Geology and Oil Prospects of Somalia, East Africa
mohamed muuse kalakaan
olume: 60 (1976)
Issue: 3. (March)
First Page: 389
Last Page: 413
Title: Geology and Oil Prospects of Somalia, East Africa
Author(s): Sydney U. Barnes (2)
Abstract:
Topographically, the southern half of Somalia is comparatively flat, with elevations of less than 350 m above sea level, whereas the northern half of the country rises northward to an elevation of 2,000 m, where it drops off sharply to the Gulf of Aden. All of Somalia was a peneplain at the end of Paleozoic time. Jurassic limestone then was deposited more or less uniformly over the entire country, but post-Jurassic erosion removed a considerable thickness of these rocks, locally exposing basement. A Cretaceous sea then covered all of Somalia. Transgressions and regressions took place during Cretaceous time, and locally the Cretaceous is disconformable with the overlying Eocene. Marine Eocene sedimentary rocks cover the northern half of Somalia, but are absent in southern omalia except for a narrow strip along the coast where marine Eocene is present in the subsurface together with marine Oligocene, Miocene, and Pliocene, the latter present locally in outcrops.
Southern Somalia was subjected to epeirogenic uplift at the end of Cretaceous time, and has remained above sea level ever since except in the coastal areas. The rift faulting of middle and late Tertiary time uplifted the northern half of the country with the result that post-Eocene marine sedimentary rocks are known only along a narrow strip of the coast. Regional structural features include the Mandera-Lugh basin in the far southwest, the Bur Acaba uplift north of it, the Tertiary coastal basin, the Somali embayment which occupies most of central Somalia, the Nogal uplift north of this basin, and the middle Tertiary uplift along the northern coast.
There is no evidence of major compressive folding in Somalia, but northeast-southwest-trending anticlines of gentle dip have been mapped in Tertiary rocks in the north, and also have been identified in the subsurface in central and southern Somalia by means of refraction-seismic profiling. The structures are believed to have been caused either by rejuvenation of Jurassic fault blocks causing arching of younger sediments, or by leaching of thick underlying evaporites.
The previously mentioned structures have been primary drilling objectives and have yielded negative results. Petroleum has been generated in Somalia as indicated by an active oil seep in former British Somaliland, and by the presence of many oil and gas shows and much staining in the many wells drilled in Somalia and Ethiopia. In southern Ethiopia a gas discovery was made in March 1973, and a well drilled offshore Ethiopia in 1969 blew out, although it subsequently was abandoned.
The most promising region for oil and gas prospecting in Somalia is believed to be the Mesozoic shelf and reef area around the Somali embayment and around the Nogal uplift. Lithofacies-isopach maps are of much assistance in determining areas of limestone buildup for subsequent geophysical surveys. Secondary prospective oil and gas regions in Somalia are the coastal and offshore marine Tertiary sedimentary rocks which have had gas shows. Stratigraphic traps in clastic sedimentary rocks caused by facies changes or overlaps against the uplift regions of Somalia, and monoclinal porosity pinchouts in carbonate sedimentary rocks are other possibilities. The total sedimentary column inSomalia is in excess of 27,000 ft (8,230 m) with 14,000 ft (4,267 m) of Tertiary strata in southern Somalia 4,600 ft (1,402 m) of Cretaceous in central Somalia, and the Sinclair 1 Obbia penetrated more than 9,000 ft (2,727 m) of marine Jurassic rocks in the Somalia embayment without reaching basement. It is the opinion of the writer that commercial oil and gas deposits are present in Somalia.
Text:
PREVIOUS WORK
East Africa, including Ethiopia, British Somaliland, Italian Somaliland, and French Somaliland, first was mapped geologically as single units by Guiseppe Stefanini (1933) for the National Research Council of Italy during 1932-1933. The geologic map accompanying this report (Fig. 2) is based in part on Stefanini's map. However, Stefanini previously had published a report, "The Geology of Southern Italian Somaliland" in 1925. This was the first geologic report of Italian Somaliland. The earliest geologic report of East Africa was made in 1870 by W. T. Blanford of the Geological Survey of India who was assigned as naturalist to a British Army which marched into Ethiopia that year (Arkell, 1956). Late 19th and early 20th century geologic reports were published on German East Africa (Tanza ia), on French Somaliland, British Somaliland, and Italian Somaliland by geologists of their respective colonial powers.
After World War II, further geologic studies of the Ogadan Province (most southerly) of Ethiopia were made by a group of Italian geologists under the direction of C. I. Migliorini for the quasigovernmental agency, AGIP Mineralia. The Somaliland Oil Exploration Company, Ltd., published
End_Page 389------------------------------
a report entitled, "A Geological Reconnaissance of the Sedimentary Deposits of the Protectorate of British Somaliland" in 1954. In Kenya, Dixey published "The Jurassic Succession of North-East Kenya and the Juba River" in 1948. Oil companies have been exploring actively for oil and gas in Somalia since the 1950s, but their geologic reports are confidential.
REGIONAL SETTING
East Somalia lies along the east coast of Africa between lats. 1 ½°S and 12°N, and between longs. 41 and 51°E. It is about 1,000 mi (1,600 km) long and 150 to 250 mi (250 to 400 km) wide (Fig. 1). As of July 1, 1960, the former Italian Somaliland and the former British Somaliland were united to form the Republic of Somalia.
The southern half of Somalia is essentially a plain with elevations of less than 350 m above the sea, except for a range of hills near Obbia (Fig. 2) which rises to an elevation of more than 400 m. From the area of Galcaio, the elevation of Somalia steadily rises north and northeast to more than 2,000 m. Topographic relief also is much greater in the north than in either the central or southern parts of Somalia. In the hills paralleling the northern coast elevations rise above 2,000 m (Surud Ad 2,400 m and Bahaia 2,200 m, peaks 250 and 525 km east of Berbera) only 10 km from the sea.
The Nogal Valley in northern Somalia also exhibits comparatively great relief, with the floor of the valley at an elevation of 350 m and the northern and southern escarpments in the same area rising to more than 800 m. The Haud Plateau, just north of the Nogal Valley, rises to an elevation of 1,000 m approximately 50 km north of Gardo but decreases to 500 m farther northeast, and again rises toward the coast ranges on the north. The only two rivers inSomalia, the Juba and the Scebeli, are in the southern half of the country and originate in Ethiopia. The marked differences in topographic expression between northern and southern Somalia are believed to be the result of middle and late Tertiary faulting in the north.
STRATIGRAPHY
Most of the formational names of the Mesozoic and Eocene strata used in this report were established by Italian geologists who assigned the name of the nearest town to the type section for that formation (Fig. 1). Most of these towns are in southern and eastern Ethiopia. However, the basal sandstone which lies on basement rocks in East Africa was named by W. T. Blanford in 1870 (Arkell, 1956). The post-Eocene formational names were given by geologists employed in Somalia by petroleum exploration companies. These men also established new names for the Mesozoic as found in the wells drilled in Somalia (Fig. 3) in consequence of the facies changes from the outcrops to wells. Descriptions of rock units, unless otherwise indicated, were made by the writer, either from outcrops or well cutt ngs.
Basement
Basement outcrops are present in two major areas in Somalia--in the Bur Acaba region in the south, and in the far north near the coast (Fig. 2). There is also a small outcrop in the Nogal Valley in former British Somaliland which may indicate the crest of the Nogal uplift (Fig. 4). In both the north and the south, basement consists of metamorphic rocks intruded by granite or granodiorite. The metamorphic rocks are gneiss, schist, quartzite, and marble. In the south banded ironstones are also present (Beltrandi and Pyre, 1973). In the north, diorite and gabbro have been identified. The strike of the basement in northern Somalia is predominantly north-south, whereas in the south it ranges from northwest-southeast to north-south.
The basement region in the south, Bur Acaba, is of very ancient origin, for it was a peneplain in Early Jurassic time. One of the intrusive bodies has been dated as Early Cambrian (615 m.y.; Beltrandi and Pyre, 1973). The basement in the north may have been exposed as a result of the separation of Africa from the Arabian Peninsula (Swartz and Arden, 1960).
Volcanic Rocks
Basalt flows are present in southern and northern Somalia. Those in the north are associated with flows that covered extensive areas of Ethiopia but which have been extensively eroded leaving only remnants of the original flow. These flows have been dated as Oligocene to Miocene (Mohr, 1963), although basalt flows in the great rift valleys have been dated as Late Cretaceous into Oligocene (Swartz and Arden, 1960). The basalt flows in the south were described first by Stefanini (1925). They are present at Lugh and along the road from Lugh to Iscia Boadia. He described a series of gypsum-bearing sedimentary rocks capped by lava, volcanic ash, and tuff, which in turn are capped by a thick basalt flow which is assigned to the late Tertiary.
In the Sinclair 1 Merca, south of Mogadishu, several flows of light-gray spilitic basalt are interbedded with Paleocene and Upper Cretaceous shales. According to Swartz and Arden (1960), the volcanism at the lower (southern) end of the Red Sea was initiated in the early Miocene and continued into the Pliocene and Quaternary.
End_Page 390------------------------------
Fig. 1. Regional geographic relations of Somalia. Type-section localities are underlined.
End_Page 391------------------------------
Fig. 2. Geologic map of Somalia, East Africa. Compiled from Stefanini (1933); information from V. Fois of AGIP Mineralia; and writer's mapping and field work, 1955-1960.
End_Page 392------------------------------
Fig. 2. Continued. See caption on page 392.
End_Page 393------------------------------
Fig. 3. Stratigraphic columns of East Africa.
End_Page 394--------------------------- ---
Paleozoic
Paleozoic sedimentary rocks are absent in Somalia.
Pre-Jurassic
Although Stefanini (1925) assigned a Triassic age to basal sandstones he found near Lugh in southern Somalia, these since have been dated as early Lias or Rhaetic in age on the basis of fossil fragments (Arkell, 1956). However, pre-Jurassic rocks named the Mansa Guda Formation by Ayers (1952) in northeast Kenya reach a thickness of 2,000 ft (600 m; Thompson and Dodson, 1960). The Mansa Guda is a series of crossbedded sandstones and conglomerates between the basement and Jurassic carbonate rocks and is in unconformable contact with both units.
Jurassic
Extensive Jurassic deposits, both in outcrop and in the subsurface, are present throughout Somalia and are divided into four units (Figs. 1, 2; Table 1). The thickness of these deposits ranges from less than 500 ft (152 m) in the Amerada 1 Burhisso near the Nogal uplift to more than 9,000 ft (2,743 m) in the Sinclair 1 Obbia in the Somalia embayment. The Jurassic thickens from north and south into the embayment (Figs. 5, 6) and from west to east (Figs. 7, 8), and the facies changes in the direction of thickening are from shallow-water (shelf) to basinal sedimentary rocks.
Adigrat Formation:
This is the basal sandstone in Somalia and is present throughout East Africa, and was named by W. T. Blanford in 1870 (Arkell, 1956). In southern Somalia it is composed of varicolored quartz sands with intercalations of gypsum and dark-red shale with a maximum thickness of 130 m. In northern Somalia, the Adigrat consists of fine to coarse-grained, varicolored quartzitic, micaceous, cross-bedded, unfossiliferous sandstones, locally grading upward into sandy limestones which have been dated as Toarcian (lower Jurassic; Swartz and Arden, 1960). Locally a pebbly conglomerate is present at the base of the Adigrat. In Ethiopia, the Adigrat is described as a fine to medium-grained sandstone, as much as 60 m thick, commonly poorly cemented, but locally quartzitic, unfossiliferous (Migliorini, 19 6).
Drilling data in Somalia indicate that the Adigrat lithology is as varied in the subsurface as in outcrop. In the AGIP 1 Cotton, the Adigrat is 345 ft (105 m) thick and consists of 105 ft (32 m) of quartzite, 90 ft (27 m) of sandstone, and 150 ft (46 m) of shale. The Amerada 1 Las Anod drilled 210 ft (64 m) of Adigrat composed of 60 ft (18 m) of sandstone, 120 ft (37 m) of shale, and 30 ft (9 m) of limestone. In Ethiopia, the Sinclair XEF-1 drilled 80 ft (24 m) of Adigrat composed of 40 ft (12 m) of sandstone, 30 ft (9 m) of shale, and 10 ft (3 m) of basal conglomerate.
Hamanlei Formation:
The type section for this formation is near the village of Hamanlei in Ethiopia (Fig. 1) and there it is described as white to buff, well-bedded, mainly oolitic and fossiliferous limestone. The lower part is Callovian and the upper is Oxfordian in age; total thickness is 210 m (Migliorini, 1956).
In southern Somalia the Hamanlei is composed of light-gray fossiliferous limestone, oolitic in the lower part and containing ammonites which date it as Callovian in age (Stefanini, 1925). In northern Somalia the Hamanlei is composed of sandy and marly fossiliferous limestone, and in former British Somaliland it is a thin-bedded gray and brown rather shaly echinoidal limestone with thin bands of coral limestone (Arkell, 1956). The Jurassic in this region thickens from east to west reaching a total thickness of 1,006 m at Bihendula (Fig. 2). However, outcrops there are restricted to small, widely separated, faulted tracts (Arkell, 1956).
In the Sinclair 1 Obbia, the Hamanlei is 7,135 ft (2,175 m) thick and composed of predominantly basinal dark-gray shale and dark-gray argillaceous limestone. The same facies is present in the Sinclair 1 Gira and 1 Marai Ascia. None of these three wells reached the base of the formation and all are in the Somalia embayment. Farther north in Somalia the AGIP wells were drilled through the Hamanlei and maximum thickness penetrated was 3,350 ft (1,021 m) in the 1 Cotton, but the formation in this region is a series of light-colored fossiliferous limestones with beds of dolomite and some anhydrite (1 Darin). These AGIP wells apparently found the northern shelf of the Somalia embayment. In former British Somaliland (Fig. 1; Table 2) the Amerada wells also were drilled through the Hamanlei a d found a comparatively thin section of less than 1,000 ft (300 m) except in the 1 Buran where more than 3,000 ft (900 m) of interbedded limestone and dolomite with some gypsum was present. The Hamanlei thins toward the Nogal uplift indicating that this also was an ancient basement region or alternatively that there was extensive post-Jurassic erosion in the area, for the Jurassic formations above the Hamanlei are missing in the Amerada and AGIP wells except for 36 ft (11 m) of Uarandab in the 1 Cotton. In Ethiopia (Fig. 1; Table 3), the Sinclair wells penetrated a limestone-dolomite-anhydrite facies as much as 3,500 ft (1,067 m) thick in one
End_Page 395------------------------------
Fig. 4. Regional structures of Somalia.
End_Page 396------------------------------
well, the 1 Galadi. This region represents apparently a part of the western shelf of the Somalia embayment during Hamanlei deposition. The southern shelf region is represented by outcrops north of the Bur Acaba region.
On the basis of foraminiferal evidence, Quinqueloculina inconstans, Epistomina mosquensi, Cristellaria contralis, Lenticulina polonica, the Hamanlei includes the Callovian Stage of the Jurassic, whereas the lower limit varies because the Jurassic seas reached different parts of East Africa at different times, but it may include Toarcian Stage deposits in some areas. Fossils collected from eastern Kenya indicate an early Toarcian transgression over basement to the west (Thompson and Dodson, 1960). In the Somalia embayment the Sinclair wells were drilled into Pliensbachian age rock at least, although none of the wells reached the Adigrat Formation.
Some of the Jurassic seas invaded East Africa from the south and reached their greatest extent during Callovian time. Remnants of Callovian limestone have been identified on the Bur Acaba uplift (Beltrandi and Pyre, 1973). However, Arkell (1956) believed that the greatest marine transgression into East Africa (Ethiopia and British Somaliland) occurred in Bathonian time, and that the seas advanced from the north along the Trans-Erythraean trough across southern Arabia into Ethiopia and British Somaliland. The presence of the Somalia embayment was not confirmed until the Sinclair 1 Gira was completed in January 1957, too late to be included in Arkell's classic "Jurassic Geology of the World" (1956). Thus it would appear that the Jurassic seas invaded East Africa from both north and sout .
Uarandab Formation:
The type section for this formation is near the village of Uarandab in Ethiopia (Fig. 1), where it is composed of 180 ft (55 m) of gray, brown, and greenish gypsum-bearing shale intercalated with gray argillaceous limestone in the middle part, and similar shale in the lower 50 ft (15 m). Fossils are common with abundant Belemnites and many ammonites (Migliorini, 1956).
In southern Somalia the Uarandab Formation is represented by yellowish, marly, fossiliferous limestone, and similar light-colored marly limestone is present in central and northern Somalia. In former British Somaliland, this formation is represented by olive, gypsum-bearing shale with thin bands of brown marl and limestone (250 m), underlain by 120 m of gray calcareous mudstone and marl, 103 m of gray thinly bedded lithographic limestone, and finally 113 m of olive to gray gypsum-bearing shale (Macfadyen, 1933).
The Uarandab Formation is present in the Sinclair Obbia, Gira, and Marai Ascia wells where it is represented by basinal dark-gray shale and gray marly limestone stringers. The greatest thickness--1,765 ft (538 m)--is in the Marai Ascia. The formation is also present in most of the Sinclair wells in Ethiopia and there it is mostly shale with a maximum thickness of 522 ft (159 m) in the 1 Gumburo. The Uarandab seems to be missing in the Amerada and AGIP wells in northern Somalia except for a 36-ft (11 m) limestone in the 1 Cotton. The absence of the Upper Jurassic in these wells may be a result of post-Jurassic erosion which was very extensive in northern Somalia and former British Somaliland. The extent of erosion is evidenced at Bihendula where more than 1,000 m of Jurassic rocks crop out and only 21 km farther south at Upper Sheikh the Jurassic is completely absent. In this region the Jurassic crops out in small faulted blocks (Hedberg, 1952).
The Uarandab appears to represent a period of quiet tectonic conditions in East Africa, with a marine neritic environment except in the Somalia embayment where basinal conditions are indicated
Table 1. SOMALIA
End_Page 397------------------------------
Fig. 5. North-south cross section A-A^prime. Scales--horizontal 1:2,000,000; vertical 1:50,000. See Figure 2 for location.
Fig. 6. North-south cross section B-B^prime. Scales--horizontal 1:2,000,000; vertical 1:50,000. See Figure 2 for location.
End_Page 398------------------------------
Fig. 5. Continued.
Fig. 6. Continued.
End_Page 399------------------------------
by the presence of Epistomina stellicostata, Cristellaria nodosa, Epistomina ornata, and Pseudocyclammina sequana.
From foraminiferal evidence the Uarandab Formation includes Oxfordian and early Kimmeridgian deposits.
Gabredarre Formation:
The type section for this formation is near the village of Gabredarre in Ethiopia (Fig. 1) and there it consists of flaggy limestone, fossiliferous in the upper 131 ft (40 m), underlain by 66 ft (20 m) of thin-bedded alternating oolitic and marly limestone with gypsum-bearing shale, overlying 98 ft (30 m) of earthy ocher-colored limestone, and finally 197 ft (60 m) of gypsum. Below the gypsum is 427 ft (130 m) of finely crystalline, yellowish, partly oolitic limestone grading downward into 131 ft (40 m) of yellowish and gray marl containing flattened ammonite impressions. The total thickness of the Gabredarre is 1,345 ft (410 m) (Migliorini, 1956).
In southwest Somalia the Gabredarre consists of tan to gray, compact, very fossiliferous limestone. Similar lithology is present in central Somalia, but the formation is missing in northern Somalia.
In the Sinclair wells drilled in the Somalia embayment, the Gabredarre consists predominantly of basinal dark-gray and dark-brown shale with some gray finely crystalline limestone, with a maximum thickness in the Obbia well of 1,140 ft (347 m). Most of the Sinclair wells in Ethiopia were drilled through the Gabredarre, and a maximum thickness was present in the 1 Gumburo of 2,064 ft (629 m) where the Gabredarre is a limestone with shale members. In western former British Somaliland the Gabredarre is represented by 244 m of fine-grained, gray to brown, fossiliferous limestone, in part cherty, the uppermost 15 m being sandy and of Tithonian age (Macfadyen, 1933).
On the basis of foraminiferal evidence the Gabredarre Formation includes the late Kimmeridgian and Tithonian age deposits.
Near the end of Jurassic time the sea began to withdraw from East Africa, probably as a result of epeirogenic movement, and it left behind a large evaporite trough. In Somalia and southern Ethiopia this phase is represented by the formation known as the "Main Gypsum" which will be described in the Lower Cretaceous section. However, in southwest Somalia this phase is represented by a sedimentary series with a lower member consisting of alternating dolomitic limestone, red cross-bedded sandstone with ripple marks, and fossiliferous calcarenite, and an upper member of dolomitic limestone, siltstone with gypsum lenses, green shale, white and reddish gypsum, and at the top a massive cross-bedded sandstone.
Fig. 7. East-west cross section C-C^prime. Scales--horizontal 1:2,000,000; vertical 1:50,000. See Figure 2 for location.
End_Page 400------------------------------
The series has been named the Garba Harre Formation by Beltrandi and Pyre (1973). The total thickness is 2,200 ft (670 m), and the age is latest Jurassic and "could be extended into the Lower Cretaceous." The depositional environment of the lower member ranges from marine neritic to littoral and the upper member from lagoonal to continental (Beltrandi and Pyre, 1973).
Lower Cretaceous
In outcrop the Early Cretaceous is represented by the Main Gypsum and in the subsurface by the Cotton Formation.
Main Gypsum:
At the type section near Gabredarre, Ethiopia, this formation consists of 200 m of gypsum with calcareous, marly, and shaly intercalations. The age of the Main Gypsum at the type locality is determined by the ages of the rocks beneath and above it and is placed between Portlandian and Barremian Stages (Migliorini, 1956). However, local limestone and shaly zones have revealed an Early Cretaceous fauna consisting of Orbitolina discoidea and Choffatella decipiens. The Main Gypsum merges by intercalation with the overlying Mustahil limestone formation, but thickens from its outcrop in the Ogadan of southern Ethiopia toward the east and south at the expense of the underlying Gabredarre (Migliorini, 1956).
In south-central Somalia around the Scebeli River, the Lower Cretaceous crops out as a series of gypsum and limestone with interbedded shale. In northern Somalia the Lower Cretaceous is present in the AGIP wells but in a different facies which will be described subsequently. The Lower Cretaceous is absent in the Amerada wells, and in southern Somalia.
In Ethiopia, the Lower Cretaceous is represented by 324 ft (98 m) of gypsum in the Sinclair 1 Gumburo, 985 ft (300 m) of anhydrite in the XC-4, and 543 ft (166 m) of gypsum in the XC-3.
Cotton Formation:
This formation was established at a joint meeting at Mogadishu of the Sinclair-Somal Corporation and the AGIP Mineralia geologists in December 1958; the Cotton Formation represents the Lower Cretaceous in the subsurface. Other formational names also were established, and they will be described when appropriate.
The type section for the Cotton Formation is the depth interval between 5,233 and 7,119 ft (1,595 and 2,170 m) in AGIP 1 Cotton. It consists of a fore-reef limestone and medium-depth neritic shale. Paleontologically, the top of the formation corresponds to the base of the Orbitolina concava zone and the extinction of other species of Orbitolina such as O. lenticularis and O. discoidea which are abundant in the Cotton and Sagaleh wells. The Sinclair Gira and Obbia wells have a deeper water facies marked by Globigerina infracretacea, Globigerina washitaensis, and abundant Robulus of Early Cretaceous affinity. The Marai Ascia well had a very thin Lower Cretaceous section.
Upper Cretaceous
Upper Cretaceous rocks are present in all the Sinclair wells in Somalia, all the AGIP wells, all the Amerada wells, and most of the Sinclair wells in Ethiopia. The Upper Cretaceous consists of the Gumburo Group which is made up of four formations.
Mustahil Formation:
The type section for this formation is near the town of Mustahil, Ethiopia (Fig. 1), and consists of alternating white to yellow limestone, marly limestone and marl, light-colored marl and clay, lenticular rudist reefs, and at the top, gypsum. The Mustahil ranges in age from Barremian into Cenomanian (Migliorini, 1956). It merges by intercalation with the overlying Ferfer Gypsum and the underlying Main Gypsum; consequently its thickness is varied but is estimated to be about 200 m. Orbitolina limestone also is present in this formation. However it is not distinguishable as a formation in all the wells listed in this report because of the changing facies of the Upper Cretaceous.
Ferfer Gypsum Formation:
The type section for this formation is at the village of Ferfer in southern Somalia near the Ethiopian border (Fig. 1). It is very similar to the Main Gypsum, but no fossils have been found in it. However, it overlies and is overlain by limestones of Cenomanian age and is intercalated with both units (Migliorini, 1956). At the type locality the Ferfer is approximately 200 m thick. This formation also is not distinguishable as a formation in the wells used in this report.
Belet Uen Formation:
The type section for this formation is near the village of Belet Uen in southern Somalia (Fig. 1). It is composed of 145 m of mainly limestone. The formation consists of, in descending order, 35 m of alternating white and yellowish gypsum-bearing limestone, shale, and sandstone with some gypsum beds passing upward into the Jesomma sandstone; 25 m of similar limestone with some shale; 15 m of siliceous limestone; 20 m of alternating brown, calcareous, locally quartzitic sandstone and arenaceous limestone; and 28 m of pseudonodular limestone with abundant mollusks and echinoids with two Orbitolina zones at the top; 11 m of compact fine-grained whitish limestone in beds 0.2 to 1.0 m thick; 3.5 m of brown calcareous sandstone abundantly fossiliferous; and at the base 7.5 m of alternating gyp um and cream to
End_Page 401------------------------------
Fig. 8. East-west cross section D-D^prime. Scales--horizontal 1:2,000,000; vertical 1:50,000. See Figure 2 for location.
End_Page 402------------------------------
buff fossiliferous limestone. The lower part of the formation is late Cenomanian and the upper part is early Turonian (Migliorini, 1956). The Belet Uen thickens southeastward from its type locality, in part because of facies changes in the Ferfer Gypsum, which resulted in environment changes toward the Somalia embayment. North from the south edge of the Somalia embayment toward the central part of this basin, the four formations of the Upper Cretaceous have much the same facies and are indistinguishable as separate units. They are shown collectively as Upper Cretaceous in cross-sections of the central basin. However, the uppermost Jesomma Formation locally retains its type-section lithology even farther north, and for this reason is shown separately from the Belet Uen in cross section accompanying this report.
Jesomma Formation:
The type section for this formation is near the village of Jesomma in southern Somalia, north of Mogadishu (Fig. 1). The Jesomma is composed of red, brown, purple, and yellow; loosely cemented to quartzitic; fine to very coarse-grained sandstone with local gypsiferous beds at the base. Cross-bedding is prevalent. It is unfossiliferous but, from the ages of the rocks beneath and above, it is believed to include deposits of the Turonian and most of the Senonian. Its thickness at the type section is about 350 to 400 m (Migliorini, 1956).
The Jesomma Formation is typical only in the Sinclair wells in Ethiopia where it reaches a thickness of 1,410 ft (430 m) in the XF-5 well. In the north and in Ethiopia, the Jesomma unconformably overlies older formations. For example, at lat. 6°53^prime, long. 44°30^prime, the Jesomma lies directly on the Main Gypsum, whereas at lat. 8°7^prime, long. 43°33^prime, lat. it lies on limestone of the Gabredarre, and at lat. 8°47^prime, long. 43°03^prime, the Jesomma lies on the Hamanlei (Migliorini, 1956). Southern Ethiopia apparently was subjected to epeirogenic uplift about the time of Hamanlei deposition so that older formations were subjected to erosion and were the sedimentary source of the Jesomma. South and northeast of its type section, the Jesomma cha ges facies and merges into the Belet Uen so that all four formations of the Upper Cretaceous are indistinguishable as separate formations and are mapped as the Gumburo Group. However, in the Amerada wells in former British Somaliland, the Jesomma Formation is present but only 50 percent or less is sandstone, with the remaining section made up mostly of shale with a few limestone stringers. The Jesomma reaches a thickness of 1,060 ft (323 m) in the Amerada 1 Yaguri (Fig. 6). The fact that the Jesomma appears here in its original lithology, although with marine shale and limestone, may indicate that the Nogal uplift was involved in the upward during Cretaceous time.
The total thickness of the Upper Cretaceous in outcrop north of the Scebeli River is more than 1,000 ft (300 m) and it is overlain by Auradu limestone of Paleocene-early Eocene age (Fig. 2). From central Somalia northward almost to the Gulf of Aden, the Cretaceous is overlain by Cenozoic sedimentary rocks. Near the former British Somaliland border, not far from the northern coast, both the Lower and Upper Cretaceous have been identified in small outcrops where the lithology is predominantly limestone with a few marly members, capped by a fine quartz sandstone. Farther west into former British Somaliland this limestone facies changes to calcareous sandstone, but the thickness increases from 2,100 ft (640 m) for the limestone to 5,000 ft (1,524 m) for the sandstone (Somaliland Oil Explo ation Co., Ltd., 1954). In Ethiopia the Upper Cretaceous is composed of the four formations previously described, showing several transgressions and regressions, the presence of rudistid reefs, and evaporites. On Socotra Island off the northeast coast of Somalia, 977 ft (298 m) of Cretaceous composed of limestone with sandstone beds overlies basement rocks.
Table 2. WELLS DRILLED IN FORMER BRITISH SOMALILAND
End_Page 403------------------------------
The Upper Cretaceous is present in all the wells drilled in Somalia and in the Amerada wells drilled in former British Somaliland. In the Sinclair 1 Gira, the Upper Cretaceous is 3,595 ft (1,096 m) thick and consists principally of very porous coquinoid limestone partly dolomitized with thin interbedded green to gray foraminiferal and ostracodal shale. The Sinclair 1 Marai Ascia penetrated almost the same thickness of Upper Cretaceous rocks but here the section consists of gray Globotruncana-bearing shale and marl of a deeper water facies. The Obbia well penetrated about 2,100 ft (640 m) of Upper Cretaceous rocks composed of light-colored, fossiliferous, porous, limestone with a few beds of dark-gray shale. The Merca well farther south did not reach the base of the Upper Cretaceous bu drilled dark-gray fossiliferous shale with many spilitic basalt flows. The AGIP wells penetrated less than 2,000 ft (610 m) of Upper Cretaceous composed of light-colored partly dolomitized limestone containing many fossils including rudistids and Gumbelinas. In the Amerada wells, the section changes from a limestone-sandstone series in the east to a shale-sandstone facies westward into Ethiopia where most of the section is made up of the Jesomma sandstone (Figs. 7, 8).
In the subsurface, the Upper Cretaceous top is identified by the last Globotruncana zone or by the base of the Globorotalia shale of the Paleocene.
In outcrop, the top of the Upper Cretaceous is the top of the Jesomma sandstone which locally shows an erosional disconformity with the overlying Paleocene. Facies changes in the Upper Cretaceous appear to be more abrupt in northSomalia, probably because of the movements of the Nogal uplift, but the presence of evaporites in the southwest and south-central part of the country suggests a much more restricted depositional environment.
Paleocene
In outcrop the Paleocene is represented by the Auradu limestone, but in the subsurface the Auradu Formation changes to a deeper water facies in its lower part. Consequently, at the joint meeting of the geologists in December 1958 in Mogadishu, Sagaleh and Marai Ascia Formations were established to represent lower Paleocene.
Sagaleh Formation:
The type section for this formation is in the AGIP 1 Sagaleh in the depth interval between 3,687 and 4,218 ft (1,124 to 1,286 m), a thickness of 531 ft (162 m). It is made up of dark-gray shale speckled white with many chalky foraminiferal tests. Silty to fine sandy inclusions are present locally. It is a deep-water facies, indicated by the presence of Globorotalia velascoensis, Globorotalia crasea, and Anomalia granosa.
Marai Ascia Formation:
The type section for this formation is in the Sinclair 1 Marai Ascia and represents a transitional zone between the shale of the Sagaleh Formation and the Auradu limestone. It is present in the depth interval between 1,430 and 2,075 ft (436 and 632 m) where medium-depth Foraminifera have been identified, i.e., Cibicides, Robulus, and various arenaceous forms.
Auradu Formation:
The type section is in northern Somalia in the Nogal Valley where the Auradu attains a thickness of up to 550 m. It is a finely crystalline, compact, hard, usually tan to light-brown limestone with local thin gray shales. It contains shallow-water Foraminifera, such as Lockhartias, Sakesarias, Alveolinas, and Nummulites. The age of the Auradu limestone in outcrop includes the Paleocene and the Ypresian and part of the Cuisian Stage of lower Eocene as indicated by the presence of Lockhartia tipperi, Nummulites somaliensis, and Daviesina danieli. The Auradu Formation is used in all cross sections of this report to represent the time intervals described previously for its outcrop. In outcrop this formation is remarkably uniform over Ethiopia and Somalia, changing to a deeper water facies in the subsurface toward the east coast of Somalia. This is seen more clearly on the southeast coast in the Sinclair 1 Merca in which the lower Eocene (1,410 ft, 430 m) and the Paleocene (3,150 ft, 960 m) consist of dark-gray to brown shale with local dark-gray to brown limestone layers, and some
Table 3. WELLS DRILLED IN ETHIOPIA
End_Page 404------------------------------
light-gray to brown, fine to medium-grained, calcareous, well-cemented quartz sandstone beds, some with tar stains, locally pyritic and micaceous. In the Lach Dera well, south of the Juba River, a similar time unit is represented by a thick series of dominantly terrigenous sedimentary rocks made up of quartzose sandstone interbedded with shale and mudstone. This unit increases in thickness to 9,000 ft (2,743 m) in the Sinclair 1 Oddo Alimo (Beltrandi and Pyre, 1973). It may be concluded on the basis of the wells drilled north of the Juba River that a Paleocene-early Eocene basin lies off the east coast of Somalia.
In former British Somaliland data from the Amerada wells indicate the Auradu Formation is lithologically similar to that of its outcrop. A maximum thickness of 1,545 ft (471 m) is present in the Yaguri well and the other four Amerada wells included in this report penetrated about 1,400 ft (427 m) of Auradu. The Sinclair wells in Ethiopia seem to have spudded in the Auradu and penetrated a limestone similar to the Auradu limestone in outcrop. A maximum thickness of 1,310 ft (399 m) was found in the XE-3A well.
Taleh Formation:
The type section for this formation is near the village of Taleh in southeast former British Somaliland (Fig. 1). It has a maximum thickness of 450 m and is composed primarily of gypsum and anhydrite with shale intercalations and some limestone and marl beds with much chert. It merges by intercalation into the overlying Karkar limestone. The Taleh is of early and middle Eocene age (Migliorini, 1956).
This formation changes facies abruptly from an almost complete evaporitic facies in the Nogal Valley to 1,190 ft (363 m) of dolomite in the AGIP 1 Sagaleh only 200 km east. North from the Sagaleh well, in the AGIP 1 Cotton, the Taleh consists of 568 ft (173 m) of dolomite, whereas farther north in the Amerada 1 Las Anod the Taleh is represented by 580 ft (177 m) of dolomite. The data from the Sinclair wells in Somalia indicate varied facies in the Taleh. In the 1 Marai Ascia the Taleh is represented by 385 ft (117 m) of pink, very fine-grained, hard, calcareous, quartz sandstone, and in the Obbia well it consists of 550 ft (168 m) of gray-green and some red shale. In the 1 Gira the upper 140 ft (43 m) of Taleh was gypsum, underlain by light-gray to light-brown, cherty, fossiliferous l mestone (240 ft, 73 m), overlying multicolored shale with sandstone members (180 ft, 55 m). In the south in the 1 Merca, the Taleh consists of 570 ft (174 m) of dark-gray to dark-green, calcareous, finely micaceous shale containing glauconite and pyrite and a few thin sandstones.
Karkar Formation:
The type section for this formation is in the Nogal Valley, where it is composed of cream to light-gray, chalky limestone with paper-thin gray shale and gypsum intercalations. It has a maximum thickness of 400 m. The top of the Karkar is upper Eocene and eroded (Migliorini, 1956).
The Sinclair wells in Somalia penetrated a variety of facies in the Karkar. In the Obbia well, this formation is represented by 1,200 ft (366 m) of interbedded white to light-gray, fine to coarse-grained, rounded, friable, pyritic, quartz sandstone and gray to gray-green, sandy, glauconitic, fossiliferous shale. The 1 Merca drilled 880 ft (268 m) of dark-gray to dark-gray-green, glauconitic, finely micaceous, pyritic, fossiliferous shale. Near the mouth of the Juba River in southern Somalia more than 9,000 ft (2,743 m) of lower Tertiary clastic rocks are preserved in a funnel-shaped embayment (Beltrandi and Pyre, 1973).
Oligocene-Miocene Undifferentiated
Undifferentiated Oligocene-Miocene littoral-lagoonal sedimentary rocks have been mapped in northern Somalia by AGIP Mineralia geologists. They named these the Scusciuban Formation and the type locality is near the village of Scusciuban in northern Somalia (Fig. 1). It consists of 200 m of marly, fossiliferous, chalky limestone with Ostrea and gastropods.
In the Sinclair 1 Obbia, about 2,400 ft (732 m) of upper Tertiary deposits were penetrated, but unfortunately because of lost circulation in the top 2,100 ft (640 m) little sample was left to describe. A few samples were obtained from Globe basket cores and from the bit, however, good samples were obtained from the lower 300 ft (91 m). From this sparse information the Oligocene-Miocene may be described as a coarse-grained, rounded, friable, sandstone with green-gray sandy shale, and some white, finely crystalline, chalky, porous limestone. In the 1 Merca, this unit is composed of 5,815 ft (1,772 m) of clastic and carbonate deposits, the upper 35 ft (11 m) is of Pliocene age. This 35 ft consists of light-gray, friable, quartz sandstone. The next 1,400 ft (426 m) is predominantly white o gray, hard, fine to medium-grained, calcareous sandstone with some gray-green and brown soft clay layers in the lower 300 ft (91 m) along with one bed of cream to white, finely crystalline, fossiliferous, gypsum-bearing limestone. Underlying this is 350 ft (107 m) of red, green, and gray silty shales; 700 ft (213 m) of sandstone; and another 300 ft (91 m) of multicolored shales. This series constitutes the Miocene. The Oligocene is represented by 3,070 ft (936 m) of predominantly white to gray, some tan, granular, fossiliferous limestone, gypsum-bearing and anhydritic in the lower 1,500 ft (457 m), changing to a shale in the lowest 400 ft (122
End_Page 405------------------------------
m). The shale is gray, silty, pyritic, calcareous, fossiliferous, in part glauconitic. Farther south, in the Sinclair 1 Brava, the Oligocene consists of 3,000 ft (914 m) of marly limestone interbedded with calcareous shale and a few sandstone beds, all unconformably overlying the Lower Cretaceous. Just south of the Juba River, in the Sinclair 1 Oddo Alim the shale content of this series present in the Brava well becomes more predominant. It is dated from early to late Miocene (Beltrandi and Pyre, 1973).
Evaporites
In Somalia and in the Ogaden Province of Ethiopia, the development of gypsum and anhydrite is one of the largest in the world and is worthy of further comment. The Lower Cretaceous Main Gypsum formation in this region has an average thickness of nearly 1,000 ft (305 m) and covers an area 420 by 470 km. The Cenomanian Ferfer gypsum averages 650 ft (198 m) in thickness and covers an area of about 290 km on a side. The Eocene gypsum and anhydrite series is as thick as 1,300 ft (396 m) and is deposited over an area 880 by 410 km. The deposition of such great masses of evaporites in approximately the same region would seem to indicate a repetition of depositional conditions over a long period of time. It has been postulated by Italian geologists that the cause of the deposition of the evap rites was primarily a very hot climate associated with a very gently shelving sea floor extending a considerable distance from the coast--up to 200 mi (120 km). Paleontologic evidence supports this hypothesis and indicates that such depositional environment lasted from the Early Jurassic to the Oligocene (Migliorini, 1956).
However, if Jurassic reefs were built up along the eastern shore of Somalia, the evaporites possibly were precipitated in enclosed or nearly enclosed basins or lagoons. Possibly also, large islands were present off the shore of Somalia, for the great thicknesses of upper Tertiary sandstone in the 1 Obbia apparently came from the east, as this formation is missing in the west; such islands also could cause restricted depositional conditions, particularly in combination with reefs. The presence of Cretaceous reefs is a distinct possibility in view of the sedimentary rocks in the AGIP wells and the 1 Obbia.
Lithofacies Maps
After the drilling of a few wells the presence of a Mesozoic deep-water embayment in Somalia became apparent. Necessary then were attempts to delineate the areas of possible shelf limestone-biohermal reef buildup around the basin for subsequent geophysical survey study. The gravity meter and the seismograph seemed to have the greatest potentialities as tools to aid in the location of limestone reefs (Agnew, 1948). From past experience in the Midland basin of West Texas, which has many similarities with the structure and stratigraphy of Somalia, it was decided to construct lithofacies-isopach maps and to concentrate on the limestone-percentage maps. The limestone-percentage map is based upon the amount of limestone, dolomite, or evaporites in any formation or geologic time unit. These ata are obtained from stratigraphic well logs, electrical logs, and measured outcrop sections, and the percentages are contoured. Because of the comparatively few wells drilled in Somalia, subsurface control is poor, so that a knowledge of the regional structure of the area is necessary.
The Upper Cretaceous limestone percentage-isopach map (Fig. 9) shows a limestone buildup along a large area of Somalia adjacent to the Ethiopian border and should be studied in more detail. Another such region is north of the AGIP wells, the Sagaleh and the Cotton. The fact that some wells have been drilled in the limestone-buildup areas without finding commercial quantities of oil should not cause dismay. There are many oil fields producing from carbonate rocks in West Texas that originally were missed by only one 40-acre spacing. Because both AGIP wells reported oil shows, the region north of them appears promising.
The Hamanlei (Lower-Middle Jurassic) limestone-isopach map (Fig. 10) shows an even larger area of limestone-buildups than the Cretaceous lithofacies map. A very large area which extends into Somalia is indicated in southern Ethiopia, and nearly all of northern Somalia appears to have potential for carbonate reservoirs.
In addition to the carbonate-lithofacies maps, sandstone-percentage maps also would be useful, particularly in the Tertiary rocks where gas shows were recorded in the Sinclair wells in southern Somalia.
STRUCTURE
Within Somalia there are six major structural features (Fig. 4). In the south is a large area of basement outcrop known as Bur Acaba. This is an ancient region which probably has remained above sea level since the beginning of sedimentation in Somalia with the exception of the Callovian of Jurassic time. Mesozoic sedimentary rocks dip away from Bur Acaba in all directions, although the dips are low (less than 5°), and the area apparently was subjected to epeirogenic uplift at least twice, once at the end of Jurassic
End_Page 406------------------------------
Fig. 9. Upper Cretaceous lithofacies map.
End_Page 407------------------------------
Fig. 10. Lower and Middle Jurassic (Hamanlei) lithofacies map.
End_Page 408------------------------------
time and again at the end of the Cretaceous; in each case the sea withdrew. However, Middle Jurassic northeast-southwest-trending faults apparently broke the southern part of Bur Acaba and the down-to-the-south block was the site of later sedimentation (Beltrandi and Pyre, 1973).
In consequence of faulting, a basin developed on the south and southwest side of Bur Acaba in which about 8,000 ft (2,438 m) of Jurassic sediments accumulated; this was named the Mandera-Lugh basin by Beltrandi and Pyre (1973). The sea withdrew in the Early Cretaceous.
Occupying most of central Somalia, but recognized only in the subsurface, is the Somalia embayment, a large region of deep-water Mesozoic deposition. The axis of this feature appears to coincide with a positive gravity anomaly which trends northeast-southwest, but the Cretaceous axis seems to be farther east than the Jurassic, and the Tertiary basin farther east than the Cretaceous. Evidence for the Jurassic basin is present in the sedimentary rocks of the Sinclair 1 Obbia, 1 Gira, and the 1 Marai Ascia; for the Cretaceous basin in the Sinclair 1 Merca and 1 Marai Ascia; and for the Somali Tertiary coastal basin in the Sinclair 1 Merca, 1 Obbia, and the 1 Oddo Alim. The Tertiary basin appears to be better developed in the southeastern part of Somalia where considerable thicknesses of arly Tertiary clastic rocks are present. The axis of the Tertiary basin is offshore but parallel with the coastline of Somalia.
The Nogal uplift in northern Somalia trends northwest-southeast and extends into former British Somaliland. A small outcrop of basement rock is present near the structural crest of the uplift (Mason, 1957). The Nogal uplift appears to be a very old feature because it influenced Mesozoic deposition (Figs. 7, 8). It also was subjected to epeirogenic uplift, once at the end of Jurassic time when considerable faulting took place creating fault blocks from which great thicknesses of Jurassic deposits were eroded, and there appears to have been an upwarp in Late Cretaceous time which involved the Ogadan (southernmost province of Ethiopia). After a period of post-Cretaceous erosion, the area downwarped so that Paleocene and Eocene seas invaded northern Somalia and southeastern Ethiopia. Agai , after Eocene deposition, the Nogal uplift was upwarped, and there has been no marine sedimentation since except along narrow areas of the coast.
Probably the youngest structural feature in Somalia is the uplift region along the northern coast which locally exposes basement rocks. This feature apparently was caused by the great movement when Africa finally separated from Arabia in early Miocene time (Swartz and Arden, 1960).
Folding
By the end of Paleozoic time, Somalia was a peneplain of ancient crystalline and metamorphic rocks. The Jurassic deposits which covered this peneplain were not folded but apparently were subjected to isostatic adjustment creating highs and lows, the highs being eroded and the lows creating basins. Because of differential movement of the blocks their erosion was irregular, amounting to thousands of feet in one area and almost none in an adjacent block.
In central Somalia, the Cretaceous strata dip gently away from the Bur Acaba region as do the Jurassic, whereas in the north there is a general southerly dip of the rocks of both periods. Therefore, as far as surface sedimentary rocks are concerned, the Mesozoic strata exhibit no major compressive folding. There may be minor folding against the faults.
On the other hand, Eocene sediments in northern Somalia show anticlinal features. In the Nogal Valley the anticlines are small, measuring only a few kilometers along the long axis, but north of the Nogal Valley near the coast AGIP geologists have mapped two lines of Oligocene-Miocene anticlines paralleling the coast and 16 to 20 km long. However, the dips are low, no more than 3° toward the west and 5 to 7° toward the east. These anticlines are believed to have been caused either by rejuvenation of Jurassic fault blocks which arched the younger sediments, or by drag along a fault that parallels the coast from Bur Acaba northward. This fault, which bounds the eastern side of Bur Acaba, is believed to have a throw of more than 14,000 ft (4,267 m; Beltrandi and Pyre, 1973). The former concept is more probable because most of the Jurassic is missing in the three AGIP wells on these anticlines. The structures in the Nogal Valley are believed to be the result of leaching of the Taleh evaporites below the Karkar Formation. There is much evidence of karst topography in the Nogal Valley.
In the subsurface, seismic exploration has resulted in the discovery of several anticlinal features which have the same trend as the surface structures of northern Somalia. They have been drilled with negative results in both northern and southern Somalia. In Kenya, Cretaceous folds have been described with both northwest-southeast and southwest trends (Dixey, 1948). On the island of Madagascar and in Arabia, Cretaceous folds are present. However, no Mesozoic folds have been found in Ethiopia.
Another type of fold was found in the Sinclair 1 Merca where about 2,000 ft (608 m) of spilitic basalt was drilled, thus accounting for the closure
End_Page 409------------------------------
that had been measured by seismic means. To ascertain whether any of the other seismic anomalies in the area were of similar origin, an aerial-magnetometer survey was made.
Faulting
Faulting in Early and Late Jurassic time took place in southern Somalia adjacent to the Bur Acaba basement outcrop, and also in northern Somalia where block-faulting is evident. Early Jurassic rift faulting in Tanzania was described by Kent (1965), but Dixey (1948) previously had shown that such faulting is not present in Kenya, although Jurassic block faulting was. The Bur Acaba uplift was faulted on its south side in Early Jurassic time, creating the Lugh-Mandera basin (Beltrandi and Pyre, 1973) with the fault trending northwest. Another major fault system bounds the Bur Acaba basement on its southeast side, trending northeast near the Somalia coast almost to the tip of the Horn of Africa. The fault is down to the southeast, and its major movement took place in the Middle Jurassic, ut movement occurred again in Tertiary time, possibly as a result of sediments accumulating on the continental shelf (Fig. 5). The Jurassic faulting in northern Somalia probably was initiated in Middle Jurassic time and is developed particularly in former British Somaliland (Swartz and Arden, 1960). These faults trend north-south, and may be related to the early breaking up of Gondwanaland (Tarling, 1971).
The upwarping of southern Somalia at the end of Cretaceous time probably resulted in faulting but none has been observed, and similarly the epeirogenic movements of northern Somalia during the Cretaceous also probably caused faulting but the land is flat and covered with thorn bushes that make field examination difficult. However, the faulting observed near the northern coast of Somalia, which is associated with Cretaceous rocks, is believed to be of Tertiary age, although some Cretaceous faults may have been rejuvenated. These faults trend east-west with downthrow to the north.
By the end of early Eocene time a paar developed between Africa and Arabia which caused tensional forces in the southern Red Sea area (Swartz and Arden, 1960). Normal faults mapped in Eocene strata in northern Somalia trend northeast and east-west, but these may be of Miocene age. On the other hand, field evidence in northern former British Somaliland suggests that old basement structural trends exerted a controlling influence on younger fault trends (Somaliland Oil Exloration Co., Ltd., 1954). North-south faults have been mapped also in Tertiary strata in northern Somalia but these may be related to Jurassic faulting which probably was rejuvenated in Tertiary time. Tensional forces were at a maximum in the early Miocene in the southern Red Sea area and resulted in the development of he Aden trough. During the Pliocene, the Gulf of Aden continued to widen and opened the Strait of Bab el Mandeb between the Red Sea and the Indian Ocean (Swartz and Arden, 1960). These movements are reflected in northern Somalia by northwest-trending faults in Tertiary sedimentary rocks, especially in the Nogal Valley where there is a displacement of 500 m with downthrow on the northeast (Migliorini, 1956). Similar faulting is present in Ethiopia in the Ogadan Province. It is believed that these late Tertiary crustal movements not only rejuvenated older faults, but are also responsible for the present morphology of Somalia, with the highest elevations in the country along the northern coast, and the only two rivers located in southern Somalia.
OIL PROSPECTS
In a discussion of the oil prospects of any region, several fundamental questions assume major importance. Has oil been generated in the area? Are there suitable reservoir rocks? Are there suitable structures? To all of these, with possibly some reservations about the last, the answers are "yes" regarding Somalia.
The earliest indications of the generation of oil in East Africa are in the large oil seep in north-central former British Somaliland at approximately 10°10½^primeN lat. and 40°17^primeE long. (Fig. 2). Stan-Vac drilled three dry holes near this seep in 1959. Azzaroli, a geologist with the National Research Council of Italy, mapped bituminous strata in limestone of the Auradu Formation in northern Somalia (A. Azzaroli, personal commun., 1957). With the coming of the drilling rigs to Somalia, many more indications of oil and gas became evident. In the AGIP 1 Sagaleh, oil staining was found in a porous limestone-dolomitic series of the Jurassic. The Sinclair 1 Gira found a small oil show in the Jurassic. In the AGIP 1 Cotton, a good gas show was found in 300 ft (91 m) of retaceous limestone, and the 1 Sagaleh found oil staining in the same zone. Auradu Paleocene limestones showed oil stains in the Sagaleh, the Cotton, and the Gira wells. More than 2,600 ft (792 m) of shelf-type limestones of Oligocene age in the 1 Merca showed no hydrocarbon traces, but the Eocene sandstones in this well not only had good oil staining but contained gas under good pressure. In a drill-stem test of the depth interval from 8,869 to 8,875 ft (2,703 to 2,705 m), salt water flowed at a rate of 15 gal (std) per minute
End_Page 410------------------------------
accompanied with about 25 to 30 Mcf of gas. Bottom-hole pressure was 5,500 psi. Another flow of salt water with gas was found from 7,858 to 7,870 ft (2,395 to 2,397 m), with bottom-hole pressure of 4,900 psi on a 3/8-in. choke. The Sinclair 1 Afgoi-1, north of the Merca well, also had good gas shows. Farther south in Somalia, the Sinclair 1 Oddo Alimo drilled about 14,000 ft (4,267 m) of Tertiary sedimentary rocks. The Gira well drilled about 4,600 ft (1,402 m) of Cretaceous beds, and the Obbia well drilled more than 9,000 ft (2,743 m) of Jurassic section without reaching basement. In total, there is a sedimentary column of more than 27,000 ft (8,230 m) in Somalia.
Since this report was initiated, a gas discovery was announced by Tenneco in its Calub 1 in southern Ethiopia (Biro, 1974, p. 2057), and a well drilled by Mobil-Esso in the Red Sea off the coast of Ethiopia experienced a gas blowout at a depth of 9,874 ft. A relief well was drilled, but blowing stopped before the relief well reached its objective (Littlefield, 1970, p. 1498). Probably the hole caved. These events stimulate interest in East Africa as a possible future oil and gas province.
There is no evidence of large-scale compressive folding in Somalia like that in Arabia and Egypt. There are anticlines in northern Somalia but these are believed to have been caused either by rejuvenation of old fault blocks or drag along major faults that parallel the Somali coast. In either case, the movements probably were associated with the Miocene separation of Africa and Arabia (Swartz and Arden, 1960) and therefore with post-Mesozoic and post-early Tertiary oil and gas accumulations. Consequently, commercial accumulations of hydrocarbons must be sought in older structures and stratigraphic traps, because those of late Tertiary age appear to be barren of petroleum. There are similar anticlines in southern Somalia, although it is apparent that some of them were developed by larg igneous intrusions. Some fault-block anticlines have been identified in central Somalia, and similar structures have been oil productive in Arabia (Baker and Henson, 1952). On the other hand, some have not been productive in Arabia as the authors state: "dry holes have been drilled in geologically favorable locations." They believe that "good results may be obtained from a stratigraphic approach to oil-finding."
All, or nearly all, of the anticlinal features in Somalia have been drilled in the last 20 years with negative results, and it is the opinion of the writer that only a stratigraphic approach to oil finding in Somalia will achieve success. The lithology and structure of the Somali embayment are similar to those of the Midland basin in West Texas. The former is a Mesozoic basin, the latter a Paleozoic basin, but they have many similarities. The Midland basin produces mostly from carbonate rocks along the shelves, although there is oil production from sandstones on the shelves and in the basin from the Spraberry Sandstone. Nearly all the oil shows in Somalia have been in carbonate rocks, although there have been shows, especially gas, in the Tertiary sandstones.
Oil fields in the Midland basin in carbonate rocks are either in porosity pinchout traps (Levelland and Slaughter fields, Texas) or in anticlinal traps (Wasson field, Texas) or in biohermal and biostromal reefs (Golden Horseshoe fields, Texas). These are all good possible types of traps in Somalia. Such areas can be localized from lithofacies-isopach maps but obviously the effectiveness of such maps is in direct proportion to the amount of information available. Two such maps have been constructed for this report (Figs. 9, 10); they show the areas of maximum limestone buildup in the Late Cretaceous and Jurassic where oil accumulations of these ages should have occurred and remained despite later Tertiary diastrophism. Location of structures and reefs in these areas can be refined furt er by seismograph and/or gravimetric methods (Agnew, 1948). The distribution of trace elements such as Ni, Co, Ba, Sr, Cr, and V has been used to identify facies in carbonate rocks (Chester, 1965) and may be useful in this area. The fact that commercial quantities of oil have not yet been found in Somalia should not cause dismay. West Texas has hundreds of dry holes, but some great oil fields. Canada also was explored unsuccessfully in the west for 20 years between oil fields, not finding the Leduc reef field until 1947.
Stratigraphic traps in clastic sedimentary rock are another good possibility for Somalia. Facies changes are common and sometimes abrupt, particularly in evaporitic strata, and especially in the clastic rocks adjacent to structural highs such as the Nogal uplift. The Cretaceous in northeast Somalia is about 90 percent limestone and 10 percent sandstone, whereas in central former British Somaliland it has changed to only 10 percent limestone but 90 percent sandstone. In the AGIP 1 Sagaleh the Cretaceous is composed of mostly carbonate rocks, whereas in the eastern Ogadan of southern Ethiopia it is about one-third limestone, one-third sandstone, and one-third gypsum. There are less marked facies changes in the Jurassic, but it changes from 1,000 m of limestone in northern former British Somaliland to about 900 m of limestone and gypsum in the eastern Ogadan (Fig. 7). In the 1 Obbia the Jurassic (Hamanlei) is mostly limestone whereas in the AGIP
End_Page 411------------------------------
1 Sagaleh it is a series of limestone, dolomite, and gypsum layers. From the Somali embayment westward into Ethiopia, there is a facies change from predominantly basinal to shallow-water sediments, thus offering possibilities for pinchouts, overlaps, and porosity traps.
There are not sufficient data on the upper Tertiary rocks to provide any information on facies changes. The lower Tertiary however provides many examples of marked facies changes. The Eocene Taleh Formation is an evaporitic series in outcrop in the Nogal Valley, but before it reaches the Indian Ocean it has changed into a limestone. In the subsurface in the Merca and Obbia wells it is a shale, whereas in the Sagaleh and Cotton wells it is a dolomite, and in the Burhisso and Buran wells in former British Somaliland it is an evaporite.
There have been good gas shows in the Tertiary sandstones along the coastal areas in southeast Somalia, which is interesting in view of the gas blowout in the Red Sea off the coast of Ethiopia. The coastal Tertiary basin of southeastern Somalia offers good possibilities for commercial oil and/or gas production because there is a thick Tertiary section of 14,000 ft (4,267 m), gas shows have been reported, and the sedimentary column may thicken seaward. Offshore exploration would appear to offer the best possibilities. It must be noted that the Miocene produces oil in Iran, and the Oligocene produces oil in Iraq, although these are carbonate reservoirs. Oil production is prolific from an Eocene-Oligocene reef complex at Kirkuk, Iraq (Baker and Henson, 1952).
In conclusion, Somalia possesses all the requirements for a petroliferous province. Hydrocarbons have been generated in Jurassic, Cretaceous, and Tertiary rocks, and the sedimentary column amounts to at least 27,000 ft (8,830 m). Many porous reservoirs are known, both in carbonate and clastic rocks, and various types of traps probably are present. In view of the gas discovery in southern Ethiopia in 1973 (Biro, 1974) it is obvious that this region is indeed a petroliferous province. However, detailed subsurface lithologic studies combined with geophysical methods are essential to further success in the discovery of commercial accumulations of hydrocarbons in Somalia.
References:
Adams, J. E., 1953, Non-reef limestone reservoirs: AAPG Bull., v. 37, p. 2566-2569.
Agnew, F. J., 1948, Geophysical exploration for limestone reefs: GSI Tech. Paper 48-01; also, Geophysics, 1949, v. 14, p. 486-500.
Arkell, W. J., 1956, Jurassic geology of the world: New York, Hafner Pub. Co., 806 p.
Ayers, F. M., 1952, Geology of the Wajir-Mandera district, north-east Kenya; parts of degree-sheets 7, 8, 15, 16, 23, 24, and 31: Kenya Geol. Survey Rept. 22, 31 p.
Baker, N. E., and F. R. S. Henson, 1952, Geological conditions of oil occurrence in Middle East fields: AAPG Bull., v. 36, p. 1885-1901.
Bandy, O. L., and R. E. Arnal, 1960, Concepts of foraminiferal paleoecology: AAPG Bull., v. 44, p. 1921-1932.
Beltrandi, M. D., and A. Pyre, 1973, Geological evolution of southwest Somalia, in Bassins sedimentaires du littoral africain--Sedimentary basins of the African coasts, pt. 2, Littoral austral et oriental--South and east coast: Paris, Assoc. Serv. Geol. Afr., p. 159-178.
Biro, P., 1974, Petroleum developments in central and southern Africa: AAPG Bull., v. 58, p. 2055-2100.
Busch, D. A., 1959, Prospecting for stratigraphic traps: AAPG Bull., v. 43, p. 2829-2843.
Chester, R., 1965, Geochemical criteria for differentiating reef from non-reef facies in carbonate rocks: AAPG Bull., v. 49, p. 258-276.
Dixey, F., 1948, Geology of northern Kenya; pt. 1, Geology and morphology of northern Kenya; pt. 2, The Jurassic succession of north-east Kenya and the Juba River: Kenya Geol. Survey Rept. 15, 43 p.
Hedberg, H. D., 1952, Petroleum developments in Africa in 1951: AAPG Bull., v. 36, p. 1395-1426.
Kent, P. E., 1965, An evaporite basin in southern Tanzania, in Salt basins around Africa: London, Inst. Petroleum, p. 41-54.
Krumbein, W. C., 1948, Lithofacies maps and regional sedimentary-stratigraphic analysis: AAPG Bull., v. 32, p. 1909-1923.
Littlefield, L. D., 1970, Petroleum developments in central and southern Africa in 1969: AAPG Bull., v. 54, p. 1484-1523.
Macfadyen, W. A., 1933, The geology of British Somaliland, pt. 1: Government of Somaliland Protectorate, 87 p.
Mason, J. E., 1957, An isolated occurrence of igneous rocks at Gorei in the Las Anod district of the Somaliland Protectorate: Geol. Magazine, v. 94, p. 498-502.
Migliorini, C. I., 1956, Geology and oil prospects of the eastern Ogadan: Rome, AGIP Mineralia (English trans.), unpublished rept.
Mohr, P. A., 1963, The Ethiopian Cainozoic lavas, a preliminary study of some trends, spatial, temporal, and chemical: Addis Ababa Univ. Geophys. Observatory Bull., v. 3, p. 103-144.
Pallister, J. W., 1959, Somaliland (British), in Petroleum developments in Africa in 1958: AAPG Bull., v. 43, p. 1671.
Somaliland Oil Exploration Co., Ltd., 1954, A geological reconnaissance of the sedimentary deposits of the Protectorate of British Somaliland: London, England, Crown Agents for the Colonies, 42 p.
Stefanini, G., 1925, Sur la constitution geol. de la Somalie Italienne meridionale: 13th Internatl. Geol. Cong., Brussels, Comptes Rendus, p. 1059.
Stefanini, G., 1933, Saggio di una carta dell' Eritrea, della Somalia e dell' Ehtiopia alla scale di 1:2,000,000: Firenze, Consiglio Nag. Ricerche, Com. Geol., 179 p.
End_Page 412------------------------------
Swartz, D. H., and D. D. Arden, Jr., 1960, Geologic history of Red Sea area: AAPG Bull., v. 44, p. 1621-1637.
Tarling, D. H., 1971, Gondwanaland, paleomagnetism, and continental drift: Nature, v. 229, p. 17-21.
Thompson, A. O., and R. G. Dodson, 1960, Geology of the Bur Mayo-Tarbaj area; degree sheet 23, N.W. and S.W. quarters: Kenya Geol. Survey Rept. 47, 49 p.
Uhl, B. F., Jr., and V. Fois, 1962, Somali Republic, in Petroleum developments in Africa in 1961: AAPG Bull., v. 46, p. 1233, 1234.
End_of_Article - Last_Page 413------------
Acknowledgments:
(2) Consultant.
Most of the material for this report was obtained by the writer during the period December 1955 to April 1960, when he was employed by the Sinclair-Somal Corporation as senior geologist in charge of well logging and subsurface geology in Somalia. The writer is indebted to V. Fois, manager and chief geologist of the AGIP Mineralia in Somalia, and to Antonio Azzaroli, a geologist of the same firm, for advice and information on the geology of Miguertinia, northeasternmost province of Somalia.
Copyright 1997 American Association of Petroleum Geologists
