New data presented herein support the previously suggested existence of an East Arabian Tectonic Block in Saudi Arabia. However, the East Arabian Tectonic Block is not exclusively bound by grabens, as was previously thought, but also by strike-slip faults and transpressive folds (in what is named here the Mughrah-Al-Kharj Transpression Zone), which laterally grade into the transtensive Central Arabian Graben System. Evidence for Plio-Quaternary strike-slip motion is presented here for the first time and supports previous contentions that graben formation occurred in Plio-Quaternary time. These new observations allow the development of a tectonic model that succinctly explains the deformation along the margins of the East Arabian Block.

With the importance of regional fault structures thus recognized, even some major faults of tectonic dimensions start to emerge in the Eastern Province of Saudi Arabia. For example, radar images from the Space Shuttle Endeavour confirm that the Nisah Fault extends eastward along Wadi As-Sahba, over a total distance of 250 kilometers, which was previously undemonstrable. The Sahba Fault is made visible beneath the Ad-Dahna Sand Sea by virtue of the sand-penetrating capability of the L-band radar. Five to eight kilometers of left-lateral strike separation is visible on either side of Wadi As-Sahba which truncates the southern end of the Ghawar field. The graben structure of the Nisah Fault System had been reported before, but its eastward continuation and any strike separation were not established until now. Furthermore, the Nisah-Sahba Fault System may extend further eastward, beneath alluvial gravels, for an additional 200 kilometers to join a known lineament at the southern end of Qatar. Further extension towards the Gulf into the 600-kilometer long Trans-Arabian Gulf Fault is also suggested here for the first time.

Field evidence presented here indicates that a substantial part of the motion of the Central Arabian Graben System took place in the Plio-Quaternary and may continue at present. This has been inferred from a study of the wadi (flood valley) system on the Jubaila and Hanifa outcrops in the vicinity of the graben structures to the south of Riyadh. In fact, an 8-kilometer long section of a mature stream channel, known as Sha’ib al-Miklaf, which was formerly discharging intermittent wadi drift and torrential run-off waters into Wadi Hanifa, has now been cut off from its former catchment area by the graben tectonics. As a result Sha’ib al-Miklaf now lies idle between the active wadis of the Awsat Graben and the Nisah Graben. The occurrence of recent fault motion in the Arabian Platform is further supported by changes in thickness of the Ad-Dahna Sand Sea across the Sahba Fault, as well as by truncation of mega-barchanoids in the Jafurah Sand Sea across the Jaub lineament. Not surprisingly, some modern seismicity in the Eastern Province of Saudi Arabia may be attributed to motion on the Nisah-Sahba Fault System.


The major, proven oil reserves of the Kingdom of Saudi Arabia are principally located below its Eastern Province, hosted in a Phanerozoic platform sequence underlain by the Precambrian basement of the Arabian Plate (Figure 1). Vigorous exploration has continued over the past sixty five years. Saudi Arabia’s proven oil reserves today amount to some 265 billion barrels, or 25% of the world’s total reserve (Petroconsultants, 1998). The entrapment of the Arabian “black gold” is due to a combination of several favorable factors: good source rocks, slow burial and migration, and huge trap development. Most of the reservoir volume occurs in several large, gentle anticlines, superimposed on a regional homocline with an average eastward dip on top of the basement of less than one degree. These structural traps include the Ghawar field, hosted in the En-Nala Anticline (Figure 1), the world’s largest oil field with about 100 billion barrels, or some 10% of total global reserves.

Our understanding of the Arabian trap formation is rapidly evolving. Complex fault patterns emerge in every major oil field as 3-D seismic uncovers details which were previously beyond resolution. With the number of faults increasing as fractal-numbers, it seems timely to re-assess the extent of Arabian oil-trap tectonics. Although the surface of the Arabian Plate is exceptionally well-exposed, most existing geological maps show little structural information. Most of our modern maps are still based on a landmark effort conceived in 1953 under joint sponsorship of Aramco and the United States Geological Survey Mission that covered the Arabian Peninsula in a series of 21 geological and geographic maps on a scale of 1:500,000, published between 1956 and 1964 (see Powers et al., 1966; Brown et al., 1989). Although the geology on these maps is still accurate, by and large, much of this early mapping work was completed under the assumption that the Arabian Platform remained tectonically stable. Relatively few faults have been demarcated and any faults seen in the field were commonly explained as local collapse features due to solution transfer of underlying evaporites (Powers et al., 1966; Vaslet et al., 1991).

In this paper, new structural elements, which outline the East Arabian Tectonic Block, are shown. These elements are based on improved geologic understanding of the Arabian Platform resulting from numerous field trips, the study of Landsat images, and public-domain radar images collected by NASA’s space shuttle Endeavour during two 1994 missions. This paper is subdivided into the following sections, each outlining a particular segment of the boundary of the East Arabian Block:

  • Central Arabian Graben System

  • Radar Image Unveils Sahba Strike-Slip Fault Below Ad-Dahna Sand Sea

  • Drainage Diversion at Sha’ib al-Miklaf Dates Graben Slip

  • Folding and Transpression Along the Central Nisah-Sahba Fault in the Mughrah-Al-Kharj Transpression Zone

  • Strike-Slip Component of the Durma, Barrah and Majma’ah Grabens

The geological observations documented in the sections listed above are used in three subsequent sections, as follows:

  • Regional Tectonic Framework: Enter Trans-Arabian Gulf Fault

  • Implications for Oil Traps

  • Critical Assessment of the Anhydrite Collapse Hypothesis

Central Arabian Graben System

Surface mapping has previously revealed the existence of a graben system which is most prominently developed in Wadi Nisah, 40 km south of Riyadh (Powers et al., 1966; Al-Kadhi and Hancock, 1980; Hancock and Al-Kadhi, 1978; Hancock et al., 1981, 1984; Breton, 1982; Manivit et al., 1985; Breton et al., 1991; Vaslet et al., 1991). Figure 1 summarizes the complex grabens and lineaments, which together form a giant arc, 500 kilometers (km) long, and concave toward the east. Previously, faulting had only been confirmed by field observations in the more or less continuous system of aligned and en-echelon grabens that stretches between the towns of Al-Kharj (40 km southeast of Riyadh) and Majma’ah (170 km northwest of Riyadh): the Nisah Graben, Awsat Graben, Durma Graben, Qaradah Graben, Barrah Graben, and Majma’ah Graben. This interconnected pattern of grabens has been termed the Central Arabian Graben System (Powers et al., 1966).

The east-west trending Nisah Graben is about 90 km long and 2 to 3.7 km wide. Dip-slip on the normal faults which bound the graben is estimated to range between 100 and 500 meters (m) (Powers et al., 1966; Vaslet et al., 1991). Most of the graben floor is obscured by alluvial fill. The Awsat Graben is also about 90 km long, and the downthrow of the graben floor has been estimated at 50 to 300 m (Powers et al., 1966). The Durma Graben is about 40 km long and 1 to 2 km wide, with a minimum downthrow of 300 to 400 m (Powers et al., 1966; Manivit et al., 1985). The Qaradah Graben is a 10 km long, 4 km wide splay of the Durma Graben with 50 to 100 m relief of faulted Tuwaiq Limestone. The fault drop is at least 400 m (Powers et al., 1966). The Barrah Graben comprises of a 16 km long, 2 km wide block of Tuwaiq Limestone, downfaulted into Dhruma rocks with fault drop estimated at 200 to 300 m (Powers et al., 1966; Vaslet et al., 1988). The Majma’ah Graben is about 1 km wide and nearly 100 km long. Normal fault movement is estimated here at 150 to 350 m at the southern end of the graben and progressively diminishes to zero in the extreme north (Powers et al., 1966; Vaslet et al., 1988).

Figure 1 includes major photo-lineaments, which have been reported before: Al-Batin Lineament (Bramkamp and Ramirez, 1959; Al-Sarawi, 1980), Ar-Rumah Lineament (Al-Khatieb, 1981), Al-Hamudiyah Lineament (Hancock et al., 1981), Az-Zilfi Lineament (Hancock et al., 1984), and As-Sahba Lineament (Powers et al., 1966; Brown, 1972).

Hancock et al. (1984) coined the term East Arabian Block for the eastern platform section that is surrounded by tectonic graben faults and lineaments (Figure 1). Below, we expand this important previous work and provide new data in support of the existence of an East Arabian Block of tectonic origin. The major contribution made here is the presentation of new field observations, which reveal that the East Arabian Block is not exclusively bound by grabens, but also by strike-slip faults and transpressive fold zones, which laterally grade into the transtensive Central Arabian Graben System. New evidence for Plio-Quaternary strike-slip motion is presented here for the first time and supports earlier suggestions of Plio-Quaternary graben formation by Vaslet et al. (1991). These new observations allow the development of a tectonic model that succinctly explains the formation of the East Arabian Block.

Radar Image Unveils Sahba Strike-Slip Fault Below Ad-Dahna Sand Sea

The on-going radar microwave experiment of the National Aeronautics and Space Administration (NASA) Space Shuttle program continues to deliver remote sensing images of geological features previously obscured from direct view by desert sands and other surficial deposits. Several new sutures and faults have already been reported from areas scanned by radar microwaves in earlier space shuttle flights (Science, 1995; Stern and Abdelsalam, 1996). The faults become visible in images produced by L-band radar waves reflecting off bedrock below surficial sand sheets. The radar waves can penetrate a few meters below the sand surface, depending upon physical conditions in the surficial deposits.

Recent public domain radar images covering parts of the Arabian Peninsula also shed new evidence supporting the extension of the Nisah Fault Zone into the Sahba Fault. Wadi Sahba is an eastward continuation of Wadi Nisah. These wadis together form an east-west striking lineament, of several hundred kms long (Figure 1). Although the significance of this lineament is subject to continued investigations, most researchers agree that it represents an important fault zone. Here new data are presented, which enhance our understanding of this fault zone.

The Space Shuttle Endeavour missions, in April and October 1994, carried a sophisticated imaging radar system, i.e. Shuttle Imaging Radar, C-band (SIR-C) and Synthetic Aperture Radar, X-band (X-SAR), designed to study the Earth and its environments. This radar system provided increased capability over previous missions by acquiring digital-image data at three microwave wavelengths; namely, L-band (24 cm), C-band (5.6 cm) and X-band (3 cm), four polarization modes HH, VV, HV, and VH modes (first ‘V’ or ‘H’ stands for vertical or horizontal transmission and second ‘V’ or ‘H’ stands for vertical or horizontal reception, together accounting for these four possible polarization modes). The radar system images a surface by emitting radiation and recording the returned echo. The ground parameters that control the radar signal returned from a surface are the orientation, roughness, compactness, and dielectric properties of the material. Comparison of the original and reflected ray yields the distance, size, orientation, roughness and characteristics of the reflector (Evans et al., 1994).

Previous studies have indicated instances of L-band radar penetration in loose, dry sand (McCauley et al., 1982, 1986; Elachi et al., 1984; Schaber et al., 1986). Laboratory measurements have shown that dry sand and soil have very low dielectric constants (3.1 to 3.7). These low dielectric values, together with other material properties, indicate a potential radar-penetration, or skin-depth, ranging from 1 to 6 m in sand sheets (Sabins, 1987). Field investigation of physical control on radar signal penetration and subsurface scattering into sand sheets in the eastern Sahara showed that radar imaging reached a maximum depth of 1.5 m (Schaber et al., 1986).

During each of the two Endeavour missions, fourteen data-takes were acquired over the Arabian Peninsula (e.g., Al-Hinai et al., 1997). Data-take number: 107.30 SRL-1 (April 1994) passed over a section of Wadi As-Sahba where it is covered by the Ad-Dahna Sand Sea (Figure 2a). Wadi As-Sahba is strategically located near the southern limit of Ghawar field. The fault trace is invisible on the corresponding Landsat image of Figure 2b, where the Ad-Dahna sands are seen covering and obscuring the radar-detected fault trace. The Ad-Dahna Sand Sea consists exclusively of Quaternary eolian deposits that migrate toward the Rub’ Al-Khali with southbound trade winds (McKee, 1979).

The radar image of Figure 2a, stretches across a section of Wadi As-Sahba where lower Tertiary strata are largely covered by the Ad-Dahna Sand Sea. However, the drainage pattern and interpreted karst depressions, carved into the bedrock below the dune sand, are clearly visible on the radargram of the area north of Wadi As-Sahba where the sand sheet is relatively thin (Figures 2a and 3). The relative absence of such drainage patterns in the area immediately to the south along the Wadi Sahba Fault suggests that the sand cover is too thick there to allow radar microwaves to reach the bedrock. Field observations have confirmed an increase in the thickness of the sand cover to the south of the Nisah-Sahba Fault trace (Figure 4). Sand thickness varies greatly and is commonly estimated from differences between interdune lows and crests, and exposure of scree horizons with bedrock fragments. Drifting sand has filled in the structurally lower area on the south side of the fault, while the northern wall is almost deflated to bedrock level.

On the north side of the fault, thickness of the sand varies between 2 m (interdune areas) to 10 m (transverse and barchan dune crests, Figure 5a), whereas sand thicknesses greater than 10 m occur on the south side of the fault (Figure 5b; sand thicknesses in accordance with estimates of US Army Map Service, 1958). Although the radar image of Figure 2a cannot be directly translated into a sand isopach map, it provides a most useful, qualitative map view of spatial variations in the sand thickness. The conclusion is that the change in bedrock depth or dip separation on either side of Wadi As-Sahba reflects a minor, vertical component of relative movement (Figure 4).

In the west of the image area (Figures 2a and 3), 8 km of left-lateral strike-separation is visible across Wadi Sahba. This off-set is marked by exposed bedrock at the base and within the Paleogene Umm Er Radhuma limestone. These rocks form a north-south trending, gentle ridge in the landscape, but younger strata are less resistant and disappear eastward beneath the eolian blanket of the Ad-Dahna Sand Sea. The ridge of exposed Paleogene rocks occurs in the basal part of the 243 m-thick Umm Er Radhuma Formation, which dips less than one degree, homoclinally, towards the east. We interpret the eastern segment of the Nisah-Sahba Fault principally as a sinistral strike-slip zone, with minimum lateral displacement of 5 to 8 km, accompanied by a minor upthrow of the northern wall of only a few meters.

In fact, the possibility of left-lateral shear at the eastern segment of the Nisah-Sahba Fault system has been suggested earlier using structure contours on the base of the Mesozoic and on top of the crystalline basement, which are offset to the west on the northern side of the Sahba lineament (Brown, 1972). The gentle eastward dip of the Paleogene rocks of less than one degree further supports this interpretation, because the large strike-separation of up to 8 km cannot be explained by a minor dip-slip of only a few meters on a dip-slip fault.

Remote sensing of Arabian Platform rocks from the Space Shuttle Endeavour has thus revealed further evidence of a major fault line transecting the Eastern Province of Saudi Arabia. Previously, the eastern extension of the Nisah Fault System was only recognized as a poorly defined photo-lineament.

Drainage Diversion at Sha’ib al-Miklaf Dates Graben Slip

Sharpness of the margins of the Central Arabian Graben System led Wolfart (1961) to suggest the occurrence of some Quaternary movement on the boundary faults. However, Hancock et al. (1984) reported that evidence for Quaternary motion could not be established, because alluvial fans sat astride boundary faults. New field evidence, reported hereinafter, supports the contention by Vaslet et al. (1991) that a substantial part of the subsidence of the Central Arabian Graben System took place in the Plio-Quaternary and may continue at present. This conclusion has been inferred from a study of the wadi system on the Jubaila and Hanifa outcrops in the vicinity of the graben structures to the south of Riyadh.

First, Figure 6 is part of a Landsat image that shows the common and characteristic dendritic drainage pattern that tends to run toward the east, off the regional homocline of the Hanifa and Jubaila formations. Throughout the region, the dendritic drainage pattern, which on the Jubaila is defined by steep-sided canyons, abruptly changes where it meets the base of the Arab Formation. The wadis of the Arab Formation are fewer and far between and the Jubaila canyons commonly collect in master streambeds following the boundary between the Jubaila and Arab formations.

Second, Figure 7a illustrates the very different drainage pattern where the Arab Formation is cut by the Nisah, Awsat, and Bu’ayja Grabens. Our geomorphological studies and field surveys have established that the Sha’ib al-Miklaf segment is no longer an active part of the regional drainage network. The flow gradients, denoted by solid blue arrows in Figure 7a, show that the wadis on the floor of the Awsat and Nisah Grabens presently serve a catchment area, which was formerly drained by the now redundant Sha’ib al-Miklaf. Similarly, subsidence of the Bu’ayja Graben floor (previously noted in Vaslet et al., 1991, p. 48) almost led to the abortion of Wadi Al-Bu’ayja by eastward draining of its own east-west trench, but Wadi Al-Bu’ayja partly managed to straddle across the graben floor, as is evident from the false-color SPOT image of Figure 7b.

Figure 8 shows the present-day graben system and the aborted, 8 km-long section of Sha’ib al-Miklaf’s mature stream channel, which was formerly discharging intermittent wadi drift and torrential run-off waters into Wadi Hanifa. Clearly, Sha’ib al-Miklaf has become disconnected from its former flow path due to the graben tectonics, and now lies idle between the active wadis of the Awsat Graben and the Nisah Graben. Sha’ib al-Miklaf’s entrance to the Awsat Graben is presently blocked by infill of reddish eolian sands. The inset in Figure 8 projects the original course of Sha’ib al-Miklaf and the interconnected Wadi Al-Bu’ayja, before they became disconnected by the formation of the Nisah and Awsat Grabens.

Figure 9 views the entrances of Sha’ib al-Miklaf and Wadi Al-Bu’ayja as seen from the cross-cutting Awsat Graben floor. These two wadi segments were formerly connected, before subsidence of the Awsat Graben caused disruption of the original drainage pattern and ensuing evolution of stream piracy. Figures 10a and 10b view the entrances of the canyons of Wadi Al-Bu’ayja where they are cut by the Bu’ayja Graben. These two canyon segments also must have been formerly connected by a continuous, steep-sided canyon stretching across the incipient Bu’ayja Graben. However, subsidence of the Bu’ayja Graben intersected and destroyed this section of the Bu’ayja canyon, but the original drainage pattern still seems to straddle across the floor of the Bu’ayja Graben, connecting the intermittent flows of the remaining Bu’ayja canyons north and south of the graben.

Stream piracy is now advancing westward by headward cutting of eastbound drainage channels following the floor of the Bu’ayja Graben, and has nearly intercepted the original stream pattern of Wadi Al-Bu’ayja. Normal faults flanking the Bu’ayja Graben juxtapose Arab mudstone of the hanging wall against Jubaila limestone in the footwall (Figures 10b, 10c and 10d).

Field studies have previously revealed the presence of a large, shallow-marine gravel fan-delta deposit emanating from Wadi As-Sahba near Haradh (Hotzl et al., 1978). One of the gravel channels in the outer lobe of the fan arch reaches into Sabkhat Matti 250 km away from the fan center. The Sahba delta deposit is only 10 km thick at maximum, and has been dated at Late Pliocene to Early Pleistocene (Hotzl et al., 1978). This implies that the Nisah-Sahba Fault, which controls the course of Wadi As-Sahba, probably had already been initiated in the Late Pliocene.

The important conclusion inferred from these observations is that most, if not all, of the movement on the three grabens of the Central Arabian Graben System, must have occurred in Plio-Quaternary times. In fact, a Plio-Quaternary age for the Central Arabian Graben System has been previously suggested by Vaslet et al. (1991). However, the importance of Sha’ib al-Miklaf’s disconnection has not been discussed before, although the Nisah and Awsat Graben segments have been mapped in detail by Bureau de Recherches Géologiques et Minières geologists (Vaslet et al., 1991; Breton et al., 1991). Previously, Powers et al. (1966) suggested that the normal faulting that created the grabens began in the Late Cretaceous and continued into the Middle Eocene, but supporting evidence was not reported and the mechanism of graben formation remained enigmatic. Hancock and Al-Kadhi (1978) and Hancock et al. (1984) suggested that the Durma and Nisah Graben segments resulted from Cenozoic stretching above a basement fault. Conclusive evidence for their estimated Cenozoic age was not given, and the principal fault motion of Plio-Quaternary age inferred here is much younger than suggested by the earlier studies quoted here.

Some motion of the Nisah Graben prior to the Plio-Quaternary is suggested by the down-faulted Biyadh sandstone, which is exposed in the hanging wall of normal faults at either side of the Nisah Graben (Figure 7a) resting against Jubaila and Arab limestones in the footwall of the Nisah Graben (Figure 11). Surface exposures of the Biyadh Formation in the footwall, outside the Nisah Graben proper, have erosionally receded much further eastward (for example, see regional map of Figure 12 discussed later); the nearest Biyadh footwall exposures show a strike separation of at least 50 km with the Biyadh exposures in the Nisah Graben proper. This 50 km recession of regional Biyadh surface outcrops seems too large to have occurred entirely within the Plio-Quaternary, but could have been achieved in about 10 million years before the present (Ma) assuming an eastward erosional migration rate of outcrop bands of only 5 mm/year.

Folding and Transpression Along the Central Nisah-Sahba Fault in the Mughrah-Al-Kharj Transpression Zone

The central segment of the Nisah-Sahba Fault System is of crucial importance for understanding the connectivity between the pure strike-slip motion in the east (As-Sahba strike-slip fault) and normal faulting and graben formation in the west (Central Arabian Graben System). Figure 12 is a field map of our observations in the Mughrah Hills north-east of Al-Kharj, where east-west trending folds form a topographic ridge of a similar east-west trend. The east-west trending Mughrah foldbelt is nearly 60 km long and 10 to 15 km wide.

Although the map pattern of Figure 12 suggests a simple doubly-plunging synformal closure, the southern limb of the structure includes antiformal closures with interlimb angles as narrow as 90°. The dip of the beds locally become as steep as 60 degrees (Figure 13). Folds are cut along strike by steep faults, which show mostly reverse dip separation, but large strike-slip separation is also inferred. Horizontal slickensides occur in several locations. All of these faulted folds occur in the north-wall of the east-west striking Nisah-Sahba main fault, which here abruptly truncates the north-south trend of the Cretaceous units in the fault’s southern wall. The map pattern of Figure 12 documents the left-lateral strike separation across the main fault, and the east-west trending folds in the north-wall of the main fault indicate a significant component of transpression (e.g. Sanderson and Marchini, 1984).

The Mughrah-Al-Kharj Transpression Zone in the Mughrah Hills was previously interpreted as a structural trough (Powers et al., 1966). However, this region (ill-termed Mughrah “trough” by Powers et al., 1966), in fact, is comprised of an elevated synformal closure which is covered by a band of Quaternary channel gravel that roughly follows the synformal hinge zone (Vaslet et al., 1991). This drainage channel has now been uplifted together with the synformal hinge zone and the modern drainage channel of the region lies to the south of the Mughrah Hills, about 100 m lower, in the east-west trending Wadi As-Sahba.

The Landsat image of Figure 14a illustrates the Mesozoic segment of the Central Arabian Graben System. Prominent in the image are the east-west trending Wadi Nisah and Wadi As-Sahba drainage channels. Structure contours constructed from the intersection of the outcrop of the top of the Khanasir Member (Aruma Formation) with the detailed topography contours (Figure 14b), reveal that the regional slope of the Tertiary Aruma Formation, which is so prominently folded in the north-wall of the Sahba Fault, is left-laterally offset by it and slopes less than one degree (0°57′) easterly, outside both north and south of the Nisah-Sahba Fault Zone proper.

Figure 15 illustrates a synoptic model of the east-west trending Mughrah-Al-Kharj Transpression Zone in the north-wall of the central segment of the Nisah-Sahba Fault. The cause of this transpression is an obvious change in the trend of the Nisah-Sahba Fault at 24°17′N, 47°55′E, which is highlighted in Figure 14b. Two regional fold hinges nucleate from the bend in the fault trace: i.e., the Mughrah Syncline and the Turabi Anticline.

In conclusion, the Nisah-Sahba Fault System comprises a left-lateral (sinistral) strike-slip zone east of 48°E longitude (i.e., the As-Sahba Fault), a transpressional foldbelt between longitude 47° and 48° (i.e., the Mughrah-Al-Kharj Transpression Zone), and a transtensional graben system west of 47° longitude (i.e., the Central Arabian Graben System). The well-known east-west trending Nisah Fault System, 40 km south of Riyadh, has now been extended further eastward to form a continuous 250 km long fault zone referred to here as the Nisah-Sahba Fault System.

Strike-Slip Component of the Durma, Qaradah, Barrah and Majma’ah Grabens

The western end of the Nisah-Sahba Fault System shows a splaying and en-echelon arrangement (in the Nisah, Awsat, and Durma Grabens) that is typical for the termination of strike-slip faults (Figure 1). The graben rocks are locally exposed in reverse relief, particularly where resistant Tuwaiq Limestone is juxtaposed to softer country rocks. At the western termination of the Nisah Graben, the throw reduces from several hundred meters to a mere 15 m (Manivit et al., 1985). The throw in the western Awsat Graben can be as much as 200 m, but diminishes when the strike of the graben swings to the northwest. The Durma Graben and connected Qaradah Graben contain Tuwaiq Limestone that is down-faulted into the Dhruma Formation (Figures 16a and 16b). Figure 16c shows a field exposure of a typical normal fault in Dhruma Limestone.

The regional importance of the Central Arabian Graben System has been widely publicized elsewhere (Hancock et al., 1981, 1984). This interest coincided with detailed field mapping and dynamic interpretations by Bureau de Recherches Géologiques et Minières (BRGM) geologists, as published in the French literature (Breton, 1982; Breton et al., 1983). Detailed 1:100,000 geologic maps of part of the Majma’ah Graben and Nisah Graben were published by the Saudi Deputy Ministry for Mineral Resources (Breton et al., 1988, 1991) together with regional 1:250,000 scale map sheets (Vaslet et al., 1988, 1991). Figures 17a and 17b prominently feature the Majma’ah Graben, a 100 km long and 2 km wide trough with a maximum dip-slip of about 350 m in the central section, where it crosses the Tuwaiq Limestone. The central and northern sections of the Majma’ah Graben contain down-faulted outliers of Aruma and Wasia rocks (Breton et al., 1988).

The occurrence of consistently south-dipping transverse, normal faults within the grabens is typical for the northwest-southeast trending Awsat and Durma Grabens, as well as for the north-south trending Majma’ah Graben (Figure 18). The geotectonic interpretation is that these grabens were formed by north-south to north-northeast to south-southwest distension (Breton, 1982; Vaslet et al., 1988), by a mechanism which is now more commonly termed transtension (Sanderson and Marchini, 1984). The Majma’ah Graben is basically a left-lateral strike-slip zone with a minor component of transtension. Left-lateral wrench faults have also been mapped subparallel to (and within several hundred meters from) the Majma’ah Graben (Breton et al., 1988). Folds, in the cover adjacent to the graben, map with hinge lines oblique to the graben walls (Vaslet et al., 1988) in a manner consistent with decollement folds formed in the cover above a basement fault with left-lateral strike-slip (Wilcox et al., 1973; Chaimov et al., 1992). The formation of the Majma’ah Graben itself is compatible with the northward shifting of the East Arabian Block at the Aruma “bump” along the Nisah-Sahba Fault Zone at 47°55′ longitude, described in the previous section (highlighted in Figure 14b).

The northern boundary of the East Arabian Block is formed by the Wadi Al-Batin photo-lineament (Figures 19a and 19b). We, however, like previous workers, have been unable to conclusively determine the tectonic significance of the Al-Batin Lineament. Similar problems occur in establishing fault slip across Wadi Ar-Rimah, which is also a known photo-lineament extending southwest of Wadi Al-Batin.

A shallow marine gravel fan, similar to that of the Wadi As-Sahba delta fan, was deposited upon the Dibdibah plain, the delta of Wadi Ar-Rimah, into which Wadi Al-Batin is now eroded (Hotzl et al., 1978). The age of the Dibdibah gravel plain (shallow-marine) is estimated as Late Pliocene to Early Pleistocene (Hotzl et al., 1978). One interpretation is that, if any faulting created the Wadi Al-Batin lineament, it post-dates the Early Pleistocene deposition of the Dibdibah gravel fan.

An alternative interpretation, however, is that even if the fault already existed along the future Wadi Al-Batin course, this fault could not yet have evolved into a stream channel in the Late Pliocene. This is because the Late Pliocene coastline lay further landward, at the mouth of Wadi Ar-Rimah and submarine erosion of the shallowly submerged Dibdibah plain was not possible until later, when regression exposed the plain. We suggest that the Batin Lineament was already initiated in the basement extension of Wadi Ar-Rimah at Pliocene times, and Wadi Ar-Rimah itself seems a fault-controlled feature. However, only after withdrawal of the Late Pliocene Arabian Gulf by reversal of the Pliocene transgression did a fault-controlled erosion channel cut down through the Dibdibah gravel fan to form the Pleistocene Wadi Al-Batin.

If any right-lateral offset occurs across Wadi Al-Batin it may have accommodated an overall westward movement of the East Arabian Block. In that case, the Az-Zilfi-Hamudiyah Lineament, located at the eastern margin of the eolian sands of the Nafud ath-Thuwayrat (Figures 17a and 17b), may represent the surface expression of an incipient westward thrust of the East Arabian Block, so that any left-lateral offset of Jurassic strata across Wadi Ar-Rimah is not required. This suggestion of westward thrust is very speculative at this stage and needs further investigation. However, it may be noted that the Zilfi Lineament (first identified by Hancock et al., 1984) coincides with the projected boundary between the Marrat and Dhruma formations, where a gypsum horizon possibly provides a stratigraphic decollement surface.

Regional Tectonic Framework: Enter Trans-Arabian Gulf Fault

The East Arabian Block is a relatively new structural element that was first suggested by Hancock and Al-Kadhi (1978) and Hancock et al. (1981, 1984). The age of the grabens along its southern and western margins was previously poorly-constrained, but new evidence reported here indicates that clockwise rotation of the East Arabian Block took place in the Plio-Quaternary, probably by episodic strike-slip events. Our observations suggest that the Nisah-Sahba Fault Zone is a major sinistral strike-slip fault originating in the basement. Irregularities along strike of the fault have resulted in local readjustments of the overburden by transtensive graben formation and transpressive folding.

The Sahba Fault is made visible beneath the Ad-Dahna Sand Sea by virtue of the sand penetrative capability of the L-band radar. As a result of this new information, a minimum length of 250 km has been established for the Nisha-Sahba Fault Zone. The fault may further continue eastward beneath alluvial gravels for at least an additional 200 km, if extended to join up with a fault displacing Neogene strata south of Qatar (Figure 20). This would make the Nisah-Sahba Fault at least 450 km long. Figure 20 speculatively extrapolates the Nisah-Sahba Fault still further eastward across the Gulf floor from the southern end of Qatar to Qeshm Island at the coast of Iran. The Gulfward extension of the Nisah-Sahba Fault is named here the Trans-Arabian Gulf Fault. Qeshm Island is cored by a southwest-northeast trending anticline of Asmari Limestone that has rotated left-laterally with respect to the northwest-southeast main trend for the fold hinges of the Fars Platform (see also Figure 21). The rotation of Qeshm’s anticline and gradual deflection of other anticlines of the Fars Platform requires a tectonic explanation and the left-lateral Trans-Arabian Gulf Fault could account for these features.

Several seismic epicenters, onshore in the Eastern Province, plot in the vicinity of the Sahba Fault (Figure 20). Taking into account the lateral inaccuracy of 30 to 50 km for the epicenter locations, these seismic events could represent left-lateral slip on sections of the Nisah-Sahba Fault. Future study of focal plane solutions is warranted and may confirm the suggested slip mechanism. Additional evidence for subrecent faulting in Eastern Arabia is provided by the Al-Jaub Lineament, which is en-echelon to the Nisah-Sahba Fault and truncates megabarchans of the Jafurah Sand Sea (Figures 21 and 22). The pattern of megabarchans on the satellite image of Figure 22 was collected in 1973 and is identical to that on topographic maps of the late 1950’s, which in turn are based on high altitude aerial photographs of 1948. This means no noticeable megabarchan migration occurred over a period of 25 years, and their truncation is due to a combination of faulting and, possibly, stream erosion along a fault-controlled drainage channel.

The Dibba Lineament in Oman, a right-lateral fault zone, is subparallel to the left-lateral Trans-Arabian Gulf Fault; but they have mutually incompatible sense of slip (Figure 20). However, the Dibba Fault is purportedly of Cretaceous age and associated with the Turonian emplacement of the Semail Ophiolite nappe (Glennie et al., 1973; Glennie, 1996). The Dibba Fault, therefore, is geodynamically related to neither the younger Nisah-Sahba Fault nor to its en-echelon Al-Jaub Lineament. In fact, numerous dextral (or right-lateral) strike-slip faults have been mapped in the United Arab Emirates coast, and all of these are attributed to Paleozoic and Mesozoic “Oman stress,” which preceded the Cenozoic “Zagros stress,” associated with the sinistral (or left-lateral) strike-slip faults in the Gulf region (Marzouk and El Sattar, 1995).

The modern convergence of the Arabian and Eurasian Plates occurs at time-averaged rates of 1 to 2 cm/year. If this convergence were to be telescoped entirely into westerly transport and slight rotation of the East Arabian Block (Figure 23), then the locally observed maximum of 8 km left-lateral strike-slip on the Nisah-Sahba Fault may have been established in a period of only 400,000 to 200,000 years. The steep change in curvature of the wrench fault in the basement beneath the Nisah-Majma’ah Graben System is not unique. For example, the Great Kavir Fault (Figure 20) and Bitlis Fault (not illustrated here) also have a marked change in curvature along strike. The occurrence of such regionally curved strike-slip faults can be explained by basement megashears, which initiate oblique to the main foliation in the basement but the preferred shear direction laterally aligns with the main foliation as dictated by the reduced effective viscosity of anisotropic rocks (Weijermars, 1992). This mechanism of megashear-curving by anisotropic alignment already has been successfully applied to explain megashears in the Brazilian basement (Davison et al., 1995).

Implications for Oil Traps

A tectonic framework for the evolution of the variety of major structural traps in the Arabian Platform is needed to explain and understand the spatial distribution of the Arabian oil traps. Many of these traps were activated by Late Paleozoic block-faulting (McGillivray and Al-Husseini, 1992), but may well have been amplified by Neogene and Quaternary compression originating from the Zagros collision front.

Figure 20 includes the regional distribution of known, giant hydrocarbon fields hosted in north-south elongated regional anticlines of the Arabian Platform. These structural traps include the Khurais and Ghawar anticlines, both with subparallel north-south trending axes of Jurassic-Cretaceous carbonate reservoirs above amplified Hercynian, fault-bound basement pop-ups. The Qatar Peninsula also represents a morphological high coinciding with another major north-south trending, gentle anticline. The Nisah-Sahba Fault Zone effectively coincides with the southern termination of these large, gentle anticlines that are superimposed on the regional homocline within the Eastern Province of Saudi Arabia and Qatar (Weijermars, 1997a, b).

Figures 20 and 21 project two major en-echelon, sinistral strike-slip faults at the northern and southern outlines of Abu Dhabi’s giant oil fields of Zakum and Ghasha. In fact, such sinistral strike-slip faults have been reported before to control the Neogene structures of these oil fields as a results of the convergence between Arabia and Asia since the Oligocene (Marzouk and El Sattar, 1994, their figure 18). These faults are here named as the Trans-Arabian Gulf Fault (north of Zakum and Ghasha) and its en-echelon Jaub Fault.

Three potentially new oilplays can be predicted on the basis of this work. First, there is the regional Turabi Anticline (Figures 14b and 15), which seems now an obvious structural trap, but was not recognized before. Second, there are faulted anticlinals in the south limb of the Mughrah Syncline (Figures 13a and 13b), which also could be of significant structural trap closure. Third, a structural basement high seems to have popped-up in the southerly trend of the Ghawar field, between the Nisah-Sahba Fault and the en-echelon Al-Jaub Lineament (Figure 21). This prospect is also of considerable potential in view of Ghawar’s known reserves. All three afore-mentioned traps would be of Plio-Quaternary age, which probably provides sufficient time for any hydrocarbon migration into these structures. The potential of all three strategic locations mentioned here have remained largely unexplored, according to leading petroleum consultants familiar with the region.

Critical Assessment of Anhydrite Collapse Hypothesis

The present model enhances the concept of the East Arabian Block, earlier suggested by Hancock and Al-Kadhi (1978) and Hancock et al. (1981, 1984), with new evidence for significant strike-slip movement along its boundaries. The Plio-Quaternary age and regional setting of this movement, as well as the implied stress regime, all suggest a tectonic relationship with the Zagros deformation, which is compatible with the shortening, reverse faulting, and thrusting of the Fars Platform. Previously, the Central Arabian Graben System had been interpreted principally as a collapse feature due to dissolution of 140 to 180 m thick Hith anhydrite at the contact between the Arab and Sulaiy formations (Vaslet et al., 1991). According to the latter authors, the Central Arabian Graben System was possibly first downthrown by the effects of Red Sea rifting (without elaboration of any further details how this east-west rifting of the Red Sea could translate into the curved pattern of the grabens in the Arabian Platform). Vaslet et al. (1991, p. 46, 47) continue to argue that a pluvial period between 3.5 and 1.1 Ma led to dissolution of evaporites, exposed near the surface along, and in the walls of, the Nisah Graben, resulting in massive collapse structures. Such anhydrite collapse must be distinguished from the well-known sinkholes in the Sulaiy Formation due to karstification of limestone, which occurs in several locations near Al-Kharj.

Contrary to Vaslet et al.’s (1991) contention of massive anhydrite dissolution, the present author has seen only concordant stratigraphic contacts between the Arab and Sulaiy formations in the walls of Wadi Nisah, without any trace of anhydrite collapse. Similar concordant sections have been reported by Meyer et al. (1996) in their studies of the Arab Formation in the north wall of Wadi Nisah. Nonetheless, Vaslet et al. (1991) insist that widespread anhydrite collapse features exist, and in view of their extensive mapping experience in the region (Vaslet et al., 1991; Breton et al., 1991), there may well be outcrops, not visited by the present author (neither by Meyer et al., 1996), where such alleged anhydrite solution features are exposed. However, to suggest that the entire Mughrah-Al-Kharj Transpression Zone would be due to regional anhydrite dissolution and subsequent collapse, is difficult to accept, for the following reasons.

The Mughrah-Al-Kharj Transpression Zone is geodynamically consistent with left-lateral strike-slip motion. Moreover, deflection of the strike of the rock units occurs only in the north wall of the Sahba Fault (due to the asymmetric bend in the Sahba Fault trend, see Figure 14b), and no such strike-deflection occurs in its south wall. But if anhydrite dissolution were the cause of regional deformation, then one must query: Why would only the north wall and not also the south wall of Wadi Sahba collapse? Moreover, solubility of anhydrite and gypsum is several orders of magnitude less than that of halite (Handbook of Chemistry). For example, field work in a well-exposed salt dome on Axel Heiberg Island, Canada, demonstrates massive solution of halite below a relatively insoluble cap rock of anhydrite. In fact, anhydrite (CaSO4) may hydrate into gypsum (CaSO4H2O) at shallow depths of less than 800 m, or below 45° to 60° centigrade (Marsal, 1952; MacDonald, 1953). And although gypsum is more mobile than anhydrite, it is still poorly soluble and any fractures tend to fill with gypsum precipitate, which clogs up the hydrothermal system and prevents any further dissolution of gypsum (Schwerdtner and Van Kranendonk, 1984). Massive dissolution of either gypsum or anhydrite, in halite-like fashion, is deemed unlikely (J. Van Berkel, personal communication, 13 August, 1998).

In addition to this solubility problem, tectonic implications of gypsum hydration and anhydrite solution at the regional scale suggested by Vaslet et al. (1991) has never been documented before (e.g. Heard and Rubey, 1965), and, therefore, would require more supporting evidence to reach beyond its current status of speculation.


The new evidence reported here supports the view that the Arabian Platform, once thought to be geologically stable, has been locally subjected to important tectonic motion, not only in the remote geological past, but continuing into Quaternary times. The Nisah-Sahba Fault seems to separate a relatively stable southern platform from a more tectonized northern platform in which large, gentle anticlines host the major Arabian oil fields. This sinistral strike-slip fault thus separates two crustal-segments subject to different tectonic stress regimes originating from the Zagros collision front. A northern stress regime seems to have attenuated the formation of north-trending basement structures into the giant oil-reservoirs of Saudi Arabia. Typical northeast-southwest principal stress (“Zagros stress”) is inferred from borehole breakouts and borehole image-logs offshore Qatar (Jorgensen et al., 1995; Akbar and Sapru, 1994). Similar northeast-southwest directed principal compressive stress has been documented from subsurface fault studies in Kuwait (Carman, 1996). However, the Arabian Platform south of the Nisah-Sahba Fault is much less affected by such a stress regime, and contains some Paleozoic reservoirs but is devoid of any giant, oil-producing structures in the Cretaceous and Jurassic carbonate units due to the lack of structural trap amplification in the relative absence of “Zagros stress.” This prognostic conclusion, which is subject to confirmation by future exploration efforts, emphasizes the author’s expectation that the northern platform region, occupied by the East Arabian Block will, continue to yield the largest hydrocarbon reserves.


I particularly wish to express sincere thanks to Dr. Moujahed Al-Husseini for numerous suggestions for improvement. Without his expertise I would have been unable to “surgically” lift this paper out of an initially much longer manuscript. The text and presentation benefited from our numerous discussions and Hassan Al-Husseini’s support is greatly appreciated. All graphics were professionally drafted by GeoArabia’s staff. Figure 3 was first interpreted by the author and complemented with suggestions of Weston Gardner. Discussions with Jean Van Berkel provided further insight in solubility of anhydrite, based on his mapping experience at Axel Heiberg Island, Canada. I further acknowledge the Research Institute of King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia, for use of the Remote Sensing Laboratory and NASA’s Jet Propulsion Laboratory (JPL) is thanked for providing the radar data and nearly all of the panchromatic Landsat images. Carol Breed of JPL gave valuable advice and support concerning NASA’s copyright policy on satellite imagery.


Ruud Weijermars is currently working as a consultant and Board Member at the Alboran Media Group in Amsterdam. He served as an Associate Professor in the Department of Earth Sciences, King Fahd University of Petroleum and Minerals in Dhahran from 1992 until mid 1998. Ruud has previously worked as a visiting research scientist at the University of Uppsala (Sweden), University of Texas at Austin (USA), and the Technical High School of Zurich (Switzerland). He holds a PhD in Geodynamics from the University of Uppsala, and BS and MS degrees in Geology and Structural Geology from the University of Amsterdam. Ruud has studied both the basement and cover rocks of Saudi Arabia in numerous field localities. He authored over sixty research articles and published two textbooks: “Structural Geology and Map Interpretation” and “Principles of Rock Mechanics,” which have been reviewed in GeoArabia, 1997, vol. 2, no. 3, p. 340, by M.I. Al-Husseini. Ruud is a member of the Editorial Board of GeoArabia.