Seismotectonics of the Hindukush and Baluchistan arc

Tec~ono~hysies, 66 (1980) 301-322 @ Elsevier Scientific ~blishing Company, Amsterdam - Printed in The Netherl~ds

301

SEISMOTECTONICSOF THE HINDUKUSHAND BALUCHISTANARC R.K. VERMA, M. MUKHOPADHYAY

and A.K. BHANJA

Indian School of Mines, Dhanbad 626004 (Received November l&1978;

(Zndia)

revised version accepted June 7,1979)

ABSTRACT Verma, R.K., Mukhopadhyay, M. and Bhanja, A.K., 1980. Seismotectonics kush and Baluchistan arc. Tectonophysics, 66: 301-322.

of the Hindu-

A seismicity map of that part of the Pakistan-Afghanistan region lying between the latitudes 28’ to 38’N and longitudes 66’ to 75OE is given using all available data for the period 1890-1970. The earthquakes of magnitude 4.5 and above were considered in the preparation of this map. On the basis of this map, it is observed that the seismicity pattern over the well-known Hindukush region is quite complex. Two prominent, mutually orthogonal, seismicity lineaments, *namely the northwestern and the northeastern trends, characterize the Hindukush area. The northwestern trend appears to extend from the Main Boundary Fault of the Kashmir Himalaya on the southeast to the plains of the Amu Darya in Uzbekistan on the northwest beyond the ~mduk~h. The Sulaiman and Kirthar ranges of Pakistan are web-defined zones of intermontane seismicity exhibiting north-south alignment. Thirty-two new focal-mechanism solutions for the above-mentioned region have been determined. These, together with the results obtained by earlier workers, suggest the predominance of strike-slip faulting in the area. The Hazara Mountains, the Sulaiman wrench zone and the Kirthar wrench zone, as well as the supposed extension of the Murray ridge up to the Karachi coast, appear to be mostly undergoing strike-slip movements. In the Hindukush region, thrust and strike-slip faulting are found to be equally prevalent. Almost all the thrust-type mechanisms belonging to the Hindubush area have both the nodal planes in the NW-SE direction for shallow as weil as intermediate depth earthquakes. The dip of P-axes for the events indicating thrust type mechanisms rarely exceeds 35’. The direction of the seismic slip vector obtained through thrust type solutions is always directed towards the northeast. The epicentral pattern together with these results suggest a deep-seated fault zone parallelmg the no~hw~terly seismic zone underneath the Hindukush. This NW-lineament has a preference for thrust faulting, and it appears to extend from the vicinity of the Main Boundary Fault of the Kashmir Himalaya on the southeast of Uzbekistan on the northwest through Hindukush. Almost orthogonal to this NW-seismic zone. there is a NE-seismic lineament in which there is a preference for strikeslip faulting. The above results are discussed from the point of view of convergence of the Indian and Eurasian plates in the light of plate tectonics theory. INTRODUCTION

Tbe Indian subcontinent has been coltiaing with the Eutxdansubcontinent during the past 40 m.y. (McE~~y, 1968; Molar and Tapponnier,

302

1975). As a result of this collision, the highest mountains in the world, the Himalaya, have been formed in the central part, the Arakan-Yoma mountains of Burma and the Naga hills of Assam towards the east, and the Baluchistan arc manifested by Kirthar and Sulaiman ranges towards the west. The consequences of this collision are also visible towards the Tien-Shan mountains in the U.S.S.R. and the Karakorum mountains in Pakistan. The Hindukush mountains have been formed at the junction of the Baluchistan arc, the Kashmir Himalaya, the Karakorum mountains and the Pamirs (Desio, 1965). These mountains are characterized by deep and concentrated seismicity through which an enormous amount of seismic energy is released every year. Seismically the area may be said to be one of the most active in the world. Some of the important problems which need to be investigated in this area are the following: (a) the causes of high seismic activity in a localized region at a depth of about 220 km under the Hindukush; (b) the nature of stress system operative in the crustal and subcrustal levels in the area, and (c) the nature of fault motions prevailing at depth. Recently, Gupta (1976) has pointed out that in spite of the recent increase in the number of seismological stations equipped with sensitive seismographs, the percentage level of detectability of earthquakes with mb = 4.5 in the Hindukush-Pamir region is extremely poor and only about 20% of earthquakes of body-wave magnitude 4.5 have been located. If this statistical measure of detectability is accepted, one might conclude that until a greater number of seismograph stations are installed, the best way to study the seismo-tectonics in the Hindukush region would be to consider all available earthquake data over a long period of time. Though the accuracy of earthquake data for the earlier years is definitely less, consideration of the data over a sufficiently long period is extremely necessary for a better representation of the true seismic characteristics of the region. Mainly from this viewpoint, an epicentral map of the Hindukush and the surrouding regions has been prepared in the present work, on the basis of all available data for the period 1890-1970. Besides the Hindukush region, the Sulaiman-Kirthar ranges of PakistanAfghanistan and adjacent regions have also been incorporated in the present study for a more complete understanding of seismo-tectonics of the entire region. For an understanding of fault motions associated with the earthquakes, and stress-system operative in the area, a number of focal-mechanism solutions for the whole region has been determined. These results, together with the results obtained by earlier workers, have been analyzed to interpret the seismo-tectonic characteristics and the nature of plate motions in the area. REGIONAL GEOTECTONIC SETTING OF THE AREA The area under study lies to the east of Daaht-E-Lut in Iran, and is restricted by the Indus plains and western section of the Kashmir Himalaya

on the east. Northward, the Pamirknot and the Rakaposbi rangedelimit the area, while to the south, the Makrancoastal rangesdefine the southernlimit. The area lies appro~ma~ly between 27”30’ to 38”N and 66’ to 75”E. The northern half of the area is occupied by the Hindukushand associatedmuuntains having an averageheight of about 3 km. A tectonic map of the area is shown in Fig. 1. The Sulaimanand the Kirtharranges of Pakistanare aligned in a northsouth direction forming the B~uch~ arc. The NW-SE trending mountains of Kashmir,which form the westernpart of the Himalayanarc, “bend” sharply to the south near “Nanga Parbat” forming the WesternHimalayan syntaxis. From there, the north-south trend of the Baluchistanarc is generally maintained along the S~~~“~h~ ranges for about 1000 km, before taking another sharp bend towards the west, after which a general east-west trend is rn~~~ along the Makranrangesand the mountainsof Southern Iran. The Makranranges of Iran and Pakistanhave been described as an active arc system by Farhoudi and Karig (1977). The structureof the Baluchistanarc, like the Himalayanarc, is dominated by tight folds and overthrusts, mainly from the north-west and west toward the Indian peninsula with axes directed essentially in a north-south direction, parallel to the generaltrend of the are (fcrishnIur, 1953). At the soFthem end of the Sulaimanranges,the mo~t~ns swingsharply towards the west, maintainan east-west trend foi nearly 300 km, and take a second sharp ‘“bend” to the south near the town of Quetta (see Fig. 1). This north-south trend continues along the Kixtharmountains(West, 1934). An almost similarsharpbend is observed in the region of Peshawarwhere the Sulaimanrangesjoin the Hazaramountains. The Salt Ranges,located to the south of the Hazaramts., also reflect,a similartrend. SEISMICITY MAP

The seismicity map of the area was prepaxedfor the period 189~1970 consideringall the earthquakesof magnitude4.5 and above. The earthquakes for which magnitudeswere not availablewere assumedto be of magnitude 5.0. The data for the map were obtained from the Catalogueof Earthquakes maintainedby the India MeteorologicalDepartment,Intemation~ Seismological Summaries, Bulletins of the International Seismological Centre, and U.S.C.G.S. reports. The epicentreswith their correspondingmagnitides were plotted on the tectonic map of the area (mostly after the Geological map of Pakistan,1: 2,000,000, Geological Surveyof Pakistan,Schreiberet al., 1972, and the UNESCO Map, 1971), and ‘are shown in Fig. 1. The important features of the map are discussedbelow. (1) The well-known H~duku~ area shows most intense seismic activity. Barring8ome scattering,two quite distinct seismic trends can be noticed in the Hind&u& area, i.e. the northwestern and the northeastern trends juxtaposing each other near 36.5”N and 70.5”E. The northeasterntrend

304

5pL +

0

T

9-t”

+

+

UlTAN +

+

+

3(

305

closely follows the syntaxial bend in the Pamir-Hindukusb region and continues southwestward. Many epicentres are located on or neeg: the surface faults which have a general northeasterly alignment. For example the northeasterly fault passing to the east of Kabul provides a clear indication of a fa~t-~~c~t~ se~icity. The epicentral ~ment associated with the fault passing to the northwest of Peshawar continues towards the Rakaposhi range on the northeast following the local arcuate trend. The epicentres having similar alignment continues to the southwest where, however, no surface faults are mapped, Almost orthogonal to this northeasterly trend, the northwesterly epicentral alignment continues from 73”E near the Kashmir Himalaya, across the main Hindukush seismic zone (located near 36.5”N, 70.5OE) towards the northwest. This no~hwes~rly seismic trend continues as far as 38”N, 66”E on the extreme northwest. The continuation of this seismic trend from near the Main Boundary Fault of the Himalaya towards the Hindukush is also corroborated by data from about 1800 micro-earthquakes reported by Armbruster et al. (1977) and Menke and Jacob (1976). As far as the present data can suggest, the width of the northwesterly seismic zone appears to be wider than the northeasterly seismic zone though the latter is better delineated on the basis of teleseismic events. It may be recalled here that Nowroozi (1971, 1972) had observed that the width of the no~h~terly seismic zone is about 100 km, having a length of over 200 km. The present data, however suggest that the length of the northeasterly zone is much more than this and may be comparable to that of the northwesterly zone. It has been mentioned earlier that Santo (1969) had noted the existence of a V-shaped seismic zone under the Hindukush, centered around 36.3”N and 70.6”E. The foci distribution suggests that the limbs of the seismic zone have slopes of about 55”, containing a seismic nest at about 225 km below the surface. It is noteworthy that the two seismic trends (i.e. the northeastern and northwestern) discussed above juxtapose each other near this location. (2) The Herat fault in Afghanistan is seismically inactive, save in its eastern part where it has a northeasterly trend while approaching the Hindukush. In comparison, the Chaman fault shows an appreciable activity over its whole length (see Fig. 1). (3) Several faults belonging to the north--south oriented S~~-K~h~ ranges are seismically quite active. The epicentral distribution within these ranges provides a clear example of intermontane seismicity. The eastern fault bordering the Sulaiman ranges from the adjoining Indus plains shows much seismic activity within the considered time period. In comparison, the western Part of the Sulaiman ranges is not so active. Only the northern tip of a fault in the western part of the Sulaiman ranges shows seismic activity where

ophiolite suites are located. (4) A very intense seismic zone, extending over a length of 450 km and having an average width of about 100 km, is noticed over the area located

306

between the SuIaiman and Kirthar ranges around the 30”N parallel which is known as the Sulaiman wrench zone. This has been described as a major zone of left-lateral shear by Abdel-Gawad (1971). The epicentral distribution suggests that the entire wrench zone is seismically very active. The east-west epicentral pattern associated with the wrench zone also extends beyond the areas of folding, and encroaches upon the Indus plains in Multan Province on the east. Towards the south many epicentres are spread over the plains. FOCAL

MECHANISM

SOLUTIONS

It has been suggested by several workers that the intense seismic activity under the Hindukush is a result of lithospheric subduction due to the convergence of the Indian landmass with that of Eurasia (Santo, 1969; Nowroozi, 1971; Billington et al., 1977). On similar grounds the seismicity of the Baluchistan arc is believed to have resulted from the northward convergence of India (Nowroozi, 1972; Chandra, 1975, 1978; Menke and Jacob, 1976). Billington et al. (1977) have noted that the seismological evidence based on the spatial distribution and focal mechanism solutions of mantle earthquakes in the Hindukush-Pamir knot indicate that a sub-oceanic lithosphere existed between the converging Eurasian and the Indian continents that has been subducted into the upper mantle. They believe that the region is the site of the final stage of subduction of the sub-oceanic lithosphere along the boundary between the two plates. To analyze further the problems associated with the converging plates in the Hindukush region and Bahrchistan arc, a number of focal mechanism solutions were determined in the present work. In all, twenty-two new solutions for the Baluchistan arc and Arabian sea region, and nine solutions for the Hindukush-Pamir area were obtained. The solutions were determined using P-wave first-motion directions reported in the I.S.S. and I.S.C. Bulletins. First-motion directions were plotted on an equal area projection of the lower hemisphere using the (i, A) tables of Nuttli (1969). A double couple source mechanism was assumed. For the purpose of determining the focal mechanism solutions, only those earthquake events were selected which were recorded by at least twenty-five to thirty seismological stations, preferably well-distributed in a.lIthe four quadrants with respect to the epicentral position. However, often not many stations were available in the southwestern quadrant because of the paucity of seismological stations in the African continent. However, sufficient care was taken in selecting the nodal planes to give least ambiguity, as far as possible, in the present data. It may be stated that the focal mechanism solutions presented here can be regarded as quite welldetermined. Further, the freedom of varying the solution parameters is restricted to a few degrees, at the most. In addition to these newly determined solutions, the solutions determined for the area by earlier workers (Ritsema, 1966; Banghar and Sykes, 1969; Nowroozi, 1972; Chandra, 1975) were also considered. The solutions are classified into two categories: (1)

307

those belonging to the Baluchistan arc, Arabian sea and adjacent regions, and (2) those belonging to the Hindukush-Pamir region. This facilitates the analysis of the results, as it is specifically known that the nature of seismic activity under the Hindukush is somewhat different from the neighbouring areas. The solution parameters for mechanism solutions belonging to the first category are given in Table I, and those for the second are listed in Table II. 1. Baluchistan arc and Arabian Sea regions

Altogether thirtyone solutions are considered for this region. All the earthquake events for which the solutions are listed in Table I occurred between 1962 and 1974. Schematic orientation of nodal planes for all the solutions are shown in Fig. 2 along with the tectonic lineaments in the area. Table I shows that the majority of the solutions suggest strike-slip movements, whereas only six normal and three thrust-type mechanism solutions are found in the region. At the places where fracture zones are known to exist, the nodal plane that strikes nearly parallel to the strike of the fracture zone was chosen as the fault plane. It should be pointed out here that though this assumption is widely-followed in the literature for the purpose of selecting the fault plane out of the two nodal planes, it is quite probable that a fault plane may strike orthogonally to the dominant structural trend, in which case the foregoing assumption is no longer valid. Such cases have been noticed by several workers, for instance Fitch (1970) and Verma et al. (1978). In both these cases the nodal planes were oriented in a direction other than the dominant structural trend. The identification of the fault plane from the two nodal planes only on the basis of P-wave data, particularly for non-shallow earthquakes, could, of course, be sometimes misleading. Mainly because of this, when selecting the fault plane, emphasis has been placed on the dominant structural trend as well as on the epicentral pattern found in the region (see later section of the text concerning the Sulaiman wrench zone and the Hindukush region). To support the reliability of the present interpretation (which is solely based on P-wave data), further investigation concerning S-wave data and more geological data are desirable particularly for the remote moutain ranges of which the geology is so poorly known. However, in the absence of other data, the following analysis is based on the above assumption. The solutions corresponding to the individual geographical zones are discussed below. Hazara mountains The mechanism solutions

4, 11, 18 and 20 belong to the earthquake events that occurred in the Hazara mountains. All the solutions suggest strike-slip faulting. The earthquake events 4, 11 and 20 are located in the syntaxial bend area close to the Main Boundary Fault of the Western Himalaya. For this reason, the ESE-WNW oriented nodal plane in the solutions 4,

308 TABLE

I

Solution Arabian

parameters for earthquake mechanisms belonging Sea region (Azimuth and plunge are in degrees)

Event

Date

Lat.

Long.

Depth

(ON)

(‘E)

(km)

to Pakistan-Afghanistan

Magnitude

Sept. Dec. Jan. Feb. Feb. May May Aug. Aug. Aug. Mar. Aug. Nov. Apr. May July May May June Sept. Nov. Dec. Jan. Feb. Aug. Aug. Sept. Nov. Nov. Mar.

1,1962 26, 1962 24,1966 2,1966 7,1966 11,1966 27,1966 1,1966 1,1966 2,1966 3,1968 3,1968 9.1968 25,1969 15,1969 28, 1970 9,197l 17,1972 27.1972 27,1972 3,1912 28,1972 24,1966 7,1966 2,1968 31,1968 1,1968 7,1969 24,1969 30.1969

31

Dec. 28,1974

25.6 23.7 32.7 34.0 30.0 34.5 24.5 29.9 29.9 30.0 34.7 25.2 23.8 30.8 34.6 36.2 35.5 33.5 29.8 34.0 34.1 34.7 29.9 29.8 27.5 33.9 34.0 27.8 37.1 21.8

69.8 68.7 68.7 68.8 68.7 72.4 62.9 64.7 70.4 70.8 68.3 71.1 71.5 70.3 72.7 69.6 70.4 69.7 69.7 60.9 59.0 58.0 60.0 71.6 62.2

0 0 43 37 10 59 5 24 31 2 43 29 15 28 49 45 82 17 18 41 38 69 12 33 62 13 15 35 123 33

35.1

72.9

22

65.2 65.2 67.5 73.0 70.0

1st Pole AZ.

Pi.

4.9 5.0 5.2 5.1 5.0 5.5 5.1 5.0 5.0 4.7 5.1 4.9 5.4 5.0 5.5 5.0 5.0 5.1 5.2 5.4 5.8 6.0 5.7 6.0 5.9 6.1 5.6 5.6

225 30 187 17 259 103 203 282 346 185 322 60 88 237 80 204 106 44 119 304 123 272 102 14 336 359 27 270 321 26

10 0 14 0 11 35 8 30 26 8 7 20 60 9 31 90 24 12 11 12 7 70 12 0 16 0 5 50 5 56

6.0

140

67

1 2 3 4 5 6 I 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

and

-

11 and 20 are chosen as the fault plane. This sense of motion is also in conformity with the northward drift of India. However; near Peshawar where event 18 is located, the trend of surface faults becomes almost ENE-WSW. The nodal plane having this orientation in solution 18 was chosen as the fault plane. These four mechanism solutions for crustal earthquakes reveal that the Hazara mountains are currently undergoing left-lateral movements along steeply dipping fault planes oriented in ESE-WNW direction.

309

P-axis

2nd Pole

T-axis

Type of faulting

Reference

present study present study present study present study Present study present study present study present study present study present study present study present study present study present study present study present study present study present study present study present study present study present study Nowroozi, 1972 Nowroozi, 1972 Nowroozi, 1972 Nowroozi, 1972 Nowroozi, 1972 Nowroozi, 1972 Nowroozi, 1972 Banghar and Sykes, 1969 Chandra, 1975

AZ.

Pl.

AZ.

Pi.

AZ.

PI.

35 120 7 107 163 283 106 102 107 95 223 330 248 132 183 24 309 151 209 212 214 92 194 104 90 89 288 0 14x 320

80 0 76 0 32 55 42 60 46 33 62 0 30 54 22 0 64 58 3 11 12 20 9 0 62 0 50 0 85 60

225 75 7 63 214 103 164 102 36 144 288 106 76 83 135 24 88 199 72 258 168 92

35 165 187 152 117 283 57 282 141 44 172 13 220 204 40 204 293 76 164 348 259 272 -

35 0 59 0 15 10 20 75 12 19 30 14 72 44 6 45 22 46

-

55 0 31 0 31 80 35 15 55 28 40 15 15 28 38 45 68 27 6 16 12 65 -

321 -

51 -

141 -

39 -

normal 8. slip thrust s. slip 6. slip normal s. slip thrust s. slip s. slip s. slip s. slip s. slip s. slip s. slip normal normal s. slip s. slip s. slip s. slip normal s. slip s. slip s. slip 8. slip s. slip s. slip normal s. slip

336

22

151

23

345

67

thrust

-

11; 4 25 -

~~la~rna~ ranges and wrench zone The solutions numbering 5, 8, 9, 10, 14, 19, 23 and 24 belong to the

earthquake events located in the Sulaimanranges and the wrench zone. All the solutions except number 8 suggest strike-slip faulting. The solution number 8 is suggestive of thrust faulting. The nodal planes belonging to events 8, 10, 14, 19, 23 and 24 oriented north--south in these solutionsare most likely to be the fault plane, because of their parallelismwith the tectonic lineation of the Sulaiman ranges. The lef-lateral sense of motion derived from these strike-slipmechanismsfor the earthquake events located in the Sulaimanranges is in agreement with the northward drift of India. In

310 TABLE

II

Solution parameters for earthquake and plunge are in degrees) Event No.

Date

mechanisms

belonging

to Hindukush

Lat.

Long.

Depth

(ON)

(OE)

(km)

Magnitude

area (azimuth

1st Pole AZ.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 46 46 47

Apr. 17,1968 May 8,1968 Jan. 21,1969 May 21,1969 Apr. 23, 1970 Aug. 21,197O Oct. 9,197o Oct. 9,197o Jan. 25,1972 Mar. 4, 1949 Apr. 6, 1956 Aug. 20,1957 Sept. 20,1957 Feb. 17,1958 Mar. 7,1958 Mar. 28, 1958 Aug. 8,1958 Sept. 18, 1958 Sept. 25,1958 Mar. 2,1969 Jan. 9,196O Feb. 16,196O Feb. 19,196O Apr. 20,196O May 19,196O July 6, 1960 June 19,196l July 6, 1962 Dec. 23, 1962 Mar. 11, 1963 Jan. 28,1964 Nov. 16,1964 Mar. 14,1965 July lo,1949 Sept. 16,1956 Nov. 14,1956 Aug. 30,1957 Jan. 6,1958 Mar. 221958 Mar. 20,1961 Sept. 5,196l Sept. 12,1962 June 1,1963 Mar. 14,1965 Jan. 25,1967 Mar. 5,1969 Aug. 8,1969

36.4 37.2 38.3 36.5 37.6 36.4 39.1 39.1 35.6 36.7 36.4 36.67 36.51 36.5 36.55 36.51 36.64 36.49 36.53 36.44 36.5 36.5 36.5 36.2 36.0 36.5 36.5 36.6 38.0 36.7 36.52 36.5 36.3 39.2 34.0 36.7 39.34 36.37 35.3 36.7 38.5 36.5 36.4 36.3 36.6 36.4 36.4

71.5 71.9 69.6 70.2 72.5 68.4 71.5 71.6 69.8 70.5 70.7 71.24 71.18 70.68 70.68 70.98 71.04 70.7 70.11 70.6 70.0 70.5 70.5 70.0 71.0 70.5 70.5 70.4 73.0 71.1 70.01 70.5 70.7 70.7 69.5 71.1 72.88 71.09 67.4 71.2 73.2 63.2 71.5 70.7 71.6 70.7 70.8

124 139 24 224 49 51 39 43 96 220 220 227 222 210 202 188 201 157 219 213 200 216 220 200 200 220 220 203 196 189 207 200 219 N 30 85 56 78 N 86 104 50 70 219 281 208 198

5.2 4.9 4.8 4.8 5.0 5.0 5.0 5.0 5.2 -. -6.6 5.7 5.9 5.8

250 254 274 256 214 49 51 162 227 20 30 225 350 30 25 338 218 218 325 35 35 35 295 35 85 30 10 7 90 37 320 340 35 211 125 135 148 108 145 300 110 140 299 21 242 29 356

P

2 3 3 2 2 2 4 4 5 2 2 1 4 3 2 1 3 2 5 2 2 5 2 2 2 3 1 z 2 3 &

311

P-axis

2nd Pole

T-axis

AZ.

Pl.

AZ.

PI.

AZ.

Pl.

70

10 52 50 11 52 0 0 1 22 74 43 25 50 65 70 60 35 35 20 82 70 65 40 45 30 70 22 75 30 22 62 47 60 40 38 40 65 10 34 61 20 58 40 70 68 76 58

250 210 94 200 225 94 96 212 227 200 240 151 170 210 188 175 7 7 145 215 215 215 308 35 59 210 61 187 164 76 140 132 215 11 125 346 328 180 291 98 331 320 86 201 -

35 72 5 34 3 0 0 24 23 29 14 55 5 20 38 22 3 3 20 37 25 20 4 0 11 25 33 30 58 9 17 17 15 9 7 10 20 45 5 15 0 13 15 25 -

70 94 274 306 126 184 186 111 47 20 345 260 350 30 45 306 272 272 325 35 35 35 200 215 172 30 330 7 60 338 320 22 35 270 305 87 148 60 194 343 61 140 342 21 -

55 15 85 17 79 0 0 24 67 61 47 12 85 70 45 60 50 50 70 53 65 70 78 90 64 55 1 60 9 44 73 47 75 48 83 47 70 27 44 58 29 77 40 65 -

209 179

33 13

29 343

57 77

124 94 158 56 139 141 72 47 200 280 105 170 210 127 208 332 332 145 215 215 215 140 215 220 210 110 187 215 289 140 90 215 333 305 25 328 213 253 65 12 320 42 201 38 209 184

Type of faulting

Reference

thrust s. slip thrust s. slip thrust s. slip s. slip s. slip thrust thrust s. slip s. slip thrust thrust s. slip s. slip s. slip s. slip thrust thrust thrust thrust s. slip thrust s. slip thrust s. slip thrust s. slip s. slip thrust s. slip thrust 8. slip thrust 9. slip thrust s. slip s. slip 6. slip s. slip thrust s. slip thrust thrust thrust thrust

present study present study present study present study present study present study present study present study present study Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritaema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Ritsema, 1966 Nowroozi, 1972 Nowroozi, 1972 Nowroozi, 1972 Nowroozi, 1972

312

Fig. 2; Schematic orientation of nodal planes and directions of P and T axes for 31 earthquakes listed in Table I. Dark area refers to the zone of compression and blank arlea to the zone of dilatation. Inward-directed arrows indicate compression, outward-&e&x! arrows indica%e extension.

313

solution number 8 both the nodal &nes are oriented NNE-SSW, which is again nearly parallel to the structural trend of the ranges. The direction of the seismic slip vector found from this thrust-type mechanism solution is directed towards the west as shown in Fig. 2. This observation implies that the underthrusting of the Indian shield is continuing below the Indus plains along the Sulaiman ranges and Quetta front. The solution 14, corresponding to an earthquake event located at the eastern margin of the Sulaiman ranges adjacent to the Indus plains, has a nodal plane oriented NNW-SSE which may be the fault plane, since faults and folds ar_egeologically known to have this orientation. This mechanism solution suggests that the fault is steeply dipping. The left-lateral motion occurring along the fault as derived from this solution also supports northward movement of India. On similar grounds, the northwesterly nodal planes in the solutions 10 and 3.9 are selected here to represent the fault plane which also calls for left-lateral motion along the Sulaiman shear zone. The solutions 23 and 24, suggestive of strike-slip mechanisms, have one of the nodal planes lying NNE-SSW. This orientation of the nodal plane is parallel to the axis of folding of the Sulaiman range in their respective localities, and hence, is most likely the fault plane. These two solutions are after Nowroozi (1972) who has suggested that the Sulaiman ranges are undergoing left-lateral movements. The three solutions (Nos. lo,14 and 19) determined in the present work also support this view as discussed above. These results are in conformity with the views held by Wilson ‘{1965) and Abdel-Gawad (1971). It may, however, be noted that the focal mechanism studies suggest steeplydipping fault planes in the area, and the Sulaiman ranges are mostly under the action of a horizontal compressive stress system (refer to Table I). The solutions 5 and 9 are for typical earthquake events belonging to the wrench zone which is characterized by an east-west trending epicentral alignment (for a fuller description of the wrench zone refer to Abdel-Gawad, 1971). These two solutions are for shallow crustal shocks. Both the solutions are suggestive of a strike-slip mechanism. The nodal planes in the ENE-WSW direction are selected here as the fault plane in both the solutions, in conformity with the epicentral alignment over the whole length of the wrench zone as well as the garland-type fold axes that typify the area in this direction. The solutions further suggest that the lateral faults belonging to the wrench zone are steeply dipping. Chaman fault zone The mechanism solutions 3, 6, 15, 17, 21,22 and 29 belong to the earthquakes associated with the Chaman, Darafshan and parallel faults. These faults are northeasterly trending, running to the west of the SulaimanKirthar ranges from about 27”N latitude to beyond 35”N. Out of this set of NE-trending faults, the Chaman fault is the longest (proved length 800 km). The north end of the Darafshan fault is about 100 km west-southwest of Kabul. The fault strikes southwest almost parallel to the Chaman fault, and

314

extends continuously for 300 km (Wellman, 1965). To the east of the Chaman fault, there are a number of northeasterly trending faults running parallel to subparallel with the Chaman fault. We designate this entire set of northeasterly trending faults as the Chaman fault zone. The solutions mentioned above correspond to the earthquakes belonging to this set of faults. To the north of Kabul, many of these faults merge with the Herat fault of Afghanistan, Out of the seven solutions, only solution No. 3 suggests thrust faulting along east-west planes, while three solutions suggest strike-slip faulting, and the rest are normal-type mechanisms. In all the six normal/strike-slip mechanism solutions, one nodal plane is always oriented in a north to northeasterly direction, which is most likely to be the fault plane. The normal-type mechanisms (solution 6,22 and 29) are found to occur in the mounta~ous areas to the north of Kabul. This may be attributed to the dominance of vertical tectonics within the elevated ranges. In all the three solutions (No. 15, 17 and 21) suggestive of strike-slip mechanism, the nodal plane oriented in north-south (in solution 15) to NE-SW direction (in solutions 17 and 21) is most likely to be the fault plane as it is parallel ‘to the strike direction of the Chaman and associated faults. An important conclusion that may be drawn from these results is that the Chaman and its parallel faults are currently undergoing left-lateral movement, and these motions are similar to those of the Sulaiman range in accordance with the northward drift of India. However, it is difficult to estimate from the present data whether there is any relative difference in amplitude of the movement among the crustal blocks located to the west of the Chaman fault, the blocks lying between the Chaman and the Sulaiman ranges, and the areas to the east of the Sulaimans. Nowroozi (1972) has suggested that the Indian plate is moving no~he~tw~d faster than the Afghanistan plate. This differential movement between these two plates produces the observed leftlateral displacement along the Kirthar-Sulaiman ranges as well as the shear zone. The present results suggest that the Chaman fault is sinistral transcurrent in sense, and the faults to the north of Herat fault (i.e. the northeasterly continuation of the Chaman and parallel faults), are also of the same type (see Henckroth and Karim, 1970). Wellman (1965) estimated an apparent displacement of about 500 km in a sinistral direction using Tertiary facies which occur on the east of the fault in southwest ~gh~~t~ to the west of the fault in southern Pakistan, Referring back to solution 3, suggestive of thrust-type mechanism for an earthquake located to the west of the Chaman fault, it may be stated that there is evidence that the fault has been upthrown on the west side (Wellman, 1969). Arabian Sea The mechanism solutions 2, 13 and 30 belong to the area of the Murray Ridge in the Arabian Sea. Solutions 2 and 13 obtained in the present work

315

are for shallow shocks, and solution 30 was determined by Banghar and Sykes (1969) for an event having a focal depth of 33 km. All the three solutions suggest strike-slip faulting. In the mechanism solution of event 30, the NNE oriented nodal plane was chosen as the fault plane by Banghar and Sykes (1969). However, if the northeasterly oriented nodal plane is regarded as the fault plane by Banghar and Sykes (1969), it represents a motion opposite to that of the general northward movement of India. Further, this would require a right-lateral motion along the Ridge, a motion opposite to that inferred along the Chaman fault, the Sulaiman-Kirthar ranges, and along the Hazara mountains. There is no geological or any other evidence to support this suggestion of right-lateral motion along the Ridge. Such motion, if occurring at all, would also oppose the motion along the Ridge envisaged by Wilson (1965). For all these reasons, the northwesterly oriented nodal plane of the solution 30 is regarded here as the fault plane. Similarly, the northwesterly oriented nodal planes in the solutions 2 and 13 are chosen as the fault plane. The inferred sense of motion’along the northwesterly trending fault plane from these three solutions is left-lateral in nature. This sense of motion is in agreement with the northward drift of India, and is also in conformity with the left-lateral motion along the Kirthar-Sulaiman shear zone as well as the Owen fracture zone. The left-lateral motion along the northwesterly nodal plane which is inferred here offers an interesting possibility: that the Murray Ridge is affected by transverse faulting in a northwesterly direction. This would also explain why the earthquake event 30 was located off the axis of the Murray Ridge (originally noted by Barker, 1966). Extrapolating from this reasoning one might observe that in the solution 12, for an earthquake event located off the Makran Coast, the northwesterly nodal plane is most likely the fault plane. If this is the case, the postulated fault would again be transverse in direction to the Murray Ridge as is possibly the case with the events 2, 13 and 30. These results indicate that the Murray Ridge is possibly offset by northwesterly faults along which left-lateral motion is currently taking place. Whether northwesterly faults are transform faults is a matter for further investigation. It may be noted that several offsets in the bathymetric and magnetic contours in Murray Ridge area provide some support for transverse faulting in the Ridge area (see figs. 2 and 3 in Barker, 1966). Eastern Persia-West Pakistan The mechanism solutions 25 and 28 are for the events located in the areas where the north-south aligned Dashte-Lut fracture zone of Eastern Persia takes an arcuate turn to meet the Siahan ranges of West Pakistan and Makran coastal ranges. To the north, events 26 and 27 belong to the localities crossed by the Herat and Ferdows faults in an east-west direction. All the four solutions are after Nowroozi (1972) who thought that the events 25 and 28 were located in a place in southeastern Persia where the structural

316

trend varies from a north-south to an east-west direction. If the nodal planes with a north-south trend are taken to be the fault plane, the sense of motion is lef-lateral for the solution 25 and right-lateral for the solution 28. If the nodal planes with the east-west trend are taken to be the fault plane, the sense of motion reverses for both solutions (Nowroozi, 1972). On the other hand, the sense of motion for solution 28 is similar to that derived from other mechanism solutions in Persia, whereas the sense of motion for the solution 25 is similar to those in the Kirthar-Sulaiman shear zone. There are a few Quarternary volcanoes and one active volcano in the vicinity of these locations. Ahmed (1969) has suggested a zone of weakness indicated by major faulting and igneous activity passing through this area (see fii. 9 in Ahmed, 1969). All these factors may suggest a change in current tectonics of southeastern Persia. The mechanism solutions 26 and 27 are for the earthquake events located in northeastern Persia. The event 27 is an aftershock of the event 26. Both the events were associated with the Ferdows fault passing north of the Lut block of Iran. The solution for the main shock (event 26) suggests pure strike-slip movement while that for event 27 is suggestive of strike-slip mechanism but with a substantial thrust component. A similar situation is observed in the solutions 8, 9 and 10 for earthqtie events that occurred in the Sulaiman ranges and the Sulaiman wrench zone where the solution for mainshock differs from the aftershocks (refer to Table I). It is intriguing that, in contrast to many other areas, the solution for the main shock (event 26) differs from that of the aftershock (event 27), when their focal depths are comparable and both are presumably associated with the Ferdows fault. However, the results probably imply that the overall tectonic regime is due to northward motion of the Lut plate as suggested by Nowroozi (1972, Fig. 8). 2. Hindukush region Nowroozi (1972) has included the Hindukush seismic zone in his Afghanistan plate whose southeastern boundary is the Kirthar-Sulaiman shear zone and the probable northern boundary is the Yasaman fracture zone and the Hissaro-Kokshaal fault zone, which pass through the south Tien-Shan mountains of the U.S.S.R. (Gubin, 1967). For the Hindukush region, altogether 47 focal mechanism solutions have been considered in the present study. The parameters for the mechanism solutions are listed in Table II, which shows that the solutions are of strike-slip or thrust type. Because of the complex nature of seismicity under the Hindukush (which has been discussed above), the results of focal mechanism studies are analyzed here from the point of view of focal depth as well as the type of faulting. For this purpose, the mechanism solutions were classified according to the type of faulting (i.e. thrust or strike-slip), and were plotted against the focal depth of the individual events irrespective of their epicentral locations. The plot is shown in Fig. 3. Several

317 THRUST 0,

STRIKE

FAULTING

- SLIP

FAULTING

I

Fig. 3. Plot of schematic orientation of nodal planes classified according to the type of faulting versus focal depths for the earthquake mechanism solutions for the events belonging to the Hindukush region listed in Table II. Note that in the case of thrust faulting, both the nodal planes are in a northwestern direction in the majority of the solutions. For the thrust mechanism solutions, the direction of seismic slip vectors (shown by arrows) are in a northeastern direction in most of the cases. In the case of strike-slip faulting, one nodal plane is in the northeastern direction .in the majority of the cases.

inferences can be made from this plot. These are: (a) Thrust as well as strike-slip faulting takes place under the Hindukush from shallow to intermediate depths. (b) Both the nodal planes of the solutions suggestive of thrust mechanisms are oriented in a northwestern direction apart from one or two exceptions. Since one of the two nodal planes represents the fault plane, it may safely be concluded that a northwesterly oriented fault zone exists under the Hindukush, which parallels the northwesterly epicentral alignment discussed above. Consequently, the seismic slip vectors (shown in Fig. 3) for thrust mechanism solutions are in a northeastern direction. This picture is consistent for nearly all shallow as well as intermediate depth earthquake events listed in Table II and is in general agreement with the northward motion of India as well as the inferred motions for the Afghanistan and Lut Block plates of Nowroozi (1972). The latter two regions are located to the south and southwest of the Hindukush.

318

(c) One of the two nodal planes in a majority of the strike-slip mechanism solutions has the orientation to the northeast. This orientation of the nodal planes parallels the northeasterly alignment of the seismic zone. This observation suggests that under the Hindukush and the Pamirs (up to 37.5”N), a northeasterly fault/fracture zone exists which may be recognized as the Pamir trend in a N45”E direction. The inferred sense of motion along this northeasterly fault zone is left-lateral in sense, and is consistent with the general northward motion of India. This sense of motion is also in conformity with that associated with the Kirthar-Sulaiman ranges and the Owen fracture zone. However, nine out of twenty-four solutions suggestive of strike-slip mechanisms indicate that strike-slip faulting also takes place along northwesterly directed faults. This suggests a relative movement of the Indian plate with respect to the Afghanistan plate in a northwesterly direction. These solutions

0

SENSE

OF

LATERAL

FAULTING

T-

20 40 60 r Y 2

00 100

c’ a.

I20

:

140

:

160

180 200 220 240

Fig. 4. Plot of left-lateral sense of motion along the northwesterly as well as the northeasterly fault planes noted in the Hindukush region against the focal depth of the events suggestive of strikeaip mechanlams listed in Table II. Note that left-lateral motion along faults in either direction takes place from shallow to intermediate depths. Thii result is contrary to the hypothesis of decoupling of stress-system with depth under the Hindukush advanced by several workers.

319

are 6, 7, 11, 25, 27, 30, 32, 39 and 40 (refer Table II). The sense of motion along the northwesterly planes derived from these nine solutions is again leftlateral. Obviously, these results show the complex nature of Hindukush seismicity. However, for purposes of presenting a composite picture of lateral faulting under the Hindukush, the sense of motion derived from all the mechanism solutions suggestive of strike-slip mechanisms were considered vis-a-vis the respective focal depth of the events. In Fig. 4, the left-lateral sense of motion along the northeasterly and northwesterly fault planes for the various events against the focal depths are shown. The plot suggests that no clear distinction can be made concerning the left-lateral sense of motion along northwesterly and northeasterly fault planes as far as the focal depth is concerned. SUMMARY

AND CONCLUSIONS

The Hindukush is the only region between the Mediterranean and the Burmese arc where large scale tectonic forces are continuously active from shallow to intermediate depths. The region of intense seismic activity is extremely localized and the seismic energy released is several orders of magnitude greater. Because of its complex seismic characteristics as well as typical geographical location, the Hindukush region probably deserves more attention than it has so far received. The analysis of earthquake data over a sufficiently long time period suggests that two distinct seismic lineaments (i.e. the northeastern Pamir trend and the northwestern trend) in mutual orthogonal directions are present under the Hindukush region. Both the seismic trends are characterized by shallow to intermediate-depth earthquakes. These two trends juxtapose one another under the Hindukush knot where most intense seismicity is observed at a depth of about 220 km, and where the activity is rather localized. Seen on a regional scale, the northeasterly seismic trend in the Hindukush region could be considered as the Pamir trend in NE direction paralleling the western Himalayan syntaxial bend. On the other hand, the northwesterly seismic trend appears to be an extension of the western Himalayan seismotectonic trend extended in a northwestern direction across the Hindukush towards the Uzbekistan area of the U.S.S.R. The microseismic data collected by the Lamont Group in the Tarbela dam area in Pakistan support the existence of the northwestern seismic trend in that locality. Menke and Jacob (1976) consider this seismic pattern in Pakistan and northwestern India to be a major feature associated with continental collision. The epicentral distribution over the Kirthar-Sulaiman ranges follows the north--south structural trend of these ranges. Over the Sulaiman wrench zone, however the epicentral alignment is in a east-west direction, following the structural hneation of the wrench zone. The Sulaiman wrench zone is found to be seismically much more active than the Sulaiman and Kirthar ranges.

320

The results of foci-me~h~ism studies discussed above suggest that both under the Hindukush and the Baluehistan arc, a large part of seismic energy is released through transcurrent faulting. In the Hindukush region, the northeasterly seismic lineament, which is recognized as the Pamir trend, appears to have a bias for strike-slip faulting. Along the northwesterly seismic zone, both thrust as well as strike-slip mechanisms are found to occur, though it appears that this seismic lineament has a preference for thrust faulting. Thrusting along the northwestern seismic zone appears to be an outcome of the effect of northward movement of the Indian peninsula as well as northeasterly movement of the Afghanistan and Lut Block plates located to the south and ~uthwest of the Hindukush. As a result of this movement, seismic slip vectors in this region trend northeasterly. The focal-mechanism studies suggest that the Baluchistan arc is a major left-lateral shear zone, located between the Lut Block of Eastern Persia and the Indian shield. Transcurrent faulting takes place along north--south directed faults within the Sulaiman-Kirthar ranges but the Hazara mts. and the Sulaiman wrench zone are ‘affected by lateral faulting along east-west oriented faults. The Sulaiman wrench zone is by far the most active part of the Baluchistan arc affected by lateral faulting. The Chaman fault located to the west of the Baluchistan arc, is a major sinistral transcurrent fault located between the Lut Block on the west and the Baluchistan arc on the east. Obviously, this raises the question whether the Afgh~~t~ plate of Nowroozi (1972) behaves as a single plate or is devided by the Chaman fault zone into two smaller plates. It is quite probable that all these small plates, viz. the Lut Block, the blocks to the east and west of the Chaman fault (conjugately forming the Afghanistan plate) and the Indian plate on the extreme east, are moving northward at different rates. As a result of this, zones of left-lateral shear have been created at the margin of the Indian plate as well as at the margins of the smaller plates further west. The Murray Ridge, which is thought to be an oceanic extension of the Baluchistan arc, is believed to be affected by lateral faults in a northwestern direction. In all probability. these left-lateral faults in a northwesterly direction are responsible for offsets in the Ridge axis where several earthquakes occur, Obviously, more data are required to prove or disprove whether these are transform faults. ACKNOWLEDGEMENTS

We are grateful to Dr. H.M. Chaudhury for permitting us to consult the I.M.D. Earthquake Catalogue. Thanks are also due to Shri S.K. Gangopadhyay for typing the manuscript and to Shri B.N.P. Singh for drawing the figures.

321

REFERENCES Abdel-Gawad, M., 1971. Wrench movements in the Baluchistan arc and relation to Himalayan-Indian Ocean tectonics. Geol. SOC. Am. Bull., 82: 1235-1250. Ahmed, S.A., 1969. Tertiary geology of part of South Makran, Baluchistan, W. Pakistan. Bull. Am. Assoc. Pet. Geol., 53 (7): 1480-1499. Armbruster, J., Seeber, L. and Jacob, K.H., 1977. The northwestern termination of the Himalayan mountain front: Active tectonics from microearthquakes. (under communication). Banghar, A.R. and Sykes, L.R., 1969. Focal mechanisms of earthquakes in the Indian Ocean and adjacent regions. J. Geophys. Res., 74 (2): 632-649. Barker, P.F., 1966. A reconnaissance survey of the Murray Ridge. Philos Trans. R. SOC. London, Ser. A, 259: 187-197. Billington, S., Isacks, B.L. and Barazangi, M., 1977. Spatial distribution and focal mechanisms of mantle earthquakes in the Hindukush-Pamir region: a contorted Benioff zone. Geology, 5: 699-704. Chandra, U., 1975. Fault-plane solution and tectonic implications of the Pattan, Pakistan, earthquake of December 28,1974. Tectonophysics, 28: T19-T24. Chandra, U., 1978. Seismicity, eachtquake mechanisms and tectonics along the Himalayan mountain range and vicinity. Phys. Earth. Planet. Int., 16: 109-131. Desio, A., 1965. On the tectonic connection between Pamirs and Hindukush. In: D.N. Wadia Commem. Vol. Min. Geol. Metall. Inst. India, pp. 716-721. Farhoudi, G. and Karig, D.E., 1977. Makran of Iran and Pakistan as an active arc system. Geology, 5: 664-668. Fitch, T.J., 1970. Earthquake mechanisms in the Himalayan, Burmese, and Andaman regions and continental tectonics in Central Asia. J. Geophys. Res., 75: 2699-2709. Gubin, I.E., 1967. Lecture notes on basic problems in seismotectonics. Int. Inst. Seismol. Earthquake Eng., Tokyo, Japan. Gupta, H.K., 1976. Seismological investigations and the tectonics of the Kashmir-Hindukushpamir region. Atti Accad. Naz. Lincei, Atti Dei Congegni Lincei, 21: 43-66. Heuckroth, L.E. and Karim, R.A., 1970. Earthquake history, seismicity and tectonics of the regions of Afghanistan. Report of Kabul Univ., Nov. 1970. Krishnan, M.S., 1953. The structural and tectonic history of India. Mem. Geol. Surv. India, 81: 137 pp. Menke, W.H. and Jacob, K.H., 1976. Seismicity patterns in Pakistan and Northwestern India associated with continental collision. Bull. Seismol. Sot. Am., 66 (5): 16951711. Molnar, P. and Tapponnier, P., 1975. Cenozoic tectonics of Asia: Effects of a continental collision. Science, 189: 419-426. McElhinny, M.W., 1968. No~hw~d drift of India- examination of recent palaeomagnetic results. Nature, 217 : 342-344. Nowroozi, A.A.. 1971. Seismotectonics of the Persian Plateau, Eastern Turkey, Caucasus, and Hindukush regions. Bull. Seismol. Sot. Am., 61 (2): 317-341. Nowroozi, A.A., 1972. Focal mechanism of earthquake in Perisa, Turkey, West Pakistan and Afghanistan and plate tectonics in the Middle East. Bull. Seismol. SOC. Am., 62 (3): 823-850. Nuttli, O.W., 1969. Tables of angles of incidence of P-waves at focus, calculated from 1968 P-tables. Earthquake Notes, 40: 21-25. Ritsema, A.R., 1966. The fault-plane solutions of earthquakes of the Hindukush Centre. Tectonophysics, 3 (2): 147-163. Santa, T., 1969. Regional study on the characteristic seismicity of the world. Part I. Hindukush region. Bull. Earthquake Res. Inst., 47: 1035-1048. Schreiber, A., Weippert, D., Wittenkindt, H. and Wolfart, R., 1972. Geology and petro-

322 leum potentials of central and south Afghanistan. Bull. Am. Assoc. Pet. Geol., 56 (8): 1494-1519. UNESCO MAP, 1971. Geological Map of Asia and the Far East. Prepared under the sponsorship of ECAFE, scale 1 : 5 million. Verma, R.K., Mukhopadhyay, M. and Bhuin, N.C., 1978. Seismicity, gravity and tectonics in the Andaman Sea. Proc. Semin. Geodynamics of Western Pacific, Tokyo, March 1978 (in press). West, W.D., 1934. The Baluchistan earthquakes of August 25th and 27th, 1931. Mem. Geol. Surv. India, 67: l-82. Wellman, H.W., 1965. Active wrench faults of Iran, Afghanistan and Pakistan. Geol. Rundsch., 55: 716-735. Wellman, H.W., 1969. Wrench (transcurrent) fault systems. In: P.J. Hart (Editor), The Earth’s Crust and Upper Mantle. Am. Geophys. Union, Monogr., 13: 544-549. Wilson, J.T., 1965. A new class of faults and their bearing on continental drift. Nature, 207: 343-347.