Oil and gas shows numerous in Nama basin, southern Africa

Malcolm Walter
Rix & Walter Pty. Ltd.
Bawley Point, N.S.W., Australia

Roger Swart
National Petroleum Corp. of Namibia (Pty.) Ltd.
Windhoek

Roger Summons
Australian Geological Survey Organisation
Canberra

The Nama basin occupies 350,000 sq km of southern Namibia, western Botswana, and northwestern South Africa¹ (Fig. 1 [86069 bytes]) and contains 10 km or more of Neoproterozoic and Cambrian sediment. National Petroleum Corp. of Namibia recently embarked on an assessment of the prospectivity of the basin, and here we report some of the first results.

Nama basin study

The Nama Group was deposited in a sag and then foreland basin setting that resulted from the opening and closing of the Damara orogen to the north and the Adamastor Ocean (now the Gariep orogen) to the west.² Locally exposed rock successions³ record older extensional terrains that underlie the Nama basin. One such is the Duruchaus formation (Fig. 2 [12811 bytes]), which includes soda-evaporites believed to have been deposited in an early rift environment.⁴

In southernmost Namibia the immediately pre-Nama succession includes two Neoproterozoic glacial units. In Botswana the Nosop sub-basin⁵ contains at least 10 km of sediment (at least 9 km of which is Proterozoic-Cambrian). Little of this is known to outcrop, and only 3 km of the Proterozoic has been penetrated by the drill.

Aeromagnetic depth-to-basement interpretations suggest that much of this must extend into South Africa and Namibia. These poorly known sub-basins are imaged in deep seismic records in Botswana⁵ and near the Vreda-1 well in Namibia.

The Nama succession is folded on its margins but away from the edges is nearly flat lying. Broad domal structures occur locally (Fig. 3 [219395 bytes]). Much of the Nama Group is now covered by the late Paleozoic-Mesozoic coal-bearing Karoo Group (with dolerite sills) and Tertiary-Recent Kalahari Group sediments. Away from the tectonized northern and western margins of the basin the thermal maturity of the basin seems to be within the oil and gas windows, apparently with much local variation resulting from the intrusion of mafic sills and dikes during the Mesozoic.

The lower Nama Group (Kuibis and Schwarzrand Subgroups) contains an "Ediacara fauna" of "soft-bodied" metazoans which are characteristic of the "Terminal" Proterozoic, and these units were derived largely from the stable craton to the east. Saylor et al.⁶ identified seven depositional sequences (designated K1-K2 and S1-S5).

The oldest two sequences (K1 and K2) are predominantly shallow marine deposits and have coarse siliciclastic bases grading up into carbonates. The third and fourth sequences (S1 and S2) comprise shale, siltstones, and sandstones deposited in an offshore environment. Deltaic sandstones are incised into the tops of both sequences.

The following two sequences (S3 and S4) are largely platform carbonate deposits. The base of the final sequence (S5) is a regional network of incised valleys with a relief of up to 30 m. This sequence is interpreted as recording a shallowing upward environment.

The first appearance of the Ediacara fauna lies above the base of the K2 sequence. The ichnofossil Phycodes pedum, the first appearance of which is taken to indicate the base of the Cambrian, first occurs above the base of the Nomtsas formation (base of S5; Fig. 2 [12811 bytes]). The overlying fluvial Fish River Subgroup comprises red, braided fluvial molasse deposits derived from the developing Damara and Gariep orogens. Initial sediment transport was from the north and west for the lower Fish River, but only from the north for the upper portion.

The maximum age of the succession (including the pre-Nama rift units) of 717±11 Ma is given by the underlying Gannakouriep dyke swarm,⁷ which does not cut the Neoproterozoic sediments. Recent U-Pb ages have better constrained the age of the succession as a whole,⁸ and some of these are given in Fig. 2.

Prior to our study, sampling of potential source rocks was very limited, being concentrated on the unsuccessful exploration wells Masetlheng Pan-1, Tses-1, and Vreda-1. As part of our assessment of petroleum prospectivity we have completed an extensive program of surface sampling of potential source rocks and have re-sampled all available well material. The samples are being screened by Rock-Eval/TOC.

No unambiguously effective or spent source rocks have been identified as yet, although one sample of dark gray limestone contained 2.2% TOC. Some samples in the Masetlheng-Pan 1 well had significant TOC contents, but there were also indications that these may have contained contaminants from drilling fluids. Further work is under way to check these results. Shallow drilling recently commenced will provide samples of black shales that are very extensively developed in the basin but which are too weathered to be sampled in outcrop.

Previous drilling

Four unsuccessful wells have been drilled into the Nama Group (Fig. 1), the first in 1927-28. None penetrated the full succession.

There are numerous petroleum shows. Previously only three have been substantially documented:⁹ solid bitumen in fissures in the Fish River sandstone, live oil in a fault breccia on farm Heigums 105 in Kuibis sandstone, and live oil in fluid inclusions in quartz crystals in veins on farm Geiaus 6.

Using reports from early this century and with advice from W. Hegenberger of the Geological Survey of Namibia, we have documented and sampled bitumen veins at several widely separated sites in the Fish River sandstone and have re-sampled the sites on Heigums 105 and Geiaus 6. Hydrocarbon-bearing quartz crystals from the latter locality have also been obtained from local collectors.

Additional reported shows are as follows: Altebaumer & Altebaumer¹⁰ note the occurrence of bitumen and "coke-like" material from the Fish River Subgroup in Vreda-1; these were not visible in the core stored in Windhoek, but portions of the core are missing. Minor methane flows were reported in Tses-1. Miller¹¹ reports "bituminous material" at a depth of 700 ft in a water borehole on the farm Plattfontein 92 southwest of Maltah"he (this is based on word-of-mouth reports from local farmers). With a reported gas blowout in Berseba-1 in 1928, this amounts to at least nine shows spread over a distance of about 400 km.

The bitumens were extracted and studied using biomarker methodology.

A number of quartz and calcite crystal samples were subjected to bitumen and fluid inclusion analysis. Bitumens of two types were observed. Some were obvious accumulations visible to the naked eye. Others were only evident on microscopic analysis as black deposits along crystal boundaries. Hydrocarbons released from the crystals by thermal evaporation at 100-300° C. had a very similar pattern to hydrocarbons in the massive vein bitumens in that there was low Pr/Ph, abundant acyclic isoprenoids, and even carbon number preference in n-alkanes.

The most consistent, reliable, and informative results were obtained by solvent extraction and detailed analysis of three samples of massive bitumen present as large veins in sandstone and siltstone, and a sample of live oil from a fault breccia. These were studied by conventional and state-of-the-art biomarker methodology. Bulk compositional analyses showed that they contained variable amounts of asphaltene (approximately 0.5-25%), and this variation probably correlates with their relative degrees of biodegradation. A paraffinic oil from farm Heigums 105 is less altered (biodegraded) than the solid bitumens and is composed of 79% saturated, 11% aromatic, 10% NSO components. Isotope data show comparatively heavy (13C-enriched) signatures. Individual compound analysis (see below) suggests that the different signatures of the solid bitumens as compared to the live oil are significant and reflect subtle source differences. These isotopic data suggested that the present set of samples contained the same oil type as occurs in fluid inclusions in quartz crystals on Farm Geiaus 6 that was reported by Kvenvolden and Roedder.¹² This was further supported by those authors' GC analysis showing low Pr/Ph and abundant acyclic isoprenoids C20, as for our samples from elsewhere in the basin.

Gas chromatography of saturated hydrocarbons showed n-alkanes (depleted by biodegradation), abundant acyclic isoprenoids, and a high molecular weight hydrocarbon peak identified as perhydro-b-carotene. A more altered sample contained no n-alkanes, low isoprenoids, and an enhanced 'hump' of unresolved components. Gammacerane and perhydro-b-carotene were prominent in the GC profile. GC-MS confirmed the pattern of biodegradation in these samples, and it was concluded that further interpretation of maturity and source parameters would be compromised by the affects of alteration, so another approach was adopted. Hydrous pyrolysis of the asphaltene and polar fractions of oil seeps and residual bitumen (but not the whole bitumen) is an established method to regenerate an original biomarker signature of a precursor oil.^13-14 The results show that the aliphatic hydrocarbons so produced are very similar in chemical composition to those in a pristine sample. Accordingly, the asphaltene or asphaltene+polar fractions of four of our samples were subject to hydrous pyrolysis and the products separated into saturated, aromatic, and polar fractions as if they were a conventional oil.

Biomarker and isotopic analyses of these four 'reconstructed' oils show very similar but not identical chemical compositions. The GC traces (Fig. 4 [31364 bytes]) show that all samples are characterized by n-alkanes with a pronounced even over odd carbon number predominance, very abundant acyclic isoprenoids ( C20) with low Pr/Ph ratio and low abundances of isoalkanes and waxy hydrocarbons. GC-MS data for steranes (Fig. 5 [35591 bytes]) for all samples are similar with similar abundances of C27 and C29 steranes indicating algal organic matter and a low ratio of diasteranes to steranes reflecting a non-clastic source lithology. C30 desmethylsteranes were absent with 4-methylsteranes (dinosteranes and 4-methylstigmastanes; data not shown) only present as trace components. These data suggest a nonmarine source environment. The ratios of 20S to 20R aaa steranes and abb to aaa steranes suggest that the oils were low to moderate maturity and that there may be maturity differences between the samples.

The triterpane patterns of all samples (Fig. 6 [32642 bytes]) had abundant gammacerane and 3b-methylhopane in addition to the ab-hopanes. Oleanane was absent.

This overall biomarker signature, together with the even carbon preference for n-alkanes, abundant acyclic isoprenoids, and low Pr/Ph ratio identified in the GC analysis confirms that the source organic matter was derived from a saline lacustrine setting. The heavy carbon isotopic signatures and presence of perhydro-b-carotene are consistent with this interpretation.

The precursor oils have a geochemical signature that matches some well-known analogs. The acyclic isoprenoid, sterane, and triterpane pattern is very similar to bitumens (including gilsonites) in the Miocene Green River formation of Wyoming and Utah,^15-16 and reported for some Tertiary-aged saline lacustrine oils and shales from China.^17-18 Notable departures from these examples, however, are the very low abundance of 4-methylsteranes (which are derived from dinoflagellates). This is a strong indication that the source of the oils predates the common occurrence of dinoflagellates and is therefore older than Late Triassic.^19-20 This is consistent with the absence of oleanane, a biomarker derived from angiosperm plants (which first became prominent in the Cretaceous).

Many of the features evident in the 'reconstructed' Nama basin oils have also been recognized in the Beatrice oil of the North Sea and traced to its co-source in Devonian lacustrine source rocks.²¹ A bitumen from this sequence resembles the Nama samples with a notable departure in the isotopic compositions (-34 to -35%) of the Devonian bitumens. Bulk carbon isotopic data for lacustrine organic matter can be highly variable due to isolation of lake-water chemistry from the moderating effect of the global ocean-atmosphere carbon reservoir.

Isotopic analysis of individual hydrocarbons^22-23 may hold a signal that could suggest a Phanerozoic age for the Nama oils. Pristane d13C values are -20 to -23.5% with n-C17 values in corresponding samples 1-3% lighter. This pattern was not present in any of the 30 Proterozoic samples analyzed by Logan et al.,²³ while it is present in the majority of Phanerozoic samples analyzed in that study and reported elsewhere.¹⁶ This isotopic observation accords with other evidence for a Phanerozoic age for the oils, such as absence of anomalous sterane preferences, low abundance of isoalkanes, and light carbon isotopic signatures that are found in many Proterozoic oils and bitumens.^19-24 However, the analysis of individual hydrocarbons is a new method with as yet little practical use in oil-source correlation. Further caution is warranted because the Proterozoic bitumens studied by Logan et al.²³ included none from saline lacustrine paleoenvironments, and also because, as noted above, isotope systematics in lacustrine systems can be very different from those of the marine realm.

Individual n-alkane and isoprenoid carbon isotope data for the four reconstructed oils track the bulk d13C values for the corresponding saturated hydrocarbon fractions. They all show a trend to lighter signatures at high carbon numbers with n-alkanes of #8633 shifting from -22% at C14 to -25% at C27 while #8634 is distinctly lighter -24% at C14 to -30% at C27; both samples are solid bitumen in veins but from different localities in the Fish River valley.

Since biodegradation does not significantly shift carbon isotopic compositions of the residual hydrocarbons, we can conclude that differences in hydrocarbon patterns indicate real differences in the source (paleoenvironment) of the bitumens despite the fact that a saline lacustrine source facies is common to all.

Both the Green River basin of the Rocky Mountain U.S. and the Devonian lacustrine sources of the North Sea include alkaline lake facies, and it may well be that the source in Namibia is not only saline, lacustrine, and low in siliciclastics, but also an alkaline, sodic system. Such facies are present in the Tertiary to Holocene Kalahari Group, but this succession has not been buried deeply enough to generate oil.

No evaporites are known from the Permian to Carboniferous Karoo Group, despite extensive outcrop and intensive drilling during the exploration for coal. There are some limited indications of marginal marine evaporative conditions in the Kuibis and Schwarzrand Subgroups, in the form of cauliflower cherts, and lenses of sugary limestone in the Nomtsas formation of the Fish River Subgroup may be after evaporites.²⁶

The pre-Nama Doornpoort and Duruchaus formations have widespread indications of saline lacustrine conditions, at least some of which were alkaline.^4-27 So, at present, these are considered the most likely source of the oil and gas. The only alternative seems to be that there was long-range (300 km) migration from an offshore source.

The Nama basin is the same age and is comparable tectonically with regions elsewhere that contain major petroleum occurrences and fields. New stratigraphic data allow relatively precise correlations with the petroliferous Neoproterozoic to Cambrian successions of Oman, Siberia, and China. Equivalent successions also occur in the Owambo basin north of the Damara orogen in Namibia and Angola, in the Passarge basin of Botswana, and in the Sao Françisco basin in Brazil.

The abundance of shows and comparisons with similarly aged petroliferous successions elsewhere suggest that the Nama basin should be included in any assessment of the petroleum prospectivity of southern Africa.

References

1. Miller, R. McG.., ed., Evolution of the Damara orogen of South West Africa/Namibia, Spec. Pub. Geol. Soc. S. Africa, Vol. 11, 1983, 515 p.

2. Germs, G.J.B., Implications of a sedimentary facies and depositional environment analysis of the Nama Group in South West Africa/Namibia, 1983, pp. 89-114, in Miller, R. McG., ed., Evolution of the Damara orogen of South West Africa/Namibia, Spec. Pub. Geol. Soc. S. Africa, Vol. 11, 1983, 515 p.

3. Hoffman, K.H., New aspects of lithostratigraphic subdivision and correlation of late Proterozoic to early Cambrian rocks of the southern Damara belt and their correlation with the central and northern Damara belt and Gariep belt, Communs Geol. Surv. Namibia, Vol. 5, 1989, pp. 59-67.

4. Behr, H.J., Ahrendt, H., Porada, H., Rohrs, J., and Weber, K., Upper Proterozoic playa and sabkha deposits in the Damara orogen, SWA/Namibia, 1983, pp. 1-20, in Miller, R. McG., ed., Evolution of the Damara orogen of South West Africa/Namibia, Spec. Pub. Geol. Soc. S. Africa, Vol. 11, 1983, 515 p.

5. Wright, J.A., and Hall, J., Deep seismic profiling in the Nosop basin, Botswana: cratons, mobile belts and sedimentary basins, Tectonophysics, Vol. 173, 1990, pp. 333-343.

6. Saylor, B.Z., Grotzinger, J.P., and Germs, G.J.B., Sequence stratigraphy and sedimentology of the Neoproterozoic Kuibis and Schwarzrand Subgroups (Nama Group), southwestern Namibia, Precambrian Research, Vol. 73, 1995, pp. 153-171.

7. Reid, D.L., Ransome, I.G.D., Onstott, and Adams, C.J., Time of emplacement and metamorphism of late Precambrian mafic dykes associated with the Pan-Africa Gariep orogeny, southern Africa: implications for the age of the Nama Group, J. African Earth Sciences, Vol. 13, 1991, pp. 531-541.

8. Grotzinger, J.P., Bowring, S.A., Saylor, B.Z., and Kaufman, A.J., Biostratigraphic and geochronologic constraints on early animal evolution, Science, Vol. 270, 1995, pp. 598-604.

9. Lawrence, S.R., Prospects for petroleum in Late Proterozoic/Early Paleozoic basins of southern-central Africa, J. Petrol. Geol., Vol. 12, 1989, pp. 231-242.

10. Altebaumer, A.M., and Altebaumer, F.J., Geochemical evaluation of the PCIAC-GSD Masetlheng Pan-1 well, Botswana, and an oil seep from Namibia, Report for Petro-Canada International Assistance Corp., Project G/S 285, 1990, 16 p.

11. Miller, R. McG., Hydrocarbons, in The mineral resources of Namibia, Ministry of Mines and Energy, Namibia, 1992.

12. Kvenvolden, K.A., and Roedder, E., Fluid inclusions in quartz crystals from South West Africa, Geochimica et Cosmochimica Acta, Vol. 35, 1971, pp. 1,209-29.

13. Jones, D.M., Douglas, A.G., and Connan, J., Hydrous pyrolysis of asphaltenes and polar fractions of biodegraded oils, Organic Geochemistry, Vol. 13, 1988, pp. 981-993.

14. Zofer, Z., Hydrous pyrolysis of Monterey asphaltenes, Organic Geochemistry, Vol. 13, 1988, pp. 939-945.

15. Collister, J.W., Summons, R.E., Lichtfouse, E., and Hayes, J.M., An isotopic biogeochemical study of the Green River oilshale, Organic Geochemistry, Vol. 19, 1992, pp. 265-276.

16. Schoell, M., Hwang, R.J., Carlson, R.M.K., and Welton, J.E., Carbon isotopic composition of individual biomarkers in gilsonites (Utah), Organic Geochemistry, Vol. 21, 1994, pp. 265-276.

17. Shi-Ji-Yang, Mackenzie, A.S., Alexander, R., Eglinton, G., Gowar, A.P., Wolff, G.A., and Maxwell, J.R., A biological marker investigation of petroleums and shales from the Shengli oilfield, The People's Republic of China, Chemical Geology, Vol. 35, 1982, pp. 1-31.

18. Fu Jiamo, Sheng G., Xu, J., Eglinton, G., Gowar, A.P., Jia, R., Fan, S., and Peng, P., Application of biological markers in the assessment of paleoenvironments of Chinese nonmarine sediments, Organic Geochemistry, Vol. 16, 1989, pp. 769-779.

19. Summons, R.E., Thomas, J., Maxwell, J.R., and Boreham, C.J., Secular and environmental constraints on the distribution of dinosterane in sediments, Geochimica et Cosmochimica Acta, Vol. 56, 1992, pp. 2,437-44.

20. Thomas, J., Marshall, J., Mann, A.L., Summons, R.E., and Maxwell, J.R., Dinosteranes (4,23,24-trimethylsteranes) and other biological markers in dinoflagellate-rich marine sediments of Rhaetian age, Organic Geochemistry, Vol. 20, 1993, pp. 91-104.

21. Peters, K.E., Moldowan, J.M., Driscole, A.R., and Demaison, G.J., Origin of Beatrice oil by co-sourcing from Devonian and Middle Jurassic source rocks, Inner Moray Firth, U.K., AAPG Bull., Vol. 73, 1989, pp. 454-471.

22. Hayes, J.M., Freeman, K.H., Popp, B.N., and Hoham, C.H., Compound-specific isotope analysis: a novel tool for reconstruction of ancient biogeochemical processes, 1990, in Durand, B., and Behar, F., eds., Advances in Organic Geochemistry, Pergamon Press, 1989, pp. 1,115-28.

23. Logan, G.A., Hayes, J.M., Hieshima, G.B., and Summons, R.E., Terminal Proterozoic reorganization of biogeochemical cycles, Nature, Vol. 376, 1995, pp. 53-56.

24. Summons, R.E., and Walter, M.R., Molecular fossils and microfossils of prokaryotes and protists from Proterozoic sediments, American Journal of Science, Vol. 290-A, 1990, pp. 212-244.

25. Hayes, J.M., Summons, R.E., Strauss, H., Des Marais, D.J., and Lambert, I.B., Proterozoic biogeochemistry, in Schopf, J.W., and Klein, C., eds., The Proterozoic biosphere: a multidisciplinary study, Cambridge University Press, 1992, pp. 81-133.

26. Germs, G.J.B., personal communication, 1996.

27. Ruxton, P., The sedimentology and diagenesis of copper-bearing rocks of the southern margin of the Damara Orogenic Belt, Namibia and Botswana, PhD thesis, University of Leeds, 1981, unpubl.

The Authors