Fentale-Dofen magmatic segment (Fig. 4.5).

The exception to the pattern of correlated seismicity and Quaternary eruptive centres

is the long-lived seismicity at the intersection of the NE-striking Miocene MER and N-

striking Oligocene Red Sea structures near Ankober (Figs. 4.1, 4.3). This intersection

zone has the highest relief in the region, with deeply incised valleys. Earthquakes are

localised in a N-S oriented cluster on the northwest margin of the rift valley at 9.5oN

39.75oE. The cluster lies at the southern end of the ~N-striking Ankober border fault

system, which is a series of closely spaced high angle normal faults and tight

monoclinal folds (Wolfenden et al., 2004). The rate of seismicity in this area was high

for the first 6 months of the experiment and characterised by frequent swarm activity

(Fig. 4.5). Focal depths are concentrated between 10 and 13 km with activity observed

down to 18 km (Fig. 4.3, 4.4).


51

˚ 39˚ 40˚

8˚

9˚

10˚

11˚


1

3838˚ 39˚ 40˚

8˚

9˚

10˚

11˚

1000

900

800

700

600

500

400

300

2

1

0Sep Nov Jan

Nov Jan Mar JulMay Sep Nov Jan

2000

1500

1000

500

2002 20032001

2002 20032001

Cum

ula

tive N

um

ber

(a)

(c)

(b)

Figure 4.5: a) Seismicity of the MER recorded by the EAGLE network with the three regions

that experienced the most activity highlighted, 1: Ankober area, 2: Fentale-Dofen magmatic

segment, and 3: south of Aluto-Gedemsa magmatic segment. b) Cumulative number of

earthquakes versus recording time of the regions 1, 2 and 3. c) Cumulative number of

earthquakes versus recording time of all the earthquakes recorded within the EAGLE network.


52

A minor, roughly E-W elongate cluster of earthquakes is located near Addis Ababa

(Fig. 4.1). The structure of this area is dominated by the E-striking Ambo lineament, a

fault zone active since the Late Miocene (Abebe et al., 1998). Isolated but relatively

deep earthquakes (15-21 km) characterise the remaining earthquake activity of the

Ethiopian plateau. The southeastern plateau shows a lack of activity except for a small

cluster on the southern margin of the Gulf of Aden rift at 9oN 40.5oE (Fig. 4.1).

4.3 Discussion

4.3.1 Pattern of seismicity on rift border faults

A striking feature of the recorded seismicity is the lack of earthquakes located on mid-

Miocene border faults that bound the MER and define the overall ~NE trend of the rift.

The inactivity of mid-Miocene border faults is reflected over longer time periods by the

minor geodetic strain on the rift flanks (Bilham et al., 1999) and lack of large magnitude

earthquakes on border faults over the last ~50 years (Ayele and Kulhánek, 1997). This

inactivity is inferred from historical records spanning the past 150 years (Gouin, 1979),

and morphology of the border faults (Boccaletti et al., 1998; Wolfenden et al., 2004).

The exception is the seismicity observed at the intersection between the N-striking Red

Sea rift and the NE-striking MER. The cluster of earthquakes is located on the N-

striking Ankober fault system that formed at ~11 Ma to link the two oblique rift systems.

Although fault and seismicity patterns show that the locus of strain has shifted to the

Quaternary magmatic segments in the central rift, this high point along the rift flank still

experiences strain (Wolfenden et al., 2004). The strike of the Ankober fault system is

oblique to the NE-trending MER and focused deformation in this complex zone of rift

intersection may be caused by flexure accommodating differential subsidence in the

Red Sea rift relative to the younger MER. Further north of Ankober, the Red Sea rift

margin is seismically active as shown in historical records, regional catalogues and

recent seismicity (Ayele et al., in press). Stress is concentrated in this area by the large

lateral density contrast and difference in lithospheric thickness between the uplifted

western Ethiopian plateau and Afar depression (e.g., Dugda et al., 2005; Tiberi et al.,

2005).


53

4.3.2 Seismicity in magmatic segments

A number of lines of evidence indicate that extensional strain is accommodated by a

combination of dyke injection and faulting within magmatic segments, as outlined

below. For the period 1960-2000, a comparison of the expected released seismic

moment and observed seismic moment shows that less than 50 % of extension across

the MER is accommodated by rapid slip on faults (Hofstetter and Beyth, 2003). At the

surface, GPS measurements show that approximately 80 % of present day extension

across the MER is localised in a ~20 km-wide zone of Quaternary faulting and

magmatism (Bilham et al., 1999). This narrow zone of localised deformation is also

observed in the brittle upper crust from patterns of seismicity. Elongate clusters of

earthquakes are associated with observed faults, fissures and active eruptive centres

in the Fentale-Dofen magmatic segment. The swarms of low magnitude earthquakes

are concentrated at 8-14 km depth which coincides with the top of the ~20-30 km-wide

zone of extensive mafic intrusions at 8-10 km depth (Keranen et al., 2004). Seismic

anisotropy of the upper crust is highest in the magmatic segments and attributed to

melt-filled cracks and dykes aligned perpendicular to the minimum stress (Keir et al.,

2005). Crustal strain across the MER is accommodated within the magmatic segments

by magma intrusion below ~10 km, and by both faulting and dyke intrusion in the brittle

seismogenic zone.

The Debre Zeit and Butajira chains of Quaternary eruptive centres located west of the

magmatic segments are largely aseismic, and they show little structural or

morphological evidence of active strain. Xenolith data and tomographic models show

these chains are underlain by hot asthenosphere (Bastow et al., 2005; Rooney et al.,

2005), but they lack the large relative positive Bouguer anomaly and high velocity crust

of the magmatic segments (e.g., Tiberi et al., 2005). These chains may be either

unfavourably oriented ‘failed’ magmatic segments, or incipient zones of strain.

In the magmatic segments of the MER, seismicity, geodetic and structural data all

show a localisation of strain in zones of Quaternary magmatism. The earthquakes in

the magmatic segments are concentrated above axial mafic intrusions and may be

induced by dyke injection. Models of the elastic stress field surrounding propagating

fluid-filled cracks show that earthquakes of magnitude > 1 can be induced ahead of a

propagating dyke if the ambient stress field is near to failure, and slip is likely to occur


54

along pre-existing fractures (Rubin and Gillard, 1998). Earthquake swarms are

assumed to occur near the crack tips due to the increasing stress caused by

concentrated internal fluids. Spatially, swarms reflect areas of magma intrusion. The

correlation we observe in the MER between seismic swarms and magma injection has

been documented near active volcanoes in other settings, suggesting the swarms are

causally linked to magma intrusion. For example, seismicity leading to the Mt. Etna

eruption of 2001 was characterised by swarms elongate parallel to surface fractures

and parallel to the maximum compressive stress determined from focal mechanisms

(Musumeci et al., 2004). This seismic activity was interpreted as being caused by dyke

emplacement prior to the eruption. By analogy to these other locales and independent

data from the MER, we propose that the observed seismicity in magmatic segments

above axial mafic intrusions is induced by magma injection into the mid- to upper crust.

4.3.3 Pattern of along-axis segmentation and episodic rifting

The along-axis segmentation of the MER is reflected at the surface by the right-

stepping en-echelon patterns of Quaternary faults and aligned cones within discrete 20

km-wide, 60 km-long magmatic segments. The pattern of seismicity interpreted in light

of other data provides clues as to the origin of this along-axis segmentation. At 8-10

km depth subsurface, the segmentation is evident as discrete axial mafic intrusions

imaged by crustal tomography (Keranen et al., 2004). These mafic bodies correlate

with along-axis velocity variations in the mid- and lower crust, implying that mafic

intrusions extend to the base of the crust (Maguire et al., 2006). Extension in the mid-

to lower crust is thus likely accommodated within a narrow zone of magma injection.

The onset of melting likely occurs in the lower crust and sub-continental lithosphere

(Rooney et al., 2005). The correlation between the orientation of lithospheric

anisotropy and the distribution of Quaternary strain and magmatism shows that

vertically oriented dykes with partial melt cross-cut the lithosphere (Kendall et al.,

2005). The concentrated seismicity in the Fentale-Dofen magmatic segment and

largely aseismic Boset-Kone and Aluto-Gedemsa magmatic segments is indirect

evidence that episodic rifting events within one magmatic segment are independent of

other magmatic segments. This suggests magma source regions are spatially and

temporally discrete.


55

The EAGLE network recorded seismicity for 15 months and thus provides a snapshot

of active deformation in the MER. During this time period, seismicity was particularly

concentrated in the Fentale-Dofen magmatic segment but the pattern of Quaternary

faults and fissures that cut recent lavas and historic earthquake data show that major

episodes of dyke injection and associated seismicity have been concentrated in other

magmatic segments in the past. Asfaw (1982) noted the development of surface

fissures following a swarm of ML<4 earthquakes near Fentale in 1981. Similar fissures

are observed in all magmatic segments along the axis of the MER and most likely

formed during previous rifting episodes (e.g. Asfaw, 1998; Williams et al., 2004). The

Boset-Kone magmatic segment was relatively inactive in 2001-2003. However, the

swarm of earthquakes reported near Nazret in 1964 (Gouin, 1979), an Mw 5.3

earthquake near Nazret in 1993 (Ayele, 2000) and fissuring of <10000 year old

ignimbrites at Kone (Williams et al., 2004) highlights the episodic nature of rifting in the

MER.

4.4 Summary

From Oct 2001 - Jan 2003, 1957 earthquakes were located within the EAGLE network

of broadband seismic stations in the northern Main Ethiopian rift and on its uplifted rift

flanks. Excluding the MER - Red Sea rift intersection zone at Ankober, seismicity within

the rift is localised to <20 km-wide, right-stepping, en echelon zones of Quaternary

magmatism. Seismicity in these magmatic segments is characterised by swarms of low

magnitude earthquakes located in clusters that parallel Quaternary faults, fissures and

chains of eruptive centres. The earthquakes in the magmatic segments are

predominantly <14 km deep and may be triggered by dyke injection. Seismic activity at

Ankober may be caused by flexure accommodating differential subsidence at the

oblique intersection of the <11 Ma MER and the older Red Sea rift. From integration of

these results with other geophysical and structural observations we propose that

present-day extension in the MER is localised to discrete <20 km-wide en echelon

magmatic segments, where extensional strain in the upper crust is accommodated by

both dyke intrusion and dyke induced faulting. The individual magmatic segments show

large spatial and temporal variations in level of seismicity over the time period of the

study, suggesting magma source regions for separate magmatic segments are

spatially and temporally discrete.

Chapter 5 - Local earthquake magnitude scale and seismicity rate

56

Chapter 5 Local earthquake magnitude scale and seismicity rate 5.1 Introduction

A calibrated local earthquake magnitude scale is essential for quantitative analyses of

seismicity. In Ethiopia, effective monitoring of earthquakes and resulting assessment of

seismic hazard are especially important as regions with seismic and volcanic activity

coincide with regions of economic significance and population growth. A new

magnitude scale is developed with the aim to quantify the size of local earthquakes in

our dataset by accurately estimating local earthquake magnitude (ML). The wealth of

broadband waveforms in the seismicity dataset allows for a direct inversion of

earthquake amplitude measurements for a local magnitude scale based on the original

definition proposed by Richter (1935, 1958). The calibrated magnitude scale is then

used to calculate the annual cumulative frequency-magnitude distribution of seismicity

in the MER.

5.2 Importance of a calibrated magnitude scale for Ethiopia

A calibrated earthquake magnitude scale based on ML is of great importance for

seismic hazard studies (Bormann, 2002). Attenuation curves that correct for the

decrease in seismic-wave amplitude with distance differ from region to region and the

use of an inappropriate curve can result in miscalculation of earthquake magnitude by

over 1 ML units, even at hypocentral distances of less than 300 km. Probabilistic hazard

analysis requires details of magnitude statistics (e.g. maximum magnitudes and the b-

value of the cumulative frequency-magnitude distribution), which require accurate

magnitude estimates to determine earthquake recurrence relationships. The

combination of the sparse station distribution, lack of a calibrated local magnitude scale

and low number of earthquakes recorded on global, regional and local catalogues has

meant that reliable earthquake magnitude statistics for the MER have not been

calculated (Ayele and Kulhánek, 1997). Earthquake magnitude is also important in

integrated seismic and geodetic studies that aim to understand lithospheric deformation

processes in rift systems by quantifying relative amounts of seismic and aseismic strain


57

(e.g. Hofstetter and Beyth, 2003; Bendick et al., 2006). The attenuation curve derived

from a local magnitude scale is also useful for risk assessment in engineering practice

as the frequency band of the Wood-Anderson seismometer (~0.8-10 Hz) is in the range

of most engineering structures. However, measurements on seismic-wave propagation

in Ethiopia are lacking (Kebede and van Eck, 1997; Mammo, 2005).

Seismic hazard assessment is important in Ethiopia as regions with seismic activity

coincide with regions of economic significance and population growth (e.g. Gouin,

1979; Kebede and Kulhánek, 1991; Kebede and Kulhánek, 1994). The potential

seismic and volcanic hazard in volcanic rift zones in Ethiopia was highlighted by the

recent rifting episode in the northern Afar rift. From 20 September - 8 October 2005,

162 earthquakes of mb > 4.0 and a volcanic eruption occurred within a ~ 60 km-long rift

segment. Disruption caused by ground-shaking, surface fissuring and ash deposits

caused the displacement of ~6000 pastoralists from the region (Yirgu et al., 2005).

Radar interferometry (InSAR) shows that ~8 m of horizontal opening occurred during

the rifting event with seismic moment release accounting for only ~9 % of extension

(Wright et al., 2006). The majority of extension was likely accommodated by dyke

intrusion.

Seismicity in Ethiopia is currently monitored by 5 permanent broadband seismic

stations, including the IRIS/GSN station FURI, maintained by the Geophysical

Observatory Addis Ababa University. The Geophysical Observatory record earthquake

coda-length, but they lack a formal method of measuring local earthquake magnitude.

In addition to the permanent stations, seismicity of north Afar is monitored by a network

of 9 three-component broadband seismic stations from October 2005 - April 2006. A

calibrated magnitude scale is critical for accurate quantitative monitoring of past,

ongoing and future seismic activity in Ethiopia.

5.3 Amplitude data

Local magnitude was originally defined by Richter (1935) using ground motions

recorded on a standard horizontal Wood-Anderson torsion seismograph. Therefore, the

EAGLE broadband data were corrected for the instrument response of the CMG-3T,

CMG-40TD and CMG-6TD seismometers. The displacement ground motions were

convolved with the standard Wood-Anderson response: magnification of 2800,


58

damping ratio of 0.8, and natural period of 0.8 sec. (Anderson and Wood, 1925;

Kanamori and Jennings, 1978). We measured the maximum absolute value of the

zero-to-peak amplitude in millimeters of the N-S and E-W horizontal component

seismograms. Stations with malfunctioning horizontal components were removed from

the dataset. The dataset of 2139 earthquakes provided 15456 amplitude

measurements on each horizontal component, a total of 30908 measurements (Figs.

5.1 and 5.2). The hypocentral distances considered range from 5 to 800 km, with the

best represented range being from 5 to 150 km (Fig. 5.1).

Figure 5.1: Distance/magnitude distribution of the data available for the horizontal components.

Magnitudes are estimated with the new distance correction terms for Ethiopia.


59

38˚ 39˚ 40˚ 41˚7˚

8˚

9˚

10˚

11˚

FURIFURI






Arboye BFArboye BF


AnkoberAnkoberBFBF

38˚ 39˚ 40˚ 41˚7˚

8˚

9˚

10˚

11˚

LEMELEME

SHEESHEE

GEWEGEWE

ANKEANKE

KOTEKOTE

MELEMELE MIEEMIEE

BEDEBEDE

MECEMECE

CHAECHAE

SENESENE

AREEAREEGTFEGTFE

BOREBORE

DONEDONE

INEEINEE

AMMEAMME

MEKEMEKEDIKEDIKE

ADEEADEE

ADUEADUE

DZEEDZEE

WOLEWOLE

ASEEASEE

BUTEBUTE

KAREKARE

HIREHIRE

38˚ 39˚ 40˚ 41˚7˚

8˚

9˚

10˚

11˚

36˚ 38˚ 40˚ 42˚ 44˚

6˚

8˚

10˚

12˚

14˚

1 2 3 4 5

1000 2000 3000 4000elevation m

Magnitude (ML)

Figure 5.2: Top panel: Distribution of the 2139 earthquakes recorded from October 2001 to

January 2003 in the MER and Afar rifts. Size of earthquake epicentres is scaled by magnitude.

The white star is the location of NEIC reported earthquake 2002/12/01/11:18 (mb PDE = 4.9).

The box encloses the location of the EAGLE network of broadband seismic stations. Lower

panel: Triangles are EAGLE stations (chapter 3). IRIS/GSN permanent broadband station FURI

is shown as a white square. Miocene border faults are shown with thick black lines and dip ticks.

Magmatic segments along the rift axis are shaded grey.


60

5.4 Methodology

We use the equation of Richter (1935, 1958)

ML = log(AWA) - log(Ao) + C, (1)

where AWA is zero-to-peak amplitude measured on a standard horizontal Wood-

Anderson seismograph, log(Ao) is a distance correction term and C are correction

terms for individual stations. We determine the attenuation curve, log(Ao), by using the

parametric approach (Bakun and Joyner, 1984). The major advantages of the

parametric form of the attenuation curve are that it considers simple expressions of

geometrical spreading and attenuation, and is represented by only a few coefficients.

This facilitates straightforward estimation of local magnitude using a single equation at

all hypocentral distances (e.g., Hutton and Boore, 1987; Kim, 1998; Kim and Park;

2005, Langston et al., 1998). On a global scale, the standardization of the local

magnitude calculation using the parametric form of the attenuation curve is

recommended (Ortega and Quintanar, 2005). A drawback of the parametric approach

is that the nonparametric expression of the attenuation curve better represents crustal

and upper-mantle complexities (e.g., Anderson and Lei, 1994; Baumbach et al., 2003;

Bragato and Tento, 2005; Savage and Anderson, 1995).

Richter’s original local magnitude scale is defined such that an earthquake of ML = 3

will cause a 1 mm zero-to-peak deflection of the Wood-Anderson seismogram at 100

km from the epicentre. Hutton and Boore (1987) point out that there are regional

differences in crustal attenuation and wave propagation that influence the attenuation

of S-waves in seismograms of local earthquakes. They suggest that local magnitudes

be normalized to motions at closer distances to avoid most of the regional differences

in wave propagation, using a 10 mm deflection of the Wood-Anderson seismogram at

17 km from the epicentre for a ML = 3 earthquake, consistent with the original definition

of the local magnitude scale. The distance correction term is thus defined as

-log(Ao) = nlog(r/17) + K(r-17) + 2, (2)

where n and K are parameters related to the geometrical spreading and attenuation of

S-waves in the region, respectively.


61

If equation (1) and (2) are combined, the observed amplitude, Aijk, is modeled by

log(Aijk) + 2 = -nlog(rij/17) - K(rij-17) + MLi - Cjk, (3)

where index i labels events, index j labels stations and index k labels the component

(NS / EW). The objective of the inversion is to determine n, K, ML and C. There are two

station factors per station corresponding to the E-W and N-S horizontal components.

The system of equations includes a constraint that the mean of station factors is zero.

The observations on the left-hand side of equation (3) are linearly related to the

unknowns, which we arrange in a model vector m. We have Ne events and Ns stations,

and thus have a total of (Ne + 2Ns) + 2 unknowns. The N observations log(Aijk) + 2 are

arranged into the N-vector d. We write the overdetermined set of equations (3) in the

form

d = Am, (4) which we solve using the conventional least-squares criterion; the optimal solution

satisfies

m = (ATA)-1ATd (5) The linear system (5) has a total of 2385 parameters and the 30908 data; it can be

solved in less than an hour on a modest workstation. Our approach leads directly to an

optimal solution and is different from the iterative procedure used to determine m (e.g.,

Langston et al., 1998). Pujol (2003) tested the direct inversion method on data from

Tanzania previously analyzed with the iterative technique (Langston et al., 1998).

Similar results were achieved but the major advantages of the direct inversion are that

the solution is independent of the starting values for the unknowns.

5.5 Results 5.5.1 Magnitude scale for the MER

The distance correction, log(Ao), term from the inversion using 17 km distance

normalization is given by:


62

-log(Ao) = 1.196997log(r/17) + 0.001066(r-17) + 2 (6)

where r is hypocentral distance in kilometers (Fig. 5.3).

The errors on the estimates of n and K can be determined from the posterior

covariance matrix. The two-by-two subsection of the entire 2385 by 2385 covariance

matrix is practically diagonal, showing that the estimates of n and K are virtually

independent. One can rigorously characterize this by calculating the eigenvectors of

the (n, K) section of the matrix; the ellipse describing the 1-standard deviation contour

has a semi-major axis of length 0.025 and semi-minor axis of length 9.7 x 10-5 and is

oriented with the semi-major axis practically parallel to the n axis (the angle between

them is 0.25o). These values for the ellipse lengths are similar to the values for the 1-

sigma standard deviations of the parameters n and K.

0

1

2

3

4

5

6

0 100 200 300 400 500 600 700 800 900 1000

Southern California (Hutton and Boore, 1987)

Ethiopia

Tanzania (Langston et al., 1998)

Southern California (Richter, 1958)

0

1

2

3

4

5

6

-lo

g(A

o)

0 100 200 300 400 500 600 700 800 900 1000

Hypocentral Distance (km) Figure 5.3: Attenuation curves for southern California (Hutton and Boore, 1987) - black

squares, California (Richter, 1935) - grey stars, Tanzania (Langston et al., 1998) - grey

triangles, and that derived for Ethiopia from our study (grey circles).


63

3

4

5

6

Mean = 4.98

mb (PDE) = 4.9

a

3

4

5

6

3

4

5

6

Mean = 4.13

mb (PDE) = 4.9

b

3

4

5

6

3

4

5

6

100 200 300 400 500 600 700

Mean = 4.79

mb (PDE) = 4.9

c

3

4

5

6

100 200 300 400 500 600 700Hypocentral Distance (km)

Ma

gn

itu

de

(M

L)

Ma

gn

itu

de

(M

L)

Ma

gn

itu

de

(M

L)

Figure 5.4: Magnitude estimated at stations of varying hypocentral distances for earthquake

2002/12/01/11:18 (mb PDE = 4.9) with three different attenuation curves, a: magnitude

estimated with the distance correction terms of Hutton and Boore (1987) for southern California,

b: magnitude estimated with the distance correction terms of Langston et al., (1998), c:

magnitude estimated with the new magnitude scale for Ethiopia. Straight lines are best-fit to the

data and show that the Hutton and Boore (1987) magnitude scale over-estimates magnitudes

with increasing hypocentral distances whereas the Langston et al. (1998) scale under-estimates

magnitude with increasing hypocentral distance. The new magnitude scale for the MER

estimates consistent magnitudes across varying hypocentral distances and the average local

magnitude (ML = 4.79) is that expected for an mb = 4.9 earthquake.


64

The new distance correction terms compensate correctly for the reduction in amplitude

with increasing distance (Figs. 5.4, 5.5). For example, earthquake 2002/12/01/11:18

(mb PDE = 4.9) was the nearest and most widely recorded earthquake on the EAGLE

network that was reported by NEIC. The earthquake is located ~200 km north of station

KARE and was recorded with a high signal-to-noise ratio by 72 EAGLE broadband

stations at hypocentral distances of 200 - 600 km (Fig. 5.4). Local magnitude was

estimated at each station using the magnitude scale for Tanzania (Langston et al.,

1998), southern California (Hutton and Boore, 1989) and Ethiopia. The magnitude

estimated at each station using the Tanzania magnitude scale decreases with

increasing hypocentral distance and the average magnitude is ML = 4.13. In contrast,

the magnitude at each station using the southern California scale increases with

increasing hypocentral distance and the average magnitude is ML = 4.98. The new

magnitude scale for Ethiopia estimates consistent magnitude for stations at different

hypocentral distances and the average magnitude is ML = 4.81.

-0.4

-0.2

0.0

0.2

0.4

0 100 200 300 400 500 600 700 800Hypocentral Distance (km)

Mean M

L R

esid

ua

l

Figure 5.5: Mean magnitude residuals per 50 km bin intervals with error bars marked by the

standard deviation in the mean magnitude residuals. Magnitude residuals are the difference

between magnitude assigned by a single station and the average magnitude of the earthquake.

The mean magnitude residual, the difference between magnitude at a single station

and average magnitute, is calculated per 50 km bin interval with error bars marked by

the standard deviation of the mean magnitude residuals (Fig. 5.5). Mean magnitude

residuals vary ± 0.1 ML to hypocentral distance of 700 km. An ML residual of 0.18 is

calculated from only 16 measurements at hypocentral distances of >700 km. The lack

of significant variation in mean magnitude residuals with distance shows that possible


65

complexities in crustal and upper mantle structure do not have a systematic effect on

variations in attenuation with distance beneath the MER. Our parametric expression of

the attenuation curve thus represents a simple model that adequately compensates for

the decay of amplitude with increasing distance.

5.5.2 Local magnitude values and station corrections The inversion procedure solved for correction factors of both N-S and E-W components

at individual stations. The N-S component correction factors vary between 0.41 to -0.34

ML units and the E-W component correction factors vary between 0.42 to -0.33 ML units

(Fig. 5.6). Most stations have similar correction factors on the two horizontal

components. Station corrections can vary dramatically over distances of ~5km and

there is no consistent difference between corrections at stations in the rift valley and on

the adjacent plateau. Thus, the spatial variation of station factors shows neither a clear

correlation to major tectonic features nor to topographic relief and suggests a strong

influence of local site effects on variations in the amplitude of ground motion.

38˚ 39˚ 40˚ 41˚

7˚

8˚

9˚

10˚

11˚38˚ 39˚ 40˚ 41˚

7˚

8˚

9˚

10˚

11˚

-0.4 -0.3 -0.2 -0.1 0.1 0.2 0.3 0.4

FURI FURI

a b

N-S E-W

Station Correction

Figure 5.6: Spatial variation of station factors on a: N-S component and b: E-W component.

Negative correction factors are shown as squares scaled by magnitude of the correction factor.

Positive correction factors are shown as circles scaled by magnitude of the correction factor.


66

Magnitude residuals were calculated with and without computed station corrections, C,

taken into account (Fig. 5.7). For magnitude residuals calculated without station

correction, the average of residuals on the N-S and E-W components is nearly zero

and the standard deviation is 0.24 (variance, σ 2, is 0.058). For magnitude residuals

calculated with station corrections, the average of residuals on the N-S and E-W

components in nearly zero, and the standard deviation is 0.18 (σ 2 is 0.032). Therefore,

adopting the station corrections reduced variance by 45%.

Figure 5.7: a: Magnitude residuals / hypocentral distance distributions for both the N-S and E-W

components. a: magnitude residuals without taking into account station corrections. The

variance is 0.054. b: magnitudes residuals with station corrections taken into account. The

variance is 0.032. Therefore, adopting the station corrections reduced variance by 45%. The

average of residuals, both with and without station corrections considered, is nearly zero.


67

The station correction factors for the permanent IRIS/GSN station FURI are -0.16 ML

units on the N-S component and -0.14 ML units on the E-W component. Future

permanent and temporary seismic array deployments in Ethiopia are likely to include

earthquake records from FURI, and such studies will now be able to use amplitude

measurements from this permanent station to calibrate new data with our magnitude

scale for the MER.

5.6 Discussion 5.6.1 Seismic attenuation

A comparison of the attenuation curves obtained from Ethiopia, southern California

(Hutton and Boore, 1987), and Tanzania shows that attenuation in Ethiopia is relatively

high. The attenuation curve computed for southern California by Hutton and Boore

(1987) is similar to Ethiopia, particularly at hypocentral distances of less than 300 km.

Rifted regions with elevated geothermal gradients such as the southwestern United

States are generally characterised by high attenuation of seismic waves (e.g., Hutton

and Boore, 1989; Richter, 1958; Savage and Anderson, 1995). In southern California,

high body-wave attenuation is attributed to a combination of high temperatures and

partial melt in the crust resulting from active rifting (Schlotterbeck and Abers, 2001).

Our results of relatively large amounts of attenuation in the MER is thus not surprising

considering the wealth of independent geophysical and geological data that shows

evidence for partial melt and magma intrusions in the crust and upper mantle beneath

the MER and adjacent Ethiopian plateau (e.g., Bastow et al., 2005; Keir et al., 2005;

Kendall et al., 2005; Rooney et al., 2005).

The high attenuation observed in Ethiopia is significantly different to the East African rift

system in Tanzania where the crust and upper mantle have had little to no modification

by rifting processes (Langston et al., 2002). In Tanzania, the combination of crystalline

Archaean and Protorozoic crust, in conjunction with low geothermal gradients typical of

Archaean craton give rise to very efficient wave propagation in the lithosphere

(Langston et al., 1998; Weeraratne et al., 2003).

5.6.2 Magnitude statistics and annual-cumulative seismicity rate


68

The new magnitude scale for Ethiopia is used to investigate seismicity of the MER for

2001-2003. Due to high attenuation in the MER, earthquakes located outside the

network but recorded on EAGLE stations are ML > ~3 (Figs. 5.1 and 5.2). Therefore,

we calculated magnitude statistics of the 1957 earthquakes located within the network

of seismic stations. This ensures we sample an earthquake catalogue that is not biased

towards large earthquakes located outside the network and also ensures our

magnitude statistics sample only earthquakes in the MER. The majority of earthquakes

are of ML 1-2 and the largest earthquake is ML 3.9 (Fig. 5.8a). The power-law

cumulative frequency-magnitude distribution shows that the seismicity catalog is

complete above ML 2.1 (Fig. 5.8b) (Gutenberg and Richter, 1954). A b-value of 1.13 ±

0.05 was estimated from earthquakes larger than the ML = 2.1 using the maximum

likelihood method (Aki, 1965) and an error estimate determined from the standard

deviation of b (Shi and Bolt, 1982). The cumulative annual seismicity rate is calculated

from an annualized dataset and follows the relation log N = 4.5 - 1.13ML. Hofstetter and

Beyth (2003) obtained a b-value of 0.83 ± 0.08 from just 16 earthquakes on global and

regional catalogues that were located across a larger area that encompasses both the

MER and southern Ethiopian Rift as far south as 5oN.

0

1

2

3

0 1 2 3 4 5

McMc

0

50

100

150

200

250

300

0 1 2 3 4Magnitude (ML) Magnitude (ML)

Nu

mb

er

of E

art

hq

ua

ke

s

Log A

nn. C

um

u. N

um

. of E

art

hquakes

a b

Figure 5.8: a: Magnitude-frequency distribution of earthquakes recorded within the network of

seismic stations. The majority of earthquakes are of magnitude ML 1-2 and the highest

magnitude earthquake is ML = 3.9. b: Gutenburg-Richter distribution of earthquakes located

within the network of seismic stations. Mc is the cut off magnitude of 2.1 and the slope shows b

= 1.13 ± 0.05. The straight line intersects the y-axis at y=4.5.


69

The relatively high b-value of 1.13 ± 0.05 for seismicity in the MER during 2001-2003

shows that seismic energy is released mostly as swarms of lower-magnitude (ML < 4)

earthquakes. Previous studies of seismicity in the MER also report lower-magnitude

seismic swarms and a similar pattern is evident in data from global and regional

catalogues that show relatively few larger magnitude (mb > 4.0) earthquakes in the

MER. (e.g. Asfaw, 1982; Gouin, 1979; Kebede and Kulhánek, 1994). The observed

lack of large magnitude earthquakes in the MER is consistent with geodetic data that

show that the majority of strain across the MER is accommodated aseismically

(Bendick et al., 2006; Hofstetter and Beyth, 2003).

An estimated b-value of 1.13 for the MER is similar to b-values of 1.05 - 1.3 calculated

for the southern Red Sea and Gulf of Aden sea-floor spreading centres (Ayele and

Kulhánek, 1997; Hofstetter and Beyth, 2003). Lower b-values of between 0.7-0.9 are

observed in the East African rift system in southern Ethiopia, Kenya and Tanzania

where moment release as large magnitude earthquakes located on rift bounding border

faults accommodate the majority of extension (e.g., Tongue et al., 1992; Langston et

al., 1998).

5.6.3 Seismic and volcanic hazards in Ethiopia

Despite the lack of large earthquakes recorded over the past ~50 years in the MER,

the recent dyke injection episode and associated swarm of earthquakes, surface

fissuring and volcanic eruption in Afar highlights the potential seismic hazard of rift

zones in Ethiopia. Seismicity within the magmatic segments of the MER is likely

controlled by episodic injection of dykes (Keir et al., 2006). Although a major rifting

event has not yet been directly observed in the MER, structural data suggest that

episodes of surface fissuring and volcanic eruptions have occurred in MER magmatic

segments during the last ~10000 years (e.g., Asfaw, 1982; Asfaw; 1998; Williams et al.,

2004). Despite the current period of quiescence, hazards associated with seismicity

and volcanic eruptions pose a serious risk to life and economy in the MER.

5.7 Summary

A local magnitude scale for Ethiopia has been developed from 30908 amplitude

measurements on simulated Wood-Anderson seismograms from 2139 earthquakes


70

recorded on 122 EAGLE broadband instruments. The new magnitude scale uses a

distance normalization of 10 mm motion at 17 km distance for a ML 3.0 earthquake and

shows that ground-motion attenuation in Ethiopia is relatively high and is consistent

with the presence of pervasive magma intrusion and partial melt beneath the MER. The catalogue of events used in this study is complete above ML 2.1 and the annual

cumulative seismicity rate in the MER is log N = 4.5 - 1.13ML. The relatively high b-

value is expressed in the observed pattern of low magnitude ML < 4 swarms of

earthquakes in the MER and lack of large magnitude earthquakes reported on global

and regional catalogues over the last ~50 years. The new calibrated magnitude scale is

critical for current and future quantitative analysis of seismicity in Ethiopia, which is

important for scientific, economic and social development.

Chapter 6 - Style of faulting and stress field orientation

71

Chapter 6

Style of faulting and stress field orientation determined from earthquake focal mechanisms

6.1 Introduction

Despite numerous structural studies (e.g., Acocella and Korme, 2002; Wolfenden et al.,

2004; Casey et al., 2006), the dominant style of faulting and direction of extension in

the MER remains controversial. Prior to EAGLE, the seismic network in Ethiopia was

sparse and thus earthquake source parameters were determined for a low number of

large magnitude earthquakes from recordings at regional and teleseismic distances.

The dense network of EAGLE seismic stations provides a wealth of three-component

broadband seismic data with which to determine well constrained earthquake source

parameters, including hypocentral location and focal mechanism, for earthquakes of

too low a magnitude to be recorded by stations at regional distances. This chapter

presents well constrained focal mechanisms of 33 earthquakes located in the MER that

are then used to invert for the regional stress tensor. Focal mechanisms are related to

structural data and the orientation of the regional stress tensor is compared to geodetic

data, as well as global and local plate kinematic models.

6.2 Determination of focal mechanisms

Focal mechanisms were computed from P- and SH- wave polarities using the grid

search algorithm FOCMEC (Snoke et al., 1984). A double-couple source type is

assumed as all the events selected are characterized by high frequency content, sharp

first-arrivals and clear S-phases at the nearest stations (Figs. 6.1, 6.2). Hypocentre co-

ordinates were determined by locating the event with the 3-D velocity model. A fault

plane solution was only attempted if an earthquake was located within the network, the

nearest station was within an epicentral distance of twice the focal depth, and the

solution had a minimum of 10 P-wave polarities located in at least 3 quadrants of the

focal sphere. Polarity errors of neither P- nor SH-waves were tolerated in the grid

search algorithm. In total, 33 well constrained and unambiguous fault plane solutions

that have a maximum 20o uncertainty in either strike or dip of both nodal planes were


72

determined (Table 6.1). This new dataset is supplemented by the 3 well-constrained

focal mechanisms in the MER determined from data at regional and teleseismic

distances (Ayele, 2000; Hofstetter and Beyth, 2003; CMT, Harvard).

b

c d

k

n

j

a

h

g

i

l

m fe

a b

c

d

e

fghi

j

k

l

m n

b

f

c

ed

a

a

b

c

d

e

f

P-wave polarities

SH-wave polarities

0 1s

0 2s Figure 6.1: Unfiltered data used to determine the focal mechanism of earthquake 2003/01/02

08:52:45.77. The earthquake is located at 40.013oN, 9.244oE and ML = 2.4. Fault slip

parameters are strike = 195o, dip = 65o, rake = -90o. Focal mechanisms are lower hemisphere

projections of the focal sphere. P-wave polarities; shaded regions show compressional (upward)

P-wave first motions and unshaded regions are dilatational (downward) P-wave first motions

measured on vertical seismogram. SH-wave polarities; the focal sphere is divided into regions

of downward and upward first motions of the S-wave measured on transverse component

seismograms.


73

w

l

x

gij

m

a

o k

u

nh

tv

p

s cq

f

er b

d

a b c

d e

f g

h i

jklo

p

r

t

v

n

q

s

u

w

m

x

o

q

cd

a

p

b

xw

n h

r

g

e

im

sv

k

u

f

t

jl

a b c

d e

f g

h i

jklo

p

r

t

v

n

q

s

u

w

m

x

0 1s

0 2s

P-wave polarities

SH-wave polarities

Figure 6.2: Data used to determine the focal mechanism of earthquake 2003/01/10

12:13:56.13. The earthquake is located at 39.447oN, 8.6112oE and ML = 3.4. Fault slip

parameters are strike = 42.64o, dip = 85.25o, rake = 13.19o. Focal mechanisms are lower

hemisphere projections of the focal sphere. P-wave polarities; shaded regions show

compressional (upward) P-wave first motions and unshaded regions are dilatational (downward)

P-wave first motions measured on vertical seismogram. SH-wave polarities; the focal sphere is

divided into regions of downward and upward first motions of the S-wave measured on

transverse component seismograms.


74

6.3 Method of inverting focal mechanisms for the regional stress tensor

The focal mechanisms were used to invert for the regional stress tensor with the linear,

least squares stress inversion method of Michael (1984) that minimizes the angle

between the predicted tangential traction on the fault plane and the observed slip

direction. The 95% confidence regions were determined with the bootstrap re-sampling

method (Michael, 1987a; 1987b) and used as an error estimate of the stress tensor.

The relatively small dataset and estimated focal mechanism errors of this study make

this method most appropriate to both accurately determine the stress orientation and

adequately estimate the confidence limits (Hardebeck and Hauksson, 2001).

The inversion procedure assumes that the four stress parameters are constant over the

spatial and temporal extent of the data set and that earthquakes slip in the direction of

the resolved shear stress on the fault plane (Michael, 1984). The uniform stress tensor

that best explains the mechanisms is expressed by the three principal stress axes

(where σ 1, σ 2 and σ 3 are the maximum, intermediate and minimum principal

stresses respectively) and the stress ratio. An average misfit angle β , which measures

the difference between the observed slip direction and the predicted direction of

maximum tangential traction, is computed and used as a measure of the success of the

inversion. The steepest nodal plane of the normal fault focal mechanisms and ~NE-

striking nodal planes of the strike-slip mechanisms were chosen as fault planes for the

inversion in accord with geological observations (e.g., Abebe et al., 1998; Wolfenden et

al., 2004; Casey et al., 2006; Pizzi et al., 2006).

6.4 Focal mechanism results

In total, 33 well constrained and unambiguous fault plane solutions that have a

maximum 20o uncertainty in either strike or dip of both nodal planes were determined

(Table 6.1, Figs. 6.3, 6.4). This new dataset is supplemented by the 3 well-constrained

focal mechanisms determined from data at regional and teleseismic distances (Ayele,

2000; Hofstetter and Beyth, 2003; CMT, Harvard) (Table 6.2, Fig. 6.4).


75

Table 6.1: Earthquake source parameters determined from EAGLE data.

Table 6.1 and 6.2 column header definitions

Date: date of earthquake as year/month/day

Time: origin time of earthquake as hour:minute:second

Lat. (oN): latitude of earthquake location

Long. (oE): longitude of earthquake location

Z (km): earthquake depth in kilometres

Strike: strike of nodal plane in degrees from north

Dip: dip of nodal plane in degrees from horizontal

Rake: rake of nodal plane in degrees

ML: local magnitude

MW: moment magnitude

Date (yr/mo/dy)

Time (hr:mn:sc)

Lat. (oN)

Long. (oE)

Z (km) Strike Dip Rake ML

2002/01/16 21:22:39.44 9.239 40.021 13.3 180.0 50.0 -90.0 1.72002/01/17 01:38:03.91 8.154 39.002 20.3 2.3 60.1 -93.5 2.012002/01/18 01:42:40.83 8.998 39.918 10.2 359.7 54.2 -97.4 2.822002/02/17 02:38:15.44 9.470 39.692 11.9 171.5 66.0 -90.0 3.212002/05/02 21:43:23.17 9.122 39.984 13.2 211.6 56.4 -80.4 2.642002/07/04 02:59:42.35 9.173 39.966 15.8 214.4 60.1 -85.4 3.542002/07/31 01:54:38.27 9.444 39.677 11.3 172.8 66.1 -85.6 2.342002/08/21 01:27:23.93 8.951 39.711 13.9 192.9 60.1 -84.2 2.142002/10/08 19:37:43.42 9.199 39.949 12.7 225.7 68.1 -85.7 2.032002/10/09 18:19:37.91 9.193 39.987 12.5 223.1 68.2 -64.0 2.142002/10/10 19:15:51.93 9.066 39.965 14.6 201.5 58.3 -66.3 1.172002/10/19 21:25:25.96 10.130 39.957 15.5 198.1 59.4 -71.3 2.832002/11/04 00:17:42.49 8.432 39.673 12.9 2.5 51.2 -83.6 1.172002/11/04 00:24:55.49 7.812 38.976 6.9 183.8 63.3 -109.1 1.712002/11/05 22:42:14.69 9.728 39.370 14.7 29.3 66.4 -79.1 1.922002/11/07 01:24:31.21 9.492 40.040 15.5 216.3 46.0 -74.6 1.892002/12/03 16:02:52.26 7.481 38.553 13.6 183.7 68.0 -92.2 2.552002/12/03 20:10:01.33 7.700 38.911 12.4 190.0 45.0 -90.0 2.342002/12/04 13:41:09.57 8.873 39.836 9.6 209.9 60.0 -90.0 1.972002/12/13 17:36:21.66 9.494 40.034 15.8 183.7 64.3 -98.9 2.22002/12/15 08:37:35.26 7.428 38.648 8.6 198.0 50.0 -90.0 3.062002/12/15 19:15:38.82 7.430 38.657 6.4 210.5 70.4 -78.3 2.892002/12/15 20:35:05.22 9.548 40.144 19.0 190.6 66.6 -103.1 1.932002/12/17 22:12:36.10 9.001 39.907 8.4 64.5 88.2 -0.8 1.42002/12/17 23:15:10.76 8.998 39.901 9.2 72.0 80.7 -3.8 1.552002/12/23 06:27:49.95 9.446 39.680 10.4 182.0 60.0 -90.0 2.45


76

2002/12/26 19:47:51.98 9.221 40.014 13.0 213.2 62.0 -87.7 3.172002/12/26 19:55:17.90 9.221 40.011 12.7 219.9 60.0 -90.0 2.412003/01/02 08:52:45.37 9.246 40.013 13.9 195.0 65.0 -90.0 2.372003/01/10 12:13:56.08 8.611 39.447 7.0 42.6 85.3 13.2 3.442003/01/13 21:06:00.76 9.491 39.681 11.2 168.5 56.2 -97.2 1.972003/01/20 21:16:22.90 7.475 38.823 11.7 206.1 43.0 -104.8 2.522003/01/21 08:08:18.85 7.495 38.822 11.4 197.8 37.2 -117.2 2.9

Table 6.2: Earthquake source parameters determined in other studies.

Date (yr/mo/dy)

Time (hr:mn)

Lat. (oN)

Long. (oE) Strike Dip Rake Mw Data Source

1983/12/28 23:08 7.03 38.60 176 51 -81 5.3 Harvard CMT 1993/02/13 02:25 8.33 39.91 221 87 -7 4.9 Ayele, 2000 1995/01/20 07:14 7.16 38.44 9 49 -119 5.0 Hofstetter & Beyth,2003

020704 02:59020704 02:59

T

P

021203 20:10021203 20:10

TP

021217 22:12021217 22:12

T

P

021217 23:15021217 23:15

TP

021226 19:47021226 19:47

T

P

021226 19:55021226 19:55

T

P

030102 08:52030102 08:52

TP

030110 12:13030110 12:13

T

P Figure 6.3: A selection of focal mechanisms from this study. Compressional P-wave first

motions are plotted as circles and dilational first motions are plotted as triangles. The

compressional quadrants of the focal sphere are shaded black. Each solution is labelled by

earthquake origin time GMT (year, month, day, hour, minute).


77

38˚

38˚

39˚

39˚

40˚

40˚

41˚

41˚

7˚ 7˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

Quaternary faultsand eruptive centresand mid-Mioceneborder faults

RedSearift

AnkoberBF

ArboyeBF

GuragheBF

AddisAbaba

Nazr

La

e

Fentale


segment AselaA -Sire BF

Bosetmagmatic

BosetDebre Zeit chainD b Z it h i

an

GadimottG

Ambo lineament

AtayeBF

Ayelu- Abida

YardiLake

AwashR.

Caldera Lakes



aAlutoAluto

Figure 6.4: Faults that cut <1.9 My lavas, and Quaternary eruptive centres comprising

magmatic segments, relative to the Miocene border faults bounding Main Ethiopian rift basins

(after Casey et al., 2006). Fault plane solutions are lower hemisphere projections. The size of

the solution is scaled to magnitude between ML 1.17 - 5.3.

Focal mechanisms of earthquakes located along the axis of the MER and in the

Ankober fault system show predominantly normal dip-slip on steep faults that strike ~N

to ~NNE (Figs. 6.4, 6.5). Focal mechanisms are sub-parallel to the dominant N10oE

orientation of Quaternary faults in the Ethiopian rift (Boccaletti et al., 1998; Wolfenden

et al., 2004; Casey et al., 2006) (Fig. 6.4). A few of the normal dip-slip focal

mechanisms have slip planes that strike ~NE, parallel to the pre-3.5 Ma, N40oE-

striking faults (Fig. 6.5). The exceptions to these normal dip-slip focal mechanisms are


78

the strike-slip earthquakes below Fentale and Boset volcanoes, interpreted as left-

lateral motion on ~NE- to ~ENE-striking faults (Fig. 6.5). However, both normal and

strike-slip focal mechanisms show near horizontal T-axes striking N80oE-N130oE (Figs.

6.3, 6.5).

0˚30˚

60˚90˚

120˚

150˚

180˚

210˚

240˚

270˚

300˚

330˚

SOUTH 9

EASTWEST

NORTH0˚

30˚

60˚90˚

120˚

150˚

180˚

210˚

240˚

270˚

300˚

330˚

SOUTH 9

EASTWEST

NORTH

0˚30˚

60˚90˚

120˚

150˚

180˚

210˚

240˚

270˚

300˚

330˚0˚

30˚

60˚90˚

120˚

150˚

180˚

210˚

240˚

270˚

300˚

330˚

σ 3

σ 1

σ22

(a)(a) (b)(b)

(c)(c) (d)(d)

Figure 6.5: a) Rose diagram of the orientation of the T-axes of earthquake focal-mechanisms.

b) Rose diagram showing the strike of earthquake slip planes. c) Lower hemisphere plot of the

trend and plunge of fault plane solution T-axes (dark circles) and P-axes (light triangles). d.)

Results of the stress tensor inversion. Circle shows σ 3, the minimum compressive stress.

Square shows σ 2, the intermediate compressive stress. Triangle shows σ 3, the maximum

compressive stress. 95 % confidence limits are shown by regions of grey shading.


79

6.5 Stress tensor results

The results of the stress inversion using the 36 focal mechanisms in the MER show

that the trend / plunge of the minimum principal stress is 283o / 6o with a mean misfit

angle ( β ) ± standard deviation of 10.9o ± 7.0o (Fig. 6.5). This mean misfit angle is

comparable to results of stress tensor inversions from focal mechanisms within uniform

stress fields in other studies: 10-17o along fault segments of the San Andreas fault

zone (Jones, 1988); and 6-24o for datasets in the Swiss Alps and northern Alpine

foreland (Kastrup et al., 2004). However, a well resolved stress tensor requires that the

dataset contains a diverse range of focal mechanisms. In our dataset, only 4 strike-slip

focal mechanisms differ from the predominant dip-slip on ~N to ~NE-striking faults.

This unavoidable lack of diversity in type of focal mechanism reduces the resolution of

the stress tensor.

6.6 Quaternary volcanoes and faults as strain indicators Earthquake focal mechanisms show the slip directions of active faults but the relatively

small size of the dataset, errors associated with computing the focal mechanism, and

assumptions made during the inversion for the stress tensor limit the resolution the

results. Quaternary deformation is observed at the surface as faulting and extrusive

volcanism. These strain indicators are thus used as independent data with which

earthquake focal mechanisms are compared and interpreted. Quaternary calderas,

aligned volcanic cones and faults were mapped using Landsat Thematic Mapper (30

m), and ASTER imagery (15 m resolution). Only faults of length greater than 200 m

that were clearly distinguishable from felsic and basaltic flow fronts were included in the

analysis. The orientation of the long-axis of the elliptical Quaternary calderas Boset,

Fentale and Dofen was also estimated.

Quaternary calderas and lava flows in the magmatic segments are predominantly cut

by N10oE to N30oE striking normal faults (Figs. 6.6, 6.7, 6.8, 6.9). A small subset of

faults striking N45-70oE link arrays of ~NNE striking faults within the magmatic

segments (Figs. 6.6, 6.8, 6.9) and may utilize ~NE striking Miocene-Pliocene faults.

Fault systems do not link the right-stepping magmatic segments. The long-axes of the

Quaternary calderas are; Boset - N107oE , Fentale - N109oE, Dofen - N110oE (Fig.

6.8).


80

39˚

39˚

40˚

40˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

RedSearift

MERAnkoberBF

ArboyeBF

AddisAbaba

et

LakeKoka

Fentale


Aluto-Gedemsamagmaticsegment AselaA -

Sire BF

Bosetmagmaticsegment

Dofenen

Kone

BosetBo toset

Gademsa

Debre Zeit chain



aAlutoAluto

Fig. 6.8. 6.8

Fig. 6.7F

37˚11˚

37˚11˚

37˚11˚

35˚ 40˚ 45˚

5˚

10˚

15˚

SS

N

AD

35˚ 40˚ 45˚

5˚

10˚

15˚

GAGA

RSRS

35˚ 40˚ 45˚

5˚

10˚

15˚

Figure 6.6: Structural map of the MER with areas enclosed within Fig. 6.7 and Fig 6.8 marked

by black boxes. From Casey et al. (2006)

Figure 6.7: Landsat TM imagery showing the close spatial coincidence of faults and eruptive

centres (fissures, cinder cones, shields) on the central Boset-Kone magmatic segment.

Location shown in Figure 6.6. Note the parallel faults and aligned scoria cones, as well as

fissural flows from some faults. Boset (B) and Kone (K). Most of the relief in this zone is

magmatic construction. From Casey et al. (2006)


81

Figure 6.8: Band 2 of Aster 15 m resolution imagery showing the Dofen volcanic edifice. The

elliptical region with roughly E-W long axis is taken to be the strained volcanic ediface (ellipse

above shows regional extension direction). Fault dips and fault slip directions from field studies

in 1990 shown in line drawing below image. Open circles indicate eruptive centres. From Casey

et al. (2006)


82

a) Fantale-Dofen

Magmatic Segment

b) Boset-Kone

Magmatic Segment

n = 164 n = 123

0˚30˚

60˚90˚

120˚

150˚

180˚

210˚

240˚

270˚

300˚

330˚

SOUTH90

EASTWEST

0˚30˚

60˚90˚

120˚

150˚

180˚

210˚240˚

270˚

300˚

330˚

SOUTH90

EASTWEST

Figure 6.9: Rose diagrams showing orientations of Quaternary faults in a) Fentale Dofen

magmatic segment and b) Boset - Kone magmatic segment. The majority of Quaternary faults

are strike ~NNE but a small population strike from ~NE to ~ENE. From Casey et al. (2006)

6.7 Discussion 6.7.1 Style of faulting

The focal mechanisms provide a uniform picture for the pattern of faulting and stress

field orientation of the Ethiopian rift. Focal mechanisms indicate predominantly normal

dip-slip on faults that strike ~N to ~NNE, parallel to the dominant N10oE strike of faults

that cut Quaternary lavas (e.g., Casey et al., 2006). Field observations and geodetic

data of volcanic rift zones in Iceland and Hawaii indicate that dyke intrusions are most

often associated with normal faulting and fracturing at the surface (Rubin, 1992). The

predominance of normal dip-slip, and resulting lack of diversity in our focal mechanism

dataset, is thus consistent with dyke-induced seismicity in the MER.

The normal, oblique, and left-lateral strike-slip displacement on NE-striking fault planes

most likely occurs on pre-3.5 Ma, N40oE-striking faults that probably formed under a

NW-SE extension direction. These have most likely been re-activated as N40oE-

striking ramps and transfer faults to link N10oE-striking fault segments formed under

the ~N105oE extension direction during the Quaternary (Wolfenden et al., 2004; Casey


83

et al., 2006). The negligible block rotations about vertical axes in zones in between

magmatic segments suggests no through-going transform faults have developed, thus

supporting our interpretation of the strike-slip focal mechanisms as left-lateral ~NE

striking faults (Kidane et al., 2006).

6.7.2 Direction of extension across the MER

The N103oE orientation of the minimum compressive stress from focal mechanisms

parallels, within errors, the extension direction determined from sea-floor spreading

data from the past 3.2 My (Chu and Gordon, 1999) and current extension direction

determined from campaign and permanent GPS data (Bilham et al., 1999; Fernandes

et al., 2004; Calais et al., 2006). The direction of the minimum compressive stress also

agrees with the N105oE oriented long axes of elliptical Quaternary calderas (Casey et

al., 2006).

Extension is perpendicular to the strike of Quaternary faults, fissures and aligned

cones and is in agreement with structural studies that show a WNW-ESE direction of

extension during Quaternary times (Boccaletti et al., 1998; Wolfenden et al., 2004;

Casey et al., 2006). The current direction of extension is thus perpendicular to the

strike of Quaternary volcanic chains and faults in the magmatic segments. The right-

stepping en echelon pattern at the surface may be induced by ~N105oE directed

extension above a ~NE striking low velocity zone in the upper mantle connecting the

MER to the triple junction in Afar (Benoit et al., 2003; Bastow et al., 2005).

6.8 Summary

Well constrained source parameters were determined for 30 earthquakes located in

the MER. Earthquake focal mechanisms show predominantly normal dip-slip on faults

striking ~N to ~NNE, parallel to the dominant N10oE strike of faults that cut Quaternary

lavas. The predominance of normal dip-slip, and resulting lack of diversity in our focal

mechanism dataset is consistent with dyke-induced seismicity in the MER. Strike-slip /

oblique slip earthquakes beneath Fentale and Boset volcanoes are interpreted as left-

lateral motion on ~NE- to ~ENE-striking faults that link N10oE-striking fault segments.

The orientation of the minimum compressive stress determined from focal mechanisms

is N103oE, consistent with geodetic data and global plate kinematic constraints. The


84

current direction of extension is thus perpendicular to the strike of Quaternary volcanic

chains and faults in the magmatic segments. The right-stepping en echelon pattern of

strain at the surface may be induced by ~N105oE directed extension above a ~NE

striking low velocity zone in the upper mantle connecting the MER to the triple junction

in Afar.

Chapter 7 - Shear-wave splitting in crustal earthquakes

85

Chapter 7

Shear-wave splitting in crustal earthquakes 7.1 Introduction Patterns of seismic anisotropy can be used to constrain style of rifting and is therefore

a useful tool to evaluate models of continental breakup. The wealth of broadband

seismic data acquired by the EAGLE network has afforded a detailed analysis of

seismic anisotropy in the upper mantle beneath the MER. However, such studies

cannot isolate anisotropy in the uppermost crust where patterns of deformation are

expressed as intrusive magmatism, and faulting and volcanism at the surface. This

chapter presents the results from a study of crustal anisotropy beneath the MER using

S-wave splitting measurements from local earthquakes. The results are compared to

independent structural and geophysical studies and this information is used to evaluate

mechanisms of deformation preceding continental break-up.

7.2 Mechanisms for seismic anisotropy in the crust Anisotropy of the shallow crust is commonly attributed to micro-cracks vertically-

oriented parallel to the direction of maximum horizontal stress (e.g. Crampin, 1994).

For example, crustal shear-wave splitting measurements in rift zones at the Mid-

Atlantic ridge and in Iceland show fast-polarization directions sub-parallel to the

maximum horizontal stress. These patterns are attributed to aligned parallel cracks and

fractures in the uppermost 3-5 km of the crust (e.g. Barclay and Toomey, 2003; Evans

et al., 1996; Menke et al., 1994). S-wave anisotropy has also been attributed to vertical

micro-cracks throughout the crust in which case S-wave splitting is accrued along the

whole ray-path (e.g. Volti and Crampin, 2003b). Fast-polarization directions at active

volcanoes are usually parallel to dykes and the maximum horizontal stress, with 90o

polarization flips observed prior to volcanic eruption due to increased pore pressure

leading to changes in crack orientation (Miller and Savage, 2001). Crustal anisotropy

has also been linked to other rock fabrics such as vertically dipping foliation of

metamorphic basement (e.g. do Nascimento et al., 2004).


86

7.3 Determination of shear-wave splitting parameters

Shear-wave splitting measurements are made on seismograms where the S-wave

incident-angle is within the shear-wave window (SWW). The SWW is the vertical cone

bound by sin-1 (Vs/Vp) where S-wave particle motions are not disturbed by S-P

conversions at the free surface (Booth and Crampin, 1985). A Vp/Vs of 1.75 was used,

calculated from P- and S-wave travel-times from earthquakes in the MER (Fig. 3.9),

which corresponds to a SWW that is a cone within 35o of the vertical.

Seismic stations are distributed with approximately equal spacing throughout the MER

but the earthquakes are spatially clustered, with 75 % of seismicity located in the

Fentale-Dofen magmatic segment and at the intersection of the MER and Red Sea rift

(Keir et al., 2006). Due to the spatial clustering of earthquakes, S-wave splitting

measurements could only be made at ~10 % of available seismic stations. 24

earthquakes located beneath 18 stations provided 26 three-component seismograms

where the S-wave incident-angle is within the shear-wave window (SWW) (Table 7.1).

The polarization direction of the fast S-wave (φ ) and the time delay between the fast

and slow S-waves (δt) is determined using the method of Silver and Chan (1991),

adapted for application to micro-earthquakes. In an isotropic radially stratified crust,

near vertically impinging S-waves should exhibit linear particle motion. This phase is

split into orthogonally polarized fast and slow S-waves when it travels through an

anisotropic medium and this splitting produces an elliptical particle motion. To remove

the effects of the anisotropy, the horizontal components are rotated by φ and their

relative positions shifted by δt, thereby linearizing the particle motion (Figure 7.1). To

estimate the splitting, a search for the correction parameters that best linearize the S-

wave motion is conducted. An F-test is performed to assess the uniqueness of the

estimated splitting parameters and thereby produce an error estimate (Silver, 1996).

The splitting parameters are well constrained. We use a cut off error criteria of ± 0.03 s

for δt and 9o forφ (Table 7.1).


87

a) SHEE

c) MELE GTFE INEE

R

R

T

T

R

R

T

T

b) E36

Figure 7.1: Examples of shear-wave splitting at EAGLE stations. Data is bandpass filtered

between 0.5-5 Hz. (a) SHEE. In the left panel the top two waveforms show observed R and T

seismograms, while the lower two waveforms show the result after the correction for splitting

using the estimated parameters (φ = 0o, δt = 0.1 s). Note that energy is minimized on the T-

component after the correction. Middle panel shows the fast (solid line) and slow (dashed line)

shear-waves and corresponding particle motions in the horizontal plane before and after

correction for shear-wave splitting. The right panel shows confidence ellipse intervals of the

solution (innermost contour is the 95% confidence interval and indicates how well the solution is

constrained. (b) Fast (solid line) and slow (dashed line) shear-waves and corresponding particle

motions at stations MELE, GTFE and INEE, before and after the splitting correction.


88

Table 7.1: Shear-wave splitting measurements.

Table 7.1 Column Header Definitions

Station: station name

Lat. (oN): latitude of earthquake location

Long. (oE): longitude of earthquake location

ML: local magnitude

θ (deg.): solid angle of shear-wave

φ (deg.): polarisation direction of the fast shear-wave

φ E (deg.): error in polarisation direction defined by the 95% confidence interval

δ t (s): delay-time between the fast and slow shear-waves

δ t E (s): error in delay-time defined by the 95% confidence interval

Stat. Lat. (oN)

Long. (oE) Z (km) ML θ

(deg.) φ (deg.)

φ E (deg.)

δ t (s)

δ tE (s)

AMME 8.279 39.096 6.80 1.78 19.44 355.0 4.5 0.10 0.020AMME 8.262 39.091 7.20 1.98 29.66 356.0 4.5 0.12 0.020ANKE 9.648 39.714 8.50 2.77 34.31 355.0 6.0 0.14 0.020ANKE 9.674 39.758 8.65 1.38 32.45 4.0 5.5 0.20 0.010AREE 8.958 39.362 8.63 1.58 34.81 54.0 4.5 0.12 0.020BORE 8.676 39.569 6.75 1.50 37.61 20.0 2.5 0.24 0.020CHAE 9.327 38.750 7.35 1.20 14.49 50.0 2.5 0.04 0.010GTFE 8.992 39.849 6.71 0.94 9.31 50.0 4.5 0.12 0.020GTFE 8.997 39.894 8.61 0.96 33.04 54.0 0.5 0.16 0.010INEE 9.927 39.236 18.00 1.20 28.07 56.0 3.0 0.10 0.010INEE 9.874 39.069 16.50 1.34 24.73 59.0 8.5 0.10 0.030MELE 9.300 40.184 7.15 1.71 14.88 10.0 1.0 0.06 0.010MELE 9.291 40.167 6.75 1.04 30.01 12.0 6.5 0.06 0.020MELE 9.255 40.195 8.15 0.98 34.01 8.0 3.5 0.10 0.010MELE 9.292 40.172 6.95 1.92 25.40 14.0 6.0 0.06 0.020SHEE 10.043 39.918 8.50 2.14 30.26 0.0 2.0 0.10 0.010E36 9.089 40.000 7.80 0.94 16.43 42.0 5.0 0.11 0.025E36 9.058 39.973 8.54 1.28 36.84 42.0 4.5 0.11 0.010E53 8.089 39.058 8.02 1.19 39.02 19.0 1.5 0.14 0.010E77 7.895 38.823 7.85 1.89 29.27 7.0 2.5 0.12 0.015E79 7.619 38.770 8.01 2.03 38.62 26.0 3.0 0.13 0.0101018 9.874 38.507 22.00 1.92 29.60 36.0 8.0 0.14 0.0151030 9.874 38.507 22.00 1.92 26.15 38.0 1.5 0.11 0.0101155 8.951 39.174 8.69 1.64 33.72 70.0 7.0 0.05 0.0151163 8.951 39.174 8.69 1.64 21.39 68.0 8.0 0.04 0.0101219 8.611 39.447 7.00 3.44 24.57 18.0 4.5 0.19 0.015


89

38˚ 39˚ 40˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

T

TT

38˚ 39˚ 40˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

38˚ 39˚ 40˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

38˚ 39˚ 40˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

5 % Anisotropy5 % Anisotropy

38˚ 39˚ 40˚

8˚ 8˚

9˚ 9˚

10˚ 10˚10181018

10301030

11551155

11631163

12191219

AMMEAMME

ANKEANKE

AREEAREE

BOREBORE

CHAECHAE

E36E36

E53E53

E77E77

E79E79

GTFEGTFE

INEEINEE

MELEMELE

SHEESHEE

38˚ 39˚ 40˚

8˚ 8˚

9˚ 9˚

10˚ 10˚





Arboye BFArboye BF


AnkoberAnkoberBFBF

Ambo FaultAmbo Fault

38˚ 39˚ 40˚

8˚ 8˚

9˚ 9˚

10˚ 10˚

35˚ 40˚ 45˚

5˚

10˚

15˚

SPSP

NPNP

APAPDPDP

35˚ 40˚ 45˚

5˚

10˚

15˚

GAGA

RSRS

35˚ 40˚ 45˚

5˚

10˚

15˚38˚ 39˚ 40˚

8˚

9˚

10˚

38˚ 39˚ 40˚

8˚

9˚

10˚

38˚ 39˚ 40˚

8˚

9˚

10˚

200 ms200 ms

38˚ 39˚ 40˚

8˚

9˚

10˚

0 2000 4000

m

NW Plateau

SE Plateau

Figure 7.2: Crustal anisotropy measurements at 18 broadband stations in Ethiopia. White

arrows show the polarization of the fast S-wave (φ ) and arrow length is scaled by % anisotropy

along ray-path. Solid black lines with dip ticks are Miocene border faults (BF) and dashed lines

are monoclines. Quaternary magmatic segments (MS) are shaded grey. Dark arrows show the

extension direction and orientation of the minimum horizontal stress (Keir et al., 2006). The

position of the along-axis profile for Figure 7.3b, d is shown by the black line. Top left inset:

Topographic map of the MER, adjacent plateau and Afar depression. NP: Nubia Plate, SP:

Somali Plate, DP: Danakil Plate, AP: Arabian Plate, RS: Red Sea, GA: Gulf of Aden. Top right

inset: White arrows show polarization of fast S-waves scaled by delay-time.


90

0.0

0.1

0.2

0.3

-200 -150 -100 -50 0 50

Rift-perpedicular distance (km)

δt (

s)

a)

NW SE

0

2

4

6

8

% A

niso

trop

y

-200 -150 -100 -50 0 50

Rift-perpedicular distance (km)

b)

NW SE

-150 -100 -50 0 50 100 150

Rift-parallel distance (km)

b)

SW NE

-150 -100 -50 0 50 100 150

Rift-parallel distance (km)

d)

SW NE

0.00

0.05

0.10

0.15

8 12 16 20 24

Ray-path length (km)

δt (

s)

e)0.0

0.1

0.2

0.3

6 7 8 9 10 11

Ray-path length (km)

f)

δt (

s)

0

20

40

60

-10 0 10 20 30

Orientation of Max Horiz Stress

φ(de

g)

g)

Figure 7.3: a) Rift-perpendicular profile of station averaged delay time (δt) versus distance from

the rift axis. The two solid lines shows the position of magmatic segments and the dashed line

shows the approximate position of the western boundary of the rift valley. b) Rift-parallel profile

of station averaged δt at stations within 20 km of the along rift-axis line on Figure 1. c) Rift

perpendicular profile of % anisotropy versus distance from the rift axis. d) Rift-parallel profile of

% anisotropy versus distance along the rift valley. e) Individual measurements of δt versus S-

wave ray-path length at stations on the western Ethiopian plateau. The dashed line is the best

straight line fit to the data. f) Individual measurements of δt versus ray-path length at stations in

the rift valley. g)φ against the average orientation of maximum horizontal stress axes of focal

mechanisms within 25 km of the splitting measurement. The dashed line is the best straight line

fit to the data. The symbols are: white squares = plateau stations; grey triangles = stations at the

Ankober fault; inverted triangles = stations in the MER but outside magmatic segments; dark

grey circles = stations in magmatic segments.


91

7.4 Results of shear-wave splitting analysis

S-wave splitting measurements from local earthquakes near the MER show large

spatial variation in both φ and δt (Figure 7.2). At stations on the NW plateau φ varies

between 36o and 70o. δt varies between 0.04 s and 0.14 s for earthquakes that

occurred at depths of 12-20 km and δt increases linearly with increased ray-path

distance (Figure 3), showing that the crust is anisotropic to at least 20 km depth. This

equates to fairly uniform anisotropy of 1.1 % on average, if splitting is assumed to be

accrued over the full ray-path length (Figures 7.2 & 7.3).

Along the Ankober fault system φ is oriented ~N, parallel to seismically active faults

(Figure 7.2). δt is 0.1 - 0.16 s, equivalent to 2.2-3.6 % S-wave anisotropy.

At stations along the rift axis φ is mostly oriented ~N to NNE (Figure 7.2). Delay times

are 0.06-0.24 s for earthquakes that are 6-9 km deep, equating to 3-6.2 % anisotropy

(Figure 7.3). The largest values of δt (0.19-0.24 s, anisotropy of 5.4-6.2 %) are

recorded at stations 1219 and BORE, both in the Quaternary Boset-Kone magmatic

segment (Figures 7.2 and 7.3).

7.5 Discussion 7.5.1 Crustal anisotropy beneath the rift-axis Near-vertically propagating S-waves from local earthquakes near the MER show clear

evidence of S-wave splitting. The anisotropy is thus most likely due to foliations, cracks

or inclusions aligned by regional and local stresses in the crust. The magnitude and

orientation of the shear-wave splitting varies dramatically across the EAGLE network,

suggesting a heterogeneous stress field or variations in the underlying cause of

anisotropy. Our results are calibrated with independent geological and seismic studies

in the MER.

Stations along the rift axis show relatively large amounts of splitting despite shallower

earthquake depths (6-9 km). Up to 0.24 s of splitting is observed beneath Boset-Kone

magmatic segment, which equates to over 6 % anisotropy. Stations within the rift valley

but located outside magmatic segments show less splitting (e.g. MELE), but the


92

average magnitude of splitting in the rift valley is still nearly 3 %, much larger than

beneath the NW plateau. The ~N to NNE orientation of φ in the magmatic segments is

parallel to Quaternary faults and aligned volcanic cones. The along axis variation of φ

correlates well with local changes in the strike of maximum horizontal stress axes from

focal mechanisms of earthquakes within 25 km of splitting measurements (Figure 7.3)

(Keir et al., 2006).

The largest amounts of upper-crustal anisotropy are in the Quaternary magmatic

segments where independent studies show evidence of pervasive dyke intrusion and

the presence of partial melt in shallow magma chambers. Mackenzie et al. (2005) and

Keranen et al. (2004) interpret cooled mafic intrusions in the mid-crust beneath these

magmatic segments using models derived from wide-angle refraction data and

controlled source tomography respectively. The magnitude of splitting under Boset

volcano is especially pronounced, where melt-related anomalies have been interpreted

in magnetotelluric data (Whaler and Hautot, 2006). The S-wave splitting observations

are consistent with anisotropy due to vertically aligned magma intrusions or melt-filled

cracks beneath the Quaternary magmatic segments, where the majority of strain is

accommodated by dyke injection (Keir et al., in press).

7.5.2 Crustal anisotropy beneath the Ethiopian plateau

The deepest earthquakes lie beneath the largely un-extended NW Ethiopian plateau,

where we observe an increase in delay time with increased ray-path length using S-

wave splitting measurements at different stations. These variations in delay-times can

be explained by relatively uniform anisotropy that extends to at least 20 km depth;

larger delay-times (0.1 - 0.14 s) at stations 1018, 1030 and INEE are caused by

splitting accrued over longer ray-paths (Figure 7.2 and 7.3). Alternatively, these

patterns may be caused by lateral variations in anisotropy of the uppermost few

kilometers with larger upper crustal anisotropy at stations 1018, 1030 and INEE.

However, controlled source seismic images of underplating (Mackenzie et al., 2005),

mid-crustal conductive anomalies in MT data (Whaler and Hautot, 2006), and

Quaternary eruptive centres as far north as Lake Tana all infer the presence of melt in

the lower crust beneath the Ethiopian plateau. Given these independent observations,

we interpret the data to show that melt induced anisotropy extends to at least 20 km

subsurface. The amount of crustal anisotropy beneath the plateau is low (1.1%),


93

consistent with melt decrease away from the rift axis. Splitting at stations on the

plateau is oriented ~NE and may also indicate a contribution from pre-existing

basement foliation or structural trends. Where exposed, Pan-African basement

foliation and Proterozoic ophiolite belts predominantly strike ~N to ~NE (e.g. Berhe,

1990; Kazmin et al., 1978). These have been used to infer a NE-SW trending suture

(Berhe, 1990) but due to limited basement outcrop their interpretation is controversial

(Church, 1991). NE to ENE oriented basement structures are evident in regional

drainage patterns along the Ambo fault, which has been reactivated in Miocene rifting

(Abebe et al., 1998). Beneath the Ethiopian plateau the crustal anisotropy may be due

to a combination of mechanisms associated with aligned melt, pre-existing basement

foliation and structural trends.

7.5.3 Model of crustal anisotropy beneath the MER

The patterns of shear-wave splitting observed in earthquakes beneath both the rift

valley and nearby plateau are most simply explained by crustal anisotropy related to

variable amounts of melt pocket alignment, with a higher degree of magma intrusion in

the crust beneath the rift. The pattern of crustal anisotropy beneath the MER can be

integrated with SKS-splitting and surface-wave anisotropy studies, which probe deeper

into the lithosphere. SKS-splitting constrains anisotropy of the uppermost 100 km.

Beneath the MER, the increased splitting in more magmatic regions and the alignment

of anisotropy along the rift-axis parallel to magmatic segments were used as evidence

to propose that the anisotropy is controlled by oriented pockets of melt (Kendall et al.,

2005). This interpretation is supported surface-wave anisotropy which is consistent with

a model of oriented melt-filled pockets as the primary mechanism for anisotropy

beneath the rift valley from 20-75 km depth (Kendall et al., 2006). The pattern of crustal

anisotropy shows that melt-induced anisotropy at 20-75 km depth (Bastow et al., 2005;

Kendall et al., 2005; Kendall et al., 2006) continues into the uppermost crust, thereby

penetrating the entire plate and facilitating continental breakup. Melt-induced

anisotropy extends from the base of the lithosphere to the upper crust, suggesting that

magma injection helps localize and facilitate extension just prior to continental breakup.

7.6 Summary


94

Along the rift-axis the orientation of the fast S-wave is ~N to NNE, parallel to

Quaternary to Recent faults, aligned cones and the current maximum horizontal stress

axis. The largest amounts of upper crustal anisotropy are in the Quaternary magmatic

segments, where the majority of strain is accommodated by magma injection;

anisotropy is most likely caused by aligned melt-filled micro-cracks and dykes. The low

amount of anisotropy beneath the Ethiopian plateau is consistent with melt decrease

away from the rift axis. These results suggest the anisotropy is related to variable

amounts of melt pocket alignment in the crust, with a higher degree of dyke intrusion in

a narrow zone of Quaternary magmatism. Melt-induced anisotropy extends from the

base of the lithosphere to the upper crust, suggesting that magma injection helps

localize and facilitate extension just prior to continental breakup.

Chapter 8 - Discussion

95

Chapter 8 Discussion

8.1 Evidence for magma-fed along-axis segmentation of the MER

The most striking feature of the recorded seismicity in the MER is the coincidence of

earthquake swarms and the magmatic segments, which are the locus of Quaternary

volcanism (Fig. 4.1). The magmatic segments are also the locus of strain, as

determined from studies of fault patterns (e.g., Williams et al., 2004; Casey et al.,

2006; Pizzi et al., 2006) and the lone geodetic profile (Bilham et al., 1999; Bendick et

al., 2006). The inactivity of mid-Miocene border faults that define the overall ~NE trend

of the MER is reflected over longer time periods by the minor geodetic strain on the rift

flanks (Bilham et al., 1999; Bendick et al., 2006) and lack of large magnitude

earthquakes and border faults over the last ~50 years and inferred from historical

records spanning ~150 years (Gouin, 1979; Ayele and Kulhánek, 1997). The

apparently inactive mid-Miocene border faults do not correlate with the spatial

arrangement of the seismically active Quaternary magmatic segments (Ebinger and

Casey, 2001; Wolfenden et al., 2004) (Fig. 2.2).

The distribution of Quaternary faulting and aligned volcanic cones defines the along-

axis segmentation of the MER near the surface. The pattern of seismicity interpreted in

light of new constraints on crustal and mantle structure from independent geophysical

studies provides clues as to the origin and maintenance of the along-axis

segmentation. Clusters of seismicity within magmatic segments are concentrated within

a <20 km-wide zone and are elongate parallel to Quaternary-Recent faults, fissures

and active eruptive centres (Fig 4.3) This pattern of seismicity is similar to dyke

induced and magma-intrusion induced seismicity in other rift zones worldwide (e.g. Pitt

et al., 2004; Hayes et al., 2004). The swarms of low magnitude earthquakes are

concentrated at 8-14 km depth which coincides with the top of the ~20-30 km-wide

zones of high seismic velocity imaged at 8-10 km depth using controlled source and

local earthquake tomography (Keranen et al., 2004; Daly et al., in review). These zones

are segmented along the axis of the rift and correlate with small wavelength highs in

the observed Bouguer anomaly (Mahatsente et al., 1999; Tiberi et al., 2005; Cornwell


96

et al., 2006); the combination of high seismic velocity and high density suggests

magmatic segments are underlain by gabbro intrusions that rise to 8-10 km subsurface

to accommodate extension (Keranen et al., 2004). The spatial correlation between

earthquake locations, faults with fissural eruptions, aligned cones and axial mafic

intrusions suggests that seismicity is induced by mafic intrusions into the mid- to upper

crust.

The comparison of seismic moment release with total geodetic strain offers additional

insights into crustal deformation processes in the MER. During the period 2001-2003,

seismic moment release within 1o of the geodetic profile across the MER

(approximately coincident with the front face of Fig. 8.1) was just 3.04 x 1014 Nm,

equivalent to one Mw 3.6 earthquake (Bendick et al., 2006). This relatively low seismic

moment release would produce surface displacements several orders of magnitude

smaller than that detectable by the geodetic array, yet an average rift opening of 4.0 ±

0.9 mm/yr was measured 1992-2003 (Bendick et al., 2006). A deficit of seismic

moment is also evident over longer time scales. For the period 1960-2000, a

comparison of the seismic moment release expected for the relative plate motion

shown by global plate kinematic models (e.g. Chu and Gordon, 1998), and the

observed seismic moment shows that less than 50 % of extension across the MER is

accommodated by rapid slip on faults (Hofstetter and Beyth, 2003). The deficit

between observed seismic moment release and that expected from geodetic

measurements and plate kinematic models provides additional evidence that dyke

injection likely accommodates more strain than faulting beneath the MER. However,

the 46 year seismicity catalogue is shorter than the 50 year recurrence intervals of M >

5.5 earthquakes in the MER, predicted from the Gutenberg-Richter relationship from

the 16 months of seismicity recorded by EAGLE. Therefore, the deficit in the scalar

moment rates from seismicity may be caused by a local delay in seismic strain release

rather than a significant contribution from aseismic strain such as dyking.

We can draw insight from theoretical studies of magma injection processes which show

that if magma is available then injection of dykes accommodates strain at lower

stresses than is required for faulting (e.g. Buck, 2004). In the MER, injection of magma

accommodates strain at lower levels of stress than is required to activate large

displacement border faults, and the locus of strain becomes the magma injection zone.

Below ~10km depth, the majority of strain is accommodated by magma injection


97

beneath magmatic segments, whereas strain in the brittle seismogenic zone is

accommodated by a combination of magma injection and faulting (Fig. 8.1).

20 40 60 80 100 120 140

5

10

B

D

Mioceneborder faultsystem seismogenic

zone

locus of seismicity

FFK

Figure 8.1: Cartoon sketch of the MER that shows abandoned Miocene border faults and

localisation of strain in <20 km-wide right stepping en echelon magmatic segments that are

zones of Quaternary magma intrusion and faulting. Front face is constrained by topographic

relief along the line of EAGLE controlled source seismic profiles. Letters mark volcanoes: B;

Boset, K; Kone, F; Fentale, D; Dofen.

The along-axis segmentation seen from the surface to 8-10 km subsurface beneath

magmatic segments correlates with the zones of higher velocity in the mid- and lower

crust, implying that mafic intrusions extend to the base of the crust (Maguire et al.,

2006; Daly et al., in review). The seismic models suggest that the segmentation

pattern observed at the surface and in the upper crust continues as discrete zones of

magma injection to the base of crust.

The pattern of seismic anisotropy provides further indirect evidence that the along-axis

segmentation of the MER is controlled by the supply of magma to discrete rift

segments arranged along the axis of the rift. The largest upper crustal anisotropy is in

Quaternary magmatic segments where independent studies show evidence of

pervasive dyke intrusion and the presence of partial melt in shallow magma chambers

(Figs 7.2, 7.3). The magnitude of splitting under Boset volcano is especially

pronounced, where melt-related anomalies have been interpreted in magnetotelluric

data (Whaler and Hautot, 2006). The S-wave splitting observations are consistent with


98

anisotropy due to vertically aligned magma intrusions or melt-filled cracks beneath the

Quaternary magmatic segments, which are the locus of strain.

Intrusion of magma into the lower crust is likely not completely restricted to the rift axis.

A sparse distribution of relatively deep source (15-21 km) earthquakes is observed

beneath the Ethiopian Plateau (Figs. 4.1, 4.3) which shows relatively small shear-wave

splitting delay-times of 0.05-0.15 s (Figs. 6.2, 6.3). The relatively low amount of

anisotropy is consistent with melt decrease away from the rift axis. However, mid-

crustal conductive anomalies in MT data (Whaler and Hautot, in press), and

Quaternary eruptive centres at latitudes of ~12oN on the Ethiopian plateau all imply the

presence of melt in the lower crust beneath the Ethiopian plateau. The patterns of

shear-wave splitting observed in earthquakes beneath both the rift valley and nearby

plateau are most simply explained by crustal anisotropy related to variable amounts of

melt pocket alignment, with a higher degree of magma intrusion in the crust beneath

the rift (Fig. 8.2).

The pattern of crustal anisotropy beneath the MER can be integrated with surface-

wave studies and SKS-splitting, which probe deeper into the lithosphere. Surface-wave

anisotropy is consistent with a model of oriented melt-filled pockets as the primary

mechanism for anisotropy beneath the rift valley from 20-75 km depth (Kendall et al.,

2006). SKS-splitting constrains anisotropy of the uppermost ~100 km. The increased

splitting beneath Quaternary magmatic segments and alignment of anisotropy parallel

to Quaternary faults, fissures and aligned volcanic cones was used as evidence by

Kendall et al. (2005) that the anisotropy is controlled by oriented pockets of melt

distributed through the lithosphere (Fig 2.10). The parallelism of SKS, and local

earthquake S-wave splitting beneath the MER indicates that melt induced anisotropy

extends from the base of the lithosphere to the upper crust. If the lithospheric thickness

varies considerably from inside the rift to the rift shoulder, as imaged in tomography

models (Bastow et al., 2005), then the large amounts of SKS-splitting along rift margins

may be caused by a combination of increased melt extraction along steep gradients at

the lithosphere-asthenosphere boundary and larger lithospheric thickness beneath the

rift flanks relative to the rift-axis (Fig. 8.2).

Observations from the MER suggest that strain is concentrated in discrete magmatic

segments, and accommodated by a combination of faulting and dyking in the upper


99

crust and by magma intrusion in the lower-mid crust. There is no evidence that crustal

scale detachment faults such as proposed by Corti et al. (2003) and Pizzi et al. (2006)

accommodate strain across the MER.

0 km 200

asthenosphere

magmaticsegments

mantlelithosphere

crust

Figure 8.2: Cartoon sketch of the MER (After Ebinger, 2005) that shows thinning of the mantle

lithosphere from heating by localized magma injection. Magma injection is localised to the rift

axis where dyke injection accommodates strain at lower stresses than that required for faulting

causing border faults to be abandoned. Localized dyke injection along the rift axis induces

faulting in the brittle upper crust.

8.2 Temporal variations of magma supply and episodic rift opening

The EAGLE network recorded seismicity for 15 months and thus provides a snapshot

of active deformation in the MER. During this time period, seismicity was particularly

concentrated in the Fentale-Dofen magmatic segment whereas the Boset-Kone

magmatic segment to the south was largely quiescent (Fig. 4.1). The pattern of

Quaternary faults and fissures that cut recent lavas and historic earthquake data,

however, show that major episodes of dyke injection and associated seismicity have

been concentrated in other magmatic segments in the past. Asfaw (1982) noted the

development of surface fissures following a swarm of ML<4 earthquakes near Fentale

in 1981. Similar fissures are observed in all magmatic segments along the axis of the

MER and most likely formed during previous rifting episodes (e.g., Asfaw, 1998;

Williams et al., 2004). The swarm of earthquakes reported near Nazret in 1964 (Gouin,

1979), an Mw 5.3 earthquake near Boset volcano in 1993 (Ayele, 2000) and fissuring of


100

<10000 year old ignimbrites at Kone caldera (Williams et al., 2004) shows that the

Boset-Kone magmatic segment has experienced episodes of rifting in the recent past.

The snapshot of seismic activity in the MER captured by the EAGLE network shows

that the pattern of seismicity mirrors the structural segmentation observed from the

pattern of Quaternary faults and aligned volcanic cones. Each magmatic segment

experiences increased rates of seismicity at different periods of time, implying that

magmatic segments deform independently of each other. If, as implied from the

integrated geological and geophysical data base from the MER that magmatic

segments are formed by repeated episodes of magma injection, then this suggests

magma source regions are spatially and temporally discrete. With improved

geochronological dating of past volcanic products and long-term seismic and volcano

monitoring, repeat times in the magma replenishment cycle can be established which

will assist seismic and volcanic hazard assessment of volcanic zones such as the

MER.

The segmented pattern of the MER, with deformation concentrated within a narrow

zone of faulting, and aligned volcanic cones is similar to that observed in the ~60 km-

long Dabbahu segment of the Afar rift. Deformation at the surface is concentrated

within a ~25 km wide zone of normal faulting, aligned volcanic cones and fissural

basalt flows. The September / October 2005 major rifting event in the Dabbahu

magmatic segment provides additional clues to the origin of the pattern of along-axis

segmentation and how magmatic segments deform. Between 14 September and 5

October 2005, 163 earthquakes (mb > 4) and a volcanic eruption occurred along the

full 60 km-long Dabbahu magmatic segment. The crisis culminated on 26 September

with the opening of a 500 m-long, 60 m-deep, N-S oriented vent, and surface fissuring

and faulting on the northeast flank of Dabbahu volcano. Radar interferometry (InSAR)

data shows that the seismic and volcanic events were accompanied by up to ~6 m of

horizontal opening at the surface, with horizontal and vertical deformation concentrated

in a ~25 km-wide zone (Wright et al., 2006).

Earthquakes during the period 14 September to 4 October, 2005 release a combined

seismic moment of 6.7 x 1018 Nm, calculated using the empirical mb - mo relationship

for Afar (Hofstetter and Beyth, 2003). This is an order of magnitude smaller than the

estimate of 7.5 x 1019 Nm for the total geodetic moment release. Rather, the observed


101

horizontal and vertical deformation along the ~60 km-length of the Dabbahu segment

is consistent with the displacement field expected for intrusion of 2.4-2.6 km3 of

magma as vertical dyke intrusion with induced faulting (Wright et al., 2006). Models of

the surface deformation indicate that magma was likely sourced from two shallow

magma chambers near the northern tip of the segment (<40 %), as well as deeper

sources beneath the 18-22 km-thick crust and located in the middle of the rift segment.

Structural data suggest that the locations of magma source(s) maintaining the

Dabbahu segment have been stable for 2-4 My (Hayward and Ebinger, 1996). The

Dabbahu rifting event thus provides observational evidence that magma intrusion into

the middle crust, and dyke intrusion and faulting in the upper crust, control and

maintain the along-axis segmentation prior to continental breakup.

8.3 Comparison with slow-spreading mid-ocean ridges

As outlined in section 8.1, the along-axis segmentation of the MER is defined at the

surface by a series of aligned volcanic cones and seismically active Quaternary fault

arrays that define ~20 km-wide, ~60 km-long right-stepping en echelon rift segments

along the axis of the rift. (e.g. Ebinger and Casey, 2001). Seismic and gravity studies

provide evidence that the discrete rift segments are underlain by axial intrusions that

extend from at least the base of the crust to ~10 km subsurface; (e.g. Keranen et al.,

2004; Tiberi et al., 2005; Mackenzie et al., 2005).

The along-axis segmentation of the MER shown in the pattern of Quaternary faults,

active aligned volcanic cones, distribution of intruded magma, along-axis variation in

seismicity, and the coincidence of seismicity with magmatism is remarkably similar to

slow- and ultra-slow spreading mid-ocean ridges such as the northern Mid-Atlantic

ridge and Gakkel ridge in the Arctic ocean. In both marine locales, the spatial and

temporal variations in melt flux preferentially delivered from mantle upwellings to

segment centres via dikes in the lower crust are believed to control the along-axis

segmentation of the ridge (e.g. Lin et al., 1990; Sparks et al., 1993; Tucholke and Lin,

1994; Magde et al., 1997; Parsons et al., 2000; Dunn et al., 2005). This similarity

provides strong evidence that the along-axis segmentation in the MER is controlled by

injection of magma sourced from discrete sources in the mantle, rather than the along-

axis pattern of border faults or their subsurface geometry.


102

Observations of the causal link between dyke injection and seismicity at slow-

spreading mid-ocean ridges provides additional insights to rift processes in the MER.

For example, seismic swarms in the Hengill volcanic area in southwestern Iceland from

1994 -1998 are concentrated at the base of the seismogenic layer and have

predominantly double-couple mechanisms (Feigl et al., 2000). Increased levels of

seismic activity correlates spatially and temporally with continuous uplift at the surface

modelled by inflation of a source due to dyke injection to ~7 km subsurface (Feigl et

al., 2000). Calculations of Coulomb failure stress suggest that inflation from magma

injection induces stresses that exceed the Coulomb failure criterion, triggering

earthquakes. The swarm activity is cyclical which shows that stresses in the brittle

crust rise slowly to failure with continued dyke injection, but drops instantaneously with

seismic swarms.

Geodetic and seismicity observations in the Asal-Ghoubbet rift, Djibouti, five years

following the 1978 volcano-seismic crisis can be explained by continuous dyke

injection into crust of visco-elastic rheology with a thermal structure determined from

heat-flow measurements and constrained by seismicity depths (Cattin et al., 2005).

Inflation from magma injection explain the localised seismicity patterns and high slip

rates on faults close to the rift axis, as well as geodetically measured ground

deformation. An abrupt decrease in opening rate after 1984 -1986 is explained by the

end of magma injection (Cattin et al., 2005).

Alternatively, seismicity can occur after dyke intrusion and likely due to release of

stress as the crust returns to equilibrium near new dykes. In the Krafla spreading

segment, northern Iceland, clusters of seismicity at 1-3 km depth were recorded 5-8

years following dyke injection and surface fissuring (Arnott and Foulger, 1994). No

evidence of geothermal activity is observed at the surface and the seismicity is thus

rather attributed to stress release as lithosphere re-equilibrates near already cooled

dykes (Arnott and Foulger, 1994).

In the MER, the spatial correlation between earthquake locations, faults with fissural

eruptions, aligned cones and axial mafic intrusions, as described in section 8.1,

suggests that seismicity is induced by magma injection into the mid-upper crust.

However, the precise relationship between the timing of magma injection and seismic


103

activity is unclear, as no constraints on ground deformation during seismic swarms are

available during the recording period of the EAGLE network.

8.4 Implications for models of continental breakup

Hayward and Ebinger (1996) suggest that during continental breakup, localisation of

strain away from border faults to intra-rift grabens is initially controlled by the decrease

in lithospheric strength within the rift from a combination of lithospheric thinning and

resultant increased heating during continued extension. The increase in magmatism

with time is caused by progressive thinning of the lithosphere and eventual onset of

adiabatic decompression melting, with dyke injection accommodating the majority of

extension in the early stages of sea-floor spreading (Hayward and Ebinger, 1996).

However, new observations from the MER suggest that injection of magma within a

narrow zone along the rift axis enables the localisation of strain to a narrow zone during

the late stages of continental breakup. Magma injection accommodates the majority of

strain in the mid- upper crust, and a combination of dyking and magma induced faulting

accommodates strain in the brittle layers. Border faults are inactive as dyke injection

accommodates strain at lower stresses than required to initiate slip on large offset

border faults. The pattern of magma fed along-axis segmentation observed in the MER

is similar to slow- and ultra-slow spreading mid-ocean ridges where supply of melt from

zones of upwelling in the asthenosphere controls the along-axis segmentation of the

rift. This suggests that magma-fed along axis segmentation of mid-ocean ridges

initiates during the early stages of continental breakup and is independent of

segmented pattern of rift bounding border faults formed during the initial stages of

extension.

Observations from the MER support a model of magma-assisted rifting whereby the

combined effects of lithospheric stretching and heating by magma injection localises

thinning of the mantle lithosphere and facilitates extension at relatively small plate

driving forces (Buck, 2004). Observations do not support detachment fault models of

lithospheric extension but provide evidence that strain is accommodated by magma

injection within narrow zones (magmatic segments) along the rift axis that mark the

eventual boundary of continental breakup.

Chapter 9 - Conclusions

104

Chapter 9 Conclusions

From Oct 2001 - Jan 2003, 1957 earthquakes were located within the EAGLE network

of broadband seismic stations in the northern Main Ethiopian rift and on its uplifted rift

flanks. Excluding the MER - Red Sea rift intersection zone at Ankober, where seismic

activity may be caused by flexure accommodating differential subsidence at the oblique

intersection of the <11 Ma MER and the older Red Sea rift, seismicity within the rift is

localised to <20 km-wide, right-stepping, en echelon zones of Quaternary magmatism.

Seismicity in these magmatic segments is characterised by swarms of low magnitude

earthquakes located in clusters that parallel Quaternary faults, fissures and chains of

eruptive centres. The earthquakes in the magmatic segments are predominantly <14

km deep and may be triggered by dyke injection.

A local magnitude scale for Ethiopia has been developed from 30908 amplitude

measurements on simulated Wood-Anderson seismograms from 2139 earthquakes

recorded on EAGLE broadband instruments. The new magnitude scale uses a distance

normalization of 10-mm motion at 17 km distance for a magnitude 3.0 earthquake. The

distance correction is given by; -logAo = 1.196997 log(r/17) + 0.001066(r-17) + 2.0,

where r is hypocentral distance in kilometres. The distance correction shows that

ground-motion attenuation in Ethiopia is relatively high. The annual cumulative

frequency - magnitude distribution for 2001-2003 follows the relation; log N = 4.5 -

1.13ML, where N is the number of earthquakes per year of local magnitude Mc or

greater. The catalogue of events used in this study is complete above ML 2.1.

Earthquake focal mechanisms show predominantly normal dip-slip on faults striking ~N

to ~NNE. The orientation of the minimum compressive stress determined from focal

mechanisms is N103oE, consistent with elongation of Quaternary Calderas, geodetic

data and global plate kinematic constraints.

Shear-wave splitting from crustal earthquakes shows that the polarization direction of

the fast S-wave along the rift-axis is ~N to NNE, parallel to Quaternary to Recent faults,

aligned cones and the current maximum horizontal stress axis. The largest amounts of

Chapter 9 - Conclusions

105

upper crustal anisotropy are in the Quaternary magmatic segments, where the majority

of strain is accommodated by magma injection; anisotropy is most likely caused by

aligned melt-filled micro-cracks and dykes. The low amount of anisotropy beneath the

Ethiopian plateau is consistent with melt decrease away from the rift axis. These

results suggest the anisotropy is related to variable amounts of melt pocket alignment

in the crust, with a higher degree of dyke intrusion beneath magmatic segments. Melt-

induced anisotropy extends from the base of the lithosphere to the upper crust,

suggesting that magma injection helps localize strain and facilitate continental breakup.

Observations from the MER suggest that injection of magma within a narrow zone

along the rift axis enables the localisation of strain to a narrow zone during the late

stages of continental breakup. Magma injection accommodates the majority of strain in

the mid- upper crust and a combination of dyking and magma induced faulting

accommodates strain in the brittle layers. Border faults are inactive as dyke injection

accommodates strain at lower stresses than required to initiate slip on large offset

border faults. The similarities in along-axis segmentation in the MER and at slow-

spreading mid-ocean ridges suggests that magma-fed along-axis segmentation

initiates during the early stages of continental breakup and is independent of

segmented pattern of rift bounding border faults formed during the initial stages of

extension.

Observations from the MER support a model of magma assisted rifting whereby the

combined effects of lithospheric stretching and heating by magma injection localises

thinning of the mantle lithosphere and facilitates extension at relatively small plate

driving forces (Buck, 2004). Observations do not support detachment fault models of

lithospheric extension but provide evidence that strain is accommodated by magma

injection within narrow zones (magmatic segments) along the rift axis that mark the

eventual boundary of continental breakup.

Appendices

106

Appendix A EAGLE broadband seismic stations

Column header definitions

Stn. code: abbreviated station code

Lat. (oN): latitude of station location

Long. (oE): longitude of station location

Elev. (m): station elevation in metres

Stn. name: station name

Start (yr., dy.): date of station deployment in year and julian day

End (yr., dy.): date of station retrieval in year and julian day

Stn. code Lat. (oN) Long. (oE) Elev. (m) Stn. name Start (yr., dy.) End (yr., dy.)

EAGLE Phase I: IRIS network YJ, Instrument type: Güralp CMG-40T LEME 8.6115 38.6095 2108 Lemen 2001, 327 2002, 292 SHEE 9.9996 39.8946 1298 Shewa Robit 2001, 298 2003, 028 GEWE 10.006 40.5743 600 Gewane 2001, 324 2002, 344 ANKE 9.5927 39.7339 2981 Ankober 2001, 297 2003, 027 KOTE 9.3875 39.3961 2872 Kotu Gabeya 2001, 295 2002, 346 MELE 9.3106 40.2008 762 Melka Werer 2001, 326 2003, 023 MIEE 9.2416 40.7581 1349 Mieso 2001, 304 2003, 023 BEDE 8.9086 40.7710 1714 Bedesa 2001, 319 2003, 024 MECE 8.5938 40.3241 1775 Mechara 2001, 317 2003, 024 CHAE 9.3118 38.7624 2646 Chancho 2001, 323 2003, 018 SENE 9.1466 39.0166 2560 Sendafa 2001, 299 2003, 018 AREE 8.9285 39.4188 1826 Areriti 2001, 303 2003, 030 GTFE 8.9934 39.8376 1036 Gudina Tumsa 2001, 305 2003, 025 NURE 8.7012 39.7956 1182 Nura Hira 2001, 305 2003, 031 BORE 8.7259 39.5540 1253 Borechota 2001, 312 2003, 031 DONE 8.5090 39.5504 1312 Doni 2001, 306 2003, 031 MEKE 8.1623 38.8330 1897 Meki 2001, 304 2003, 031 DIKE 8.0627 39.5566 2754 Diksis 2001, 303 2003, 023 ADUE 8.5404 38.9019 1750 Adulala 2001, 306 2003, 030 DZEE 8.7803 38.9959 1907 Debre Zeit 2001, 328 2002, 338 WOLE 8.5339 37.9822 2058 Woliso 2002, 103 2003, 019 ASEE 7.9729 39.1317 2333 Asela 2002, 106 2003, 018 AWAE 8.9895 40.1659 956 Awash 2002, 107 2003, 022 BUTE 8.1170 38.3824 2093 Butajira 2002, 109 2003, 031 KARE 10.422 39.9349 1774 Kara Kore 2002, 120 2003, 028 HIRE 9.2221 41.1059 1845 Hirna 2002, 128 2002, 310 DEBE 8.7803 38.9959 1907 Debre Zeit 2001, 302 2001, 329 EAGLE Phase I: IRIS network YJ, Instrument type: Güralp CMG-3T INEE 9.8954 39.1431 2686 Inewari 2001, 296 2003, 029 AMME 8.3031 39.0934 1670 Amudi 2001, 304 2003, 032 ADEE 7.7909 39.9068 2485 Adele 2001, 313 2003, 019 EAGLE Phase II, IRIS network XJ, Instrument type: Güralp CMG-6TD

Appendices

107

E31 8.7774 39.8644 1009 Abadir Farm 2002, 277 2003, 038 E32 8.8473 40.0089 970 Nat. P.K.H.Q. 2002, 277 2002, 361 E33 8.9255 39.9290 979 Fentale South 2002, 278 2003, 033 E34 7.2143 38.5989 1934 Shashamene 2002, 289 2003, 042 E35 9.1336 40.1682 846 Awash Arba 2002, 277 2003, 033 E36 9.1075 40.0136 771 Hot Springs 2002, 278 2003, 033 E37 8.1744 38.6964 1799 Mukia 2002, 290 2003, 042 E38 8.9253 39.8432 1011 Elala 2002, 281 2002, 282 E39 9.2415 40.1334 770 Melka Sedi 2002, 277 2003, 033 E40 9.3616 40.2172 743 Melka Werer 2002, 277 2003, 033 E41 8.0096 38.5325 1905 Koshe 2002, 289 2003, 042 E42 8.8790 40.0961 1057 Kereyou 2002, 277 2003, 033 E43 9.2549 39.5023 3294 Ilekase 2002, 296 2003, 035 E44 9.6722 39.5249 2831 Debre Birhan 2002, 283 2003, 035 E45 7.7881 38.7948 1932 Aluto Power 2002, 289 2003, 003 E46 8.7090 39.6909 1242 Meki Dela 2002, 278 2003, 034 E47 8.4649 39.4532 1448 Bofa 2002, 280 2003, 038 E48 7.6239 38.9907 2613 Ego 2002, 295 2003, 042 E49 8.3144 39.3212 1711 Dera 2002, 282 2003, 038 E50 8.2739 39.4958 2074 Sire 2002, 282 2003, 038 E51 8.1485 39.3478 2080 Huruta 2002, 288 2003, 039 E52 8.1380 39.2405 2200 Iteya 2002, 282 2003, 039 E53 8.0433 39.0114 1704 Ogolcha 2002, 290 2003, 041 E54 8.1182 39.1385 2084 Danisa 2002, 288 2003, 039 E55 8.2950 38.9492 1683 Alem Tena 2002, 291 2003, 042 E56 8.4622 39.0637 1637 Ejersa 2002,288 2003, 042 E57 8.5848 39.1320 1823 Mojo 2002,280 2003, 036 E58 8.6909 39.1826 2057 Gogli 2002,282 2003, 037 E59 8.7059 39.3516 1680 Wolenchiti 2002, 278 2003, 034 E60 8.6215 39.4488 1626 Boset Track 2002, 282 2003, 034 E61 8.8974 39.6229 1155 Melka Jilo 2002, 278 2003, 034 E62 8.8313 39.7301 1443 Kone 2002, 281 2002, 282 E63 8.2600 39.2359 1781 Dawero 2002, 287 2003, 039 E64 8.5680 39.2907 1753 Nazret 2002, 280 2003, 037 E65 8.4031 39.2114 1550 Wonji Shoa 2002, 283 2003, 037 E66 9.0326 39.5300 1720 Aroge Minjar 2002, 330 2003, 037 E67 8.3844 39.6810 2144 Arboye 2002, 281 2003, 038 E68 8.7821 39.2628 2290 Ejere 2002, 282 2003, 037 E69 7.9296 38.7216 1675 Ziway 2002, 288 2003, 041 E70 8.8823 39.1544 2227 Tulu Dimtu 2002, 291 2003, 036 E71 8.6935 38.8955 1980 Dire 2002, 283 2003, 036 E72 8.4871 39.8329 1577 Abomsa 2002, 281 2003, 038 E73 7.7375 39.0262 2496 Lolee Abosere 2002, 295 2003, 042 E74 8.3564 38.8456 1714 Kile Doyo 2002, 331 2003, 042 E75 7.9141 38.9512 1745 Kiyensho 2002, 294 2003, 041 E76 7.7245 38.6537 1674 Bulbula 2002, 289 2003, 041 E77 7.8643 38.7914 1668 Chefe Jila 2002, 290 2003, 041 E78 8.5931 39.6977 1220 Merti 2002, 280 2003, 038 E79 7.6294 38.7056 1588 Langano 2002, 289 2003, 041 E80 8.4781 39.3106 1655 Adulala Koshe 2002, 280 2003, 037 E81 8.7938 39.6683 1330 Road Camp 2002, 365 2003, 032 E82 8.8459 40.0096 969 Awash Park 2002, 361 2003, 033 E83 7.8019 38.7919 1905 Geotherm 2003, 003 2003, 042 E84 8.6983 39.4005 1536 Eyaya 2002, 341 2003, 016 E85 8.4572 39.5926 1323 Tibila 2002, 336 2003, 016 EAGLE Phase III, IRIS network XM, Instrument type: Güralp CMG-6TD 1001 9.9845 38.2801 2524 1001 2002, 329 2003, 019 1004 9.9600 38.2973 2554 1004 2002, 329 2003, 019 1008 9.9318 38.3222 2544 1008 2002, 329 2003, 019 1011 9.9106 38.3419 2578 1011 2002, 329 2003, 019 1014 9.8891 38.3613 2554 1014 2002, 328 2003, 019 1018 9.8552 38.3819 2584 1018 2002, 328 2003, 019 1023 9.8152 38.4045 2564 1023 2002, 329 2003, 019 1026 9.8005 38.4165 2574 1026 2002, 329 2003, 019 1030 9.7773 38.4588 2588 1030 2002, 332 2003, 019

Appendices

108

1037 9.7380 38.5142 2674 1037 2002, 330 2003, 018 1042 9.6897 38.5300 2557 1042 2002, 330 2003, 018 1046 9.6577 38.5177 2478 1046 2002, 330 2003, 018 1054 9.6150 38.5853 2669 1054 2002, 330 2003, 020 1058 9.5805 38.5966 2555 1058 2002, 330 2003, 019 1062 9.5487 38.5824 2450 1062 2002, 330 2003, 016 1069 9.4889 38.5840 1506 1069 2002, 346 2003, 016 1077 9.4399 38.6462 2385 1077 2002, 345 2003, 019 1081 9.3990 38.6631 2426 1081 2002, 345 2003, 019 1085 9.3667 38.6711 2484 1085 2002, 325 2003, 019 1089 9.3391 38.6930 2602 1089 2002, 325 2003, 019 1094 9.3062 38.7293 2578 1094 2002, 327 2003, 019 1097 9.3025 38.7516 2614 1097 2002, 325 2003, 018 1101 9.2831 38.7820 2717 1101 2002, 327 2003, 018 1105 9.2595 38.8064 2889 1105 2002, 345 2003, 019 1110 9.2552 38.8513 3262 1110 2002, 327 2003, 018 1114 9.2122 38.8926 2968 1114 2002, 340 2003, 017 1120 9.1814 38.9275 2644 1120 2002, 340 2003, 018 1125 9.1478 38.9611 2548 1125 2002, 327 2003, 018 1132 9.1313 39.0029 2538 1132 2002, 340 2003, 018 1138 9.1189 39.0490 2523 1138 2002, 326 2003, 018 1141 9.0946 39.0664 2531 1141 2002, 327 2003, 018 1146 9.0560 39.0930 2525 1146 2002, 326 2003, 018 1151 9.0196 39.1200 2484 1151 2002, 326 2003, 018 1155 8.9825 39.1244 2403 1155 2002, 326 2003, 018 1157 8.9679 39.1301 2348 1157 2002, 338 2003, 018 1163 8.9301 39.1470 2309 1163 2002, 338 2003, 018 1171 8.9084 39.2359 2356 1171 2002, 342 2003, 017 1179 8.8447 39.2442 2415 1179 2002, 342 2003, 017 1182 8.8183 39.2546 2377 1182 2002, 342 2003, 017 1189 9.3391 38.6930 2602 1189 2002, 342 2003, 017 1195 8.7259 39.3352 1746 1195 2002, 341 2003, 016 1204 8.6983 39.4005 1536 1204 2002, 341 2003, 016 1209 8.6562 39.4281 1460 1209 2002, 336 2003, 018 1219 8.5801 39.4548 1870 1219 2002, 341 2003, 016 1226 8.5289 39.4794 1817 1226 2002, 337 2003, 016 1231 8.4882 39.4628 1479 1231 2002, 337 2003, 016 1235 8.4731 39.4822 1449 1235 2002, 337 2003, 016 1238 8.4819 39.5074 1408 1238 2002, 343 2003, 016 1242 8.4961 39.5398 1304 1242 2002, 341 2003, 016 1246 8.5038 39.5704 1267 1246 2002, 341 2003, 016 1252 8.4572 39.5926 1323 1252 2002, 336 2003, 016 1258 8.4177 39.6326 1803 1258 2002, 341 2003, 017 1262 8.3894 39.6572 2087 1262 2002, 341 2003, 017 1266 8.3633 39.6719 2170 1266 2002, 330 2003, 018 1270 8.3269 39.6711 2591 1270 2002, 341 2003, 017 1274 8.2853 39.6596 2606 1274 2002, 342 2003, 017 1278 8.2712 39.6898 2646 1278 2002, 342 2003, 017 1281 8.2488 39.6946 2708 1281 2002, 342 2003, 017 1285 8.2122 39.698 2686 1285 2002, 342 2003, 017 1290 8.1775 39.7186 2687 1290 2002, 342 2003, 017 1296 8.1272 39.7164 2616 1296 2002, 342 2003, 017 1301 8.0962 39.6904 2602 1301 2002, 342 2003, 017 1306 8.0494 39.6922 2549 1306 2002, 342 2003, 017 1310 8.016 39.6868 2531 1310 2002, 342 2003, 017 1315 7.9763 39.6889 2478 1315 2002, 338 2003, 017 1320 7.9671 39.7338 2472 1320 2002, 338 2003, 018 1324 7.9340 39.7520 2476 1324 2002, 338 2003, 018 1329 7.9015 39.7844 2477 1329 2002, 338 2003, 018 1333 7.8732 39.8074 2465 1333 2002, 338 2003, 018 1337 7.8508 39.8356 2489 1337 2002, 338 2003, 018 1340 7.8338 39.8570 2479 1340 2002, 338 2003, 018 1344 7.8059 39.8842 2474 1344 2002, 338 2003, 017 1346 7.7950 39.8979 2472 1346 2002, 338 2003, 019 1351 7.7810 39.9368 2462 1351 2002, 338 2003, 017 1356 7.7539 39.9743 2450 1356 2002, 338 2003, 019 1360 7.7259 40.0091 2458 1360 2002, 340 2003, 019

Appendices

109

1366 7.7252 40.0676 2483 1366 2002, 337 2003, 019 1370 7.706 40.1015 2485 1370 2002, 337 2003, 019 1373 7.7068 40.1353 2490 1373 2002, 337 2003, 019 1377 7.6889 40.1664 2498 1377 2002, 337 2003, 016 1381 7.6726 40.2010 2496 1381 2002, 337 2003, 016 1384 7.3815 40.0146 2342 1384 2002, 332 2003, 017 1387 7.3837 40.0420 2380 1387 2002, 332 2003, 017 1394 7.3886 40.1083 2380 1394 2002, 333 2003, 017 1400 7.3826 40.1661 2383 1400 2002, 333 2003, 017 1403 7.3718 40.1924 2388 1403 2002, 333 2003, 017 1416 7.3163 40.2653 2412 1416 2002, 333 2003, 017 1430 7.2323 40.3608 2361 1430 2002, 333 2003, 017 1435 7.2252 40.4086 2313 1435 2002, 333 2003, 017 1437 7.2309 40.4216 2273 1437 2002, 333 2003, 017 1443 7.2592 40.4743 2214 1443 2002, 333 2003, 017 Permanent Stations, Geophysical Observatory Addis Ababa University FURI 8.9030 38.6880 Mount Furi AAE 9.0350 38.7670 Addis Ababa WNDE 7.0980 38.6350 Wendo Genet

Appendices

110

Appendix B Catalogue of earthquake epicentres located with 1-D velocity model Column header definitions

Date (yr.-mo.-dy.): date of earthquake in year, month and day

Time (hr.:min.:sec.): time of earthquake in hour, minutes and seconds

Lat. (oN): latitude of earthquake epicentre

Long. (oE): longitude of earthquake epicentre

ML: local magnitude

Date (yr.-mo.-dy.) Time (hr.:min.:sec.) Lat. (oN) Long. (oE) ML

2001-10-26 20:33:55.32 10.0128 39.8163 2.658 2001-10-26 20:44:33.50 9.5370 39.6125 1.575 2001-10-26 22:00:11.80 9.4445 39.6820 1.774 2001-10-26 22:39:51.20 9.4432 39.6877 1.607 2001-10-27 00:46:35.03 9.6278 39.5843 1.236 2001-10-27 09:07:52.98 9.6327 39.5693 2.419 2001-10-27 15:08:23.81 9.5663 39.6112 2.075 2001-10-27 22:04:51.21 9.6057 39.5875 1.456 2001-10-27 23:20:48.20 9.5542 39.5983 0.689 2001-10-27 23:39:33.15 9.6157 39.5797 1.884 2001-10-28 02:20:42.12 9.6090 39.5793 1.424 2001-10-28 05:24:50.33 9.6010 39.5855 2.113 2001-10-28 17:18:15.05 9.4493 39.6763 2.428 2001-10-28 17:31:28.02 9.7740 39.2892 3.184 2001-10-28 18:17:38.93 9.4668 39.6628 2.099 2001-10-28 19:40:44.93 9.6242 39.5745 1.562 2001-10-28 23:23:21.39 9.6175 39.5810 0.937 2001-10-29 01:47:05.34 9.5792 39.6787 0.912 2001-10-29 01:55:16.40 9.6158 39.5872 1.442 2001-10-29 05:24:50.56 9.5603 39.5948 1.044 2001-10-29 06:33:17.84 9.4457 39.6807 2.154 2001-10-29 10:46:24.87 9.4517 39.6950 3.507 2001-10-29 19:32:07.33 9.6153 39.5785 2.098 2001-10-29 20:44:21.98 9.5977 39.5360 1.198 2001-10-30 09:44:48.96 9.6332 39.5793 2.758 2001-10-30 18:12:30.25 9.5970 39.4623 1.586 2001-10-30 18:34:33.61 9.4638 39.6757 1.632 2001-10-30 19:56:27.23 9.6087 39.5778 1.639 2001-10-30 21:01:19.75 9.4585 39.6687 1.551 2001-10-30 21:27:36.95 9.4592 39.6665 1.891 2001-10-30 22:18:43.44 9.6107 39.5832 1.454 2001-10-30 23:27:43.47 9.4592 39.6733 1.647 2001-10-30 23:45:45.98 9.4618 39.6702 1.799 2001-10-30 23:51:13.19 9.4648 39.6698 1.419 2001-10-31 00:05:04.19 9.6033 39.5843 1.704 2001-10-31 00:46:04.16 9.6090 39.5455 0.923 2001-10-31 01:12:22.06 9.4602 39.6715 2.062 2001-10-31 19:53:07.38 9.4595 39.6730 1.798 2001-10-31 20:32:48.86 9.6073 39.5807 0.716 2001-10-31 20:51:55.14 9.4363 39.6848 1.990 2001-11-01 19:40:12.19 9.5960 39.5278 1.153 2001-11-01 21:02:21.93 9.5778 39.4940 1.621 2001-11-01 21:54:51.76 9.4717 39.6725 1.657 2001-11-01 22:11:25.73 9.4760 39.6698 1.245 2001-11-01 22:46:17.34 9.5563 39.6747 1.029 2001-11-02 01:04:23.78 9.4555 39.6712 1.458 2001-11-02 01:16:27.57 9.5592 39.6242 1.698 2001-11-02 06:35:40.04 9.5290 39.5998 2.472 2001-11-02 08:21:44.54 9.4482 39.7127 1.906 2001-11-02 16:23:44 11.7900 43.1900 5.180 2001-11-02 21:31:22.10 9.6155 39.5592 1.289 2001-11-02 23:04:23.95 9.4733 39.6598 2.297 2001-11-03 00:21:30.21 9.5817 39.6037 1.379 2001-11-03 00:22:35.45 9.8325 39.5002 1.512 2001-11-03 00:42:28.69 9.6198 39.5767 1.296 2001-11-03 08:43:06.26 9.6178 39.5863 1.966

Appendices

111

2001-11-03 16:43:41.17 9.6142 39.5827 1.884 2001-11-03 17:20:19.10 9.4573 39.6718 2.399 2001-11-03 17:46:38.88 9.5973 39.5352 1.637 2001-11-04 01:05:52.41 9.6095 39.5902 1.258 2001-11-04 11:00:31.31 9.6110 39.5850 2.154 2001-11-04 13:32:04.18 9.6133 39.5830 1.885 2001-11-04 15:08:29.60 9.4587 39.6853 1.774 2001-11-04 16:24:15.79 9.9320 39.8035 2.268 2001-11-04 17:41:24.42 9.6202 39.5780 1.389 2001-11-04 19:16:03.31 9.5975 39.5470 1.126 2001-11-04 20:07:50.00 9.6188 39.5837 1.029 2001-11-04 22:23:16.94 9.4433 39.6780 0.939 2001-11-04 23:30:55.98 9.6120 39.5768 1.468 2001-11-04 23:31:48.45 9.4648 39.6582 1.791 2001-11-04 23:51:10.87 9.4633 39.6697 1.670 2001-11-05 00:05:44.33 9.4645 39.6743 0.955 2001-11-05 15:29:11.72 9.6120 39.5783 2.433 2001-11-05 15:55:32.59 9.4578 39.6743 2.354 2001-11-05 16:10:15.17 9.6070 39.5433 1.580 2001-11-05 17:20:56.27 9.6168 39.5795 1.939 2001-11-05 17:52:14.36 9.4640 39.6773 2.654 2001-11-05 20:13:27.90 9.4475 39.6760 1.242 2001-11-05 22:26:31.10 9.4715 39.6800 1.315 2001-11-05 22:36:28.04 9.5963 39.6005 0.753 2001-11-06 01:08:32.59 9.4680 39.6507 1.447 2001-11-06 01:09:03.16 9.5368 39.6007 1.470 2001-11-06 01:10:32.28 9.4793 39.6463 1.571 2001-11-06 01:15:54.37 9.6123 39.5680 1.217 2001-11-06 05:38:15.51 9.4550 39.6753 1.983 2001-11-06 14:21:01.06 9.4533 39.6760 2.231 2001-11-06 18:38:45.04 9.3948 39.8765 2.008 2001-11-06 19:03:21.02 9.6032 39.5590 1.125 2001-11-06 19:23:46.20 9.6645 39.7608 0.872 2001-11-06 19:57:38.45 9.4900 39.7233 1.139 2001-11-06 20:27:40.82 9.6108 39.5853 1.122 2001-11-06 21:38:28.29 9.6213 39.5770 0.822 2001-11-07 18:14:59.39 9.6130 39.5785 1.574 2001-11-07 23:23:11.12 9.6140 39.5777 1.377 2001-11-08 01:36:22.74 9.2408 39.9162 1.253 2001-11-08 07:02:43.45 9.4625 39.6680 2.387 2001-11-08 17:56:17.70 9.6115 39.5767 1.343 2001-11-09 17:54:27.28 9.4622 39.6602 2.049 2001-11-09 22:03:35.52 9.4572 39.6738 1.718 2001-11-10 02:35:10.41 9.6202 39.5777 1.446 2001-11-10 04:38:19.18 9.4573 39.6690 2.484 2001-11-10 09:32:53.39 9.1593 39.8742 2.570 2001-11-10 13:35:29.75 9.2393 39.4203 1.942 2001-11-11 02:19:22.28 9.4475 39.6852 2.200 2001-11-11 21:05:21.95 9.4560 39.6672 2.373 2001-11-11 21:21:51.69 9.6027 39.5855 1.264 2001-11-11 22:32:40.81 9.4563 39.6808 3.253 2001-11-11 22:45:25.24 9.6055 39.5933 1.695 2001-11-11 22:57:13.40 9.4510 39.6847 1.933 2001-11-11 23:00:46.97 9.4653 39.6700 1.999 2001-11-11 23:14:22.98 9.4600 39.6757 2.109 2001-11-11 23:35:43.89 9.4903 39.6752 2.320 2001-11-13 00:50:06.29 9.6517 39.5173 1.007 2001-11-14 01:58:11.77 9.7070 39.7370 1.550 2001-11-14 02:18:25.19 9.4595 39.6828 1.892 2001-11-14 20:57:39.00 9.5108 39.6740 1.786 2001-11-14 21:14:39.19 9.6640 39.5657 1.070 2001-11-14 22:16:00.34 9.5748 39.7055 1.096 2001-11-14 22:53:15.98 9.4545 39.6868 1.927 2001-11-15 00:30:55.96 9.4628 39.6932 1.886 2001-11-15 01:04:34.91 9.2865 39.9927 1.411 2001-11-15 01:15:45.45 9.4283 39.6925 1.238 2001-11-15 02:54:01.97 9.4482 39.7000 1.701 2001-11-15 14:33:40.64 9.3250 40.0337 2.195 2001-11-15 18:22:05.12 9.3513 40.0202 1.705 2001-11-15 20:33:30.74 9.3168 40.0033 1.827 2001-11-16 00:49:56.73 9.2965 39.9970 2.754 2001-11-16 01:54:12.74 9.3148 40.0243 1.538 2001-11-16 02:12:07.81 9.3118 40.0032 2.407 2001-11-16 09:17:37.11 9.3157 40.0058 3.290 2001-11-16 14:39:54.09 9.4293 39.6807 2.415 2001-11-17 03:28:16.74 9.2918 39.9875 2.766 2001-11-17 03:56:21.77 9.5900 39.5490 1.568 2001-11-17 11:34:09.77 9.3130 39.9973 2.624 2001-11-17 13:02:26.13 9.3257 39.9863 1.967 2001-11-17 13:21:13.64 9.2218 40.0310 2.029 2001-11-17 19:17:43.42 10.2715 40.4717 2.317 2001-11-18 00:09:25.77 9.3792 40.0433 0.968 2001-11-18 05:19:59.06 9.0655 39.7825 2.501 2001-11-18 16:29:16.02 9.3097 40.0042 2.266 2001-11-18 18:56:15.33 9.8808 39.5093 1.635 2001-11-18 19:12:10.30 9.3168 40.0037 2.325 2001-11-18 19:57:23.71 9.3158 39.9965 1.846 2001-11-18 21:54:49.21 9.2695 40.0087 1.657 2001-11-18 22:03:37.98 9.2950 39.9922 1.589 2001-11-19 02:13:41.06 9.3268 40.0153 2.240 2001-11-19 03:08:10.42 9.3388 40.0105 2.120 2001-11-19 03:58:20.17 9.3115 39.9777 2.508 2001-11-19 15:44:52.54 9.2990 39.9992 2.426 2001-11-19 21:32:21.89 9.3023 39.9990 2.044

Appendices

112

2001-11-20 02:08:53.41 9.3083 39.9760 1.690 2001-11-20 18:39:27.42 9.2135 39.9978 2.369 2001-11-21 21:12:11.97 9.4093 39.6282 1.105 2001-11-22 21:51:34.84 9.4382 39.6862 0.868 2001-11-23 05:20:37.98 9.4128 39.6788 2.171 2001-11-23 09:04:36.95 9.3058 39.9890 2.791 2001-11-23 14:45:32.93 9.3053 39.9908 3.245 2001-11-23 14:59:13.54 9.3110 39.9825 2.121 2001-11-23 18:33:23.79 9.2980 39.9877 2.510 2001-11-23 18:36:29.37 9.3132 39.9808 2.118 2001-11-23 22:29:55.44 9.2995 39.9898 2.254 2001-11-24 01:12:55.70 9.3068 39.9985 2.197 2001-11-24 01:54:50.02 9.3973 39.5812 1.103 2001-11-24 02:25:26.47 9.2697 40.0212 1.580 2001-11-24 13:03:38.84 9.5492 38.5650 3.121 2001-11-25 00:28:08.81 9.2873 39.9695 1.934 2001-11-25 00:52:16.54 9.2935 40.0183 1.896 2001-11-25 01:30:00.56 9.3227 39.9813 1.608 2001-11-25 09:20:28.80 9.3045 39.9848 2.808 2001-11-25 20:27:16.42 9.3970 39.5330 0.975 2001-11-25 20:42:06.77 9.4477 39.6772 1.644 2001-11-25 23:08:17.00 9.7065 39.5408 1.095 2001-11-26 00:31:39.27 9.2927 40.0200 1.529 2001-11-26 09:18:16.37 9.3845 40.0327 1.064 2001-11-26 18:02:41.69 9.2807 39.9943 1.243 2001-11-26 00:51:43.83 9.2937 39.9955 2.247 2001-11-26 01:31:03.18 9.4963 38.6192 1.670 2001-11-26 13:33:05.68 9.3492 39.6470 2.116 2001-11-26 20:59:49.73 7.5885 38.0228 1.785 2001-11-27 00:46:21.86 9.4438 39.6963 1.127 2001-11-27 01:08:08.18 9.4537 39.6873 1.508 2001-11-27 01:20:30.20 9.4577 39.6827 1.486 2001-11-27 01:21:31.35 9.4612 39.6862 1.570 2001-11-27 10:28:54.56 9.2698 40.1928 1.949 2001-11-27 10:31:17.13 9.2642 40.1965 1.452 2001-11-27 12:31:37.67 8.8783 38.8128 1.443 2001-11-27 19:53:25.27 9.2705 40.1453 1.871 2001-11-27 19:58:13.61 9.2833 40.1690 1.339 2001-11-27 20:02:27.52 9.2557 40.2102 0.932 2001-11-27 20:11:12.37 9.3230 40.1267 0.982 2001-11-27 21:13:26.72 9.4702 39.6863 1.214 2001-11-27 21:35:35.77 9.4652 39.6770 2.648 2001-11-27 22:02:37.28 9.4333 39.6672 0.902 2001-11-27 22:15:26.21 9.4658 39.6647 1.058 2001-11-27 23:51:51.88 9.2940 39.9918 1.723 2001-11-27 23:55:11.40 9.2973 39.9912 1.332 2001-11-28 00:30:27.35 9.2883 39.9918 1.371 2001-11-28 01:08:44.10 9.2645 40.2040 0.808 2001-11-28 05:17:35.04 9.2807 40.0187 2.455 2001-11-28 09:46:07.81 9.2843 39.9877 1.830 2001-11-28 14:18:36.39 9.2830 39.9788 1.740 2001-11-28 16:53:01.46 9.4032 39.6957 1.302 2001-11-28 22:35:10.77 9.4302 39.6625 1.094 2001-11-28 22:49:43.22 9.4560 39.6813 1.273 2001-11-29 01:35:41.85 9.2855 39.9852 1.913 2001-11-29 01:41:27.70 9.2775 39.9762 1.199 2001-11-29 19:06:06.29 9.2895 39.9790 2.563 2001-11-29 19:05:54.63 9.2673 39.9722 2.573 2001-11-29 19:11:42.06 9.2875 39.9835 2.065 2001-11-29 19:50:34.94 9.2822 39.9785 1.764 2001-11-29 19:50:40.71 9.2890 39.9760 1.847 2001-11-29 20:11:27.85 9.3035 39.9870 2.102 2001-11-29 20:27:13.63 9.2837 39.9792 3.080 2001-11-29 21:06:48.79 9.2903 39.9772 1.874 2001-11-29 21:14:26.47 9.2817 39.9815 1.890 2001-11-29 21:51:39.37 9.2820 39.9767 1.457 2001-11-29 21:56:57.51 9.2745 39.9770 1.290 2001-11-29 21:57:22.67 9.2812 39.9867 1.153 2001-11-29 22:09:53.70 9.2775 39.9833 1.834 2001-11-29 22:22:51.01 9.2693 39.9692 1.217 2001-11-29 22:25:12.73 9.4722 39.6560 1.033 2001-11-29 23:46:36.49 9.2797 39.9638 1.174 2001-11-30 00:35:02.39 9.2848 39.9830 1.993 2001-11-30 00:42:59.51 9.2767 39.9808 2.238 2001-11-30 00:44:33.20 9.2853 39.9797 2.407 2001-11-30 00:45:01.64 9.2928 40.0003 1.989 2001-11-30 00:48:22.06 9.2885 39.9863 1.533 2001-11-30 01:01:02.72 9.2938 39.9822 1.392 2001-11-30 01:16:02.84 9.2967 39.9807 1.671 2001-11-30 02:29:15.64 9.2853 39.9727 2.284 2001-11-30 02:58:52.23 9.2705 39.9808 1.538 2001-11-30 06:39:07.95 9.2915 39.9937 2.400 2001-11-30 11:44:06.93 9.2840 39.9758 2.383 2001-11-30 20:21:46.86 9.4232 39.6855 1.342 2001-11-30 23:28:52.33 9.4425 39.6605 1.034 2001-12-01 00:12:38.15 9.4280 39.6575 1.034 2001-12-01 00:23:47.84 9.5793 39.7070 0.759 2001-12-01 00:41:03.71 9.4288 39.6907 0.977 2001-12-01 02:03:42.75 9.4010 39.6333 0.953 2001-12-01 02:26:28.13 7.6967 38.8080 1.275 2001-12-01 03:17:31.05 9.2808 39.9815 1.989 2001-12-01 03:48:18.18 9.1757 39.9100 1.567 2001-12-01 09:17:22.36 9.4045 39.7045 1.540 2001-12-01 12:59:47.38 10.2762 39.8440 2.922

Appendices

113

2001-12-01 18:15:49.60 9.3917 39.6162 0.897 2001-12-01 18:16:21.80 9.4337 39.6982 1.398 2001-12-01 20:43:04.28 9.8803 41.2137 1.440 2001-12-01 21:53:29.19 9.4330 39.6912 1.331 2001-12-03 09:35:24.93 8.9822 38.6978 1.184 2001-12-05 15:52:37 12.6700 40.5300 4.310 2001-12-05 18:49:14.27 9.1790 40.1328 1.811 2001-12-05 19:44:20.97 10.1205 39.7112 1.589 2001-12-05 22:00:16.78 9.3833 39.6648 1.199 2001-12-05 22:41:50.53 9.5202 39.6325 0.810 2001-12-05 22:47:28.42 9.6392 39.5772 0.656 2001-12-06 19:09:29.75 9.4328 39.5148 1.425 2001-12-07 00:52:00.27 9.4633 39.6742 1.781 2001-12-07 01:18:44.88 9.5847 39.5880 0.979 2001-12-08 08:58:40.20 9.6107 39.7973 1.817 2001-12-08 13:51:17.91 9.0723 40.4512 2.215 2001-12-08 15:00:50.06 9.4032 39.3202 1.068 2001-12-08 16:27:43.66 9.4503 39.6865 2.405 2001-12-08 18:01:37.54 10.6725 41.0637 2.268 2001-12-08 21:53:46.94 9.3905 39.6757 1.068 2001-12-08 23:23:56.35 9.4480 39.6853 1.678 2001-12-09 12:03:34.14 9.7773 40.4950 2.320 2001-12-09 19:37:04.96 9.4462 39.6903 1.734 2001-12-09 20:00:52.63 9.1837 40.1412 2.050 2001-12-09 21:48:18.63 10.4868 39.7678 1.769 2001-12-09 22:07:39.93 9.4105 39.6302 1.012 2001-12-10 15:50:52.97 9.1668 40.1642 1.393 2001-12-10 20:06:19.97 9.6528 39.7318 1.304 2001-12-10 20:19:14.74 9.4622 39.6883 2.392 2001-12-11 00:08:45.97 9.4600 39.6820 1.939 2001-12-11 00:27:34.07 9.4545 39.6922 2.002 2001-12-11 00:52:08.90 9.4090 39.6682 1.269 2001-12-11 07:02:13.24 9.4097 39.6903 2.149 2001-12-11 19:32:12.60 9.5848 39.5233 1.048 2001-12-11 21:06:34.89 9.4532 39.6918 1.896 2001-12-11 21:13:55.50 9.4442 39.6890 1.523 2001-12-11 21:36:17.19 9.4425 39.6860 1.747 2001-12-11 23:11:05.09 9.4588 39.6940 2.019 2001-12-11 23:12:25.00 9.4525 39.6785 1.505 2001-12-12 09:40:14.65 9.3503 39.3255 1.267 2001-12-12 13:33:53.10 9.4380 39.6940 2.414 2001-12-12 13:46:34.43 9.4437 39.7007 2.276 2001-12-12 14:03:24.83 9.4235 39.6837 1.652 2001-12-12 14:27:20.99 9.4373 39.6802 2.657 2001-12-12 15:20:37.91 9.4530 39.6707 1.631 2001-12-12 21:45:40.05 9.6072 40.2932 2.044 2001-12-12 22:42:07.03 9.4462 39.6822 1.644 2001-12-12 22:56:15.21 9.4520 39.6892 1.352 2001-12-13 00:39:02.02 9.4938 39.6407 1.118 2001-12-13 01:12:10.49 9.4565 39.6933 2.111 2001-12-13 01:26:29.89 9.4383 39.6838 1.297 2001-12-13 02:14:38.85 9.4565 39.6963 3.214 2001-12-13 03:42:20.48 9.5602 39.6635 1.202 2001-12-13 04:25:45.49 9.4290 39.6933 2.037 2001-12-13 04:33:57.33 8.3562 39.0417 1.930 2001-12-13 21:31:12.71 9.4330 39.6872 1.830 2001-12-13 22:42:17.32 9.3842 39.6372 0.998 2001-12-14 03:18:32.58 9.5093 40.2172 1.838 2001-12-14 03:22:24.25 9.4728 40.2235 1.828 2001-12-14 03:29:40.46 9.5390 39.6050 1.284 2001-12-14 03:39:13.03 9.4258 39.6572 1.300 2001-12-14 04:22:29.30 9.4767 40.2212 1.778 2001-12-14 10:28:44.24 9.4387 39.6988 2.289 2001-12-15 00:59:14.29 9.2568 39.9888 1.157 2001-12-15 01:08:40.08 9.4552 39.6912 2.000 2001-12-15 03:34:39.05 8.4248 39.8313 1.691 2001-12-15 14:07:30.60 9.3940 39.6907 1.764 2001-12-15 15:34:55.00 9.4118 39.6782 1.916 2001-12-15 19:55:14.74 8.8172 40.4670 1.187 2001-12-15 22:40:46.54 8.8260 39.7888 0.736 2001-12-15 22:50:35.59 9.4172 39.6867 1.262 2001-12-15 23:01:55.16 9.4460 39.6868 1.498 2001-12-15 23:44:04.20 9.0022 40.7607 0.617 2001-12-16 00:04:31.20 9.4210 39.6705 1.087 2001-12-16 07:33:00.54 9.3307 39.6743 1.491 2001-12-16 12:44:49.72 9.2632 40.2073 1.166 2001-12-16 20:55:58.58 9.1335 40.0515 0.909 2001-12-16 22:53:14.95 9.4247 39.6862 1.216 2001-12-17 00:01:04.04 9.9058 41.3273 1.481 2001-12-17 02:22:19.69 9.4138 39.6770 1.369 2001-12-17 05:33:45.42 9.4345 39.6955 1.698 2001-12-17 05:41:44.12 9.4443 39.6742 1.744 2001-12-17 14:55:57.86 9.4343 39.6758 1.522 2001-12-17 18:26:13.73 9.4477 39.6823 2.036 2001-12-17 18:36:43.50 9.4330 39.6838 1.337 2001-12-17 20:49:17.97 10.1150 41.5278 2.318 2001-12-17 22:24:13.82 7.4593 38.8908 1.023 2001-12-18 20:01:16.55 9.4373 39.6920 1.859 2001-12-18 20:01:55.52 9.4368 39.6850 1.914 2001-12-18 20:02:32.55 9.4457 39.6782 1.469 2001-12-18 22:07:30.81 9.4398 39.6900 1.373 2001-12-18 23:28:54.62 9.8457 41.1018 2.140 2001-12-19 00:53:11.21 9.4258 39.6865 1.258 2001-12-19 02:55:35.03 9.4452 39.6820 1.490

Appendices

114

2001-12-19 03:03:25.70 9.4397 39.6903 1.378 2001-12-19 05:28:01.66 9.8937 40.7317 2.374 2001-12-19 18:12:48.36 9.4607 39.6667 1.411 2001-12-19 21:03:08.64 8.1573 39.1337 1.176 2001-12-20 00:02:40.75 9.3998 39.6787 1.303 2001-12-20 00:56:04.86 9.4235 39.6922 1.204 2001-12-20 02:45:56.29 7.7353 38.7370 1.550 2001-12-20 04:31:33.10 9.4690 39.7005 2.837 2001-12-20 04:56:16.34 9.4378 39.6980 1.939 2001-12-20 05:29:40.21 9.4413 39.6860 1.503 2001-12-20 07:54:04.59 9.1808 40.1425 2.082 2001-12-20 11:07:40.88 9.4480 39.6978 2.618 2001-12-20 12:35:27.80 9.4498 39.7018 2.459 2001-12-20 19:45:23.10 9.4415 39.6740 1.305 2001-12-20 19:50:27.82 9.4372 39.6800 1.563 2001-12-20 20:57:01.87 7.7493 38.8012 1.436 2001-12-20 21:45:04.36 9.4247 39.6905 1.355 2001-12-20 22:59:01.68 9.4462 39.6877 1.477 2001-12-21 00:36:10.46 7.7445 38.7400 1.529 2001-12-21 00:38:12.66 8.2743 39.0292 0.949 2001-12-21 00:53:53.69 9.4527 39.6927 2.770 2001-12-21 01:13:14.03 9.4345 39.6828 1.408 2001-12-21 01:34:45.96 9.4455 39.7002 2.479 2001-12-21 01:41:54.31 9.4162 39.6875 1.240 2001-12-21 01:52:09.32 9.4532 39.7032 1.619 2001-12-21 02:01:38.26 7.4157 38.6125 1.441 2001-12-21 13:41:59.86 9.7072 39.8110 2.237 2001-12-21 16:47:35.79 9.4333 39.6977 1.742 2001-12-21 20:17:45.04 9.4305 39.6795 1.308 2001-12-21 22:58:41.26 9.4212 39.6882 0.925 2001-12-22 00:57:20.10 9.4678 39.7045 2.376 2001-12-22 01:10:11.14 9.4232 39.6897 1.282 2001-12-22 01:13:03.90 9.4157 39.6877 1.318 2001-12-22 06:46:45.03 9.4153 39.6892 2.070 2001-12-22 06:47:32.38 9.4162 39.6912 2.123 2001-12-22 16:15:33.02 9.4548 39.6510 1.721 2001-12-22 22:07:06.27 9.4607 39.6838 0.909 2001-12-22 23:18:23.15 9.4788 39.6607 0.993 2001-12-23 01:27:43.55 9.2800 40.0893 1.008 2001-12-23 01:45:35.38 9.6297 39.7242 1.419 2001-12-23 02:46:21.63 9.1732 39.9983 1.165 2001-12-23 10:27:27.80 9.0323 39.9222 1.158 2001-12-23 19:25:08.21 9.4218 39.6925 1.377 2001-12-23 22:33:35.60 7.7465 38.7202 1.822 2001-12-23 22:52:56.83 9.0333 39.9713 1.050 2001-12-23 23:42:29.60 7.4982 38.6402 1.447 2001-12-24 01:03:16.26 8.6865 39.9617 0.635 2001-12-24 01:30:11.11 9.9215 40.0057 2.275 2001-12-24 18:56:38.22 9.5480 39.6675 1.766 2001-12-24 23:26:20.37 8.5683 40.0165 0.915 2001-12-25 03:52:51.85 9.6485 39.7138 2.771 2001-12-25 16:41:18.12 9.3365 39.9012 2.044 2001-12-25 22:20:11.90 10.1365 40.4205 2.089 2001-12-26 00:19:54.40 7.6145 38.7655 2.835 2001-12-26 00:20:36.52 7.6547 38.7727 2.801 2001-12-26 00:35:42.80 7.6228 38.8767 1.624 2001-12-26 01:42:18.33 8.0277 38.9963 1.058 2001-12-27 05:35:37.59 7.7667 38.7497 2.792 2001-12-27 08:48:39.55 7.7597 38.7615 2.887 2001-12-27 16:44:41.18 7.7405 38.7290 2.281 2001-12-27 17:59:01.69 9.7210 41.4242 1.977 2001-12-27 18:31:36.91 9.6850 41.4905 2.054 2001-12-28 02:55:17.13 9.4372 39.6935 1.332 2001-12-28 03:00:55.28 9.4598 39.6907 1.517 2001-12-28 15:45:31.66 9.4643 39.6665 1.581 2001-12-28 16:39:23.13 9.4287 39.6820 1.928 2001-12-28 17:01:52.01 9.4305 39.6715 1.271 2001-12-28 21:17:51.02 9.4422 39.6723 1.728 2001-12-28 21:34:50.49 9.4593 39.7062 1.807 2001-12-28 21:37:39.89 9.4630 39.7010 2.340 2001-12-28 21:47:09.60 9.4510 39.6935 1.823 2001-12-28 21:49:14.88 9.4438 39.6987 1.701 2001-12-28 23:39:11.79 9.4712 39.7032 1.006 2001-12-28 23:43:27.46 9.4385 39.6913 1.144 2001-12-28 23:54:06.42 9.4595 39.6847 1.591 2001-12-29 00:15:10.83 9.4485 39.6857 0.830 2001-12-29 01:22:00.71 9.4567 39.6965 2.397 2001-12-29 01:57:51.91 7.6223 38.8315 1.297 2001-12-29 02:39:38.55 9.4533 39.6933 1.569 2001-12-29 03:07:31.28 9.4402 39.6917 1.270 2001-12-29 03:48:58.82 9.4518 39.6993 2.118 2001-12-29 09:53:04.11 9.4522 39.7002 2.058 2001-12-29 13:49:38.37 9.4972 40.0268 2.370 2001-12-30 01:13:45.20 9.2115 40.1253 1.333 2001-12-30 06:16:15.66 9.4682 39.6945 2.988 2001-12-30 18:12:12.37 9.4430 39.6903 1.867 2001-12-30 18:59:57.10 9.4360 39.6953 2.381 2001-12-30 19:05:27.55 9.4478 39.6887 1.544 2001-12-30 22:54:03.97 9.3327 40.1780 1.181 2001-12-30 23:34:20.89 9.2257 40.0810 0.921 2001-12-31 00:07:06.10 9.4790 39.6957 1.725 2001-12-31 01:58:16.45 9.4550 39.7000 2.115 2001-12-31 02:17:11.60 9.4382 39.6985 2.007 2001-12-31 06:07:49.92 9.4387 39.6985 1.923

Appendices

115

2001-12-31 06:09:03.39 9.4332 39.6950 2.127 2001-12-31 06:14:33.31 9.4320 39.6703 1.661 2001-12-31 07:08:37.05 9.4410 39.6608 1.452 2001-12-31 07:18:53.87 7.5970 38.8682 2.454 2001-12-31 08:41:07.44 9.4277 39.6848 1.946 2001-12-31 11:53:09.63 9.4242 39.6767 1.911 2001-12-31 13:32:23.21 9.4333 39.6942 2.008 2001-12-31 20:29:02.74 9.4553 39.6727 1.134 2001-12-31 20:30:02.24 9.4550 39.6978 1.299 2001-12-31 20:36:55.48 9.4542 39.6803 1.066 2001-12-31 22:33:48.76 9.4308 39.7002 1.730 2001-12-31 23:50:46.32 9.4438 39.6982 1.367 2002-01-01 00:20:54.48 9.4445 39.6813 1.256 2002-01-01 00:42:36.84 9.4062 39.6653 1.690 2002-01-01 01:41:09.61 8.2842 39.1502 1.232 2002-01-01 09:56:11.75 9.5590 38.4038 1.635 2002-01-01 23:24:50.47 7.5950 38.7683 1.620 2002-01-02 02:02:12.54 9.2458 40.4685 1.448 2002-01-02 06:24:46.14 9.0787 40.3257 2.016 2002-01-02 10:41:43.07 8.9328 38.3667 1.518 2002-01-02 15:09:58.91 9.0723 40.4527 2.393 2002-01-02 17:40:26.01 10.1590 39.5255 2.240 2002-01-02 21:04:00.34 7.7440 38.7622 1.924 2002-01-02 21:11:19.34 7.7480 38.7345 1.902 2002-01-02 22:42:07.41 7.7135 38.7170 1.993 2002-01-03 03:25:37.74 7.5828 38.7518 1.966 2002-01-03 03:52:29.14 9.4430 39.6915 2.455 2002-01-03 04:03:44.55 9.4492 39.7145 1.559 2002-01-03 10:34:09.84 8.6637 39.5432 2.664 2002-01-03 14:57:06.19 8.6783 39.5753 2.801 2002-01-03 16:05:21.04 9.8263 41.1220 2.477 2002-01-04 03:27:17.44 8.2780 39.0987 1.782 2002-01-04 16:50:29.20 8.2588 39.0927 1.982 2002-01-05 21:06:21.67 9.8955 41.1377 2.364 2002-01-06 19:45:56.29 9.4393 39.6785 1.858 2002-01-07 19:21:47.82 11.2768 39.7498 2.897 2002-01-07 19:54:50.78 9.4325 39.6873 1.612 2002-01-07 21:47:16.17 9.4400 39.6987 1.396 2002-01-07 22:17:23.11 8.6843 39.5807 1.746 2002-01-08 00:09:41.28 11.4885 39.8000 2.819 2002-01-09 22:18:41.09 9.7625 39.7952 2.434 2002-01-11 15:26:28.60 9.8495 40.5462 2.540 2002-01-11 21:37:46.35 9.0300 40.7542 1.173 2002-01-12 16:00:09.91 9.4693 39.7083 1.790 2002-01-12 20:14:48.59 7.2812 38.5368 2.063 2002-01-12 22:31:14.00 10.0387 41.1908 1.865 2002-01-12 23:17:57.51 9.3763 39.6485 1.082 2002-01-12 23:32:12.29 9.4528 40.3417 1.673 2002-01-13 00:05:19.33 9.4562 39.6938 1.917 2002-01-13 00:52:53.56 9.4480 39.6977 1.322 2002-01-13 01:19:19.78 9.4320 39.9713 2.233 2002-01-13 02:07:30.99 9.4077 39.6962 1.225 2002-01-13 23:21:59.81 9.4672 39.6870 1.712 2002-01-14 00:30:38.87 9.4523 39.6945 1.163 2002-01-14 00:36:26.34 9.4538 39.6840 1.065 2002-01-14 07:56:46.58 9.4580 39.6937 2.451 2002-01-14 08:32:12.31 8.9185 40.5888 2.605 2002-01-14 08:42:35.24 8.9140 40.5853 1.996 2002-01-14 17:48:27.74 9.4417 39.6787 2.205 2002-01-14 22:14:08.40 9.4728 39.6995 1.944 2002-01-14 22:16:14.60 9.4740 39.7012 1.458 2002-01-14 22:45:57.50 9.6117 39.5225 2.303 2002-01-14 23:31:31.48 9.4447 39.7035 1.427 2002-01-14 23:34:24.02 9.4548 39.6965 2.212 2002-01-14 23:39:45.10 9.4517 39.6978 1.381 2002-01-14 23:52:09.54 9.4455 39.7037 1.935 2002-01-15 00:22:32.07 9.4705 39.6728 1.212 2002-01-15 00:55:46.93 9.4658 39.6917 2.188 2002-01-15 01:02:40.37 9.4677 39.6855 1.374 2002-01-15 01:14:51.90 9.4722 39.6810 1.393 2002-01-15 03:11:11.83 7.3717 38.6417 1.435 2002-01-15 04:17:22.92 9.4533 39.7033 1.557 2002-01-15 12:37:47.93 9.4460 39.6778 2.086 2002-01-15 21:40:55.86 9.4690 39.6940 1.803 2002-01-16 20:07:48.26 9.4712 39.7017 1.658 2002-01-16 21:22:39.08 9.2332 40.0438 1.698 2002-01-17 01:13:18.82 10.0063 39.9905 2.195 2002-01-17 01:38:04.08 8.1490 39.0142 2.014 2002-01-17 02:43:49.60 9.4332 39.7065 1.624 2002-01-17 03:15:25.53 8.1443 39.0308 2.252 2002-01-17 14:18:13.63 9.4442 39.7062 2.842 2002-01-17 14:20:36.14 9.3843 39.7060 2.167 2002-01-17 14:26:59.16 9.4413 39.6962 2.857 2002-01-17 17:48:20.41 9.4607 39.7045 2.823 2002-01-17 18:35:03.51 9.4493 39.7087 1.492 2002-01-17 19:16:31.08 9.4612 39.6943 2.331 2002-01-17 20:01:05.21 7.4538 38.1222 2.150 2002-01-17 20:07:35.13 9.4762 39.6935 2.398 2002-01-17 20:24:21.07 9.4742 39.6970 2.940 2002-01-17 20:55:55.12 9.4975 39.6565 1.222 2002-01-17 21:30:32.53 9.4607 39.6792 0.992 2002-01-17 22:33:01.48 9.4777 39.6902 1.354 2002-01-17 22:43:58.93 9.4588 39.6982 2.391 2002-01-17 23:00:27.38 9.4512 39.6870 1.578

Appendices

116

2002-01-17 23:16:46.89 9.4472 39.6830 1.303 2002-01-17 23:34:57.53 9.4492 39.6962 1.593 2002-01-17 23:45:55.79 9.4272 39.6635 1.141 2002-01-17 23:53:08.22 9.4800 39.6885 2.045 2002-01-18 00:42:28.03 9.4582 39.6887 1.243 2002-01-18 01:32:50.94 9.4813 39.6780 1.380 2002-01-18 01:42:40.62 8.9935 39.9450 2.824 2002-01-18 01:50:10.15 8.9895 39.9265 0.684 2002-01-18 01:58:11.64 8.9905 39.9232 0.955 2002-01-18 15:43:33.87 7.7647 38.6082 2.894 2002-01-18 21:46:19.95 9.4888 39.6547 1.147 2002-01-19 00:51:10.24 9.4822 39.6895 1.547 2002-01-19 01:02:19.47 9.4648 39.6790 1.043 2002-01-19 06:11:11.58 8.6803 39.5810 1.986 2002-01-19 19:08:34.20 9.5312 39.6923 1.695 2002-01-19 22:21:35.12 11.5325 40.1017 2.702 2002-01-20 02:55:52.29 8.5357 39.3222 1.468 2002-01-20 03:41:08.04 8.9755 39.8607 1.329 2002-01-20 05:06:21.42 9.4093 39.4323 1.774 2002-01-20 22:11:05.26 9.4747 39.6903 2.441 2002-01-20 22:23:31.09 9.4665 39.6812 1.193 2002-01-20 23:34:01.38 7.1230 38.6868 2.175 2002-01-21 12:36:52.97 9.3637 39.6052 2.400 2002-01-21 12:46:50.33 9.0517 40.4282 2.377 2002-01-23 17:36:13.86 9.6663 39.2162 2.152 2002-01-26 21:55:09.62 8.1543 39.0190 1.837 2002-01-27 00:45:36.98 10.2655 40.4722 1.872 2002-01-27 21:20:01.52 9.4430 39.6777 1.396 2002-01-28 03:06:25.75 9.4185 40.7413 1.521 2002-01-28 18:50:13.54 9.4602 39.2795 1.907 2002-01-29 11:21:02.43 7.9335 38.8392 2.242 2002-01-29 13:20:58.10 10.0005 39.8052 2.126 2002-01-30 09:08:07.06 8.9473 38.6045 1.099 2002-01-30 20:13:47.88 6.9048 38.5472 2.115 2002-01-30 21:40:43.00 9.2773 39.9692 1.572 2002-01-30 21:52:05.23 9.2962 40.0013 1.895 2002-01-31 00:04:41.95 9.2882 39.9918 1.417 2002-01-31 00:06:31.88 9.2743 39.9808 1.557 2002-01-31 02:05:55.25 9.4555 39.6967 1.853 2002-01-31 02:15:37.16 9.2953 40.0000 1.712 2002-01-31 02:25:44.71 9.2633 40.0082 1.485 2002-01-31 02:50:59.82 8.9058 39.9340 1.186 2002-01-31 13:09:31.31 9.2328 40.0158 1.753 2002-01-31 15:33:37.65 9.2807 39.9913 1.973 2002-01-31 16:35:56.78 9.2410 39.9720 1.579 2002-01-31 16:44:20.94 9.2743 40.0020 2.207 2002-01-31 19:25:21.68 9.2983 40.0003 1.503 2002-01-31 20:41:45.16 9.4347 39.6932 1.475 2002-01-31 23:54:41.15 9.2763 39.9842 1.751 2002-02-01 08:51:18.22 8.9685 38.8252 2.256 2002-02-01 09:33:22.45 9.0095 38.3762 1.303 2002-02-01 11:05:14.17 8.9238 38.6088 1.368 2002-02-01 19:49:47.74 9.5140 41.8380 2.376 2002-02-01 23:38:11.39 10.3018 39.8520 1.587 2002-02-03 14:06:47.79 9.7732 39.7343 2.118 2002-02-03 22:07:44.61 9.4730 39.5378 1.277 2002-02-04 01:43:12.55 9.1093 40.0150 1.118 2002-02-04 17:56:45.19 9.2143 41.0450 2.482 2002-02-05 00:14:27.85 9.4230 40.6098 1.288 2002-02-05 07:23:42.49 8.8273 39.7952 1.569 2002-02-06 01:10:48.60 7.9695 38.1205 1.628 2002-02-06 18:33:40.41 9.6870 41.3592 2.288 2002-02-07 06:34:38.90 9.4303 39.6697 1.927 2002-02-07 18:44:39.90 10.4383 39.6997 4.333 2002-02-08 01:53:01.59 9.0630 39.9840 1.537 2002-02-08 08:42:20.30 9.1882 38.2562 1.396 2002-02-09 16:52:03.81 9.4413 39.6937 2.170 2002-02-09 23:34:27.89 9.4288 39.7035 1.253 2002-02-10 09:41:34.73 9.0230 38.7002 1.124 2002-02-10 14:18:01.10 10.0530 39.2965 2.191 2002-02-10 14:33:09.95 9.4358 39.7175 1.765 2002-02-10 16:13:20.46 9.4573 39.6858 1.471 2002-02-11 01:47:16.01 9.4472 39.6800 1.429 2002-02-11 01:51:54.07 9.4448 39.6950 2.202 2002-02-11 01:57:19.76 9.4667 39.6808 1.742 2002-02-11 02:54:01.35 9.4327 39.6830 1.376 2002-02-11 04:33:28.06 9.4112 39.6115 1.817 2002-02-11 05:09:10.83 9.4063 39.6962 1.958 2002-02-11 10:16:52.54 9.4157 39.6990 2.006 2002-02-11 17:09:07.70 9.4222 39.6777 2.040 2002-02-11 18:48:27.12 9.4193 39.6840 1.546 2002-02-11 20:19:43.02 9.4850 39.6500 1.043 2002-02-11 21:37:29.47 9.9950 41.6228 3.062 2002-02-11 21:42:32.10 9.4377 39.6758 1.407 2002-02-12 00:07:23.46 9.4503 39.6777 1.134 2002-02-12 00:41:35.62 9.4377 39.6840 1.418 2002-02-12 00:49:53.46 9.4485 39.6940 1.322 2002-02-12 00:50:39.69 9.4365 39.6890 1.205 2002-02-12 01:02:21.57 9.4525 39.6647 1.241 2002-02-12 12:54:06.03 10.3375 39.6272 2.460 2002-02-12 17:18:50.90 9.4322 39.6850 1.947 2002-02-12 21:49:15.81 9.4845 39.6102 1.533 2002-02-13 02:37:28.92 9.2117 39.6730 1.527 2002-02-13 09:11:20.39 9.2657 39.4118 2.107

Appendices

117

2002-02-14 00:04:10.52 7.9128 38.9257 0.870 2002-02-14 00:47:19.85 9.4547 39.7103 1.563 2002-02-14 01:18:35.02 9.2868 39.9953 1.619 2002-02-14 03:08:13.49 11.9815 39.7823 3.205 2002-02-14 22:05:45.72 9.1085 40.0308 1.420 2002-02-15 00:16:54.79 9.1137 40.0260 1.262 2002-02-15 01:11:32.90 8.1122 39.0437 1.001 2002-02-15 01:21:56.77 8.1208 39.0392 1.036 2002-02-15 03:01:53.06 9.4203 39.6782 1.755 2002-02-15 22:30:30.58 8.1062 39.0555 0.979 2002-02-16 22:58:26.00 9.4467 39.7113 1.368 2002-02-16 23:10:17.20 9.4645 39.7000 1.686 2002-02-17 02:38:15.46 9.4695 39.7025 3.211 2002-02-17 02:40:52.33 9.4552 39.7138 1.961 2002-02-17 02:42:18.03 9.4442 39.6920 2.701 2002-02-17 03:00:14.72 9.4753 39.7035 1.663 2002-02-17 03:13:41.41 9.4402 39.7063 1.651 2002-02-17 04:08:49.55 10.3293 39.9337 2.247 2002-02-17 10:38:19.25 7.4592 38.6795 2.516 2002-02-17 11:15:46.75 7.4837 38.7120 2.358 2002-02-17 13:26:22.44 9.4580 39.7070 2.782 2002-02-17 13:45:58.56 9.4663 39.7067 2.122 2002-02-17 14:16:13.46 9.4648 39.6965 2.441 2002-02-17 14:17:31.01 9.4693 39.7067 2.371 2002-02-17 14:16:49.66 9.4688 39.7083 2.010 2002-02-17 14:24:44.63 9.4318 39.6840 2.180 2002-02-17 14:24:18.84 9.4542 39.7067 2.297 2002-02-17 14:29:54.81 9.4698 39.6907 2.394 2002-02-17 15:52:10.27 9.4315 39.6557 1.788 2002-02-17 22:07:19.73 9.4312 39.7023 1.773 2002-02-17 23:33:50.00 9.4778 39.7105 1.645 2002-02-18 00:48:31.93 9.4673 39.6720 1.428 2002-02-18 21:25:28.13 9.4478 39.6810 1.214 2002-02-19 02:04:13.11 8.4500 39.2868 1.271 2002-02-19 03:26:19.03 8.4898 39.2925 1.235 2002-02-19 22:47:35.80 9.4685 39.6998 2.062 2002-02-19 22:57:10.41 8.5228 38.9408 1.261 2002-02-20 00:53:06.35 9.5237 39.6470 1.113 2002-02-20 01:24:10.76 9.2695 40.2053 0.986 2002-02-20 21:03:17.45 9.4702 39.7030 2.018 2002-02-20 21:27:35.02 9.7642 39.7742 1.493 2002-02-20 21:35:04.25 9.4640 39.6430 0.861 2002-02-21 08:21:58.33 7.4360 38.6103 3.212 2002-02-21 08:56:11.35 7.4242 38.6112 2.813 2002-02-21 09:02:06.65 9.2607 39.0002 0.455 2002-02-21 09:26:49.58 9.8492 40.9240 2.179 2002-02-21 12:48:59.11 7.3852 38.6820 2.763 2002-02-21 17:00:28.57 7.4247 38.5782 2.127 2002-02-21 19:20:31.00 7.3633 38.7118 2.031 2002-02-21 22:14:51.47 9.2058 40.0058 1.378 2002-02-21 22:25:25.41 7.4095 38.6078 2.102 2002-02-21 23:25:58.01 8.4237 40.1162 1.027 2002-02-21 23:53:12.07 7.4227 38.8078 1.459 2002-02-22 00:34:11.28 9.2502 39.9793 1.128 2002-02-22 01:55:22.81 8.9515 39.6470 0.974 2002-02-22 02:18:52.35 9.4740 39.7007 0.968 2002-02-22 07:27:22.29 7.3762 38.7157 2.751 2002-02-22 07:30:59.54 7.4137 38.6645 2.693 2002-02-22 09:16:27.14 7.3802 38.7143 2.369 2002-02-22 13:45:32.01 9.4395 39.6910 1.729 2002-02-22 15:01:46.59 9.4363 39.6928 1.626 2002-02-23 02:03:03.94 7.5080 38.6628 2.125 2002-02-23 06:49:13.02 9.4492 39.7588 1.838 2002-02-23 09:07:59.20 9.4773 40.0375 2.027 2002-02-23 15:19:13.07 7.3978 38.6040 2.927 2002-02-23 16:07:08.25 9.1183 39.9508 1.467 2002-02-24 07:12:39.45 7.3275 38.7295 2.285 2002-02-24 10:18:32.27 8.9805 38.7012 1.043 2002-02-24 20:35:42.13 9.4867 38.6248 1.597 2002-02-24 21:35:17.09 9.2685 40.1835 0.797 2002-02-25 03:36:05.78 9.7617 41.4870 2.170 2002-02-25 20:45:34.85 9.4607 39.6968 2.175 2002-02-25 23:57:06.58 9.4645 39.6953 1.228 2002-02-26 00:24:27.18 9.4763 39.7023 1.388 2002-02-26 03:55:15.66 9.4317 39.6507 1.521 2002-02-26 11:09:17.27 8.8888 38.6025 1.480 2002-02-26 23:44:27.03 10.4903 39.5752 1.980 2002-02-27 08:18:06.57 8.2465 39.0702 1.907 2002-02-28 22:32:00.86 9.3965 39.6715 1.064 2002-03-01 12:09:02.95 9.4390 40.0018 1.755 2002-03-01 18:00:07.05 9.4627 39.6950 3.458 2002-03-01 18:01:41.44 9.4687 39.7100 3.890 2002-03-01 19:14:56.05 9.4357 39.6855 2.119 2002-03-01 19:17:16.10 9.4077 39.6150 1.086 2002-03-01 20:27:49.39 9.4493 39.6952 1.924 2002-03-01 20:29:52.51 9.4362 39.6945 1.742 2002-03-01 20:45:17.04 9.4355 39.6763 1.131 2002-03-01 20:56:19.95 9.4557 39.6848 2.877 2002-03-01 20:56:31.90 9.1563 39.7223 2.675 2002-03-01 20:57:47.25 9.4718 39.6870 3.812 2002-03-01 20:59:28.49 9.4093 39.6910 2.912 2002-03-01 21:04:06.96 9.3732 39.6492 1.392 2002-03-01 21:05:00.42 9.4648 39.6578 1.586 2002-03-01 21:06:17.45 9.4473 39.6673 1.968

Appendices

118

2002-03-01 21:07:05.69 9.4452 39.7013 1.975 2002-03-01 21:09:02.02 9.4352 39.6935 2.353 2002-03-01 21:09:07.90 9.4515 39.7040 2.416 2002-03-01 21:13:22.03 9.4463 39.6663 1.098 2002-03-01 21:16:20.05 9.4388 39.6890 2.232 2002-03-01 21:30:08.14 9.4507 39.6952 2.422 2002-03-01 21:35:09.68 9.4710 39.7028 1.263 2002-03-01 22:16:44.13 9.4427 39.6993 1.477 2002-03-01 22:27:25.29 9.5970 39.5597 0.792 2002-03-01 22:54:48.88 9.4795 39.6728 1.559 2002-03-01 23:07:36.13 9.4007 39.6987 1.277 2002-03-01 23:29:10.83 8.9325 40.5678 1.094 2002-03-01 23:37:46.15 9.4742 39.6885 2.931 2002-03-01 23:40:12.44 9.4577 39.6875 2.067 2002-03-01 23:44:16.39 9.4333 39.6683 1.108 2002-03-01 23:49:57.42 9.4542 39.6995 1.647 2002-03-01 23:55:38.37 9.4058 39.6952 0.992 2002-03-02 00:15:17.62 9.4552 39.7017 2.328 2002-03-02 00:17:26.31 9.5648 39.4625 0.943 2002-03-02 00:23:03.48 9.4323 39.6890 1.142 2002-03-02 00:37:04.70 9.4385 39.6467 0.851 2002-03-02 01:15:41.46 9.4618 39.6997 1.685 2002-03-02 02:17:03.20 9.4247 39.6575 1.157 2002-03-02 03:17:08.36 9.4223 39.6648 1.408 2002-03-02 04:12:24.99 9.4517 39.6925 2.470 2002-03-02 04:31:34.24 9.4020 39.6963 2.005 2002-03-02 05:36:46.69 8.4733 40.0897 1.756 2002-03-02 11:45:55.37 9.4295 39.6697 1.706 2002-03-02 14:00:31.56 9.4288 39.6933 1.764 2002-03-02 20:04:08.02 9.4120 39.6027 0.977 2002-03-02 21:35:39.70 9.4103 39.6138 0.834 2002-03-02 21:48:38.22 9.4067 39.5997 1.153 2002-03-02 22:09:59.52 9.2590 39.6133 0.994 2002-03-02 23:00:41.26 9.3780 39.6518 1.046 2002-03-02 23:22:35.93 9.4263 39.7005 1.358 2002-03-03 00:31:12.89 9.4135 39.6598 0.976 2002-03-03 00:32:13.26 9.4375 39.6778 1.542 2002-03-03 03:25:05.79 9.2585 39.5532 1.149 2002-03-03 14:07:56.25 9.4400 39.6967 1.800 2002-03-03 14:21:47.89 9.4237 39.6800 2.071 2002-03-03 14:30:13.76 9.4493 39.6847 1.790 2002-03-03 15:13:54.01 9.9737 40.4965 2.504 2002-03-03 19:53:11.70 9.4543 39.7058 1.395 2002-03-03 19:54:04.70 9.5085 39.3478 0.613 2002-03-03 21:52:08.47 9.3737 39.6440 0.874 2002-03-03 21:52:58.40 9.5972 39.7297 0.931 2002-03-04 01:50:23.17 9.3392 40.1290 1.020 2002-03-04 02:58:51.85 9.3913 39.6373 1.451 2002-03-04 03:10:19.01 9.3913 39.6485 1.289 2002-03-04 07:05:37.96 9.4435 39.7105 2.032 2002-03-04 14:59:00.54 9.4008 39.6912 1.710 2002-03-04 15:13:21.30 9.3523 39.6063 1.448 2002-03-04 16:18:33.33 9.2962 40.1848 1.678 2002-03-04 16:25:56.84 9.2858 40.1768 2.360 2002-03-04 16:29:15.11 9.2797 40.1900 1.784 2002-03-04 16:29:29.21 9.3035 40.1817 1.844 2002-03-04 16:49:34.67 9.2792 40.1685 1.682 2002-03-04 18:12:01.18 9.3008 40.1798 1.296 2002-03-04 18:32:10.30 9.2955 40.1907 1.285 2002-03-04 21:17:09.61 9.2828 40.1995 1.679 2002-03-04 22:24:40.67 9.2877 40.1820 1.106 2002-03-04 23:08:53.95 9.4457 39.6900 1.844 2002-03-04 23:54:39.15 9.1913 40.0217 1.230 2002-03-04 23:57:55.59 9.4235 39.6772 1.300 2002-03-05 00:06:53.78 9.3145 40.1752 1.281 2002-03-05 00:16:12.58 9.3332 40.1768 0.970 2002-03-05 00:37:41.78 9.4903 39.6328 1.110 2002-03-05 00:45:57.24 9.2947 40.1830 1.133 2002-03-05 00:57:59.88 9.2907 40.1912 0.763 2002-03-05 01:35:27.91 9.2490 40.2125 0.826 2002-03-05 01:57:47.58 9.2935 39.5722 1.040 2002-03-05 02:09:20.28 9.3060 40.1715 0.951 2002-03-05 17:13:25.10 9.3435 39.6158 1.281 2002-03-05 18:02:50.39 9.4443 39.6953 1.437 2002-03-06 02:48:40.80 7.4017 38.8447 1.885 2002-03-06 03:45:10.51 9.3958 39.7050 1.759 2002-03-06 16:56:26.53 9.1933 39.9540 1.570 2002-03-06 18:42:50.89 9.1980 39.9310 1.882 2002-03-06 18:56:19.71 9.1798 39.9780 1.529 2002-03-06 19:09:08.70 9.2913 40.1713 1.384 2002-03-06 20:16:54.92 9.1970 39.9478 1.547 2002-03-07 01:07:39.43 9.1970 39.9247 1.425 2002-03-07 10:46:17.40 9.3053 40.1820 2.221 2002-03-07 11:26:08.38 9.4352 39.6915 1.809 2002-03-07 12:03:09.34 9.4427 39.6893 2.133 2002-03-07 12:22:06.34 9.4248 39.6845 1.727 2002-03-07 21:57:13.51 9.4570 39.7093 1.379 2002-03-07 22:44:33.64 9.4482 39.7088 1.778 2002-03-07 23:27:50.43 9.3142 40.1840 1.342 2002-03-08 00:17:14.42 9.4082 39.6552 1.208 2002-03-08 02:00:17.70 9.4300 39.6627 1.198 2002-03-08 06:04:02.59 9.4403 39.6948 2.459 2002-03-08 07:11:45.76 9.4002 39.6345 1.549 2002-03-08 15:06:35.39 9.4483 39.7085 1.858

Appendices

119

2002-03-08 15:28:09.39 9.3678 39.6248 1.431 2002-03-08 18:22:54.09 9.4617 39.6985 2.395 2002-03-08 19:24:24.01 9.4060 39.6632 1.468 2002-03-08 20:45:14.83 9.4397 39.6948 2.099 2002-03-08 23:44:01.74 9.2655 39.3933 1.344 2002-03-09 01:11:04.82 9.4582 39.7087 1.676 2002-03-09 07:50:49.44 9.1462 39.9635 2.259 2002-03-09 09:43:05.15 9.3988 39.6602 2.003 2002-03-09 10:19:22.85 9.4323 39.6825 2.594 2002-03-09 10:19:29.37 9.4998 39.6432 2.586 2002-03-09 10:21:10.35 9.4468 39.6893 2.487 2002-03-09 10:37:13.69 9.5780 39.4722 1.510 2002-03-09 10:38:06.40 9.5042 39.3352 1.317 2002-03-09 10:58:38.68 9.4162 39.7038 2.086 2002-03-09 12:22:50.99 9.4422 39.6627 2.185 2002-03-09 17:50:14.42 9.2450 40.1262 1.325 2002-03-09 23:02:47.73 9.4702 39.6573 0.968 2002-03-09 23:10:30.19 9.4465 39.6868 1.546 2002-03-09 23:46:18.26 9.4493 39.6672 1.478 2002-03-09 23:47:51.63 9.4365 39.6822 0.913 2002-03-09 23:48:57.83 9.4392 39.6612 1.558 2002-03-10 00:01:59.42 9.4088 39.6555 0.958 2002-03-10 00:17:59.77 9.4393 39.6840 1.550 2002-03-10 00:18:36.67 9.4252 39.6758 1.354 2002-03-10 00:19:59.80 9.3887 39.6393 0.982 2002-03-10 01:01:11.02 9.8770 39.3175 2.797 2002-03-10 01:07:04.41 9.8183 39.3055 2.002 2002-03-10 08:41:13.51 9.1497 39.9653 2.029 2002-03-10 20:30:03.92 9.4667 39.6560 0.751 2002-03-10 23:49:20.84 9.4517 39.6878 1.570 2002-03-10 23:51:37.24 9.1087 39.9935 1.470 2002-03-11 00:01:56.34 9.4112 39.6033 0.770 2002-03-11 00:13:00.62 9.1242 39.9938 1.093 2002-03-11 21:54:53.60 9.4422 39.6890 1.682 2002-03-11 22:07:08.94 9.4088 39.6757 1.273 2002-03-11 22:49:46.90 9.4062 39.6718 1.723 2002-03-12 00:12:51.50 9.4502 39.6997 1.788 2002-03-12 01:14:59.22 9.4225 39.6820 1.280 2002-03-12 08:44:10.34 9.4385 39.7000 2.693 2002-03-12 22:28:34.59 9.3985 39.6267 1.169 2002-03-12 22:41:57.33 9.6967 39.7307 1.740 2002-03-12 23:02:52.24 9.7088 39.6962 1.182 2002-03-13 03:32:51.13 9.4515 39.7063 1.697 2002-03-13 10:37:12.24 9.1080 40.0085 1.993 2002-03-13 22:52:11.40 9.5900 39.7293 0.976 2002-03-14 00:48:02.53 8.9938 39.9153 1.542 2002-03-14 12:43:22.02 9.5175 39.6197 1.496 2002-03-14 13:25:05.20 9.4145 39.2953 1.853 2002-03-14 18:16:37.36 9.4562 39.6757 1.400 2002-03-15 01:34:39.83 9.4573 39.6562 1.113 2002-03-15 01:49:39.65 8.9952 39.9283 1.423 2002-03-15 23:22:39.29 11.1723 39.7282 2.721 2002-03-16 19:43:07.21 8.0963 39.0490 1.296 2002-03-16 20:15:44.93 7.5035 38.8523 1.812 2002-03-16 21:02:23.39 9.1213 39.9862 0.830 2002-03-16 21:52:52.51 9.4687 39.2903 1.750 2002-03-17 13:51:17.71 7.8397 38.6665 2.254 2002-03-17 19:24:17.13 8.9190 39.7287 0.818 2002-03-18 01:36:32.79 9.4867 40.0138 1.686 2002-03-18 06:50:00.25 9.4803 39.7062 2.392 2002-03-18 19:18:57.02 10.0648 39.8828 1.608 2002-03-19 10:04:09.56 8.4425 39.2363 1.787 2002-03-19 23:12:43.95 9.8043 39.1758 1.718 2002-03-20 08:06:34.93 9.4542 39.6985 1.848 2002-03-20 23:13:43.42 8.9548 39.9292 0.648 2002-03-21 01:41:49.72 9.4483 39.7012 1.847 2002-03-21 01:51:06.73 9.4777 39.6997 2.933 2002-03-21 02:26:39.09 9.4588 39.6825 2.148 2002-03-21 02:39:53.83 9.3922 39.6440 1.467 2002-03-21 05:04:47.78 9.4518 39.6928 2.129 2002-03-21 05:31:39.74 9.4548 39.6977 2.862 2002-03-21 05:36:23.38 9.4557 39.7088 2.288 2002-03-21 09:38:42.75 9.4440 39.6523 1.526 2002-03-21 09:53:39.17 9.4617 39.6962 2.415 2002-03-21 11:38:40.13 9.4627 39.7083 1.813 2002-03-21 19:30:30.90 9.4555 39.7018 1.632 2002-03-22 08:55:48.91 9.4222 39.6535 1.533 2002-03-22 09:51:59.36 9.3602 39.3112 1.183 2002-03-22 12:47:00.27 9.4323 39.6798 2.994 2002-03-22 12:48:28.32 9.3363 39.6100 2.061 2002-03-23 16:44:04.43 9.4308 39.6900 1.312 2002-03-24 08:34:16.83 10.7410 39.7432 2.573 2002-03-24 16:33:43.38 10.3308 40.5295 2.595 2002-03-24 17:14:31.87 9.4473 39.6855 1.447 2002-03-24 19:50:50.68 9.3968 39.6157 0.755 2002-03-24 20:42:29.63 9.8342 39.2988 2.265 2002-03-24 21:07:17.29 9.2898 40.1785 2.032 2002-03-24 21:16:57.33 9.2747 40.2172 0.673 2002-03-24 21:29:28.42 9.3050 40.1630 1.247 2002-03-24 21:44:46.28 10.1327 39.8230 1.650 2002-03-24 23:07:31.40 9.2757 40.2197 0.643 2002-03-25 12:55:39.40 9.3230 40.1820 1.524 2002-03-25 20:47:09.76 9.4092 39.6385 0.910 2002-03-25 21:48:12.86 9.2560 40.1995 1.129

Appendices

120

2002-03-26 01:30:31.32 9.2952 40.1975 0.809 2002-03-28 09:03:35.94 9.4943 40.0270 2.515 2002-03-28 16:53:11.90 9.3890 39.6403 1.360 2002-03-28 20:11:43.88 8.1773 39.1607 1.576 2002-03-28 23:02:44.61 9.6170 39.4225 0.629 2002-03-29 21:09:56.18 9.4178 39.6852 1.459 2002-03-29 22:22:11.41 9.3568 39.6428 1.333 2002-03-30 05:46:24.96 9.7768 39.7497 2.311 2002-03-30 10:21:16.52 8.8247 38.8292 1.107 2002-04-01 18:02:32.01 9.6822 39.7665 2.038 2002-04-02 01:15:00.19 9.4282 39.6860 1.221 2002-04-02 20:22:41.91 10.0387 41.6420 2.736 2002-04-03 06:09:58.29 9.3802 39.6583 1.353 2002-04-04 00:37:11.69 9.1335 39.9757 1.194 2002-04-04 20:55:38.64 9.4358 39.6958 1.373 2002-04-04 22:06:42.43 9.3560 39.6517 1.094 2002-04-05 08:49:01.39 9.4105 39.6570 1.715 2002-04-05 08:50:21.79 9.4092 39.6518 1.811 2002-04-06 14:31:00.83 9.4015 39.6568 1.653 2002-04-06 19:39:10.82 9.4143 39.6457 1.107 2002-04-07 00:19:44.07 9.4187 39.6555 0.880 2002-04-07 03:19:00.88 9.4283 39.6948 1.600 2002-04-07 12:58:17.98 9.4837 39.6915 1.656 2002-04-07 19:08:02.85 9.4045 39.6640 1.241 2002-04-07 20:38:41.82 9.5592 39.6602 1.208 2002-04-08 01:41:37.26 9.5078 39.7343 1.114 2002-04-09 04:56:39.87 9.4193 39.6752 1.750 2002-04-09 07:28:53.58 8.9900 39.8840 2.147 2002-04-09 12:25:56.21 9.4460 39.6967 2.623 2002-04-09 16:41:17.29 9.4243 39.6817 1.476 2002-04-09 17:17:21.16 9.3735 39.6237 1.345 2002-04-09 21:25:30.71 8.9868 39.8667 0.689 2002-04-09 21:28:03.83 8.9915 39.8557 0.943 2002-04-09 23:52:01.67 9.4750 40.0058 1.099 2002-04-10 22:53:20.58 9.4140 39.6330 1.144 2002-04-11 04:50:24.55 8.9463 39.2513 2.815 2002-04-11 22:44:40.74 9.2338 40.1747 1.006 2002-04-12 09:16:46.60 9.4160 39.6875 1.942 2002-04-12 10:31:47.75 9.4333 39.7032 2.088 2002-04-12 18:06:33.04 8.9747 39.1418 1.327 2002-04-12 22:41:48.25 9.2532 39.5962 1.073 2002-04-13 00:23:46.14 9.4462 39.6810 1.268 2002-04-13 01:25:56.81 9.4257 39.7035 1.393 2002-04-13 01:30:30.21 9.4027 39.7087 1.219 2002-04-13 14:28:33.30 9.4600 39.6582 2.474 2002-04-13 14:42:00.28 9.5245 39.3705 1.213 2002-04-13 18:42:53.70 9.3592 39.6522 1.487 2002-04-13 20:31:27.04 9.3913 39.6270 0.891 2002-04-13 21:39:44.69 9.4890 39.7425 1.934 2002-04-13 22:57:43.67 9.1373 40.0100 1.252 2002-04-13 23:17:14.18 9.3570 39.3115 0.726 2002-04-13 23:17:26.63 9.4065 39.6688 1.356 2002-04-13 23:27:49.03 9.1132 40.0072 0.949 2002-04-14 00:20:11.96 9.4328 39.6840 0.946 2002-04-14 00:27:39.06 9.4387 39.6788 1.842 2002-04-14 00:31:01.25 9.1322 39.9950 1.340 2002-04-14 01:53:44.96 9.4375 39.6863 1.364 2002-04-14 21:13:36.48 7.7023 38.9710 2.021 2002-04-14 21:59:42.58 9.4713 38.5900 2.407 2002-04-14 22:56:12.06 9.1190 39.9995 0.975 2002-04-15 10:21:59.99 9.4513 39.6947 2.868 2002-04-15 11:01:35.23 9.4155 39.6693 2.145 2002-04-15 11:44:51.88 9.3732 39.6352 1.532 2002-04-15 11:46:08.12 9.4678 39.6858 1.478 2002-04-15 18:53:37.07 9.4200 39.7020 1.203 2002-04-15 20:11:40.75 9.2368 39.5427 0.993 2002-04-15 21:26:00.62 9.4833 39.3310 0.780 2002-04-15 22:41:04.66 9.1675 39.9553 0.967 2002-04-16 00:21:58.96 9.5905 39.7583 2.019 2002-04-16 00:24:03.27 9.5780 39.7605 1.996 2002-04-16 03:21:33.97 9.1660 39.9385 1.617 2002-04-16 08:26:49.93 9.4215 39.6752 1.634 2002-04-17 02:20:31.55 9.4322 39.6792 1.761 2002-04-17 02:42:36.46 9.3782 39.6583 1.345 2002-04-17 10:44:54.72 9.1013 40.0073 1.820 2002-04-17 12:16:31.84 9.1210 40.0073 1.601 2002-04-17 19:30:39.28 9.3033 40.1852 1.713 2002-04-17 20:31:49.77 9.5278 39.3753 0.522 2002-04-17 20:49:13.11 9.5243 39.3655 0.733 2002-04-17 21:41:09.31 9.4073 39.6063 0.843 2002-04-17 23:54:08.45 9.4920 40.0070 1.312 2002-04-18 00:31:30.66 9.2927 40.1678 1.041 2002-04-18 03:53:10.23 10.1995 40.7278 2.520 2002-04-18 10:04:31.63 7.3687 38.7040 2.562 2002-04-18 17:46:19.79 7.3878 38.6552 2.135 2002-04-18 17:58:43.92 9.4298 39.6920 1.951 2002-04-18 18:50:55.77 9.3862 39.6502 1.552 2002-04-18 20:12:18.83 9.4557 39.7037 1.712 2002-04-18 20:42:44.44 9.4452 39.6988 2.317 2002-04-18 21:22:19.31 9.4190 39.6788 1.468 2002-04-18 23:05:42.84 7.0463 38.1053 2.098 2002-04-19 00:06:00.94 9.2847 40.2027 0.916 2002-04-19 03:51:40.24 9.4298 39.6920 1.936 2002-04-19 04:08:59.66 9.4072 39.6948 2.463

Appendices

121

2002-04-19 04:10:32.41 9.4132 39.6895 2.143 2002-04-19 04:54:32.43 9.3595 39.3102 1.309 2002-04-19 05:14:16.95 9.3830 39.6407 2.053 2002-04-19 20:14:17.40 9.3997 39.6482 1.170 2002-04-19 22:34:31.94 9.4232 39.6285 1.115 2002-04-19 23:42:12.62 9.4435 39.6910 2.167 2002-04-19 23:43:23.64 9.4408 39.6695 1.349 2002-04-20 02:03:17.75 9.4415 39.6857 2.215 2002-04-20 02:05:01.25 9.4410 39.6885 1.515 2002-04-20 02:06:00.66 9.4222 39.6872 1.255 2002-04-20 05:19:42.09 9.3672 39.6288 1.617 2002-04-20 14:39:15.77 9.4327 39.6973 1.797 2002-04-20 14:59:22.82 9.3820 39.6400 1.897 2002-04-20 17:29:42.83 9.4083 39.6522 1.461 2002-04-20 21:29:09.98 9.6583 39.2825 1.663 2002-04-20 22:56:42.53 9.1352 39.9887 1.213 2002-04-20 22:57:26.91 9.1115 39.9967 1.102 2002-04-21 19:00:41.28 9.3860 39.6920 1.145 2002-04-22 01:48:53.51 9.1965 39.3870 0.903 2002-04-22 22:58:40.04 9.3638 39.6647 1.040 2002-04-23 00:15:30.83 9.4510 39.6863 2.404 2002-04-23 12:26:40.91 9.2972 40.2030 1.838 2002-04-23 18:01:09.69 9.3653 39.6792 1.142 2002-04-23 18:02:42.64 9.4067 39.6002 1.261 2002-04-23 18:48:15.07 9.4357 39.6885 1.309 2002-04-24 01:12:44.85 7.4642 38.6360 1.275 2002-04-24 23:24:05.36 7.3270 38.4633 2.722 2002-04-25 14:36:50.32 9.5443 40.3888 1.759 2002-04-26 01:10:04.25 7.7792 38.2758 1.529 2002-04-26 09:09:00.05 8.9145 38.8122 1.417 2002-04-26 17:41:52.40 9.2273 39.5350 1.082 2002-04-26 19:48:52.45 9.4358 39.9650 1.503 2002-04-27 01:30:38.50 9.2745 40.2215 0.931 2002-04-27 01:54:45.31 9.9387 39.9025 1.572 2002-04-27 09:57:24.66 9.3427 40.0652 1.618 2002-04-27 17:23:27.94 8.9770 39.8770 1.006 2002-04-27 17:30:23.00 9.0102 39.9197 1.344 2002-04-27 21:32:37.69 9.4192 39.7058 1.198 2002-04-27 23:24:10.00 8.9963 40.6563 0.850 2002-04-28 07:24:09.11 8.9018 39.8062 1.883 2002-04-28 07:32:35.15 9.1452 40.0050 2.616 2002-04-28 09:03:16.38 9.1353 40.0037 1.867 2002-04-29 01:12:32.90 9.4947 39.6285 1.285 2002-04-29 20:26:44.22 9.5272 39.3727 0.642 2002-04-29 20:55:13.15 9.2917 39.5780 1.183 2002-04-29 21:05:52.81 9.4215 39.6833 1.527 2002-04-29 22:13:00.65 8.9100 40.6650 1.052 2002-04-30 10:21:32.52 8.9127 38.6353 1.582 2002-05-01 15:57:43.74 9.9367 39.1345 3.021 2002-05-01 16:46:32.42 9.4293 39.6892 1.743 2002-05-01 20:34:30.75 9.4405 39.6845 1.702 2002-05-02 11:55:50.28 9.1157 39.9972 3.019 2002-05-02 21:43:23.05 9.1302 40.0003 2.643 2002-05-02 22:22:35.64 9.1217 39.9780 1.301 2002-05-03 22:23:04.88 9.4940 39.3420 0.604 2002-05-04 01:33:27.78 9.4655 40.0365 1.366 2002-05-04 06:15:45.08 7.6282 39.0287 2.265 2002-05-04 22:54:00.92 8.1187 39.0512 1.019 2002-05-05 01:55:34.42 9.0945 39.9822 1.114 2002-05-05 16:54:32.65 9.1492 39.9858 1.696 2002-05-06 20:18:09.68 9.0960 39.9610 1.154 2002-05-06 20:19:09.51 9.1295 39.9947 1.041 2002-05-06 23:44:44.46 9.4538 39.6777 1.072 2002-05-07 01:30:42.57 9.5503 39.6010 0.822 2002-05-07 19:22:39.92 9.2915 40.1727 1.320 2002-05-07 19:27:51.29 9.2760 40.1958 1.490 2002-05-07 22:34:17.19 9.4745 39.2795 1.827 2002-05-07 22:43:36.76 9.4547 39.7198 1.525 2002-05-07 22:43:37.91 9.4777 39.6710 1.501 2002-05-07 23:03:56.09 9.4477 39.6968 1.752 2002-05-07 23:19:04.14 9.3867 39.6435 0.846 2002-05-08 20:02:51.57 9.2807 40.1698 1.437 2002-05-09 01:42:43.13 9.2980 40.1743 0.729 2002-05-09 14:04:20.48 7.4872 38.9898 2.457 2002-05-09 21:08:34.99 9.3742 39.6375 0.932 2002-05-09 21:17:28.34 9.3828 39.6363 0.738 2002-05-09 21:18:31.80 9.4210 39.6807 0.852 2002-05-09 21:25:35.50 9.2950 40.1723 1.577 2002-05-09 21:26:35.12 9.3038 40.1772 1.856 2002-05-09 21:27:09.70 9.2935 40.1750 1.778 2002-05-09 22:11:25.62 9.4713 39.6405 1.732 2002-05-09 22:11:49.04 9.2777 40.1897 1.893 2002-05-09 22:19:38.80 9.2578 40.1420 1.273 2002-05-09 22:49:06.71 9.2957 40.1772 1.662 2002-05-09 23:01:38.86 9.3053 40.1688 0.988 2002-05-09 23:08:46.27 9.2940 40.1557 1.013 2002-05-10 02:06:10.01 9.2827 40.1910 0.905 2002-05-10 11:19:43.15 8.8903 38.8072 1.579 2002-05-10 22:16:37.98 9.3842 39.6327 1.230 2002-05-10 22:25:18.57 9.4728 39.6483 0.999 2002-05-10 22:34:05.46 9.4240 39.6792 0.875 2002-05-11 20:46:04.24 9.6903 39.8283 1.235 2002-05-11 21:26:11.65 9.3493 39.3185 0.607 2002-05-11 21:29:43.43 9.1253 39.9993 0.836

Appendices

122

2002-05-11 21:44:34.26 9.4275 39.6902 1.128 2002-05-12 05:21:16.79 9.5142 39.8377 1.733 2002-05-12 09:12:00.81 9.4312 39.6712 1.816 2002-05-12 10:01:24.05 9.4160 39.6657 1.846 2002-05-12 17:23:59.05 11.6993 39.7117 3.541 2002-05-12 21:02:17.23 9.4018 39.6703 1.099 2002-05-12 21:15:30.99 9.4630 39.6523 1.279 2002-05-13 01:48:20.11 8.9980 39.8948 0.965 2002-05-13 16:05:15.83 9.3998 39.6532 1.448 2002-05-14 09:24:28.79 9.0308 38.7682 1.309 2002-05-14 18:49:27 12.5900 41.1300 4.400 2002-05-14 22:23:52.58 8.1132 39.0617 1.141 2002-05-14 23:01:06.73 8.3132 39.1813 0.471 2002-05-15 00:18:13.29 8.3117 39.0890 0.516 2002-05-15 00:27:47.17 8.2657 39.1605 0.458 2002-05-15 01:09:00.55 8.1168 39.0420 1.041 2002-05-15 10:10:01.47 8.8747 38.5845 1.817 2002-05-15 20:34:24.55 8.1210 39.0448 1.042 2002-05-15 22:30:50.76 8.1607 39.0357 1.026 2002-05-15 22:56:39.52 8.1092 39.0473 0.906 2002-05-15 23:57:38.25 8.1583 39.0360 0.821 2002-05-16 00:50:19.98 8.1587 39.0385 1.183 2002-05-16 02:23:59.54 9.0268 39.9730 1.059 2002-05-16 20:52:08.75 10.2012 40.4232 2.079 2002-05-16 23:27:06.76 9.6193 39.2143 1.033 2002-05-17 02:10:34.92 9.2673 39.4307 0.976 2002-05-17 19:10:24.06 9.4615 39.6690 1.217 2002-05-17 21:13:22.82 9.4353 39.6875 1.134 2002-05-18 00:50:39.37 9.4238 39.6930 1.227 2002-05-18 01:20:50.78 8.9733 40.7367 1.806 2002-05-18 07:51:01.70 9.5465 40.3893 1.824 2002-05-18 09:39:43.78 8.0972 38.5175 2.252 2002-05-18 18:35:15.74 9.3288 38.7470 1.199 2002-05-19 00:09:17.48 9.4363 39.6767 0.668 2002-05-19 14:31:15.80 8.9060 39.7925 2.254 2002-05-19 17:29:53.96 9.8678 40.3975 1.871 2002-05-19 18:10:51.68 9.4380 39.6757 1.945 2002-05-19 18:52:45.54 9.4422 39.6982 1.573 2002-05-19 22:06:59.39 9.4158 39.7683 0.842 2002-05-20 03:02:33.31 9.4165 39.7322 1.881 2002-05-20 08:17:09.71 8.9292 38.7758 0.979 2002-05-20 16:58:48.93 9.1100 40.0137 1.382 2002-05-20 17:24:33.77 9.0292 39.7215 1.315 2002-05-20 21:28:47.46 9.4340 39.6920 1.493 2002-05-21 01:01:11.49 9.0220 40.6822 1.046 2002-05-21 02:32:26.86 9.4323 39.7000 2.069 2002-05-21 02:33:02.11 9.4410 39.6938 2.223 2002-05-21 06:53:56.36 9.4552 39.6960 1.793 2002-05-21 07:05:33.89 9.4235 39.6828 2.037 2002-05-21 09:08:37.12 9.3692 39.6295 1.348 2002-05-21 13:40:41.79 9.3663 39.6235 1.512 2002-05-21 23:41:40.11 9.4967 39.3438 0.639 2002-05-21 23:48:40.83 9.5318 39.3727 0.683 2002-05-22 00:17:09.82 9.4518 39.6853 2.563 2002-05-22 00:37:13.37 9.3983 39.6462 1.213 2002-05-22 00:42:15.27 9.4608 39.6735 1.126 2002-05-22 01:17:01.45 8.8982 40.6110 1.495 2002-05-22 01:24:01.44 9.4703 39.6603 1.203 2002-05-22 01:26:44.23 9.4177 39.6627 1.046 2002-05-22 01:34:33.34 9.5583 39.5872 1.027 2002-05-22 09:36:06.86 8.9868 38.6077 1.522 2002-05-22 20:45:18.74 9.4797 39.6967 1.177 2002-05-22 23:26:15.67 9.4615 39.6668 1.500 2002-05-23 02:00:23.00 9.3735 39.6230 1.055 2002-05-23 02:12:23.72 9.4638 39.6728 1.023 2002-05-23 03:22:03.52 9.4158 39.6745 2.056 2002-05-24 09:00:06.13 8.8718 39.9225 1.944 2002-05-24 10:03:36.91 8.9665 38.7680 1.041 2002-05-24 16:42:45.10 9.1147 39.6987 1.735 2002-05-24 23:06:33.06 9.2690 40.2135 1.154 2002-05-25 19:59:24.69 9.1003 39.9943 1.727 2002-05-25 21:01:04.59 9.2915 40.1720 0.970 2002-05-25 21:28:02.72 9.0925 39.9873 1.104 2002-05-26 22:00:01.41 9.1305 39.9982 2.605 2002-05-27 10:16:40.56 9.1167 40.0023 2.050 2002-05-28 21:00:26.59 9.3895 39.6250 1.074 2002-05-29 02:52:30.08 9.4142 39.6707 1.778 2002-05-29 02:54:47.44 9.4473 39.6863 1.803 2002-05-29 03:31:39.69 9.3807 39.6350 1.378 2002-05-30 09:41:52.50 8.9635 38.7732 0.624 2002-05-31 11:46:00.25 9.0428 38.6812 0.873 2002-06-01 00:59:29.63 9.2517 39.9498 1.254 2002-06-01 13:38:45.03 7.9947 38.9377 2.531 2002-06-01 21:51:51.80 9.4068 39.6560 1.469 2002-06-01 22:04:11.02 9.3853 39.6412 0.811 2002-06-01 22:54:10.57 9.4538 40.0202 1.353 2002-06-01 22:56:54.25 9.4665 40.0255 1.362 2002-06-02 02:21:07.92 9.1203 40.0037 1.476 2002-06-02 10:41:31.87 7.5223 38.7082 2.415 2002-06-02 10:44:13.39 7.5450 38.6987 2.389 2002-06-02 18:42:59.00 7.5690 38.6478 2.035 2002-06-03 00:42:25.37 8.8917 40.6255 0.783 2002-06-03 01:22:01.40 9.0777 40.0155 1.133 2002-06-03 08:11:02.85 8.8700 40.6372 2.286

Appendices

123

2002-06-03 09:00:08.87 9.0568 38.7163 0.696 2002-06-04 00:48:48.99 7.2350 38.8412 1.788 2002-06-04 15:51:53.30 9.4387 39.6830 2.453 2002-06-04 15:54:07.14 9.4527 39.6818 2.482 2002-06-04 15:58:58.84 9.3802 39.6385 1.502 2002-06-04 16:13:47.41 9.4280 39.6888 2.358 2002-06-04 16:29:36.50 9.3698 39.6377 1.804 2002-06-04 16:30:21.06 9.4310 39.6913 2.082 2002-06-04 17:36:56.22 9.4478 39.7032 1.819 2002-06-05 05:33:21.49 8.9975 39.9438 1.764 2002-06-07 08:00:31.66 9.4563 39.6935 1.816 2002-06-07 08:00:59.90 9.4532 39.6957 1.918 2002-06-07 08:07:28.25 9.4585 39.7077 1.943 2002-06-08 12:32:13.84 8.8693 40.6368 2.556 2002-06-08 21:11:35.65 9.4098 39.5997 0.750 2002-06-08 22:12:25.19 8.9038 40.6595 0.833 2002-06-09 01:28:45.30 7.5913 38.7557 2.281 2002-06-09 02:45:56.78 8.9072 40.6048 1.537 2002-06-09 07:48:09.60 9.2737 40.1938 1.751 2002-06-09 08:50:42.85 9.2658 40.2103 1.285 2002-06-09 08:55:01.88 9.2700 40.1985 1.774 2002-06-09 09:26:09.88 9.2802 40.1917 1.476 2002-06-09 09:28:35.83 9.2748 40.2008 1.603 2002-06-09 09:33:47.98 9.0565 39.9907 1.325 2002-06-09 11:37:19.05 9.2513 40.1813 1.265 2002-06-09 13:15:52.96 9.2872 40.1953 1.736 2002-06-09 15:55:52.75 9.2825 40.1927 2.295 2002-06-09 17:11:25.78 9.2750 40.1830 2.173 2002-06-09 22:51:28.31 9.2632 40.2180 1.328 2002-06-09 22:53:09.33 9.2552 40.2037 1.296 2002-06-10 03:17:26.95 9.2707 40.2115 1.577 2002-06-10 05:56:23.27 9.0600 39.9805 1.468 2002-06-10 09:47:17.49 9.0167 38.5765 1.069 2002-06-10 15:08:55.87 9.0973 39.9942 1.496 2002-06-11 10:17:01.70 8.8700 38.7913 1.058 2002-06-11 10:29:41.55 8.7922 38.8087 1.035 2002-06-11 21:17:09.37 9.3790 39.6283 1.154 2002-06-12 00:28:52.13 8.8935 40.6648 1.330 2002-06-12 08:14:20.66 9.5092 39.7148 3.121 2002-06-12 09:06:38.24 9.4635 39.6955 2.082 2002-06-12 12:25:12.64 8.9023 39.7990 2.028 2002-06-12 17:38:46.63 9.3725 39.6503 1.660 2002-06-13 01:33:06.00 9.3800 39.6267 1.186 2002-06-13 05:51:19.51 9.4308 39.6740 1.588 2002-06-13 14:50:42.21 9.4105 39.7073 1.680 2002-06-13 22:59:19.25 7.8693 38.8347 1.695 2002-06-14 02:30:15.92 9.3918 39.6553 1.432 2002-06-14 10:09:59.76 9.4182 39.6730 1.639 2002-06-14 19:04:06.18 8.7852 40.4478 1.340 2002-06-14 19:35:29.05 8.8660 40.6190 1.356 2002-06-14 20:22:21.62 9.5115 39.7062 2.593 2002-06-16 06:15:17.73 9.4525 39.6678 1.608 2002-06-16 15:24:29.02 9.4355 39.6908 2.034 2002-06-16 16:49:31.61 9.4215 39.6790 1.510 2002-06-16 17:57:44.37 9.4373 39.6902 1.397 2002-06-17 06:39:53.59 9.4240 39.6825 1.845 2002-06-17 09:26:19.60 8.9815 38.8063 1.618 2002-06-17 17:33:51.89 9.4292 39.6760 1.708 2002-06-17 18:49:15.36 8.8728 40.6243 1.177 2002-06-17 23:14:26.25 8.8597 40.6243 1.760 2002-06-18 02:15:36.55 8.8397 40.6630 1.344 2002-06-18 10:44:15 14.5200 42.1100 4.500 2002-06-18 16:41:54.39 8.8892 40.6727 1.423 2002-06-18 19:32:11.25 8.8910 40.6218 1.462 2002-06-18 20:38:06.78 7.9970 38.9538 1.298 2002-06-18 20:50:30.12 8.9035 40.6295 0.896 2002-06-18 22:16:12.23 7.9487 38.8548 2.223 2002-06-18 23:17:10.33 8.8488 40.6543 1.005 2002-06-20 20:44:29.96 7.7990 38.8353 1.266 2002-06-20 21:44:27.73 7.8683 38.7745 1.221 2002-06-20 22:00:21.96 8.7903 39.3158 1.104 2002-06-21 00:43:32.73 9.2822 39.3440 1.413 2002-06-21 09:31:27.18 9.3545 39.6197 1.549 2002-06-21 10:56:19.19 9.4298 39.6817 2.542 2002-06-21 11:10:19.92 9.4193 39.6718 1.833 2002-06-21 21:09:04.99 9.4108 39.5960 0.920 2002-06-21 22:10:50.97 9.3985 39.6412 0.834 2002-06-21 23:23:32.79 9.8263 41.0388 1.805 2002-06-21 23:26:38.84 9.8650 41.0802 2.590 2002-06-21 23:50:29.16 8.9033 40.6355 2.327 2002-06-22 22:32:33.36 9.3227 39.5942 0.893 2002-06-22 22:35:05.13 9.4020 39.6508 1.154 2002-06-22 23:28:41.40 9.6793 40.5837 1.556 2002-06-23 00:08:01.93 9.3932 40.3785 1.335 2002-06-23 00:50:32.31 10.0047 41.1250 2.327 2002-06-23 14:21:09.34 9.2310 39.5377 1.529 2002-06-23 20:07:15.02 9.2393 40.1367 1.452 2002-06-23 20:56:04.23 9.2473 40.1337 1.309 2002-06-23 21:34:20.71 8.2577 39.1813 0.946 2002-06-23 22:36:02.62 9.4412 39.6505 1.214 2002-06-24 11:04:33.47 9.4880 39.6718 1.851 2002-06-24 20:19:20.09 9.2085 40.0153 1.539 2002-06-25 09:38:59.33 8.9848 38.7438 0.732 2002-06-25 17:38:49.32 9.3723 39.6258 1.421

Appendices

124

2002-06-25 17:42:33.41 9.4007 39.6060 1.163 2002-06-25 21:34:40.69 9.3662 39.6263 1.126 2002-06-26 02:32:28.40 9.4453 39.6627 1.186 2002-06-26 07:40:27.72 9.4695 39.6912 1.571 2002-06-26 20:42:43.90 9.7880 40.8977 1.955 2002-06-27 05:43:18.78 10.0658 39.9205 2.143 2002-06-28 23:43:13.27 9.4192 39.6575 1.323 2002-06-29 00:00:44.95 7.5067 38.7492 1.705 2002-06-29 18:33:54.68 9.4538 39.6800 1.377 2002-06-29 18:40:46.28 9.4238 39.6535 1.166 2002-06-29 20:27:45.68 9.3633 39.6865 1.212 2002-06-29 23:16:59.29 9.4583 39.6905 1.836 2002-06-30 05:37:35.86 9.3755 39.2380 1.552 2002-06-30 20:21:19.42 9.1120 39.9732 1.027 2002-06-30 23:50:52.02 9.4238 39.6795 1.650 2002-07-01 09:22:58.38 8.9660 38.7695 0.902 2002-07-01 09:25:06.86 9.0293 38.7542 0.866 2002-07-01 18:13:12.77 9.5987 39.6787 1.719 2002-07-01 21:57:59.10 9.4877 39.6873 1.566 2002-07-01 23:04:17.43 9.6832 40.5928 1.561 2002-07-01 23:24:22.93 9.5015 39.7077 1.583 2002-07-01 23:29:51.81 9.4947 39.7085 1.198 2002-07-01 23:40:39.74 9.8108 41.0107 2.239 2002-07-02 21:45:03.50 9.4507 39.6880 1.453 2002-07-02 21:48:18.92 9.4220 39.6223 0.742 2002-07-03 19:32:12.73 8.8282 39.7940 1.022 2002-07-03 19:36:08.14 9.1827 39.9797 1.300 2002-07-03 19:37:52.99 9.1453 39.9615 1.027 2002-07-03 19:47:54.24 9.1727 39.9745 2.173 2002-07-03 19:48:21.63 9.1690 39.9845 2.177 2002-07-03 20:13:38.36 9.1252 39.9408 1.083 2002-07-03 20:17:42.09 9.1755 39.9797 1.402 2002-07-03 20:57:30.21 9.3133 40.1917 0.763 2002-07-03 21:32:58.70 9.1647 39.9708 1.437 2002-07-03 21:38:16.97 9.0082 39.9448 1.201 2002-07-03 21:47:02.47 9.1833 39.9712 1.780 2002-07-03 22:52:19.22 9.1668 39.9545 1.106 2002-07-03 22:52:52.96 9.1995 39.9837 1.145 2002-07-03 23:11:23.61 9.1833 39.8868 0.835 2002-07-03 23:51:56.97 9.1767 39.9808 1.929 2002-07-03 23:59:19.49 9.1912 39.9690 1.300 2002-07-04 00:10:55.13 8.8833 39.8605 0.820 2002-07-04 01:35:51.13 9.1712 39.9888 1.113 2002-07-04 01:45:13.19 9.1652 39.9478 1.126 2002-07-04 02:53:01.18 9.1778 39.9827 3.023 2002-07-04 02:53:31.13 9.1605 39.9743 2.658 2002-07-04 02:54:55.57 9.1708 39.9785 1.759 2002-07-04 02:55:47.79 9.1440 39.9408 1.433 2002-07-04 02:58:35.20 9.1547 39.9598 2.737 2002-07-04 02:59:42.24 9.1770 39.9933 3.543 2002-07-04 03:03:26.46 9.1713 39.9673 3.475 2002-07-04 05:59:02.07 9.1678 39.9848 1.682 2002-07-04 06:00:11.72 9.4190 39.6657 1.741 2002-07-04 10:20:59.00 9.3847 39.6312 1.462 2002-07-05 03:29:11.58 9.1908 39.9615 1.410 2002-07-05 23:05:04.85 9.4665 39.6915 2.275 2002-07-06 00:28:21.84 9.4670 39.6863 1.640 2002-07-06 08:41:40.26 9.0470 38.8178 0.775 2002-07-06 17:05:18.73 10.2210 40.4283 1.974 2002-07-07 00:47:09.52 9.2563 40.2500 1.307 2002-07-07 01:52:41.72 10.4277 40.1647 2.382 2002-07-07 10:17:38.67 9.3923 39.6333 1.378 2002-07-07 19:22:27.11 9.3713 39.6478 0.942 2002-07-07 19:58:12.87 9.4010 39.6123 1.070 2002-07-07 20:02:10.16 9.4365 39.6765 1.760 2002-07-07 20:18:18.32 9.5028 39.3672 0.467 2002-07-07 21:47:47.14 9.4247 39.6488 0.939 2002-07-07 23:52:04.45 9.4317 39.6478 0.860 2002-07-08 00:19:07.95 9.4470 39.6818 2.360 2002-07-08 00:30:08.28 9.4268 39.6857 1.194 2002-07-08 03:10:10.50 9.4647 39.6972 1.844 2002-07-09 01:52:19.77 9.1152 39.9660 1.102 2002-07-09 11:04:09.28 8.9755 39.9298 1.146 2002-07-09 19:20:04.44 9.8505 41.0880 3.634 2002-07-09 21:38:29.26 9.2132 40.1392 0.953 2002-07-09 23:45:41.68 9.9058 41.1118 2.440 2002-07-10 13:12:18.30 8.8950 39.9075 2.090 2002-07-10 16:38:37.05 9.1008 40.0038 1.880 2002-07-10 19:38:53.92 9.0072 39.9317 1.642 2002-07-10 21:14:38.81 9.4535 39.6492 1.007 2002-07-10 21:16:31.45 9.4973 39.6237 0.973 2002-07-11 03:15:12.74 8.9585 39.9035 1.620 2002-07-11 16:53:48.85 9.4273 39.6993 1.645 2002-07-11 16:56:51.22 9.4242 39.6488 1.667 2002-07-11 19:52:04.08 9.8193 40.9952 2.440 2002-07-12 00:17:58.98 9.8195 41.0112 1.828 2002-07-12 00:47:12.33 9.1983 39.9557 1.171 2002-07-12 15:07:28.46 8.8607 39.8455 1.614 2002-07-12 20:09:59.12 9.5347 39.6938 1.377 2002-07-12 21:28:26.10 8.1623 39.1090 1.323 2002-07-13 03:23:24.09 9.4790 39.6878 1.696 2002-07-13 03:25:19.42 9.4820 39.6803 1.535 2002-07-13 05:08:52.08 9.4373 39.6820 2.065 2002-07-14 01:31:19.30 11.8915 39.3928 3.146

Appendices

125

2002-07-14 01:46:08.68 7.3560 38.6148 1.650 2002-07-14 19:54:07.45 8.1203 39.0893 1.108 2002-07-15 07:32:30.10 8.8437 40.6870 1.464 2002-07-15 19:04:25.45 9.0130 39.9232 1.870 2002-07-16 02:27:11.57 9.1653 40.0252 1.109 2002-07-16 08:48:39.87 9.0015 38.8057 1.225 2002-07-16 12:20:45.49 9.3220 40.0137 1.855 2002-07-16 20:07:00.22 7.4943 38.7445 1.630 2002-07-16 20:07:54.66 7.4967 38.7522 1.597 2002-07-16 20:20:15.93 7.3053 38.4578 2.360 2002-07-16 22:19:23.58 9.4283 39.6715 1.273 2002-07-17 00:25:55.79 9.4405 39.6925 2.000 2002-07-17 00:29:16.63 9.4500 39.6343 1.195 2002-07-17 01:08:25.72 9.4133 39.6833 1.188 2002-07-17 01:13:51.88 9.4297 39.6783 1.670 2002-07-17 02:14:28.68 9.4363 39.6885 1.270 2002-07-17 02:35:11.54 9.3875 39.6617 1.853 2002-07-17 02:35:25.63 9.5858 39.4918 1.840 2002-07-17 02:54:25.91 9.4557 39.6992 1.826 2002-07-17 22:26:58.50 9.4005 39.6692 1.117 2002-07-17 22:29:02.82 9.4325 39.6757 1.269 2002-07-17 22:50:05.73 9.3680 39.6330 0.955 2002-07-18 00:50:25.17 9.8522 41.0892 1.982 2002-07-18 22:37:10.58 8.8908 40.6222 0.887 2002-07-19 23:50:44.51 9.0153 39.9382 0.831 2002-07-19 23:51:58.34 9.0507 39.9765 1.148 2002-07-20 11:57:56.12 9.0467 39.2783 1.635 2002-07-20 13:27:02.66 9.4640 39.6892 2.311 2002-07-20 13:44:35.40 8.8563 39.8395 1.902 2002-07-20 21:54:16.94 9.4075 39.6403 1.069 2002-07-20 22:04:28.95 9.4603 39.6890 1.419 2002-07-20 22:28:19.99 9.4580 39.6948 1.706 2002-07-20 22:31:52.86 9.3495 40.0498 1.251 2002-07-21 02:35:16.97 9.4445 39.6867 1.865 2002-07-22 00:09:54.76 9.4415 39.6840 0.972 2002-07-22 16:30:54.44 9.1395 39.9810 1.455 2002-07-23 17:25:17.23 9.4783 39.6967 1.598 2002-07-24 00:58:00.36 7.6098 38.8928 1.745 2002-07-24 17:13:56.03 9.4868 39.6582 1.679 2002-07-24 17:46:08.67 9.5230 40.7475 1.916 2002-07-25 16:36:49.42 9.4508 39.7012 1.957 2002-07-26 01:21:47.95 10.6105 39.7585 2.348 2002-07-26 18:42:01.17 9.0968 39.9830 1.488 2002-07-27 23:13:37.61 8.2945 39.9725 1.240 2002-07-28 04:05:06.68 7.0552 38.5750 2.670 2002-07-29 01:46:27.42 9.4723 39.6973 1.511 2002-07-29 18:17:29.17 9.2245 40.5072 1.172 2002-07-30 02:15:10.60 9.6893 41.4115 2.100 2002-07-30 22:25:04.38 9.3790 39.6527 1.047 2002-07-30 23:42:48.89 9.4402 39.6677 0.963 2002-07-30 23:46:13.21 9.4495 39.6757 1.287 2002-07-31 01:54:38.01 9.4515 39.6832 2.339 2002-07-31 13:55:42.39 8.7135 38.5218 2.141 2002-08-01 13:37:42.58 8.8912 39.8187 2.007 2002-08-01 21:48:25.18 9.5495 39.7365 1.088 2002-08-02 05:27:24.97 9.5650 39.1835 1.546 2002-08-02 07:05:52.64 9.5750 39.2253 1.959 2002-08-02 09:23:12.76 8.9265 38.5180 1.612 2002-08-02 09:41:59.13 9.0395 38.6048 2.275 2002-08-02 12:35:32.91 7.0967 38.5407 2.661 2002-08-03 00:24:15.64 9.8430 41.0823 3.349 2002-08-04 00:20:31.63 9.2887 40.1798 1.266 2002-08-04 00:24:31.16 9.2647 40.2152 1.058 2002-08-04 00:30:35.15 9.2848 40.1742 1.046 2002-08-05 05:39:35.06 9.4647 39.7017 1.864 2002-08-05 10:50:20.16 9.4615 39.6928 2.211 2002-08-05 11:11:22.25 9.4587 39.6935 2.196 2002-08-05 13:48:16.01 9.4647 39.6958 1.960 2002-08-08 01:50:33 14.0100 39.9400 4.870 2002-08-08 02:08:10.77 8.7830 40.4002 2.068 2002-08-08 17:10:59.60 8.8855 40.6260 1.131 2002-08-08 21:17:11 13.6500 40.0000 5.050 2002-08-09 02:52:02.31 7.1392 38.4015 1.996 2002-08-09 16:42:54.00 10.3273 40.5013 2.224 2002-08-09 22:08:42 11.8200 43.6500 4.990 2002-08-10 05:04:04.46 9.4228 39.6622 2.030 2002-08-10 09:45:41 12.1300 43.8800 4.740 2002-08-10 12:01:20 13.9200 39.9000 4.650 2002-08-10 15:56:02 13.6500 39.8100 6.110 2002-08-10 16:45:56 13.7100 39.8900 4.970 2002-08-11 01:11:13.06 9.0112 39.9732 1.137 2002-08-11 02:12:38.89 9.0257 39.9427 1.270 2002-08-11 07:17:34.31 8.8802 40.6180 1.987 2002-08-11 20:29:38 12.0200 43.8700 4.470 2002-08-12 08:54:21.13 8.9337 38.7817 0.982 2002-08-12 11:52:00.14 8.9778 39.6992 1.748 2002-08-12 13:59:50.22 8.9792 39.6915 1.783 2002-08-12 14:08:46.87 8.9807 39.7013 1.634 2002-08-13 08:01:02.41 9.4482 39.4565 1.479 2002-08-13 12:20:02.82 8.9298 38.5517 1.451 2002-08-13 23:36:45.63 7.9535 38.9412 1.138 2002-08-14 17:51:52.13 9.8280 40.5430 2.733 2002-08-14 20:33:05.29 8.1917 39.1520 1.615 2002-08-14 20:49:25 13.4100 39.9900 4.480

Appendices

126

2002-08-17 00:16:33.89 9.6093 39.7152 2.536 2002-08-17 02:25:13.20 8.7630 40.4102 1.394 2002-08-18 02:00:47.01 9.0902 39.9790 1.622 2002-08-18 03:10:09.29 9.7977 40.8792 2.245 2002-08-18 22:25:53.23 8.1583 39.1708 1.446 2002-08-19 18:47:40.80 9.4312 39.6923 1.789 2002-08-20 01:04:46.06 9.0435 40.0000 0.884 2002-08-20 02:08:05.43 9.0997 39.9840 0.897 2002-08-20 09:25:54.46 8.9877 38.7357 1.188 2002-08-20 20:25:58.81 9.1113 39.9887 1.172 2002-08-21 01:27:23.87 8.9513 39.7162 2.144 2002-08-22 01:09:36.64 6.8693 39.9947 1.748 2002-08-22 16:44:55.43 9.4382 39.6895 1.571 2002-08-22 18:47:52.93 9.4418 39.6690 1.367 2002-08-22 19:50:44.67 7.5010 38.8265 1.888 2002-08-22 21:31:17.49 9.0927 40.0127 1.089 2002-08-24 08:42:03.82 9.0623 38.7085 0.996 2002-08-24 14:35:38.76 9.1123 39.8450 2.013 2002-08-24 18:30:11.37 9.2503 39.9210 1.295 2002-08-24 20:50:20.36 9.2612 39.9215 1.744 2002-08-25 03:34:23.66 9.2513 39.9258 1.818 2002-08-25 06:12:37 13.6700 39.8600 5.120 2002-08-26 04:25:45.06 9.3872 39.6355 2.091 2002-08-26 15:58:18.50 9.4787 39.6815 1.766 2002-08-26 18:24:00.34 9.1130 39.9768 1.250 2002-08-27 01:27:59.80 9.2180 39.9412 1.051 2002-08-27 16:14:30.61 9.7038 39.6808 2.430 2002-08-27 22:59:23.76 9.2607 39.9333 1.125 2002-08-27 23:08:13.05 9.2653 39.9148 1.212 2002-08-27 23:08:12.85 9.2575 39.9208 1.211 2002-08-28 10:35:51.85 8.9568 39.3622 1.578 2002-08-28 21:40:57.62 9.5130 41.0783 1.513 2002-08-29 13:10:39.37 9.4280 38.3957 2.059 2002-08-29 23:09:35.03 9.8763 41.6808 1.824 2002-08-30 11:43:33.22 9.0715 40.8428 1.847 2002-08-30 12:53:58.32 9.4467 39.6670 1.616 2002-08-31 04:40:14.72 9.4562 39.3083 1.867 2002-08-31 16:19:56.18 9.1135 39.9695 1.448 2002-08-31 23:33:15.64 9.9620 40.4593 2.176 2002-09-01 23:04:52.94 10.0898 40.4165 2.250 2002-09-01 23:09:40.05 10.0895 40.4335 1.510 2002-09-01 23:25:40.33 10.1405 40.4458 2.234 2002-09-01 23:33:46.47 8.9195 40.6117 3.237 2002-09-01 23:38:03.64 8.8915 40.6113 3.372 2002-09-01 23:57:34.33 8.8865 40.6492 0.967 2002-09-01 23:57:49.24 8.8902 40.6202 1.102 2002-09-02 00:30:07.68 8.8918 40.6337 1.154 2002-09-02 00:43:52.49 8.9142 40.6500 0.772 2002-09-02 01:05:32.79 8.8960 40.6342 0.770 2002-09-02 01:11:19.44 8.9098 40.6035 0.786 2002-09-02 01:18:10.63 8.9080 40.6120 0.799 2002-09-02 02:54:19.55 8.8983 40.6205 0.851 2002-09-02 04:43:00.60 8.8922 40.6223 1.334 2002-09-02 05:55:43.00 8.8975 40.6337 1.401 2002-09-03 00:38:50.38 9.0097 40.7215 1.389 2002-09-03 00:49:40.09 8.8990 40.6123 1.528 2002-09-03 00:52:37.20 8.9278 40.5932 1.030 2002-09-03 03:43:51.98 8.8418 40.6462 1.557 2002-09-03 07:07:09.38 8.9027 40.6138 1.787 2002-09-03 07:28:41.92 8.9190 40.5975 1.317 2002-09-03 18:11:53.64 9.2013 39.7623 2.039 2002-09-03 20:14:16.58 9.4598 40.0452 1.362 2002-09-03 20:51:12.19 9.5085 39.9853 1.735 2002-09-03 22:25:13.44 8.8835 40.6097 1.602 2002-09-03 23:22:45.34 9.2995 40.1802 1.069 2002-09-03 23:30:30.81 9.2850 40.1928 0.858 2002-09-03 23:31:05.24 9.2910 40.1963 1.036 2002-09-04 00:47:01.35 9.1010 40.0037 1.043 2002-09-04 05:44:39.83 9.1870 40.0203 1.963 2002-09-04 09:15:09.15 8.9787 38.6488 1.139 2002-09-04 19:08:10.34 8.8903 40.6162 2.152 2002-09-04 19:54:00.39 8.9102 40.6115 1.575 2002-09-05 01:08:44.67 7.5782 38.5932 1.595 2002-09-05 01:50:45.58 9.1513 39.7585 1.226 2002-09-05 10:15:43.05 9.4300 39.7103 1.763 2002-09-06 00:47:53.58 9.1003 39.9947 0.957 2002-09-06 00:59:09.90 9.1075 39.9953 1.553 2002-09-06 02:29:37.33 9.0562 40.0198 0.735 2002-09-06 19:42:05.87 9.8600 40.4443 1.280 2002-09-06 21:38:10.10 9.5705 40.0677 1.425 2002-09-07 20:07:40.81 9.4777 39.6617 1.074 2002-09-07 23:08:37.31 8.9087 40.6328 1.044 2002-09-07 23:16:23.04 8.9003 40.6308 1.238 2002-09-08 15:40:15.51 7.8267 38.9423 2.468 2002-09-08 21:48:39.19 9.7763 39.8153 1.770 2002-09-09 02:34:12.44 9.4297 39.6703 1.623 2002-09-09 03:41:30.98 7.7877 38.8737 2.071 2002-09-09 03:50:53.78 9.4537 39.6830 1.941 2002-09-09 04:03:54.56 9.4413 39.6785 1.758 2002-09-09 04:12:20.36 9.4328 39.6858 1.893 2002-09-09 21:41:34.69 9.7178 41.3618 1.730 2002-09-09 23:33:14.57 9.5967 39.7002 1.053 2002-09-10 02:40:44.28 7.6657 38.9097 1.527 2002-09-10 22:54:56.06 9.7208 39.2213 1.979

Appendices

127

2002-09-11 00:07:48.01 9.1963 39.9032 1.096 2002-09-11 00:18:42.03 7.7710 38.9997 1.227 2002-09-11 07:38:01.03 9.4402 39.6983 1.928 2002-09-11 21:13:40.25 6.3560 37.9905 1.975 2002-09-11 21:22:08.45 9.0013 40.4663 0.620 2002-09-12 00:07:18.97 8.2057 39.1185 0.515 2002-09-13 13:15:18.61 9.0942 39.9892 1.322 2002-09-13 19:07:45.09 9.9228 41.6967 2.531 2002-09-13 21:06:17.22 9.4355 39.6865 1.654 2002-09-13 21:37:02.33 9.4052 39.6647 1.228 2002-09-13 23:04:23.12 8.9248 40.5567 0.965 2002-09-14 01:37:07.47 8.8168 40.6643 0.910 2002-09-15 10:37:27.14 9.7938 39.8290 2.730 2002-09-15 14:22:22.99 8.8813 40.6297 1.468 2002-09-16 15:07:49.28 9.3768 39.6362 1.719 2002-09-17 00:25:55.93 9.4403 39.6765 1.112 2002-09-17 02:07:55.32 9.5935 40.2245 2.200 2002-09-17 04:53:28.14 10.1368 40.6380 2.181 2002-09-17 05:51:45.61 10.1032 40.6247 2.303 2002-09-17 05:54:09.37 9.9418 40.5317 1.955 2002-09-17 14:50:16.74 10.1090 40.6782 2.738 2002-09-17 21:49:37.15 10.9893 40.8538 2.379 2002-09-17 22:25:16.91 10.9960 40.8230 2.697 2002-09-17 22:28:48.27 10.8765 40.7492 1.829 2002-09-18 00:04:53.61 10.9462 40.8360 2.831 2002-09-18 00:07:15.14 10.9653 40.8442 2.806 2002-09-18 02:30:47.39 10.1402 39.2793 2.340 2002-09-18 03:19:35.05 11.1300 41.2320 3.300 2002-09-18 06:02:13.92 11.2662 37.7387 3.400 2002-09-19 01:06:53.10 9.5125 40.0537 2.258 2002-09-19 01:27:11.25 9.4745 40.0393 1.202 2002-09-19 07:19:35.72 8.8722 40.6353 1.941 2002-09-19 07:55:35.99 9.5007 40.0478 1.958 2002-09-19 22:48:02.42 7.9412 38.8155 0.705 2002-09-20 09:11:47.99 8.9552 38.5162 1.593 2002-09-20 17:29:56.32 9.4470 39.6687 1.675 2002-09-20 19:21:14.43 10.9047 41.5912 2.688 2002-09-20 19:36:45.93 11.9785 42.0013 3.333 2002-09-21 15:48:02.07 9.5437 39.7060 2.004 2002-09-21 17:53:00.27 9.6033 39.2830 1.905 2002-09-21 18:16:37.04 8.8787 40.6190 2.275 2002-09-21 18:44:00.95 9.4448 39.6687 1.375 2002-09-21 18:46:01.69 8.8903 40.6215 2.258 2002-09-22 00:30:28.08 8.8885 40.6093 1.147 2002-09-22 10:55:54.23 9.2848 40.1742 1.718 2002-09-22 10:57:29.56 9.2898 40.1732 1.787 2002-09-22 15:59:26.45 9.3508 39.6840 1.786 2002-09-22 16:19:36.91 9.1613 39.9735 2.103 2002-09-22 20:03:13.15 8.8290 40.6907 1.190 2002-09-23 00:14:33.84 8.4803 39.3262 0.968 2002-09-23 01:57:30.65 9.4445 39.6887 1.121 2002-09-23 16:59:35.12 8.5325 39.3468 1.206 2002-09-23 17:16:45.44 9.4242 39.6822 1.725 2002-09-23 20:28:43.99 9.4062 39.6615 1.299 2002-09-23 22:02:34.09 9.4297 39.6902 1.473 2002-09-23 22:40:32.25 9.0213 38.5398 1.624 2002-09-23 23:45:51.77 9.4235 39.6680 1.381 2002-09-23 23:47:38.29 9.2900 40.1843 1.260 2002-09-24 01:02:12.32 9.2907 40.1870 1.087 2002-09-24 01:06:56.78 9.3040 40.1828 1.292 2002-09-24 06:44:41.17 9.3733 39.6292 1.703 2002-09-24 06:49:53.90 9.4367 39.6857 2.348 2002-09-24 10:03:43.61 9.4910 40.0458 2.368 2002-09-24 18:57:29.79 11.1270 39.8633 4.206 2002-09-24 22:07:36.48 9.3247 39.4478 1.217 2002-09-24 22:39:34.34 10.2418 40.4943 1.847 2002-09-24 22:39:48.74 8.9327 39.6868 1.369 2002-09-25 00:13:06.07 8.9200 39.7262 0.924 2002-09-25 08:58:51.78 9.0197 38.5292 1.621 2002-09-25 16:49:38.12 9.0198 38.5310 1.306 2002-09-25 22:42:10.50 9.0082 38.5172 2.527 2002-09-26 00:13:21.87 10.2760 39.8242 1.942 2002-09-26 01:03:17.44 9.0040 38.5237 1.462 2002-09-26 11:58:51.68 9.1035 39.9785 1.466 2002-09-26 13:20:16.50 8.9168 40.6075 1.782 2002-09-27 00:22:45.07 8.1813 39.0895 0.825 2002-09-27 00:56:00.61 9.4350 39.6708 1.227 2002-09-27 00:59:01.11 9.3930 39.6693 1.035 2002-09-27 02:30:29.81 9.4268 39.6783 1.538 2002-09-27 03:13:15.26 9.4397 39.6597 1.205 2002-09-27 22:45:29.22 8.9080 40.5928 0.946 2002-09-27 23:22:37.33 8.1617 39.0340 0.920 2002-09-28 13:12:56.07 9.4188 39.7060 1.587 2002-09-28 22:42:19.15 9.4178 39.6487 0.958 2002-09-28 23:00:25.90 9.9313 39.2310 1.197 2002-09-29 02:04:38.15 9.4320 39.6775 1.319 2002-09-29 02:06:30.07 10.5105 39.7612 1.717 2002-09-29 18:55:40.59 9.4378 39.6728 1.497 2002-09-30 00:59:42.33 8.2025 39.1513 0.928 2002-10-01 00:25:08.95 9.2355 40.1935 0.754 2002-10-01 02:21:57.16 9.3717 39.6512 1.197 2002-10-02 01:30:22.09 9.5007 40.0298 2.469 2002-10-02 01:50:00.92 7.7762 39.0053 1.304 2002-10-02 23:50:47.60 9.0742 39.9875 0.871

Appendices

128

2002-10-03 22:31:19.84 8.1797 39.1648 0.909 2002-10-03 23:07:11.47 11.3485 40.6997 2.349 2002-10-04 07:22:33.24 8.9985 40.5828 1.569 2002-10-04 08:34:42.79 11.1937 40.8820 2.767 2002-10-04 16:19:00.35 11.1318 41.0248 3.107 2002-10-04 20:11:57.13 9.2218 40.0318 1.309 2002-10-04 20:13:54.44 9.1993 40.0388 1.216 2002-10-04 20:18:35.86 7.9483 38.8173 1.868 2002-10-04 20:20:40.21 7.9925 38.9242 0.852 2002-10-04 21:15:52.38 10.8105 39.7997 2.619 2002-10-05 00:49:13.61 10.0608 39.6048 2.252 2002-10-06 09:19:56.84 10.3135 40.4927 2.898 2002-10-06 17:29:56.98 9.5490 39.7527 1.205 2002-10-06 21:39:17.56 9.3927 39.7195 1.038 2002-10-06 22:02:18.12 9.6560 39.2718 1.186 2002-10-07 02:25:34.38 10.2492 40.4388 2.311 2002-10-07 02:51:14.73 7.9292 38.8907 0.915 2002-10-07 20:31:13.86 9.8528 40.4793 1.307 2002-10-07 21:28:43.20 9.1885 39.9400 1.039 2002-10-07 21:47:47.21 9.0805 39.9355 0.791 2002-10-07 22:23:53.01 9.4313 39.6682 1.097 2002-10-07 23:01:15.52 9.4352 39.6720 1.370 2002-10-08 14:58:17.07 8.8847 39.8367 2.158 2002-10-08 19:37:43.23 9.1915 39.9612 2.026 2002-10-08 22:10:55.62 9.1825 39.9600 1.202 2002-10-08 23:51:32.61 9.1637 39.9675 0.834 2002-10-09 02:18:50.49 9.2007 39.9708 1.420 2002-10-09 02:23:35.88 9.1887 39.9668 1.825 2002-10-09 18:19:37.82 9.1803 40.0008 2.137 2002-10-09 23:41:43.46 8.1443 39.0630 0.713 2002-10-10 00:15:19.77 9.0993 39.9932 0.844 2002-10-10 00:59:07.82 8.1442 39.0453 0.763 2002-10-10 19:15:51.28 9.0518 39.9773 1.173 2002-10-10 20:04:20.00 12.9163 40.9795 4.034 2002-10-10 20:17:10.61 9.6788 39.7570 1.382 2002-10-11 01:17:05.42 9.4990 39.8013 1.558 2002-10-11 02:57:43.00 9.3977 39.7143 1.869 2002-10-11 15:29:40.42 9.4320 38.3885 1.938 2002-10-11 23:51:35.52 10.1093 40.7738 1.702 2002-10-12 02:07:15.17 7.4897 38.5888 1.619 2002-10-12 09:16:53.68 9.2018 40.0275 1.957 2002-10-13 00:11:30.60 9.3648 39.7418 1.501 2002-10-13 07:59:23.82 10.6632 40.9100 2.735 2002-10-14 00:28:58.03 8.8723 40.6213 0.946 2002-10-14 02:01:03.66 7.4215 38.6383 1.399 2002-10-14 21:13:16.17 7.6578 39.2627 1.347 2002-10-14 22:40:21.59 9.4645 40.0695 1.624 2002-10-15 11:04:50.84 10.3120 40.5017 3.134 2002-10-15 15:38:34.13 12.5065 39.3562 2.715 2002-10-15 17:41:31.27 9.5620 39.6830 1.327 2002-10-15 21:38:37.21 8.9220 40.6442 0.667 2002-10-15 21:38:54.99 8.9163 40.6423 0.684 2002-10-15 21:41:24.04 8.8972 40.6267 0.521 2002-10-15 21:42:08.59 9.5633 39.7875 0.830 2002-10-15 22:53:48.53 9.1642 39.9838 1.068 2002-10-16 02:55:50.75 8.9040 40.6210 1.368 2002-10-16 18:40:43.26 7.7833 38.9452 2.489 2002-10-17 04:58:06.72 10.2385 39.9700 1.954 2002-10-17 15:38:33.54 12.4667 39.1118 3.607 2002-10-17 20:15:07.38 8.1678 39.1720 0.642 2002-10-18 01:28:09.22 9.2795 40.2228 1.140 2002-10-18 17:11:53.90 11.6880 37.7320 3.206 2002-10-18 20:21:12.04 9.1748 39.9902 1.149 2002-10-19 00:36:15.24 9.5197 39.5987 1.305 2002-10-19 02:58:44.17 9.3953 39.7392 1.564 2002-10-19 21:25:25.86 10.1198 39.9775 2.844 2002-10-19 22:08:46.72 9.4753 39.6455 0.913 2002-10-19 22:31:32.77 10.2602 40.4488 1.939 2002-10-19 23:30:10.07 10.0707 39.9720 1.534 2002-10-20 00:29:11.25 9.7593 39.2217 1.384 2002-10-20 03:02:48.63 9.8702 39.8862 1.869 2002-10-20 05:36:38.50 8.1635 39.1070 1.672 2002-10-20 23:48:43.76 9.8212 40.5328 1.594 2002-10-21 02:41:18.29 9.2890 40.1853 1.165 2002-10-21 17:57:24.81 9.4592 40.0278 1.501 2002-10-21 20:26:36.98 9.5838 39.3292 1.451 2002-10-22 05:51:28.56 11.6483 41.5418 3.451 2002-10-22 22:13:49.55 9.4470 39.6457 1.289 2002-10-22 22:18:02.99 7.6618 38.8497 1.592 2002-10-23 01:50:22.88 9.5003 39.9457 1.167 2002-10-23 10:06:59.51 9.3902 40.0503 1.664 2002-10-23 23:15:37.81 9.1900 39.6725 1.092 2002-10-24 21:16:51.78 10.7538 39.6948 2.101 2002-10-24 22:45:17.94 10.1582 40.7535 1.534 2002-10-24 23:23:48.91 9.9322 39.8037 1.786 2002-10-24 23:26:15.41 9.9308 39.8187 1.224 2002-10-24 23:27:51.79 9.9498 39.8338 1.944 2002-10-24 23:40:26.66 9.9083 39.8017 1.031 2002-10-25 02:54:14.62 9.2290 40.2348 1.522 2002-10-25 21:22:01.19 9.9168 39.8080 1.153 2002-10-26 01:32:37.33 9.5330 39.7520 1.421 2002-10-27 01:43:26.73 7.9998 38.8322 1.183 2002-10-27 06:16:17.24 9.8198 41.5903 2.604 2002-10-27 22:06:26.98 9.2922 40.2022 0.344

Appendices

129

2002-10-28 09:07:57.39 7.6253 38.7807 2.803 2002-10-28 22:24:38.41 9.3275 38.7515 0.377 2002-10-28 22:46:18.56 9.0878 39.9958 0.942 2002-10-29 00:33:03.21 9.2757 40.2277 0.442 2002-10-29 17:15:38.56 9.9813 41.6210 2.969 2002-10-30 09:08:15.22 8.7328 38.5843 1.590 2002-10-30 18:16:14.56 9.4758 40.0235 1.625 2002-10-31 01:00:05.71 9.2263 40.2213 0.799 2002-10-31 17:30:15.36 9.8282 40.4542 1.996 2002-11-01 00:41:09.71 9.6713 40.2182 1.344 2002-11-01 18:42:39.84 9.4583 39.9430 1.033 2002-11-02 09:10:49.51 8.6995 38.5907 1.592 2002-11-02 21:08:20.76 9.4595 39.6812 1.171 2002-11-02 22:48:01.77 8.6762 39.0103 0.958 2002-11-02 22:53:05.19 10.1885 41.6537 2.109 2002-11-03 22:51:45.41 9.4397 39.6638 1.495 2002-11-03 22:52:12.27 9.4518 39.6748 1.550 2002-11-03 22:59:20.33 9.4848 39.6652 1.171 2002-11-04 00:17:42.30 8.4275 39.6783 1.540 2002-11-04 00:24:55.79 7.7953 38.9923 1.712 2002-11-04 01:04:39.00 7.4090 38.5600 1.589 2002-11-04 01:53:12.47 10.1303 38.7072 1.425 2002-11-04 02:46:18.05 7.7820 38.0920 1.934 2002-11-05 22:42:13.87 9.7362 39.7615 1.922 2002-11-05 22:47:59.61 9.7558 39.7437 1.306 2002-11-06 01:00:30.29 7.4493 38.6913 2.017 2002-11-06 01:58:19.59 9.8392 39.2843 1.402 2002-11-06 09:56:04.00 9.3953 40.0805 1.419 2002-11-06 16:44:34.41 9.8765 40.0062 2.491 2002-11-07 01:24:30.96 9.4952 40.0512 1.885 2002-11-07 03:47:15.79 10.0428 39.8077 1.529 2002-11-07 08:52:15.95 9.0365 40.7723 1.769 2002-11-08 08:38:45.11 9.2480 39.0413 1.780 2002-11-08 08:44:20.35 9.0605 38.6332 1.006 2002-11-08 11:58:06.79 9.7440 39.1817 1.556 2002-11-08 14:13:32.66 9.1495 40.8397 2.305 2002-11-08 21:03:09.60 9.4612 39.6817 1.309 2002-11-08 21:13:27.86 9.4048 39.5995 0.803 2002-11-09 08:24:18.70 9.0228 39.9327 1.735 2002-11-09 08:25:17.11 9.0153 39.9280 1.366 2002-11-09 14:21:17.42 9.2537 40.2088 1.606 2002-11-09 18:50:47.48 9.2533 40.2053 1.265 2002-11-09 23:38:11.21 9.6010 39.6623 0.689 2002-11-11 20:27:33.92 9.6413 39.7265 0.808 2002-11-11 20:33:50.44 9.2568 40.1930 0.682 2002-11-11 23:25:31.54 9.6728 39.8043 1.483 2002-11-11 23:50:02.29 9.3090 40.1938 0.134 2002-11-12 00:14:09.51 9.6387 38.1745 1.175 2002-11-12 00:39:49.51 11.2815 41.8433 3.305 2002-11-12 05:52:52.13 9.5575 39.8412 1.753 2002-11-12 08:04:10.51 8.9657 38.7623 1.016 2002-11-12 18:39:44.50 8.1632 39.1350 1.787 2002-11-12 23:16:20.34 8.9098 40.6257 1.333 2002-11-13 01:43:28.68 9.6378 39.8063 1.332 2002-11-13 14:39:44.57 9.3245 40.1692 1.010 2002-11-13 18:59:51.60 10.7562 39.5437 2.159 2002-11-14 00:19:24.98 9.5387 39.6277 0.835 2002-11-14 08:32:05.77 10.8357 41.2047 3.120 2002-11-14 16:46:38.77 9.8803 39.9980 2.477 2002-11-15 18:00:27.45 9.4952 40.0497 2.377 2002-11-15 21:30:43.71 10.0975 39.5678 0.830 2002-11-15 23:24:30.10 9.0570 39.9735 1.277 2002-11-16 01:36:33.86 9.8002 42.4710 2.913 2002-11-16 19:59:19.67 10.8652 40.2617 1.785 2002-11-17 01:17:31.76 9.4990 39.5947 0.759 2002-11-17 10:31:29.74 10.0970 40.0095 1.745 2002-11-18 00:50:54.77 10.7792 41.3563 2.752 2002-11-18 14:26:23.78 9.3822 38.3895 1.748 2002-11-18 20:17:56.51 9.0583 40.7642 0.966 2002-11-18 21:09:32.78 10.2518 39.9062 1.392 2002-11-18 22:00:20.72 9.0623 40.7620 1.159 2002-11-18 22:39:50.88 9.0827 40.7582 2.211 2002-11-18 22:52:41.02 9.0668 40.7633 0.971 2002-11-18 23:26:06.24 9.0500 40.8323 2.185 2002-11-18 23:26:37.84 9.2742 40.1888 1.920 2002-11-18 23:27:32.70 9.2977 40.1740 1.920 2002-11-18 23:35:44.86 9.0612 40.7677 1.074 2002-11-18 23:44:02.27 9.0640 40.7458 1.122 2002-11-19 00:15:39.44 9.3158 40.1992 1.786 2002-11-19 00:37:42.42 9.2953 40.2258 0.835 2002-11-19 00:41:44.87 9.3222 40.1573 0.786 2002-11-19 00:52:11.44 9.3028 40.2012 1.299 2002-11-19 00:56:00.00 9.3215 40.1518 0.607 2002-11-19 01:09:57 11.9100 44.1400 4.930 2002-11-19 01:21:54.42 9.3258 40.1550 0.940 2002-11-19 02:14:41.66 10.9792 42.1715 3.989 2002-11-19 02:25:29.34 9.2847 40.2310 0.674 2002-11-19 06:25:01.14 9.0630 40.7948 1.764 2002-11-19 08:24:09 11.8700 44.0400 5.020 2002-11-19 08:46:04.80 9.0043 38.7573 1.142 2002-11-19 20:22:01.21 7.4827 38.4957 1.391 2002-11-19 23:38:07.96 9.3005 40.1970 0.628 2002-11-19 03:45:23 11.8700 44.2100 4.760 2002-11-20 19:39:52.94 7.4068 38.5760 1.466

Appendices

130

2002-11-20 23:01:52.54 8.2382 39.0380 1.382 2002-11-20 23:11:36.81 8.2470 39.0458 0.788 2002-11-20 23:34:39.92 10.0312 39.2822 1.316 2002-11-21 15:37:42.64 9.0603 40.8032 1.809 2002-11-21 23:06:27.57 7.8665 39.0307 1.147 2002-11-22 13:32:35.53 8.8742 39.8503 2.023 2002-11-22 23:18:02.38 9.3320 39.8915 0.802 2002-11-23 04:25:05.66 9.4472 39.7055 1.483 2002-11-23 14:06:34.61 9.3128 40.1910 1.005 2002-11-24 15:42:28.37 9.3163 40.1898 1.128 2002-11-24 16:14:29.58 9.3208 40.1810 1.505 2002-11-24 16:25:10.24 9.3042 40.1923 1.576 2002-11-24 16:26:12.85 9.3162 40.1975 1.434 2002-11-24 16:46:25.32 9.3240 40.1830 1.712 2002-11-24 16:58:03.53 9.2967 40.2080 0.442 2002-11-24 17:12:42.27 9.2923 40.2128 0.171 2002-11-24 17:40:46.61 9.3803 40.1033 1.219 2002-11-24 18:21:42.18 9.3018 40.2022 0.388 2002-11-24 19:11:11.55 9.3227 40.1897 2.279 2002-11-24 19:16:37.11 9.3293 40.1862 0.938 2002-11-24 19:18:05.94 9.2995 40.2120 0.708 2002-11-24 19:22:51.83 9.3117 40.1933 0.809 2002-11-24 19:23:17.39 9.2805 40.1637 1.100 2002-11-24 19:35:28.90 9.3247 40.1890 1.233 2002-11-24 19:37:27.60 9.2685 39.8343 1.196 2002-11-24 19:40:16.51 9.3183 40.1840 1.717 2002-11-24 20:08:44.52 9.3480 40.1807 1.277 2002-11-24 20:19:21.26 9.3177 40.1998 0.729 2002-11-24 20:43:44.97 9.3080 40.1950 2.065 2002-11-24 20:52:47.14 9.3307 40.1760 1.653 2002-11-24 21:08:12.21 9.3190 40.1768 0.632 2002-11-24 21:49:53.23 9.4927 39.7820 1.125 2002-11-24 22:52:07.14 9.3258 40.1842 2.054 2002-11-24 23:08:35.30 9.4812 39.7917 1.551 2002-11-24 23:25:23.67 9.3180 40.1722 0.575 2002-11-25 00:33:55.07 9.3117 40.1917 0.358 2002-11-25 00:35:17.23 9.3122 40.1918 0.618 2002-11-25 00:51:25.95 9.2950 40.2128 0.518 2002-11-25 02:05:53.12 9.0205 40.7653 0.634 2002-11-25 02:31:55.07 10.7637 39.7623 2.269 2002-11-25 15:49:30.67 9.4533 40.0120 1.697 2002-11-25 21:01:51.15 9.3047 40.2167 0.524 2002-11-26 01:10:24.52 9.3213 40.1823 0.915 2002-11-26 02:16:35.29 9.3148 40.1925 0.218 2002-11-26 05:22:32.74 10.2680 39.9073 2.443 2002-11-26 11:07:26.30 9.2523 38.9018 0.208 2002-11-26 20:39:13.41 9.1743 40.0042 1.492 2002-11-26 21:30:27.32 9.1772 40.0098 2.082 2002-11-26 22:43:21.41 7.5050 38.6637 1.297 2002-11-26 23:47:20.68 9.1787 40.0222 1.245 2002-11-27 20:04:03.41 9.2993 40.2250 0.564 2002-11-27 20:05:07.43 9.2952 40.2282 0.689 2002-11-27 21:07:37.17 9.3003 40.2150 0.359 2002-11-27 22:39:24.21 10.3227 39.6252 1.384 2002-11-28 02:12:06.73 9.3198 40.1867 1.319 2002-11-28 02:29:49.86 9.2967 40.2068 0.317 2002-11-28 06:02:34.96 9.3817 39.9030 2.077 2002-11-28 06:19:46.29 9.3737 39.8887 2.125 2002-11-28 08:53:51.66 9.4033 39.9040 2.364 2002-11-28 08:57:17.92 9.3700 39.8833 1.494 2002-11-28 18:19:16.33 9.4050 39.9053 2.436 2002-11-28 20:10:46.09 9.2945 40.2185 0.013 2002-11-29 01:24:36.39 9.3873 39.8963 1.455 2002-11-29 01:25:19.84 9.3607 39.8860 1.363 2002-11-29 20:07:39.86 9.9940 37.6080 1.107 2002-11-30 04:08:49.71 10.0220 39.2817 1.560 2002-11-30 07:58:46.62 9.3992 39.9203 1.704 2002-11-30 09:26:06.86 8.1638 39.1442 2.517 2002-11-30 18:22:11.94 9.2950 40.2108 0.153 2002-11-30 20:51:27.74 9.5143 39.8937 1.274 2002-11-30 22:11:02.35 9.4153 39.7062 1.268 2002-11-30 22:32:46.20 9.2680 40.2358 0.607 2002-11-30 22:53:07.76 8.1693 39.1260 0.491 2002-11-30 23:53:53.18 9.4528 39.6910 1.012 2002-12-01 07:34:29.87 9.3003 40.2048 0.507 2002-12-01 10:48:31.42 9.3008 40.2073 0.498 2002-12-01 11:18:18.85 13.0030 40.1447 5.197 2002-12-01 11:18:32 12.2800 39.7400 4.810 2002-12-01 12:09:45.70 12.7647 39.8018 4.735 2002-12-01 14:36:32.76 9.8725 38.5030 1.917 2002-12-01 15:57:25.83 12.5472 39.8560 4.195 2002-12-02 01:01:59.10 9.4058 39.8965 1.189 2002-12-02 07:38:40.62 11.6337 40.6212 3.725 2002-12-02 10:14:28.39 9.5562 39.6802 1.117 2002-12-02 16:38:50.16 9.5137 39.6328 1.125 2002-12-02 18:08:37.54 7.4288 38.7028 1.832 2002-12-02 18:29:16.81 9.4465 39.6948 1.602 2002-12-03 00:58:57.49 7.4175 38.7043 2.034 2002-12-03 02:57:42.82 13.0205 39.5148 3.211 2002-12-03 03:53:06.99 13.2507 40.3548 3.714 2002-12-03 16:02:52.50 7.4677 38.5692 2.547 2002-12-03 16:36:16.39 8.3113 38.3683 2.223 2002-12-03 19:38:26.33 9.4162 39.3188 0.420 2002-12-03 20:10:01.52 7.7043 38.9212 2.341

Appendices

131

2002-12-03 22:13:59.17 9.9408 37.9357 1.446 2002-12-03 22:55:57.82 9.4545 39.6818 1.434 2002-12-03 23:20:36.73 7.5978 38.8033 1.366 2002-12-03 23:54:46.17 7.5742 38.7928 1.875 2002-12-04 13:41:06.71 8.8727 39.8352 1.974 2002-12-04 18:58:37.49 9.5702 39.7815 0.993 2002-12-05 17:38:29.87 9.3027 40.2138 0.652 2002-12-05 20:23:53.51 9.4455 39.6962 1.246 2002-12-05 20:24:46.22 9.4468 39.6708 0.915 2002-12-06 00:47:29.74 10.4980 39.7915 1.688 2002-12-06 01:42:09.93 9.0793 39.3518 1.771 2002-12-06 03:28:38.58 9.5512 40.7083 1.112 2002-12-06 08:21:02.52 9.0967 38.5305 1.059 2002-12-07 11:24:43.94 9.9297 41.0435 2.690 2002-12-07 18:40:57.30 9.3082 40.1918 0.781 2002-12-07 22:56:04.21 11.7032 39.7447 3.116 2002-12-08 18:56:33.67 7.5087 38.6593 2.633 2002-12-09 01:41:58.89 9.3190 40.1960 0.477 2002-12-09 10:29:48.40 10.7493 39.7477 3.254 2002-12-10 02:39:47.38 9.4178 39.3650 1.231 2002-12-10 08:06:23.98 9.4238 39.6225 1.429 2002-12-10 08:16:26.83 9.4277 39.6345 1.500 2002-12-11 22:20:10.81 13.7403 39.8545 3.887 2002-12-11 23:01:33.08 9.4698 39.6842 1.404 2002-12-12 01:40:13.26 9.4875 40.0258 1.595 2002-12-12 02:23:30.30 9.5628 39.6560 0.998 2002-12-12 09:18:11.58 8.9660 38.7780 1.056 2002-12-12 09:22:31.68 9.0577 40.0635 1.021 2002-12-12 17:02:04.02 9.3857 40.0400 1.630 2002-12-12 17:24:10.87 8.9478 39.1713 1.643 2002-12-12 17:49:16.11 8.9060 39.7817 1.391 2002-12-12 21:23:04.48 8.2368 39.2083 0.951 2002-12-13 09:00:47.73 8.9720 38.7780 1.048 2002-12-13 17:36:22.22 9.4432 40.0248 2.196 2002-12-13 22:54:00.20 9.4832 39.9747 1.029 2002-12-13 23:08:59.63 11.9642 39.5197 2.592 2002-12-15 08:37:35.47 7.4277 38.7117 3.063 2002-12-15 10:06:06.01 9.5792 39.7260 1.804 2002-12-15 16:29:11.56 9.0565 40.7662 1.402 2002-12-15 19:15:39.08 7.4235 38.7173 2.886 2002-12-15 20:04:08.72 7.4232 38.7163 3.144 2002-12-15 20:07:04.70 7.4262 38.7088 2.591 2002-12-15 20:14:53.03 7.4117 38.7252 2.073 2002-12-15 20:18:51.34 7.3567 38.7405 3.065 2002-12-15 20:35:05.22 9.5522 40.1488 1.932 2002-12-15 21:18:38.01 7.4247 38.7085 2.580 2002-12-15 21:39:09.28 7.3735 38.7317 1.721 2002-12-15 23:17:12.47 7.4125 38.7258 1.833 2002-12-16 00:27:01.90 7.4125 38.7352 1.424 2002-12-16 01:57:03.25 6.8895 38.6420 2.291 2002-12-16 16:28:11.58 9.3043 40.1960 0.554 2002-12-16 16:28:23.11 9.2918 40.2173 0.556 2002-12-16 16:29:21.28 9.2903 40.2212 0.769 2002-12-16 19:35:04.79 9.6325 38.4758 1.107 2002-12-17 00:28:39.46 9.4710 40.0033 2.291 2002-12-17 00:34:35.52 9.4607 39.9843 0.940 2002-12-17 21:43:08.40 9.0045 39.9062 1.149 2002-12-17 22:02:28.66 9.0205 39.9103 0.881 2002-12-17 22:12:35.75 9.0095 39.9058 1.400 2002-12-17 23:15:10.61 8.9987 39.9052 1.546 2002-12-18 01:31:41.14 8.9575 39.8837 0.329 2002-12-18 08:15:30.35 8.9525 38.7783 0.997 2002-12-18 18:04:58.52 7.3987 38.7593 2.342 2002-12-19 00:08:26.10 9.3167 40.1927 0.316 2002-12-20 13:27:42.13 8.8840 39.8412 2.049 2002-12-20 22:28:10.97 10.0295 39.2943 1.731 2002-12-21 09:47:10.45 9.1435 39.0735 1.246 2002-12-21 19:29:23.71 7.4178 38.7052 1.602 2002-12-21 19:39:38.41 7.3808 38.7670 1.507 2002-12-21 22:00:21.29 9.2898 40.2433 1.031 2002-12-21 22:00:39.02 9.3023 40.1978 0.696 2002-12-21 23:34:44.79 7.4042 38.7142 1.351 2002-12-21 23:43:26.92 9.8803 39.0697 1.344 2002-12-22 02:07:29.90 9.4888 38.3662 0.733 2002-12-22 11:14:08.44 9.1562 39.9868 1.513 2002-12-22 22:36:10.81 7.3783 38.7847 1.285 2002-12-23 00:05:53.04 9.4348 39.6802 1.389 2002-12-23 00:47:28.12 9.0160 39.9390 0.738 2002-12-23 01:05:50.40 9.4355 39.6638 1.056 2002-12-23 06:27:49.96 9.4405 39.6802 2.448 2002-12-23 07:52:51.21 9.4513 39.6145 1.162 2002-12-23 11:04:17.22 9.3962 39.6913 2.023 2002-12-23 12:04:54.69 9.4505 39.5990 1.166 2002-12-23 12:40:43.45 8.8678 39.8288 1.479 2002-12-23 14:51:37.20 9.4615 40.0045 2.213 2002-12-23 15:46:17.57 9.4467 39.6908 2.148 2002-12-23 22:52:01.99 9.4537 39.6792 0.902 2002-12-24 01:36:08.24 10.5980 39.7855 2.583 2002-12-24 02:48:24.01 9.5298 39.6498 1.078 2002-12-24 04:42:06.75 9.4400 39.6787 2.309 2002-12-25 01:10:55.64 9.3427 40.1885 1.066 2002-12-25 01:12:15.65 9.1780 40.2733 0.883 2002-12-25 02:01:37.48 9.4297 39.6328 1.164 2002-12-25 02:05:51.96 9.4443 39.6898 1.331

Appendices

132

2002-12-25 10:53:37.96 8.9905 38.7660 1.232 2002-12-25 17:36:57.96 9.3140 40.1922 0.634 2002-12-25 20:32:28.52 7.3837 38.7032 1.937 2002-12-26 19:47:52.03 9.2187 40.0227 3.167 2002-12-26 19:55:17.94 9.2080 40.0145 2.409 2002-12-26 19:58:00.77 9.2030 40.0165 1.902 2002-12-26 20:17:00.50 9.2148 40.0043 0.891 2002-12-26 20:17:20.21 9.2175 40.0067 1.375 2002-12-26 20:43:26.13 9.2115 40.0302 1.232 2002-12-26 20:55:39.35 9.2053 40.0183 1.654 2002-12-26 21:18:53.97 9.2098 40.0038 1.297 2002-12-26 21:21:39.61 9.1920 40.0318 0.925 2002-12-26 21:22:40.25 9.1985 40.0232 1.154 2002-12-26 21:26:53.22 9.1972 40.0362 1.006 2002-12-26 21:59:17.44 9.2073 40.0300 1.346 2002-12-27 00:03:43.51 7.3535 38.7765 1.697 2002-12-27 02:11:37.51 8.6760 39.5695 1.499 2002-12-27 15:31:13.71 9.1618 40.0492 1.503 2002-12-27 15:33:21.02 9.4342 39.6473 1.529 2002-12-27 17:59:06.61 9.3463 41.1755 2.133 2002-12-27 21:06:50.14 9.2178 39.9113 1.422 2002-12-27 21:48:25.74 8.9280 39.4487 0.679 2002-12-27 22:34:09.41 9.3137 40.1938 0.786 2002-12-27 22:38:03.52 9.3030 40.2050 0.264 2002-12-28 00:56:35.21 10.7395 39.7878 2.365 2002-12-28 23:53:46.41 9.3930 39.6685 1.667 2002-12-29 08:33:54.80 8.9558 38.8313 0.748 2002-12-30 07:19:57.49 9.4778 39.2175 1.479 2002-12-31 23:42:32.59 7.3518 38.6890 1.380 2003-01-01 07:45:26.37 10.1123 38.5638 1.453 2003-01-01 20:31:42.06 9.9210 37.9790 1.391 2003-01-01 21:37:55.02 9.4372 39.5972 1.451 2003-01-01 23:36:37.29 11.5872 41.9362 4.668 2003-01-01 23:36:45 11.2000 41.7500 4.520 2003-01-02 06:24:26.59 10.4027 40.8593 2.996 2003-01-02 07:02:09.43 8.8505 38.5288 1.945 2003-01-02 08:52:45.28 9.2343 40.0212 2.371 2003-01-02 10:18:18.48 8.9192 38.7645 1.151 2003-01-02 21:44:56.45 9.4693 39.6172 1.773 2003-01-02 22:41:56.54 9.7635 39.1167 1.995 2003-01-03 01:45:31.73 9.4410 39.6760 1.898 2003-01-03 03:05:42.14 10.4098 39.8330 2.373 2003-01-03 05:17:25.55 10.2028 41.2275 2.681 2003-01-03 23:26:08.55 9.2867 40.2175 0.070 2003-01-05 16:46:36.55 9.1512 38.9375 2.083 2003-01-05 21:54:52.19 8.0893 39.0567 1.185 2003-01-06 22:51:06.60 9.3167 40.1697 0.798 2003-01-07 01:19:51.11 10.5607 39.7348 2.502 2003-01-07 02:00:00.09 9.2897 40.2228 0.753 2003-01-07 06:53:28.32 9.6712 39.7022 2.096 2003-01-07 07:53:27.16 9.4347 39.6090 1.708 2003-01-07 16:09:52.12 9.2840 40.2230 0.259 2003-01-07 23:29:50.43 9.2855 40.1627 0.219 2003-01-07 23:55:48.37 9.3053 40.1450 0.232 2003-01-08 00:42:29.47 9.2142 40.2998 0.726 2003-01-08 00:52:56.76 9.0302 39.9293 0.932 2003-01-08 01:21:58.59 9.2490 40.2017 0.855 2003-01-08 16:49:33.05 9.2482 40.2063 1.147 2003-01-08 18:15:29.94 9.2920 40.1650 0.905 2003-01-08 20:24:09.65 9.6745 39.7047 1.831 2003-01-08 21:59:07.37 9.6333 39.7037 1.286 2003-01-09 00:37:13.64 9.6498 39.6738 0.898 2003-01-09 02:01:11.14 9.2275 42.1173 2.516 2003-01-09 05:47:34.99 9.5827 39.6488 1.331 2003-01-09 12:38:16.42 8.6690 40.0582 1.592 2003-01-09 20:36:57.97 9.7260 39.7643 3.113 2003-01-09 22:39:25.65 10.5750 39.8335 2.527 2003-01-10 12:13:56.13 8.6125 39.4472 3.435 2003-01-10 22:50:26.98 9.1712 39.7788 1.756 2003-01-11 14:00:21.28 9.7102 39.7368 2.136 2003-01-11 23:57:23.98 9.3017 40.1822 0.231 2003-01-12 00:40:11.95 9.4073 39.6535 1.346 2003-01-12 07:03:38.79 8.8023 38.9638 1.415 2003-01-13 12:03:23.15 9.4963 38.4185 1.403 2003-01-13 21:06:00.73 9.4922 39.6808 1.969 2003-01-13 22:15:29.97 9.1012 39.9322 1.125 2003-01-14 20:02:21.04 5.4248 37.4658 3.904 2003-01-15 21:57:10.80 9.6438 38.5377 1.573 2003-01-15 23:14:16.95 9.6502 38.5002 0.919 2003-01-16 08:52:39.62 8.9903 38.8217 1.331 2003-01-16 12:02:08.98 9.3658 39.6758 2.054 2003-01-16 14:22:50.19 9.2030 39.9437 1.552 2003-01-16 17:31:14.46 9.4382 39.6807 2.092 2003-01-17 18:53:49.10 9.4572 39.6835 1.642 2003-01-17 18:54:45.02 9.4535 39.6687 1.633 2003-01-17 18:56:30.34 9.4432 39.6898 2.263 2003-01-17 19:16:08.83 9.4462 39.6863 1.962 2003-01-17 20:14:43.69 9.4192 39.6847 1.337 2003-01-17 21:22:42.35 9.4685 39.7075 1.172 2003-01-17 21:32:54.04 7.6113 38.7727 2.027 2003-01-17 23:44:36.51 9.4265 39.6867 1.413 2003-01-18 01:20:36.40 9.4878 39.9888 1.294 2003-01-18 02:45:37.29 7.8930 38.8210 1.889 2003-01-18 11:45:59.57 9.4312 39.6882 2.549

Appendices

133

2003-01-18 19:55:23.91 9.4652 39.6728 1.728 2003-01-19 23:21:55.25 10.2570 41.1270 3.077 2003-01-20 21:07:23.72 7.4668 38.8208 3.885 2003-01-20 21:16:22.91 7.4752 38.8233 2.523 2003-01-20 23:39:29.94 8.8852 40.6385 0.866 2003-01-21 08:08:18.83 7.4922 38.8208 2.901 2003-01-22 02:33:42.68 9.4342 39.6890 2.452 2003-01-22 02:34:08.05 9.4107 39.6960 2.569 2003-01-22 07:51:30.37 8.7133 38.5560 2.081 2003-01-22 08:01:24.74 8.7497 38.8413 2.141 2003-01-22 10:24:49.20 9.0530 39.9697 1.428 2003-01-22 21:54:51.77 9.4625 40.0062 2.059 2003-01-23 02:08:59.81 9.4425 39.6820 2.065 2003-01-25 18:26:59.48 4.9363 37.5817 4.512 2003-01-27 09:42:37.48 11.4227 39.4412 2.972

Appendices

134

Appendix C Publications in peer reviewed journals resulting from Ph.D. research The following publications in peer reviewed journals resulted from the author’s research

during the course of the Ph.D. First authored articles that have been published are

highlighted in the list below and full texts presented with the pagination independent of

the thesis body.

Kendall, J-M., G. W. Stuart, C. J. Ebinger, I. D. Bastow and D. Keir, Magma assisted

rifting in Ethiopia, Nature, 433, 146 - 148, 2005.

Keir, D., J-M. Kendall, C. J. Ebinger and G. W. Stuart, Variations in late syn-rift

melt alignment inferred from shear-wave splitting in crustal earthquakes beneath the Ethiopian rift, Geophys. Res. Lett., 32, L23308,

doi:10.1029/2005GL024150, 2005. Casey, M., C. Ebinger, D. Keir, R. Gloaguen and F. Mohammed, Strain

accommodation in transitional rifts: Extension by magma intrusion and faulting

in Ethiopian rift magmatic segments, in G. Yirgu, C. Ebinger and P. Maguire,

The Afar Volcanic Province within the East African Rift System, Geol. Soc.

Lond. Spec. Pub., 259, 143-164, 2006.

Kendall, J-M. S. Pilidou, D. Keir, I. Bastow, A. Ayele, and G. W. Stuart, Mantle

upwellings, melt migration and the rifting of Africa: Insights from seismic

anisotropy, in G. Yirgu, C. Ebinger and P. Maguire, The Afar Volcanic Province

within the East African Rift System, Geol. Soc. Lond. Spec. Pub., 259, 55-72

2006.

Ayele, A., G. W. Stuart, I. D. Bastow and D. Keir, The August 2002 earthquake

sequence in north Afar: Insights into the neotectonics of the Danakil microplate,

J. Afr. Earth Sci., in press (2006).

Appendices

135

Keir, D., C. J. Ebinger, G. W. Stuart, E. Daly and A. Ayele, Strain accommodation by magmatism and faulting as rifting proceeds to breakup: Seismicity of the northern Ethiopian rift, J. Geophys. Res., 111, B05314,

doi:1029/2005JB 003748, 2006.

Wright, T., C. J. Ebinger, J. Biggs, A. Ayele, G. Yirgu, D. Keir and A. Stork, Magma-

maintained rift segmentation at continental rupture in the 2005 Afar dyking

episode, Nature, 442, doi:10.1038/nature04978, 2006.

Keir, D., G. W. Stuart, A. Jackson and A. Ayele, Local earthquake magnitude scale and seismicity rate of the northern Ethiopian rift, Bull. Seism. Soc.

Am., 96, doi:10.1785/0120060051, 2006.

Daly, E., D. Keir, C. Ebinger, G. Stuart, I. Bastow, and A. Ayele, Crustal tomographic

imaging of a transitional continental rift: the Ethiopian Rift, submitted to

Geophys. J. Int., 2006.

Variations in late syn-rift melt alignment inferred from shear-wave

splitting in crustal earthquakes beneath the Ethiopian rift

Derek Keir,1 J-M. Kendall,2,3 C. J. Ebinger,1 and G. W. Stuart2

Received 21 July 2005; revised 12 October 2005; accepted 25 October 2005; published 8 December 2005.

[1] The northern Main Ethiopian rift (MER) marks thetransition from continental rifting to incipient seafloorspreading. We constrain anisotropy of the upper-crust in theMER and its uplifted rift flanks using shear-wave splittingfrom 24 earthquakes located beneath 18 broadband stations.Along the axis of the MER the fast polarization direction isoriented between N and NNE, parallel to Quaternary-Recent faults, aligned cones and maximum horizontalstress. Delay times are highest (0.24 s) where independentseismic studies show evidence of shallow partial melt. Weattribute anisotropy along the rift axis to aligned melt-filledmicro-cracks and dikes. At stations flanking the rift, the fastpolarization direction is oriented NE and delay-times aresmaller (0.04–0.14 s). The lower amount of anisotropy isconsistent with reduced melt away from the rift axis. Theseresults show melt-induced anisotropy persists into the crust,and magma injection localizes and accommodates strain justprior to continental break-up. Citation: Keir, D., J-M.

Kendall, C. J. Ebinger, and G. W. Stuart (2005), Variations in

late syn-rift melt alignment inferred from shear-wave splitting in

crustal earthquakes beneath the Ethiopian rift, Geophys. Res. Lett.,

32, L23308, doi:10.1029/2005GL024150.

1. Introduction

[2] Strain localizes as rifting proceeds to continentalbreakup, but the partitioning of strain between mechanicalfailure and magma injection remains controversial. Thevolcanically active northern Main Ethiopian rift (MER) istransitional between continental and incipient oceanic rift-ing [e.g., Ebinger and Casey, 2001], affording the opportu-nity to actively observe rift processes just prior to break-up.[3] The Miocene-Recent MER constitutes the northern

part of the East African rift system and forms the youngestarm of the Afar triple junction, which developed in theEocene-Oligocene flood basalt province (Figure 1, inset).The MER is bounded by NE-trending Miocene border faults.Since Quaternary times extensional strain is localized in<20 km-wide right stepping en-echelon magmatic segmentswhich are zones of NNE-striking fissures, faults and alignedvolcanic cones [Bilham et al., 1999; Ebinger and Casey,2001]. These magmatic segments are the locus of seismicityand magmatism (D. Keir et al., Strain accommodation bymagmatism and faulting as rifting proceeds to breakup:Seismicity of the northern Ethiopian rift, submitted to

Journal of Geophysical Research, 2005, hereinafter referredto as Keir et al., submitted manuscript, 2005) (Figure 1). Thecurrent extension direction is N105E [Wolfenden et al.,2004; Keir et al., submitted manuscript, 2005].[4] Anisotropy provides further constraints on style of

rifting and breakup. SKS-splitting dominantly reflects up-per-mantle anisotropy, and measurements in the MER showa rift-parallel (NNE) fast anisotropic orientation thatparallels the aligned eruptive centers, fissures and activefaults. The magnitude of splitting and cross-rift variation inthe orientation of the fast S-wave were used to propose thatpartial melt beneath the MER rises through melt-filledcracks that penetrate the thinned lithosphere [Kendall etal., 2005a]. Sv and Sh velocity models determined frominversion of surface-wave dispersion curves show faster Svvelocities than Sh velocities below 20 km along the rift axis.The results are consistent with anisotropy at 20–75 kmdepth due to oriented melt-filled pockets [Kendall et al.,2005b]. Bastow et al. [2005] show, by comparing P-andS-wave relative arrival-time data, that upper mantle lowvelocity anomalies beneath the MER are likely due to high-temperatures and partial melt.[5] Anisotropy of the shallow crust is commonly attrib-

uted to micro-cracks vertically-oriented parallel to thedirection of maximum horizontal stress [e.g., Crampin,1994]. For example, crustal shear-wave splitting measure-ments in rift zones at the Mid-Atlantic ridge and in Icelandshow fast-polarization directions sub-parallel to the maxi-mum horizontal stress. These patterns are attributed toaligned parallel cracks and fractures in the uppermost 3–5 km of the crust [e.g., Barclay and Toomey, 2003; Evans etal., 1996; Menke et al., 1994]. S-wave anisotropy has alsobeen attributed to vertical micro-cracks throughout the crustin which case S-wave splitting is accrued along the wholeray-path [e.g., Volti and Crampin, 2003]. Fast-polarizationdirections at active volcanoes are usually parallel to dikesand the maximum horizontal stress, with 90 polarizationflips observed prior to volcanic eruption due to increasedpore pressure leading to changes in crack orientation [Millerand Savage, 2001]. Crustal anisotropy has also been linkedto other rock fabrics such as vertically dipping foliation ofmetamorphic basement [e.g., do Nascimento et al., 2004].We use measurements of S-wave splitting from local earth-quakes to study crustal anisotropy in the MER. We compareour results to independent studies and use this informationto evaluate mechanisms of deformation preceding continen-tal break-up.

2. Data and Methodology

[6] From October 2001 to January 2003, seismicity wasrecorded by 29 broadband instruments that covered a

GEOPHYSICAL RESEARCH LETTERS, VOL. 32, L23308, doi:10.1029/2005GL024150, 2005

1Department of Geology, Royal Holloway University of London,Egham, UK.

2School of Earth and Environment, University of Leeds, Leeds, UK.3Now at Department of Earth Sciences, University of Bristol, Bristol,

UK.

Copyright 2005 by the American Geophysical Union.0094-8276/05/2005GL024150$05.00

L23308 1 of 4

250 km 350 km area of the MER and adjacent upliftedplateau [Bastow et al., 2005]. A further 150 broadbandinstruments operated for the final 2-4 months of the exper-iment (Keir et al., submitted manuscript, 2005). Earth-quakes were located using a 3-D velocity modeldetermined from local earthquake tomography (Keir et al.,submitted manuscript, 2005). 75% of seismicity occurred inthe Fentale-Dofen magmatic segment and at the intersectionof the MER and Red Sea rift (Keir et al., submittedmanuscript, 2005). Due to this severe spatial clustering ofearthquakes, S-wave splitting measurements could only bemade at 10% of available seismic stations (Figure 1).24 earthquakes located beneath 18 stations provided 26

three-component seismograms where the S-wave incident-angle is within the shear-wave window (SWW). The SWWis the vertical cone bound by sin1 (Vs/Vp) where S-waveparticle motions are not disturbed by S-P conversions at thefree surface [Booth and Crampin, 1985]. We use a Vp/Vs of1.75, calculated from P- and S-wave travel-times fromearthquakes in the MER, which corresponds to a SWW thatis a cone within 35 of the vertical.[7] The polarization direction of the fast S-wave (f) and

the time delay between the fast and slow S-waves (dt) isdetermined using the method of Silver and Chan [1991],adapted for application to micro-earthquakes. In an isotropicradially stratified crust, near vertically impinging S-wavesshould exhibit linear particle motion. This phase is split intoorthogonally polarized fast and slow S-waves when ittravels through an anisotropic medium and this splittingproduces an elliptical particle motion. To remove the effectsof the anisotropy we rotate the horizontal components by fand shift their relative positions by dt, thereby linearizingthe particle motion (Figure S1, auxiliary material1). Toestimate the splitting we search for the correction parame-ters that best linearize the S-wave motion. An F-test isperformed to assess the uniqueness of the estimated splittingparameters and thereby produce an error estimate [Silver,1996]. The splitting parameters are well constrained. We usea cut off error criteria of ± 0.03 s for dt and 9 for f(Table S1, auxiliary material).

3. Results

[8] S-wave splitting measurements from local earth-quakes near the MER show large spatial variation in bothf and dt (Figure 1). At stations on the NW plateau f variesbetween 36 and 70. dt varies between 0.04 s and 0.14 sfor earthquakes that occurred at depths of 12–20 km anddt increases linearly with increased ray-path distance(Figure 2), showing that the crust is anisotropic to at least20 km depth. This equates to fairly uniform anisotropy of1.1 % on average, if splitting is assumed to be accrued overthe full ray-path length (Figures 1 and 2).[9] Along the Ankober fault system f is oriented N,

parallel to seismically active faults (Figure S1). dt is 0.1–0.16 s, equivalent to 2.2–3.6 % S-wave anisotropy.[10] At stations along the rift axis f is mostly oriented

N to NNE (Figure 1). Delay times are 0.06–0.24 s forearthquakes that are 6–9 km deep, equating to 3–6.2 %anisotropy (Figure 2). The largest values of dt (0.19–0.24 s,anisotropy of 5.4–6.2 %) are recorded at stations 1219 andBORE, both in the Quaternary Boset-Kone magmatic seg-ment (Figures 1 and 2).

4. Discussion

[11] Near-vertically propagating S-waves from localearthquakes near the MER show clear evidence of S-wavesplitting. The anisotropy is thus most likely due to folia-tions, cracks or inclusions aligned by regional and localstresses in the crust. The magnitude and orientation of theshear-wave splitting varies dramatically across the EAGLEnetwork, suggesting a heterogeneous stress field or varia-

Figure 1. Crustal anisotropy measurements at 18 broad-band stations in Ethiopia. White arrows show the polariza-tion of the fast S-wave (f) and arrow length is scaled by %anisotropy. Solid black lines with dip ticks are Mioceneborder faults (BF) and dashed lines are monoclines.Quaternary magmatic segments (MS) are shaded grey. Darkarrows show the extension direction and orientation of theminimum horizontal stress (Keir et al., submitted manu-script, 2005). The position of the along-axis profile forFigures 2b and 2d is shown by the black line. Top left inset:Topographic map of the MER, adjacent plateau and Afardepression. NP: Nubia Plate, SP: Somali Plate, DP: DanakilPlate, AP: Arabian Plate, RS: Red Sea, GA: Gulf of Aden.Top right inset: White arrows show polarization of fastS-wave and arrow length is scaled by delay-time.

1Auxiliary material is available at ftp://ftp.agu.org/apend/gl/2005GL024150.

L23308 KEIR ET AL.: VARIATIONS IN LATE SYN-RIFT MELT ALIGNMENT L23308

2 of 4

tions in the underlying cause of anisotropy. We calibrate ourresults with independent geological and seismic studies inthe MER.[12] Stations along the rift axis show relatively large

amounts of splitting despite shallower earthquake depths(6–9 km). Up to 0.24 s of splitting is observed beneathBoset-Kone magmatic segment, which equates to over 6 %anisotropy. Stations within the rift valley but located outsidemagmatic segments show less splitting (e.g. MELE), but theaverage magnitude of splitting in the rift valley is still nearly3%, much larger than beneath the NW plateau. The N toNNE orientation of f in the magmatic segments is parallelto Quaternary faults and aligned volcanic cones. The alongaxis variation of f correlates well with local changes in thestrike of maximum horizontal stress axes from focal mech-anisms of earthquakes within 25 km of splitting measure-ments (Figure 2) (Keir et al., submitted manuscript, 2005).[13] The largest amounts of upper-crustal anisotropy are

in the Quaternary magmatic segments where independentstudies show evidence of pervasive dike intrusion and thepresence of partial melt in shallow magma chambers.

Mackenzie et al. [2005] and Keranen et al. [2004] interpretcooled mafic intrusions in the mid-crust beneath thesemagmatic segments using models derived from wide-anglerefraction data and controlled source tomography respec-tively. The magnitude of splitting under Boset volcano isespecially pronounced, where melt-related anomalies havebeen interpreted in magnetotelluric data [Whaler andHautot, 2005]. The S-wave splitting observations are con-sistent with anisotropy due to vertically aligned magmaintrusions or melt-filled cracks beneath the Quaternarymagmatic segments, where the majority of strain is accom-modated by dike injection (Keir et al., submitted manu-script, 2005).[14] The deepest earthquakes lie beneath the largely un-

extended NW Ethiopian plateau, where we observe anincrease in delay time with increased ray-path length usingS-wave splitting measurements at different stations. Thesevariations in delay-times can be explained by relativelyuniform anisotropy that extends to at least 20 km depth;larger delay-times (0.1–0.14 s) at stations 1018, 1030 andINEE are caused by splitting accrued over longer ray-paths

Figure 2. (a) Rift-perpendicular profile of station averaged delay time (dt) versus distance from the rift axis. The two solidlines shows the position of magmatic segments and the dashed line shows the approximate position of the western boundaryof the rift valley. (b) Rift-parallel profile of station averaged dt at stations within 20 km of the along rift-axis line on Figure1. (c) Rift perpendicular profile of % anisotropy versus distance from the rift axis. (d) Rift-parallel profile of % anisotropyversus distance along the rift valley. (e) Individual measurements of dt versus S-wave ray-path length at stations on thewestern Ethiopian plateau. The dashed line is the best straight line fit to the data. (f) Individual measurements of dt versusray-path length at stations in the rift valley. (g) f against the average orientation of maximum horizontal stress axes of focalmechanismswithin 25 km of the splittingmeasurement. The dashed line is the best straight line fit to the data. The symbols are:white squares = plateau stations; grey triangles = stations at the Ankober fault; inverted triangles = stations in the MER butoutside magmatic segments; dark grey circles = stations in magmatic segments.


3 of 4

(Figures 1 and 2). Alternatively, these patterns may becaused by lateral variations in anisotropy of the uppermostfew kilometers with larger upper crustal anisotropy atstations 1018, 1030 and INEE. However, controlled sourceseismic images of underplating [Mackenzie et al., 2005],mid-crustal conductive anomalies in MT data [Whaler andHautot, 2005], and Quaternary eruptive centers as far northas Lake Tana all infer the presence of melt in the lower crustbeneath the Ethiopian plateau. Given these independentobservations, we interpret the data to show that meltinduced anisotropy extends to at least 20 km subsurface.The amount of crustal anisotropy beneath the plateau is low(1.1%), consistent with melt decrease away from the riftaxis. Splitting at stations on the plateau is oriented NE andmay also indicate a contribution from pre-existing basementfoliation or structural trends. Where exposed, Pan-Africanbasement foliation and Proterozoic ophiolite belts predom-inantly strike N to NE [e.g., Berhe, 1990; Kazmin et al.,1978]. These have been used to infer a NE-SW trendingsuture [Berhe, 1990] but due to limited basement outcroptheir interpretation is controversial [Church et al., 1991]. NEto ENE oriented basement structures are evident in regionaldrainage patterns along the Ambo fault, which has beenreactivated in Miocene rifting [Abebe et al., 1998]. Beneaththe Ethiopian plateau the crustal anisotropy may be due to acombination of mechanisms associated with aligned melt,pre-existing basement foliation and structural trends.[15] The patterns of shear-wave splitting observed in

earthquakes beneath both the rift valley and nearby plateauare most simply explained by crustal anisotropy related tovariable amounts of melt pocket alignment, with a higherdegree of magma intrusion in the crust beneath the rift. Ourstudy shows that melt-induced anisotropy at 20–75 kmdepth [Bastow et al., 2005; Kendall et al., 2005a, 2005b]continues into the uppermost crust, thereby penetrating theentire plate and facilitating continental breakup.

5. Conclusions

[16] Along the rift-axis the orientation of the fast S-waveis N to NNE, parallel to Quaternary to Recent faults,aligned cones and the current maximum horizontal stressaxis. The largest amounts of upper crustal anisotropy are inthe Quaternary magmatic segments, where the majority ofstrain is accommodated by magma injection; anisotropy ismost likely caused by aligned melt-filled micro-cracks anddikes. The low amount of anisotropy beneath the Ethiopianplateau is consistent with melt decrease away from the riftaxis. These results suggest the anisotropy is related tovariable amounts of melt pocket alignment in the crust,with a higher degree of dike intrusion in a narrow zone ofQuaternary magmatism. Melt-induced anisotropy extendsfrom the base of the lithosphere to the upper crust, suggest-ing that magma injection helps localize and facilitateextension just prior to continental breakup.

[17] Acknowledgments. We thank SEIS-UK and A. Brisbourne forthe use of instruments and assistance in the field. The support provided byA. Ayele and L. Asfaw of the Geophysical Observatory, Addis AbabaUniversity is much appreciated. The input provided by E. Daly, I. Bastow,D. Cornwell, P. Maguire and K. Whaler is also gratefully acknowledged.We thank Stuart Crampin and the anonymous reviewer who helped improvethis manuscript. This research was supported by NERC grant NER/A/S/2000/01004 and NERC studentship NER/S/A/2002/10547.

ReferencesAbebe, T., F. Mazzarini, F. Innocenti, and P. Manetti (1998), The Yerer-Tullu Wellel volcanotectonic lineament: A transtensional structure incentral Ethiopia and the associated magmatic activity, J. Afr. Earth Sci.,26(1), 135–150.

Barclay, A. H., and D. R. Toomey (2003), Shear wave splitting and crustalanisotropy at the Mid-Atlantic Ridge, 35N, J. Geophys. Res., 108(B8),2378, doi:10.1029/2001JB000918.

Bastow, I. D., G. W. Stuart, J-M. Kendall, and C. J. Ebinger (2005), Uppermantle seismic structure in a region of incipient continental breakup:Northern Ethiopian rift, Geophys. J. Int., 162, 479–493.

Berhe, S. M. (1990), Ophiolites in Northeast and East Africa: Implicationsfor Proterozoic crustal growth, J. Geol. Soc., 147, 41–57.

Bilham, R., R. Bendick, K. Larson, P. Mohr, J. Braun, S. Tesfaye, andL. Asfaw (1999), Secular and tidal strain across the Main Ethiopian Rift,Geophys. Res. Lett., 26, 2789–2792.

Booth, D. C., and S. Crampin (1985), Shear-wave polarizations on a curvedwavefront at an isotropic free-surface, Geophys. J. R. Astron. Soc., 83,31–45.

Church, W. R., S. M. Berhe, M. G. Abdelsalam, and R. J. Stern (1991),Discussion of ophiolites in Northeast and East Africa: Implications forProterozoic crustal growth, J. Geol. Soc. London, 148, 600–606.

Crampin, S. (1994), The fracture criticality of crustal rocks,Geophys. J. Int.,118, 428–438.

do Nascimento, A. F., F. H. R. Bezerra, and M. K. Takeya (2004), DuctilePrecambrian fabric control of seismic anisotropy in the Acu dam area,northeastern Brazil, J. Geophys. Res., 109, B10311, doi:10.1029/2004JB003120.

Ebinger, C., and M. Casey (2001), Continental breakup in magmatic pro-vinces: An Ethiopian example, Geology, 29, 527–530.

Evans, J. R., G. R. Foulger, B. R. Julian, and A. D. Miller (1996), Crustalshear-wave splitting from local earthquakes in the Hengill triple junction,southwest Iceland, Geophys. Res. Lett., 23, 455–458.

Kazmin, V., A. Shifferaw, and T. Balcha (1978), The Ethiopian basement:Stratigraphy and possible manner of evolution, Geol. Rundsch., 67, 531–546.

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

Kendall, J-M., S. Pilidou, D. Keir, I. D. Bastow, A. Ayele, and G. W. Stuart(2005b), Mantle upwellings, melt migration and the rifting of Africa:Insights from seismic anisotropy, in Structure and Evolution of the EastAfrican Rift in the Afar Volcanic Province, edited by G. Yirgu,C. Ebinger, and P. Maguire, Geol. Soc. Spec. Publ., in press.

Keranen, K., S. L. Klemperer, R. Gloaguen, and EAGLE Working Group(2004), Three-dimensional seismic imaging of a protoridge axis in theMain Ethiopian rift, Geology, 32, 949–952.

Mackenzie, G. H., H. Thybo, and P. Maguire (2005), Crustal velocitystructure across the Main Ethiopian Rift: Results from 2–dimensionalwide-angle seismic modelling, Geophys. J. Int., 162, 994–1006.

Menke, W., B. Brandsdottir, S. Jakobsdottir, and R. Stefansson (1994),Seismic anisotropy in the crust at the mid-Atlantic plate boundary insouth-west Iceland, Geophys. J. Int., 119, 783–790.

Miller, V., and M. Savage (2001), Changes in seismic anisotropy aftervolcanic eruptions: Evidence from Mount Ruapehu, Science, 293,2231–2233.

Silver, P. G. (1996), Seismic anisotropy beneath the continents, Annu. Rev.Earth Planet. Sci., 24(385), 432.

Silver, P. G., and W. W. Chan (1991), Shear-wave splitting and subconti-nental mantle deformation, J. Geophys. Res., 96, 16,429–16,454.

Volti, T., and S. Crampin (2003), A four-year study of shear-wave splittingin Iceland: 2. Temporal changes before earthquakes and volcanic erup-tions, in New Insights Into Structural Interpretation and Modeling, editedby D. A. Nieuwland, Geol. Soc. Spec. Publ., 212, 135–149.

Whaler, K. A., and S. Hautot (2005), The electrical resistivity structure ofthe crust beneath the northern Ethiopian rift, in Structure and Evolution ofthe East African Rift in Afar Volcanic Province, edited by G. Yirgu,C. Ebinger, and P. Maguire, Geol. Soc. Spec. Publ., in press.

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 Planet. Sci. Lett., 224, 213–228.

C. J. Ebinger and D. Keir, Department of Geology, Royal Holloway

University of London, Egham TW20 0EX, UK. ([email protected])J-M. Kendall, Department of Earth Sciences, University of Bristol,

Bristol BS8 1RJ, UK.G. W. Stuart, School of Earth and Environment, University of Leeds,

Leeds LS2 9JT, UK.


4 of 4

Strain accommodation by magmatism and faulting as rifting

proceeds to breakup: Seismicity of the northern Ethiopian rift

Derek Keir,1 C. J. Ebinger,1 G. W. Stuart,2 E. Daly,3 and A. Ayele4

Received 24 March 2005; revised 1 December 2005; accepted 2 February 2006; published 26 May 2006.

[1] The volcanically active Main Ethiopian rift (MER) marks the transition fromcontinental rifting in the East African rift to incipient seafloor spreading in Afar. We usenew seismicity data to investigate the distribution of strain and its relationship withmagmatism immediately prior to continental breakup. From October 2001 to January2003, seismicity was recorded by up to 179 broadband instruments that covered a250 km 350 km area. A total of 1957 earthquakes were located within the network, aselection of which was used for accurate location with a three-dimensional velocity modeland focal mechanism determination. Border faults are inactive except for a cluster ofseismicity at the structurally complex intersection of the MER and the older Red Sea rift,where the Red Sea rift flank is downwarped into the younger MER. Earthquakes arelocalized to 20-km-wide, right-stepping en echelon zones of Quaternary magmatism andfaulting, which are underlain by mafic intrusions that rise to 8–10 km subsurface.Seismicity in these ‘‘magmatic segments’’ is characterized by low-magnitude swarmscoincident with Quaternary faults, fissures, and chains of eruptive centers. All but threefocal mechanisms show normal dip-slip motion; the minimum compressive stress isN103E, perpendicular to Quaternary faults and aligned volcanic cones. The earthquakecatalogue is complete above ML 2.1, and the estimated b value is 1.13 ± 0.05. Theseismogenic zone lies above the 20-km-wide intrusion zones; intrusion may triggerfaulting in the upper crust. New and existing data indicate that during continental breakup,intrusion of magma beneath 20-km-wide magmatic segments accommodates themajority of strain and controls the locus of seismicity and faulting in the upper crust.

Citation: Keir, D., C. J. Ebinger, G. W. Stuart, E. Daly, and A. Ayele (2006), Strain accommodation by magmatism and faulting as

rifting proceeds to breakup: Seismicity of the northern Ethiopian rift, J. Geophys. Res., 111, B05314, doi:10.1029/2005JB003748.

1. Introduction

[2] Strain localizes as rifting proceeds to continentalbreakup, but the partitioning of strain between faults andmagmatic intrusion remains controversial [e.g., Lister et al.,1986; Ebinger and Casey, 2001]. Models of continentalbreakup that assume purely mechanical stretching predictstrain localization along preexisting or new shear zones thatmay accommodate large displacements [e.g., Lister et al.,1986; Dunbar and Sawyer, 1989]. Magmatic processessuperposed on the mechanical deformation pose additionalcomplications to our understanding of continental breakup.The magma-assisted rifting model of Buck [2004] showsthat if a steady supply of magma is available, the release ofstress and overall decrease in lithospheric strength due to

diking will prevent the stress level reaching those requiredto activate the border faults of a rift zone. As a result, borderfaults become inactive and extension localizes to the zone ofdiking. Extension near the surface, where the brittle short-term rheology allows rapid fault slip, is accommodated by acombination of faulting and dike injection. Analysis ofseismicity in a volcanically active rift setting that is nearbreakup provides a means to study the pattern of strainlocalization and assess how strain is partitioned betweenfaulting and dike injection.[3] The seismically and volcanically active northern Main

Ethiopian rift (MER) and Afar rifts are virtually the onlyplaces worldwide where the transition between continentaland oceanic rifting is exposed on land. The multidisciplin-ary project EAGLE (Ethiopia Afar Geoscientific Litho-spheric Experiment) provides fundamental constraints oncrust and upper mantle structure beneath the MER, setwithin a strong regional tectonic framework [e.g., Maguireet al., 2003; WoldeGabriel et al., 1990; Wolfenden et al.,2004]. The MER is thus an ideal natural laboratory to studycontinental breakup processes.[4] The EAGLE network, the largest array of seismic

instruments yet deployed on the African continent, is usedto analyze the distribution of local earthquakes in this

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 111, B05314, doi:10.1029/2005JB003748, 2006

1Department of Geology, Royal Holloway University of London,Egham, UK.

2School of Earth and Environment, University of Leeds, Leeds, UK.3Department of Earth and Ocean Sciences, National University of

Ireland, Galway, UK.4Geophysical Observatory, Addis Ababa University, Addis Ababa,

Ethiopia.

Copyright 2006 by the American Geophysical Union.0148-0227/06/2005JB003748$09.00

B05314 1 of 17

transitional rift zone [Maguire et al., 2003] (Figure 1). Ourseismicity study aims to evaluate the accommodation ofstrain by faulting and magmatic processes in the MER. Weuse accurately located hypocenters to map out variations inthickness of the seismogenic layer. Patterns of seismicity arethen compared with the distribution of Quaternary faultsand magmatism to distinguish between models for strainlocalization prior to continental breakup. Earthquake focalmechanisms are used to determine fault slip parameters andinverted for the extension direction across the Ethiopian rift.These results are then compared to global and local platekinematic models. Our new seismicity data are interpretedin light of structural, seismic refraction/wide-angle reflec-tion, gravity, anisotropy, and crustal and mantle tomographicstudies to propose that extension via magma injection andminor faulting characterizes the late stages of continentalrifting prior to breakup.

2. Tectonic Setting

[5] The Ethiopian rift system is on the Ethiopia-Yemenplateau that is thought to have developed above a mantle

plume [e.g., Schilling, 1973; Ebinger and Sleep, 1998;George et al., 1998]. A 2-km-thick sequence of floodbasalts and rhyolites erupted across the Ethiopia-Yemenplateau region between 45 and 22 Ma [e.g., George et al.,1998; Kieffer et al., 2004]. The majority erupted at 30 Maalong the Red Sea margins [e.g., Hofmann et al., 1997;Ukstins et al., 2002] coincident with the opening of the RedSea and Gulf of Aden [Wolfenden et al., 2005]. Anoma-lously low P wave velocities exist in the mantle beneathAfar to depths of at least 410 km, but their connection withthe profound low-velocity zone in the lower mantle beneathsouthern Africa is debated [e.g., Debayle et al., 2001;Benoit et al., 2003; Montelli et al., 2004].[6] The MER forms one arm of the complex Afar triple

junction zone (Figure 2). Rifting initiated in the southernand central Main Ethiopian rift between 18 and 15 Ma butthe northern Main Ethiopian rift only developed after11 Ma [WoldeGabriel et al., 1990; Wolfenden et al.,2004]. Between 12 and 10 Ma, the southern Red Sea marginpropagated southward as the MER propagated NE, effec-tively linking the southern Red Sea and Ethiopian rifts, and

Figure 1. EAGLE permanent broadband seismic stations used for earthquake location with respect tomajor border faults and magmatic segments of the Main Ethiopian rift (MER). Grey triangles are phase 1stations (October 2001 to January 2003), white triangles are phase 2 stations (October 2002 to January2003), white circles are phase 3 stations (November 2002 to January 2003), and white squares are theIRIS GSN permanent stations FURI and permanent stations AAE and WNDE. The inset shows thetopographic relief, plates, and rift zones: A, Arabia; D, Danakil; N, Nubian Plate; S, Somalian Plate; RS,Red Sea rift; GA, Gulf of Aden rift.

B05314 KEIR ET AL.: ETHIOPIAN RIFT SEISMICITY

2 of 17

B05314

forming a triple junction for the first time [Wolfenden et al.,2004].[7] The northern Main Ethiopian rift is a series of linked

half grabens bounded by steep NE striking Miocene borderfaults [WoldeGabriel et al., 1990; Wolfenden et al., 2004](Figure 1). Structural patterns suggest a change fromN130E to N105E directed extension sometime in theinterval 6.6 to 3 Ma [Boccaletti et al., 1998; Wolfenden etal., 2004]. During this time period extensional strain mi-grated from border faults to smaller offset approximatelyN10E striking faults and aligned eruptive centers in thecentral rift valley [Wolfenden et al., 2004]. The <20 kmwide, right-stepping, en echelon zones of magmatism andfaulting are referred to as magmatic segments [Ebinger andCasey, 2001]. GPS measurements show that approximately80% of present-day extension across the MER is localizedwithin these magmatic segments [Bilham et al., 1999]. Themagmatic segments in the center of the rift are underlain by20-km-wide, high-velocity (Vp > 6.5 km/s) elongatebodies that are interpreted as cooled mafic intrusions[Keranen et al., 2004; Mackenzie et al., 2005]. Thesemagmatic segments are characterized by relative positiveBouguer anomalies [Mahatsente et al., 1999; Tiberi et al.,2005]. Historic fissural basalt flows at Fentale and Konevolcanoes as recently as 1810 indicate ongoing volcanismin magmatic segments [Harris, 1844].[8] The northern Main Ethiopian rift shows a northward

increase in crustal extension and magmatic modification[Tiberi et al., 2005; Maguire et al., 2006; Stuart et al.,

2006]. Crustal thickness beneath the MER decreases from38 km in the south beneath the caldera lakes to 24 kmbeneath Fentale volcano in the southern Afar depression[Dugda et al., 2005; Maguire et al., 2006] (Figure 1). Thealong-axis thinning is consistent with a northward along-axis decrease in effective elastic thickness and seismogeniclayer thickness [Ebinger and Hayward, 1996]. Seismicrefraction/wide-angle reflection data show 40-km-thickcrust beneath the southeastern plateau, whereas the westernside of the rift is underlain by 45- to 50-km-thick crust witha 10- to 15-km high-velocity lower crust believed to beunderplate [Mackenzie et al., 2005].[9] Geochemical and seismic data provide constraints on

melting and melt emplacement beneath the MER. The majorelement compositions of Quaternary mafic lavas from theMER show the onset of melting occurs in the lower crustand upper subcontinental lithospheric mantle [Rooney et al.,2005]. This is consistent with P and S wave tomographicmodels that show anomalous low-velocity zones in theupper mantle beneath the rift, attributed to a combinationof higher temperatures and the presence of partial melt[Bastow et al., 2005]. Both the large amount of SKSsplitting and the rift parallel orientation of the fast polari-zation direction led Kendall et al. [2005] to propose thatpartial melt beneath the MER rises through dikes thatpenetrate through the thinned lithosphere. Shear wavesplitting in local earthquakes beneath the MER shows thatanisotropy is highest in zones of diking, and it suggests that

Figure 2. Past and present constraints on plate kinematics in the Afar triple junction zone. A, Arabia;D, Danakil; N, Nubian Plate; S, Somalian Plate; RS, Red Sea rift; GA, Gulf of Aden rift; MER, MainEthiopian rift; TGD, Tendaho-Goba’ad Discontinuity. (a) Pre-3.2 Ma tectonics of the Afar triple junction.Relatively rigid blocks are shaded. Rift propagation directions are shown by light grey arrows. Thin darkgrey arrows show pre-3.2 Ma extension directions. (b) The 3.2 Ma to present and current plate motionswith respect to the Nubian plate; vector length scaled to extensional velocity. Along-axis propagationdirection is shown by light grey arrows.


3 of 17

B05314

melt-induced anisotropy deeper in the lithosphere continuesinto the upper crust [Keir et al., 2005].[10] The orientation of present-day extension across the

Ethiopian rift remains controversial. Laser ranging and GPSdata show that the northern Ethiopian rift over the period1969–1997 extended in a direction of N108E ± 10 at 4.5 ±0.1 mm/yr [Bilham et al., 1999] (Figure 2). The velocityfield calculated from permanent GPS stations on Africasince 1996 shows opening of 6–7 mm/yr at an azimuth ofapproximately N95E [Fernandes et al., 2004; Calais et al.,2006] (Figure 2). Global and regional plate tectonic modelsby Jestin et al. [1994] and Chu and Gordon [1999] averageplate kinematic indicators from the past 3.2 Ma and findsimilar extension directions and extensional velocities ofN102E at 5 ± 1 mm/yr and N96E ± 9 at 6.0 ± 1.5 mm/yr,respectively. Active Quaternary volcanoes in the MER haveelliptical shapes with their long axes in the directionN105E [Casey et al., 2006]. Source parameters of tele-seismically recorded earthquakes show normal, normal left-

oblique and sinistral strike-slip motions with the horizontalcomponent of T axes between N135E and N90E inorientation [e.g., Ayele and Arvidsson, 1998; Foster andJackson, 1998; Ayele, 2000; Hofstetter and Beyth, 2003](Figure 3). Kinematic indicators on Quaternary faults thatdip 70–75 and strike N10–35E indicate a principal dip-slip normal movement with a mean direction of approxi-mately N95E [Pizzi et al., 2006]. However, Acocella andKorme [2002] matched pairs of asperities along the sides ofQuaternary extension fractures to show a mean extensiondirection of N128E ± 20. Korme et al. [1997] used theorientation of extension fractures to determine an extensiondirection of NW-SE, similar to Wolfenden et al.’s [2004]N130E estimate of Miocene-Pliocene extension direction.

3. Seismic Activity

[11] Seismicity data are lacking from the Ethiopianrift due to previous sparse station coverage [Ayele and

Figure 3. Seismic activity of the Horn of Africa since 1960. Earthquake locations and magnitudes arefrom Ayele [1995] for the time period 1960–1997 and the NEIC catalogue (1997–2005). Earthquakefocal mechanisms are from Harvard CMT catalogue, Foster and Jackson [1998], Ayele and Arvidsson[1998], Ayele [2000], and Hofstetter and Beyth [2003]. Quaternary volcanoes in the MER are shown bytriangles.


4 of 17

B05314

Kulhanek, 1997]. The earliest documented seismic event inthe Ethiopian rift is a swarm of earthquakes in 1841–1842near Debre Birhan which caused the destruction of the townof Ankober by landslides [Gouin, 1979] (Figure 3). Histor-ical records spanning the past 150 years show that largemagnitude earthquakes are rare in the MER [Gouin, 1979].The record of seismicity from 1960 to 2000 compiled fromteleseismic and regional catalogues complete down to ML 4 shows that the majority of earthquakes are located alongthe highly eroded southern Red Sea escarpment north of9.5N, 38.7E [e.g., Kebede et al., 1989; Ayele, 1995; Ayeleand Kulhanek, 1997; Hofstetter and Beyth, 2003; Ayele etal., 2006b] (Figure 3).[12] An estimate of seismic moment release since 1960

shows that more than 50% of extension across the MER isaccommodated aseismically [Hofstetter and Beyth, 2003].During this period swarms of low-magnitude events havebeen located near Debre Birhan [Gouin, 1979], and nearFentale volcano in 1981 and 1989 where NNE strikingsurface fissures developed following earthquake swarms ofML < 4 [Asfaw, 1982]. Similar fissures oriented N20E andN45E are observed elsewhere along the axis of the MERand attributed to tectonic processes [Asfaw, 1982, 1998].Tension fractures cut welded tuffs at Fentale and Konevolcanoes and suggest a fissuring episode within the past7000 years [Williams et al., 2004]. In the year preceding thisstudy, the seismicity was concentrated in the Fentale-Dofenand Angelele magmatic segments [Ayele et al., 2006a].From mid-October 2003, after removal of the EAGLEseismic network, a 1 month long earthquake swarm witha main shock of ML 5 was recorded by the GeophysicalObservatory and reported by local inhabitants near Dofenvolcano (Geophysical Observatory, Addis Ababa University,personal communication). The epicenter of the main shockis estimated to be 9.2N 40.1E from the locations ofdamaged buildings and trees, reported scree slides in thearea, and personal accounts of ground shaking (Figure 3).[13] Hypocenter depths of 5–10 km have been reported

for seismic swarms in the MER and southern Afar [Asfaw,1982; Ayele et al., 2006a]. Teleseismically recorded earth-quakes on the eastern side of the MER have been locatedbetween 8 and 12 km depth [Ayele, 2000].

4. EAGLE Seismic Data

4.1. Seismic Network

[14] Local earthquakes were recorded on 29 broadbandseismic stations (EAGLE phase I) that were operationalbetween late October 2001 and January 2003 (Figure 1).These three-component Guralp CMG-3T and CMG-40TDinstruments recorded data at 50 Hz. Additional data over thistime period was acquired from the permanent broadbandstations AAE, FURI, and WNDE maintained by the Geo-physical Observatory, Addis Ababa University. The numberof seismic stations was increased during the final 4 monthsof the experiment with the deployment of an additional 50CMG-6TD instruments recording at 100 Hz. These stationswere deployed at15 km spacing in the rift valley and in theAnkober region, and were operational between October2002 and February 2003 (EAGLE phase II). For securityreasons the broadband stations were located in compoundsattached to schools, clinics and plantation offices. A further

100 CMG-6TD instruments deployed at 5 km spacingacross the rift valley and adjacent plateaus were operationalbetween November 2002 and January 2003 (EAGLE phaseIII). Three local earthquakes were recorded on up to 600single-component, short-period, Reftek ‘‘Texan’’ instru-ments deployed for 8 days at 1 km spacing both alongand across the rift valley. During the daytime, the level ofhigh-frequency cultural noise (>1 Hz) could be high. Atnight, however, noise levels were significantly reduced.

4.2. Arrival Time Analysis

[15] Earthquakes were detected on the continuous seismicdata of the phase I array with a short-term amplitude/long-term amplitude (STA/LTA) trigger algorithm with windows1 s and 60 s in length respectively. The algorithm scannedthe continuous data, filtered using a Butterworth filter withcorner frequencies of 2–15 Hz, and flagged time windowswhen an STA/LTA ratio of 20 was exceeded within a 120 stime window on two or more stations. Arrival times of Pand S phases were initially measured on phase I data filteredusing the same Butterworth filter. Arrival times from phaseII and phase III seismic stations were added to earthquakesthat occurred during the respective operation periods ofthese arrays. Arrival times of P phases were assigned aquality factor of 0, 1, 2, or 3 according to estimatedmeasurement errors of 0.05 s, 0.1 s, 0.15 s, and 0.2 s,respectively, and S wave quality factors of 0, 1, 2, and 3were assigned to arrivals with estimated measurement errorsof 0.1 s, 0.175 s, 0.25 s, and 0.3 s, respectively. A total of13,388 P wave and 12,725 S wave arrivals were pickedfrom 2139 local earthquakes.

5. Methodology

5.1. Earthquake Location and Magnitude

[16] In total, 2139 local earthquakes recorded at four ormore stations were located with the Hypo2000 algorithm[Klein, 2002]. A one-dimensional (1-D) P wave velocitymodel and Vp/Vs ratio of 1.75, calculated from P and S wavetraveltimes, were used for the initial earthquake locations.The weighting of arrival times was dependent on the qualityfactor assigned to the phase, with P wave quality factors of0, 1, 2, and 3 given full (1), 0.75, 0.5, and 0.25 weightsrespectively. S waves were given half weighting relative toP waves of the same quality factor.[17] The 1-D P wave velocity model and station correc-

tions were determined by simultaneously relocating earth-quakes and inverting for velocity structure with VELEST[Kissling et al., 1995]. Only earthquakes with eight or more Parrivals, an azimuthal gap of less than 180, and an epicentraldistance to the nearest station of less than twice the focaldepth were used to invert for the 1-D Pwave velocity model.280 earthquakes satisfied the selection criteria and can beconsidered as ‘‘well-located’’ earthquakes. Additional con-straints on the 1-D P wave model were provided by thecontrolled source experiment [Mackenzie et al., 2005].[18] The 280 well-located earthquakes were subsequently

relocated using a 3-D P wave velocity model determinedwith SIMULP [e.g., Eberhart-Philips and Michael, 1998;Haslinger et al., 1999]. Hypocenter accuracy of the earth-quakes was tested by relocating shots and randomly adjust-ing horizontal and vertical positions of hypocenters. From


5 of 17

B05314

these tests we estimate hypocenter accuracy for earthquakesof about ±600 m in horizontal directions and ±2000 m indepth.[19] Local magnitude was estimated using the maximum

body wave displacement amplitudes (zero to peak) measuredon a simulated Wood-Anderson seismograph and distancecorrection terms ofHutton and Boore [1987]. The power lawcumulative frequency-magnitude distribution [Gutenbergand Richter, 1956] is used to describe the size distributionof earthquakes recorded within the EAGLE network in theMER. The b value is calculated using the maximum likeli-hood method [Aki, 1965]. The standard deviation of the bvalue is used as an error estimate [Shi and Bolt, 1982].

5.2. Focal Mechanisms and Stress Inversion

[20] Focal mechanisms were computed from P and SHwave polarities using the grid search algorithm FOCMEC[Snoke et al., 1984]. A double-couple source type isassumed as all the events selected are characterized by

high-frequency content, sharp first arrivals and clear Sphases at the nearest stations. Hypocenter coordinates weredetermined by locating the event with the 3-D velocitymodel. A fault plane solution was only attempted if anearthquake was located within the network, the neareststation was within an epicentral distance of twice the focaldepth, and the solution had a minimum of 10 P wavepolarities located in at least three quadrants of the focalsphere. Polarity errors of neither P nor SH waves weretolerated in the grid search algorithm. In total, 33 well-constrained and unambiguous fault plane solutions that havea maximum 20 uncertainty in either strike or dip of bothnodal planes were determined. This new data set is supple-mented by the three well-constrained focal mechanismsdetermined from data at regional and teleseismic distances[Ayele, 2000; Hofstetter and Beyth, 2003] (CMT, Harvard).[21] The focal mechanisms were used to invert for the

regional stress tensor with the linear, least squares stressinversion method ofMichael [1984] that minimizes the angle

Figure 4. Seismicity of the MER from October 2001 to January 2003. Earthquakes were located withthe minimum 1-D P wave velocity model determined from local earthquake tomography. Only eventsrecorded by at least four stations and located within the array of seismic stations are displayed. Heavyblack lines show major border faults; ellipses mark Quaternary magmatic segments. The star shows thelocation of the October 2003 earthquake swarm near Dofen volcano.


6 of 17

B05314

between the predicted tangential traction on the fault planeand the observed slip direction. The 95% confidence regionswere determined with the bootstrap resampling method[Michael, 1987a, 1987b] and used as an error estimate. Therelatively small data set and estimated focal mechanismerrors of this study make this method most appropriate toboth accurately determine the stress orientation and ade-quately estimate the confidence limits [Hardebeck andHauksson, 2001].[22] The inversion procedure assumes that the four stress

parameters are constant over the spatial and temporal extentof the data set and that earthquakes slip in the direction of theresolved shear stress on the fault plane [Michael, 1984]. Theuniform stress tensor that best explains the mechanisms isexpressed by the three principal stress axes (where s1, s2 ands3 are the maximum, intermediate, and minimum principalstresses, respectively) and the stress ratio. An average misfitangle b, which measures the difference between the observedslip direction and the predicted direction of maximumtangential traction, is computed and used as a measure ofthe success of the inversion. The steepest nodal plane of thenormal fault focal mechanisms and approximately NE strik-ing nodal planes of the strike-slip mechanisms were chosenas fault planes for the inversion in accord with geologicalobservations [e.g., Abebe et al., 1998a; Wolfenden et al.,2004; Casey et al., 2006; Pizzi et al., 2006].

6. Results

6.1. Hypocenter Distribution

[23] From October 2001 to January 2003, 2139 localearthquakes were recorded by the EAGLE network. Of

these, 1957 earthquakes were located within the networkof seismic stations (Figure 4). Concentrated seismic activityoccurs in the Fentale-Dofen magmatic segment, which is a20-km-wide, 70-km-long zone that extends from Fentalecaldera to Dofen volcano (Figure 4). Earthquakes arelocated in a 10-km-wide, NNE trending zone that extends40 km north of Fentale volcano where the pattern ofseismicity is mirrored by the surface expression of theclosely spaced, small offset Quaternary faults and fractures(Figure 5). Three distinct earthquake clusters are locatednear the Pliocene–Recent Dofen volcano (Figures 4 and 6).The distribution of earthquakes located with the 3-D P wavevelocity model show that these clusters are elongate ap-proximately north to approximately NNE, parallel to thesurface expression of major Quaternary fault systems thatcut lavas erupted from fissures (Figure 6).[24] The frequency-depth distribution of earthquakes

within the Fentale-Dofen magmatic segment located withthe 3-D P wave velocity model shows most earthquakes are8–14 km deep (Figure 6 and 7). Hypocenter depths are 8–10 km deep near Fentale and Dofen volcanoes but are up to16 km deep in between these major eruptive centers(Figure 6). The temporal distribution of seismicity in theFentale-Dofen magmatic segment is characterized by earth-quake swarms that punctuate largely aseismic periods(Figure 8).[25] Minor seismicity is located within the Boset and

Aluto-Gedemsa magmatic segments (Figure 4). Regionsbetween the right stepping en echelon magmatic segmentsare largely aseismic.[26] Seismic activity south of the Aluto-Gedemsa mag-

matic segment is more diffuse than to the north (Figure 4).

Figure 5. Example of seismicity located near Quaternary eruptive volcanic centers and faults of theFentale-Dofen magmatic segment plotted on a gray scale Landsat 741 image. The inset shows theposition of the image with respect to border faults and magmatic segments in the MER.


7 of 17

B05314

This rift sector lacks the narrow zone of localized faults anderuptive centers characterizing the magmatic segments(Figure 10). Epicenters are located within a 30- to 40-km-wide zone of Quaternary faults along the eastern side of therift valley. The amount of seismicity in this rift sector isrelatively low and lacks the periods of swarm activity

observed further north in the Fentale-Dofen magmaticsegment (Figure 8).[27] The exception to the pattern of correlated seismicity

and Quaternary eruptive centers is the long-lived seismicityat the intersection of the NE striking Miocene MER andnorth striking Oligocene Red Sea structures near Ankober

Figure 6. Earthquake locations determined using the 3-D P wave velocity model in the Ankober regionand Fentale-Dofen magmatic segment, plotted on 90 m resolution SRTM topographic data. Theearthquakes were recorded with eight or more P wave arrivals and have an azimuthal gap of less than180 and an epicentral distance to the nearest station of less than twice the focal depth. From Ankober(grey triangle) south, the uplifted rift flank of the 30 Ma southern Red Sea was warped southeastwardinto the northern MER after 11 Ma. Profiles A-A0 and B-B0 project earthquakes within 30 km of the lineof section onto the profile. The thickened portions of the profiles show where the profile crosses theFentale-Dofen magmatic segment.


8 of 17

B05314

(Figures 4 and 6). This intersection zone has the highestrelief in the region, with deeply incised valleys. Earthquakesare localized in a N-S oriented cluster on the northwestmargin of the rift valley at 9.5N 39.75E. The cluster lies atthe southern end of the approximately north strikingAnkober border fault system, which is a series of closelyspaced high-angle normal faults and tight monoclinal folds[Wolfenden et al., 2004]. The rate of seismicity in this areawas high for the first 6 months of the experiment andcharacterized by frequent swarm activity (Figure 8). Focaldepths are concentrated between 10 and 13 km with activityobserved down to 18 km (Figures 6 and 7).[28] A minor, roughly E-W elongate cluster of earth-

quakes is located near Addis Ababa (Figure 4). The struc-ture of this area is dominated by the east striking Ambolineament, a fault zone active since the late Miocene[Abebe et al., 1998b]. Isolated but relatively deep earth-quakes (15–21 km) characterize the remaining earthquakeactivity of the Ethiopian plateau. The southeastern plateaushows a lack of activity except for a small cluster on thesouthern margin of the Gulf of Aden rift at 9N 40.5E(Figure 4).

6.2. Seismicity Rate

[29] The annual cumulative frequency-magnitude distri-bution of the 1957 earthquakes recorded within the EAGLEnetwork shows that the seismicity catalogue is completeabove ML 2.1 (Figure 9). The largest magnitude earthquakeis only ML 3.9. The estimated b value using the maximumlikelihood method of Aki [1965] is 1.13 ± 0.05, and thisslope intercepts the y axis at 4.5. This is the first reliable bvalue estimate for the MER as the historic record is toosparse for a reliable estimate [Ayele and Kulhanek, 1997].Hofstetter and Beyth [2003] obtained a b value of 0.83 ±0.08 for a larger area that encompasses both the MER andsouthern Ethiopian Rift to 5N.[30] The estimated b value of 1.13 ± 0.05 for the MER is

similar to b values of between 1.05 and 1.3 calculated forthe oceanic southern Red Sea and Gulf of Aden rift systems[Ayele and Kulhanek, 1997; Hofstetter and Beyth, 2003].Lower b values of between 0.7 and 0.9 are observed in theless evolved continental rifts in Kenya and Tanzania [e.g.,

Tongue et al., 1992; Langston et al., 1998; Ibs-von Seht etal., 2001].

6.3. Focal Mechanisms and Stress Inversion

[31] In total, 33 well-constrained and unambiguous faultplane solutions that have a maximum 20 uncertainty ineither strike or dip of both nodal planes were determined(Table 1 and Figures 10 and 11). This new data set issupplemented by the three well-constrained focal mecha-nisms determined from data at regional and teleseismicdistances [Ayele, 2000; Hofstetter and Beyth, 2003] (CMT,Harvard) (Table 2 and Figure 10).[32] Focal mechanisms of earthquakes located along the

axis of the MER and in the Ankober fault system showpredominantly normal dip slip on steep faults that strikeapproximately north to approximately NNE (Figures 10and 12). Focal mechanisms are subparallel to the dominantN10E orientation of Quaternary faults in the Ethiopian rift[Boccaletti et al., 1998; Wolfenden et al., 2004; Casey et al.,2006] (Figure 10). A few of the normal dip-slip focalmechanisms have slip planes that strike approximatelyNE, parallel to the pre-3.5 Ma, N40E striking faults(Figure 12). The exceptions to these normal dip-slip focalmechanisms are the strike-slip earthquakes below Fentaleand Boset volcanoes, interpreted as left-lateral motion onapproximately NE to approximately ENE striking faults(Figure 12). However, both normal and strike-slip focalmechanisms show near horizontal T axes striking N80E–N130E (Figures 11 and 12).[33] The results of the stress inversion using the 36 focal

mechanisms in the MER show that the trend/plunge of theminimum principal stress is 283/6 with a mean misfitangle (b) ± standard deviation of 10.9 ± 7.0 (Figure 12).This mean misfit angle is comparable to results of stresstensor inversions from focal mechanisms within uniformstress fields in other studies: 10–17 along fault segmentsof the San Andreas fault zone [Jones, 1988]; and 6–24 fordata sets in the Swiss Alps and northern Alpine foreland[Kastrup et al., 2004]. However, a well-resolved stresstensor requires that the data set contains a diverse rangeof focal mechanisms. In our data set, only four strike-slipfocal mechanisms differ from the predominant dip slip onapproximately north to approximately NE striking faults.

Figure 7. Histograms of number of earthquakes per 1 km depth bin interval for the (a) Fentale – Dofenmagmatic segment and (b) Ankober region. The hypocenters were located with the 3-D P wave velocitymodel and are displayed on Figure 6.


9 of 17

B05314

This unavoidable lack of diversity in type of focal mecha-nism reduces the resolution of the stress tensor.

7. Discussion

7.1. Distribution of Seismicity

[34] This study recorded seismicity for 15 months andthus provides a snapshot of active deformation in the MER.However, the pattern of Quaternary faults and fissures thatcut recent lavas and historic seismicity data show that ourresults are representative of the longer-term brittle strainpatterns in the rift [e.g., Asfaw, 1982, 1998; Williams et al.,2004; Wolfenden et al., 2004; Ayele et al., 2006a; Casey etal., 2006].[35] The most striking feature of the recorded seismicity

in the MER is the coincidence of earthquake swarms andthe magmatic segments, which are the locus of Quaternaryvolcanism. The inactivity of mid-Miocene border faults thatdefine the overall approximately NE trend of the rift isreflected in the minor geodetic strain on the rift flanks[Bilham et al., 1999] and lack of large magnitude earth-quakes on border faults over the last 50 years [Ayele andKulhanek, 1997]. This inactivity is inferred from historicalrecords spanning the past 150 years [Gouin, 1979], andmorphology of the border faults [Boccaletti et al., 1998;Wolfenden et al., 2004].[36] The exception is the seismicity observed at the

intersection between the north striking Red Sea rift andthe NE striking MER. The cluster of earthquakes is locatedon the north striking Ankober fault system that formed at11 Ma to link the two oblique rift systems. Although faultand seismicity patterns show that the locus of strain hasshifted to the Quaternary magmatic segments in the centralrift, this high point along the rift flank still experiencesstrain [Wolfenden et al., 2004]. The strike of the Ankoberfault system is oblique to the NE trending MER, and

Figure 9. Log annual cumulative number of earthquakesagainst magnitude plot of earthquakes recorded within theEAGLE network. Mc marks ML 2.1, above which thecatalogue is complete. The slope of the straight line (bvalue) is 1.13 ± 0.05.

Figure 8. (a) Seismicity of the MER recorded by theEAGLE network with the three regions that experienced themost activity highlighted: 1, Ankober area; 2, Fentale-Dofen magmatic segment; and 3, south of Aluto-Gedemsamagmatic segment. (b) Cumulative number of earthquakesversus recording time of the regions 1, 2, and 3.(c) Cumulative number of earthquakes versus recording timeof all the earthquakes recorded within the EAGLE network.


10 of 17

B05314

focused deformation in this complex zone of rift intersectionmay be caused by flexure accommodating differentialsubsidence in the Red Sea rift relative to the youngerMER. Further north of Ankober, the Red Sea rift marginis seismically active as shown in historical records, regionalcatalogues and recent seismicity [Ayele et al., 2006b]. Stressis concentrated in this area by the large lateral densitycontrast and difference in lithospheric thickness betweenthe uplifted western Ethiopian plateau and Afar depression[e.g., Dugda et al., 2005; Tiberi et al., 2005].[37] A number of lines of evidence indicate that exten-

sional strain is accommodated by a combination of dikeinjection and faulting within magmatic segments, as out-lined below. For the period 1960–2000, a comparison of theexpected released seismic moment and observed seismicmoment shows that less than 50% of extension across theMER is accommodated by rapid slip on faults [Hofstetterand Beyth, 2003]. At the surface, GPS measurements showthat approximately 80% of present-day extension across theMER is localized in a 20-km-wide zone of Quaternaryfaulting and magmatism [Bilham et al., 1999]. This narrowzone of localized deformation is also observed in the brittleupper crust from patterns of seismicity. Elongate clusters ofearthquakes are associated with observed faults, fissures andactive eruptive centers in the Fentale-Dofen magmaticsegment. The swarms of low-magnitude earthquakes areconcentrated at 8–14 km depth which coincides with thetop of the 20- to 30-km-wide zone of extensive maficintrusions at 8–10 km depth [Keranen et al., 2004]. Seismicanisotropy of the upper crust is highest in the magmaticsegments and attributed to melt-filled cracks and dikesaligned perpendicular to the minimum stress [Keir et al.,

2005]. Crustal strain across the MER is accommodatedwithin the magmatic segments by magma intrusion below10 km, and by both faulting and dike intrusion in thebrittle seismogenic zone.[38] The Debre Zeit and Butajira chains of Quaternary

eruptive centers located west of the magmatic segments arelargely aseismic, and they show little structural or morpho-logical evidence of active strain. Xenolith data and tomo-graphic models show these chains are underlain by hotasthenosphere [Bastow et al., 2005; Rooney et al., 2005],but they lack the large relative positive Bouguer anomalyand high-velocity crust of the magmatic segments [e.g.,Tiberi et al., 2005]. These chains may be either unfavorablyoriented ‘‘failed’’ magmatic segments, or incipient zones ofstrain.[39] In the magmatic segments of the MER, seismicity,

geodetic and structural data all show a localization of strainin zones of Quaternary magmatism. The earthquakes in themagmatic segments are concentrated above axial maficintrusions and may be induced by dike injection. Modelsof the elastic stress field surrounding propagating fluid-filled cracks show that earthquakes of magnitude >1 can beinduced ahead of a propagating dike if the ambient stressfield is near to failure, and slip is likely to occur alongpreexisting fractures [Rubin and Gillard, 1998]. Earthquakeswarms are assumed to occur near the crack tips due to theincreasing stress caused by concentrated internal fluids.Spatially, swarms reflect areas of magma intrusion. Thecorrelation we observe in the MER between seismic swarmsand magma injection has been documented near activevolcanoes in other settings, suggesting the swarms arecausally linked to magma intrusion. For example, seismicity

Table 1. Earthquake Source Parameters Determined From EAGLE Data

Event Date, year/month/day Time, UT Latitude, N Longitude, E Depth, km Strike Dip Rake ML

1 2002/01/16 2122:39.44 9.239 40.021 13.25 180.00 50.00 90.00 1.72 2002/01/17 0138:03.91 8.154 39.002 20.29 2.27 60.05 93.46 2.013 2002/01/18 0142:40.83 8.998 39.918 10.23 359.67 54.23 97.40 2.824 2002/02/17 0238:15.44 9.470 39.692 11.86 171.52 66.00 90.00 3.215 2002/05/02 2143:23.17 9.122 39.984 13.16 211.58 56.38 80.38 2.646 2002/07/04 0259:42.35 9.173 39.966 15.84 214.40 60.08 85.38 3.547 2002/07/31 0154:38.27 9.444 39.677 11.25 172.76 66.06 85.62 2.348 2002/08/21 0127:23.93 8.951 39.711 13.85 192.88 60.13 84.23 2.149 2002/10/08 1937:43.42 9.199 39.949 12.65 225.74 68.06 85.69 2.0310 2002/10/09 1819:37.91 9.193 39.987 12.52 223.13 68.19 64.02 2.1411 2002/10/10 1915:51.93 9.066 39.965 14.59 201.49 58.30 66.30 1.1712 2002/10/19 2125:25.96 10.130 39.957 15.47 198.07 59.38 71.32 2.8313 2002/11/04 0017:42.49 8.432 39.673 12.91 2.54 51.18 83.58 1.1714 2002/11/04 0024:55.49 7.812 38.976 6.88 183.82 63.32 109.10 1.7115 2002/11/05 2242:14.69 9.728 39.370 14.65 29.29 66.39 79.08 1.9216 2002/11/07 0124:31.21 9.492 40.040 15.48 216.30 46.04 74.63 1.8917 2002/12/03 1602:52.26 7.481 38.553 13.63 183.71 68.01 92.16 2.5518 2002/12/03 2010:01.33 7.700 38.911 12.43 190.00 45.00 90.00 2.3419 2002/12/04 1341:09.57 8.873 39.836 9.57 209.92 60.00 90.00 1.9720 2002/12/13 1736:21.66 9.494 40.034 15.79 183.69 64.27 98.89 2.221 2002/12/15 0837:35.26 7.428 38.648 8.61 197.95 50.00 90.00 3.0622 2002/12/15 1915:38.82 7.430 38.657 6.42 210.49 70.38 78.31 2.8923 2002/12/15 2035:05.22 9.548 40.144 19.01 190.55 66.56 103.10 1.9324 2002/12/17 2212:36.10 9.001 39.907 8.44 64.49 88.17 0.81 1.425 2002/12/17 2315:10.76 8.998 39.901 9.17 71.95 80.73 3.78 1.5526 2002/12/23 0627:49.95 9.446 39.680 10.43 181.99 60.00 90.00 2.4527 2002/12/26 1947:51.98 9.221 40.014 12.96 213.15 62.02 87.74 3.1728 2002/12/26 1955:17.90 9.221 40.011 12.65 219.89 60.00 90.00 2.4129 2003/01/02 0852:45.37 9.246 40.013 13.91 195.00 65.00 90.00 2.3730 2003/01/10 1213:56.08 8.611 39.447 7.00 42.64 85.25 13.19 3.4431 2003/01/13 2106:00.76 9.491 39.681 11.17 168.45 56.21 97.23 1.9732 2003/01/20 2116:22.90 7.475 38.823 11.74 206.12 42.96 104.76 2.5233 2003/01/21 0808:18.85 7.495 38.822 11.41 197.78 37.16 117.15 2.9


11 of 17

B05314

leading to the Mount Etna eruption of 2001 was character-ized by swarms elongate parallel to surface fractures andparallel to the maximum compressive stress determinedfrom focal mechanisms [Musumeci et al., 2004]. Thisseismic activity was interpreted as being caused by dikeemplacement prior to the eruption. By analogy to theseother locales and independent data from the MER, wepropose that the observed seismicity in magmatic segmentsabove axial mafic intrusions is induced by magma injectioninto the midcrust to upper crust (Figure 13).[40] The along-axis segmentation of the MER is reflected

at the surface by the right-stepping en echelon patterns ofQuaternary faults and aligned cones within discrete 20-km-wide, 60-km-long magmatic segments. The pattern ofseismicity interpreted in light of other data provides cluesas to the origin of this along-axis segmentation. At 8–10 kmdepth subsurface, the segmentation is evident as discreteaxial mafic intrusions imaged by crustal tomography[Keranen et al., 2004]. These mafic bodies correlate withalong-axis velocity variations in the midcrust and lowercrust, implying that mafic intrusions extend to the base ofthe crust [Maguire et al., 2006]. Extension in the midcrustto lower crust is thus likely accommodated within a narrowzone of magma injection. The onset of melting likely occurs

in the lower crust and subcontinental lithosphere [Rooney etal., 2005]. The correlation between the orientation oflithospheric anisotropy and the distribution of Quaternarystrain and magmatism shows that vertically oriented dikeswith partial melt crosscut the lithosphere [Kendall et al.,2005]. The concentrated seismicity in the Fentale-Dofenmagmatic segment and largely aseismic Boset-Kone andAluto-Gedemsa magmatic segments is indirect evidencethat episodic rifting events within one magmatic segmentare independent of other magmatic segments. This suggestsmagma source regions are spatially and temporally discrete.[41] The pattern of seismicity observed in the MER is

strikingly similar to patterns in oceanic rift zones whereseismic swarms are induced in already stressed lithosphereby injection of magma. For example, seismic swarms in theHengill volcanic area in southwestern Iceland are concen-trated at the base of the seismogenic layer and havepredominantly double-couple mechanisms [Feigl et al.,2000]. Calculations of Coulomb failure stress suggest thatmagma injection to 7 km subsurface is sufficient to triggerearthquakes in the overlying crust. Clusters of seismicitymarked a narrow zone parallel to fissure swarms in theKrafla spreading segment of northern Iceland 5–8 yearsafter a dike injection episode [Arnott and Foulger, 1994].

Figure 10. Faults that cut <1.9 Ma lavas, and Quaternary eruptive centers comprising magmaticsegments, relative to the Miocene border faults bounding Main Ethiopian rift basins [after Casey et al.,2006]. Fault plane solutions are lower hemisphere projections. The size of the solution is scaled tomagnitude between ML 1.17 and 5.3.


12 of 17

B05314

This pattern was attributed to the release of stress as thecrust returns to equilibrium in the neighborhood of the newdikes [Arnott and Foulger, 1994]. Alternatively, Cattin etal. [2005] reproduce geodetic observations constrained byheat flow and seismicity in the Asal-Ghoubbet rift using aviscoelastic model of semicontinuous dike intrusion in anarrow zone at depth. Episodes of magma injection explainthe localized seismicity patterns and high slip rates on faultsclose to the rift axis, as well as geodetically measuredground deformation.

7.2. Style of Faulting and Extension Direction

[42] The focal mechanisms provide a uniform picture forthe pattern of faulting and stress field orientation of theEthiopian rift. Focal mechanisms indicate predominantlynormal dip slip on faults that strike approximately north toapproximately NNE, parallel to the dominant N10E strikeof faults that cut Quaternary lavas [e.g., Casey et al., 2006].Field observations and geodetic data of volcanic rift zonesin Iceland and Hawaii indicate that dike intrusions are mostoften associated with normal faulting and fracturing at thesurface [Rubin, 1992]. The predominance of normal dipslip, and resulting lack of diversity in our focal mechanismdata set, is thus consistent with dike-induced seismicity inthe MER.[43] The normal, oblique, and left-lateral strike-slip dis-

placement on NE striking fault planes most likely occurs onpre-3.5 Ma, N40E striking faults that probably formedunder a NW-SE extension direction. These have most likelybeen reactivated as N40E striking ramps and transfer faultsto link N10E striking fault segments formed under theapproximately N105E extension direction during the Qua-ternary [Wolfenden et al., 2004; Casey et al., 2006]. Thenegligible block rotations about vertical axes in zones inbetween magmatic segments suggests no throughgoingtransform faults have developed, thus supporting our inter-pretation of the strike-slip focal mechanisms as left-lateralapproximately NE striking faults [Kidane et al., 2006].[44] The N103E orientation of the minimum compres-

sive stress from focal mechanisms parallels, within errors,the geodetically determined extension direction averagedover the past 3.2 Myr [Jestin et al., 1994; Chu and Gordon,1999] and current extension direction determined fromcampaign and permanent GPS data [Bilham et al., 1999;Fernandes et al., 2004; Calais et al., 2006]. Extension isperpendicular to the strike of Quaternary faults, fissures andaligned cones and is in agreement with structural studiesthat show a WNW-ESE direction of extension duringQuaternary times [Boccaletti et al., 1998; Wolfenden etal., 2004; Casey et al., 2006]. The current direction ofextension is thus perpendicular to the strike of Quaternaryvolcanic chains and faults in the magmatic segments. Theright-stepping en echelon pattern at the surface may beinduced by approximately N105E directed extension abovean approximately NE striking low-velocity zone in theupper mantle connecting the MER to the triple junction inAfar [Benoit et al., 2003; Bastow et al., 2005].

8. Conclusions

[45] 1. From October 2001 to January 2003, 1957 earth-quakes were located within the EAGLE network of broad-

Figure 11. A selection of focal mechanisms from thisstudy. Compressional P wave first motions are plotted ascircles and dilational first motions are plotted as triangles.The compressional quadrants of the focal sphere are shadedblack. Each solution is labeled by earthquake origin timeGMT (year, month, day, hour, minute).


13 of 17

B05314

band seismic stations in the northern Main Ethiopian riftand on its uplifted rift flanks. The earthquake catalogue iscomplete above ML 2.1 and the b value is 1.13 ± 0.05.[46] 2. Excluding the MER-Red Sea rift intersection

zone at Ankober, seismicity within the rift is localized to<20-km-wide, right-stepping, en echelon zones of Quater-nary magmatism. Seismicity in these magmatic segments ischaracterized by swarms of low-magnitude earthquakeslocated in clusters that parallel Quaternary faults, fissuresand chains of eruptive centers. The earthquakes in themagmatic segments are predominantly <14 km deep andmay be triggered by dike injection.[47] 3. Seismic activity at Ankober may be caused by

flexure accommodating differential subsidence at the

oblique intersection of the <11 Ma MER and the olderRed Sea rift.[48] 4. Earthquake focal mechanisms show predominantly

normal dip slip on faults striking approximately north toapproximately NNE. The orientation of the minimum com-pressive stress determined from focal mechanisms isN103E, consistent with geodetic data and global platekinematic constraints.[49] 5. From integration of these results with other

geophysical and structural observations we propose thatpresent-day extension in the MER is localized to discrete<20-km-wide en echelon magmatic segments, where exten-sional strain in the upper crust is accommodated by bothdike intrusion and dike induced faulting. The individual

Table 2. Earthquake Source Parameters Determined in Other Studies

Event Date, year/month/day Time, UT Latitude, N Longitude, E Strike Dip Rake Mw Data Source

34 1983/12/28 2308 7.03 38.60 176 51 81 5.3 Harvard CMT35 1993/02/13 0225 8.33 39.91 221 87 7 4.9 Ayele [2000]36 1995/01/20 0714 7.16 38.44 9 49 119 5.0 Hofstetter and Beyth [2003]

Figure 12. (a) Rose diagram of the orientation of the T axes of earthquake focal mechanisms. (b) Rosediagram showing the strike of earthquake slip planes. (c) Lower hemisphere plot of the trend and plungeof fault plane solution T axes (dark circles) and P axes (light triangles). (d) Results of the stress tensorinversion. Circle shows s3, the minimum compressive stress. Square shows s2, the intermediatecompressive stress. Triangle shows s3, the maximum compressive stress. The 95% confidence limits areshown by regions of grey shading.


14 of 17

B05314

magmatic segments show large spatial and temporal varia-tions in level of seismicity over the time period of the study,suggesting magma source regions for separate magmaticsegments are spatially and temporally discrete.

[50] Acknowledgments. We thank SEIS-UK for the use of instru-ments and A. Brisbourne for assistance in the field, with data managementand analysis. We thank A. Page, C. Tiberi, and A. Intawong for help withdata acquisition and processing. D. Cornwell, I. Bastow, and M. Casey arethanked for their significant contributions to this study. Laike Asfaw,Bekele Abebe, Dereje Ayalew, Gezahegn Yirgu, and Tesfaye Kidane ofAddis Ababa University are thanked for support throughout the project. Wethank Stephanie Prejean, Kevin Furlong, and an anonymous reviewer whohelped improve this manuscript. Our research was supported by NERCgrant NER/A/S/2000/01004 and NERC studentship NER/S/A/2002/10547.

ReferencesAbebe, B., M. Boccaletti, R. Mazzuoli, M. Bonini, L. Tortorici, and T. Trua(1998a), Geological map of the Lake Ziway-Asela region (Main Ethio-pian Rift), scale 1:50,000, Cons. Naz. delle Ric., A. R. C. A. Florence,Italy.

Abebe, T., F. Mazzarini, F. Innocenti, and P. Manetti (1998b), The Yerer-Tullu Wellel volcanotectonic lineament: A transtensional structure incentral Ethiopia and the associated magmatic activity, J. Afr. Earth Sci.,26(1), 135–150.

Acocella, V., and T. Korme (2002), Holocene extension direction along theMain Ethiopian Rift, East Africa, Terra Nova, 14, 191–197.

Acocella, V., T. Korme, and F. Salvini (2003), Formation of normal faultsalong the axial zone of the Ethiopian Rift, J. Struct. Geol., 25, 503–513.

Aki, K. (1965), Maximum likelihood estimate of b in the formula log N =a bM and its confidence limits, Bull. Earthquake Res. Inst. Univ.Tokyo, 43, 237–239.

Arnott, S. K., and G. R. Foulger (1994), The Krafla spreading segment,Iceland: 1. Three-dimensional crustal structure and the spatial and tem-poral distribution of local earthquakes, J. Geophys. Res., 99(B12),23,801–23,825.

Asfaw, L. M. (1982), Development of earthquake induced fissures, Nature,286, 551–553.

Asfaw, L. M. (1998), Environmental hazard from fissures in the MainEthiopian Rift, J. Afr. Earth Sci., 27(3/4), 481–490.

Ayele, A. (1995), Earthquake catalogue of the Horn of Africa for theperiod 1960–1993, Rep. 3-95, pp. 1–9, Seismol. Dep., Uppsala Univ.,Uppsala.

Ayele, A. (2000), Normal left-oblique fault mechanisms as an indication ofsinistral deformation between the Nubia and Somalia plates in the MainEthiopian Rift, J. Afr. Earth Sci., 31(2), 359–368.

Ayele, A., and R. Arvidsson (1998), Fault mechanisms and tectonic impli-cation of the 1985–1987 earthquake sequence in south western Ethiopia,J. Seismol., 1, 383–394.

Ayele, A., and O. Kulhanek (1997), Spatial and temporal variation of theseismicity in the Horn of Africa from 1960–1993, Geophys. J. Int., 130,805–810.

Ayele, A., G. W. Stuart, I. D. Bastow, and D. Keir (2006a), The August2002 earthquake sequence in north Afar: Insights into the neotectonics ofthe Danakil microplate, J. Afr. Earth Sci., in press.

Ayele, A., A. A. Nyblade, C. A. Langston, M. Cara, and J.-J. Leveque(2006b), New evidence for Afro-Arabian plate separation in southernAfar, in The Structure and Evolution of the East African Rift System inthe Afar Volcanic Province, edited by G. Yirgu, C. J. Ebinger, and P. K.H. Maguire, Geol. Soc. Spec. Publ., 259, in press.

Bastow, I. D., G. W. Stuart, J.-M. Kendall, and C. J. Ebinger (2005), Uppermantle seismic structure in a region of incipient continental breakup:Northern Ethiopian rift, Geophys. J. Int., 162, 479–493.

Benoit, M. H., A. A. Nyblade, J. C. VanDecar, and H. Gurrola (2003),Upper mantle P wave velocity structure and transition zone thicknessbeneath the Arabian Shield, Geophys. Res. Lett., 30(10), 1531,doi:10.1029/2002GL016436.

Bilham, R., R. Bendick, K. Larson, P. Mohr, J. Braun, S. Tesfaye, andL. Asfaw (1999), Secular and tidal strain across the Main Ethiopian Rift,Geophys. Res. Lett., 26, 2789–2792.

Boccaletti, M., M. Bonini, R. Mazzuoli, B. Abebe, L. Piccardi, andL. Tortorici (1998), Quaternary oblique extensional tectonics in the Ethio-pian Rift (Horn of Africa), Tectonophysics, 287, 97–116.

Buck, W. R. (2004), Consequences of asthenospheric variability on con-tinental rifting, in Rheology and Deformation of the Lithosphere at Con-tinental Margins, edited by G. D. Karner et al., pp. 1–30, ColumbiaUniv. Press, New York.

Calais, E., C. DeMets, and J.-M. Nocquet (2003), Evidence for a post-3.16–Ma change in Nubia-Eurasia-North America plate motions, EarthPlanet. Sci. Lett., 216, 81–92.

Calais, E., C. Ebinger, C. Hartnady, and J. M. Nocquet (2006), Kinematicsof the East African rift from GPS and earthquake slip vector data, in TheStructure and Evolution of the East African Rift System in the AfarVolcanic Province, edited by G. Yirgu, C. J. Ebinger, and P. K. H.Maguire, Geol. Soc. Spec. Publ., 259, in press.

Casey, M., C. Ebinger, D. Keir, R. Gloaguen, and F. Mohamed (2006),Strain accommodation in transitional rifts: Extension by magma intrusionand faulting in Ethiopian rift magmatic segments, in The Structure andEvolution of the East African Rift System in the Afar Volcanic Province,edited by G. Yirgu, C. J. Ebinger, and P. K. H. Maguire, Geol. Soc. Spec.Publ., 259, in press.

Cattin, R., C. Doubre, J.-B. de Chabalier, G. King, C. Vigny, J.-P. Avouac,and J.-C. Ruegg (2005), Numerical modelling of Quaternary deformationand post-rifting displacement in the Asal-Ghoubbet rift (Djibouti, Africa),Earth Planet. Sci. Lett., 239, 352–367.

Chu, D., and R. G. Gordon (1999), Evidence for motion between Nubia andSomalia along the Southwest Indian Ridge, Nature, 398, 64–67.

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 Planet. Sci. Lett., 193,423–436.

Figure 13. Cartoon sketch of the MER that shows abandoned Miocene border faults and localization ofstrain in <20-km-wide right-stepping en echelon magmatic segments that are zones of Quaternary magmaintrusion and faulting. Front face is constrained by topographic relief along the line of EAGLE controlledsource seismic profiles. Letters mark volcanoes: B, Boset; K, Kone; F, Fentale; D, Dofen.


15 of 17

B05314

Dugda, M., A. Nyblade, J. Julia, C. Langston, C. A. Ammon, and S. Simiyu(2005), Crustal structure in Ethiopia and Kenya from receiver functionanalysis: Implication for rift development in eastern Africa, J. Geophys.Res., 110, B01303, doi:10.1029/2004JB003065.

Dunbar, J., and D. Sawyer (1989), How pre-existing weaknesses control thestyle of continental breakup, J. Geophys. Res., 94, 7278–7292.

Eberhart-Philips, D., and A. J. Michael (1998), Seismotectonics of theLoma Prieta, California, region determined from three-dimensional Vp,Vp/Vs, and seismicity, J. Geophys. Res., 103, 21,099–21,120.

Ebinger, C., and M. Casey (2001), Continental brekaup in magmatic pro-vinces: An Ethiopian example, Geology, 29, 527–530.

Ebinger, C. J., and N. Hayward (1996), Soft plates and hot spots: Viewsfrom Afar, J. Geophys. Res., 101, 21,859–21,876.

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

Feigl, K. L., J. Gasperi, F. Sigmundsson, and A. Rigo (2000), Crustaldeformation near Hengill volcano, Iceland 1993–1998: Couplingbetween magmatic activity and faulting inferred from elastic modelingof satellite radar interferograms, J. Geophys. Res., 105, 25,655 –25,670.

Fernandes, R., B. Ambrosius, R. Noomen, L. Bastos, L. Combrinck,J. Miranda, and W. Spakman (2004), Angular velocities of Nubia andSomalia from continuous GPS data: Implications on present-day relativekinematics, Earth Planet. Sci. Lett., 222, 197–208.

Foster, A. N., and J. A. Jackson (1998), Source parameters of large Africanearthquakes: Implications for crustal rheology and regional kinematics,Geophys. J. Int., 134, 422–448.

George, R., N. Roger, and S. Kelley (1998), Earliest magmatism in Ethio-pia: Evidence for two mantle plumes in one flood basalt province, Geol-ogy, 26, 923–926.

Gouin, P. (1979), Earthquake History of Ethiopia and the Horn of Africa,258 pp., Int. Dev. Res. Centre, Ottawa, Ont.

Gutenberg, B., and C. F. Richter (1956), Earthquake magnitude, intensity,energy, and acceleration, Bull. Seismol. Soc. Am., 46, 105–145.

Hardebeck, J. L., and E. Hauksson (2001), Stress orientations obtainedfrom earthquake focal mechanisms: What are appropriate uncertaintyestimates, Bull. Seismol. Soc. Am., 91, 250–262.

Harris, W. C. (1844), The Highlands of Ethiopia, vol. 3, Longman, NewYork.

Haslinger, F., E. Kissling, J. Ansorge, D. Hatzfeld, E. Papadimitriou,V. Karakostas, K. Makropoulos, H.-G. Kahle, and Y. Peter (1999), 3Dcrustal structure from local earthquake tomography around the Gulf ofArta (Ionian region, NW Greece), Tectonophysics, 304, 201–218.

Hayward, N., and C. Ebinger (1996), Variations in along-axis segmentationof the Afar rift system, Tectonics, 15, 244–257.

Hofmann, C., V. Courtillot, F. Feraud, and P. Rochette (1997), Timing ofthe Ethiopian flood event: Implications for plume birth and globalchange, Nature, 389, 838–841.

Hofstetter, R., and M. Beyth (2003), The Afar Depression: Interpretation ofthe 1960–2000 earthquakes, Geophys. J. Int., 155, 715–732.

Hutton, L. K., and D. M. Boore (1987), The ML scale in southern Califor-nia, Bull. Seismol. Soc. Am., 77, 2074–2094.

Ibs-von Seht, M., S. Blumenstein, R.Wagner, D. Hollnack, and J.Wohlenberg(2001), Seismicity, seismotectonics and crustal structure of the southernKenya Rift: New data from the Lake Magadi area, Geophys. J. Int., 146,439–453.

Jestin, F., P. Huchon, and J. M. Gaulier (1994), The Somalia plate and theEast African Rift System: Present-day kinematics, Geophys. J. Int., 116,637–654.

Jones, L. M. (1988), Focal mechanisms and the state of stress on theSan Andreas fault in southern California, J. Geophys. Res., 93, 8869–8891.

Kastrup, U., M. L. Zoback, N. Deichmann, K. F. Evans, D. Giardini, andA. J. Michael (2004), Stress field variation in the Swiss Alps and thenorthern Alpine foreland derived from inversion of fault plane solutions,J. Geophys. Res., 109, B01402, doi:10.1029/2003JB002550.

Kebede, F., W.-Y. Kim, and O. Kulhanek (1989), Dynamic source para-meters of the March–May 1969 Serdo earthquakes in central Afar, Ethio-pia, deduced from teleseismic body waves, J. Geophys. Res., 94, 5603–5614.

Keir, D., J.-M. Kendall, C. J. Ebinger, and G. W. Stuart (2005), Variationsin late syn-rift melt alignment inferred from shear-wave splitting in crus-tal earthquakes beneath the Ethiopian rift, Geophys. Res. Lett., 32,L23308, doi:10.1029/2005GL024150.

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.

Keranen, K., S. L. Klemperer, R. Gloaguen, and Eagle Working Group(2004), Three-dimensional seismic imaging of a protoridge axis in theMain Ethiopian rift, Geology, 32, 949–952.

Kidane, T., E. Platzman, C. Ebinger, B. Abebe, and P. Rochette (2006),Paleomagnetic constraints on Continental Breakup: Observations fromthe Main Ethiopian rift, in The Structure and Evolution of the EastAfrican Rift System in the Afar Volcanic Province, edited by G. Yirgu,C. J. Ebinger, and P. K. H. Maguire, Geol. Soc. Spec. Publ., 259, in press.

Kieffer, B., et al. (2004), Flood and shield basalts from Ethiopia: Magmasfrom the African superswell, J. Petrol., 45(4), 793–834.

Kissling, E., U. Kradolfer, and H. Maurer (1995), VELEST user’s guide:Short introduction, Inst. of Geophys. and Swiss Seismol. Serv., ETH,Zurich.

Klein, F. W. (2002), User’s guide to Hypoinverse-2000, a FORTRAN pro-gram to solve for earthquake locations and magnitudes, U.S. Geol. Surv.Open File Rep., 02-171, 1–123.

Korme, T., J. Chorowicz, B. Collet, and F. F. Bonavia (1997), Volcanicvents rooted on extension fractures and their geodynamic implications inthe Ethiopian Rift, J. Volcanol. Geotherm. Res., 79, 205–222.

Langston, C. A., R. Brazier, A. A. Nyblade, and T. J. Owens (1998), Localmagnitude scale and seismicity rate for Tanzania, East Africa, Bull. Seis-mol. Soc. Am., 88, 712–721.

Lister, G., M. Etheridge, and P. Symonds (1986), Detachment faultingand the evolution of passive continental margins, Geology, 14, 246–250.

Mackenzie, G. H., G. H. Thybo, and P. Maguire (2005), Crustal velocitystructure across the Main Ethiopian Rift: Results from 2-dimenionalwide-angle seismic modelling, Geophys. J. Int., 162, 996–1006.

Maguire, P. K. H., et al. (2003), Geophysical project in Ethiopia studiescontinental breakup, Eos Trans. AGU, 84, 337, 342–343.

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, in The Structure and Evolution ofthe East African Rift System in the Afar Volcanic Province, edited byG. Yirgu, C. J. Ebinger, and P. K. H. Maguire, Geol. Soc. Spec. Publ.,259, in press.

Mahatsente, R., G. Jenztsch, and T. Jahr (1999), Crustal structure of theMain Ethiopian Rift from gravity data: 3-dimensional modelling, Tecto-nophysics, 313, 363–382.

Michael, A. J. (1984), Determination of stress from slip data: Faults andfolds, J. Geophys. Res., 89, 11,517–11,526.

Michael, A. J. (1987a), The use of focal mechanisms to determine stress: Acontrol study, J. Geophys. Res., 92, 357–368.

Michael, A. J. (1987b), Stress rotation during the Coalinga aftershocksequence, J. Geophys. Res., 92, 7963–7979.

Montelli, R., G. Nolet, F. A. Dahlen, G. Masters, E. R. Engdahl, and S.-H.Hung (2004), Finite-frequency tomography reveals a variety of plumes inthe mantle, Science, 303, 338–343.

Musumeci, C., O. Cocina, P. De Gori, and D. Patane (2004), Seismologicalevidence of stress induced by dike injection during the 2001 Mt. Etnaeruption, Geophys. Res. Lett., 31, L07617, doi:10.1029/2003GL019367.

Pizzi, A., M. Coltorti, B. Abebe, L. Disperati, G. Sacchi, and R. Salvini(2006), The Wonji Fault Belt (Main Ethiopian Rift, Ethiopia): Structuraland geomorphological constraints and GPS monitoring, in The Structureand Evolution of the East African Rift System in the Afar Volcanic Pro-vince, edited by G. Yirgu, C. J. Ebinger, and P. K. H. Maguire, Geol. Soc.Spec. Publ., 259, in press.

Rooney, T. O., T. Furman, G. Yirgu, and D. Ayalew (2005), Structure of theEthiopian lithosphere: Xenolith evidence in the Main Ethiopian Rift,Geochem. Cosmochim. Acta, 69, 3889–3910.

Rubin, A. M. (1992), Dike-induced faulting and graben subsidence involcanic rift zones, J. Geophys. Res., 97, 1839–1858.

Rubin, A. M., and D. Gillard (1998), Dike-induced earthquakes: Theore-tical considerations, J. Geophys. Res., 103, 10,017–10,030.

Schilling, J.-G. (1973), Afar mantle plume: Rare earth evidence, Nature,242, 2–5.

Shi, Y., and B. A. Bolt (1982), The standard error of the magnitude-frequencyb-value, Bull. Seismol. Soc. Am., 72, 1677–1687.

Snoke, J. A., A. C. Munsey, A. C. Teague, and G. A. Bollinger (1984), Aprogram for focal mechanism determination by combined use of polarityand SV-P amplitude ratio data, Earthquake Notes, 55(3), 15.

Stuart, G. W., I. D. Bastow, and C. J. Ebinger (2006), Crustal structure ofthe northern Main Ethiopian rift from receiver function studies, in TheStructure and Evolution of the East African Rift System in the AfarVolcanic Province, edited by G. Yirgu, C. J. Ebinger, and P. K. H.Maguire, Geol. Soc. Spec. Publ., 259, in press.

Tiberi, C., C. Ebinger, V. Ballu, G. Stuart, and B. Oluma (2005), Inversemodels of gravity data from the Red Sea - Aden - East African rifts triplejunction zone, Geophys. J. Int., 163(2), 775–787, doi:10.1111/j.1365-246X.2005.02710.x.

Tongue, J. A., P. K. H. Maguire, and P. Burton (1992), Seismicity distribu-tion from temporary earthquake recording networks in Kenya, Tectono-physics, 204, 71–79.


16 of 17

B05314

Ukstins, I., P. Renne, E. Wolfenden, J. Baker, and M. Menzies (2002),Matching conjugate volcanic rifted margins: 40Ar/39Ar chrono-stratigra-phy of pre- and syn-rift bimodal flood volcanism in Ethiopia and Yemen,Earth Planet. Sci. Lett., 198, 289–306.

Williams, F. M., M. A. J. Williams, and F. Aumento (2004), Tensionalfissures and crustal extension rates in the northern part of the MainEthiopian Rift, J. Afr. Earth Sci., 38, 183–197.

WoldeGabriel, G., J. Aronson, and R. Walter (1990), Geology, geochronol-ogy, and rift basin development in the central sector of the Main Ethio-pian rift, Geol. Soc. Am. Bull., 102, 439–458.

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 Planet. Sci. Lett., 224, 213–228.

Wolfenden, E., C. Ebinger, G. Yirgu, P. Renne, and S. P. Kelley (2005),Evolution of the southern Red Sea rift: Birth of a magmatic margin, Geol.Soc. Am. Bull., 117, 846–864.

A. Ayele, Geophysical Observatory, Addis Ababa University, P.O. Box

1176, Addis Ababa, Ethiopia.E. Daly, Department of Earth and Ocean Sciences, National University of

Ireland, Galway, UK.C. J. Ebinger and D. Keir, Department of Geology, Royal Holloway

University of London, Egham TW20 0EX, UK. ([email protected])G. W. Stuart, School of Earth and Environment, University of Leeds,

Leeds LS2 9JT, UK.


17 of 17

B05314

Name /mea_ssa966_605411/966_06051/Mp_1 09/21/2006 10:57AM Plate # 0 pg 1 #

1

Bulletin of the Seismological Society of America, Vol. 96, No. 6, pp. 1–, December 2006, doi: 10.1785/0120060051

Local Earthquake Magnitude Scale and Seismicity Rate

for the Ethiopian Rift

by Derek Keir, G. W. Stuart, A. Jackson,* and A. Ayele

Abstract A calibrated local earthquake magnitude scale is essential for quanti-tative analyses of seismicity. In Ethiopia, effective monitoring of earthquakes andresulting assessment of seismic hazard are especially important as regions with seis-mic and volcanic activity coincide with regions of economic significance and popu-lation growth. We have developed a local magnitude (ML) scale for the northernMain Ethiopian rift (MER) using earthquake data collected during 2001–2003 on 122three-component broadband seismic stations. Waveform data from 2139 local earth-quakes were corrected for instrument response and convolved with the nominalWood–Anderson torsion seismograph response appropriate for the original definitionof local magnitude. The hypocentral distances considered are 5 to 800 km, with thebest represented range from 5 to 150 km. A total of 30,908 maximum zero-to-peakamplitudes (AWA) were incorporated into a direct linear inversion for individual earth-quake local magnitudes (ML), 244 station factors (C), and 2 linear distance-dependentfactors (n, K) in the distance correction term, log (Ao), of the equation for localmagnitude: ML log(AWA) log(Ao) C. The resulting distance correction isgiven by log(Ao) 1.196997log(r/17) 0.001066(r 17) 2, which impliesthat ground-motion attenuation is relatively high, consistent with ongoing magmaintrusion and the presence of shallow magma reservoirs beneath the MER. Stationcorrections significantly reduce ML residuals, which range between 0.42 ML units.The catalog of earthquakes is complete above ML 2.1 and the annual cumulativeseismicity rate follows the relation log N 4.5 1.13 ML. Our results are criticalfor accurate routine quantitative analysis of past, current and, future seismicity inEthiopia.

Introduction

The northern Main Ethiopian rift (MER) and Afar riftsare virtually the only places worldwide where the transitionbetween continental and oceanic rifting is subaerially ex-posed. As part of project EAGLE (Ethiopia Afar Geoscien-tific Lithospheric Experiment), up to 179 broadband seismicstations were deployed across a 250 350 km area of theEthiopian rift and adjacent uplifted plateaus (e.g., Bastow etal., 2005 Maguire et al., 2006) (Fig. 1). The data were col-lected from the EAGLE network and seismicity was analyzedto study the pattern of strain localization just prior to con-tinental breakup (Keir et al., 2006). We aim to quantify thesize of local earthquakes in our dataset by accurately esti-mating local earthquake magnitude (ML). The wealth ofbroadband waveforms in the seismicity dataset allows us toundertake a direct inversion of earthquake-amplitude mea-

*Present address: Institut fur Geophysik, ETHZ Honggerberg, CH-8093Zurich, Switzerland: [email protected].

surements for a local magnitude scale based on the originaldefinition proposed by Richter (1935, 1958). The calibratedmagnitude scale is then used to calculate the annual cumu-lative frequency-magnitude distribution of seismicity in theMER.

A calibrated earthquake-magnitude scale based on ML

is of great importance for seismic-hazard studies (Bormann,2002). Attenuation curves that correct for the decrease inseismic-wave amplitude with distance differ from region toregion and the use of an inappropriate curve can result inmiscalculation of earthquake magnitude by over 1 ML units,even at hypocentral distances of less than 300 km (Fig. 2).Probabilistic hazard analysis requires details of magnitudestatistics (e.g., maximum magnitudes and the b-value of thecumulative frequency-magnitude distribution), which re-quire accurate magnitude estimates to determine earthquake-recurrence relationships. The combination of the sparse sta-tion distribution, lack of a calibrated local-magnitude scale,and low number of earthquakes recorded on global, regional,


2 D. Keir, G. W. Stuart, A. Jackson, and A. Ayele

Figure 2. Attenuation curves for southern Cali-fornia (Richter, 1935, Hutton and Boore, 1987), forTanzania (Langston et al., 1998), and for Ethiopia(from our study).

Figure 1. (Top) Distribution of the 2139 earth-quakes recorded from October 2001 to January 2003in the MER and Afar rifts. Size of earthquake epicen-ters is scaled by magnitude. The white star is the lo-cation of NEIC-reported earthquake 1 December 200211:18 (mb Preliminary Determination of Epicenter[PDE] 4.9). The box encloses the location of theEAGLE network of broadband seismic stations. (Bot-tom) Location of EAGLE network broadband seismicstations. Dark triangles with station names are CMG-3T and CMG-40TD stations that operated from Oc-tober 2001 to January 2003. Light triangles are CMG-6TD stations that operated from October 2002 toJanuary 2003. Circles are CMG-6TD stations that op-erated from November 2002 to January 2003. IRIS/GSN permanent broadband station FURI is shown asa white square. Rift bounding Miocene border faults(BF) are shown with thick black lines and dip ticks.Right-stepping, en echelon magmatic segments (MS)along the rift axis are shaded grey.

and local catalogs has meant that reliable earthquake-magnitude statistics for the MER have not been calculated(Ayele and Kulhanek, 1997). Earthquake magnitude is alsoimportant in integrated seismic and geodetic studies that aimto understand lithospheric deformation processes in rift sys-tems by quantifying relative amounts of seismic and ase-ismic strain (e.g., Hofstetter and Beyth, 2003; Bendick et al.,2006). The attenuation curve derived from a local magnitudescale is also useful for risk assessment in engineering prac-tice as the frequency band of the Wood–Anderson seismom-eter (0.8–10 Hz) is in the range of most engineering struc-tures. However, measurements on seismic-wave propagationin Ethiopia are lacking (Kebede and van Eck, 1997;Mammo, 2005).

Seismic-hazard assessment is important in Ethiopia be-cause regions with seismic activity coincide with regions ofeconomic significance and population growth (e.g., Gouin,1979; Kebede and Kulhanek, 1991, 1994). The potentialseismic and volcanic hazard in volcanic rift zones in Ethi-opia was highlighted by the recent rifting episode in thenorthern Afar rift. From 20 September to 8 October 2005,162 earthquakes of mb 4.0 and a volcanic eruption oc-curred within a 60-km-long rift segment. Disruptioncaused by ground shaking, surface fissuring, and ash depos-its caused the displacement of 6000 pastoralists from theregion (Yirgu et al., 2005). Radar interferometry (InSAR)shows that up to 8 m of horizontal opening occurred dur-ing the rifting event with seismic-moment release accountingfor less than 7% of observed deformation (Wright et al.,2006). Most extension was likely accommodated by dikeintrusion.

Seismicity in Ethiopia is currently monitored by fivepermanent broadband seismic stations, including the Incor-porated Research Institutions for Seismology/Global Seis-


Local Earthquake Magnitude Scale and Seismicity Rate for the Ethiopian Rift 3

mographic Network (IRIS/GSN) station FURI, maintained bythe Geophysical Observatory Addis Ababa University. TheGeophysical Observatory records earthquake coda lengthand uses the relationship Md 1.97 log s 0.0008D 1.28, where s is coda length in seconds, and D is hypocentraldistance in kilometers to determine duration magnitude (Md)(Asfaw, 1988). No relationship between displacement am-plitude and local magnitude (ML) has yet been determined,however. In addition to the permanent stations, seismicity ofnorth Afar has been monitored by a network of eight three-component broadband seismic stations since October 2005.A calibrated magnitude scale is critical for accurate quanti-tative monitoring of past, ongoing, and future seismic activ-ity in Ethiopia.

Tectonic Setting

The Miocene-Recent MER constitutes the northern partof the East African rift system and forms the youngest armof the Afar triple junction, which developed in an Eocene-Oligocene flood basalt province (Fig. 1). The MER isbounded by northeast-trending Miocene border faults. SinceQuaternary times extensional strain has localized in 20-km-wide right-stepping en echelon magmatic segments,which are zones of north-northeast-striking fissures, faults,and aligned volcanic cones (Bilham et al., 1999; Ebingerand Casey, 2001).

Most of the seismicity from October 2001 to January2003 occurred at the oblique intersection between the south-ern Red Sea rift and MER near Ankober, and within theFentale–Dofen magmatic segment (Keir et al., 2006)(Fig. 1). Earthquakes were predominantly 8–14 km deep(Keir et al., 2006). Records of seismicity over the past 150years show a similar pattern with seismicity concentratednear Ankober and within the en echelon magmatic segmentsalong the axis of the MER (e.g., Gouin, 1979; Asfaw, 1982;Kebede and Kulhanek, 1994; Ayele and Kulhanek, 1997;Hofstetter and Beyth, 2003). The spatial correlation betweenseismicity and aligned cones in magmatic segments suggeststhat earthquakes are induced by dike injection, and the tem-poral patterns of seismicity suggest that magmatic segmentsare fed from discrete sources of magma and that extensionoccurs mainly in episodic dike-injection events (Keir et al.,2006).

Geodetic, structural, and seismic studies show that themajority of strain across the MER is accommodated by dikeinjection beneath magmatic segments. A comparison of seis-mic-moment release and total opening strain predicted fromplate separation rates shows that 50% of strain is accom-modated aseismically (Hofstetter and Beyth, 2003). Keranenet al. (2004) interpret high-velocity anomalies in controlledsource tomographic images as 20-km-wide axial maficintrusions at 10 km depth beneath magmatic segments.S-wave splitting in local earthquakes shows the fast-polarization direction in magmatic segments parallel toaligned volcanic cones with the amount of splitting highest

in regions of probable partial melt. From these observationsKeir et al. (2005) proposed that anisotropy of the upper10 km beneath the rift is caused by aligned melt and dikes.Magma intrusions are likely not restricted to crust beneathrift valley magmatic segments as midcrustal conductiveanomalies in MT data indicate the presence of partial meltin the lower midcrust beneath the Ethiopian plateau (Whalerand Hautot, 2006).

Seismic studies that probe mantle structure provide con-straints on the distribution of partial melt at depth. Anoma-lous low-velocity zones in the upper asthenosphere are seg-mented beneath the rift valley and also impinge beneath theEthiopian plateau. Comparison of relative P- and S-arrivaltimes shows that they are likely due to a combination of hightemperatures and partial melt (Bastow et al., 2005). Both thelarge amount of SKS splitting and the rift parallel orientationof the fast polarization direction led Kendall et al. (2005) topropose that partial melt beneath the MER rises throughdikes that penetrate the thinned lithosphere. In addition, Svand Sh velocity models derived from surface-wave disper-sion curves are consistent with a model of anisotropy due toaligned melt-filled pockets from 20–75 km depth beneaththe rift (Kendall et al., 2006).

Amplitude Data

Local earthquakes were recorded on 29 broadband seis-mic stations (EAGLE Phase I) that were operational betweenlate October 2001 and January 2003 (Fig. 1). These three-component Guralp CMG-3T and CMG-40TD instrumentsrecorded continuous data at 50 Hz. Additional data duringthis period were acquired from the permanent IRIS/GSNbroadband station FURI. The number of seismic stations wasincreased during the final 4 months of the experiment withthe deployment of an additional 50 CMG-6TD broadbandinstruments recording at 100 Hz. These stations were de-ployed at 15-km spacing mainly in the rift valley, and theywere operational between October 2002 and February 2003(EAGLE Phase II). A further 100 broadband CMG-6TD in-struments deployed at 5-km spacing across the rift valleyand adjacent plateaus were operational between November2002 and January 2003 (EAGLE Phase III).

During the 16 months of the EAGLE passive experiment,2139 earthquakes recorded at four or more stations werelocated with the Hypo2000 algorithm (Klein et al., 2002)(Fig. 1). The earthquakes were located using a 1D velocitymodel determined by simultaneously relocating earthquakesand inverting for velocity structure (Daly et al., in press);1957 earthquakes were located within the EAGLE network.We estimate hypocenter accuracy of 600 m in horizontaldirections and 1500 m in depth for earthquakes recordedat eight or more stations (Keir et al., 2006). Earthquakeswithin the network recorded by between four and eight sta-tions have estimated hypocental uncertainties of 5 km inboth horizontal directions and depth. Earthquakes located



Figure 3. Distance/magnitude distribution of thedata available for the horizontal components. Mag-nitudes are estimated with the new distance-correc-tion terms for Ethiopia.

outside the network have estimated hypocental uncertaintiesof up to 20 km in horizontal directions and 10 km depth.

Local magnitude was originally defined by Richter(1935) using ground motions recorded on a standard hori-zontal Wood–Anderson torsion seismograph. Therefore, theEAGLE broadband data were corrected for the instrumentresponse of the CMG-3T, CMG-40TD, and CMG-6TD seis-mometers. The displacement ground motions were con-volved with the standard Wood–Anderson response: mag-nification of 2800, damping ratio of 0.8, and natural periodof 0.8 sec (Anderson and Wood, 1925; Kanamori and Jen-nings, 1978). We measured the maximum absolute value ofthe zero-to-peak amplitude in millimeters of the north–southand east–west horizontal component seismograms. Stationswith malfunctioning horizontal components were removedfrom the dataset. The dataset includes 15,456 amplitudemeasurements on each horizontal component, a total of30,908 measurements (Fig. 3). The hypocentral distancesconsidered range from 5 to 800 km, with the best representedrange being from 5 to 150 km (Fig. 3).

Methodology

We use the equation of Richter (1935, 1958)

M log(A ) log(A ) C, (1)L WA o

where AWA is zero-to-peak amplitude measured on a standardhorizontal Wood–Anderson seismograph, log(Ao) is a dis-tance correction term, and C is a correction term for indi-vidual stations. We determine the attenuation curve, log(Ao),by using the parametric approach (Bakun and Joyner, 1984).The major advantages of the parametric form of the atten-uation curve are that it considers simple expressions of geo-metrical spreading and attenuation, and is represented byonly a few coefficients. This facilitates straightforward es-timation of local magnitude using a single equation at allhypocentral distances (e.g., Hutton and Boore, 1987; Kim,1998; Langston et al., 1998; Kim and Park; 2005). On aglobal scale, the standardization of the local magnitude cal-culation using the parametric form of the attenuation curveis recommended (Ortega and Quintanar, 2005). A drawbackof the parametric approach is that the nonparametric expres-sion of the attenuation curve better represents crustal andupper-mantle complexities (e.g., Anderson and Lei, 1994;Savage and Anderson, 1995; Baumbach et al., 2003; Bragatoand Tento, 2005).

Richter’s original local magnitude scale is defined suchthat an earthquake of ML 3 will cause a 1-mm zero-to-peakdeflection of the Wood–Anderson seismogram at 100 kmfrom the hypocenter. Hutton and Boore (1987) observed thatif the attenuation within the first 100 km has a large geo-graphic variation, earthquakes in different regions with thesame ML may have very different ground motions near thesource, thus making ML a poor measure of source size. Toavoid this difficulty, they suggest that local magnitudes be

normalized to motions at closer distances to avoid most ofthe regional differences in wave propagation, using a 10-mm deflection of the Wood–Anderson seismogram at 17 kmfrom the hypocenter for a ML 3 earthquake, consistent withthe original definition of the local magnitude scale. The shiftto normalization nearer the source should only be used if thedistribution of data is such that the attenuation curve can beevaluated with sufficient precision down to the new refer-ence distance (Alsaker et al., 1991). This is the case withour dataset, which includes 2564 amplitude measurements(8% of the dataset) at hypocentral distances of less than 25km (Fig. 3). The distance correction term is thus defined as

log(A ) n log(r/17) K(r 17) 2, (2)o

where r is hypocentral distance in kilometers, and n and Kare parameters related to the geometrical spreading and at-tenuation of S waves in the region, respectively.

If equations (1) and (2) are combined, the observed am-plitude, Aijk, is modeled by

log(A ) 2 n log(r /17)ijk ij

K(r 17) M C , (3)ij Li jk

where index i labels events, index j labels stations, and indexk labels the component (north–south or east–west). The ob-jective of the inversion is to determine n, K, ML, and C.There are two station factors per station corresponding to theeast–west and north–south horizontal components. The sys-tem of equations includes a constraint that the mean of sta-



Figure 4. Magnitude estimated at stations of vary-ing hypocentral distances for earthquake 1 December2002 11:18 (mb PDE 4.9) with three different at-tenuation curves (a) magnitude estimated with thedistance-correction terms of Hutton and Boore (1987)for southern California; (b) magnitude estimated withthe distance correction terms of Langston et al.(1998); (c) magnitude estimated with the new mag-nitude scale for Ethiopia. Straight lines are best-fit tothe data and show that the Hutton and Boore (1987)magnitude scale overestimates magnitudes with in-creasing hypocentral distances, whereas the Langstonet al. (1998) scale underestimates magnitude with in-creasing hypocentral distance. The new magnitudescale for the MER estimates consistent magnitudesacross varying hypocentral distances and the averagelocal magnitude (ML 4.79) is that expected for an mb

4.9 earthquake.

tion factors is zero. The observations on the left-hand sideof equation (3) are linearly related to the unknowns, whichwe arrange in a model vector m. We have Ne events and Ns

stations, and thus have a total of (Ne 2Ns) 2 unknowns.The N observations log(Aijk) 2 are arranged into the N-vector d. We write the overdetermined set of equations (3)in the form

d Am, (4)

which we solve using the conventional least-squares crite-rion; the optimal solution satisfies

T 1 Tm (A A) A d (5)

The linear system (5) has a total of 2385 parameters and30,908 data; it can be solved in less than an hour on a modestworkstation. Our approach leads directly to an optimal so-lution and is different from the iterative procedure used todetermine m (e.g., Langston et al., 1998). Pujol (2003) testedthe direct inversion method on data from Tanzania previ-ously analyzed with the iterative technique (Langston et al.,1998). Similar results were achieved but the major advan-tages of the direct inversion are that the solution is indepen-dent of the starting values for the unknowns.

Results

Magnitude Scale for the MER

The distance correction, log(Ao), term from the inver-sion using 17-km distance normalization is given by:

log(A ) 1.196997 log(r/17)o

0.001066(r 17) 2 (6)

where r is hypocentral distance in kilometers (Fig. 2).The errors on the estimates of n and K can be determined

from the posterior covariance matrix. The two-by-two sub-section of the entire 2385 by 2385 covariance matrix is prac-tically diagonal, showing that the estimates of n and K arevirtually independent. One can rigorously characterize thisby calculating the eigenvectors of the (n, K) section of thematrix; the ellipse describing the one standard deviation con-tour has a semimajor axis of length 0.025 and semiminoraxis of length 9.7 105 and is oriented with the semimajoraxis practically parallel to the n axis (the angle between themis 0.25). These values for the ellipse lengths are similar tothe values for the one-sigma standard deviations of the pa-rameters n and K.

The new distance-correction terms compensate cor-rectly for the reduction in amplitude with increasing distance(Figs. 4 and 5). For example, earthquake 1 December 200211:18 (mb PDE 4.9) was the nearest and most widelyrecorded earthquake on the EAGLE network that was re-ported by National Earthquake Information Center (NEIC).



Figure 5. Mean magnitude residuals calculatedper 50-km bin intervals with error bars marked by thestandard deviation in the mean magnitude residuals.Magnitude residuals were computed as the differencebetween magnitude assigned by a single station andthe average magnitude of the earthquake. The lack ofsignificant variation in mean magnitude residualswith distance shows that possible complexities incrustal and upper-mantle structure do not have a sys-tematic effect on variations in attenuation with dis-tance.

Figure 6. Magnitude residuals/hypocentral dis-tance distributions for both the north–south and east–west components. (a) Magnitude residuals withouttaking into account station corrections. The standarddeviation is 0.24 and variance r2 is 0.058. (b) Mag-nitude residuals with station corrections taken into ac-count. The standard deviation is 0.18 and variance r2

is 0.032. Therefore, adopting the station correctionsreduced variance by 45%. The average of residuals,both with and without station corrections considered,is nearly zero.

The earthquake is located 200 km north of station KAREand was recorded with a high signal-to-noise ratio by 72EAGLE broadband stations at hypocentral distances of 200–600 km (Fig. 4). Local magnitude was estimated at eachstation using the magnitude scale for Tanzania (Langston etal., 1998), southern California (Hutton and Boore, 1987),and Ethiopia. The magnitude estimated at each station usingthe Tanzania magnitude scale decreases with increasing hy-pocentral distance and the average magnitude is ML 4.13. Incontrast, the magnitude at each station using the southernCalifornia scale increases with increasing hypocentral dis-tance and the average magnitude is ML 4.98. The new mag-nitude scale for Ethiopia estimates consistent magnitude forstations at different hypocentral distances and the averagemagnitude is ML 4.81.

Also, magnitude residuals for the whole dataset werecomputed as the difference between magnitude assigned bya single station for a given earthquake and the average mag-nitude of the same earthquake. The mean magnitude residualis calculated per 50-km bin interval with error bars markedby the standard deviation of the mean magnitude residuals(Fig. 5). Mean magnitude residuals vary 0.1 ML units tohypocentral distance of 700 km. An ML residual of 0.18 iscalculated from only 16 measurements at hypocentral dis-tances of 700 km. The lack of significant variation in meanmagnitude residuals with distance shows that possible com-plexities in crustal and upper-mantle structure do not have asystematic effect on variations in attenuation with distancebeneath the MER. Our parametric expression of the attenu-ation curve thus represents a simple model that adequatelycompensates for the decay of amplitude with increasing dis-tance.

Local Magnitude Values and Station Corrections

The inversion procedure solved for correction factors ofboth north–south and east–west components at individualstations. Magnitude residuals were calculated with and with-out computed station corrections, C, taken into account(Fig. 6). For magnitude residuals calculated without stationcorrection, the average of residuals on the north–south andeast–west components is nearly zero and the standard de-viation is 0.24 (variance, r2, is 0.058). For magnitude resid-uals calculated with station corrections, the average of re-siduals on the north–south and east–west components innearly zero, and the standard deviation is 0.18 (r2 is 0.032).Therefore, adopting the station corrections reduced varianceby 45%.

The north–south component correction factors vary be-tween 0.41 and 0.34 ML units and the east–west compo-nent correction factors vary between 0.42 and 0.33 ML

units (Fig. 7). Most stations have similar correction factorson the two horizontal components. Station corrections canvary dramatically over distances of 5 km and there is noconsistent difference between corrections at stations in therift valley and on the adjacent plateau. Thus, the spatial vari-



Figure 7. Spatial variation of station factors on north–south component (a) and east–west component (b). Negative correction factors are shown as squares scaled by mag-nitude of the correction factor. Positive correction factors are shown as circles scaledby magnitude of the correction factor.

ation of station factors shows neither a clear correlation tomajor tectonic features nor to topographic relief and suggestsa strong influence of local site effects on variations in theamplitude of ground motion.

The station correction factors for the permanent IRIS/GSN station FURI are 0.16 ML units on the north–southcomponent and 0.14 ML units on the east–west compo-nent. Future permanent and temporary seismic-array deploy-ments in Ethiopia are likely to include earthquake recordsfrom FURI, and such studies will now be able to use ampli-tude measurements from this permanent station to calibratenew data with our magnitude scale for the MER.

Discussion

A comparison of the attenuation curves obtained fromEthiopia, southern California (Hutton and Boore, 1987), andTanzania shows that attenuation in Ethiopia is relativelyhigh. The attenuation curve computed for southern Califor-nia by Hutton and Boore (1987) is similar to Ethiopia, inparticular, at hypocentral distances of less than 300 km.Rifted regions with elevated geothermal gradients such asthe southwestern United States are in general characterizedby high attenuation of seismic waves (e.g., Richter, 1958;Hutton and Boore, 1987; Savage and Anderson, 1995). Insouthern California, high body-wave attenuation in the lowercrust beneath the Salton Trough and San Gabriel Mountainsis attributed to a combination of high temperatures and par-tial melt (Schlotterbeck and Abers, 2001). Our results of

relatively large amounts of attenuation in the MER are thusnot surprising considering the wealth of independent geo-physical and geological data that show evidence for partialmelt and magma intrusions in the crust and upper mantlebeneath the MER and adjacent Ethiopian plateau (e.g., Bas-tow et al., 2005; Keir et al., 2005; Kendall et al., 2005;Rooney et al., 2005).

The high attenuation observed in Ethiopia is signifi-cantly different from the East African rift system in Tanzaniawhere the crust and upper mantle have had little to no mod-ification by rifting processes (Langston et al., 2002). In Tan-zania, the combination of crystalline Archaean and Proto-rozoic crust, in conjunction with low geothermal gradientstypical of Archaean craton give rise to very efficient wavepropagation in the lithosphere (Langston et al., 1998; Weer-aratne et al., 2003).

The new magnitude scale for Ethiopia is used to inves-tigate seismicity of the MER for 2001–2003. Because of highattenuation in the MER, earthquakes located outside the net-work but recorded on EAGLE stations are ML 3 (Figs. 1and 2). Therefore, we calculated magnitude statistics of the1957 earthquakes located within the network of seismic sta-tions. This ensures we sample an earthquake catalog that isnot biased toward large earthquakes located outside the net-work and also ensures our magnitude statistics sample onlyearthquakes in the MER. Most earthquakes are of ML 1–2and the largest earthquake is ML 3.9 (Fig. 8a). The power-law cumulative frequency-magnitude distribution shows thatthe seismicity catalog is complete above ML 2.1 (Mc 2.1)



Figure 8. (a) Magnitude-frequency distribution of earthquakes recorded within thenetwork of seismic stations. Most of the earthquakes are of magnitude ML 1–2 and thehighest-magnitude earthquake is ML 3.9. (b) Gutenburg–Richter distribution of earth-quakes located within the network of seismic stations. Mc is the cutoff magnitude of2.1 and the slope shows b 1.13. The straight line intersects the y axis at y 4.5.

(Fig. 8b) (Gutenberg and Richter, 1954). A b-value of 1.13 0.05 was estimated from earthquakes larger than the ML

2.1 using the maximum-likelihood method (Aki, 1965) andan error estimate determined from the standard deviation ofb (Shi and Bolt, 1982). The cumulative annual seismicityrate is calculated from an annualized dataset and follows therelation log N 4.5 1.13 ML. Hofstetter and Beyth(2003) obtained a b-value of 0.83 0.08 from just 16 earth-quakes on global and regional catalogs that were locatedacross a larger area that encompasses both the MER andsouthern Ethiopian Rift as far south as 5 N.

The relatively high b-value of 1.13 for seismicity in theMER during 2001–2003 is consistent with seismic energybeing released mostly as swarms of lower magnitude (ML

4) earthquakes (Keir et al., 2006). This pattern of seis-micity is likely representative of longer-term deformationpatterns as previous studies of seismicity in the MER alsoreport lower-magnitude seismic swarms and a similar patternis evident in data from global and regional catalogs that showrelatively few larger magnitude (mb 4.0) earthquakes inthe MER (e.g., Gouin, 1979; Asfaw, 1982; Kebede and Kul-hanek, 1994). The observed lack of large-magnitude earth-quakes in the MER is consistent with geodetic data that showthat the majority of strain across the MER is accommodatedaseismically (Hofstetter and Beyth, 2003 Bendick et al.,2006). Studies of seismically active magmatic systems showthat Mc and b-value can vary in space and time (e.g., Wiemeret al., 1998 Murru et al., 1999). The b-value obtained in ourstudy is based on only 16 months of data from a 250 km 350 km area and may thus be biased by short-term spatialand temporal variations in the pattern of seismic activity. A

detailed appraisal of the recurrence interval of large-magnitude earthquakes can only be achieved with longer-term seismic monitoring in Ethiopia.

An estimated b-value of 1.13 for the MER is similar tob-values of 1.05–1.3 calculated for the southern Red Sea andGulf of Aden seafloor spreading centers (Ayele and Kulha-nek, 1997; Hofstetter and Beyth, 2003). Lower b-values ofbetween 0.7 and 0.9 are observed in the East African riftsystem in southern Ethiopia, Kenya, and Tanzania wheremoment release as large-magnitude earthquakes located onrift-bounding border faults accommodates the majority ofextension (e.g., Tongue et al., 1992; Langston et al., 1998).

Despite the lack of large earthquakes recorded over thepast 50 years in the MER, the recent dike-injection episodeand associated swarm of earthquakes, surface fissuring, andvolcanic eruption in Afar highlights the potential seismichazard of rift zones in Ethiopia. Seismicity within the mag-matic segments of the MER is likely controlled by episodicinjection of dikes (Keir et al., 2006). Although a major rift-ing event has not yet been directly observed in the MER,structural data suggest that episodes of surface fissuring andvolcanic eruptions have occurred in MER magmatic seg-ments during the last 10,000 years (e.g., Asfaw, 1982,1998; Williams et al., 2004). Despite the current period ofquiescence, hazards associated with seismicity and volcaniceruptions pose a serious risk to life and economy in the MER.

Conclusions

A local magnitude scale for Ethiopia has been devel-oped from 30,908 amplitude measurements on simulated



Wood–Anderson seismograms from 2139 earthquakes re-corded on 122 EAGLE broadband instruments. The newmagnitude scale uses a distance normalization of 10 mmmotion at 17 km distance for a ML 3.0 earthquake. The dis-tance correction is given by;

log(A ) 1.196997 log(r/17)o

0.001066(r 17) 2.0, (7)

where r is hypocentral distance in kilometers. The distancecorrection shows that ground-motion attenuation in Ethiopiais relatively high and is consistent with the presence of per-vasive magma intrusion and partial melt beneath the MER.

The catalog of events used in this study is completeabove ML 2.1. The annual cumulative seismicity rate in theMER is log N 4.5 1.13ML. The relatively high b-valueis consistent with the observed pattern of low magnitudeML 4 swarms of earthquakes in the MER and lack of large-magnitude earthquakes reported on global and regional cat-alogs over the past 50 years.

The recent swarm of 162 mb 4.0 earthquakes andvolcanic eruption and surface fissuring in the Dabbahu mag-matic segment in Afar highlights the potential seismic andvolcanic hazard in volcanic rift zones such as the MER. Ourresults are critical for current and future quantitative analysisof seismicity in Ethiopia, which is important for scientific,economic, and social development.

Acknowledgments

We thank SEIS-UK for the use of instruments and especially AlexBrisbourne for assistance in the field, with data management and analysis.Cindy Ebinger, Eve Daly, Dave Cornwell, and Ian Bastow are thanked fortheir significant contributions to this study. Laike Asfaw of the GeophysicalObservatory Addis Ababa University and Bekele Abebe, Dereje Ayalew,Gezahegn Yirgu, and Tesfaye Kidane are thanked for support throughoutthe project. Research was supported by NERC Grant NER/A/S/2000/01004and NERC Studentship NER/S/A/2002/10547.

References

Alsaker, A., L. B. Kvamme, R. A. Hansen, and A. Dahle (1991). The ML

scale in Norway, Bull. Seism. Soc. Am. 81, 379–398.Aki, K. (1965). Maximum likelihood estimate of b in the formula logN

a-bM and its confidence limits, Bull. Earthquake Res. Inst. 43, 237–239.

Anderson, J. A., and H. O. Wood (1925). Description and theory of thetorsion seismometer, Bull. Seism. Soc. Am. 15, 1–72.

Anderson, J. G., and Y. Lei (1994). Nonparametric description of peakacceleration as a function of magnitude, distance and site in Guerrero,Mexico, Bull. Seism. Soc. Am. 84, 1003–1017.

Asfaw, L. M. (1982). Development of earthquake induced fissures, Nature286, 551–553.

Asfaw, L. M. (1988). Catalogue of Ethiopian earthquakes, earthquake pa-rameters, strain release and seismic risk, in Proceedings of theSAREC-ESTC Conference on Research Development and CurrentResearch Activities in Ethiopia, G. Woldeyes (Editor).

Asfaw, L. M. (1998). Environmental hazard from fissures in the Main Ethi-opian rift, J. Afr. Earth Sci. 27, no. 3/4, 481–490.

Ayele, A., and O. Kulhanek (1997). Spatial and temporal variation of the

seismicity in the Horn of Africa from 1960–1993, Geophys. J. Int.130, 805–810.

Bakun, W. H., and W. B. Joyner (1984). The ML scale in central California,Bull. Seism. Soc. Am. 74, 1827–1843.

Bastow, I. D., G. W. Stuart, J-M. Kendall, and C. J. Ebinger (2005). Uppermantle seismic structure in a region of incipient continental breakup:northern Ethiopian rift, Geophys. J. Int. 162, 479–492.

Baumbach, M., D. Bindi, H. Grosser, C. Milkereit, S. Parolai, R. Wang, S.Karakisa, S. Zunbul, and J. Zschau (2003). Calibration of an ML scalein northwestern Turkey from 1999 Izmit aftershocks, Bull. Seism. Soc.Am. 93, 2289–2295.

Bendick, R., S. McClusky, R. Bilham, L. Asfaw, and S. Klemperer (2006).Distributed Nubia-Somalia relative motion and dike intrusion in theMain Ethiopian rift, Geophys. J. Int. 165, 303–310.

Bilham, R., R. Bendick, K. Larson, P. Mohr, J. Braun, S. Tesfaye, and L.Asfaw (1999). Secular and tidal strain across the Main Ethiopian rift,Geophys. Res. Lett. 26, 2789–2792.

Bormann, P. (2002). Magnitude of seismic events, in IASPEI New Manualof Seismological Observatory Practice. Vol. 1, P. Bormann (Editor),GeoForschungsZentrum, Potsdam, 16–50.

Bragato, P. L., and A. Tento (2005). Local magnitude in northeastern Italy,Bull. Seism. Soc. Am. 95, 579–591.

Daly, E., D. Keir, C. J. Ebinger, G. W. Stuart, A. Ayele, and I. D. Bastow(2006). Crustal tomographic imaging of a transitional continental rift:the Ethiopian rift, Africa, Geophys. J. Int. (in press).

Ebinger, C., and M. Casey (2001). Continental breakup in magmatic prov-inces, an Ethiopian example, Geology 29, 527–530.

Gouin, P. (1979). Earthquake History of Ethiopia and the Horn of Africa,International Development Research Centre (IDRC), Ottawa, Ontario,258 pp.

Gutenberg, B., and C. F. Richter (1956). Earthquake magnitude, intensity,energy, and acceleration, Bull. Seism. Soc. Am. 46, 105–145.

Hofstetter, R., and M. Beyth (2003). The Afar Depression: interpretationof the 1960–2000 earthquakes, Geophys. J. Int. 155, 715–732.

Hutton, L. K., and D. M. Boore (1987). The ML scale in southern California,Bull. Seism. Soc. Am. 77, 2074–2094.

Kanamori, H., and P. C. Jennings (1978). Determination of local magni-tude, ML, from strong motion accelerograms, Bull. Seism. Soc. Am.68, 471–485.

Kebede, F., and O. Kulhanek (1991). Recent seismicity of the East Africanrift system and its implications, Phys. Earth Planet. Interiors 68,259–273.

Kebede, F., and O. Kulhanek (1994). Spatial and temporal variations in b-values along the East African rift system and the southern Red Sea,Phys. Earth Planet. Interiors 83, 249–264.

Kebede, F., and T. van Eck (1997). Probabilistic seismic hazard assessmentfor the Horn of Africa based on seismotectonic regionalization, Tec-tonophysics 270, 221–237.

Keir, D., C. J. Ebinger, G. W. Stuart, E. Daly, and A. Ayele (2006). Strainaccommodation by magmatism and faulting as rifting proceeds tobreakup: seismicity of the northern Ethiopian rift, J. Geophys. Res.111, no. B5, B05314, doi 1029/2005JB003748.

Keir, D., J-M. Kendall, C. J. Ebinger, and G. W. Stuart (2005). Variationsin late syn-rift melt alignment inferred from shear-wave splitting incrustal earthquakes beneath the Ethiopian rift, Geophys. Res. Lett. 32,L23308, doi 10.1029/2005GL024150.

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, A. Ayele, and G. W. Stuart(2006). Mantle upwellings, melt migration and the rifting of Africa:insights from seismic anisotropy, in The Afar Volcanic Provincewithin the East African Rift System, G. Yirgu, C. J. Ebinger, andP. K. H. Maguire (Editors), Geol. Soc. Spec. Publ. 259, 55–72.

Keranen, K., S. L. Klemperer, R. Gloaguen, and Eagle Working Group(2004). Three-dimensional seismic imaging of a protoridge axis in theMain Ethiopian rift, Geology 32, 949–952.



Kim, W-Y. (1998). The ML scale in eastern North America, Bull. Seism.Soc. Am. 88, 935–951.

Kim, S. K., and M. A. Park (2005). The local magnitude scale in the KoreanPeninsula, Pure Appl. Geophys. 162, 875–889.

Klein, F. W. (2002). User’s Guide to Hypoinverse-2000, a Fortran programto solve for earthquake locations and magnitudes, U.S. Geol. Surv.Open File Rept. 02-171, 1–123.

Langston, C. A., A. A. Nyblade, and T. J. Owens (2002). Regional wavepropogation in Tanzania, East Africa, J. Geophys. Res. 107, doi10.1029/2001JB000167.

Langston, C. A., R. Brazier, A. A. Nyblade, and T. J. Owens (1998). Localmagnitude scale and seismicity rate for Tanzania, East Africa, Bull.Seism. Soc. Am. 88, 712–721.

Maguire, P. K. H., G. R. Keller, S. L. Klemperer, G. D. Mackenzie, K.Keranen, S. Harder, B. O’Reilly, H. Thybo, L. Asfaw, M. A. Khan,and M. Amha (2006). Crustal structure of the northern main EthiopianRift from the EAGLE controlled source survey; a snapshot of incipientlithospheric break-up, in The Afar Volcanic Province within the EastAfrican Rift System, G. Yirgu, C. J. Ebinger, and P. K. H. Maguire(Editors), Geol. Soc. Spec. Publ. 259, 269–292.

Mammo, T. (2005). Site-specific ground motion simulation and seismicresponse analysis at the proposed bridge sites within the city of AddisAbaba, Ethiopia, Eng. Geol. 79, 127–150.

Murru, M., C. Montouri, M. Wyss, and E. Privitera (1999). The locationsof magma chambers at Mt. Etna, Italy, mapped by b-values, Geophys.Res. Lett. 26, 2553–2556.

Ortega, R., and L. Quintanar (2005). A study of local magnitude scale inthe basin of Mexico: mutually consistent estimates of log Ao andground motion scaling, Bull. Seism. Soc. Am. 95, 605–613.

Pujol, J. (2003). Determination of a local magnitude scale: a generalizedinverse solution, Bull. Seism. Soc. Am. 93, 2758–2761.

Richter, C. F. (1935). An instrumental earthquake magnitude scale, Bull.Seism. Soc. Am. 25, 1–31.

Richter, C. F. (1958). Elementary Seismology, W. H. Freeman and Co., SanFrancisco, 578 pp.

Rooney, T. O., T. Furman, G. Yirgu, and D. Ayalew (2005). Structure ofthe Ethiopian lithosphere: xenolith evidence in the Main Ethiopianrift, Geochem. Cosmochim. Acta 69, 3889–3910.

Savage, M. K., and J. G. Anderson (1995). A local-magnitude scale for thewestern Great Basin-Eastern Sierra Nevada from synthetic Wood-Anderson seismograms, Bull. Seism. Soc. Am. 85, 1236–1243.

Schlotterbeck, B. A., and G. A. Abers (2001). Three-dimensional attenua-tion variations in southern California, J. Geophys. Res. 106, 30,719–30,735.

Shi, Y., and B. A. Bolt (1982). The standard error of the magnitude-frequency b-value, Bull. Seism. Soc. Am. 72, 1677–1687.

Tongue, J. A., P. K. H. Maguire, and P. Burton (1992). Seismicity distri-bution from temporary earthquake recording networks in Kenya, Tec-tonophysics 204, 71–79.

Weeraratne, D. S., D. W. Forsyth, K. M. Fischer, and A. A. Nyblade (2003).Evidence for an upper mantle plume beneath the Tanzanian cratonfrom Rayleigh wave tomography, J. Geophys. Res. 108, no. B9, 2427,doi 10.1029/2002JB002273.

Wiemer, S., S. R. McNutt, and M. Wyss (1998). Temporal and three-dimensional spatial analyses of the frequency-magnitude distributionnear Long Valley Caldera, California, Geophys. J. Int. 134, 409–421.

Whaler, K. A., and S. Hautot (2006). The electrical resistivity structure ofthe crust beneath the northern Ethiopian rift, in The Afar VolcanicProvince within the East African Rift System, G. Yirgu, C. J. Ebinger,and P. K. H. Maguire (Editors), Geol. Soc. Spec. Publ. 259, 293–306.

Williams, F. M., M. A. J. Williams, and F. Aumento (2004). Tensionalfissures and crustal extension rates in the northern part of the MainEthiopian rift, J. Afr. Earth Sci. 38, 183–197.

Wright, T. J., C. Ebinger, J. Biggs, A. Ayele, G. Yirgu, D. Keir, and A.Stork (2006). Dyke-maintained rift segmentation at continental rup-ture—the 2005 Dabbahu (Afar) rifting episode, Nature (in press).

Yirgu, G., A. Ayalew, A. Asrat, A. Ayele, and A. Philpotts (2005).www.volcano.si.edu/reports/bulletin.

Department of GeologyRoyal Holloway University of LondonEgham, Surrey, TW20 0EX, United [email protected]

(D.K.)

School of Earth and EnvironmentUniversity of LeedsLeeds, LS2 9 JT, United [email protected]

(G.W.S., A.J.)

Geophysical ObservatoryAddis Ababa UniversityAddis Ababa, P.O.B. 1176., [email protected]

(A.A.)

Manuscript received 9 March 2006.

References

136

References

Abebe, B., M. Boccaletti, R. Mazzuoli, M. Bonini, L. Tortorici and T. Trua, Geological Map of the

Lake Ziway-Asella region (Main Ethiopian Rift), 1: 50 000 scale, Consiglio Nazionale

delle Ricerche , A.R.C.A. Firenze, 1998.

Abebe, T., Manetti, P., Bonini, M., Corti, G., Innocenti, F., Mazzarini, F., and Pecksay, Z,

Geological map (1:200000) of the northern Main Ethiopian rift and its implications for

the volcano-tectonic evolution of the rift. Geol. Soc. Amer. Chart and Map Series

MCH094, 20 pp, 2005.

Abebe, T., F. Mazzarini, F. Innocenti and P. Manetti, The Yerer-Tullu Wellel volcanotectonic

lineament: a transtensional structure in central Ethiopia and the associated magmatic

activity, J. Afr. Earth Sci., 26(1), 135-150, 1998.

Acocella, V. and T. Korme, Holocene extension direction along the Main Ethiopian Rift, East

Africa, Terra Nova, 14, 191-197, 2002.

Acocella, V., T. Korme and F. Salvini, Formation of normal faults along the axial zone of the

Ethiopian Rift, J. Struct. Geol., 25, 503-513, 2003.

Aki, K., Maximum likelihood estimate of b in the formula logN = a-bM and its confidence limits,

Bull. Earthq. Res. Inst., 43, 237-239, 1965.

Anderson, J. A. and H. O. Wood, Description and theory of the torsion seismometer, Bull.

Seism. Soc. Am., 15, 1-72, 1925.

Anderson, J. G., and Y. Lei, Nonparametric description of peak acceleration as a function of

magnitude, distance and site in Guerrero, Mexico, Bull. Seism. Soc. Am., 84, 1003-

1017, 1994.

Arnott, S. K. and G. R. Foulger, The Krafla spreading segment, Iceland, 1. Three-dimensional

crustal structure and the spatial and temporal distribution of local earthquakes, J.

Geophys. Res., 99(B12), 23801-23825, 1994.

Asfaw, L. M., Development of earthquake induced fissures, Nature, 286, 551-553, 1982.

Asfaw, L. M., Environmental hazard from fissures in the Main Ethiopian Rift, J. Afr. Earth Sci.,

27(3/4), 481-490, 1998.

Ayele, A., Earthquake catalogue of the Horn of Africa for the period 1960-1993, Seismological

Dept. , Uppsala Univ. , Report, 3-95, 1-9, 1995.

Ayele, A., Normal left-oblique fault mechanisms as an indication of sinistral deformation

between the Nubia and Somalia Plates in the Main Ethiopian Rift, J. Afr. Earth Sci., 31

(2), 2000.

References

137

Ayele, A. and R. Arvidsson, Fault mechanisms and tectonic implication of the 1985-1987

earthquake sequence in south western Ethiopia, J. Seismol., 1, 383-394, 1998.

Ayele, A. and O. Kulhánek, Spatial and temporal variation of the seismicity in the Horn of Africa

from 1960-1993, Geophys. J. Int., 130, 805-810, 1997.

Ayele, A., G. W. Stuart, I. D. Bastow and D. Keir, The August 2002 earthquake sequence in

north Afar: Insights into the neotectonics of the Danakil microplate, J. Afr. Earth Sci., in

press.

Ayele, A., A. A. Nyblade, C. A. Langston, M. Cara and J-J. Leveque, New evidence for Afro-

Arabian plate separation in southern Afar, in G. Yirgu, C. Ebinger and P. Maguire,

Structure and Evolution of the East African Rift in the Afar volcanic province, Geol. Soc.

Lond. Spec. Pub., 2006.

Bakun, W. H. and W. B. Joyner, The ML scale in central California, Bull. Seism. Soc. Am., 74,

1827-1843, 1984.

Barclay, A. H. and D. R. Toomey, Shear wave splitting and crustal anisotropy at the Mid-

Atlantic Ridge, 35oN, J. Geophys. Res., 108(B8), doi:10.1029/2001JB000918, 2003.

Bastow, I. D., G. W. Stuart, J-M. Kendall and C. J. Ebinger, Upper mantle seismic structure in a

region of incipient continental breakup: northern Ethiopian rift, Geophys. J. Int. 162,

479-493, 2005.

Baumbach, M., D. Bindi, H. Grosser, C. Milkereit, S. Parolai, R. Wang, S. Karakisa, S. Zünbül,

and J. Zschau, Calibration of an ML scale in northwestern Turkey from 1999 Izmit

aftershocks, Bull. Seism. Soc. Am., 93, 2289-2295, 2003

Bendick, B., S. McClusky, R. Bilham, L. Asfaw, S. Klemperer, Distributed Nubia-Somalia relative

motion and dike intrusion in the Main Ethiopian rift, Geophys. J. Int., in press (2006).

Benoit, M., J. Nyblade, J. Vandecar and H. Gurrola, Upper Mantle P wave velocity structure

and transition zone thickness beneath the Arabian shield, Geophys. Res. Lett., 30(10),

2003.

Berhe, S. M., Ophiolites in Northeast and East Africa: implications for Proterozoic crustal

growth, J. Geol. Soc., 147, 41-57, 1990.

Bilham, R., R. Bendick, K. Larson, P. Mohr, J. Braun, S. Tesfaye and L. Asfaw, Secular and

tidal strain across the Main Ethiopian Rift, Geophys. Res. Lett., 26, 2789-2792, 1999.

Boccaletti, M., M. Bonini, R. Mazzuoli, B. Abebe, L. Piccardi and L. Tortorici, Quaternary

oblique extensional tectonics in the Ethiopian Rift (Horn of Africa), Tectonophysics, 287,

97-116, 1998.

References

138

Booth, D. C. and S. Crampin, Shear-wave polarizations on a curved wavefront at an isotropic

free-surface, Geophys. J. Royal Ast. Soc., 83, 31-45, 1985.

Bormann P., Magnitude of seismic events, in IASPEI New Manual of Seismological Observatory

Practice, P. Bormann (Editor), V 1, GeoForschungsZentrum, Potsdam, 16-50, 2002.

Bragato, P. L. and A. Tento, Local magnitude in northeastern Italy, Bull. Seism. Soc. Am., 95,

579-591, 2005.

Buck, W. R., Modes of continental lithospheric extension: Pure and simple, J. Geophys. Res.,

96(12), 20161-20178, 1991

Buck, W. R., Consequences of asthenospheric variability on continental rifting, Rheology and

Deformation of the Lithosphere at Continental Margins, in G. D. Karner, B. Taylor, N. W.

Driscoll and D. L. Kohlstedt, pp. 1-30, Columbia Univ. Press, New York, 2004.

Calais, E., C. DeMets and J.-M. Nocquet, Evidence for a post-3.16-Ma change in Nubia-

Eurasia-North America plate motions, Earth Planet. Sci. Lett., 216, 81-92, 2003.

Calais, E., C. Ebinger, C. Hartnady and J. M. Nocquet, Kinematics of the East African rift from

GPS and earthquake slip vector data, in G. Yirgu, C. Ebinger and P. Maguire, Structure

and Evolution of the East African Rift in the Afar volcanic province, Geol. Soc. Lond.

Spec. Pub., 2006.

Casey, M., C. Ebinger, D. Keir, R. Gloaguen, F. Mohamed, Strain accommodation in

transitional rifts: Extension by magma intrusion and faulting in Ethiopian rift magmatic

segments, in G. Yirgu, C. Ebinger and P. Maguire, Structure and Evolution of the East

African Rift in the Afar volcanic province, Geol. Soc. Lond. Spec. Pub., 2006.

Cattin, R., C. Doubre, J-B. de Chabalier, G. King, C. Vigny, J-P. Avouac and J-C. Ruegg,

Numerical modelling of Quaternary deformation and post-rifting displacement in the

Asal-Ghoubbet rift (Djibouti, Africa), Earth Planet. Sci. Lett., 239, 352-367, 2005.

Chu, D. and R. G. Gordon, Evidence for motion between Nubia and Somalia along the

Southwest Indian ridge, Nature, 398, 64-67, 1999.

Church, W. R., S. M. Berhe, M. G. Abdelsalam and R. J. Stern, Discussion of ophiolites in

Northeast and East Africa: Implications for Proterozoic crustal growth, J. Geol. Soc.

Lon., 148, 600-606, 1991.

Crampin, S., The fracture criticality of crustal rocks, Geophys. J. Int., 118, 428-438, 1994.

Daly, E., D. Keir, C. J. Ebinger, G. W. Stuart, A. Ayele, and D. Waltham, Along-axis

segmentation of the Main Ethiopian rift from tomographic inversion of local earthquakes,

submitted to J. Geophys. Res., 2006.

References

139

Debayle, E., J-J. Lévêque and M. Cara, Seismic evidence for a deeply rooted low velocity

anomaly in the upper mantle beneath the northeastern Afro / Arabian continent, Earth

Planet. Sci. Lett., 193, 423-436, 2001.

De Chabalier, J. B. and J. P. Avouac, Kinematics of the Asal rift Djibouti determined from the

deformation of the Fieale volcano, Science, 265, 1677-1681, 1994

do Nascimento, A. F., F. H. R. Bezerra and M. K. Takeya, Ductile Precambrian fabric control of

seismic anisotropy in the Açu dam area, northeast Brazil, J. Geophys. Res.,

109(B10311), doi:10.1029/2004JB003120, 2004.

Dugda, M., A. Nyblade, J. Julia, C. Langston, C.A. Ammon and S. Simiyu, Crustal structure in

Ethiopia and Kenya from receiver function analysis: Implication for rift development in

eastern Africa, J. Geophys. Res., 110, B01303, doi:10.1029/2004JB003065, 2005.

Dunbar, J. and D. Sawyer, How pre-existing weaknesses control the style of continental

breakup, J. Geophys. Res., 94, 7278-7292, 1989.

Eberhart-Philips, D. and A. J. Michael, Seismotectonics of the Loma Prieta, California, region

determined from three-dimensional Vp, Vp/Vs, and seismicity, J. Geophys. Res., 103 ,

21099-21120, 1998.

Ebinger, C., Continental break-up: the East African perpective. Astronomy and Geophysics, 46,

16-21, 2005.

Ebinger, C. and M. Casey, Continental brekaup in magmatic provinces: an Ethiopian example,

Geology, 29, 527-530, 2001.

Ebinger, C. J. and N. Hayward, Soft plates and hot spots: Views from Afar, J. Geophys. Res.,

101, 21859-21876, 1996.

Ebinger, C. and N. Sleep, Cenozoic magmatism throughout East Africa resulting from impact of

a single plume, Nature, 395, 788-791, 1998.

Ebinger, C. J., J. A. Jackson, A. N. Foster, and N. J. Hayward, Extensional basin geometry and

the elastic lithosphere. Phil. Trans. R. Soc. Lond. Series A - Mathemetical, Physical and

Engineering Sciences, 150, 99-108, 1999.

Evans, J. R., G. R. Foulger, B. R. Julian and A. D. Miller, Crustal shear-wave splitting from local

earthquakes in the Hengill triple junction, southwest Iceland, Geophys. Res. Lett., 23(5),

455-458, 1996.

Feigl, K. L., J. Gasperi, F. Sigmundsson and A. Rigo, Crustal deformation near Hengill volcano,

Iceland 1993-1998: Coupling between magmatic activity and faulting inferred from

elastic modelling of satellite radar interferograms, J. Geophys. Res., 105, 25655-25670,

2000.

References

140

Fernandes, R., B. Ambrosius, R. Noomen, L. Bastos, L. Combrinck, J. Miranda and W.

Spakman, Angular velocities of Nubia and Somalia from continuous GPS data:

implications on present-day relative kinematics, Earth Planet. Sci. Lett., 222, 197-208,

2004.

Foster, A. N. and J. A. Jackson, Source parameters of large African earthquakes: implications

for crustal rheology and regional kinematics, Geophys. J. Int., 134, 422-448, 1998.

Gashawbeza, E. M., Klemperer, S. L., Nyblade, A. A., Walker, K. T., and K. M. Keranen, Shear-

wave splitting in Ethiopia: Precambrian mantle anisotropy locally modified by Neogene

rifting, Geophys. Res. Lett., 31, doi:10.1029/2004GL020471, 2004.

George, R., N. Roger and S. Kelley, Earliest magmatism in Ethiopia: Evidence for two mantle

plumes in one flood basalt province, Geology, 26, 923-926, 1998.

Gouin, P., Earthquake History of Ethiopia and the Horn of Africa, International Developement

Research Centre (IDRC), Ottawa, Ont., pp. 258, 1979.

Gutenberg, B. and C. F. Richter, Earthquake magnitude, intensity, energy, and acceleration,

Bull. Seism. Soc. Am., 46, 105-145, 1956.

Hardebeck, J. L. and E. Hauksson, Stress orientations obtained from earthquake focal

mechanisms: What are appropriate uncertainty estimates, Bull. Seism. Soc. Am., 91(2),

250-262, 2001.

Harris, W. C., The Highlands of Ethiopia, vol. 3, Longman, Brown and Longman, London, 1844.

Haslinger, F., E. Kissling, J. Ansorge, D. Hatzfeld, E. Papadimitriou, V. Karakostas, K.

Makropoulos, H.-G. Kahle and Y. Peter, 3D crustal structure from local earthquake

tomography around the Gulf of Arta (Ionian region, NW Greece), Tectonophysics, 304,

201-218, 1999.

Hayward, N. and C. Ebinger, Variations in along-axis segmentation of the Afar rift system,

Tectonics, 15, 244-257, 1996.

Hofmann, C., V. Courtillot, F. Feraud and P. Rochette, Timing of the Ethiopian flood event:

Implications for plume birth and global change, Nature, 389, 838-841, 1997.

Hofstetter, R. and M. Beyth, The Afar Depression: interpretation of the 1960-2000 earthquakes,

Geophys. J. Int., 155, 715-732, 2003.

Holbrook, W., and P. Kelemen, Large igneous province on the US Atlantic margin and

implications for magmatism during breakup, Nature, 364, 433-436, 1993.

Hutton, L. K. and D. M. Boore, The ML scale in Southern California, Bull. Seism. Soc. Am., 77,

2074-2094, 1987.

References

141

Ibs-von Seht, M., S. Blumenstein, R. Wagner, D. Hollnack and J. Wohlenberg, Seismicity,

seismotectonics and crustal structure of the southern Kenya Rift - new data from the

Lake Magadi area, Geophys. J. Int., 146, 439-453, 2001.

Jestin, F., P. Huchon and J. M. Gaulier, The Somalia plate and the East African Rift System:

present-day kinematics, Geophys. J. Int., 116, 637-654, 1994.

Jones, L. M., Focal mechanisms and the state of stress on the San Andreas fault in Southern

California, J. Geophys. Res., 93, 8869-8891, 1988.

Kanamori, H. and P. C. Jennings, Determination of local magnitude, ML, from strong motion

accelerograms, Bull. Seism. Soc. Am., 68, 471-485, 1978.

Kastrup, U., M. L. Zoback, N. Deichmann, K. F. Evans, D. Giardini and A. J. Michael, Stress

field variation in the Swiss Alps and the northern Alpine foreland derived from inversion

of fault plane solutions, J. Geophys. Res., 109, B01402, doi:10.1029/2003JB002550,

2004.

Kazmin, V., A. Shifferaw and T. Balcha, The Ethiopian basement: stratigraphy and possible

manner of evolution, Int. J. Earth Sci. (Geologische Rundschau), 67, 531-546, 1978.

Kebede, F. and O. Kulhánek, Recent seismicity of the East African rift system and its

implications, Phys. Earth Planet. Inter., 68, 259-273, 1991.

Kebede, F. and O. Kulhánek, Spatial and temporal variations in b-values along the East African

rift system and the southern Red Sea, Phys. Earth Planet. Inter., 83, 249-264, 1994.

Kebede, F. and T. van Eck, Probabilistic seismic hazard assessment for the Horn of Africa

based on seismotectonic regionalization, Tectonophysics, 270, 221-237, 1997.

Kebede, F., W.-Y. Kim and O. Kulhánek, Dynamic source parameters of the March-May 1969

Serdo earthquakes in Central Afar, Ethiopia, deduced from teleseismic body waves, J.

Geophys. Res., 94, 5603-5614, 1989.

Keir, D., C. Ebinger, G. Stuart, E. Daly and A. Ayele, Strain accomodation by magmatism and

faulting as rifting proceeds to breakup: Seismicity of the northern Ethiopian rift, J.

Geophys. Res., in press (2006).

Keir, D., J-M. Kendall, C. J. Ebinger and G. W. Stuart, Variations in late syn-rift melt alignment

inferred from shear-wave splitting in crustal earthquakes beneath the Ethiopian rift,

Geophys. Res. Lett., 32, L23308, doi 10.1029/2005GL024150, 2005.

Kendall, J-M., G. W. Stuart, C. J. Ebinger, I. D. Bastow and D. Keir, Magma assisted rifting in

Ethiopia, Nature, 433, 146-148, 2005.

Kendall, J-M., S. Pilidou, D. Keir, I .D. Bastow, A. Ayele, G. W. Stuart, Mantle upwellings, melt

migration and the rifting of Africa: Insights from seismic anisotropy, in G. Yirgu, C.

References

142

Ebinger, and P. Maguire, Structure and Evolution of the East African Rift in the Afar

volcanic province, Geol. Soc. Lond. Spec. Pub., in press.

Keranen, K., S. L. Klemperer, R. Gloaguen and Eagle working group, Three-dimensional

seismic imaging of a protoridge axis in the Main Ethiopian rift, Geology, 32, 949-952,

2004.

Kidane, T., E. Platzman, C. Ebinger, B. Abebe, and P. Rochette, Paleomagnetic constraints on

Continental Breakup: Observations from the Main Ethiopian rift, in G. Yirgu, C. Ebinger

and P. Maguire, Structure and Evolution of the East African Rift in the Afar volcanic

province, Geol. Soc. Lond. Spec. Pub., 2006.

Kieffer, B., N. Arndt, H. Lapierre, F. Bastien, D. Bosch, A. Pecher, G. Yirgu, D. Ayalew, D. Weis,

D. A. Jerram, F. Keller and C. Meugniot, Flood and shield basalts from Ethiopia:

Magmas from the African superswell, J. Petrology, 45(4), 793-834, 2004.

Kim, W-Y., The ML scale in eastern North America, Bull. Seism. Soc. Am., 88, 935-951, 1998.

Kim, S. K. and M. A. Park, The local magnitude scale in the Korean Peninsula, Pure appl.

geophys., 162, 875-889, 2005.

Kissling, E., U. Kradolfer and H. Maurer, VELEST user's guide - Short introduction, Institute of

Geophysics and Swiss Seismological Service, ETH, Zurich, 1995.

Klein, F. W., User's Guide to Hypoinverse-2000, a Fortran program to solve for earthquake

locations and magnitudes, USGS Open File Report, 02-171, 1-123, 2002.

Korme, T., J. Chorowicz, B. Collet and F. F. Bonavia, Volcanic vents rooted on extension

fractures and their geodynamic implications in the Ethiopian Rift, J. Volcanol. Geotherm.

Res., 79, 205-222, 1997.

Langston, C. A., A. A. Nyblade and T. J. Owens, Regional wave propogation in Tanzania, East

Africa, J. Geophys. Res., 107, DOI 10.1029/2001JB000167, 2002.

Langston, C. A., R. Brazier, A. A. Nyblade and T. J. Owens, Local magnitude scale and

seismicity rate for Tanzania, East Africa, Bull. Seism. Soc. Am., 88, 712-721, 1998.

Lister, G., M. Etheridge and P. Symonds, Detachment faulting and the evolution of passive

continental margins, Geology, 14, 246-250, 1986.

Mackenzie, G. H., G. H. Thybo and P. Maguire, Crustal velocity structure across the Main

Ethiopian Rift: results from 2-dimenional wide-angle seismic modelling, Geophys. J. Int.,

162, 996-1006, 2005.

Maguire, P. K. H., C. J. Ebinger, G. W. Stuart, G. D. Mackenzie, K. A. Whaler, J.-M. Kendall, M.

A. Khan, C. M. R. Fowler, S. L. Klemperer, G. R. Keller, S. Harder, T. Furman, K.

References

143

Mickus, L. Asfaw, A. Ayele and B. Abebe, Geophysical project in Ethiopia studies

continental breakup, Eos (Trans. , Am. Geophys. Union), 84, 337-343, 2003.

Maguire, P. K. H., G. R. Keller, S. L. Klemperer, G. D. Mackenzie, K. Keranen, S. Harder, B.

O’Reilly, H. Thybo, L. Asfaw, M. A. Khan and M. Amha, Crustal Structure of the

northern Main Ethiopian Rift from the EAGLE controlled source survey; a snapshot of

incipient lithospheric break-up, in G. Yirgu, C. Ebinger and P. Maguire, Structure and

Evolution of the East African Rift in the Afar volcanic province, Geol. Soc. Lond. Spec.

Pub., 2006.

Mahatsente, R., G. Jenztsch and T. Jahr, Crustal structure of the Main Ethiopian Rift from

gravity data: 3-dimensional modelling, Tectonophysics, 313, 363-382, 1999.

Mammo, T., Site-specific ground motion simulation and seismic response analysis at the

proposed bridge sites within the city of Addis Ababa, Ethiopia, Engineering Geology, 79,

127-150, 2005.

McKenzie, D., Some remarks on the development of sedimentary basins, Earth Plan. Sci. Lett.,

40, 25-32, 1978.

Menke, W., B. Brandsdottir, S. Jakobsdottir and R. Stefansson, Seismic anisotropy in the crust

at the mid-Atlantic plate boundary in south-west Iceland, Geophys. J. Int., 119, 783-790,

1994.

Michael, A. J., Determination of stress from slip data: Faults and folds, J. Geophys. Res., 89,

11517-11526, 1984.

Michael, A. J., The use of focal mechanisms to determine stress: A control study, J. Geophys.

Res., 92, 357-368, 1987.

Michael, A. J., Stress rotation during the Coalinga aftershock sequence, J. Geophys. Res., 92,

7963-7979, 1987.

Miller, V. and M. Savage, Changes in seismic anisotropy after volcanic eruptions: evidence from

Mount Ruapehu, Science, 293, 2231-2233, 2001.

Montelli, R., G. Nolet, F. A. Dahlen, G. Masters, E. R. Engdahl and S-H. Hung, Finite-frequency

tomography reveals a variety of plumes in the mantle, Science, 303, 338-343, 2004.

Musumeci, C., O. Cocina, P. De Gori and D. Patanè, Seismological evidence of stress induced

by dike injection during the 2001 Mt. Etna eruption, Geophys. Res. Lett., 31, 2004.

Nyblade, A. A., C. Birt, C. Langston, T. Owens, and R. Last, Seismic experiment reveals rifting

of craton in Tanzania, Eos, 77, 517-521, 1996.

References

144

Ortega, R., and L. Quintanar, A study of local magnitude scale in the basin of Mexico: Mutually

consistent estimates of log Ao and ground motion scaling, Bull. Seism. Soc. Am., 95,

605-613, 2005.

Phipps-Morgan, J. and J. Chen, Dependence of ridge-axis morphology on magma supply and

spreading rate, Nature, 364, 706-708, 1993.

Pizzi, A., M. Coltorti, B. Abebe, L. Disperati, G. Sacchi, R. Salvini, The Wonji Fault Belt (Main

Ethiopian Rift, Ethiopia): Structural and geomorphological constraints and GPS

monitoring, in G. Yirgu, C. Ebinger and P. Maguire, Structure and Evolution of the East

African Rift in the Afar volcanic province, Geol. Soc. Lond. Spec. Pub., 2006.

Pujol, J., Determination of a local magnitude scale: a generalized inverse solution, Bull. Seism.

Soc. Am., 93, 2758-2761, 2003.

Richter, C. F., An instrumental earthquake magnitude scale, Bull. Seism. Soc. Am., 25, 1-31,

1935.

Richter, C. F., Elementary Seismology, W. H. Freeman and Co. , San Francisco, 578 pp., 1958.

Rooney, T. O., T. Furman, Yirgu.G. and D. Ayalew, Structure of the Ethiopian lithosphere:

Xenolith evidence in the Main Ethiopian rift, Geochem. Cosmochim. Acta, 69, 3889-

3910, 2005.

Rubin, A. M., Dike-induced faulting and graben subsidence in volcanic rift zones, J. Geophys.

Res., 97, 1839-1858, 1992.

Rubin, A. M. and D. Gillard, Dike-induced earthquakes: Theoretical considerations, J. Geophys.

Res., 103, 10017-10030, 1998.

Ruegg, J. C., and M. Kasser, Deformation across the Asal-Ghoubbet rift, Djibouti, uplift and

crustal extension 1979-1986, Geophys. Res. Lett., 14, 745-748, 1987.

Savage, M. K. and J. G. Anderson, A local-magnitude scale for the western Great Basin-

Eastern Sierra Nevada from synthetic Wood-Anderson seismograms, Bull. Seism. Soc.

Am., 85, 1236-1243, 1995.

Schilling, J-G., Afar mantle plume: Rare earth evidence, Nature, 242, 2-5, 1973.

Schlotterbeck, B. A. and G. A. Abers, Three-dimensional attenuation variations in southern

California, J. Geophys. Res., 106, 30719-30735, 2001.

Shi, Y. and B. A. Bolt, The standard error of the magnitude-frequency b-value, Bull. Seism.

Soc. Am., 72, 1677-1687, 1982.

Silver, P. G., Seismic anisotropy beneath the continents, Annu. Rev. Earth Planet. Sci.,

24(385), 432, 1996.

References

145

Silver, P. G. and W. W. Chan, Shear-wave splitting and subcontinental mantle deformation, J.

Geophys. Res., 96, 16429-16454, 1991.

Snoke, J. A., A. C. Munsey, A. C. Teague and G. A. Bollinger, A program for focal mechanism

determination by combined use of polarity and SV-P amplitude ratio data, Earthquake

Notes, 15(3), 1984.

Steckler, M.S., and A. B. Watts, Subsidence history and tectonic evolution of Atlantic-type

continental margins, in R. A. Scrutton, ed. Dynamics of Passive Margins, AGU

Geodynamics Series, Washington DC: American Geophysical Union, 6, 184-196, 1981.

Stein, R. S., P. Briole, J. C. Ruegg, P. Tapponier, and F. Gasse, Contemporary Holocene, and

Quaternary deformation of the Asal rift, Djibouti – implications for the mechanisms of

slow-spreading ridges, J. Geophys. Res., 96, 21789-21806, 1991.

Stuart, G. W., I. D. Bastow and C. J. Ebinger, Crustal structure of the northern Main Ethiopian

rift from receiver function studies, in G. Yirgu, C. Ebinger and P. Maguire, Structure and

Evolution of the East African Rift in the Afar volcanic province, Geol. Soc. Lond. Spec.

Pub., 2006.

Tiberi, C., C. Ebinger, V. Ballu, G. Stuart and B. Oluma, Inverse models of gravity data from the

Red Sea - Aden - East African rifts triple junction zone, Geophys. J. Int.,

doi:10.1111/j.1365-246X.2005.02736.x, 2005.

Tongue, J. A., Maguire P.K.H. and P. Burton, Seismicity distribution from temporary earthquake

recording networks in Kenya, Tectonophysics, 204, 71-79, 1992.

Ukstins, I., P. Renne, E. Wolfenden, J. Baker and M. Menzies, Matching conjugate volcanic

rifted margins: 40Ar/39Ar chrono-stratigraphy of pre- and syn-rift bimodal flood volcanism

in Ethiopia and Yemen, Earth Planet. Sci. Lett., 198, 289-306, 2002.

Volti, T. and S. Crampin, A four-year study of shear-wave splitting in Iceland: 2. Temporal

changes before earthquakes and volcanic eruptions, in New insights into structural

interpretation and modeling, ed. Nieuwland, D. A., Geol. Soc. Lond., Spec. Publ., 212,

135-149, 2003b.

Weeraratne, D. S., D. W. Forsyth, K. M. Fischer, and A. A. Nyblade, Evidence for an upper

mantle plume beneath the Tanzanian craton from Rayleigh wave tomography, J.

Geophys. Res., 108 (B9), 2427, doi:10.1029/2002JB002273, 2003.

Whaler, K. A. and S. Hautot, The electrical resistivity structure of the crust beneath the northern

Ethiopian rift, in G. Yirgu, C. Ebinger, and P. Maguire, Structure and Evolution of the

East African Rift in Afar volcanic province, Geol. Soc. Lond. Spec. Pub., in press.

References

146

Williams, F. M., M. A. J. Williams and F. Aumento, Tensional fissures and crustal extension

rates in the northern part of the Main Ethiopian Rift, J. Afr. Earth Sci., 38, 183-197,

2004.

WoldeGabriel, G., J. Aronson, R. Walter, Geology, geochronology, and rift basin development in

the central sector of the Main Ethiopian rift, Geol. Soc. Am. Bull., 102, 439-458, 1990.

Wolfenden, E., C. Ebinger, G. Yirgu, A. Deino and D. Ayalew, Evolution of the northern Main

Ethiopian rift: birth of a triple junction, Earth Planet. Sci. Lett., 224, 213-228, 2004.

Wolfenden, E., C. Ebinger, G. Yirgu, P. Renne and S. P. Kelley, Evolution of the southern Red

Sea rift: Birth of a magmatic margin, Bull. Geol. Soc. Am., 117, 846-864, 2005.

Wright, T., C. Ebinger, J. Biggs, A. Ayele, G. Yirgu, A. Stork and D. Keir, Magma-fed

segmentation of an incipient oceanic rift revealsed by InSar in Afar, submitted to Nature

December 2005.

Yirgu, G., A. Ayalew, A. Asrat, A. Ayele and A. Philpotts,

http://www.volcano.si.edu/reports/bulletin (2005).

Documents

Keir_thesis Seismicity of the Northern Ethiopian Rift