hydrocarbon potential of the meso-cenozoic turkana depression

Hydrocarbon potential of the Meso-Cenozoic Turkana Depression,

northern Kenya. II. Source rocks: quality, maturation, depositional

environments and structural control

M.R. Talbota,*, C.K. Morleyb, J.-J. Tiercelinc,d, A. Le Herissec,d, J.-L. Potdevine, B. Le Gallc,d

aGeological Institute, University of Bergen, Allegt. 41, 5007 Bergen, NorwaybUniversiti Brunei Darussalam, Jalan Tungku Link, Gadong BE1410, Brunei

cUMR CNRS 6538 ‘Domaines Oceaniques’, Institut Universitaire Europeen de la Mer, Place Nicolas Copernic, 29280 Plouzane, FrancedUniversite de Bretagne Occidentale, 6, avenue Le Gorgeu, B.P. 809, 29285 Brest, France

eUniversite des Sciences et Technologies de Lille, UFR des Sciences de la Terre, UMR CNRS Processus et Bilans des

Domaines Sedimentaires, 59655 Villeneuve d’Ascq Cedex, France

Received 14 March 2003; received in revised form 12 November 2003; accepted 14 November 2003

Abstract

Source rocks identified by drilling (Loperot-1 well) in the Lokichar Basin of the northern Kenya Rift were evaluated for source potential

using RockEval pyrolysis and palynofacies analysis. Results indicate moderate to good oil source potential from algal-prone source rocks.

The original sediments accumulated in a large freshwater lake similar to some modern rift lakes. In the Loperot-1 well, the section with the

highest source potential lies within the Lokhone Shale Member (up to 17% total organic carbon (TOC), average 2.4% TOC from well-log

calculation). The other potential source interval is the deeper Loperot Shale Member, which is highly mature to post-mature (R0 ¼ 1:1%;

average 1.2% TOC from well-log calculation). The poorer quality of the lower shale interval while partly due to maturity, is thought to be

largely a result of the basin evolution. The Loperot Shale Member was restricted to the Lokichar Basin, while the Lokhone Shale Member

was deposited during a time when displacement on the main boundary fault system was greater, and adjacent basins had been established.

Consequently, lacustrine conditions appear to have extended into adjacent basins, thus allowing the accumulation of relatively higher

concentrations of algal material in sediment-starved parts of the basin.

q 2003 Elsevier Ltd. All rights reserved.

Keywords: Northern Kenya Rift; Lokichar Basin; Lacustrine source rock; RockEval; Palynofacies

1. Introduction

Hints that significant quantities of hydrocarbons can be

generated in the East African Rift System (EARS) have

come from several sources over the years. Oil seeps and the

presence of organic-rich lacustrine shales in an onshore well

at Lake Albert (western branch of EARS) were perhaps the

first evidence (Harris, Pallister, & Brown, 1956; Schull,

1988). Subsequently oil seeps and natural tar balls have been

described from Lake Tanganyika (Simoneit, Aboul-Kassim,

& Tiercelin, 2000; Tiercelin et al., 1993), and the organic-

rich Poi Shales have been identified in the Miocene

fluvio-lacustrine Ngorora Formation of central Kenya Rift

(Morley, 1999c; Renaut, Ego, Tiercelin, Le Turdu, & Owen,

1999; see Fig. 1 in companion paper). Source rocks have

also been encountered in the Lokichar Basin, northern

Kenya Rift, where they can be seen at outcrop, and have

been sampled in shallow and deep exploration wells, and

can also traced on seismic reflection profiles. A general

description of the Lokichar Basin in terms of structure,

sedimentary infill, quality of reservoir, and source rocks has

been given in Morley et al. (1999) and in the companion

paper to this work. The aim of this paper is to provide more

details on the geochemistry of the source rocks, the evidence

for their depositional setting, and to discuss the link between

basin structure and source-rock quality. Two sets of shale

samples from the Loperot-1 well, cleaned cuttings and

sidewall cores, have been used in this work.

0264-8172/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.

doi:10.1016/j.marpetgeo.2003.11.008

Marine and Petroleum Geology 21 (2004) 63–78

www.elsevier.com/locate/marpetgeo

* Corresponding author. Tel.: þ47-55583390/3520; fax: þ47-

55589416/9417.

E-mail address: [email protected] (M.R. Talbot).

http://www.elsevier.com/locate/marpetgeo

Fig. 1. (A) Location map of the N–S-oriented string of Palaeocene–Miocene basins in the northern Kenya Rift: Lokichar, North Kerio, North Lokichar and

Lothidok (modified from Hung (1996)). The Loperot-1 well is located a few km north of the TVK-12 seismic line. The black star indicates the location of shale

outcrops. (B) Schematic 3D view of geological cross-sections illustrating the main structural and stratigraphic elements of the Lokichar, North Lokichar and

Lothidok basins, and correlation with the Loperot-1 well (projected). The top Oligocene time line cuts across the volcanic rocks filling the Lothidok Basin.

Based on outcrop, seismic reflection (TVK-2, -12 and -13 lines; see box A) and well data.

M.R. Talbot et al. / Marine and Petroleum Geology 21 (2004) 63–7864

2. Source rocks in the Lokichar Basin

Hydrocarbon exploration interest in the Lokichar Basin

(Fig. 1(A); see also Figs. 2(B) and 5(A) in companion paper)

was first sparked by gravity data, which revealed a large,

negative Bouger anomaly with a typical half-graben

geometry (Morley et al., 1992). Subsequent reconnaissance

seismic reflection data demonstrated the presence of a deep

half-graben and tilted fault blocks (Morley et al., 1992,

1999). The presence of a section some 6–7 km thick

indicated that any source rocks located in the lower half of

the basin fill section were likely to be mature, assuming

normal or elevated crustal temperatures. Consequently, a

key subsequent step in exploration was to establish the

presence of source rocks in the basin. Typically, where rift

sections outcrop they tend to be representative of

flexural margins, which are sand-prone areas (Lambiase &

Bosworth, 1995). Major episodes of fine-grained lacustrine

deposition, where present, tend to be located closer to the

boundary fault, in areas of maximum subsidence, and

remain buried unless the basin is subsequently subjected to

significant inversion. Consequently, considerable optimism

was generated when a 100-m thick black, organic-rich shale

(now called the Lokhone Shale Member) was found as

discontinuous outcrops close to the Lokhone Horst at lat.

02823.0050N; long. 35855.8500E (Fig. 1(A); see also

Fig. 5(A) in companion paper).

Further definition of the Lokhone Shale Member was

made using a water-well rig to drill a series of shallow wells

(maximum depth about 100 m), along selected seismic lines

Fig. 2. (A) Seismic-well tie between seismic line TVK-12 and Loperot-1 well (projected). Note high-amplitude, continuous reflections from the top and base of

the Lokhone Shale Member, a typical response associated with the impedance contrast between organic-rich shales and fluvio-deltaic sandstones. (B)

Stratigraphic log of the Loperot-1 well, and sonic and gamma ray curves (modified from Hung (1996) and Morley et al. (1999)). (C) Distribution of TOC in the

Loperot Shale Member, Loperot-1 well, based on sidewall core samples (after Morley et al. (1999)).

M.R. Talbot et al. / Marine and Petroleum Geology 21 (2004) 63–78 65

between the strike-projection of the shale outcrops

(Fig. 1(A)). This drilling program established that the

shale interval is at least 100 m thick, thermally immature

ðR0 ¼ 0:32–0:37%Þ; with average total organic carbon

(TOC) content for cuttings samples ranging between 0.05

and 8.6%. Selected pieces of shale were as rich as 11%

(Morley et al., 1999). Following a standard continental rift,

half-graben sedimentation model (Katz, 1995; Lambiase &

Bosworth, 1995), it was anticipated that the source rocks

would thicken considerably towards the basin centre, and

that their source quality would be maintained or improve in

the same direction.

The Loperot-1 exploration well (Figs. 1 and 2(A) and

(B)) was a very important test of the assumed source

potential of the basin. The well lay about 7 km closer to the

basin centre than the water wells and penetrated the

Lokhone Shale Member between 925 and 1385 m. Signifi-

cant basinward thickening of the shale interval, from about

100 to 460 m, occurs over the 7 km, confirming the

predictions based upon a half-graben depositional model.

Well logs clearly record the presence of this thick shale

interval. For example, the predominantly high gamma ray,

low resistivity and SP readings are indicative of a

consistently shale-prone section (Fig. 2(B)). Within this

section, the most organic-rich zone, defined by the TOC

content of shale samples from sidewall cores (1–17% TOC,

average 2.4% TOC; Hung, 1996; Morley et al., 1999), lies

between 1100 and 1385 m (Morley et al., 1999; Fig. 2(C)).

Correlation of the Loperot-1 well with seismic reflection

data demonstrates that high-amplitude, continuous to

shingled reflections characterize the top and bottom of the

Lokhone Shale Member, which internally displays continu-

ous low amplitude events (Fig. 2(A) and (B)). Thus, the

lacustrine facies can be extrapolated with some confidence

into the deeper part of the basin (Figs. 3 and 4).

In other rift basins it has been demonstrated that organic-

rich lacustrine sequences tend to produce such distinctive

reflection packages due to their internal homogeneity and a

marked reduction in seismic velocity passing from adjacent

sand-prone units into the shale (Liro & Pardus, 1990).

Vitrinite reflectance values for the Loperot-1 well are

0.6–0.65% for the depth interval 1050–1390 m, indicating

a rapid transition to maturity from the outcrops to the well.

The main depocentre lies a further 6–16 km west of the

Loperot-1 well; seismic data indicate that the Loperot Shale

Member here reaches thicknesses in excess of 1 km (Fig. 4).

The Loperot-1 well yielded a surprise when it encoun-

tered a second, deeper potential source interval called the

Loperot Shale Member, of questionable Eocene to Oligo-

cene age (Morley et al., 1999; Fig. 1(B)). The Loperot Shale

Member is found between 2325 and 2950 m (T.D.), and in

comparison with the Lokhone Shale Member is a much less

discrete and less obvious shale package on the well logs

(Fig. 2(B)). The gamma ray and sonic logs are very ragged

(Fig. 2(B)), and suggest that silt- and sandstones occur

interbedded within the shale. The most organic-rich interval

lies between 2410 and 2600 m, with TOC values of between

0.2 and 3.3%. Burial has rendered the section highly mature

to post-mature for oil (R0 ¼ 1:1%; depth 2410–2600 m;

Morley et al., 1999; Fig. 4).

3. Basin structure

The Lokichar Basin has a classic half-graben geometry,

with the basin fill thickening westwards towards the east-

dipping Lokichar Fault (Fig. 1(B); see also Fig. 5(A) and (B)

in companion paper). The fault shows some evidence for

early linkage of three separate faults, although the dominant

displacement pattern that affects the basin fill is a simple one

where the displacement maximum lies at the centre of the

fault (Morley, 1999a). Isopach maps of the two source rock

intervals (Lokhone and Loperot Shale Members) show three

isopach maxima located in similar areas for both intervals

(Figs. 5 and 6). For example, the strata are relatively thin

between seismic lines TVK 107 and 109 (Fig. 1(A)), within

the main depocentre for both shale intervals. This isopach

minimum might represent an early linkage point between

two faults that amalgamated to form the boundary fault.

The top of the basin fill is marked by lava flows of the

Auwerwer Hills (Auwerwer Basalts), which are of middle

Miocene age (Morley et al., 1992; Fig. 1(B); see also

Fig. 5(A) and (B) in companion paper). Subsequent to their

extrusion, the flexural margin of the basin was uplifted and

Fig. 3. Time-structure map (ms) of the base of the Lokhone Shale Member,

Lokichar Basin (from Morley et al. (1999)).


Fig. 4. (A, B) Isopach maps for the Loperot and Lokhone Shale Members, based on extrapolation of seismic reflections bounding the units from the Loperot-1

well, creating time-structure maps of the horizons, depth converting the maps and deriving isopach maps from the time-structure maps. R0 contours are based

on the Loperot-1 well data, extrapolated to time-structure maps. (C) Burial history plots for the Loperot-1 well, and pseudo-well locations on seismic lines

using the depth to mapped horizons. See Fig. 1(A) for seismic lines location.


eroded. This erosional event might be due to footwall uplift

in response to late Miocene–Pliocene motion on the

Lokhone Fault (which bounds the North Kerio Basin on

the eastern side of the Lokhone Horst; Fig. 1(A)). The

amount of uplift can be estimated from extrapolation of

seismic horizons, and suggests about 650 m of erosion.

Maturity data from the Loperot-1 well also indicate that

erosion has occurred. Assuming the normal vitrinite

reflectance value at the Earth’s surface is 0.2% (Dow,

1977), and considering a geothermal gradient value of

4.2 8C/100 m in this area, extrapolation of the subsurface

vitrinite reflectance values to the 0.2% value indicates about

600 m of uplift has occurred.

4. The Lokichar black shales

Cleaned cuttings from three shale intervals in the

Loperot-1 well were sampled at depths of 925–1385 m

(corresponding to the Lokhone Shale Member), 1770–

1818, and 2325–2950 m (Loperot Shale Member) at the

National Oil Corporation of Kenya laboratory, Nairobi, in

December 1998. These three units are informally named the

‘Lower’, ‘Intermediate’ and ‘Upper’ Black Shale Intervals,

Fig. 5. Regional isopach map (grey shaded area) for the Lokhone Shale

Member in the Lokichar Basin (contours in km), illustrating the structural

configuration of the west Turkana area during the late Oligocene–early

Miocene. Contours in the North Kerio Basin show estimated thickness of

Palaeogene–middle Miocene section in km. The extent of the Lokhone

Shale Member was limited to the north by an extensive region of

Oligocene–early Miocene volcanics, and the termination of the Lokichar

Fault at an inferred convergent overlapping transfer zone, with the

boundary fault to the Lothidok Basin. To the north-east, the Lokhone Shale

Member probably extended into the North Kerio Basin.

Fig. 6. Isopach map (grey shaded area, contours in km) for the Loperot

Shale Member. The Loperot Shale Member is restricted to the Lokichar

Basin. Changes in isopach thickness within the basin and the hanging wall

splays from the Lokichar Fault suggest the fault linkage and segmentation

pattern influenced the subsidence pattern during Loperot Shale Member

deposition. Areas of unshaded contours show estimated thickness of

Palaeogene–middle Miocene section in km.


respectively (Fig. 6(A)). In addition, green, dark grey and

black shales were sampled in the Lokichar Basin near

Lokhone (lat. 02823.0050N; long. 35855.8500E; Fig. 1(A)).

TOC values from the well cuttings and field samples were

obtained using RockEval pyrolysis at the Institut Francais

du Petrole. Data are summarised in Table 1. Palynofacies

studies were performed at the Laboratory of Micropaleon-

tology, UMR 6538 ‘Domaines Oceaniques’ in Brest.

4.1. Organic matter content and hydrocarbon source

potential

TOC values for the exposed shales near Lokhone are

uniformly low (,1%), in agreement with the values cited

by Morley et al. (1999). Values obtained on our samples

(cuttings) of the Lokhone Shale Member from the Loperot-1

well range from 0.1 to 7.15% TOC. However, values of up

to 17% are cited for selected samples from sidewall cores in

the well (Morley et al., 1999; Fig. 2(C)). High TOC contents

(.1%) are largely confined to the depth interval between

1130 and 1385 m (Lokhone Shale Member), and a thin

interval with 2% average TOC just above 2500 m (within

the Loperot Shale Member; Fig. 7(A)). Hydrogen Indices

(HI) vary from ,90 to 611. A few relatively high HI values

(.300) were obtained from samples with very low TOC

content (,0.2%). As HI determinations from rocks with

such low organic matter (OM) contents are known to be

unreliable (Peters, 1986), these results were rejected. Shales

with HI suggesting moderate to good oil source potential

(.300) all occur in the high TOC section of the Loperot-1

well (between 1130 and 1385 m; Lokhone Shale Member;

Fig. 7(B)). This is confirmed by source rock quality as

measured by the petroleum potential, S2: Many samples

from the OM-rich section have S2 . 5 and several exceed

S2 ¼ 10 mg HC gm21 rock, indicating good to very good

source quality (Peters, 1986).

More detailed insight into the HC potential is provided

by plotting TOC versus S2 (Langford & Blanc-Valleron,

1990). Fig. 7(D) shows that the OM-rich samples define one

highly correlated ðr ¼ 0:99Þ linear population, while a

second group is defined by samples with very low S2 and

low to moderate TOC. The slope of each line represents the

mean HI of the two groups and in the case of the HC-rich

shales indicates that the pyrolizable OM is today still highly

oil-prone with a mean HI of 670 (this mean is higher than

any of the measured HI values because it defines the

character of the pyrolizable fraction. Individual HI values

are influenced by the presence of dead carbon with no

generation potential). The most OM-rich horizons within

this group represent high-quality oil source rocks, and as the

section as a whole is of the order of 300 m thick (Fig. 7(A)),

this interval clearly represents a potentially major source of

hydrocarbons within the Lokichar Basin. The second OM

population has a very low HI (mean ca. 90), with

insignificant oil source potential.

A further parameter of significance to potential

hydrocarbon production is the maturity. One measure of

source-rock maturity is provided by the RockEval Tmax

(Peters, 1986; Tissot & Welte, 1984). The main shale

section has a Tmax of 438–452 8C (Table 1; Fig. 7(C)) and

the lower, HC-poor shales a Tmax mainly in the range of

463–478 8C, with one measurement of 523 8C (Table 1;

Fig. 7(C)), suggesting the presence of reworked or highly

degraded OM. Both sets of Tmax are in generally good

agreement with vitrinite reflectances ðR0Þ of 0.6–0.65

between 1050 and 1390 m, and 1.1 between 2410 and

2600 m (Shell Exploration and Production Kenya B.V.,

unpublished report). The maturity of the OM-rich shales is

relatively high in relation to their present burial depth

(1.0–1.4 km), suggesting either a recent period of uplift or

above-average heat flow in the region, or a combination of

both. The latter is certainly a possibility, given the magma-

rich rift setting (see below) and a modern geothermal

Table 1

RockEval data for shale samples (cleaned cuttings) from the Loperot-1 well

sampled at the National Oil Corporation of Kenya laboratory, Nairobi (see

text and Fig. 7 for further details)

Depth (m) TOC (wt%) S1 S2 S3 Tmax HI OI

1062 0.04 0 0.13 0.18 531 321 438

1074 0.17 0 0.11 0.24 438 63 143

1086 0.73 0.01 0.97 0.49 444 134 68

1098 0.51 0 0.57 0.16 446 111 32

1104 0.84 0.01 1.52 0.35 444 181 41

1128 1.22 0.02 3.76 0.27 447 308 22

1140 1.2 0.01 3.37 0.34 446 281 28

1152 2.3 0.02 9.42 0.54 449 409 23

1164 5.21 0.18 31.65 0.7 445 607 13

1170 7.15 0.41 43.54 0.73 444 609 10

1188 4.38 0.18 20.73 0.71 443 473 16

1206 1.81 0.04 6.01 0.61 448 332 34

1224 3.42 0.2 16.39 0.14 444 479 4

1236 2.35 0.11 8.68 0.37 446 369 16

1242 3.51 0.12 15.19 0.73 446 433 21

1254 1.3 0.03 2.54 0.67 444 195 51

1266 2.78 0.1 13.67 0.72 451 492 26

1278 3.32 0.22 20.3 0.63 447 611 19

1290 1.26 0.11 3.3 0.89 443 261 71

1314 1.93 0.11 8.67 0.69 451 448 36

1326 3.5 0.2 20.9 0.75 452 596 21

1338 1.3 0.04 4.25 0.65 448 327 50

1350 0.55 0.02 0.84 1.00 448 153 182

1380 1.21 0.03 3.98 0.6 448 328 50

1770 0.24 0.01 0.49 0.1 451 208 42

1776 0.41 0.02 0.57 1.16 447 139 283

1809 0.6 0.15 0.67 0.31 462 113 52

1818 1.87 0.32 1.7 0.19 463 91 10

2487 2.05 0.38 1.9 0.59 463 93 29

2496 1.38 0.2 1.16 0.42 463 84 31

2499 0.5 0.06 0.32 0.19 465 63 38

2502 0.38 0.02 0.22 0.48 478 59 126

2607 0.33 0.02 0.18 0.22 476 54 67

2613 0.2 0.02 0.1 0.18 465 49 87

2622 0.4 0.02 0.17 0.16 467 43 41

2847 0.35 0.04 0.24 0.13 459 69 37

2856 0.34 0.02 0.55 0.45 523 163 133

2862 0.12 0 0.76 0.02 471 644 14


Fig. 7. (A) Total organic carbon content of the ‘Upper’ (Lokhone), ‘Intermediate’, and ‘Lower’ (Loperot) black shale intervals in the Loperot-1 well. Samples

are cleaned cuttings provided by National Oil Corporation of Kenya (see Table 1). (B) Hydrogen Index (HI) versus depth for shales in the Loperot-1 well.

Shales with good oil source potential are confined to the depth interval between 600 and 1100 m (‘Upper’ Lokhone interval). (C) Tmax versus depth in the

Loperot-1 well. (D) Plot of total organic carbon (TOC) versus RockEval S2 for shales from the Loperot-1 well. Note the two very different, but highly

correlated populations, one rich in oil-prone organic matter, the other of no significance as a hydrocarbon source rock. For comparison, similar plots are shown

for core MPU-10 taken from a water depth of 443 m in the southern basin of Lake Tanganyika (Mondeguer, 1991), and core M98-3P from 392 m water depth in

the northern basin of Lake Malawi (Barry, Filippi, Talbot, & Johnson, 2002) (RockEval data for Lakes Malawi and Tanganyika from Talbot et al., in

preparation). (E) Hydrogen Index versus Tmax plot showing RockEval data from the Lokhone Shale Member (solid squares) and Loperot Shale Member (open

squares) in the Loperot-1 well. Curves I–III show the maturation trajectories for, respectively, Types I–III kerogens. Also plotted are corresponding data from

OM in modern African lake sediments (Filippi & Talbot, in press; Talbot & Lærdal, 2000; Talbot et al., in preparation). The field defined by the latter represents

the range of likely precursor compositions for OM that accumulated in the Lokichar palaeolake. Extrapolation to the immature precursors (arrows) suggests

that some of the Lokhone Shales were extremely rich in oil-prone OM at the time of deposition (see text for further discussion). Kerogen trajectories and

vitrinite ðR0Þ contours courtesy of Birger Dahl (personal communication, 2003).


gradient in the well of 42 8C km21. The oil window is

conventionally regarded as beginning at a Tmax of 430–

435 8C (Peters, 1986; Tissot & Welte, 1984), the shales can

thus be regarded as mature with respect to oil generation, a

Tmax of 438–452 8C suggesting that they are probably

close to their maximum generative capacity (Tissot &

Welte, 1984). Evidence of possible oil generation from the

Loperot shales was provided by traces of oil encountered at

several levels in the Loperot-1 well (Shell Exploration and

Production Kenya B.V., unpublished report), although

geochemical fingerprinting would be required to confirm

the source. The deeper shale section is already close to the

point of transition from oil to wet gas generation (Peters,

1986).

Despite being close to peak generation today, the upper

(Lokhone) shale member still has relatively high HI,

implying that it was probably extremely rich in oil-prone

OM when it entered the oil window (see discussion on

kerogen types below). It is thus probable that significant

volumes of liquid hydrocarbons have already been gener-

ated from this unit. In addition, today’s moderate to high HI

indicates that the unit still has considerable generative

potential. The lower unit lacks significant potential today,

and even when its high maturity is taken into account (see

below), it seems likely that it never had the oil-generation

capacity of the upper shale.

4.2. Use of well logs to calculate TOC content

The response of sonic velocity to OM content and the

resistivity response to the level of maturation can be used to

estimate TOC content of source intervals. The method

proposed by Passey, Creaney, Kulla, Moretti, and Stroud

(1990) was applied to the log data for the Loperot-1 well, to

test the method against the Lokhone Shale Member where

the TOC content has been measured, and to estimate the

TOC content of the Loperot Shale Member which was not

cored (Fig. 8). Gamma ray, sonic and resistivity logs were

digitized and scaled to 100 m/s/ft per two logarithmic

cycles. According to the method of Passey et al. (1990), the

TOC content is proportional to the separation Dlog R

between the sonic and resistivity logs:

Dlog R ¼ log10ðR=RbaselineÞ þ 0:02ðDt 2 DtbaselineÞ

where Rbaseline and Dtbaseline are the values of the fine-grained

non-source rocks chosen after superimposing the two logs:

Dtbaseline ¼ 127 ms=ft; Rbaseline ¼ 1:2 Vm: Then, the TOC

values were calculating using:

TOC ¼ ðDlog RÞ £ 10ð2:29720:1688£LOMÞ

where LOM is the level of maturation.

Based on the available vitrinite reflectance values, LOM

was chosen as 8.5 for the Lokhone Shale Member. The

calculated values were calibrated with the TOC data from

sidewall cores from the Loperot-1 well, and resulted in a

good match. The separation of calculated TOC and sidewall

core data below 1325 m (Fig. 8) is probably due to the poor

quality of the resistivity curve over that interval. The

Loperot Shale Member is a less discrete shale interval than

the Lokhone Shale Member, with more numerous sandy and

silty intervals. The organic-rich interval of the shale lies

between 2410 and 2600 m. It has relatively low TOC

content (0.2–3.3%) estimated from cuttings only, and has a

vitrinite reflectance ðR0Þ of 1.1.

Calculation of the TOC from sonic and resistivity logs

for the Loperot Shale Member was conducted using the

following values: Dtbaseline ¼ 90 ms=ft; Rbaseline ¼ 3 Vm;

LOM ¼ 12. The calculated TOC values shows the high

TOC intervals tend to be narrow bands 10–30 m thick that

spike to values of about 2–3% TOC. Some high TOC

intervals are calculated to be present below 2600 m. The

average log-calculated TOC values are 4.5% for the Loperot

Shale Member and 1.2% for the Lokhone Shale Member.

However, the 4.5% value for the Loperot Shale appears to

be an overestimate due to the problems with the poor quality

logs over the lower interval of the shale (Fig. 8).

Consequently, when log calculated values for the upper

part of the interval are combined with average sidewall core

values for the lower part of the interval, the average TOC for

the shales is 2.4%.

4.3. Palynofacies analysis

Palynofacies analyses were performed on cleaned cut-

tings from the three ‘black shale’ intervals identified in the

Loperot-1 well, in relation to the study of potential source

rocks (Fig. 2(B)). In addition, the Lokhone surface samples

(near the Lokhone Horst at lat. 02823.0050N and long.

35855.8500E; Fig. 1(A)) have also been analysed. Optical

studies were carried out on kerogen separated by standard

palynological techniques and provide the following results:

Upper black shale interval (Lokhone Shale Member:

920–1385 m) (Fig. 7(A); Table 2): the upper part of this

interval (920–1164 m) is characterized by mixed palyno-

facies with abundant palynomacerals (cuticle, woody

fragments), exoskeleton remains, and structureless OM,

associated with spores and pollen. The OM is yellow in

colour and very well preserved. The palynofacies is

consistent with the correspondingly low TOC values

(0.04–0.7%; Fig. 7(A); Table 2). From 1164 to 1385 m,

the organic facies is dominated by amorphous organic

matter (AOM; Table 2), with proportions that follow the

relatively high TOC content (Fig. 7(A)), and are thus

consistent with a good HC source potential. Some

fluctuations in the TOC content at the base of this interval

(1290–1385 m; Fig. 7(A); Table 2) may reflect variations in

the abundance of preserved OM versus the volume of

siliciclastic sediment. The unstructured OM is accompanied

by fungal remains, multicellular algae, Botryococcus spp.

(Fig. 9(A)), Pediastrum spp. (Fig. 9(B)), and some cysts of

Prasinophycean algae (Fig. 9(C)). Pediastrum as well as

Botryococcus are unequivocal indicators of fresh water


conditions, associated with lakes, ponds or lowland rivers

(Batten, 1996; Fritsch, 1961; Reynolds, 1980). The

petroleum generation potential of chlorophycean algae,

particularly Botryococcus and Pediastrum, has frequently

been underlined (Batten & Grenfell, 1996; Hutton, 1988;

Peniguel, Couderc, & Seyve, 1989; Tyson, 1995; Wang

et al., 1994; Wood & Miller, 1997). These algae, whose

modern counterparts are commonly associated with

eutrophic to hypertrophic conditions in lakes (Reynolds,

1984; Zippi, 1998), have here clearly contributed to the

excellent HC source potential of these levels, but their

percentages are generally limited compared to the AOM.

Intermediate black shale interval (1770 – 1818 m)

(Fig. 7(A); Table 2): This interval contains variably

Fig. 8. (A) TOC calculated from sonic and resistivity logs for the Lokhone Shale Member (depth in metre). Black dots are spot values for TOC obtained from

sidewall cores from the Loperot-1 well for comparison with calculated results. (B) TOC calculated from sonic and resistivity logs for the Loperot Shale

Member (depth in m).


degraded, dispersed AOM, with locally abundant pyrite and

associated heavy minerals, and disseminated structured

elements (Fig. 9(E)). Terrestrial microfossils (spores and

pollen) are more abundant than in the overlying interval,

and TOC content is very low.

Lower black shale interval (Loperot Shale Member:

2325–2950 m) (Fig. 7(A); Table 2): Samples from this

lower interval contain abundant AOM (Fig. 9(F)),

although it is less important in volume compared to the

upper black shale interval. The palynofacies are also rich

in fresh water coenobial algae, fungal spores and hyphae,

and few spores and pollens (Fig. 9(G) and (H)). The

dominance of Pediastrum and Botryococcus, together

with abundant AOM in the palynomorph assemblages at

2418 m, is an indication of eutrophication. These algae

can have contributed to the generation of hydrocarbons,

but the OM in this interval is at a higher level of

maturity, probably corresponding to a stage of gas

generation (Fig. 9(H)). This observation is clearly

corroborated by the RockEval data (Fig. 7(E)).

Outcrop samples. The surface samples collected at

Lokhone (Fig. 1(A)) belong to the upper Lokhone Shale

Member. They contain degraded AOM with few structured

elements. One sample has a palynofacies dominated by

small spores and organic fragments of uniform size,

suggesting sorting during transport and deposition. The

OM appears to be slightly degraded.

4.4. Kerogen types

A mean HI of 670 for the OM-rich shales suggests a Type

I/II composition for the productive kerogen (Cornford,

1990). Their TOC versus S2 character compares well with

phytoplankton-dominated OM assemblages accumulating

in some large modern rift lakes such as Lakes Tanganyika

and Malawi (Fig. 7(D)), suggesting the existence of a

similar water body in the Lokichar Basin at the time of

source rock accumulation. Palynofacies analysis reveals

relatively abundant remains of the green algae Pediastrum

and Botryococcus, both of which are prominent components

of the phytoplankton flora in many of the modern African

rift lakes (Hecky & Kling, 1987), and are relatively common

in sediments that accumulated during Pleistocene–Holo-

cene dilute phases in some East African lakes such as Lakes

Bogoria and Victoria (Grall, 2000; Talbot, 1988; Talbot &

Lærdal, 2000; Tiercelin, Perinet, Le Fournier, Bieda, &

Robert, 1982; Tiercelin & Vincens, 1987; see Fig. 1 in

companion paper). Further insights into the probable nature

of the original OM can be obtained by plotting HI against

Tmax and then extrapolating along the probable maturation

trend back to the precursor OM (Fig. 7(E)). As a baseline for

the likely starting material we have used late Pleistocene to

recent OM in sediments from modern African lakes (Filippi

& Talbot, in press; Talbot & Lærdal, 2000; Talbot, Lærdal,

Jensen, & Filippi, submitted; Fig. 7(C)), which probably

Table 2

Percentages of the different palynofacies elements from shale samples (cleaned cuttings) of the Upper (Lokhone), Intermediate, and Lower (Loperot) ‘black

shale’ intervals in the Loperot-1 well

Depth (m) AOM (%) CE (%) DW (%) Pyrite Sp.-Poll. (%) Fungi (%) Algae HM OI HI Tmax TOC

1062 9 83.60 5.30 0.40 0.80 U—0.50% 438 321 531 0.04

1086 42 53.30 2 2.10 0.80 P 68 134 444 0.73

1140 55 42 P 1.70 P–B 28 281 446 1.2

1164 87 9.50 0.80 P 1.60 P–B 13 607 445 5.21

1170 98 1.30 P 0.70 P–B P 10 609 444 7.15

1188 67.50 32 0.50 16 473 443 4.38

1242 96.80 2.90 0.20 21 433 446 3.51

1290 99.80 0.20 P–B 71 261 443 1.26

1314 99.70 0.30 36 448 451 1.93

1326 97.50 1.40 0.20 P P 0.80 B 21 596 452 3.5

1350 96 0.20 1.60 A: 1.2% 0.90 B 182 153 448 0.55

1380 96 2.60 P 0.10 0.60 B—0.3% P 50 328 448 1.21

1770 96 1.80 0.20 A 0.10 0.20 P, B—0.4% A 42 208 451 0.24

1776 87.50 8.50 1.20 A: 2% 0.10 0.30 P—0.1% P 283 139 447 0.41

1809 92.80 6.40 0.30 P 0.08 0.10 B—0.1% A 52 113 462 0.6

2418 86 1.40 6.90 A: 3.4% 0.10 P, B—2.2% A

2571 97 1.20 1.60 P 0.20 P

2622 92.80 3.70 1.80 0.06 0.20 P—1.3% P 41 43 467 0.4

2847 87.60 6.80 4.30 0.20 P—0.8% 37 69 459 0.35

2856 85.60 12 1.90 0.04 0.20 P 133 163 523 0.34

2862 97.60 0.70 0.90 0.20 U—0.40% P 14 644 471 0.12

AOM, Amorphous Organic Matter; CE, Cuticles, epiderms; DW, dark woody remains. Pyrite: A, Abundant; P, Present. Sp.-Poll.: Palynomorphs, spores

and pollens. P, Present; Fungi, Fungal remains. Algae: P, Pediastrum; B, Botryococcus; U, Undifferentiated. HM, Heavy Minerals; A, Abundant; P, Present;

OI, HI, Oxygen and Hydrogen Indices. Tmax; Temperature max. (8C); TOC, Total Organic Carbon (%).


provide close analogues to the sediments that accumulated

in the Lokichar palaeolake (see preceding discussion). This

baseline suggests that OM in much of the upper shale unit

had HI values that were over 600 and, at some levels, in

excess of 800 and was thus closer to a pure Type I kerogen

(Fig. 7(D)). The relatively high maturity of the deeper shales

precludes precise identification of their OM type from the

pyrolysis data alone, but the palynofacies analyses indicate

the presence of relatively abundant cuticle and woody

material in these rocks, suggesting a Type III kerogen, a

conclusion that is supported by the HI versus Tmax plot (Fig.

7(E)). Poor source potential in these intervals is thus

probably due to a dominance of highly degraded OM.

4.5. Basin temperature

The temperature at 2950 m, at the bottom of the Loperot-

1 well, was 149 8C (corrected), which using a surface

temperature of 25 8C, yields an average geothermal gradient

of 4.2 8C/100 m. Vitrinite reflectance values for the well are

as follows: 1050 m, R0% 0.6; 1110 m, R0% 0.64; 1390 m,

R0% 0.65; 2500 m, R0% 1.1, and suggest that about 650 m

of strata have subsequently been removed (see above). Heat

flow values from the Kenya Rift are highly variable,

generally ranging between 40 and 110 mW/m2, with

geothermal gradients between 2.5 and 6.6 8C/100 m

(Wheildon, Morgan, Williamson, Evans, & Swanberg,

1994). The Loperot-1 well indicates that the Lokichar

Basin today lies in the middle range of the data, however,

during the active rifting and volcanic phase, it is quite likely

that heat flow, and thus temperatures, were higher.

4.6. Generating potential of the basin

The high geothermal gradient and vitrinite reflectance

data gathered from the Loperot-1 well indicate the

following approximate depths of maturity for the basin:

1100 m early mature (0.5–0.8% R0); 2100 m mid-late

mature (0.8–1.2% R0); 2800 m overmature (.1.2% R0).

Using structure maps, derived from seismic data, of the

Loperot and Lokhone Shale Members, the volume of

source rock for the two intervals, at the different maturity

levels has been calculated. The results are as follows:

Loperot Shale Member, early mature 246 106 acre-ft; mid-

late mature 135.3 106 acre-ft; post-mature 38.7 106 acre-ft;

Lokhone Shale Member, early mature 13 106 acre-ft, mid-

late mature 46 106 acre-ft, post-mature 197 106 acre-ft.

Using an average value of S2 ¼ 5 mg HC gm21; and a

kinetic model based on Tissot and Espitalie (1975), the

potential generating capacity of the Lokhone Shale

Member (for the areas above if they were fully mature)

is estimated at 10.5 billion barrels. Given that about 40%

of the shale volume is at mid to post mature levels this

suggests the Lokhone Shale Member could have generated

about 4 billion barrels of oil.

5. Consequences of regional structure on source

rock-trap relationships

The presence of source rocks in the Lokichar Basin is

encouraging for exploration in Northern Kenya, however, it

is well established that in rifts thick, high-quality source

rocks can be restricted to a single half-graben basin. So what

do the Lokichar source rocks indicate about regional

prospectivity?

Despite the differences in maturity, the Lokichar Basin

source rocks have clear similarities with OM-rich sediments

that have accumulated during the Holocene in the deep,

offshore waters of Lakes Tanganyika and Malawi

(Fig. 7(D)). In both the latter examples, accumulation has

occurred below the chemocline in permanently anoxic

waters and it is likely that similar conditions prevailed

during deposition of at least the oil-prone population of

Lokichar samples. The OM-poor population with no

hydrocarbon potential (Fig. 7(D)) also has its analogues in

modern rift lakes and is typical of areas subject to fluvial

influence where highly degraded OM enters the lake

(Talbot, in preparation). In addition, the palynofacies

suggests a freshwater environment, conditions that are

also compatible with a lake basin similar to the modern,

hydrologically open giant rift lakes. Consequently, a strong

structural control on source-rock distribution in the

Lokichar Basin is implied, where displacement on the

boundary fault caused subsidence rates to outpace sediment

supply. The resulting accommodation space coupled with a

favourably humid climate (rainfall perhaps five times higher

than today) (Roche, 2002; Vincens, pers. comm.) enabled

the development of prolonged periods of deep anoxic

lacustrine conditions.

Fig. 9. Palynofacies elements from shale samples (cleaned cuttings) of the ‘black shale’ intervals in the Loperot-1 well. (A) Botryococcus brauni; size

69 £ 50 mm2, 1170 m, Loperot-1 well. (B) Pediastrum spp.; coenobium left 67 mm in diameter; right 33 mm in diameter, 1164 m, Loperot-1 well. (C)

Phycoma stage of Prasinophycean algae; Diameter 28 mm. Remark: Fossil Prasinophyceae are rarely associated with freshwater environments. Their presence

here could indicate a lagoonal/brackish environment, 1164 m, Loperot-1 well. (D) Palynofacies rich in AOM and containing also some fungal spores, cuticle

and membrane tissues; magnification approximately 200 £ , 1242 m, Loperot-1 well. (E) Palynofacies dominated by fluffy AOM and rich in framboidal pyrite.

Magnification approximately 200 £ . (F) Palynofacies dominated by grey particulate amorphous matter, with rare fungal remains and heavy minerals;

magnification approximately 200 £ , 2571 m, Loperot-1 well. (G) Palynofacies including brown woody remains, cuticle and other membranous tissues, AOM

and heavy minerals; magnification approximately 200 £ , 1776 m, Loperot-1 well. (H) Pediastrum spp.; coenobium left 75 mm in diameter. Compare with

photograph B for organic matter maturity. The brown colour indicates higher maturity than in the Upper (Lokhone) black shale interval in the well, 2622 m,

Loperot-1 well.

R


Despite the presence of two shale members separated by

fluvio-deltaic sediments close to the flexural margin of the

Loperot-1 well, it is likely that closer to the basin centre

these two shales merge into a thicker, vertically more

continuous sequence (Fig. 1(B)). The Loperot Shale

Member is not seen at outcrop, and is interpreted to onlap

the flexural margin west of the Lokhone Horst. Conse-

quently, the Loperot Shale Member was restricted to the

Lokichar Basin. The Loperot Shale Member lies deep in the

section, and formed at a relatively early stage in the basin

history, hence the early stage of fault linkage and

propagation is reflected in the limited distribution of the

Loperot Shale Member (Morley & Wescott, 1999—Chapter

13, Fig. 4). The early linkage of three faults is also reflected

in the presence of three isopach maxima near the

boundary fault, each coincident with splays in the Lokichar

Fault (Fig. 4).

The Lokhone Shale Member thins from about 460 m in

the Loperot-1 well to 100 m at outcrop near the Lokhone

Horst. The presence of source rocks so close to the (eroded)

flexural margin suggests that at least during one depositional

phase, presumably during a high stand of the Lokichar Lake,

anoxic lacustrine conditions extended eastwards into the

neighbouring North Kerio Basin (Fig. 5). This is encoura-

ging for extending hydrocarbon plays beyond the Lokichar

Basin. The northwards extent of the Lokhone Shale Member

is more problematic. During the Oligocene–lower Miocene,

the basin north of the Lokichar Basin, called the Lothidok

Basin, deepened eastwards (Morley et al., 1999; Fig. 1(A)).

The uplifted and eroded hanging wall of this basin, now

exposed in the Lothidok Hills (Figs. 1(A) and 5) reveals an

Oligocene–middle Miocene section dominated by volcanic

flows, pyroclastic and volcaniclastic rocks (Boschetto,

Brown, & McDougall, 1992). Consequently, the Lokhone

source rock interval passes northwards into a volcanic-

dominated sequence across a conjugate convergent transfer

zone, into the neighbouring Lothidok Basin. A major

volcanic centre lay at the northern margin of the Lothidok

Basin (Figs. 1(A) and 5). The final infilling phase of the

Lokichar Basin is marked by the presence of volcanic flows,

following the propagation of the volcanic centres to the

south, into the Napedet Hills area (Fig. 1(A)).

Accumulation of lacustrine source-rock precursors in the

Lokichar Basin was terminated by two effects: (1) the

development of volcanic centres just north of the Lokichar

Basin, and (2) a major structural reorganization of the region

in the middle–late Miocene. The southern part of the North

Kerio Basin, the Lokichar Basin and the Lothidok Basin all

became inactive in the middle Miocene. Active extension

switched to the North Lokichar Basin, the northern part of

the North Kerio Basin and the Turkana Basin (Morley et al.,

1999; Fig. 1). Consequently, new depocenters and new areas

of lacustrine sedimentation were established during the late

Miocene–Pliocene. No high-quality source rock intervals

related to these new depocenters have been identified in

outcropping late Miocene–Pliocene deposits in the Napedet

Hills or at Lothagam Hill (Fig. 1(A)). However, such

negative evidence does not eliminate the possibility of good

quality source rocks existing at depth, closer to the basin

depocenter, or offshore beneath Lake Turkana. Indeed the

Lokichar Basin suggests that such source rocks might occur

in the region.

The Lokichar Basin appears to be little affected by

inversion structures, which developed elsewhere in Western

Turkana during the Plio-Pleistocene (Morley, 1999b). This

late inversion occurred after subsidence ceased in the

Lokichar Basin during the middle Miocene. Hence hydro-

carbon generation and migration occurred before inversion.

Since inversion creates new traps and tends to destroy older

traps, the late timing of inversion could potentially have

destroyed older hydrocarbon accumulations and requires

late hydrocarbon generation to fill the inversion structures.

From this perspective, it is fortuitous for the petroleum

potential of the Lokichar Basin that the effects of inversion

are minimal.

6. Conclusions. Implications for exploration

The Lokichar Basin contains lacustrine shales, which

exhibit many of the characteristic features of source rocks

developed in continental rifts. The highest quality source

rock is the Lokhone Shale Member which accumulated in

freshwater, and is a Type I algal-prone source with an

average TOC of 2.4% (based upon well-log calculations and

sidewall cores) and containing intervals with up to 17%

TOC (Hung, 1996; Morley et al., 1999). Towards the

boundary fault, the Lokhone Shale Member probably attains

a maximum thickness of 1 km. The deeper Loperot Shale

Member has only marginal source potential, with an average

TOC of 1.2% from well-log calculation. Only one thin

interval at ca. 2500 m yielded high TOCs of around 2%. The

poorer quality of the lower potential source interval, while

partly due to higher maturity, is thought to be largely a result

of the basin evolution. The Loperot Shale Member was

restricted to the Lokichar Basin. Hence the lake was

relatively small (50-km long by 20-km wide), and the algal

OM subject to dilution by the influx of siliciclastics and

terrestrial plant material, as shown by palynofacies analysis

(Table 2). The Lokhone Shale Member was deposited

during a time when displacement on the boundary fault

system was greater, and adjacent basins had also been

established. Subsidence occurred over a broad area and

exceeded sediment supply, thus favouring the establishment

of a deep, anoxic waterbody, possibly similar to modern

Lakes Tanganyika or Malawi. Palaeoclimatic conditions

characterized by precipitation amounts that may have been

as much as five times higher than today (modern mean

annual precipitation is ca. 200 mm) (Roche, 2002; Vincens,

pers. comm.) also contributed to the development of a large,

deep freshwater body that extended into adjacent half-

graben basins, and was favourable to the development of


Pediastrum, Botryococcus and other phytoplankton blooms.

The highest concentrations of algal material accumulated in

sediment-starved parts of the basin. Regional structure,

through its control of regional drainage networks, was

probably partly responsible for sediment starvation, but

dense vegetation cover (humid montane forest) in the

watershed, may also have reduced sediment input to the

basin, and thus dilution of the autochthonous OM.

The Lokichar Basin source rocks are important in

demonstrating the potential for some East African Rift

basins to have generated billions of barrels of oil from

geographically restricted, but thick lacustrine shale deposits,

analogous to the older but extremely prolific rift basins of

southern Sudan and Chad (Genik, 1993; Mohamed, Pearson,

Ashcroft, Iliffe, & Whiteman, 1999; Schull, 1988). In such

cases, tectonic control on source-rock distribution is a major

key, but prevailing palaeoclimatic conditions during the

development of the Lokichar Basin also clearly played a key

role in the generation of such an oil potential.

Acknowledgements

This study represents part of a cooperative program

conducted with the National Oil Corporation of Kenya

(NOCK), and funded by the 3D-3G Project supported by Elf

Petroleum Norge AS (grant No. 2231-01 ELF to J.-J.

Tiercelin) as well as grants from SUCRI-2E and UMR

CNRS 6538 ‘Domaines Oceaniques’, European Institute of

Marine Studies, University of Western Brittany. Research

authorization was provided by the Office of the President of

the Republic of Kenya. Special thanks to Institut Francais du

Petrole (IFP) for their help in the RockEval procedure. The

authors express their thanks to NOCK Managing Director

for permission to publish this work, and to Dr F.M. Mbatau,

NOCK Exploration and Production Manager, for scientific,

administrative and logistic support. Two anonymous

reviewers made a number of helpful suggestions, which

helped improve an earlier version of this paper. Special

thanks to B. Coleno for patient assistance with illustrations.

This is publication No. 150 of the International Decade of

East African Lakes (IDEAL) programme.

References

Barry, S., Filippi, M. L., Talbot, M. R., & Johnson, T. C. (2002).

Sedimentology and chronology of Late Pleistocene and Holocene

sediments from northern Lake Malawi. In E. O. Odada, & D. O. Olago

(Eds.), The East African great lakes: Limnology, palaeoclimatology and

biodiversity (pp. 369–391). Dordrecht: Kluwer Academic Publishers.

Batten, D. J. (1996). Colonial chlorococcales. In J. Jansonius, & D. C. Mc

Gregor (Eds.), Palynology: Principles and applications (pp. 191–203).

American Association of Stratigraphic Palynologists Foundation 1,

Chapter 7C.

Batten, D. J., & Grenfell, H. R. (1996). Botryococcus. In J. Jansonius, &

D. C. Mc Gregor (Eds.), Palynology: Principles and applications

(pp. 205–226). American Association of Stratigraphic Palynologists

Foundation 1, Chapter 7D.

Boschetto, H. B., Brown, F. H., & McDougall, I. M. (1992). Stratigraphy of

the Lothidok Range, northern Kenya, and K/Ar ages of its Miocene

primates. Journal of Human Evolution, 22, 47–71.

Cornford, C. (1990). Source rocks and hydrocarbons of the North Sea. In

K. W. Glennie (Ed.), Introduction to the petroleum geology of the North

Sea (pp. 294–361). Oxford: Blackwell.

Dow, W. G. (1977). Kerogen studies and geological interpretation. Journal

of Geochemical Exploration, 7, 77–99.

Filippi, M. L., & Talbot, M. R. (in press) Lake Malawi: The

palaeolimnology of the North Basin over the last 25 kA based upon

the elemental and stable isotopic composition of sedimentary organic

matter. Quaternary Science Reviews.

Fritsch, F. E. (1961) (Vol. 1). The structure and reproduction of the algae,

London: Cambridge University Press, p. 791.

Genik, G. J. (1993). Petroleum geology of Cretaceous–Tertiary rift basins

in Niger, Chad and Central African Republic. AAPG Bulletin, 77,

1405–1434.

Grall, M. (2000). Le lac Bogoria, Rift Central du Kenya: Reconstitution

paleogeographique et interet petrolier d’un bassin de rift a partir d’une

serie sedimentaire Pleistocene superieur-Holocene. DEA Universite de

Bretagne Occidentale, 49.

Harris, N., Pallister, J. W., & Brown, J. M. (1956). Oil in Uganda.

Geological Survey of Uganda, 9, 33.

Hecky, R. E., & Kling, H. J. (1987). Phytoplankton ecology of the great

lakes in the rift valleys of central Africa. Ergebnisse der Limnologie, 25,

197–228.

Hung, L. V (1996). Hydrocarbon generation assessment of Lokichar Basin,

Kenya. MSc Thesis, Universiti Brunei Darussalam, p. 85.

Hutton, A. C. (1988). The lacustrine Condor oil shale sequence. In A. J.

Fleet, K. Kelts, & M. R. Talbot (Eds.), (Vol. 40) (pp. 329–340).

Lacustrine petroleum source rocks, London: Geological Society

Special Publication.

Katz, B. J. (1995). A survey of rift basin source rocks. In J. J. Lambiase

(Ed.), (Vol. 80) (pp. 213–242). Hydrocarbon habitat in rift basins,

London: Geological Society Special Publication.

Lambiase, J. J., & Bosworth, W. (1995). Structural controls on

sedimentation in continental rifts. In J. J. Lambiase (Ed.), (Vol. 80)

(pp. 117–144). Hydrocarbon habitat in rift basins, London: Geological

Society Special Publication.

Langford, F. F., & Blanc-Valleron, M.-M. (1990). Interpreting Rock-Eval

pyrolysis data using graphs of pyrolizable hydrocarbons vs. total

organic carbon. AAPG Bulletin, 74, 799–804.

Liro, L. M., & Pardus, Y. C. (1990). Seismic facies analysis of fluvial-

deltaic lacustrine systems-Upper Fort Union Formation (Paleocene),

Wind River Basin, Wyoming. In B. J. Katz (Ed.), Lacustrine basin

exploration-case studies and modern analogues (pp. 225–242). AAPG

Memoir 50.

Mohamed, A. Y., Pearson, M. J., Ashcroft, W. A., Iliffe, J. E., & Whiteman,

A. J. (1999). Modeling petroleum generation in the southern Muglad rift

basin, Sudan. AAPG Bulletin, 83, 1943–1964.

Mondeguer, A (1991). Bassins sedimentaires en contexte extensif et

decrochant: L’exemple du ‘Complexe des fosses sud-Tanganyika, Rift

Est-africain’. Morphostructures et sedimentation. These Universite de

Bretagne Occidentale, p. 273.

Morley, C. K. (1999a). Marked along-strike variations in dip of normal

faults—the Lokichar fault, N. Kenya Rift: A possible cause for

metamorphic core complexes. Journal of Structural Geology, 21,

479–492.

Morley, C. K. (1999b). Tectonic inversion in East African rifts. In C. K.

Morley (Ed.), Geoscience of rift systems-evolution of East Africa (Vol.

44) (pp. 193–210). AAPG Studies in Geology.

Morley, C. K. (1999c). Comparison of hydrocarbon prospectivity in

rift systems. In C. K. Morley (Ed.), Geoscience of rift systems-

evolution of East Africa (Vol. 44) (pp. 233–242). AAPG Studies in

Geology.


Morley, C. K., & Wescott, W. A. (1999). Sedimentary environments and

geometry of sedimentary bodies determined from subsurface studies

in East Africa. In C. K. Morley (Ed.), Geoscience of rift systems-

evolution of East Africa (Vol. 44) (pp. 211–231). AAPG Studies in

Geology.

Morley, C. K., Wescott, W. A., Stone, D. M., Harper, R. M., Wigger, S. T.,

& Karanja, F. M. (1992). Tectonic evolution of the northern Kenyan

Rift. Journal of the Geological Society, London, 149, 333–348.

Morley, C. K., Wescott, W. A., Stone, D. M., Harper, R. M., Wigger, S. T.,

Karanja, F. M., & Day, R. A. (1999). Geology and geophysics of the

western Turkana basins. In C. K. Morley (Ed.), Geoscience of rift

systems-evolution of East Africa (Vol. 44) (pp. 19–54). AAPG Studies

in Geology.

Passey, Q. R., Creaney, S., Kulla, J. B., Moretti, F. J., & Stroud, J. D.

(1990). A practical model for organic richness from porosity and

resistivity logs. AAPG Bulletin, 74, 1777–1794.

Peniguel, G., Couderc, R., & Seyve, C. (1989). Les microalgues actuelles et

fossiles. Interet stratigraphique et petrolier. Bulletin des Centres de

Recherches Exploration-Production Elf Aquitaine, 13, 455–482.

Peters, K. E. (1986). Guidelines for evaluating petroleum source rocks

using programmed pyrolysis. AAPG Bulletin, 70, 318–329.

Renaut, R. W., Ego, J., Tiercelin, J.-J., Le Turdu, C., & Owen, B. (1999).

Saline, alkaline palaeolakes of the Tugen Hills-Kerio Valley region,

Kenya Rift Valley. In P. Andrews, & P. Banham (Eds.), Late cenozoic

environments and hominid evolution: A tribute to bill bishop (pp.

41–58). London: Geological Society.

Reynolds, C. S. (1980). Phytoplankton assemblages and their periodicity in

stratifying lake systems. Holartic Ecology, 3, 61–159.

Reynolds, C. S. (1984). The ecology of freshwater phytoplankton,

Cambridge: Cambridge University Press, p. 384.

Roche, H. (2002). Paleoenvironnement des Hominides Mio-Plio-Pleisto-

cenes en Afrique Centrale et Orientale. Colloque ECLIPSE (Envir-

onnement et Climat du Passe: Histoire et Evolution), Paris, October

21–22, 2002.

Schull, T. J. (1988). Rift basins of interior Sudan: Petroleum exploration

and discovery. AAPG Bulletin, 72, 1128–1162.

Simoneit, B. R. T., Aboul-Kassim, T. A. T., & Tiercelin, J.-J. (2000).

Hydrothermal petroleum from lacustrine sedimentary organic matter in

the East African Rift. Applied Geochemistry, 15, 355–368.

Talbot, M. R. (1988). The origins of lacustrine oil source rocks: Evidence

from the lakes of tropical Africa. In A. J. Fleet, K. Kelts, & M. R. Talbot

(Eds.), (Vol. 40) (pp. 29–43). Lacustrine petroleum source rocks,

London: Geological Society Special Publication.

Talbot, M. R., & Lærdal, T. (2000). The Late Pleistocene–Holocene

palaeolimnology of Lake Victoria, East Africa, based upon elemental

and isotopic analyses of sedimentary organic matter. Journal of

Paleolimnology, 23, 141–164.

Talbot, M. R., Lærdal, T., Jensen, N. -B., & Filippi, M. -L. The geochemical

signature of major transgressions in giant lakes. Geochemical et

Cosmochimica Acta, submitted.

Talbot, M. R., Tiercelin, J.-J., Filippi, M.L., Jensen, N.B., Lærdal, T., &

Øvstedal, B. (in preparation). Sedimentary organic matter in the rift

lakes of East Africa: Sources, Accumulation and Palaeo environmental

Significance.

Tiercelin, J.-J., Perinet, G., Le Fournier, J., Bieda, S., & Robert, P. (1982).

Lacs du rift est-africain, exemples de transition eaux douces-eaux

salees: Le lac Bogoria, rift Gregory, Kenya. Memoire Societe

geologique de France, 64, 217–230.

Tiercelin, J.-J., Pflumio, C., Castrec, M., Boulegue, J., Gente, P., Rolet, J.,

Coussement, C., Stetter, K. O., Huber, R., Buku, S., & Mifundu, W.

(1993). Hydrothermal vents in Lake Tanganyika, East African Rift

System. Geology, 21, 499–502.

Tiercelin, J.-J., & Vincens, A. (1987). Le demi-graben de Baringo-Bogoria,

Rift Gregory, Kenya. 30000 ans d’histoire hydrologique et sedimen-

taire. Bulletin des Centres de Recherches Exploration-Production Elf

Aquitaine, 11, 249–540.

Tissot, B., & Espitalie, J. (1975). L’evolution thermique de la matiere

organique des sediments: Applications d’une simulation mathematique.

Revue de l’Institut Francais du Petrole, 5(IFT-75023), 743–777.

Tissot, B. P., & Welte, D. H. (1984). Petroleum formation and occurrence,

Berlin: Springer, p. 699.

Tyson, R. V. (1995). Sedimentary organic matter. Organic facies and

palynofacies, London: Chapman & Hall, p. 615.

Wang, K.-F., Zhang, Y.-L., Wu, G.-X., Li, Y.-Y., He, C., & Zhang, H.-S.

(1994). Study on the thermal simulation for generating oil by

Pediastrum. Journal of Thonghi University, 22(2), 184–191.

Wheildon, J., Morgan, P., Williamson, K. H., Evans, T. R., & Swanberg,

C. A. (1994). Heat flow in the Kenya Rift zone. Tectonophysics, 236,

131–149.

Wood, G., & Miller, M. (1997). Stratigraphic, paleoecologic and petroleum

generating significance of Chlorophyta (chlorococcalean algae) in the

Cretaceous of Western Africa and South America. Africa Geoscience

Review, 4, 499–510.

Zippi, P. A. (1998). Freshwater algae from the Mattagami Formation

(Albian), Ontario: Paleoecology, botanical affinities, and systematic

taxonomy. Micropaleontology, 44, 1–78.


hydrocarbon potential of the meso-cenozoic turkana depression

Documents