Petroleum Prospects of Lamu Basin, South-Eastern Kenya

Transcription

1 414 MOGAKA JOURNAL D. GEOLOGICAL NYABERI AND SOCIETY BERNARD OF INDIA K. ROP Vol.83, April 2014, pp Petroleum Prospects of Lamu Basin, South-Eastern Kenya MOGAKA D. NYABERI 1 and BERNARD K. ROP 2 1 School of Earth Sciences, South Eastern University College (A Constituent College of University of Nairobi), P.O. Box , Kitui, Kenya 2 Department of Mining and Mineral Processing Engineering, Jomo Kenyatta University of Agriculture and Technology, P.O. Box , Voi, Kenya drropben@yahoo.com; mogdan07@yahoo.com Abstract: The hydrocarbon potential of the sub-surface Lamu basin (SE Kenya) offshore sedimentary rock sequences of Mesozoic age formed the premise of this study. Major similarities and some differences in structural styles can be seen between the offshore Lamu and the Gondwana basins along the margins of Indian Ocean and Carnarvon basin along Australia s North West Shelf, where oil pools have been discovered. The existing well results and recent 2-D seismic data have been interpreted to identify various structural styles and play fairway segments, which bolster the possibility that the Karroo to late Tertiary sedimentary mega-sequences (~ m thick), suitable for hydrocarbon exploration, could be visualized in both the onshore and offshore Lamu basin areas. Similarly, major reservoir-seal and potential source intervals have been identified in the present study. The hydrocarbon indicator from the well-log data shows that oil potential in complex multiple petroleum systems, ranging in age from Triassic to Tertiary, have tested gas deposits. Well control of only one exploratory well per 25,000 sq km in the offshore Lamu basin shows evidence of the existence of at least two active petroleum systems. The Lamu basin has evolved consequent to a complex tectonic activity related to continental rifting and block faulting of the Lamu-Anza and Central African rift systems. An attempt has been made to recommend the probable prognostic structural leads, which are controlled by NW-SE trending faults sympathetic to the Anza-Lamu rift systems, for future essential sub-surface features of source rocks, reservoir rocks and the cap rocks in the Lamu basin. Keywords: Sedimentary rock, Hydrocarbon, Geophysical and seismic data, Lamu Basin, Karoo, Kenya. INTRODUCTION The Lamu basin (bounded by the Equator and 4 o 30'S latitude and by longitudes 39 o 00'E and 44 o 00'E) covers a large area of southeastern coastal Kenya with an aerial extent of 132,720 square kilometers, extending offshore from a relatively narrow onshore graben to cover most of the continental shelf and slope of Kenya (Fig. 1). The hydrocarbon resource potential of the sub-surface offshore segment of the Lamu basin (with sediment thickness ranging from 3 km onshore to 13 km offshore) can be assessed by comparing with other Gondwana basins on the margins of the Indian Ocean. The formation of the Lamu basin went through initial rifting during Carboniferous (Nyagah and Cloeter, 1992). The sediments deposited during the rift stage have been interpreted on the basis of geophysical data and facies, and subsequently inferred from geotectonic environment and their outcrop evidence (Rabinowitz and Coffin, 1982). The Gondwana break-up history of East Africa is reviewed to identify major reservoir-seal pairs and potential source intervals within the basin. The stratigraphic column of the offshore Lamu basin is compared with that of the Carnarvon basin, along Australia s Northwest shelf, in order to identify analogue plays and potential resource volumes (Rabinowitz, 1982). Historical well results and recent 2-D seismic data have been interpreted to identify various structural styles and play fairway segments. Four mega-sequences from the Karroo to the late Tertiary are identified and all seem to have potential reservoir-seal pairs. Major similarities and some differences in structural styles are pragmatic between the offshore Lamu and Carnarvon basins. Yet, the Carnarvon basin has shown reserves in excess of three billion barrels of oil from the drilling of about 50 wells per 25,000 square kilometers of productive area of the basin. The results of the comparison indicate a significant potential resource in a number of trap types, within a core area of the offshore play fairway in the Lamu basin. However, scant, a control of only one well per 25,000 square kilometers in the offshore / /$ 1.00 GEOL. SOC. INDIA

2 PETROLEUM PROSPECTS OF LAMU BASIN, SOUTH-EASTERN KENYA 415 SUDAN? LOTIKIPI BASIN UGANDA LOKICHAR (LOPEROT) TROUGH LAKE VICTORIA SOUTH KERIO TROUGH?? NYANZA TROUGH KILOMETERS? TURKANA BASIN SUGUTA Sedimentary Basins Tertiary Rift Basin Anza Basin Mandera Basin Lamu Basin MAGADI TROUGH KENYA NAIROBI ETHIOPIA ANZA BASIN MANDERA BASIN LAMU EMBAYMENT Lamu basin shows evidence of the existence of at least two active petroleum systems. Considering the Lamu basin, a total of 16 wells have been drilled in the sub-surface as shown in Fig.2, many of which were poorly located but some tested questionable structures. The initial drilling focused primarily on Mesozoic, marine Jurassic and Cretaceous rocks that are visible in a few poorly exposed outcrops and large block faulted structures on intra-basinal highs and platform areas. Only in the later stages did the drilling activity move in the Tertiary depocentres. Petroleum exploration is an arduous job involving interpretation and integration of geoscientific information emanating from indirect sources (Rop, 2003). The surface geologic information in both onshore and offshore of Lamu basin is mainly gathered from exploratory wells, geophysical surveys and wireline well logs. Meticulous data from these sources are required. Projections for probability of sources, reservoir and cap rock combinations over an area with favorable entrapment conditions lead to the identification of prospective structures and locales for exploration drilling. Over the past five decades there have been spectacular scientific advances in the quality and interpretation of geophysical data for meeting geological goals. The National Oil Corporation of Kenya (NOCK, 1995) is keeping pace with the modern techniques and technology and has procured matching computer-aided facilities (Rop, 2003). SOMALIA INDIAN OCEAN Fig.1. Map showing sedimentary basins in Kenya (Lamu basin shaded yellow) To date, about 33 wells have been drilled in the sedimentary basins of Kenya but so far not a single viable hydrocarbon field has been discovered. However, efforts are still underway to demarcate drillable structures in the sub-surface basins. Much of the present work depends upon accomplishing extensive geophysical and geochemical data and their logical interpretation, where the existing information on drill cores from the onshore and offshore Lamu basin have been studied and interpreted. The much needed geophysical (gravity and seismic) data was made available for the present work by NOCK (1995) in order to interpret and understand the sub-surface geology and structures. Though, exploration in the offshore area of the Lamu basin was dormant from 1954 until the 1980s, the Cities Services Group, which included Marathon and Union Oil, completed two dry holes: Maridadi-1 (4,196m deep) and Kofia-1 (3,629 m deep ) in 1982 and 1985, respectively. While the activities in the offshore area were disappointing, they revealed the presence of a possible Jurassic salt basin that had probably undergone significant halo kinetic movement thus forming diapiric structures (NOCK, 1995). The under-explored oil reserves and ample evidence of a sizeable hydrocarbon deposit in the basin instigated Woodside Petroleum Company, that started exploration of the offshore Lamu basin in 2003, acquiring approximately 3600 km of 2-D seismic data, between the months of November 2004 and January 2005; eventually drilling one well by early 2006, although dry. Fig.2. Location of wells drilled in the Lamu basin.

3 416 MOGAKA D. NYABERI AND BERNARD K. ROP GEOLOGY, LITHOLOGY AND GEOPHYSICAL SUBSURFACE STRUCTURES The Lamu basin formed a failed arm of a major Paleozoic tri-radial rift system with the southwestern arm extending further south up to Malawi (Rabinowitz and Coffin, 1982). The separation of Madagascar along a strike slip transverse fault (Davie fracture zone) in late Mesozoic times created an initial transform margin configuration. Passive margin development accompanied progradational sediment deposition on a continental margin that underwent periodic rifting and subsidence until recent times (Karanja, 1982; NOCK, 1995). Outcrop studies in the coastal onshore areas of Lamu basin confirm the presence of potential reservoirs in the Jurassic carbonate sequence (Nyagah, 1992). In Triassic and early-mid Jurassic prospective Jurassic reservoirs may be overlain by evaporitic units on the regionally developed marine shale. Jurassic source rock potential indicates hydrocarbon fluid inclusions found in Jurassic oolitic limestone samples from the Lamu basin. In the Cretaceous the reservoir comprises deltaic clastics of the Ewaso, Kofia sands and the Frere town limestone which has well developed vuggy porosity. Walu shale which is related to a regional transgression and the pro-deltaic sequences provide adequate sealing potential. The seismic reflection mapping is concerned with delineation of sub-surface sedimentary structures and lithologic variations in the upper 5 km of the crust (BEICIP, 1984). Located in the northern part of the basin is a NW-SE trending structural basement high (the Garissa High), while in the southern part is an N-S trending basement uplift (the Walu and Tana synclines), which is believed to be an onshore continuation of the Davie Fracture Zone and the Cap Saint Andre Axis (NOCK, 1995). This basement high separates the Morondava and Majunga basins of Madagascar. The western boundary of the basin is marked by N-S trending fault that separates the Karroo rocks from the outcropping Precambrian basement, which shows a distinct tectonic direction, exhibited by structural elements in the basin. An example of a structural element is the Hargaso anticline (section of 181 km running NW SE), which includes Hargaso well in Dodori area in the basin. In the southeast of the faulted Hargaso anticline, all horizons undergo progressive but irregular deepening in the direction of the coast (Fig. 3). There are few unconformities in this southeasterly dipping and deepening sedimentary basin. The on lap of Paleocene and late Eocene on the Cretaceous period is very clear. The assortment of deep horizons (the top and the base Cretaceous limestone s, possible tops of Jurassic and Triassic periods) is more difficult as one proceeds towards the southeast. Hagarso is a structure related to basement tectonics but, R 94 line movements relate to deep intrusions apparently rooted below the bottom of the section (below 4.5 second t.w.t that is, km). The intrusions are either crystalline, which originated within the basement or salt diapirs. The Fig.3. Burial history curve of Hagarso-1 well

4 PETROLEUM PROSPECTS OF LAMU BASIN, SOUTH-EASTERN KENYA 417 Jurassic salts resulted from the opening of the sea between Madagascar and Africa, forming the base of Jurassic at around 9 km (Nyagah and Cloeter, 1992). The Karroo sediments are moderately thick. The total offshore line 140 begins at about 25 km SE of Malindi and extends offshore for 140 km with water depth increasing from 300 2,400 m (BEICIP, 1984). In the southeast a lenticular sequence poor in reflection and possibly identified as Jurassic salt rests on about 2,000 m of flat energetic series, considered as belonging to the Karroo group. The thickness of the possible salt may vary towards the southeast from 4,000 to few hundreds meters. Above this the Cretaceous and Tertiary sediments are rather constant in thickness and the bottom itself is approximately parallel to the top salt. In the northeast, the Cretaceous, the Eocene and the Miocene are heavily faulted and may be very thick in a syncline centered in shot point This tectonism may be related to salt diapirism but deep reflections are too poor and too discontinuous to allow thorough interpretation. Here the Karroo and the basement are probably very deep. The profile ends in the northwest on the flank of a probably diapiric, anticline culminating 15 km southeast of Malindi. The transition between these two parts of profile corresponds to strike-slip fault which are related to Davie ridge, along which Madagascar is known to have shifted towards the south during the Jurassic and early Cretaceous, where the related horizons on both sides of the fault exhibit clear change of facies (Simiyu,1989; NOCK, 1995). In shallow horizons, for instance, at the top of Cretaceous, the fault corresponds to the Tembo anticline and at depth it appears to be down to the northwest. The roughly northsouth line of culmination, which extends from the Tembo anticline to the Kipini structure, through the Kifaru diapir, is considered as shallow expression of the strike-slip fault. Four major groups of stratigraphic mega-sequences (Nyagah, 1992; NOCK, 1995) have been distinguished for the strata of Lamu basin sub-surface. The local lithological and sedimentological assemblages are delimited by major unconformities and grouped into mega-sequences (Fig. 4). The mega-sequence boundaries were seismically defined. Mega-sequence 1 belongs to the Karroo Group of rocks (Permo-Carboniferous to early Jurassic). The sequence comprises in ascending order, the Taru Grifts, Maji ya Chumvi, Mariakani sandstones and the Mazeras sandstone. These rocks were deposited in graben and half graben structures. The axes of these grabens are mainly oriented in north-south direction. These grabens are a result of regional extension that was the precursor to the break-up of Gondwanaland (Miller, 1852; NOCK, 1995). The Taru Member consists of primary sedimentary structures such as rare water escape and slump structure (Odada E.O, 1995). The Maji Ya Chumvi Formation comprises lacustrine regime that was sustained by renewed faulting during late Permian times (Cannon et al., 1981). A faulted contact exists between the Mazeras and Mariakani formations. Movements (middle Triassic) gently folded the Mariakani sandstone producing broad and shallow synclines and anticlines (Caswell, P.V, 1953, p.35). The beds have a regional dip to the northeast throughout the greater part of the area. Upper Jurassic to Cretaceous had a Tertiary passive margin of developing Indian Ocean Complex and sediment evolution (BEICIP 1984; NOCK, 1995). The major structural elements in the Lamu basin include; the ENE-WSW structural high trending from Garissa to Wal Merer High, with N-S strike (Precambrian basement direction with block faulting tectonics but also strike slip faulting which originated during the formation of proto- Indian Ocean) and NE-SW structural low (Simiyu, 1989; Nyagah, 1992). In Triassic, characteristic traps are mainly block faulted anticlinal structures with associated antithetic faults. Various large intra-basinal anomalies have been mapped, seismically and by aeromagnetic, along the near shore areas of the coast. The largest of these structures covers 400 sq km. In Cretaceous, structural-stratigraphic traps associated with the Ewaso sand and Kofia sands are present within the sequence. In the Paleogene, there is a large structure with 1000 m of vertical closure, in well positioned in delta lobe and down the southwest plunge of the Walu- Kipini High (NOCK, 1995). Another large structural complex located on the delta fair way has been informally termed the Hackberry play (Fig. 5). It is a direct geologic analogue to the productive trend in the Texas Gulf Coast. Back-reef and shoal-water oolitic carbonate facies overlying basement horsts or pinching out on the Walu Kipini basement ridge, provide favorable trap geometries. Several small reefs that are just above the oil window have been mapped, seismically, on the southwest flank of the Walu High. A very large Miocene reefal build-up, extending for 40 km in water (900 m to 1000 m depth), is seismically identified. Although the reef trend may be above the top of the oil window that starved ocean basin conditions, following basin subsidence may have been critical in limiting compaction and thermal conductivity which influenced the development of favorable thermal regimes. PROGNOSTIC EVALUATION OF HYDROCARBON POTENTIAL Exploratory drilling in Tertiary depocentres has shown encouraging wet gas shows in the basin. Recent time

5 418 MOGAKA D. NYABERI AND BERNARD K. ROP Fig.4. Stratigraphy of Lamu basin.

6 PETROLEUM PROSPECTS OF LAMU BASIN, SOUTH-EASTERN KENYA 419 Fig.5. Seismic profile showing the Hackberry Play. temperature index determinations for coastal wells indicate that the oil window is generally located at depths greater than 3000 m, due to the low geothermal gradient associated with the great sediment thickness in this part of the Lamu basin (Rop, 2003; NOCK, 1995). The present day knowledge indicates that hydrocarbon discoveries in the world have been made in basins located within or at the edge of cratons (BEICIP, 1984), not overlooking the continental margin where the Lamu basin is located. The rates of sedimentation in the basin average 30 to 50 meters per million years, during rifting and post rifting with Neogene period yielding a higher value (around 100m / 10 5 y) for Lamu embayment. The rifting stage affected the Lamu basin during late Jurassic and was responsible for the generation and maturation of source rocks. High geothermal gradients and early migration into paleostructures or pinch are associated with rifting system. The failed rift and the post rift stages were characterized by low geothermal gradients. The second phase of subsidence which resulted in the generation of hydrocarbons, applies to Karroo sediments of the basin of Jurassic, for example oil shows that were encountered in Kipini and Dodori wells (upper Cretaceous and lower Tertiary beds) are related to Tertiary subsidence which triggered a late phase of oil generation from Jurassic source rocks. The Tertiary period rocks have low degree of maturation while the upper Cretaceous rocks prove immature except in the deepest parts of studied basins. The maturity levels in lower Cretaceous are variable with Garissa High (Wal Merer 1 well) probably being overmature, whereas, the Garissa flanks (Bahat Anza graben, Kipini Dodori) show mature to over-mature. Those of the Jurassic period (from the present day geothermal gradient 3 O C/100 m) whose Cretaceous sections are very thick (4,000 m), the maturation is likely to be over-mature. The occurrence of oil window in most of these wells did not coincide with strata possessing well developed porosities. Saturated porous sands of the Kipini sands at terminal depth of the oil window were seen to be deeper than previously determined. It is possible that some well-developed source rock sequences from Carboniferous to mid-jurassic are likely to exist in Lamu embayment. CHARACTERISTICS OF THE SOURCE ROCKS The assessment of the hydrocarbon potential of the basin hinges on the establishment of time stratigraphic framework interpretation of deposition conditions of different facies, and regional paleogeographic reconstruction. The eastern onshore of Lamu east (forming extension of Davie Fracture Zone) is characterized by reduced rift sequence continuation of Bur ridge in rift stage. The Tertiary sandstones in this area has 5 20 percent porosity which is comparable to structural high areas of lower Oligocene deposits of Maragh Trough, southeastern Sirt basin, Lybia (Reeveset al, 1986). Middle Cretaceous carbonates between Wal Merer

7 420 MOGAKA D. NYABERI AND BERNARD K. ROP and Walu have porosity of 4 7 percent with depths of 2000 meters; late Cretaceous sandstone and quartzite have a porosity of 4 14 percent at depths of meters between Hargaso and Walu. The late Tertiary beds are evidently containing rich potential source rock. Hydrocarbon trapping areas in this formation are in anticlines of block faulting type which are associated with compaction of faults. The clastic fans of Eocene Paleocene sequence are up to 10 meters with porosity of 20 percent, while those of upper Cretaceous (1 2 meters thick), have percent porosity. The reefal limestone in the Tertiary sequence is meters thick, and the Cretaceous Tertiary section presents fair to good source rock. Source rock might well be down in the oil window of the basin and hydrocarbon migration from hypothetical section, or the infra evaporatic sequence might also have filled Tertiary reservoirs. The Tembo Walu culmination, situated in the Tembo Kipini Walu trend, has better possibilities of hydrocarbon migration from deep source rock as well as in potential structural area in the Malindi high, with limited closure. The other good prospects of hydrocarbon potential are in Eocene to Oligocene deltaic clastics and shelf carbonate facies underlying the present coastal area which also constitute good prospective reservoirs (Karanja, 1982). Shelf edge patch reefs, commonly associated with these shelf facies, could also be the targets in the shallow offshore area of the Pate embayment. The pattern of alternating periods of subsidence and sedimentation should have brought about the source and reservoirs facies into spatial and temporal coincidence (Miller, 1952; NOCK, 1995). There is a frequent occurrence of the Karroo euxinic sediments (BEICIP, 1984) with a possible existence of rich and well-developed lacustrine and/or marine source rocks. The organic matter present in the sediments is coaly, humic Type III, algae-sapropelic Type II and there may be the enriched Type I under favorable conditions of deposition. The Karroo source rocks sequences, in the present study, have similarities to Gondwanaland depocentres in other parts of the world, such as the Carnarvon basin where oil reserves in excess of three billion barrels of oil have been discovered. The sedimentation, in the initial rifting stage, took place from late Triassic to mid-jurassic under the anoxic marine environmental conditions. These sedimentary sequences, deposited under the same geotectonic conditions, have shown well-developed and matured source beds, due to the high geothermal fluxes and rapid sediments deposition in the rift environment. The evaporate beds observed in the Lamu basin act as excellent cap rocks for hydrocarbons generated during the syn-rift or initial rifting tectonics. The Lamu embayment underwent a typical evolution in the post-rift stage (mid-jurassic to Quaternary) causing continental passive marine depositions with good source rocks in the passive marginal sediments. The organic matter in these formations may contain Type II kind of kerogen, but more complex type (maybe Type II) may be predominant if sporadic and/or deltaic facies were present during sediment deposition. Maturation in the post-rift stage may not have been rapid owing to the low geothermal gradients of passive margin. Deep burial is, therefore, required in order to attain the oil window. Exploration data indicates that the Kipini-1 well has 400 m plus of porous and permeable sands while the Pate-1 is known to have flowed wet gas, water and mud from sand unit of lower Eocene age. Shale deposited during the intermittent sea level cycles of the Paleogene are effective caps. Shale sections associated with the Kipini Formation and the Pate limestone demonstrate fair to good source potential. Similar formation, for example oil-prone shales of the Eocene age with good source richness, has been identified in the Pemba-5 well on Pemba Island in Tanzania just to the south of the Kenya Tanzania border. Drilling of Nyuni-1 well has confirmed an active petroleum system in the area as evidenced by gas and oil shows in Tanzania. They posses oil and gas prone Type II to Type III Kerogen similar to source rocks of Niger grabens (Suggate, 1998). In the late Tertiary (Neogene), bioclastic debris associated with traps generated from salt diapirsm, also provide viable prospects. The Simba shales provide good seals for the deep water prospects with possible source rocks being shale sections in the Kipini sands and Kipini Formation (NOCK, 1995). GEOCHEMICAL STUDIES OF SOURCE ROCKS Based on geochemical analysis from the Simba-1Well (NOCK, 1995), the late Cretaceous and Paleocene sections have fair to good source richness with an average Total organic content of 1.4 % and Type III Kerogen, which is comparable to the source rocks of Cretaceous sequence at the Abu Gharadig oil and gas field, northwestern desert, Egypt (Kent, 1982; Khaleb 1999). The late Cretaceous rocks carry source rock characteristics similar to those of Tertiary sections, with rich marginal organic matter (Type IV) and a fair (4kg Hc/t. rock) mature source rock interval at 3450 m depth in Kipini-1 well, which may be conducive for gas production. Some local marginal source rock sequences whose organic matter could be detrital residual Type IV, often associated with humic or gas-prone kerogen Type III, have been realized in the lower Cretaceous. Those of the

8 PETROLEUM PROSPECTS OF LAMU BASIN, SOUTH-EASTERN KENYA 421 Jurassic sequences contain Type IV kerogen and poor source rock petroleum potential (1kg.Hct/rock) with fair reservoir capabilities. In the Tertiary sequence, immature source intervals were identified in the Kipini 1 and Dodori 1 wells which contained one fair (3kg. Hc/t. rock, m) interval in the Kipini (1,900 m) upper Eocene and a good interval (15kg.Hc/t. rock) with intercalated coal seams of mainly sandy section (20 30 m) in middle Eocene (2160 m). Those of the Dodori 1 well have a good interval (48kg.Hc/ t rock) in middle Eocene (1450 m depth) with also intercalations of several thin coal or lignite streaks in the sandy sequence. In 1983 BEICIP Labs analyzed 42 surface samples, which included 10 silty-shaley samples from Cretaceous period, 3 silty samples from Permian period, and 29 samples; all from Jurassic lithologies (sandstones, siltstones and shale). No significant petroleum show was recorded from these zones. KARROO SOURCE ROCK POTENTIAL The sediments of Karroo were laid down during the initial rifting of Lamu embayment under continental lacustrine to restricted marine conditions during the late Paleozoic to early Triassic. Such conditions (anoxic environment) are often favorable to the preservation of organic matter, with rich and well developed source rocks containing either Type III or Type II, or even Type I under favorable conditions and depth. Such source beds are likely to be associated with excellent carrier beds. The source characteristics of the Karoo sediments in the depocenters of Gondwanaland basins contain specks of bitumen in cores, as well as carbonaceous intercalations in the Ria-Kalui well, rich in lenticular black shale of Duruma outcrops. Those to the south of Taru Formation contain fair amount of organic carbon (0.5 1 percent) and fair to rich Total Organic Content ( percent). The Permian sandstones tested in the Calul well of Southern Ethiopia (the extension of Mandera Basin in northeastern Kenya) have produced gas of 1 million at m 3 / d from m depth. The Madagascar Isalo I sandstones hydrocarbons, contained in tar sands, seem to have similar stratigraphic equivalents to those of the Taru Grits of Maji Chumvi Karroo beds in Kenya. The impressive thicknesses of coal seams (several meters thick) have been reported in lower Permian coal measures (Rabinowitz, 1982) of India. The prolific source rocks deposited during initial rifting, which led to the separation of Southern Africa from South America, have also been encountered in Neocemican lacustrine prodeltaic shales of the coastal Brazilian fields; the abundant source rocks of early Cretaceous shale and lacustrine clays in the Gabo Congo coastal basin (Mayanga Marls, Melanie formation, and Black shale) also form potential source rocks (BEICIP, 1984). The Karroo sediments (4,000 m thick) could have undergone a rapid maturation during the late Paleozoicmiddle Jurassic rifting stage, resulting in high geothermal fluxes (gradients 3 o C/100 m). Thus in such geotectonic environments, the maturation may have occurred in greater depths during the post-rift or failed-rift stages, with burial becoming considerably deeper during late Triassic Cretaceous and Tertiary sedimentation. This has given rise to the top of the Karroo at 5,000 10,000 m depth in Lamu embayment. The Karroo sediments increasingly become over-mature or super-mature as evidenced in the thermal and burial history (Fig. 3). The hydrocarbons generated before the present burial may have been preserved in traps which were not subjected to burial at destructive depths. The syn rift sequences of the Jurassic sedimentation in other parts of the world, with similar geotectonic conditions and facies characteristics, are apparently similar to those belonging to the Jurassic outcrops in Kenya. The central Lamu embayment deposits constitute good reservoirs that are devoid of source potential, comprising shaley interbedded limestones and anhydrites intercalations which gradually become more calcareous and subsequently terminate as evaporitic sequences. They have a well developed shaley or carbonate source rocks containing Type II kerogen. The argillaceous shale or evaporate form good cap rocks in the Karroo, although they are poor quality reservoirs. Under favorable structural conditions the underlying sandstone sections of the late Triassic to early Jurassic may be excellent carrier beds. Evidence of source rocks in syn rift sequences include the Miocene Globigerina marls of Egyptian fields and Gulf of Suez forming prolific source rocks. However, some transitional euxinic shale associations of Aptian age are considered to be good source rocks; for example, those in the Brazilian coastal fields and Jurassic limestone of Ganale Doria river in Ethiopia. The geothermal gradients of 3 C/100 m are far less in the post-rift sequences, hence slowing down the sediment maturation. The maturity level is likely to be high (overmature to super mature) in the Jurassic. The oil generation window obtained by use of computer-aided modeling indicate a burial depth ranging from 3,300 m to 3,400 m at the top of the oil zone and 4,700 m in the top of gas zone. The paleogeographic position of this region (what is presently Lamu embayment in eastern Kenya) was much to the south of the Equator during the Triassic-Jurassic/

9 422 MOGAKA D. NYABERI AND BERNARD K. ROP Cretaceous time (Bloom, 2002). Luxuriant vegetation on land, swampy grounds, humid climate and good rainfall were some of the prevailing environmental conditions. The organic matter that got buried could only generate the Type I, Type II or Type III kerogen whose initial product could be waxy crude or gas. The sub-surface strati-graphical and sedimentological studies indicate that the Cretaceous sediments of Lamu embayment show the tendency to become more pelagic in the direction of Deep Ocean, with the presence of source rocks dwindling at depths. Computer-aided model indicates the presence of mature upper Cretaceous sediments in the Walu Kipini Pate Dodori wells, whose top of oil window vary in depths from 2,400 3,600 m with the maturity levels gradually diminishing offshore (where they are probably immature). The lower Cretaceous sediments in the present study area are mostly over-mature, except in the proximity of the offshore areas (along Kenya coastal shores), where they may be mature at depths ranging between 3,300 4,700 m. DISCUSSIONS, PROGNOSTIC EVALUATION AND CONCLUSION The prognostic evaluation of potential hydrocarbons in Lamu basin have been examined and recommended, although they are not of any commercial quantities. It is also evident that there is poor well siting in the area under study. Literature shows that Tertiary episodes of subsidence in the Lamu basin (cyclic reciprocal sedimentation), halokinesis and faulting are important from the stand point of hydrocarbon formation and migration. Basinwards, at greater depths, the immature intervals of Kipini and Dodori wells might be in the oil zone window. From the interpretation of seismic data, the reservoirs along which many wells in the area have been drilled appear to be compartmentalized by a combination of structural lead features (faults) and sub-surface stratigraphic elements (geothermal highs). Faulting at favorable locations may have trapped the generated oil and gas although high heat may lead to over-maturation of hydrocarbons. Thus, on analysis and interpretation of the data from the basin, the potential effects of sub-seismic faults in the area should be of critical significance. For improved reservoir characterization, a model should be developed which integrates seismic, petrophysical and transient pressure data yields. This will prognostically direct the drilling of the wells at proper points of the reservoir, away from the major fault zones and zones of high thermal gradients. References (Received: 29 May 2011; Revised form accepted: 6 March 2013) BEICIP (1984) Petroleum potential of Kenya 1984 follow-up, Ministry of Energy and Regional Development, 100 p. BLOOM, A.L. (2002) Geomorphology, Third Edition, Prentice-Hall India Pvt. Ltd., New Delhi, 482 p. CANNON, R.T.; SIAMBI, W.M.N. and KARANJA, F.M. (1981) The proto Indian Ocean and probable Paleozoic Mesozoic Triradial rift system in east Africa. Earth Planet. Sci. Lett., v.52, pp KARANJA, F.M. (1982) Report on the geology of the Kilifi, Gede and Sokoke area. Geol. Surv. Kenya, Nairobi, 32p. KENT, P.E. (1982) The Somali ocean basin and the continental margin of east Africa. In: A.E.M. Nairn and F.G. Stehli (Eds.), The Ocean Basins and Margins, The Indian Ocean, Plenum Press, New York, v.6, pp KHALEB, K.A. (1999) Cretaceous source rocks at the Abu Gharadig oil and gas field, Northern western desert, Egypt. Jour. Petrol. Geol., v.22. pp MILLER, J.M. (1952) Geology of the Mariakani MacKinnon road area, Report No. 20, Geol. Surv. Kenya, 45p. NATIONAL OIL CORPORATION OF KENYA (NOCK, 1995) Hydrocarbon potential of the coastal onshore and offshore Lamu Basin of southeast Kenya, Integrated Report, 97 p. NYAGAH, K. (1992) Hydrocarbon exploration in the sedimentary basins of Kenya. In: P.S. Plummer, Ed., Proceedings of the Indian Ocean-First Regional Seminar on Petroleum Exploration, United Nations Department of Technical Cooperation for Development, Seychelles, pp NYAGAH, K. and CLOETER, J.J. (1992) International exploration opportunities in Kenya. Bull. Houston Geol. Soc., v.35, pp ODADA, E.O. (1995) Geological evolution of coast Kenya as Inferred from sedimentary sequences and marine terraces, Geol. Surv. Kenya, Nairobi, 24 p. RABINOWITZ, P.D. and COFFIN, M.F. (1982) Salt diapirs bordering the continental margin of northern Kenya and southern Somalia. Science, v.215, pp RABINOWITZ, P.D. COFFIN M.F. and FALVEY, D. (1983) The separation of Madagascar and Africa. Science., v.220, pp RAIS-ASSA, R. (1988) Stratigraphy and geodynamics of the Mombasa basin (Kenya) in relation to the genesis of the proto- Indian Ocean, Geol. Mag., v.125, pp REEVES, C.V. KARANJA, F.M. and MACLEOD, I.N. (1986) Geophysical evidence for a Jurassic triple-junction in Kenya, Earth Planet. Sci. Lett., v.81, pp ROP, B.K. (2003) Subsurface stratigraphical studies of Cretaceous- Tertiary basins of northwest Kenya, (Unpublished Ph.D. Thesis, University of Pune), 174p. SIMIYU, S.M. (1989) Geophysical studies in Lamu Embayment to determine its structures and stratigraphy, (Unpublished M.Sc. Thesis, University of Nairobi), 126p. SUGGATE, R.P. (1998) Relationships between depth of burial, vitrinite reflectance and geothermal gradient. Jour. Petrol. Geol., v.21, pp.5 32.