Tectonostratigraphic development in the eastern Lower Congo Basin, offshore Angola, West Africa

Tectonostratigraphic development in the eastern Lower Congo Basin, offshore Angola, West Africa

Marine and Petroleum Geology 18 (2001) 909±927 www.elsevier.com/locate/marpetgeo Tectonostratigraphic development in the eastern Lower Congo Basin, ...

3MB Sizes 5 Downloads 63 Views

Marine and Petroleum Geology 18 (2001) 909±927

www.elsevier.com/locate/marpetgeo

Tectonostratigraphic development in the eastern Lower Congo Basin, offshore Angola, West Africa Paul J. Valle a,b,*, John G. Gjelberg b, William Helland-Hansen a a

Department of Geology, University of Bergen, AlleÂgt 41, N-5007 Bergen, Norway b Norsk Hydro ASA, E&P, Box 7190, 5020 Bergen, Norway

Received 19 February 2001; received in revised form 2 July 2001; accepted 9 July 2001

Abstract As in all Aptian Salt Basins off western Africa, the post-rift evolution of the Lower Congo Basin offshore Angola was greatly in¯uenced by raft tectonics. We suggest that rafting in the Lower Congo Basin took place during relatively short periods (7±10 Ma), characterized by high strain rates separated by longer periods (15±35 Ma), characterized by low strain rates. The high strain rate periods are dated as: Aptian±Late Cretaceous, Late Eocene±Late Oligocene and Late Miocene±Recent. With respect to the Tertiary development we have obtained a positive correlation between sedimentary thickness and cumulative stretching, suggesting sedimentary loading as an important driving mechanism for raft tectonics. During the Tertiary two different types of depocentres developed. These are (i) broad ®rst order depocentres governed by regional subsidence and (ii) narrower elongated second order depocentres governed by growth on active faults. Within the study area the depocentres seem to have migrated from the west towards the east. q 2001 Elsevier Science Ltd. All rights reserved. Keywords: Lower Congo Basin; Salt tectonics; Sedimentary loading

1. Introduction The Lower Congo Basin is located on the West African continental margin between latitudes 58S and 8830 0 S of Angola (Fig. 1). The area is a part of the `Aptian Salt Basins' (Clifford, 1986) that extend between the Cameroon Volcanic Line to the north and the Walvis Ridge to the south (Fig. 1). Since the Cretaceous the Aptian Salt Basins have been subjected to `raft tectonics', a term ®rst introduced by Burollet (1975) in his summary of the Kwanza Basin (Fig. 1). More recently this deformation process has been described more fully by other workers in the area (e.g. Lundin, 1992; Mauduit & Brun, 1998; Vendeville & Jackson, 1992a,b). Raft tectonics is a process whereby rigid blocks of strata separate along listric faults that sole out in a ductile detachment horizon. In the Lower Congo Basin the decoupling horizon consists of Aptian salt. The open areas between the separating blocks are successively ®lled with syn-kinematic strata (Fig. 2). The Lower Congo Basin is a quickly growing hydrocarbon province where turbidity deposits are important reservoir units (Anderson, Cartwright, Drysdall, & Vivian, * Corresponding author. Address: Norsk Hydro ASA, E&P, Box 7190, 5020 Bergen, Norway. Tel.: 147-5599-5643; fax: 147-5599-6510. E-mail address: [email protected] (P.J. Valle).

2000). It has been shown that fault growth due to salt tectonics has acted in periods as the fundamental control on the ¯ow patterns of the turbidity currents (Anderson et al., 2000; Valle, Sperrevik, & Gjelberg, 2001). Therefore one of our objects for the present work has been to achieve precise time constraints for the raft tectonics and the associated tectonostratigraphic development. In the Kwanza Basin immediately to the south of the Lower Congo Basin (see Fig. 1) Duval, Cramez and Jackson (1992) recognized two distinct phases of raft tectonics (rupturing in the Late Early Cretaceous and in the Early Eocene) separated by an inactive period (Late Cretaceous±Early Tertiary). During the inactive period the Cretaceous raft-blocks were `yoked' together by Late Cretaceous and Early Tertiary sediments (Duval et al., 1992). On the other hand, Lundin (1992), in his study of the northern Kwanza Basin, described rafting as a continuous process that started in the Cretaceous and is active today. The driving mechanism of raft tectonics has also been debated. Most authors emphasize the importance of extension in the initiating of raft tectonics, and it is a common opinion that the necessary extension is governed by gravity processes on the inclined slope (Lundin, 1992; Vendeville & Jackson, 1992b). In the literature the gravity-induced deformation is attributed to two principal factors: gravity

0264-8172/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved. PII: S 0264-817 2(01)00036-8

910

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

Fig. 1. Location of the study area within the southern part of the Lower Congo Basin offshore Angola (upper left). In the lower part 2D seismic grid, wells and seismic sections cited in text are marked.

gliding and gravity spreading (Duval et al., 1992; Lundin, 1992). Duval et al. (1992) de®ne gravity gliding as: ªtranslation of fault blocks down a gentle slope driven by the downslope shear stressº and gravity spreading as ªvertical collapse and lateral spreading of a rock mass under gravityº. Shultz-Ela (2001) reviews the use of these terms, and states that if there is rigid translation down a slope then it is gravity gliding, but if there is any internal strain within the sliding sheets, then it is gravity spreading. According to Lundin (1992), it is dif®cult to determine whether gravity gliding dominated over gravity spreading or vice versa off

the Angolan margin. However, in the present work we show that gravity spreading probably was the dominant mechanism. The present study covers an area of approximately 15,500 km 2 of the Lower Congo Basin, and it is interpreted 204 2D seismic sections with a total length of about 10,000 km (Fig. 1). The post-rift package has been subdivided by means of biostratigraphic markers identi®ed in exploration wells. For the Tertiary interval the biostratigraphic markers were tied to the seismic sections in order to obtain a set of structural maps for the different time

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

911

Fig. 2. The principle of raft tectonics (Duval et al., 1992). Pre-rafts remain in mutual contact while rafts separate so far that there is no longer contact between the original footwalls and hanging walls. The basement remains undisturbed by faulting since the salt layer acts as a decoupling horizon.

surfaces. The time surfaces form the basis for isochron maps which allow `monitoring' of fault growth and depocentre distribution along the faults through time.

2. Tectonic setting The Lower Congo basin belongs to a family of sedimentary basins occurring both onshore and offshore on both sides of the Southern Atlantic Ocean. The modern South Atlantic margins started their evolution as a part of a continental rift system that developed in the southern parts of the Gondwana super-continent at the Jurassic±Cretaceous boundary (Ala & Selley, 1997; Brice, Cochran, Pardo, & Edwards 1982). According to Nuernberg and Mueller (1991), rifting started at approximately 150 Ma at the southernmost tip of the Southern Atlantic and propagated northwards, to reaching the area immediately south of the Walvis Ridge at 126.5 Ma (Fig. 1), and the Benue Trough at 118.7 Ma. The active rifting has been further divided into three separate phases: Berriasian, Hauterivian and Late Barremian±Early Aptian (Karner & Driscoll, 1999), the ®rst phase being regarded as the most important (Coward, Purdy, Ries, & Smith 1999). Subsequently southern Gondwana split into the South American and African continental plates. The margins became passive when active sea¯oor spreading started between the two continents. The post-rift period was dominated by thermal subsidence as from the Late Early Cretaceous (Ala & Selley, 1997; Brice et al., 1982). The post-rift±present interval is dominated by regional subsidence associated with deposition of a thick regressive package, especially during the Oligocene and Miocene (Ala & Selley, 1997; Brice et al., 1982). Seranne, Seguret and Fauchier (1992) subdivided the post-rift package into an

aggradational super-unit (Albian±Eocene) and progradational super-unit (Oligocene±present), separated by an Oligocene erosional event of 10±20 Ma duration. It is a common opinion among workers in the area that the progradation and erosion were governed by epiorogenic motions. Sahagian (1988) relates them to hot spot activity re¯ecting the Late Cretaceous±present mantle convection regime underlying the African lithosphere. The latter author has quanti®ed the motion by means of ancient shoreline deposits, and suggests uplift in the range 0.5±1.0 km for the continental areas east of the study area. However, the time constraint for the uplift is poor (in the range Cenomanian±Pliocene). Uplift calculations, based on maturity data from onshore exploration wells, suggest 1.0± 2.0 km of uplift of the western African continent from Miocene to present (Lunde, Aubert, Lauritzen, & Lorange, 1992). 3. Stratigraphical and structural description 3.1. Stratigraphy The sedimentary in®ll of the Aptian salt basins is traditionally divided into three megasequences (Ala & Selley, 1997; Brice et al., 1982; Teisserenc & Villemin 1990): (1) `the syn±rift megasequence' (Late Jurassic±Early Cretaceous), (2) `the transitional/early drift megasequence' (Aptian) and (3) `the drift megasequence' (Albian± Holocene). A generalized stratigraphic column for the Lower Congo Basin is shown in Fig. 3. In accordance with the general subdivision for the Aptian Salt Basins, the column is subdivided in a pre-salt unit (syn±rift megasequence), a transitional unit and a drift unit. The stratigraphic information is compiled from Abilio (1986),

912

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

Ala and Selley (1997), Brognon and Verrier (1966), Coward et al. (1999) and Franks and Nairn (1973). In addition we have had access to data from unpublished exploration wells and associated oil company reports (see Fig. 1 for location of wells). Among the available wells, the shallow water well C (Fig. 4) offers the most complete general overview of

the Lower Congo stratigraphy since it has fully penetrated all the megasequences mentioned above and terminates in basement. In well C the lowermost 40 m above the termination depth consists of basement, or more speci®c crystalline and metamorphic rocks of early±mid Proterozoic age (2000±1100 Ma) (Teisserenc & Villemin, 1990). The basement is unconformably

Fig. 3. Generalized stratigraphic column for the Lower Congo Basin. The seismic markers cited in text are marked in the right column. (Compiled from Abilio (1986), Ala and Selley (1997), Brognon and Verrier (1966) and Franks and Nairn (1973) and unpublished oil company reports).

Litho.

O1 E1

?

Deep marine channel deposits

?

Seabed

O2

?

?

re B_P

Salt

Sandstone

Shaly sandstone Salt

PERIOD GROUP

Legend Mudstone

e Iab B_ inda e P B_ Loem B_

(m)

3000

2750

2500

2250

2000

1750

1500

1250

1000

750

500

?

250

Depth Litho.

Basement

Calcareous shale Limestone

Gamma GR (GAPI) 0,0 150,0

Well C

?

?

Well D

Gamma GR (GAPI) 0,0 150,0 (m)

3750

3500

3250

3000

2750

2500

2250

2000

1750

1500

1250

1000

750

500

250

Depth Litho.

?

?

Well E

Gamma GR (GAPI) 0,0 150,0

Fig. 4. Exploration wells from the Angolan shelf. (A) and (B) are located in the deepwater area, while C±E all are located in shallower water (see Fig. 1 for location of wells).

?

Litho.

Well B

Landana

NEOGENE Molembo LA TE CRETACEOUS Lobe

Loeme Pinda L AT E CRETACEOUS Pre salt

PERIOD GROUP

NEOGENE Malembo PALEOGENE

LATE CRETACEOUS Iabe

Pinda Loeme EARLY CRETACEOUS Pre Salt ? ?

PERIOD GROUP

TERTIARY Landana Malembo L AT E CRETACEOUS Iabe

Pinda EARLY CRETACEOUS Presalt

Well A Depth (m)

3250

3000

2750

2500

2250

2000

1750

1500

1250

1000

750

500

250

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927 913

Litho.

914

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

Fig. 5. Structural domains governed by post-rift deformation. Domain 1: Cretaceous raft blocks and grabens, Domain 2: Tertiary raft blocks and grabens, Domain 3: Tertiary pre-rafts and Domain 4: Tertiary salt diapirs. Faults that were active during the Cretaceous are marked with thin solid line while the Tertiary faults affecting the E1 level are shown with thick solid line.

overlain by the Pre-salt Group that contains syn±rift (and pre-rift?) sediments. Lithologically, the Pre-salt Group is dominated by lacustrine shales interrupted by several ¯uviatile sand deposits, especially near the base. The base of the transitional unit is marked by an approximately 100 m thick sandstone that appears immediately underneath the Loeme Group. In well C the Loeme Group is circa 150 m thick and consists basically of halite occasionally interrupted by 1±5 m thick layers of silt and clay. However, due to post depositional movement, the salt package shows extreme thickness variations, from near zero thickness in the east to thousands of meters further west due to doming (see Fig. 12). The drift unit consists of the Pinda, Iabe, Landana and Malembo groups of Late Early Cretaceous (Albian)±present age (Fig. 3). In well C the Pinda Group consists of a 200 m thick bipartite package with limestone in the lower half and shale in the upper half. The limestones represent a shallow carbonate platform that developed during a relative sea level rise along the margin of west Africa during the Aptian

(Ala & Selley, 1997), while the shale in the upper half marks the establishment of more open marine conditions. The remaining 1300 m of well C (Iabe, Landana and Malembo groups) shows a very uniform lithology consisting of marine shale interrupted by minor silt layers. The upper part of the Iabe and the Landana Groups are associated with a distinct gamma ray peak. This, together with the very small thickness of the Landana Group (approximately 20 m) suggests that the Upper Cretaceous and Lower Tertiary should be regarded as a condensed section. The deepwater well A (Fig. 4) terminates within the Iabe Group, here represented by approximately 300 m of sediments. Like in well C Iabe consists of a mixture of limestone and shale, and the upper part of the Iabe together with the lower part of Malembo is also associated with a distinct gamma ray peak. However, the Landana Group is better developed in the deepwater well and has a thickness of almost 400 m. Well A is totally dominated by the Malembo Group that is more than 3000 m thick, consisting of marine shale, with several intervals of up to 25 m thick sandstone

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

packages. The sandstone packages represent debris ¯owand turbidite units (Dupont, Masse, Moron, Pourtoy, & Gerard, 2000) deposited under relatively deep marine conditions. 3.2. Structural domains The main post-rift deformation mechanism in the area is raft tectonics (Fig. 2). Following the terminology of Jackson and Talbot (1991) the term `raft' is applied on normal faulted blocks that are so widely separated that there is no longer contact between the original footwall and hanging wall after faulting. On the other hand, `pre-rafts' represent an early stage of rafting where the hanging wall is still resting on the original footwall after faulting (Fig. 2). Based on the post-rift structuring we sub-divide the study area into four domains where the names of the domains refer to the dominant structural elements within each of them. From the east towards the west they are (1) Cretaceous raft blocks/grabens, (2) Tertiary raft blocks/grabens, (3) Tertiary pre-rafts and (4) Salt diapirs (Fig. 5). 3.2.1. Domain 1: Cretaceous raft blocks and grabens In the northern part of the study area, domain 1 stretches from the coastline about 30 km towards the west. Southwards it reaches a maximum width of approximately 50 km (Fig. 5). A typical cross section is shown in Fig. 6. Above the top salt re¯ector the strata are heavily in¯uenced by listric faults that sole out above the base salt re¯ector. The youngest detectable re¯ector pre-dating the onset of the deformation seems to be the near top Albian (nt_Alb) re¯ector, that marks the upper boundary of the Pinda Group. The re¯ector is clearly rotated and downfaulted, and occasionally the nt_Alb re¯ector is so heavily downfaulted that the hanging wall is totally separated from the footwall. In such a scenario the separated fragments of the Pinda Group consti-

915

tute small raft blocks typically 1±2.5 km wide (dip section) in most parts of the platform area. The separation between the individual raft blocks varies normally from just a few hundred meters to several kilometres measured parallel to dip (see Fig. 5). In the lower part of the overlying Iabe Group, re¯ectors display divergent patterns towards the faults (Fig. 6), which is a typical criterion for syn-sedimentary faulting. By identifying the age when the re¯ectors again turn into parallelism, we observe that the syn-sedimentary faulting stopped at different times within the Upper Cretaceous, from the intra Coniacian in the vicinity of well E (Fig. 7(a)), to intra Maastrictian in the vicinity of well D (Fig. 7(b)). Above the diverging pattern the re¯ectors turn into parallelism, suggesting that (varying) parts of the Upper Cretaceous together with the entire Tertiary sequence is undisturbed by syn-sedimetary faulting within domain 1 (see Fig. 6). In the seismic sections several unconformities can be identi®ed throughout the Cretaceous and Tertiary sequence within domain 1. Cretaceous movement of raft blocks was associated with block rotation, and some of the elevated crests were eroded. For example, in Fig. 7(a) where Turonian and Coniacian re¯ectors (marked by arrows) terminate against a strong and continuous re¯ector. The strong re¯ector is subsequently onlaped by Maastrictian re¯ectors. In well E (Fig. 4) the onlap surface correlates with a distinct rise in the gamma ray readings, indicating slow sedimentation and stagnant conditions during the Latest Cretaceous. The Cretaceous±Tertiary transition is characterized by multiple erosional events. Between CDP2950±3050 in Fig. 8, re¯ectors from the near top Cretaceous are onlapped by re¯ectors representing Paleocene sediments, and appear as an erosive contact. Further to the west in Fig. 8 (CDP2650±2450) this erosive contact itself is truncated and onlapped by mid-Oligocene re¯ectors (O1). The latter erosional surface is the E1 re¯ector (see Fig. 3). Along strike

Fig. 6. Seismic section from structural domain 1 (see Fig. 1 for map location). Thin salt forms minor pillows. The listric faults sole out within the salt package and the syn-sedimentary faulting ceases within the Upper Cretaceous (see Fig. 7 for details. See Fig. 3 for identi®cation of labeled horizons.

916

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

Fig. 7. Re¯ector patterns within the Cretaceous and Lower Tertiary, structural domain 1. In the vicinity of well E (a) we record divergent pattern from the nt_Alb re¯ector to intra Coniacian times, while the divergent pattern continues well into the Maastrichtian in the vicinity of well D (b). See Fig. 3 for identi®cation of labelled horizons.

Fig. 8. Erosional surfaces within the Upper Cretaceous/Lower Tertiary (see Fig. 1 for map location). Between CDP2950±3050, re¯ectors from the near top Cretaceous (nt_Cret.) are onlapped by re¯ectors representing Paleocene sediments, and appear as an erosive contact. Further to the west (CDP2650±2450) this erosive contact itself is truncated and onlapped by mid-Oligocene re¯ectors (O1).

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

917

Fig. 9. Strike section showing intra-Tertiary erosional surfaces (See Fig. 1 for map location). The E1 and M3 surfaces both show considerable relief (100± 500 m).

sections the E1 surface often appears with relief in the range of 100±500 m (Fig. 9). The youngest unconformity within the scope of this study is the M3 re¯ector (Fig. 3), which also displays considerable relief in strike sections (see Fig. 9). To the east the seabed truncates several of the mentioned re¯ectors, including the M3 re¯ector (Figs. 6 and 8). This indicates Late Miocene (or younger) uplift and tilting of the coastline. 3.2.2. Domain 2: Tertiary raft blocks and grabens In plan view, domain 2 is wedge-shaped. Its width increases from zero in the south east, to about 100 km in the north, where it covers almost the entire width of the

study area (see Fig. 5). The well coverage within this domain is rather sparse, and the interpretation is based on correlation of biostratigraphic markers taken from available deep-water wells in the western part of the study area (Fig. 4). Fig. 10 is an E±W striking seismic section that is tied to well B. The section intersects two fault systems. The easternmost system is a full graben (CDP4300±5200) between west- and east facing faults, whereas the westward facing fault in the west (around CDP3500) delineates a halfgraben. In both systems, faults are strongly listric and sole out at top salt horizon. From Fig. 10 it is evident that the different faults have been active at different time intervals, which are expressed as different intervals for growth against

Fig. 10. Cross section from structural domain 2 (see Fig. 1 for map location). The section is tied to well B and displays a half graben in the west (CDP3100± 3500) at E1 and O1 level and a full graben further east (CDP4300±5200) that is active until post M2 level. Around 5.0 s twt and between CDP5200±5400 remnants of Cretaceous raft blocks are visible. See Fig. 3 for identi®cation of labeled horizons.

918

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

the faults. In addition the faults terminate at different levels in the stratigraphy (early post O1 and early post M2 for the western and eastern faults, respectively). In the footwall of both the western and eastern growth faults are remnants of the deformed carbonate platform (the Pinda Group). This is demonstrated by the rotated nt_Alb re¯ectors and the overlying wedge-shaped package deposited during the Upper Cretaceous, as described for domain 1. The wedge-shaped re¯ector package terminates against the relatively ¯at near top Cretaceous re¯ector, while the re¯ectors between the near top Cretaceous and the E1 re¯ector are parallel. With respect to termination of faults the westernmost fault dies out near the base of the O1±O2 package, while the faults in the easternmost fault system reach well into the M2±M3 package. The package between the E1 and O1 re¯ector expands towards the westernmost fault, while the corresponding package shows even thickness when approaching the easternmost fault. The thickness increase against the easternmost fault takes place within the package enveloped by the O1 and O2 re¯ectors. The packages enveloped by the re¯ectors O2±M2 generally show even thicknesses despite a minor thickness increase in the vicinity of the eastwards facing antitetic fault (CDP4400±4500). This antithetic fault represents a very common accommodation structure opposing major parts of the Tertiary raft grabens within the study area. The M3±P1 package together with the P1Seabed package are not affected by the two described

fault systems, however both the packages shows thickness increase from the west towards the east. The described fault terminations plus the growth pattern both indicate a deformation history where the deformation starts in the west and progresses eastwards trough time. 3.2.3. Domain 3: Tertiary pre-rafts In plan view the pre-raft domain is wedge-shaped and widens to about 45 km in the south (Fig. 5). In the northwestern direction the belt of pre-rafts gradually turns into raft blocks and raft grabens. In the south the pre-raft domain merges directly into the salt diapir domain in a westerly direction. The seismic section in Fig. 11 cross cuts the pre-raft domain and proceeds westwards through the southern tip of the southernmost graben in domain 2. The deformed remnants of the Early Cretaceous carbonate platform can also be observed within the pre-raft blocks. The faults in Fig. 11 generally show a listric shape and they sole out near the base salt re¯ector. The faults generally offset all mapped re¯ectors. Some of the faults rise almost to the seabed, and sometimes offset it (see Fig. 11). The thickness of the individual layers between the E1 and M3 re¯ectors is more or less constant, even in the vicinity of the faults. However, the package above the M3 re¯ector shows a marked thickness increase and assumes a wedge shape. This is especially evident near the easternmost parts of the pre-raft domain. The faults within the pre-raft domain

Fig. 11. Cross section from structural domain 3 (see Fig. 1 for map location). The Tertiary sub-packages generally show even thickness except for the M3seabed package that shows marked thickness increase in the vicinity of the eastern faults (CDP2000±2300). Cretaceous raft blocks are visible in the lower part of the section. See Fig. 3 for identi®cation of labeled horizons.

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

919

Fig. 12. Cross section from structural domain 4 (see Fig. 1 for map location), which is dominated by a near 3.5 km high salt diapir (CDP2400±2600). The seismic re¯ectors are generally parallel except up to 5 km from the diapir where the re¯ector pattern is highly divergent. See Fig. 3 for identi®cation of labeled horizons.

generally strike NE±SW (see Fig. 5), parallel to faults in the southern part of the adjacent raft- block and graben domain. The overall fault system, running through both the pre-raft and raft- block and graben domain, thus describes a curved pattern. 3.2.4. Domain 4: salt diapirs Domain 4 lies immediately west of the pre-raft domain in the south, but borders on to domain 2 further north (see Fig. 5). The dominant salt structures within the study area are located in the north-eastern part of Domain 4 (see Fig. 5) in the form of two major salt diapirs. An E±W pro®le running through the southern diapir is presented in Fig. 12. The diapirs rise from the base salt level and reach a height of more than 3000 m. In plan view the diapirs strike NW±SE (see Fig. 5), and appear as slightly elongate features. In 3D the diapirs assume a conical shape, and the diameter decreases from around 10 km measured at the E1 level to approximately 3 km measured at the M3 level. The diapir displayed in Fig. 12 is slightly asymmetric (westwards leaning), and altogether the cross section has an asymmetric appearance. The base salt re¯ector is relatively strong, and displays a characteristic velocity pull-up underneath the massive salt body. The re¯ectors from E1 to M3 all terminate against the diapir, while the P1 re¯ector is continuous and drapes the structure. When approaching the diapir from the west the different re¯ectors are relatively ¯at lying and parallel. However, within the last 5 km the different packages are deformed and assume a synclinal

geometry. Except for the package between the t_Salt and the E1 re¯ector the different packages show a slight thickness increase when approaching the diapir. This is especially evident for the package between the O2 and M1 re¯ectors and between the M2 and M3 re¯ectors. To the east of the diapir the synclinal geometry is more weakly developed, and the packages between the O2 and M3 consist of nearly horizontal parallel re¯ectors. Graben structures are observed both in the northern and southern extension of the major diapirs (see Fig. 5), and Fig. 13 represents an E±W striking cross section running through the northern of these (see Fig. 1 for map location of line). In addition to the graben structure (CDP4400±5200), a relatively narrow salt diapir that stretches more than 2500 m above the base salt re¯ector can be seen at CDP6500. When approaching the salt structure from the west the re¯ectors are relatively ¯at lying and parallel until they terminate against the salt structure. In the area between the salt structure and the graben the re¯ector pattern is more complicated. The E1±Ox re¯ectors are folded monoclinally in the vicinity of the graben and bend slightly upwards when approaching the salt structure. The packages between these re¯ectors assume a wedge shape that increase in thickness towards the west. The Ox±M3 re¯ectors display a synclinal geometry when approaching the graben structure and bend sharply upwards in the vicinity of the salt diapir. The packages between the latter re¯ectors also show a wedge shape, but the thickness increases towards the east.

920

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

Fig. 13. Cross section (see Fig. 1 for map location) through graben (CDP4400±5200) in the northern part of structural domain 4. Between the narrow diapir in the east (CDP6300±6400) and the west facing listric fault (CDP4300±4600) the E1±O1 package has diverging re¯ectors towards the west, while O1-recent package diverges towards the east. See Fig. 3 for identi®cation of labeled horizons.

4. Tectonostratigraphy In Section 4.1 we synthesise the post-rift tectonostratigraphic development in the light of the presented structural and stratigraphical data. 4.1. Cretaceous and Early Paleogene 4.1.1. Late Aptian±Albian Above the evaporites (deposited in Middle and Late Aptian) a sallow marine carbonate platform was deposited (the lower part of the Pinda Group). At its base, the Pinda Group often contains minor evaporite deposits (e.g. in well C, Fig. 4), and these deposits might be regarded as transitional from the massive evaporites (beneath) to the massive carbonates (above). Locally the massive carbonates in the Lower Albian are replaced by ¯uvial sandstone (well E, Fig. 4) which testi®es to the shallow water conditions that generally prevailed at this time. Normal marine conditions were established during the Late Aptian. This is re¯ected in well C (Fig. 4), where the upper Pinda Group consists of oolitic limestone that grades into sandy mudstone and packstone near the top. The deformation governed by salt tectonics started around the Late Albian, as demonstrated by means of the down faulted nt_Alb re¯ector (96 Ma) that is registered in domains 1, 2 and 3 (Figs. 6, 10 and 11). 4.1.2. Late Cretaceous±Paleogene The salt-induced deformation persisted into the Late Cretaceous. The lower part of the Iabe Group was deposited in a syn-tectonic setting as can be demonstrated in the seismic sections by means of wedge-shaped re¯ector packages

in domains 1, 2 and 3 (see Figs. 6, 10 and 11). Relative sea level generally continued to rise throughout this period and the shallow water wells (wells C±E Fig. 4) show ®ning and deepening upward. The sediments consist of siltstone and claystone interbedded with limestone near the base. Limestone occurs generally in the lower part of the package only (Fig. 4). The upper part of the Iabe Group, above the described Late Cretaceous/Early Tertiary unconformity (Fig. 6), consists mainly of black and brown shale interbedded with carbonate-cemented layers less than 2 m thick. The thin carbonate-cemented layers, the relatively high content of organic material in the shale (high gamma ray readings) together with the parallel re¯ector pattern seen in the seismic sections (Fig. 6) indicate stagnant conditions and little tectonic activity in the Latest Cretaceous and Early Tertiary. The conditions described for the Latest Cretaceous seem to have persisted into the Lower Tertiary (Paleogene) since this part of the sequence is also represented by parallel re¯ector patterns on seismic sections in all the structural domains (Figs. 6, 10±12). However, according to the unconformity of Early Tertiary age (see Fig. 7), parts of the Lower Tertiary might be missing due to erosion or non-deposition. Lithologically the Paleogene deposits (Landana Group) look quite similar to the deposits from the Upper Cretaceous, and consist basically of marine shales. The Landana Group displays marked thickness variations in the shallow water wells (from less than 20 m in well C, to more than 200 m in well D, see Fig. 4). Much of this thickness variation is probably due to erosion since the top of the Landana Group, represented by the E1 re¯ector, is a clear erosional surface (see Figs. 8 and 9).

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

4.2. Late paleogene and Neogene 4.2.1. Late Eocene±Early Oligocene Lithologically this interval consists predominantly of non-calcareous claystone and siltstone with minor intercalations of calcareous claystone and sandstone (see Fig. 4). In seismic sections the E1 and O1 re¯ectors delimit the Late Eocene±Early Oligocene interval, and the E1 re¯ector was above described as a regional unconformity (see Fig. 9). Generally the package consists of parallel re¯ectors within all the structural domains (see Figs. 6 and 10±12), but mildly diverging patterns are observed in the westernmost parts of domain 2 and within domain 4 (see Figs. 10 and 13). In the eastern part of domain 1, re¯ectors from the lowest part of the package onlap the E1 re¯ector (Fig. 8). The `nondiverging' character of the E1±O1 re¯ector package is also evident from the E1±O1 isochron map (Fig. 14(a)). Here the package appears as a blanket of fairly even thickness over large areas, and the maximum thickness (250±500 m twt) is observed in domains 2, 3 and 4. Hence, it is not possible to obtain any clear candidate for a depocentre within the study area for this time interval. 4.2.2. Late Oligocene In seismic sections intervals representing the Late Oligocene package is bounded by the O1 and O2 re¯ectors, and is a package with highly diversi®ed re¯ection pattern. Within domains 1 and 3 the re¯ectors are fairly parallel, whereas they are highly divergent around the large diapirs and along the grabens in the extension of the diapirs in domain 4, and also in the vicinity of the faults in domain 2 (see Figs. 6 and 10±13). The isochron map representing the O1±O2 package (Fig. 14(b)) shows considerable differences in time±thickness. The time±thickness is moderate in domains 1 and and 3 and larger in domains 2 and 4. We observe substantial thickness increase in the neighbourhood of the diapirs and along the grabens in domain 4, and also along the faults in domain 2. This bears witness to considerable tectonic activity in the form of salt withdrawal and diapirism and also growth faulting in the Late Oligocene. Corresponding to the observed increase in tectonic activity, we note thicker and more frequent turbidite and debris ¯ow deposits (see Fig. 4). A plausible source for this increased sediment in¯ux could be gravity ¯ows shed from a prograding delta governed by the Congo River north east of the study area (Fig. 1). This river system probably was an important drainage pathway also in the Oligocene (Amaral & Seyve 2000). 4.2.3. Early and Middle Miocene The re¯ectors that represent the Early and Middle Miocene package (enveloped by the O2 and M3 re¯ectors) display a much higher degree of parallelism compared to the underlying Late Oligocene package. In domains 1 and 3 the package contains parallel re¯ectors, whereas diverging patterns are restricted to the grabens in the extension of the diapirs are observed in domain 4 (Fig. 13). In domain

921

2 re¯ectors are fairly parallel in the west but show a weak diverging pattern towards the east (Fig. 10). In the O2±M3 isochron map (Fig. 14(c)) the study area still appears bipartite with great time-thickness for domains 2 and 4 relative to domains 1 and 3. However the contrast is smaller than seen in the O1±O2 time-thickness map (see Fig. 14(b)). It is still possible to de®ne a depocentre within the study area, which is now basically restricted to the eastern part of domain 2 associated with growth on the north-eastern faults. However, small isolated depocentres in the neighbourhood of the diapirs in domain 4 and areas in the extension of the diapirs in domain 4 were also active in the Miocene (see Fig. 14(c)). 4.2.4. Late Miocene±present Within this package, bounded by the M3 and the seabed re¯ector, the re¯ectors are relatively parallel in domains 4, 1 and in the western part of domain 2. The E3 re¯ector was above described as a distinct unconformity (see Fig. 9). In the eastern part of domain 2 and 3 the re¯ectors assume wedge-shaped patterns (see Figs. 10 and 11). In the M3present isochron map (Fig. 14(d)) the greatest thicknesses are found along the faults in the north-eastern part of domain 2, in the western part of domain 1 and along the faults in the eastern part of domain 3. In addition extensive growth can be identi®ed along the faults stretching southwards into domain 3. 5. Discussion 5.1. Spatial and temporal development of rafting and depocentres From the foregoing description, we conclude that post-rift deformation in the Lower Congo Basin has broadly evolved through three periods of large strain rate, separated by periods of very low strain rate. This development, as observed in domains 1 and 2, is summarized in Fig. 15. In principle this is the same development as described by Duval et al. (1992), who reported two separate periods of high strain rate for the Kwanza Basin further south. However, we have achieved a ®ner time resolution for the Tertiary±present interval. In the Lower Congo Basin the ®rst `high strain rate phase' began within the Aptian (96 Ma), i.e. somewhat later than reported from the Kwanza Basin (110 Ma). In the Lower Congo Basin the Cretaceous deformation was accompanied by the opening of 1±2 km wide grabens that were contemporaneously ®lled by clastic sediments, forming restricted NNE±SSW striking depocentres. The development is quite similar to what has been described for the Kwanza Basin (Duval et al., 1992). In the Lower Congo Basin, the Cretaceous `high strain rate period' ends at varying stages within the Upper Cretaceous, and is followed by an inactive/low strain rate period lasting from the Upper Cretaceous until the Early/Late Oligocene

Fig. 14. Isochron maps for different packages within the Tertiary displaying depocentres in the form of thickness anomalies, where the thickness refers to two way travel time (twt). (a): E1±O1 (approximately 36±28 Ma), (b): O1±O2 (approximately 28±21 Ma), (c): O2±M3 (approximately 21±10 Ma), (d): M3-recent (approximately 10 Ma-present). Index map showing the different structural domains in upper right.

922 P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

923

Fig. 15. Schematic development of the post-rift development in structural domains 1 and 2 in the Lower Congo Basin. (a): Late Albian, (b): Late Albian±Late Cretaceous. (c): Late Cretaceous±Late Eocene. (d): Late Eocene±Recent.

transition (marked by the <28 Ma O1 re¯ector). Thereafter a second time interval of high strain rate began in the Lower Congo Basin, and persisted until the earliest Miocene marked by the O2 re¯ector (<21 Ma). The deformation was accompanied by frequent deposition of turbidites in a well-de®ned and thick depocentre, especially in the northwestern part of the study area (see Fig. 14(b)). Again, the high strain period appears to have started later in the Lower Congo Basin than in the Kwanza Basin, where it is reported to have begun at the Paleocene±Eocene transition (55 Ma)

(Duval et al., 1992). After the second interval of high strain rate in the Lower Congo Basin (28±21 Ma) a period of markedly reduced strain rate and sediment accumulation followed, lasting until the Middle Miocene (see Fig. 14(c)). This low strain rate period was followed by a third interval of high strain rate (between the M3 time line and present) documented by growth faulting in domain 3 (Fig. 14(d)). When comparing the two Tertiary high strain rate intervals in the Lower Congo Basin (Figs. 14(b) and (d)) it is obvious that the depocentre migrated from the west

924

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

towards the east during the Late Paleogene±Neogene. This is in accordance with observations from the Kwanza Basin (Duval et al., 1992) but opposite to what is reported from the northern Kwanza Basin by Lundin (1992). However, the latter author admits that his results might be a consequence of insuf®cient sampling of strata. 5.2. Driving mechanism Introductorily gravity gliding and gravity spreading was put forward as the two possible driving mechanisms for raft tectonics. In the following we will argue that gravity spreading, as de®ned by Shultz-Ela (2001), was the main driving mechanism of Tertiary salt tectonics seen in the Lower Congo Basin, and that it resulted from differential loading of the margin. We are aware that any discussion concerning the relative in¯uence of sedimentary loading versus tectonic processes, easily becomes a question of `the chicken and the egg' as long as we rely on sediment rate and growth along faults to monitor tectonic development. However, in the following we will reason that strong sedimentation and differential loading impacted salt movement and contributed signi®cantly to post-rift basin evolution. As mentioned above, there seems to be a good correlation between the beginning of both high strain rate periods within the Tertiary and the formation of unconformity surfaces E1 and M3. The ®rst of these is not completely obvious, since there is a time gap of approximately 8 Ma between the formation of E1 (36 Ma) and the onset of the ®rst clear salt movement in domain 4 at post O1 times (post 28 Ma) (see Fig. 14(a) and (b)). Both the unconformities are associated with onlap patterns from in the succeeding sediment packages (see Fig. 9). The formation of the E1 surface is associated with substantial uplift of the African continent during the Late Cretaceous±Tertiary. For the present study area the uplift was estimated to be 0.5±1.0 km by Sahagian (1988) and 1.0±2.0 km by Lunde et al. (1992). Erosion associated with this continental uplift was most likely the reason for the considerable build up of sediments in the basins off West Africa during the Late Paleogene/Neogene. The mentioned time correlation with the beginning of the high strain rate periods thus indicates that increased loading is an important driving mechanism of the raft tectonics. Analogue experiments performed to simulate raft tectonics also identi®ed the importance of loading. Mauduit, Guerin, Brun and Lecanu (1997) state that ªsyn-kinematic sedimentation rates exert control on both displacement rate and extensional patternsº, and Ge, Jackson and Vendeville (1997) maintain that their experiments ªillustrate the importance of (sedimentary) progradation as a trigger for salt tectonics and the formation of allochthonous sheets.º On the other hand the uplift most probably, in addition to encourage erosion and increased sediment input, was also responsible for a basinward tilting of the continental margin, that in turn promoted gravity gliding as suggested by Lunde

et al. (1992). A way to discriminate between these two driving mechanisms (gravity spreading and gravity gliding, respectively), would be to compare the stretching in different parts of the study area. If sedimentary loading had a crucial in¯uence on salt tectonics, one should expect a positive correlation between the sediment thickness and stretching factor. In Fig. 16 we present an isochron map representing the sediment thickness between E1 and seabed (Fig. 16(a)) and a depth map for the base salt horizon (Fig. 16(b)). In both Fig. 16(a) and (b) fault polygons drawn at the E1 level plus three ENE±WSW striking pro®le lines are shown. From Fig. 16(a) it is obvious that the pile of sediments become markedly thicker from the south towards the north. From south to north the calculated b factors are 1.09, 1.33 and 1.28 in the ESE±WNW striking pro®les, and this suggests a positive correlation between sediment thickness and stretching. However, we still need information concerning the gradient of the detachment plane. If an equivalent positive correlation between the basinward gradient of this plane and the stretching can be proven, the foregoing reasoning for loading as a principle driving mechanism would be inconclusive. Therefore, we have calculated the dip of the base salt re¯ector for the ESE±WNW striking pro®les (Fig. 16(b)). The dip angles are 2.15, 1.38 and 1.238 denoted the south towards the north. Hence, in our study area there exist a negative correlation between the base salt gradient and the amount of stretching. We thus conclude that increased loading is one of the principal driving mechanisms for the raft tectonics in the Lower Congo Basin, and that the increased loading probably is at least as important as the gradient of the detachment plane. This is emphasized by the positive correlation between stretching and sediment thickness and lack of correlation between stretching and gradient of the detachment plane. This is in accordance with the results of Elliot (1976). In his study of deformation mechanisms of thrust sheets, he concluded that it is the slope of the top surface of the thrust sheet that acts as the primary control on emplacement, and that the top surface slope probably outweighs the in¯uence of the slope of the detachment surface. As mentioned above, the ancient Congo Delta (see Fig. 1) was probably the main sediment source for our study area, an assumption that is supported by the increasing thickness of the Tertiary package towards the north (Fig. 16(a)). The difference in relative thickness of sediment overburden probably caused differences in sea¯oor topography, which in turn acted as driving mechanism for the Tertiary rafting. This is then probably the main reason for the observed north to south decrease in Tertiary horizontal displacement (Fig. 16(a)). The latter is in accordance with the work of Anderson et al. (2000) who studied data from Block 4, somewhat to the east of the present study area. Among other things these authors conclude that ªThe late Miocene extension direction was approximately parallel to the progradation direction of the Congo fan and was driven

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

925

Fig. 16. (a) Thickness of E1-Recent package (thickness in ms twt) and (b) depth of the base salt re¯ector (m) compared to the cumulative Tertiary stretching represented by width of fault polygons. Black lines indicate the positions of pro®les where the stretching-factors (b ) and gradient (a ) are calculated. The maps indicate a positive correlation between sedimentary thickness and cumulative stretching.

by gravity loading as pro-delta and shelf sediments prograded into the area.º As pointed out earlier, there exists a certain time gap between the time of the formation of the E1 surface (36 Ma) and the earliest recorded fault growth (post 28 Ma). If this time gap is real, and not caused by erroneous seismic correlation or biostratigraphic dating, it might signify that substantial amounts of sediments by-passed the study area, and that the depocentre was located further to the west during the Late Paleocene±Early Oligocene. This scenario would be in accordance with the suggested west to east migration of the depocentre during the Late

Paleogene±Neogene. However, this speculation cannot be tested without results from future exploration wells to the west of the present study area. 6. Conclusions ² The raft tectonics in the Lower Congo basin can in broad terms be attributed to three `high strain rate intervals' separated by intervals associated with very low strain rates. The high strain rate periods initiated at approximately 96, 28 and 10 Ma. The ®rst persisted for 6±26 Ma

926

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927

the second for approximately 7 Ma, while the latter remains active. ² Based on the structuring caused by the raft tectonics we have sub-divided the study area into four domains where the typical structures for the different domains are: Cretaceous raft blocks and grabens (domain 1), Tertiary raft blocks and grabens (domain 2), Tertiary pre-rafts (domain 3), salt diapirs (domain 4). While the principal structuring in domain 1 more or less was accomplished already in the Cretaceous the development of the remaining domains continued throughout the Tertiary, and domain 3 is still active. ² The salt induced deformation and corresponding depocentre seems to migrate from the west towards the east during Late Paleogene±Neogene times. ² The main driving mechanism for Tertiary raft-tectonics in the Lower Congo Basin seems to be gravity spreading.

Acknowledgements The authors wish to thank Norsk Hydro ASA for making the data available and for ®nancing this study. Dr Ian Sharp is thanked for valuable comments on an earlier version of the manuscript. We also thank Dr Jake Hossack and one anonymous referee whose comments clearly improved the quality of this paper. References Abilio, M. S. (1986). The geology and hydrocarbon potential of Angola. In SADCC Energy Sector Technical and Administrative Unit, TAU, 129± 150. Ala, M. A., & Selley, R. C. (1997). The West African coastal basins. In R. C. Selley, African basins. Sedimentary Basins of the World, Vol. 3. (pp. 173±186). Amsterdam: Elsevier. Amaral, J., & Seyve, C. (2000). Angola Ð The Lower Congo lithostratigraphic chart. Geoluanda 2000, International conference (p. 16). Luanda: Department of Geology, University of Agostino. Anderson, J. E., Cartwright, J., Drysdall, S. J., & Vivian, N. (2000). Controls on turbidite sand deposition during gravity-driven extension of a passive margin: examples from the Miocene sediments in Block 4, Angola. Marine and Petroleum Geology, 17, 1165±1203. Brice, S. E., Cochran, M. D., Pardo, G., & Edwards, A. D. (1982). Tectonics and sedimentation of the South Atlantic rift sequence; Cabinda, Angola. Studies in Continental Margin Geology, 34, 5±18. Brognon, G. P., & Verrier, G. R. (1966). Oil and geology in Cuanza Basin of Angola. AAPG Bulletin, 50 (1), 108±158. Burollet, P. F. (1975). Tectonique en radeaux en Angola. Bulletin de la SocieÂte GeÂologique de France, 17 (4), 503±504 Les marges continentales et leur inteÂret eÂconomique. Clifford, A. C. (1986). African oil Ð past, present, and future. In M. T. Halbouty (Ed.), Future petroleum provinces of the world. Am. Assoc. Pet. Geol. Mem. 40. 339±372. Coward, M. P., Purdy, E. G., Ries, A. C., & Smith, D. G. (1999). The distribution of petroleum reserves in basins of the South Atlantic margins. In N. R. Cameron, R. H. Bate & V. S. Clure, The oil and gas habitats of the South Atlantic Geological Society Special Publications, Vol. 153. (pp. 101±131). London: The Geological Society.

Dupont, G., Masse, P., Moron, J. -M., Pourtoy, D., & Gerard, J. (2000). Integrated high resolution biostratigraphy in Oligo-Miocene turbiditic series from the Lower Congo Basin (Angola): a constrain of seismic and sedimentological interpretations at chronostratigraphic reservoir scales. In Geoluanda 2000, International Conference, Department of Geology, University of Agostino, Luanda, 59. Duval, B. C., Cramez, C., & Jackson, M. P. A. (1992). Raft tectonics in the Kwanza Basin, Angola. In M. P. A. Jackson, Special issue; salt tectonicsMarine and petroleum geology, Vol. 9. (pp. 389±404). Surrey: Butterworth, Geological Society. Elliot, D. (1976). The energy balance and deformation mechanisms if thrust sheets. Philosophical Transactions of the Royal Society of London, A280, 529±539. Franks, S., & Nairn, A. E. M. (1973). The equatorial marginal basins of west Africa. The Ocean basins and margins; Vol. 1, The South Atlantic. New York: Plenum Press, pp. 301±350. Ge, H., Jackson, M. P. A., & Vendeville, B. C. (1997). Kinematics and dynamics of salt tectonics driven by progradation. Association of American Petroleum Geologist Bulletin, 81 (3), 398±423. Jackson, M. P. A., & Talbot, C. J. (1991). A glossary of salt tectonics, Austin: University of Texas, Bureau of Economic Geology. Karner, G. D., & Driscoll, N. W. (1999). Tectonic and stratigraphic development of the West African and eastern Brazilian Margins: insights from quantitative basin modelling. In N. R. Cameron, R. H. Bate & V. S. Clure, The oil and gas habitats of the South Atlantic Geological Society Special Publications, Vol. 153 (pp. 11±41). London: The Geological Society. Lunde, G., Aubert, K., Lauritzen, O., & Lorange, E. (1992). Tertiary uplift of the Kwanza Basin in Angola. In R. Curnelle, Geologie Africaine; 1er colloque de Stratigraphie et de paleogeographie des bassins sedimentaires ouest-africains; 2e colloque africain de Micropaleontologie. Bulletin des Centres de Recherches Exploration-Production Elf-Aquitaine. Memoire 13 (pp. 99±117). France: Societe Nationale Elf-Aquitaine. Lundin, E. R. (1992). Thin-skinned extensional tectonics on a salt detachment, northern Kwanza Basin, Angola. In M. P. A. Jackson, Special issue; salt tectonics. Marine and petroleum geology, Vol. 9. (pp. 405± 411). Surrey: Butterworth, Geological Society. Mauduit, T., & Brun, J. P. (1998). Growth fault/rollover systems; birth, growth, and decay. Journal of Geophysical Research, B, Solid Earth and Planets, 103 (8), 18119±18136. Mauduit, T., Guerin, G., Brun, J. P., & Lecanu, H. (1997). Raft tectonics; the effects of basal slope angle and sedimentation rate on progressive extension. Journal of Structural Geology, 19 (9), 1219±1230. Nuernberg, D., & Mueller, R. D. (1991). The tectonic evolution of the South Atlantic from Late Jurassic to present. Tectonophysics, 191 (1-2), 27± 53. Sahagian, D. (1988). Epiorogenic motions of Africa as inferred from Cretaceous shoreline deposits. Tectonics, 7 (1), 125±138. Seranne, M., Seguret, M., & Fauchier, M. (1992). Seismic super-units and post-rift evolution of the continental passive margin of southern Gabon. Bulletin de la Societe Geologique de France, HuitieÁme Serie, 163 (2), 135±146. Shultz-Ela, D. D. (2001). Excursus on gravity gliding and gravity spreading. Journal of Structural Geology, 23, 725±731. Teisserenc, P., & Villemin, J. (1990). Sedimentary basin of Gabon; geology and oil systems. In J. D. Edwards & P. A. Santogrossi, Divergent/ passive margin basins. Association of American Petroleum Geologists Memoir 48 (pp. 117±199). Tulsa: American Association of Petroleum Geologists. Valle, P. J., Sperrevik, S., & Gjelberg, J. (2001). Structural impact on deep marine sedimentary systems in the Lower Congo Basin, offshore Angola. In A. M. Evans, A. J. Fraser, S. I. Fraser & H. D. Johnson, Petroleum geology of deepwater depositional systems Ð advances in understanding 3D sedimentary architecture. London: The Geological Society.

P.J. Valle et al. / Marine and Petroleum Geology 18 (2001) 909±927 Vendeville, B. C., & Jackson, M. P. A. (1992a). The fall of diapirs during thin-skinned extension. In M. P. A. Jackson, Special issue; salt tectonics. Marine and petroleum geology, Vol. 9. (pp. 354±371). Surrey: Butterworth, Geological Society.

927

Vendeville, B. C., & Jackson, M. P. A. (1992b). The rise of diapirs during thin-skinned extension. In M. P. A. Jackson, Special issue; salt tectonics. Marine and Petroleum Geology, Vol. 9. (pp. 331±353). Surrey: Butterworth, Geological Society.