Katz, B. J., and M. R. Mello, 2000, Petroleum systems of South Atlantic Marginalbasins—an overview, in M. R. Mello and B. J. Katz, eds., Petroleum systems ofSouth Atlantic margins: AAPG Memoir 73, p. 1–13.

Chapter 1

Petroleum Systems of South Atlantic Marginal Basins—An Overview

Abstract

The marginal basins along the South Atlantic have developed into one of the most active regions forpetroleum exploration. The increase in the level of industry interest has resulted from numerous recentsuccesses along both the eastern and western continental margins of the South Atlantic, the evolution ofthe region’s political character, and an increase in the rate of permitting in deep and ultradeep waters. Thisheightened industry interest provided the rationale for a Hedberg Research Symposium on the petroleumsystems of South Atlantic marginal basins.

Use of the petroleum system concept in South Atlantic marginal basins provides an effective means ofclassifying and characterizing the diversity of the systems and a way to aid in the selection of appropriateexploration analogs. South Atlantic marginal basins provide some of the best examples of how petroleumsystems evolve through time with respect to both their levels of certainty and their areal and stratigraphiclimits. Acomparison of three basins from the South Atlantic—the Niger Delta, Lower Congo, and CamposBasins—provide examples of both the common traits that exist throughout the region as well as the differ-ences among the individual basins. Differences are clear when the source and reservoir couplets are exam-ined. In the Niger Delta, shallow water sands are charged from a Tertiary source with an important higherplant contribution. In the Lower Congo Basin, the lacustrine Bucomazi Formation (Neocomian–Barremian) charges primarily shelfal carbonates and sandstones. In the Campos Basin, the lacustrineLagoa Feia (Barremian) Formation charges primarily Upper Cretaceous–Tertiary deep-water turbiditesandstones. A common trait appears to be the nature of the migration network which typically incorpo-rates both normal faults and regional unconformities. The relative importance of vertical and lateralmigration does differ among the basins, with vertical migration and short-distance lateral migration beingdominant in the Campos and Gabon Basins and longer lateral distance migration being more importantin the Niger Delta.

The basins of the South Atlantic also provide an excellent opportunity to examine the variety of lacus-trine source rock settings. The depositional settings of these lakes range from freshwater to hypersaline.Source quality within these units also varies in response to their different depositional conditions andother factors that control or influence organic productivity and preservation.

INTRODUCTIONOver the past few years, the basins along the South

Atlantic continental margins (Figures 1 and 2) haveundergone a surge of expanded exploratory interest. Thishas in part been fueled by recent giant field discoveriessuch as the Bonga, Zafiro, N’Kossa, Girassol, Dalia, Rosa,and Kuito fields along the west African margin and theRoncador, Marlim, South Marlim, Barracuda, andAlbacora Leste fields in the Campos Basin along the

South American margin. Furthering this growth in activ-ity has been an increase in political stabilization and otherpolitical changes along coastal Africa, denationalizationand an end to the governmental exploration monopoly inthe petroleum industry in Brazil and Argentina, and anoverall increase in the rate of permitting activity in deepand ultradeep waters throughout the region. Thesefactors have culminated in providing the industry with aunique opportunity to expand exploration into one of thelast remaining high-potential frontier regions on earth.

1

B. J. Katz

Texaco Group Inc.Houston, Texas, U.S.A.

M. R. Mello

Petrobrás Research CenterRio de Janeiro, Brazil

As activity has increased, it became clear through thenumerous partnerships and working groups that therewould be universal benefit to review and examine thecurrent state of understanding of the petroleum systemspresent along the South Atlantic margins. Such an exer-cise would potentially provide the following:

1. A mechanism to relate the nature and effectivenessof the petroleum systems along the two conjugatemargins;

2. A series of basin and/or system analogs as explo-ration extends into deeper waters and into basinswhere drilling density has been low;

3. Information on the general character and diversityof largely lacustrine-sourced petroleum systemsthat dominate much of the region’s reserve base;and

4. A better understanding of the presence of marinepetroleum systems in ultradeep waters of westAfrica and possibly the southern Brazilian basins.

To accomplish these goals, a joint AAPG-ABGP ResearchConference on the Petroleum Systems of the SouthAtlantic Margins was organized and held in Rio de

Janiero, Brazil, on November 16–19, 1997. This volumerepresents a collection of papers presented at this meet-ing, with this chapter providing a general overview of themeeting’s themes and conclusions. Most of the details arepresented in the individual contributions.

PETROLEUM SYSTEM CONCEPT

The petroleum system is viewed in various ways bydifferent organizations. Also, like many geoscience terms,although a rigorous definition has been proposed(Magoon, 1988), this definition has not been universallyaccepted or it has been modified to better fit individualorganizational needs and circumstances. However, thereis a common theme among the concept’s numerousproponents. The petroleum system is viewed as a means offormalizing the relationship between the geologicelements in time and space that are required for the devel-opment of a commercial petroleum accumulation. Thekey elements necessary for the presence of a viable petro-leum system are the source rock, reservoir, seal, trap, andnecessary overburden for hydrocarbon generation to

2 Katz and Mello

Foz do AmazonasPara-Maranhao

Barreirinhas

Potiguar

Sergipe/Alagoas

Bahia Sul

Espirito Santo

Campos

Santos

Pelotas

Salado

Colorado

San Jorge

San JulianPiedrabuena

Falkland

Malvinas

Ceara

~`

SOUTH AMERICAJatobaTucano

Reconcavo

Valdez

`

`

^

Figure 1—Distribution of basins along the western SouthAtlantic margin.

Niger Delta

AFRICA

Cameroon

Gabon

Congo

Kwanza

Mocamedes

Walvis

Orange River

Figure 2—Distribution of basins along the eastern SouthAtlantic margin.

proceed. Furthermore, these individual elements mustshare the appropriate temporal and spatial relationshipsto permit hydrocarbons to accumulate and ultimately bepreserved. The temporal, structural, and areal relation-ships among these different elements are often presentedin a series of diagrams displaying the relative timing ofdepositional and other critical events such as trap devel-opment, along with cross-sections establishing strati-graphic and structural relationships among the differentelements and maps defining the geographic limits of thesystem and its components.

The study of petroleum systems requires a paradigmshift from an emphasis on a basin’s sedimentary fill to thefluids that it contains. The examination of a petroleumsystem differs further from a basin study in that it focuseson the hydrocarbons generated by a single source rock.Thus, by definition, a petroleum system is limited to asingle geochemical oil family, although oil properties mayvary as a function of alteration and thermal maturity. Infact, it has been proposed that a petroleum system benamed for the source and dominant reservoir, that is, thereservoir unit in which the majority of the hydrocarbonsare contained. In a basin, multiple oil families or petro-leum systems may be present. One such basin is theOgooué delta, where two distinct petroleum systems arepresent, each with its own source and primary reservoirunits (Katz et al., Chapter 18, this volume). Within a basin,these different systems may overlap both stratigraphi-cally and geographically, but often they are discrete andthe presence of multiple petroleum systems extends boththe stratigraphic and areal extent of a basin’s exploratoryopportunities.

In part, the interest in the use of the petroleum systemconcept as an exploration tool has developed as a directconsequence of the refinement of the geochemical andstatistical tools necessary to establish the genetic relation-ships among oils (Mello et al., Chapter 4, this volume).For example, the combined use of stable isotopes andbiomarker geochemistry has now become common.Geochemical data serve several roles in petroleum systemassessment. First, these indices establish both similaritiesand differences among oils that define individual oilfamilies or systems. Second, when samples of effectivesource rocks are present, they can establish genetic link-ages with oil families. (Source rocks are defined as rocksthat are thermally mature and contain sufficient quanti-ties of the appropriate type of organic matter for petro-leum generation to occur.) Third, the knowledge baseassociated with key geochemical marker (biomarker)compounds has expanded so that the chemical composi-tion of an oil can be used to infer the nature of a sourcerock’s depositional setting and in some cases chrono-stratigraphic position.

Geochemical information also establishes the level ofcertainty that exists for each system. Magoon (1988)suggested three levels of geochemical certainty: known,hypothetical, and speculative. In a known system, a defin-itive geochemical correlation can be established betweensource rock and oil family. In a hypothetical system, thepresence of a source rock system can be established

through its organic richness, generation potential, andkerogen character, but a definitive correlation with oilshas yet to be established. Alternatively, the geochemicalcharacteristics of an oil family can be used to establish thenature and stratigraphic position of a source even whenno correlation has been established. In general, the lack ofdefinitive correlation results either from a lack of sampleavailability or sample quality. In a speculative system, thenecessary supporting data to establish the presence andcharacter of a source are lacking and its presence isinferred through either geologic or geophysical data. Thelevel of certainty of a system can and will be upgraded asaccess to new samples and data are made available.

The defining of a petroleum system along with its levelof certainty can therefore be used to establish theexploratory risks within a region more effectively.Exploration risks increase with increasing geographicand stratigraphic displacement from a known petroleumsystem. This has been clearly shown by Demaison (1984)for the largely vertically migrating petroleum systempresent in the North Sea where exploration success isclosely related to the areal limits of the generative(mature) Kimmeridge Clay.

A caveat, however, does exist in that the limits of apetroleum system are effectively controlled by drillingdensity. Numerous examples can be cited in which thelimits of a petroleum system have been expanded signif-icantly both stratigraphically and areally by additionaldrilling, as has been the case in the Campos Basin(Guardado et al., 1989). In the Campos Basin, initialexploration objectives were Albian carbonates limitedmainly to the outer continental shelf and to water depthsof less than 200 m. Today, these reservoirs account for lessthan 10% of the oil in place, with the largest discoverieshaving been made in Tertiary sandstone reservoirs inwater depths typically greater than 500 m (Figure 3). Inpart, this expansion and redefining of the petroleumsystem has come about as a result of technological devel-opments that have permitted exploration to be extendedinto deeper water. Note that, although the source forhydrocarbons has remained unchanged in the CamposBasin, the primary reservoir and hence petroleum systemname has changed with time. Clearly, this is an examplewhere the petroleum system has grown and evolved in apositive fashion.

At the same time, the lack of a key element or elementsin a speculative petroleum system in either a basin or partof a basin results in exploratory failure. Such a criticalflaw occurred in the Rawson Basin, Argentina (Otis andSchneiderman, Chapter 28, this volume) where drillingfailed to prove the existence of two critical elements in thespeculated petroleum system. Drilling revealed that bothdiscrete reservoir sandstones and organic-rich shaleswere absent. Thus as known and hypothetical systemsmay grow and expand as information becomes available,speculative systems either will evolve into a hypotheticalor known system depending on the nature of data thatbecomes available or will disappear if critical flaws asso-ciated with the absence of key elements are discoveredthrough drilling.

Chapter 1—Petroleum Systems of South Atlantic Marginal Basins—An Overview 3

SOUTH ATLANTIC CONJUGATEMARGINS

The sedimentary basins along the South American andAfrican margins are traditionally considered to be inde-pendent basins. In part, this view has been fostered as aresult of the lack of a common stratigraphic nomenclatureand regional integration. An expanding knowledge base,however, allows both margins and their associated basinsto be viewed as part of a larger single regional, structural,stratigraphic, and geochemical entity, upon which localcharacteristics can be overlain. This commonality amongbasins can be seen best in the geochemical characteristicsof many of the oils. Available data show that althoughseveral oil types exist, there are common themes withrespect to source rock depositional setting and effective-ness (Figure 4).

Such a view leads to a better understanding of the rela-tionships among the different petroleum systems and theappropriateness of a basin or a specific petroleum systemto be used as an analog for the less explored portions ofthe margins, including deep and ultradeep waters. Forexample, this approach demonstrates how the success inthe Lagoa Feia–Carapebus system of the Campos Basin(which includes the Marlim field with 8.2 Bbbl of oil inplace) (Guardado et al., 1989; Mello et al., 1994) can beused to explore more effectively in the more outerportions of the Kwanza and Lower Congo Basins whereexploratory drilling is less mature. However, as Szatmari(Chapter 6, this volume) has pointed out, differences inbasin development are such that the distribution of oilfields along the two margins should not be viewed assimple mirror images. It appears that the differences in

basin evolution have created an asymmetric distributionof hydrocarbon resulting in an apparent alternationbetween productive and nonproductive portions in thetwo margins.

Although the presence of diapiric salt along much ofthe South Atlantic margin has complicated the imagingand analysis of synrift structures, merged geophysicaldata, including both seismic and satellite-derived gravitydata, indicate a common tectonostratigraphic evolutionfor the two conjugate margins. Karner et al. (1997) notedthat the South Atlantic margins formed as a result of riftpropagation across the region during three riftingepisodes: Berriasian–Hauterivian, Hauterivian–middleBarremian, and late Barremian–early Aptian. Each riftingevent resulted in the formation of a series of basins. Freshto brackish to saline lacustrine water conditions devel-oped within these basins. Often these conditions led toboth high levels of organic productivity and preservation.Each of these lacustrine basins was filled with an overallregressive package as a consequence of basin shallowingthrough in-filling.

The last rifting event resulted in the emplacement ofoceanic crust and is consistent with the M0 magneticanomaly along the Brazilian and Angolan margins. Twotectonic hinge zones have been defined along each of themargins. The inner hinge marks the limit of continentalextension. The outer hinge consists of a series ofsegmented en echelon blocks. The en echelon character ofthese blocks plays a key role in controlling river drainageand hence the character of each rift segment’s fill (Karneret al., 1997).

As noted above, the petroleum potential of the differ-ent marginal basins appears to be asymmetric, reflectingthe nature of the South Atlantic’s original rifting (Mello et

4 Katz and Mello

9 11 10 17 27 34 29 32 24 28 15 13 16 22 15 14 14 15 19 30 21 15 13

Number of Exploratory Wells

100

80

60

40

20

0

74 76 78 80 82 84 86 88 90 92 94 96

PER

CEN

T EX

PLO

RA

TORY

WEL

LS

Cretaceous/Tertiary Turbidites

AlbianCarbonates

Rift Carbonates

YEAR

Figure 3—Evolution of primary drilling objectives in the Campos Basin. After Guardado et al. (1997).

al., 1991). The asymmetric character of rifting along themargin is suggested by differences in the width of riftedcontinental crust. For example, the Campos and SantosBasins both display a wide zone of extended continentalcrust, while in the Bahia Sul and Sergipe–Alagoas Basins,the extended crust is restricted to a narrow band. Theconverse appears to be present along the African margin,where the wide zone of extended crust occurs in the northand the narrow band is present in the south. These differ-ences in extent of the extended crust ultimately influencebasin subsidence, basin fill, and thermal history, thusdirectly impacting the distribution and effectiveness ofthe margin’s petroleum systems.

The onset of sea floor spreading was coincident withan apparent increase in salinity. Open marine conditionswere established across the entire region by early Albiantime. Initially, intermittent marine incursions from thesouth resulted in development of hypersaline conditionsand deposition of thick evaporite sequences. These evap-oritic conditions evolved into more normal marine condi-tions with time. During this initial open marine episode,the widespread deposition of carbonates occurred onboth sides of the South Atlantic. From the middleCretaceous onward, the deepening of the South Atlanticresulted in the accumulation of sediments over a widebathymetric range (neritic to abyssal).

When viewed in broad terms, the stratigraphiccolumns of the South Atlantic marginal salt basins appear

similar and can be divided into five megasequences (Figure5): prerift, rift, transitional (or evaporitic), transgressivemarine, and regressive marine. These megasequences arerepresented by time-equivalent successions along the twoconjugate margins representing similar depositional envi-ronments. These five megasequences are separated byfour major regional unconformities: a regional unconfor-mity separating Paleozoic from Mesozoic (mainly UpperJurassic) prerift sedimentary beds; a Lower Cretaceousunconformity between the prerift and rift sequences; apre-Aptian unconformity that corresponds to the“breakup” unconformity; and a lower Tertiary unconfor-mity associated with a Paleocene sea level drop (Henry etal., 1996). The Niger Delta, as a consequence of its differ-ent origin, does not fully share this common tectonostrati-graphic succession (Doust and Omastsola, 1989).

Although the megasequence concept is useful, it isimportant to note that the distribution and geometry ofthe sedimentary fill differs from basin to basin. For exam-ple, along the Brazilian margin, the thickness of the driftsequence varies tremendously. The thickness of thissequence is much greater in the southern basins than inthe northern basins. These differences can also beextended to the west African margin, where rift andprerift sedimentary rocks are exposed in outcrops close toPrecambrian rocks in the Kwanza, Cabinda, and GabonBasins. Such exposures do not exist within the conjugatesalong the southern Brazilian margin. Only in the

Chapter 1—Petroleum Systems of South Atlantic Marginal Basins—An Overview 5

Figure 4—Oil types along the South Atlantic margins.

6 Katz and Mello

regionalvolcanism

BARREIRAS

MA

RITB

A

MO

SQUE

IRO

CALUMBI

RIO DANDE

GRATIDAO

CUNGA

MARINE(REGRESSIVE)

MARINE(TRANSGRESSIVE)

TEBA

N GOLOME

ITOMBE

CABO LEDO

MARINE(CARBONATE)

ARACAJU

SAPUCARI

CO

TIN

TAQUARI

AGUILHADA

RIACHUELOANGIC

O

MARUIM

IBURA TRANSITIONAL (EVAPORITIC)

M.CHAVESPENEDO

BARRADE ITIUBA

COQUEIRO SECO

MACEIO

CARMOPOLIS

SERRARIA

BANANEIRAS

ARACARE

BATINGA

ESTANCIA

RISING SEALEVEL

BASEMENT-INVOLVEDFAULT

LOWERINGSEALEVEL

COMPRESSIONALTECTONICS

VOLCANICEPISODE

BASEMENT-DETACHEDFAULT

PALEOZOIC / PRECAMBRIAN

MAJORUNCONFORMITY

PROGRADATIONEPISODE

SHALE

MARL

SANDSTONE

LIMESTONE

EVAPORITE

SILICICLASTICS

IGNEOUS ROCKS

META SEDIMENTS

SOURCE ROCKS

W. Mohriak, June 1996.

NOVA

M VONELUCULA

LOWERBUCOMAZI

UPPERBUCOMAZI

REDCUVO

GRAYCUVO

MELANIATOCA

CRABE

LOEME

QU

ISSO

ND

E

DO

ND

OTUEN

ZA

PINDA

TEBA

regionalvolcanism

volcanism

TECTONO-STRATIGRAPHIC CHARTS

0QUATERN.

PLIOCENE

PLEIST.

MIO

CEN

E

NEO

GEN

EO

LIG

OC

ENE

EOC

ENE

PALE

OG

ENE

TER

TIA

RY

PALE

OC

ENE

66

SEN

ON

IAN

CENOMANIAN

TURONIAN

AP

TIA

N

BARREM.

CR

ETA

CEO

US

ALB

IAN

NEO

CO

MIA

N

144

180 JUR

AS.

570

TRIAS.

PERM.

CARB.

PRE-CAMBRIAN

QUATERN.

PLIOCENE

PLEIST.

MIO

CEN

E

NEO

GEN

E

OLI

GO

CEN

EEO

CEN

E

PALE

OG

ENE

TER

TIA

RY

PALE

OC

ENE

SEN

ON

IAN

CENOMANIAN

TURONIAN

AP

TIA

N

BARREM.

CR

ETA

CEO

US

ALB

IAN

NEO

CO

MIA

N

JUR

AS.

TRIAS.

PERM.

CARB.

PRE-CAMBRIAN

SERGIPE–ALAGOAS AND KWANZA BASINSMa

volcanism

Figure 5—Comparison of the tectonostratigraphic frameworks of the Sergipe–Alagoas and Kwanza Basins.

Sergipe–Algoas and Recôncavo Basins can the prerift andrift sequence be mapped. These differences are partially areflection of the previously noted asymmetric riftingalong the margins (Mello et al., 1991).

SYSTEM DIVERSITY AND EFFECTIVENESS

The petroleum systems in the South Atlantic arehighly diverse in character. This diversity is manifested instructural style, source rock depositional setting, reservoirlithology, and timing of fundamental events. The sixdistinct oil types present along the margin partially reflectthe diversity of the petroleum systems. Each oil typereflects a unique source rock depositional environmentthat developed as the margins evolved. It is important tonote, however, that individual oil accumulations mayform through the mixing of various oil types. In strati-graphic succession (oldest first), the six oil types arebriefly described here, following Mello et al. (1996).

Oil Types

The first oil type observed in the Ceará, Potiguar,Sergipe–Alagoas, Bahia Sul, Douala, Lower Congo, andGabon Basins was derived from Neocomian source rocksdeposited under lacustrine freshwater to brackish condi-tions (e.g., Pendência, Candeias, Bucomazi, and MelaniaFormations in the Potiguar, Bahia Sul, Lower Congo, andGabon Basins, respectively). Much of this oil appears tobe contained within Neocomian sandstones. Lateral andvertical migration distances appear limited. These oils areparaffinic (saturates >60%, paraffins >40%) and have lowsulfur (<0.1%) and low naphthene contents and lowgas–oil ratios (<60 m3/m3).

The second oil type, which is observed in the Santos,Campos, Espirito Santo, and Sergipe–Alagoas Basinsand along much of west Africa from Angola toCameroon, was derived from Neocomian–Barremiancalcareous black shales deposited under brackish tosaline conditions (e.g., Lagoa Feia, Bucomazi, andMelania Formations in the Campos, Lower Congo, andGabon Basins, respectively). These oils are often reser-voired in turbiditic sandstones ranging from UpperCretaceous to Miocene and to a lesser degree inCretaceous carbonates. Lateral migration distances awayfrom the generative “kitchen” appear limited. Verticalmigration associated with normal fault systems andthrough windows in the overlying salt may, however, besignificant. These oils are naphthenic (saturates >60%,naphthenes >40%) and have low to medium sulfurcontents (0.1–0.5%) and low to medium gas–oil ratios(<200 m3/m3).

The third distinct oil type identified in the Ceará,Potiguar, Sergipe–Alagoas, Bahia Sul, and Gabon Basinsis derived from source rocks deposited under marineevaporitic conditions (e.g., the Paracuru, Upanema,

Muribeca, and Gamba Formations in the Ceará, Potiguar,Sergipe–Alagoas, and Gabon Basins, respectively). Thelimited occurrence of this oil type along the west Africanmargin is probably due to the margin’s more limitedvolume of the evaporitic oil-prone source facies. Alongthe west African margin, this stratigraphic interval isdominated by shallow-water carbonates. Reservoirs forthese oils range in age from late Aptian to early Eoceneand are predominantly delta-front sandstones. These oilsmigrated vertically within the normal fault system andlaterally along regional unconformities. They are naph-thenic (saturates >60%, naphthenes >40%), with mediumsulfur contents (0.1–0.5%) and low to medium gas–oilratios (<200 m3/m3).

The fourth oil type is associated with accumulations inthe Sergipe–Alagoas, Bahia Sul, Espirito Santo, LowerCongo, Kwanza, and Gabon Basins. These oils arederived from Albian–Aptian marls and calcareous blackshales (e.g., Regencia and Iabe Formations of the EspiritoSanto and Lower Congo Basins, respectively). The migra-tion pathways of these oils appear torturous due tocombined components of both vertical and lateral migra-tion associated with transform fault systems. These oilsare largely pooled in Upper Cretaceous shallow-watercarbonates and sandstones. They are classified as naph-thenic-aromatic (saturates <50%, naphthenes >25%), andthey contain moderate quantities of sulfur (0.1–0.5 %) anddisplay low gas–oil ratios (<60 m3/m3).

The fifth oil type is present in the Santos, EspiritoSanto, and Sergipe–Alagoas Basins along the Brazilianmargin and in Gabon, Lower Congo, and Kwanza Basinsalong the west African margin. These oils can be corre-lated to Cenomanian–Turonian marine black shales (e.g.,Itajai Formation of the Santos Basin and the Azile andAnguille Formations of the Gabon Basin). They are foundin reservoirs ranging from Albian carbonates toEocene–Oligocene deep-water turbidites. They migrateprincipally along regional unconformities. These oils arecharacterized by their moderate sulfur (0.1–0.5%) andnaphthene contents and by high gas–oil ratios (>200m3/m3). In some basins along the Brazilian margin, thelack of significant accumulations of these oils appears toresult from a lack of necessary overburden for generationto have occurred. Data indicate, however, that in somebasins the viability of this potential petroleum system isenhanced and has not yet been tested by the limiteddrilling.

The sixth oil type has been recovered from the EspiritoSanto and Lower Congo Basins and the Niger Delta. Itcan be correlated with rocks deposited within lowerTertiary marine deltaic deposits (e.g., UrucutucaFormation of the Espirito Santo Basin and the AkataFormation of the Niger Delta). These oils are pooled inEocene and younger sandstone reservoirs. Hydrocarbonmigration involves both vertical and lateral componentsthat incorporate both growth fault systems and regionalunconformities. These oils are paraffinic (saturates >60%,paraffins >40%) and have low sulfur contents (<0.1%)and high gas–oil ratios (>200 m3/m3).

Chapter 1—Petroleum Systems of South Atlantic Marginal Basins—An Overview 7

The nature and magnitude of South Atlantic petro-leum system diversity can also be observed through asimple and brief comparison of the petroleum systemspresent in three of the South Atlantic basins: the NigerDelta, Lower Congo, and Campos Basins. Even before thedetails of the geology are examined, differences in thesesystems become apparent when field size distributionsare examined (Figure 6a). For example, within the NigerDelta, the USGS (Klett et al., 1997) has estimated oilreserves of approximately 35 billion barrels contained in491 oil fields, with a modal field size between 16 and 32million barrels of oil equivalent. In contrast, they esti-mated current reserves in the Campos Basin at approxi-mately 10 billion barrels of oil in 60 oil fields, with amodal field size between 128 and 256 million barrels of oilequivalent (Figure 6b).

Niger Delta

The Niger Delta began its development during thelate Paleocene–Eocene as sediments progaded beyondthe horst and graben associated with the breakupbetween Africa and South America (Doust andOmatsola, 1989). On the basis of newly generatedgeochemical data obtained on both oils and rocks, Haacket al. (Chapter 16, this volume) report the presence ofthree petroleum systems within the Niger Delta: a LowerCretaceous lacustrine derived system, an UpperCretaceous–lower Paleocene derived system, and aTertiary (late Eocene–Pliocene) derived system. It is theTertiary-sourced petroleum system that dominates thedelta’s reserve base. The Niger Delta oils display affini-ties typical of a mixed marine–terrestrial source rocksystem consistent with the recently identified sourcerocks. It is becoming apparent that this dominance by aTertiary source rock is atypical among the South Atlanticbasins, where Cretaceous lacustrine and marine sourcerocks are the general rule (Schiefelbein et al., Chapter 2,this volume).

Within the delta, all current hydrocarbon production isfrom the growth section above the regional dêcollement.The primary reservoirs are from the Oligocene–Mioceneportions of the Agbada Formation, which is composed ofa series of interbedded shallow marine and fluvial sand-stones, siltstones, and claystones typical of most paralicsettings. Barrier bar sandstones are typically more later-ally continuous than those deposited in distributarychannels. In general, the section is so sand rich that thereservoir is often considered a secondary risk to the seal(Doust and Omastsola, 1989).

The depositional setting of the Niger Delta reservoir isalso different from the other two basins to be examined,which favor either deeper water clastics (Campos Basin)or carbonate reservoirs (Lower Congo Basin). Most of thehydrocarbon traps are structural and developed as aresult of synsedimentary deformation, such as rolloveranticlines associated with growth faults (Doust andOmatsola, 1989). Hydrocarbon generation is time trans-gressive across the delta as a result of the delta’s progra-dation and appears to postdate the synsedimentaryfaulting. It also appears that hydrocarbon generation hasbeen episodic, possibly as a result of depositional lobeswitching. Petroleum generation is thought to havebegun as early as late Eocene–early Oligocene in themost proximal parts of the delta (Ekweozor andDaukoru, 1994) and is still actively proceeding today.Hydrocarbon migration within the basin appears to belateral along regional unconformities. Growth faults inthe basin result in the redistribution of hydrocarbons,creating the common occurrence of multipay fields(Demaison and Huizinga, 1994).

Lower Congo Basin

The Lower Congo Basin began with Early Cretaceousrifting, and unlike the Niger Delta, its early rifting history

8 Katz and Mello

10

8

4

2

0

Nu

mb

er o

f Fi

eld

s

6

12

14

16

Field Size (MMBOE)

2,04

8 to

< 4

,096

1,02

4 to

< 2

,048

512

< 1,

024

256

< 51

2

128

to <

256

64 to

< 1

28

32 to

< 6

4

16 to

< 3

2

8 to

< 1

6

4 to

< 8

2 to

< 4

1 to

< 2

Gas

Oil

80

60

40

20

0

Nu

mb

er o

f Fi

eld

s

100

120

Field Size (MMBOE)

2,04

8 to

< 4

,096

1,02

4 to

< 2

,048

512

< 1,

024

256

< 51

2

128

to <

256

64 to

< 1

28

32 to

< 6

4

16 to

< 3

2

8 to

< 1

6

4 to

< 8

2 to

< 4

1 to

< 2

Gas

Oil

A

B

NIGER DELTA

CAMPOS BASIN

Figure 6—Field size distributions of (a) the Niger Delta and(b) the Campos Basin.

(a)

(b)

has played a major role in controlling the distribution ofhydrocarbons in the basin. The initial sediments werealluvial, fluvial, and lacustrine. Postrifting subsidenceresulted in a marine incursion and deposition of theLoeme Evaporites. Following evaporite deposition, openmarine conditions developed, leading first to the deposi-tion of extensive carbonates that eventually evolved intoa clastic-dominated sequence. Postrift structuring hasbeen controlled by both salt tectonics and faulting withinthe Tertiary sequence.

As in the Niger Delta, three potential petroleumsystems have been identified in the Lower Congo Basin:a Lower Cretaceous Bucomazi Formation sourcedsystem, an Upper Cretaceous Iabe Formation sourcedsystem, and Tertiary Malembo Formation derivedsystem (Cole et al., Chapter 23, this volume; Schoellkopfand Patterson, Chapter 25, this volume). The systemassociated with the lacustrine Bucomazi Formationdominates in the basin. Bucomazi-derived oils arepresent in both presalt reservoirs (e.g., sandstones of theLucala and Erva Formations and the Toca carbonates)and postsalt reservoirs (e.g., Pinda carbonates, shelf andbeach sandstones of the Vermelha and Iabe Formations,and deep-water sandstones of the Malembo Formation).The Bucomazi Formation developed in synrift lakes.Burwood et al. (1995) noted that the source rock attri-butes of the Bucomazi are highly varied with respect toboth quality and character. For example, Burwood notedthat the basin fill portion of the Bucomazi Formationdisplays generally higher hydrocarbon generationpotentials and different generation kinetics than theoverlying sheet drape portion. In part, these differencesappear to reflect changes in lake salinity and oxygencontent through time. Hydrocarbon generation in theBucomazi began during the Late Cretaceous and contin-ues to the present day.

Unlike the Niger Delta, two other identified petroleumsystems significantly contribute to the Lower CongoBasin’s reserves. The outer shelf and slope facies of theIabe Formation have also fed multiple reservoirs, includ-ing both the Pinda and Malembo Formations. Theorganic carbon content of the Iabe Formation averages2–3%, ranging upward to ~5%. In the Iabe Formation,there is a general upward increase in source qualityreflecting the transgressive character of the unit. Often,fields that have been charged from the Iabe have alsoreceived some contribution from the deeper lacustrineBucomazi Formation. Generation from the Iabe beganduring the middle Miocene and continues to the present.

The source potential of the Malembo Formation islimited, with maximum organic carbon contentsapproaching about 4%. The limited source quality in theunit decreases up-section as a result of the regressivecharacter of the unit. Generation from the MalemboFormation also appears to be restricted areally to areaswhere a thick Tertiary depocenter exists. As with the Iabe-derived oils, commercial accumulations that includeMalembo-derived oils have also received a contributionfrom the Bucomazi Formation.

Campos Basin

As for the Lower Congo Basin, the Campos Basin alsobegan with Early Cretaceous rifting and has a strati-graphic succession very similar to that of the LowerCongo, including an evaporitic facies in the upper part ofthe Lagoa Feia Formation, a carbonate sequence,followed by an open marine clastic sequence. Similarly,the basins share a similar postrifting structural historyincorporating both salt tectonics and faulting. Theamount of postrifting subsidence and sedimentation isless significant in the Campos than in the Lower CongoBasin.

The Campos Basin contains only a single active petro-leum system. Although organic-rich Cenomanian–Turonian blacks shales are present, all of the known oilsare derived from the synrift, brackish to saline, lacustrineblack shales of the Barremian Lagoa Feia Formation. Theorganic carbon content of the Lagoa Feia Formationranges typically from 2 to 6%, with values as high as 9%.Hydrocarbon generation potential typically exceeds 25kg/ton rock and often exceeds 40 kg/ton. The geochem-ical attributes of the Lagoa Feia are also quite diverse,reflecting changes in depositional conditions such as anupward increase in lake salinity caused by intermittentmarine incursions (Mello and Hessel, 1998).

Hydrocarbon generation from the Lagoa Feia beganduring the Late Cretaceous, but its apex was during thelate Miocene and it continues to the present. This rela-tively late onset of hydrocarbon generation in compari-son to the Lower Congo Basin is largely the result ofdifferences in the amount of postrifting overburden thatthe two margins have received. The more limited amountof postrifting sediment also appears to explain the lack ofeffectiveness of the Cenomanian–Turonian source rocks.The Campos Basin oil reservoirs range in age fromNeocomian to Miocene. Unlike the Lower Congo Basin,most of the Campos Basin reserves are located higher inthe stratigraphic sequence in the Upper Cretaceous andOligocene–Miocene deep-water, turbiditic sandstones.Migration is largely vertical through salt windows andalong unconformities and faults.

Systems Summary

These three petroleum systems clearly reflect thedifferent tectonostratigraphic positions of the effectivesource and primary reservoirs within the three basins.These three basins also show variations in system effec-tiveness that may exist mainly because of differences inthermal maturity resulting from variations in the amountof overburden. From this assessment, it is clear that,although the basins share many characteristics such asgeneral tectonic and sedimentologic histories, local differ-ences dominate the geologic details and control hydro-carbon generation and migration mechanisms. Thesedifferences must be clearly understood prior to attempt-ing to extend an identified petroleum system or using oneas an exploration analog.

Chapter 1—Petroleum Systems of South Atlantic Marginal Basins—An Overview 9

SOUTH ATLANTIC LACUSTRINESYSTEMS

Much of the available geologic and geochemical dataindicate that a significant portion of the region’s crudeoil has been derived from lacustrine source rocks.Current estimates suggest that as much as 95% ofdiscovered Brazilian oil and at least 15% of west Africanoil can be related to lacustrine source rocks. It should benoted that the west African reserves are heavily domi-nated by the Tertiary marine and deltaic sequence of theNiger Delta.

Although much of the known and potential futureSouth Atlantic oils are derived from lacustrine sourcerocks, they display significant variability. These differ-ences are largely the result of differences in lake watersalinity (Figure 7) caused by a combination of factorssuch as climate changes and the magnitude and extentof marine incursions. Lake water salinity impacts thelevel and type of organic productivity, as well as itspreservation and early diagenesis. These differences arereflected in the molecular chemistry of the source rocksystems.

In addition to differences in oil chemistry and associ-ated molecular chemistry in the source rocks, their bulkchemistry varies, reflecting their organic richness, oilproneness, and hydrocarbon generation kinetics.Available data indicate a long-term trend and oscillatorycycling of facies (Figures 8 and 9). The long-term trend istoward improved quality up-section, which maximizes

prior to the development of lake-fill or drape facies. Thistrend in clearly shown by the Lagoa Feia Formation in theCampos Basin (Figure 8). The oscillatory changes are wellrepresented in the Kwanza Basin (Figure 9). These cycles,which represent expansions and contractions of the lakebody, most likely reflect climatic variations directlyimpacting water level and chemistry, as well as theamount of both organic and inorganic terrestrial input.Lake level low stands reflect intervals of poorer sourcequality largely due to poorer organic preservation.

Further examination of available data suggests that theareal distribution of oil-prone source material from salineversus freshwater lakes differs. The freshwater lakesappear to have been mainly restricted geographically tobasin deeps. In contrast, the saline lakes tended to be shal-lower and areally more extensive, their distribution beingstrongly influenced by intermittent transgressions ofmarine waters from the south. These transgressions alsointroduced fresh nutrients into the systems, resulting isblooms of cyanobacteria. Such controls on lake characterexplain why the saline water oil type is volumetricallymore important than the freshwater oil type and thusmust be considered the dominant petroleum system inultradeep water.

The complex anatomy of these synrift lacustrinesystems has become apparent as additional geochemicaland stratigraphic data become available. These data,however, reveal some consistent trends and patterns thatcan be effectively translated to less explored parts of theregion as well as to other similar settings from elsewherearound the world.

10 Katz and Mello

Lacustrinefresh tobrackish

Lacustrinefresh tosaline

Marineevaporitic

Marinedrift

Marinedeltaic

24-nor/24-nor + nor dia + regular cholestanesC30

-Din

ost

era

ne

s/C

30-D

ino

ste

ran

es

+ C

29ααα

R-S

tera

ne

s

E. Cretaceous Oil - Campos BasinE. Cretaceous Oil - Camamu BasinE. Cretaceous Oil - CabindaLate Cretaceous Oil - Santos BasinLate Cretaceous Oil - Kwanza BasinTertiary Oil - Niger DeltaTertiary Oil - Offshore - BrazilAptian - Oil Ceara’ BasinAlbian - Oil - Foz do Amazonas’ Basin

60504030201000

10

20

30

40

50

60

70

80

90

Figure 7—Geochemical differentiation of South Atlantic oil types.

CONCLUSIONSThe petroleum system concept appears to be an effec-

tive tool in evaluating the remaining exploration poten-tial of the South Atlantic. Such a multidisciplinaryapproach, focusing on the nature and distribution ofhydrocarbon fluids, places the discovered hydrocarbonsin a clear framework and provides a potential road mapfor future exploration. Available data reveal a generalsimilarity between the South American and westAfrican marginal basins with respect to their deposi-tional sequences, including source rock facies and

consequently oil types. Asymmetric rifting, however,has resulted in different sedimentary and subsidencehistories, which has, in turn, created major differences inthe distribution of oil types along the margins. As aresult, the Brazilian marginal basins are dominated bylacustrine oils, whereas the west African margin isdominated by marine oils. Although marine sourcerocks are present in Brazilian marginal basins, theiroverburden is insufficient for generation to proceed. Inconstrast, sufficient overburden is present, at leastlocally, along the west African margin for generation totake place.

Chapter 1—Petroleum Systems of South Atlantic Marginal Basins—An Overview 11

Figure 8—Geochemical log of the lacustrine sequence of the Kwanza Basin.

Consequently, when viewing the petroleum systemsof the South Atlantic as possible exploration analogs, caremust be taken. For example, this overview suggests that,although the deep-water success of the Brazilian CamposBasin associated with the Lagoa Feia–Carapebus systemmay be satisfactorily transferred to the west Africanmargin, it does not appear that the Iabe–Pinda system ofthe Lower Congo will translate to the South Americanbasins. This study also indicates that the synrift lacustrinefacies is heterogeneous, reflecting differences in climate,water chemistry, and lake basin maturity. Yet, there do

appear to be regular patterns that may be useful in theless well explored and documented parts of the SouthAtlantic as well as in other tropical to subtropical lacus-trine settings.

Acknowledgments—The authors wish to thank their respec-tive managements for their permission to publish this introduc-tory overview. Financial support for the joint AAPG/ABGPResearch Conference on the Petroleum Systems of the SouthAtlantic Margins was provided by Petrobras, Texaco, Chevron,and Exxon.

12 Katz and Mello

Figure 9—Geochemical log of the lacustrine sequence of the Campos Basin.

REFERENCES CITED

Burwood, R., S. M. De Witte, B. Mycke, and J. Paulet, 1995,Petroleum geochemical characterisation of the LowerCongo coastal basin Bucomazi Formation, in B. J. Katz,ed., Petroleum source rocks: Berlin, Springer-Verlag,p. 235–263.

Demaison, G., 1984, The generative basin concept, in G.Demaison and R. J. Murris, eds., Petroleum geochemistryand basin evaluation: AAPG Memoir 35, p. 1–14.

Demaison, G., and B. J. Huizinga, 1994. Genetic classificationof petroleum systems using three factors: charge, migra-tion, and entrapment, in L. B. Magoon and W. G. Dow,eds., The petroleum system—from source to trap: AAPGMemoir 60, p. 73–89.

Doust, H., and E. Omatsola, 1989, Niger Delta, in J. D.Edwards and P. A. Santogrossi, eds., Divergent/passivemargin basins: AAPG Memoir 48, p. 201–238.

Ekweozor, C. M., and E. M. Daukoru, 1994, Northern deltadepobelt portion of the Akata–Agbada(!) petroleumsystem, Niger Delta, Nigeria, in L. B. Magoon and W. G.Dow, eds., The petroleum system—from source to trap:AAPG Memoir 60, p. 599–613.

Guardado, L. R., L. A. P. Gamboa, and C. F. Lucchesi, 1989,Petroleum geology of the Campos basin, Brazil, a modelfor a producing Atlantic type basin, in J. D. Edwards andP. A. Santogrossi, eds., Divergent/passive margin basins:AAPG Memoir 48, p. 3–79.

Guardado, L. R., B. Wolff, and J. A. S. L. Brandao, 1997,Campos basin, Brazil, a model for producing Atlantic typebasins: Proceedings of the 29th Annual OffshoreTechnology Conference, v. 3, p. 457–462.

Henry, S. G., W. U. Mohriak, and M. R. Mello, 1996, SouthAtlantic sag basins: new petroleum system components(abs.): AAPG Bulletin, v. 80, p. 1300.

Karner, G. D., N. W. Driscoll, J. P. McGinnis, W. D.Brumbaugh, and N. R. Cameron, 1997, Tectonic signifi-cance of syn-rift sediment packages across theGabon–Cabinda continental margin: Marine andPetroleum Geology, v. 14, p. 973–1000.

Klett, T. R., T. S. Ahlbrandt, J. W. Schmoker, and G. L. Dolton,1997, Ranking of the world’s oil and gas provinces byknown petroleum volumes: USGS Open-File Report 97-463, 207p.

Magoon, L. B., 1988, The petroleum system—a classificationscheme for research, exploration, and resource assessment,in L. B. Magoon, ed., Petroleum systems of the UnitedStates: USGS Bulletin, v. 1870, p. 2–15.

Mello, M. R., W. U. Mohriak, E. A. M. Koutsoukos, and J. C.A. Figueira, 1991, Brazilian and west African oils: genera-tion, migration, accumulation, and correlation:Proceedings of the 13th World Petroleum Congress, v. 2,153–164.

Mello, M. R., E. A. M. Koutsoukos, W. U. Mohriak, and G.Bacoccoli, 1994, Selected petroleum systems of Brazil, inL. B. Magoon and W. G. Dow, eds., The PetroleumSystem—from Source to Trap: AAPG Memoir 60, p.499–512.

Mello, M. R., E. A. M. Koutsoukos, W. U. Mohriak, and G.Bacoccoli, 1996. Petroleum systems of the Brazilian SouthAtlantic Margin (abs.): AAPG Bulletin, v. 80, p. 1314.

Mello, M. R., and M. H. Hessel, 1998, Biological marker andpaleozoological characterization of the early marine incur-sion in the lacustrine sequences of the Campos Basin,Brazil (abs.): AAPG Annual Convention, ExtendedAbstracts, v. 2, p. A455.

Chapter 1—Petroleum Systems of South Atlantic Marginal Basins—An Overview 13