ArticlePDF Available

Crustal framework of Namibia derived from magnetic and gravity data

Authors:
  • Corner Geophysics Namibia

Figures

Content may be subject to copyright.
2
An Integrated Geophysical and Geological
Interpretation of the Southern African
Lithosphere
Branko Corner and Raymond J. Durrheim
Abstract
Southern Africa, here taken as the region that comprises
the Kalahari and southern Congo Cratons, and the
important orogenic belts that surrounded or separated
them during the assembly of Gondwana, was situated in
the heart of the supercontinent. As such, the region is an
ideal site to study the lithospheric structure, composition
and evolution of the supercontinent. A plethora of data
sets, both geological and geophysical, are available in the
public domain, including outcrop mapping, drilling
results, aeromagnetic, gravity and magnetotelluric sur-
veys, allowing mapping of extensive regions under cover.
Deeper penetrating seismic reection, refraction and
teleseismic data, and also magnetotelluric data, have
allowed the lithospheric interpretation to be extended to
the middle and lower crust, and to the upper mantle.
Interpretation has included, inter alia, mapping, or
renement of existing mapping, of the craton boundaries
and associated terranes; major faults, structural linea-
ments and ring structures; specic features which have a
geophysical expression, such as the Witwatersrand Basin,
the Xade Complex and the tectonostratigraphic zones of
the Damara-Ghanzi-Chobe Orogenic Belt; the
Namaqua-Natal Belt and extensions thereof as the Maud
Belt in Antarctica, as well as associated features such as
the Beattie Magnetic Anomaly and Southern Cape
Conductivity Belt; and the Namibian volcanic passive
margin. Interpretations of the large-scale seismic, electri-
cal resistivity, geomagnetic induction and magnetotelluric
data by many workers have revealed conductive zones,
terrane boundaries and continental-scale shear zones
concealed by younger strata, and yielded important
insights into the deep structure and evolution of the
subcontinent. The topography of the Moho and the
LithosphereAsthenosphere Boundary has also been
mapped, showing that the Archaean Kaapvaal, Zimbabwe
and Congo Cratons have deep roots that are relatively
cold. As far as possible, the interpreted features honour
the geological and geophysical data sets within the
resolution of the data. Integration of these results in the
unied interpretation map presented here brings new
insights into both the disposition of selected geological
features under cover, and the evolution of the Precam-
brian geology of southern Africa, extending into Antarc-
tica within a Gondwana framework.
Keywords
Southern Africa Geophysical studies Integrated
interpretation Continental crust Lithosphere
Craton Gondwana Aeromagnetic Gravity
Magnetotelluric Seismic Kaapvaal Zimbabwe
Congo Maud Belt Namaqua-Natal Belt
Damara-Ghanzi-Chobe Belt Beattie Magnetic Anomaly
Southern Cape Conductive Belt
2.1 Introduction and Chapter Layout
Southern Africa is an ideal site to study the structure,
composition and evolution of the Gondwana supercontinent.
Geophysical methods, integrated with the results of geo-
logical eld mapping and exploratory drilling, play an
increasingly important role in mapping rocks and structures,
concealed by younger cover, from the near surface to the
upper mantle. The Archaean-to-Palaeoproterozoic Kaapvaal,
Zimbabwe and Congo Cratons are surrounded by Protero-
zoic metamorphic belts, platforms and basins, preserving
more than 3600 million years of the Earths history (Hunter
et al. 2006). These terranes are penetrated by kimberlite
intrusions, which provide samples of the lower crust and
B. Corner (&)
Manica Minerals Ltd., Swakopmund, Namibia
e-mail: branko@iafrica.com.na
R. J. Durrheim
School of Geosciences, University of the Witwatersrand,
Johannesburg, South Africa
e-mail: Raymond.Durrheim@wits.ac.za
©Springer International Publishing AG, part of Springer Nature 2018
S. Siegesmund et al. (eds.), Geology of Southwest Gondwana, Regional Geology Reviews,
https://doi.org/10.1007/978-3-319-68920-3_2
19
upper mantle (Skinner and Truswell 2006). Hart et al. (1981,
1990a,b) propose that the 2023 Ma Vredefort meteorite
impact near the centre of the Kaapvaal Craton turned the
crust on edge, providing a window into the deeper crys-
talline basement. Geophysical studies of the structure of the
continental crust and upper mantle have also been driven by
the search for hydrocarbons, diamonds, base metals, and
precious metals. While the resolution and accuracy of geo-
physical images and models has improved as technology and
knowledge have advanced, early studies still remain relevant
because they provide important constraints on lithospheric
models, especially in areas where surveys have not been
repeated. De Beer (2015a,b,c) provides a comprehensive
review of the history of geophysics in South Africa. We
briey also review the historic investigations that were
conducted 20 or more years ago. More recent work is dis-
cussed in greater detail, as well as new aspects of interpre-
tation of magnetic and gravity data sets. The scope of
coverage is enormous, thus the mapping is selective,
focusing on geophysically evident features and selected
geological units that may not have a geophysical expression
but which are relevant to the interpretation.
The early regional aeromagnetic and gravity data sets
covering southern Africa have proved invaluable in mapping
regional structure. More recently, higher-resolution data sets
have become available in Namibia and Botswana, as well as
magnetotelluric (MT), seismic reection and teleseismic
array studies. The interpretation presented here follows the
norm of working from known, mapped geology, as pub-
lished by the various geological surveys and geoscience
councils, to mapping extensions thereof under cover using
the geophysical data sets presented here, either by us or as
referenced. Emphasis is placed on the mapping of large-scale
structures, including major faults, lineaments and ring
structures, and extensions of specic lithologies and
tectonostratigraphic zones, many of which were not recog-
nized in past studies of these geoscience data sets. Much of
what is presented here is, in the rst instance, observational,
identifying and mapping many new features and thus cre-
ating a basis for further research on their nature, genesis and
geological evolution. Although potential eld data has
formed the basis of the presented interpretation, and includes
consideration of the MT and seismic results, we recognize
that complete integration of the data sets has not been fully
achieved. The prime reason for this is that the various geo-
physical studies had different key questions, and depths of
investigation that did not always overlap, ranging from the
near surface to the mantle. In addition, aspects of the chapter
address potentially different readership interests, particularly
in Sects. 2.2 and 2.3. Figure 2.1 shows the extent of the
interpretation area within southern Africa.
The coverage of the chapter, bearing the above comments
in mind, is based on the type of geophysical data usedthat is,
commencing with potential eld data integrated with known
geology, progressing to published magnetotelluric and seis-
mic data and interpretations, as summarized briey below:
Interpretative mapping of specic Archaean and
Proterozoic geological features and their boundaries, in
areas covered by Phanerozoic or younger sediments,
largely using magnetic and gravity data sets integrated
with published outcrop geologythat is:
Kaapvaal, Zimbabwe, Kalahari and southern Congo
Cratons, their boundaries and intracratonic belts or
zones (e.g., Limpopo Terrane (LP) and Magondi Belt);
aspects of crustal magnetization;
the Witwatersrand Basin and its extensions;
the Xade Complex in Botswana;
the Meso-Proterozoic Sinclair-Rehoboth Groups, and
Grootfontein Metamorphic Complex;
tectonostratigraphic zones of the
Damara-Ghanzi-Chobe Belt, and the Gariep Belt;
the Rehoboth Terrane, and deep Neo- and Late Meso-
Proterozoic basins in Namibia, Botswana and Angola;
the Karas Impact Structure;
the offshore Namibian passive volcanic margin;
the Namaqua-Natal Belt, the Khoisan Province, the
Beattie Magnetic Anomaly and the Southern Cape
Conductivity Belt;
extensions of the Namaqua-Natal Belt within Gond-
wanathe Maud Belt, Antarctica;
Fig. 2.1 Location of the interpretation area. The red outline delineates
the Proto-Kalahari Craton (Jacobs et al. 2008). See Sect. 2.2.10.3 for a
discussion of the Gondwana reconstruction used. DCB Damara-Chobe
Belt; KVC Kaapvaal Craton; LT Limpopo Terrane; NNB
Namaqua-Natal Belt; ZC Zimbabwe Craton; CC Congo Craton; GC
Grunehogna Craton
20 B. Corner and R. J. Durrheim
major, geophysically evident faults and structural
lineaments.
Electrical resistivity, magnetotelluric, and regional seis-
mic investigations which complement the above inter-
pretation and extend it to the deeper crust and upper
mantle:
reection seismic data sets which have been used to
search for reservoirs containing oil and gas on land
and at sea, and also to map and explore for extensions
of the Witwatersrand Basin and Bushveld Complex;
electrical and seismic characteristics which have been
used to determine the thickness of the crust and
lithosphere and to map the boundaries between cra-
tons and mobile belts;
several important regional zones of anomalous con-
ductivity and magnetization have been discovered, as
noted also abovethat is, the Southern Cape Con-
ductivity Belt, Damara-Chobe Conductivity Belt and
the Beattie Magnetic Anomaly. Considerable research
has been conducted to map, and determine the cause
of, these anomalous zones.
2.2 Potential Field Data Sets: An Integrated
Interpretation
2.2.1 Magnetic and Gravity Data
The regional aeromagnetic data sets covering southern
Africa, mostly acquired pre-1990, have proved invaluable in
mapping stratigraphic units, specic lithologies and regional
structure under the cover sequences. These data, shown as a
reduced-to-the-pole (RTP, of the total magnetic intensity)
image in Fig. 2.2, were acquired at ight-line spacings
varying from mostly 14 km, under contract to the various
geological surveys and geoscience councils of the countries
covered by the map. The more recent medium-resolution
Botswana aeromagnetic data (200250 m line spacing,
degraded to a 500 m grid interval) was merged into this grid.
The image also includes an offshore aeromagnetic survey,
own under contract to the National Petroleum Corporation
of Namibia (Pty) Ltd (NAMCOR) over the continental shelf
area at a line spacing of 25 km. Some of the interpreted
Fig. 2.2 Reduced-to-the Pole
regional aeromagnetic image of
southern Africa, including
offshore aeromagnetic data over
the continental shelf of Namibia.
A grey-scale image is shown
here, in preference to colour,
because it best shows the
magnetic relief of some of the
features. (With acknowledgement
to Fugro/CGG airborne Surveys,
the Council for Geoscience of SA
(formerly Geological Survey), the
Geological Surveys of Namibia,
Botswana, Swaziland,
Mozambique, Zambia and
Zimbabwe, and NAMCOR Pty
Ltd. No data is available in the
public sector for Angola, Lesotho
or portions of Mozambique.)
2 An Integrated Geophysical and Geological Interpretation 21
geological features are derived from the higher-resolution
data sets in Botswana and Namibia. Also used in the inter-
pretation was the Worldwide Earth Magnetic Anomaly data
set, compiled from merged satellite, airborne and marine
magnetic data (EMAG2; Maus et al. 2009). Given the rel-
atively low 2 arc-min grid spacing, the higher-frequency
(shallower) anomalies are degraded but the grid preferen-
tially enhances some of the regional structures (Fig. 2.3).
The offshore areas of this data set, compiled from satellite
and marine magnetic data sets, provide a wealth of structural
and seaoor spreading information.
The Bouguer Anomaly image shown in Fig. 2.4, to which
a regional-residual separation lter has been applied, was
derived from mostly ground-based national gravity data sets
of Botswana, Namibia, Zambia, Mozambique, Swaziland
and Lesotho, as well as from an older-generation South
African data set of the Geological Survey (now the Council
for Geoscience). Also shown in Fig. 2.4 is the
satellite-derived NAMCOR offshore Free-Air anomaly data
set covering the continental shelf of Namibia. Further gravity
data sets used for interpretation were extracted from the
World Gravity Map and related products, compiled by the
Bureau Gravimetrique International (Balmino et al. 2012;
Bonvalot et al. 2012) from satellite, airborne and ground
data. Four data products were available in XYZ format at a 2
arc-min grid spacingthat is, the Bouguer Anomaly data
shown in Fig. 2.5, and the Isostatic Anomaly, surface
Free-Air Anomaly data and the ETOP-1 Topographic data,
which are not shown here owing to space constraints. These
data sets were gridded at a 3500 m interval. Although the
resolution is reduced with the Bureau Gravimetrique Inter-
national gravity data sets, the regional structure is relatively
clear, as with the aeromagnetic data. Of particular value is
the gravity data over Angola, where more detailed private
sector ground and airborne data sets are not available.
2.2.2 Interpretation Methodology
The interpretation maps in Figs. 2.6 and 2.7 show geological
units, stratigraphy and tectonostratigraphic zones that have
been derived from either (1) published outcrop mapping, as
indicated in the caption to the interpretation map (Fig. 2.6)
or (2) from geophysical interpretation by the authors and
their co-authors, or other authors as referenced in each case.
Interpretation was aided in part by numerous magnetotelluric
(MT) surveys discussed in Sect. 2.3. A number of lters
were applied to the above potential eld data sets prior to
Fig. 2.3 Total Magnetic
Intensity Worldwide Earth
Magnetic Anomaly image,
derived from satellite, airborne
and ship-track magnetic data
(Maus et al. 2009)
22 B. Corner and R. J. Durrheim
interpretation so as to enhance structure, lithological fabric
and contact locations. In the case of the magnetic data,
Reduction-to-the-Pole (RTP) of the Total Magnetic Intensity
(TMI) was applied for specic local-scale interpretations.
Filters applied to the RTP data set included the First Vertical
Derivative, Gaussian residual lters, Analytical Signal, and
Total Horizontal Derivative. In the case of the gravity data
sets, a Gaussian regional-residual separation lter was
applied to each of the three Bureau Gravimetrique Interna-
tional gravity grids in order to reduce high-frequency noise,
evident in the data, before further ltering to enhance
lithological anomalies. The interpretation methodology fol-
lowed the norm of working from the known, mapped geol-
ogy and structure to a projection thereof under areas of cover
using the above geophysical data sets. The derivative and
residual lters provided the resolution required to map both
local- and regional-scale structures. However, owing to the
scale of presentation, many of the smaller, local structures
have been excluded from the interpretation maps presented
here. All features shown in the maps honour both the
mapped geology and the geophysical data sets as best pos-
sible within the resolution of the data sets.
The interpretation maps (Figs. 2.6 and 2.7) show
numerous geophysically evident faults and lineaments, many
of which have been previously identied and discussed (e.g.,
Corner 2008). The term lineamentis used here, in the
denition of Richards (2000), to denote a large-scale fault
zone, or structural corridor, having a much broader swathe of
manifestation up to 50 km in width, either continuous or
disrupted. Geophysically, lineaments reect approximately
linear structures or zones with anomalous physical properties
that range in depth from surface to 5 km or more, depending
on the size of the source bodies and their physical property
contrast. The lineaments, shown as lines on the interpretation
maps, thus reect the locus of a much broader structural
corridor.
A number of ring features have also been identied
(Corner 2000,2008), which are considered to be the mani-
festation of ring fractures or faults, or alteration aureoles.
Their origins may be varied and possibly include:
ring fractures or faults associated with magmatism and
associated intrusions;
alteration aureoles associated with intrusions;
alteration aureoles associated with exhalative vents;
meteorite impact;
craton-scale ring structures resulting from plate rotation
or changes in lithospheric thickness.
Fig. 2.4 Residual-ltered
Bouguer Gravity image and Free
Air data offshore Namibia. The
onshore data was derived from
ground surveys and the offshore
Free Air data from satellite
measurements
(Acknowledgements as in
Fig. 2.2)
2 An Integrated Geophysical and Geological Interpretation 23
Fig. 2.5 Bouguer Gravity image
of southern Africa derived from
the World Gravity Map (Balmino
et al. 2012; Bonvalot et al. 2012)
Fig. 2.6 aIntegrated interpretation of southern Africa. Legend and
text key are in Fig. 2.6,bLegend to Fig. 2.6a. Note that the
stratigraphic units marked with an asterisk denote mapping published
by the geological surveys, and geoscience councils or institutes, of the
countries covered by the map. The following abbreviations are used
ASZ(WA) Amanzimtoti Shear Zone (Williston Anomaly); Au-L Autseib
Lineament; BA, Bloemfontein Arch; Bb Brandberg Complex; BDSL
Botswana dyke swarm, main limits; Be-L Bethlehem Lineament; BGH
Bethlehem Gravity High; BIC Bushveld Igneous Complex; BK-L
Barberton-Kimberley Lineament; BMA Beattie Magnetic Anomaly; CC
Congo Craton; Cc Cape Cross Complex; CFB Cape Fold Belt; Co-L
Colesberg Lineament; Cs Cape Seal Complex; ddeep features;
DDaneib intrusion; EF Elliot fault; EI Epupa Inlier; Er Erongo
Complex; Et Etosha pan; Ga-L Gam Lineament; GD Great Dyke; GI
Grootfontein Inlier; GMB Grootfontein Mac Body; GSP Gordonia
Subprovince; Gu Gunib Intrusions; HZ Hinge Zone of seaward dipping
seismic reectors; Ka Soekor borehole; Kal-L Kalahari Lineament;
KB-L Kuboos-Bremen Lineament; KVC Kaapvaal Craton; KVC-OB
Kaapvaal Craton Okwa Block; Kc Soekor borehole; Kf Koegelfontein
Complex; KG-L Khorixas-Gaseneirob Lineament; KH Karas Horst;
Kh-L Kheis Lineament; KIC Kunene Igneous Complex; KIS Karas
Impact Structure; KI Kamanjab Inlier; KP Khoisan Province; Ku-L
Kudu Lineament; LB Lebombo Belt; LMSZ Lilani-Matigulu Shear
Zone; LSZ Lovat Shear Zone; LT Limpopo Terrane; M2,4 Offshore
magnetic anomalies; MMessum Complex; MAB Matchless Amphibo-
lite Belt; Mf Molopo Farms Complex; MGB Magondi-Gweta Belt;
MGB-L Magondi-Gweta Belt limit; MIS Morokweng Impact Structure;
MkF Makgadikgadi Fault; MSZ Mwembeshi Shear Zone; MT Melville
Thrust; NNaukluft Nappe Structure; NamqP Namaqua Province; NatP
Natal Province; Naq-L Namaqua Lineament; NcB Ncojane Basin; NNF
Namaqua-Natal Front; NoB Nosob Basin; Ok-L Okahandja Lineament;
Om-L Omaruru Lineament; OR Omatako remanent anomalies; Op-L
Opuwo Lineament; OtM Otjiwarongo Massif; Ov Okavango delta;
PPilanesberg Complex; Pa Paresis Complex; PB Passarge Basin;
Pof-L Pofadder Lineament; PaSZ Palala Shear Zone; PuSZ Purros
Shear Zone; PV Phoenix Volcano; QI Quangwadum Inlier; Qu Soekor
borehole; RGFZ Rio Grande Fracture Zone; RT Rehoboth Terrane;
SMA Steinhausen Magnetic Anomaly; SRZ Sinclair-Rehoboth Zone; ST
Sesfontein Thrust; St-L Strydenburg Lineament; TA Tugela Allochthon;
TC Trompsburg Complex; TI Tsumkwe Inlier; TK-L Trans-Kalahari
Lineament; T-G-C-B Tsumis-Ghanzi-Chobe-Belt; TM-L
Thabazimbi-Murchison Lineament; Ts-L Tsodilo Lineament; Tsh
Tshane Complex; Tst Tsetseng Complex; TTF Tugela Thrust Front;
VIS Vredefort Impact Structure; Wb Walvis Bay Complex; We Soekor
borehole; We-L Welwitschia Lineament; WFZ Walvis Fracture Zone;
WR Walvis Ridge; WT Waterberg Thrust; XC Xade Complex (SL, NL
South, North Lobes); ZFZ Zoetfontein Fault Zone; ZC Zimbabwe
Craton
c
24 B. Corner and R. J. Durrheim
Examples of ring features are provided and discussed by
Corner (2008) and Corner et al. (1997). For their arcuate
geometry to be preserved, which may cut across older
structural fabric, the ring features would have to be post- or
late-tectonic in the rst instance. It is also considered pos-
sible, as with many major faults, that the above foci or
causative sources may have been reactivated through geo-
logical time owing to crustal weakness. In this sense, a ring
feature may be evidence of an older reactivated focal source.
A craton-scale source (last item in list above) is best
exemplied by the Kaapvaal Ring Structure (KVRS;
Fig. 2.7), which may have resulted from rheological or
structural variations in the deep crust or upper mantle. It is
evident in the south as the arcuate Namaqua-Natal Front,
separating the Kaapvaal Craton and the Namaqua-Natal
Belt, whereas its northern arcuate sector is evidenced
through fault-trace analysis of the aeromagnetic data. Fig-
ure 2.16 shows an overlay of the KVRS on the 200 km
P-wave velocity model depth slice. It clearly encompasses
the high-velocity root of the Kaapvaal Craton and is inter-
preted here to arise from relative movement between this
root and the surrounding lower velocity zones, perhaps ini-
tiated during plate movement along the zones of competency
contrast.
2 An Integrated Geophysical and Geological Interpretation 25
Examples of ring structures that are interpreted to arise
from meteorite impact events are, rstly, due to the Mor-
okweng Impact event (MIS; Fig. 2.6; MRS; Fig. 2.7; e.g.
Corner 1994a; Andreoli et al. 1995; Corner et al. 1997) and,
secondly, due to the interpreted Karas Impact event (KIS;
Fig. 2.6; KRS; Fig. 2.7; Corner 2008; Section 2.2.7.2)
2.2.3 Archaean and Palaeoproterozoic Cratons
2.2.3.1 Introduction to the Interpretation
of the Archaean and Proterozoic Geology
The various Archaean and Proterozoic stratigraphic units
shown in the interpretation map (Fig. 2.6a, Legend 2.6b) are
derived, rst, from mapped geology as published by the
relevant geological surveys, geoscience councils and insti-
tutes of the countries covered, and, second, from the inter-
pretation of magnetic and gravity data sets, constrained by
the outcrop data and limited published borehole data in the
areas of cover. Cover sequences that have not been shown in
this interpretation are dened here as being Phanerozoic in
age, ranging approximately from the Cambrian to recent,
with one exceptionthat is, the magnetic basalt ows and
dolerite sills of Karoo age, including the Etendeka basalts.
The distribution of these rocks is shown so as to indicate
those areas where interpretation of the underlying
suboutcrop geology is compromised as a result of the pres-
ence of these highly magnetized strata in the cover sequence.
Much geophysical research has been conducted in the
southern African region. Aspects that are highlighted and
referenced in this review include (1) interpretation of gravity
and magnetic data, particularly of Precambrian features
which have a clear expression in this data; (2) geomagnetic
induction, Magnetotelluric (MT) and deep electrical resis-
tivity studies; (3) deep seismic reection surveys, conducted
both by industry in its quest to locate extensions to the
Witwatersrand Basin and by the South African National
Geophysics Programme; (4) deep seismic refraction surveys;
and (5) teleseismic studies of the cratons, underlying mantle,
and adjacent polymetamorphic terranes. Many geological
publications on the Precambrian geology of southern Africa
show variable boundaries or delineations of the cratons and
surrounding terranes in the areas of cover, as well as of key
regional structures. Often, the detail which geophysical data
can and does give has been ignored or at best loosely
interpreted. We have sought to be rigorous in the mapping of
these features, carefully honouring the detail of geological,
magnetic and gravity data sets on both regional and local
scales. The latter interpretations are not shown here owing to
the scale of this presentation, but have been used to constrain
the regional mapping shown here, as referenced where
appropriate. The lines indicating craton and terrane
Fig. 2.6 (continued)
26 B. Corner and R. J. Durrheim
boundaries should be considered to represent broader com-
plex swathes of varying structural style and dip, drawn at
their shallowest manifestation.
2.2.3.2 Kaapvaal Craton
The Kaapvaal and Zimbabwe cratons were formed and have
grown through accretion, thus comprising crustal blocks of
different ages with different structural styles, from the
Archaean to the Proterozoic. The boundaries thereof are
mostly clearly revealed by the gravity and magnetic data. The
assumption is made here that major structures or structural
zones, as evidenced in the geophysical data, with geological
control where available, constitute the craton boundaries. This
applies similarly to the surrounding polymetamorphic ter-
ranes. De Beer and Meyer (1984) were the rst to geophysi-
cally map a portion of the Kaapvaal Craton boundary covered
by Phanerozoic rocks, delineating the arcuate southern craton
margin by modelling a number of gravity proles across it,
where it is juxtaposed against the Namaqua-Natal Belt.
The Kalahari Lineament (Kal-L; Fig. 2.6; Kalahari Line,
Reeves 1978), interpreted as the western Meso-,
Palaeo-Proterozoic boundary of the Kaapvaal craton, is one
of the most dramatic features in the aeromagnetic image of
southern Africa (Figs. 2.2 and 2.6), separating relatively
Fig. 2.7 Regional ring or arcuate structures are superimposed on the interpretation map of Fig. 2.6.CBRS Chameis Bay ring structure; KRS Karas
ring structure; KVRS Kaapvaal ring structure; MRS Morokweng ring structure; ORS Omatako ring structure; PRS Phoenix ring structure
2 An Integrated Geophysical and Geological Interpretation 27
shallow basement to the east (with mostly less than 1 km of
cover) from extremely deep magnetic basement beneath the
Rehoboth Terrane (RT) to the west, where cover thicknesses
have been determined to vary from 6 to 10 km (Reeves
1978; Corner 2008). The magnetic signature of the Kal-L
changes towards the south, where it bounds the Kheis Pro-
vince (comprising the Olifantshoek Supergroup rocks) in the
west, but the craton boundary is nevertheless clear as map-
ped in Fig. 2.6. Moen (1999) places the western limit of the
Olifantshoek Supergroup at the Dabeep fault, which lies
roughly centrally between the craton boundary and the Kheis
Lineament (Kh-L) in Fig. 2.6. No justication for a major
structural boundary is seen in the geophysical data in this
central area. Recent gravity data (Botswana Geoscience
Institute) indicates that a gravity high is associated with the
Kal-L, although not ubiquitously. This suggests that the
Kal-L magnetic signature results in part from mac intru-
sions, probably mostly of late Mesoproterozoic Umkondo
age (c. 1.1 Ga; Meixner and Peart 1984; Hanson et al. 2006;
Cornell et al. 2011), and in part from magnetization inter-
preted to arise from hydrothermally altered granite-gneiss,
associated with this major suture, where no clear gravity
high is evident (Corner 2008). Localized high magnetic
anomalies evident within the anomalous gravity zone are
most likely associated with smaller-scale mac intrusions,
such as the Tshane Complex.
AnumberoffeatureshallmarktheKal-Lasamajorcrustal
structure, including the enormous change in depths to base-
ment, as described above, and apparent separation of Precam-
brian stratigraphy and terranes of differing structural styles,
from its east to its west. It has beeninferred to be a zone of major
collision (Meixner and Peart 1984) or a zone of major trans-
pression (Cornell et al. 2011). Its age has been inferred to be
post-Waterberg (Olifantshoek Supergroup equivalent)that
is, post-Eburnean (Reeves 1978). Early interpretations of the
Kal-L (e.g., Reeves 1978) identify a separate crustal terrane, the
Okwa basement, north of the Makgadikgadi Fault, where it cuts
the Kal-L. Many authors interpret the Zoetfontein fault to be the
northern boundary of the Kaapvaal craton, constituting the
southern boundary of the larger Okwa terrane north thereof
(e.g., De Wit and Tinker 2004), a view most likely based on the
original interpretation by Reeves (1978), who separated the
basement north of the Zoetfontein fault from cratonic basement
to the south. Corner (1998) does not support the interpretation
of the larger Okwa terrane (i.e., north of the Zoetfontein fault,
Kaapvaal Craton Okwa Block; Fig. 2.6) as a crustal entity
separate from the Kaapvaal Craton, since it is bounded in the
west by the uninterrupted Kal-L. That the Zoetfontein fault is a
major early fault, with post-Karoo faulting as its youngest
manifestation, is beyond question. However, as with the
Makgadikgadi fault, there is no disruption of the Kal-L by the
Zoetfontein fault. The continuity of the Kal-L thus suggeststhat
the crustal blocks accreted north and south of the Zoetfontein
fault are at least of Meso-, Palaeo-Proterozoic age, if not earlier.
Nevertheless, interpretation of the aeromagnetic data conrms
a clear change in basement fabric north and south of the
Zoetfontein fault, the two most dramatic examples being the
truncation of the north-northwest-trending Pilanesberg dykes
against it, with minor continuation north thereof beneath Karoo
basalt, and the east-west disposition of the interpreted feeder
dykes to the Xade Complex to its north (Figs. 2.2 and 2.6;
Corner et al. 2012). A change in crustal level within the
Kaapvaal Craton is thus inferred across the Zoetfontein fault
zone. The Okwa terrane, as described above, is referred to as the
Okwa Block here, and interpreted to be part of the Kaapvaal
Craton. Other authors also recognize the possibility that the
Okwa Block is one of a mosaic of blocks making up the
Kaapvaal Craton (e.g., Eglington and Armstrong 2004). Fur-
ther subdivisions internal to the Kaapvaal Craton have been
published. For example, Schmitz et al. (2004) recognize sepa-
rate entities west (Witwatersrand Block) and east (Kimberley
Block) of the Colesberg Lineament.
Reeves (1978) identied two sub-basins west of the
Kal-L, overlying the Rehoboth Terranethat is, the Ncojane
Basin north of the Makgadikgadi fault, and the Nosob Basin
to the south. The distinction between these two sub-basins is
not recognized here, other than a probable change in
deformation northwards as the Tsumis-Ghanzi-Chobe Belt is
approached (T-G-C-B; Fig. 2.6). The northern margin of the
Kaapvaal Craton, adjoining the Kal-L, is clearly evidenced
by a similar dramatic change in depth to basement beneath
the Passarge Basin (PB; Fig. 2.6), which is underlain by the
T-G-C-B. Further east, the Palala shear zone (PaSZ;
Fig. 2.6) is seen to trend north-westwards towards the PB, in
both the gravity and magnetic data sets, constituting the
boundary between the Kaapvaal Craton and the Limpopo
Terrane (LT; Fig. 2.6). The eastern craton margin is well
delineated by the Lebombo Belt (LB; Fig. 2.6), both geo-
physically and geologically.
The southeastern boundary of the Kaapvaal Craton,
extending through Lesotho to the Tugela Thrust Front
(TTF; Fig. 2.6), is a broad complex tectonic zone of inherited
Archaean and Mesoproterozoic crust separating the craton
from the Natal Province. This zone has been termed the
Tugela Allochthon (TA; Fig. 2.6), based on geophysical and
isotopic studies by Barkhuizen and Matthews (1990), De Wit
and Tinker (2004), Eglington and Armstrong (2004); Schmitz
and Bowring (2004). Although no magnetic data is available
for Lesotho, the full gravity coverage has been merged into the
residual Bouguer Gravity image of Fig. 2.4. Two gravity
highs are evident in Lesotho, roughly parallel to the craton
margin. At rst sight these might be taken to be part of the
Kaapvaal Craton, roughly on-strike with, although detached
from, the gravity high (and coincident magnetic high) anking
the southern Witwatersrand Basin in the south. Alternatively,
these highs might be interpreted to be relicts of the (unknown)
28 B. Corner and R. J. Durrheim
source of the Bethlehem Gravity High (BGH; Fig. 2.6).
However, the latter is of much higher amplitude and lacks any
magnetic expression. Schmitz and Bowring (2004) have
shown, from geochronological and isotopic data on lower
crustal xenoliths from the Lesotho kimberlites located just
northeast of the two Lesotho gravity highs, that granulitization
of the lower crust was a relatively young phenomenon, c. 1.0
1.1 Ga, which affected pre-existing Archaean to Mesopro-
terozoic crust. The gravity highs are thus assumed to be within
the allochthonous zone (e.g., of De Wit and Tinker 2004), as
are the smaller-scale gravity highs further east (excluding the
known mac intrusive sources). The Tugela Thrust Front
clearly demarcates the northern boundary of the TA in the
northern sector of the Natal Province. A possible continuation
of the Kaapvaal Craton into Dronning Maud Land, Antarc-
tica, where the Grunehogna Craton has been identied, is
discussed in Sect. 2.2.10.3.
2.2.3.3 Zimbabwe Craton, Limpopo Terrane,
Magondi Belt and the Kalahari Craton
The disposition of the Zimbabwe Craton (ZC; Fig. 2.6), and
associated Magondi-Gweta Belt (MGB; Fig. 2.6) and Lim-
popo Terrane (LT; Fig. 2.6), has been revisited here in as
much detail as the mapped geology and geophysical data
sets allow. The boundary between the ZC and LT is rela-
tively well dened from geological mapping in southern
Zimbabwe, northern South Africa and eastern Botswana,
and from a clear change in magnetic fabric across the
boundary. Working westwards, this magnetic fabric loses
clarity under the cover of Karoo basalts, although it is still
evident in places. The MGB has a semilinear magnetic fabric
that is not dissimilar to that of the LT. A faulted contact is
suggested between the two belts in Fig. 2.6, but this may be
a local feature, and the possibility thus exists that the LT and
MGB constitute a continuous deformation zone encom-
passing the ZC in the south and west. Mapping of the eastern
boundary of the ZC is limited by the both the paucity and the
coarseness of the geophysical data.
The Mwembeshi Shear Zone in southern Zambia (MSZ;
Fig. 2.6), a major roughly eastwest dislocation zone, has been
extended using the aeromagnetic data, as shown in Fig. 2.6 (see
also Fig. 2.2)that is, in the east it is seen to curve south-
eastwards toward the ZC boundary. Two other regional struc-
tures, with a similar roughly eastwest trend, are mapped to its
south. Fault mapping, derived from both the geological map-
ping and the geophysical interpretation, shows a continuation
oftheseregionalstructureseastwardintoZimbabwe.Oneof
these (in northern Zimbabwe) is a major post-Karoo fault, at its
youngest manifestation, preserving a deep Karoo basin to the
norththat is, the Lower Zambezi Zone of the Cabora Bassa
Basin. These regional fault zones appear to curve into the ZC
boundary, suggesting possible later transpressional movement
along the boundary. Of interest is that the northeast-trending
magnetic fabric of the Magondi-Gweta Belt in Zimbabwe is
seen to change direction dramatically, truncating against the
above eastwest-striking fault zone, and trending westwards
into Zambia south of the Mwembeshi Shear Zone, initially
following the eastwest trend of this zone. Deep-seated mag-
netic sources are also shown within the basement in Zambia
(Fig. 2.6), which mimic the interpreted westward trend of the
Magondi-Gweta Belt. We nd no evidence in the geophysical
data for any continuity between the LT and the so-called Okwa
Block, as inferred by a number of workers.
Jacobs et al. (2008) discuss the evolution of the Kalahari
Craton, describing it as having been spawned from a small
Archaean core which grew by prolonged crustal accretion in
the Palaeoproterozoic to form the Proto-Kalahari Craton
by 1750 Ma. They include the Grunehogna Craton in
Dronning Maud Land, Antarctica, in their denition of this
Archaean core (as discussed in Sect. 2.2.10.3). From
c. 1400 Ma, all margins of the Proto-Kalahari Craton
recorded intense tectonic activity, and by c. 1050 Ma the
Proto-Kalahari nucleus was almost completely rimmed by
voluminous Mesoproterozoic crust, becoming a larger entity,
the Kalahari Craton (Jacobs et al. 2008). The outline of the
Proto-Kalahari Craton, as dened by Jacobs et al. (2008)
above, is shown in Fig. 2.1.
2.2.3.4 Congo Craton in Namibia and Botswana
McCourt et al. (2013) describe the Angolan Shield,
c. 2.0 Ga, as being a Palaeoproterozoic basement terrane
dominated by granitoids, together with a limited amount of
Neo-Archaean crust, which extends from south of Lubango
in Angola into Namibia, and eastward under cover into
Zambia. This denes the southwest section of the Congo
Craton. The craton was intruded by the c. 1385 Ma Kunene
Complex, the areal extent of which indicates an extensive
period of Mesoproterozoic crustal extension (McCourt et al.
2013). Delineation of the southern boundary of the Congo
Craton in northern Namibia, and its continuation into
Botswana and Zambia, has been the subject of much spec-
ulation in view of, rstly, the extensive Karoo and Kalahari
cover in the central and eastern areas of northern Namibia
and, secondly, the complex structural evolution that hall-
marks the Kaoko Zone in the northwest (see Sect. 2.2.6.3).
One is left with geophysical signatures (magnetic and
magnetotelluric), and in places a lack thereof, which have
often been variably interpreted, to map the craton boundary.
A signicant crustal-scale magnetic anomaly, character-
ized by deep-seated high-amplitude anomalies arising from
the crust beneath the Namibian Northern Platform and a
southern dominant magnetic low, strikes northeastwards
across northern Namibia, as is readily evident in Fig. 2.2.
Eberle et al. (1995) modelled a number of magnetic proles
across key structural features in Namibia, of which three
traversed this regional magnetic anomaly. Their modelling,
2 An Integrated Geophysical and Geological Interpretation 29
based on dipping prism-shaped bodies with induced (nor-
mal) magnetization, consistently suggested the presence of a
regional-scale antiformal structure situated beneath the
southern portion of the carbonate platform. This may indi-
cate the southern abutment of the Congo Craton (CC). We
thus interpret this regional magnetic anomaly and associated
structure to delineate the boundary of the Congo Craton, as
shown in Fig. 2.6, which is a renement of an earlier
interpretation which placed the boundary further north
(Corner 2008). The magnetotelluric work of Khoza et al.
(2013a,b), and seismic tomography studies of Raveloson
et al. (2015), provide support for the boundary mapped here.
The Khorixas-Gaseneirob Lineament (KG-L; Miller 2008a;
Fig. 2.6) is thus considered to constitute the near-surface
manifestation of the southern boundary of the Congo Craton.
The continuity of the southern eastwest-trending portion of
the craton boundary has been disrupted by a number of
major structural lineaments, as mapped in Fig. 2.6, including
the Welwitschia (We-L) and Kudu (Ku-L) Lineaments in
particular. This relatively well-dened regional magnetic
signature of the southern Congo Craton boundary disappears
northwestwards, where the craton boundary is inferred to
continue along the north-northwest-trending Purros Shear
Zone (PuSZ; Fig. 2.6). Similarly, in the east, the Congo
Craton boundary, as evidenced in the magnetic data, appears
to terminate abruptly in northwestern Botswana against what
is newly interpreted here as the north-trending Tsodilo
Lineament (Ts-L; see also Sect. 2.2.6.3).
The Mesoproterozoic Grootfontein Inlier (GI; also known as
the Grootfontein Metamorphic Complex; Miller 2008b)is
located in north-northeastern Namibia and extends eastward into
Botswana, where it is inferred to continue as the Quangwadum
Inlier (QI; Fig. 2.6) and to include the Chihabudum Complex.
A singular combination of geophysically evident regional-scale
features occurs within the Inlierthat is, an annular zone of high
magnetization encompasses a low-magnetic central zone, refer-
red to as the Omatako remanent anomalies (OR; Fig. 2.6; Corner
2000), which suggest deep, probably remanent, sources. These,
and the encompassing annular high-magnetic zone, are collec-
tively referred to as the Omatako Ring Structure (ORS; Fig. 2.7).
The ORS occurs at the intersection of a number of major linea-
ments, including the Omaruru and Kudu Lineaments, and lies
within the swathe of the west-northwest-trending Botswana dyke
swarmthatcrossesthesubcontinent(BDSL;Fig. 2.6). A number
of deep, roughly linear magnetic anomalies radiate to the east,
north and west of the OR/ORS, forming part of a much larger
regional structural feature. The Grootfontein Inlier is mostly
characterized by strongly magnetic gneisses. The radial
anomalies are thus interpreted to arise from radial faults within
the Complex, with a focus on the ORS. The area is covered by
Kalahari and Karoo sediments, as well as by Karoo basalts, thus
the origin of the ORS and associated features is unknown.
Possible causes include a major volcanic eruptive centre, or even
a meteorite impact site.
McCourt and Jelsma (this volume) discuss the Angolan
Craton and, of relevance here, age aspects of the Congo
Craton in Namibia. They describe the continental crust
forming the Grootfontein Inlier as comprising plutonic rocks
of dominantly alkaline/calc-alkaline composition. The pro-
tolith of granitic gneisses from the related Tsumkwe and
Quangwadum Inliers (TI, QI; Fig. 2.6) have been dated at
2022 ±15 Ma (Hoal et al. 2000) and at 2051 ±1Ma
(Singletary et al. 2003), which are compatible with inherited
zircon grains at 2052 ±44 Ma and 1987 ±4 Ma in granite
and felsic lava in the Kamanjab Inlier (KI; Fig. 2.6). They
take this as evidence of correlation between the older crust,
interpreted to be present at a depth below the KI, and the
granite gneisses exposed in the Grootfontein, and related
Quangwadum and Tsumkwe, Inliers. Archaean ages within
the Palaeoproterozoic basement inliers of northern Namibia
and Botswana have been reported in the Epupa Inlier (e.g.,
2585.4 ±1.2 Ma and 2645 ±6 Ma; Seth et al. 1998; EI;
Fig. 2.6) and in the Tsodilo Hills area associated with the QI
(2548 ±65 Ma, Gaisford 2010). The Archaean localities
are shown as red stars in Fig. 2.6.
2.2.3.5 Crustal Magnetization
Interpretation of the aeromagnetic and gravity data covering
the Witwatersrand Basin (Figs. 2.6), and its potential
extensions to the south and west of the main basin, led to
some fundamental new insights into the underlying craton
(Corner et al. 1986a,b; Corner et al. 1990). As part of these
studies, long-wavelength magnetic anomalies within the
craton, relating to deep sources, were interpreted through
forward modelling. The Vredefort Impact Structure (VIS;
Fig. 2.6) has an annular highly magnetic zone in its gneissic
basement core which, accepting the crust-on-edge model for
the structure, would thus lie some 8 km beneath the West
Rand Group of the Witwatersrand Basin. This zone was
interpreted, using a typical P-wave velocity for the base-
ment, to correspond with seismic reectors in the
granite-gneissic basement evident in a reection seismic line
traversing the basin west of VIS (Durrheim et al. 1991).
These reectors are parallel, or subparallel to the base of the
basin. Projection of this intermediate-crustal magnetic,
reective zone, through forward modelling of the magnetic
data, to its suboutcrop beneath the Karoo Sequence to the
southwest of the Witwatersrand Basin, suggested that the
deep, high-amplitude semilinear magnetic anomalies trend-
ing northward from Colesberg (Co-L; Fig. 2.6) were due to
this same level of magnetized basement. Drilling by a
mining company (Goldelds SA, pers. comm.) conrmed
this interpretation. Corner et al. (1986c) named this
anomalous belt the Colesberg Trend, here termed Lineament
30 B. Corner and R. J. Durrheim
(Co-L; Fig. 2.6). This is inferred to be the locus of an early
orogenic belt west of the Witwatersrand Basin, along which
a major section of the upper crust has been eroded, possibly
being the main source of the sediments that lled the western
portion of the basin.
The conrmation of this interpretation led Corner (1998)
to propose a model that attributed many of the
high-amplitude, regional long-wavelength magnetic anoma-
lies to arise from intermediate-crustal magnetization, partic-
ularly to the west of the Witwatersrand Basin, including the
Colesberg Lineament (Figs. 2.2 and 2.6). Comparisons were
made with other cratonic areas, particularly with studies of the
structure of Russian cratons based on numerous deep crustal
seismic refraction proles (Pavlenkova 1987). These studies
led Pavlenkova to propose a generalized three-layer craton
model, distinguished on the basis of geological structure,
seismic boundaries and P-wave velocities. Corner (1998)
compared the above magnetization model with both the
Pavlenkova three-layer craton model and the deep electrical
resistivity results for southern Africa (Van Zijl 1978; De Beer
and Meyer 1984; De Beer and Stettler 1988). Corner (1998)
noted that the three-layer craton model could fully explain the
extremely low resistivities observed in the deeper crust,
which would thus occur below the upper, brittle, resistive
crust, in an underlying hydrofractured, aseismic, probably
uid-lled middle crust (in the three seismic-layer model)
where horizontal displacement stresses predominate. It is in
this electrically conductive intermediate-crustal zone that
Corner proposed the development of magnetization, through
the growth of magnetite from iron in the protolith at elevated
temperatures (below the Curie temperature for magnetite), in
the presence of uid.
Invoking the presence of an intermediate-crustal zone of
magnetization, not necessarily ubiquitous throughout the
craton but certainly prevalent in many areas particularly
along major structures that may have allowed uid move-
ment, readily allows the interpretation, through forward
modelling of both magnetic and gravity data sets, to explain
many of the deeper features, such as the long-wavelength
anomalies west of the Witwatersrand Basin near Wol-
maransstad and Vryburg, the Colesberg Lineament (Co-L;
Fig. 2.6), the Strydenburg Lineament to the west of the
Colesberg Lineament (St-L; Fig. 2.6) and possibly portions
of the Kal-L in Botswana (Corner 1998).
2.2.4 The Witwatersrand Basin
The main Witwatersrand (Wits) Basin is situated roughly
centrally on the Kaapvaal Craton (Fig. 2.6). Economic
exploitation since the discovery of gold in 1886 has seen the
development of more than 150 mines, some of which have
yielded uranium as a by-product. In terms of value of metal
recovered, the basins mineralization must rank as one of the
most valuable mineral deposits ever found. However, more
than 90% of the basin is covered by younger sequences,
ranging from Neo-Archaean, through the Palaeoproterozoic,
to Phanerozoic. Geophysics played a critical role, at an early
stage, in the discovery of new mines, in particular through
the application of magnetic and gravity techniques. More
recently, reection seismic techniques have also been very
successfully applied, both to locate new extensions and to
map, in 3D, structures on a mine scale (e.g., Pretorius et al.
1989). This technology was further applied to mapping and
evaluation of the Bushveld Igneous Complex in the search
for platinum (e.g., Pretorius et al. 2010).
In terms of mapping the basin under cover, Borchers
(1964) produced the rst geological map based on outcrop,
mining and drilling data. A more recent map, based mostly on
additional geological data acquired since that time, partially
constrained by geophysical data, was published by Pretorius
(1986). Using this map and associated data as a base to work
from, Corner and Wilsher (1989) conducted a rigorous rein-
terpretation of the basin, using both the aeromagnetic and
ground gravity data to further upgrade the mapping. Corner
et al. (1986a,b) also published the rst ever digitally printed
colour images, in the public sector, of the aeromagnetic and
gravity data covering the basin. The mapping of Corner and
Wilsher (1989) is replicated in Fig. 2.6. Although older than
the overlying Ventersdorp and Transvaal Supergroups, the
main Wits Basin and outliers are placed on top of these in
Fig. 2.6 so as to provide the reader with a view of their actual
extent and disposition. Corner et al. (1986c) extended their
interpretation southwards, covering the southern portion of
the Kaapvaal Craton. They identied the possible presence of
outliers of Witwatersrand rocks southeast of the main basin in
the Bethlehem area, and south of Bloemfontein, conrming
what many geologists had proposed in the past. High gold
prices at the time resulted in extensive exploration activities
commencing in these areas.
The ensuing quest, by major and junior mining companies,
for extensions of the Wits Basin in the Bethlehem area and south
of Bloemfontein, using aeromagnetic, gravity and extensive
seismic reection surveys, resulted in many successful boreholes
being drilled which intersected Witwatersrand Supergroup
rocks, and Waterberg-equivalent rocks, in previously untested
areas. Unfortunately, most of the intersections were in the barren
West Rand Group and, in time, these activities were terminated.
One such exploration programme was conducted by AfriOre
(Pty) Ltd, which built on the data available from other compa-
nies that had withdrawn from the area. The rst author (Corner)
was a member of a team that interpreted these data as well as a
more recent aeromagnetic survey own by AfriOre (McCarthy
et al. in preparation). Drilling intersected both Witwatersrand
and Transvaal Supergroup rocks east and southeast of the
Trompsburg Complex. Integrated interpretation, in a team
2 An Integrated Geophysical and Geological Interpretation 31
approach, of all data resulted in the delineation of major Wits
and Transvaal sub-basins underlying the Trompsburg Complex
south of Bloemfontein, extending eastwards to Lesotho and to
the Bethlehem area north of Lesotho (Fig. 2.6;McCarthyetal.
in preparation). Both the Wits and the Transvaal sub-basins are
separated from the main Wits Basin by the Bloemfontein Arch
(BA; Fig. 2.6). This is a major contribution, built on all past
exploration programmes, to the mapping of Archaean and
Palaeoproterozoic rocks beneath an extensive cover of Karoo
Supergroup rocks, in excess of 1000 m in thickness, in the
southern portion of the Kaapvaal Craton.
2.2.5 Xade Complex
The Xade Complex (XC), situated in central Botswana, is a
large singularly anomalous feature in the aeromagnetic and
gravity images of southern Africa (Figs. 2.2,2.4 and 2.6),
occurring under a complete cover of sediments of the Kalahari
Group and Karoo Supergroup, including Karoo volcanics in
places, with a combined thickness that varies from 220 to
1000 m, probably extending to greater depths in the north
beneath the Passarge Basin (PB). As such, it has drawn much
attention both academically and from a minerals exploration
point of view. It was rst identied during the regional
aeromagnetic survey of the country in 19751977 (Reeves
1978; Meixner et al. 1984; Figs. 2.2 and 2.4). The Xade
Complex was originally interpreted to comprise a
high-amplitude kidney-shaped zoned magnetic anomaly with
two semilinear anomalies extending to the northwest and
northeast in a Y-shaped form (e.g., Meixner et al. 1984). It is
also evidenced by a coincident Bouguer gravity anomaly.
Historical work was limited, with only three cored-boreholes
having been drilled. Two of these were drilled as part of the
Kalahari Drilling Project in the early 1980s, following inter-
pretation of the aeromagnetic data (Meixner et al. 1984). One
borehole intersected gabbroic rocks at 815 m, and the other a
weathered basalt at 419 m, passing into dolerite. A third
borehole was drilled by the Anglo American Corporation
(Ambot 1998), which held exploration licences over the
complex in the late 1990s. Amygdaloidal lava was intersected
at 621 m, passing into dolerite, and shales assigned to the
Waterberg Group. An U-Pb zircon age of 1109.0 ±1.3 Ma,
which is coeval with the Umkondo Igneous episode, has been
published for the gabbroic unit intersected in the rst borehole
(Hanson et al. 2004).
A junior exploration company, Manica Minerals Ltd, held
prospecting licences over the Xade Complex from 2005. Its
exploration activities, conducted in joint venture partner-
ships with two other companies, included use of both the
medium-resolution aeromagnetic data, acquired under con-
tract to the Botswana Geoscience Institute at a 250 m line
spacing, and the Institutes ground Bouguer gravity data. An
additional higher-resolution aeromagnetic survey was con-
ducted over a portion of the complex. Detailed ground
gravity surveys and time domain electromagnetic soundings
were conducted on selected proles traversing the complex.
Forward modelling of the magnetic and gravity proles
helped constrain the zoning and structure of the complex.
Three boreholes were subsequently drilled (Corner et al.
2012). A parallel interpretation was conducted using the
Institutes data, but without the benet of the further
exploration data, by Pouliquen and Key (2007).
The interpretation of the work of Corner et al. (2012)
showed for the rst time that the Xade Complex comprises two
lobes: a Southern Lobe (XC-SL; Fig. 2.6), which is the his-
torically identied kidney-shaped zoned magnetic anomaly, as
well as a hitherto unrecognized large Northern Lobe (XC-NL;
Figs. 2.2 and 2.6). The NL is mostly deeply buried in the north
and northwest beneath the Neo-Proterozoic PB, as evidenced
by deep magnetic and gravity anomalous sources, but its
southern and eastern margins partially suboutcrop beneath
Karoo sediments, forming the Y-shaped anomalies north of the
SL. An apparent transgressive contact between the two lobes is
indicated by the aeromagnetic data, suggesting that the NL may
be slightly younger. Inversion depths to the complex range
from 220 to 1000 m beneath the Kalahari and Karoo sediments,
and greatly in excess of this beneath the PB. Forward mod-
elling, of both magnetic and gravity data along a number of
sections traversing the both lobes, indicates that they are
lopolithic features with a depth extent of approximately 4 km.
This is supported by the interpretation of Pouliquen and Key
(2007) for the SL, as well as from dips derived from the drill
cores. The total of four boreholes drilled into the SL shows that
it comprises a volcanic sequence with subordinate gabbro
(Corner et al. 2012). The basalts are partly highly magnetic,
giving rise to the zoned high-amplitude anomalies of the larger
kidney-shaped anomaly, and partly magnetically subdued
owing to less magnetic basalts that appear to underlie the rocks
of the main SL anomaly. The three boreholes drilled into the NL
margins, as published to date, indicate that it comprises a tex-
turally heterogeneous and magmatically differentiated
sequence of gabbroic rocks, with minor dioriticand monzonitic
rocks, as well as basalt (Corner et al. 2012).
The interpretation of Corner et al. (2012) has also iden-
tied a dyke system associated with the NL, which may
represent either feeder or exit magmatic conduits. The
interpretation further shows that the Xade Complex is
located in a craton margin settingthat is, the SL lies on the
northern margin of the Kaapvaal Craton, whereas the eastern
suboutcrop of the NL extends along the margins of the
Kaapvaal and Zimbabwe cratons (Fig. 2.6). The combined
extent of both lobes of the Xade Complex is approximately a
third the size of the Bushveld Complex, making it the largest
Late-Mesoproterozoic magmatic complex in southern
Africa, with a potential for nickel-copper mineralization.
32 B. Corner and R. J. Durrheim
2.2.6 Tectonostratigraphic Zones
of the Damara-Chobe Orogenic Belt
2.2.6.1 Regional Aspects of the Interpretation
The tectonostratigraphic zones of the Namibian Damara Belt
and its Mesoproterozoic basement, extending eastward into
Botswana as the Damara-Chobe Orogenic Belt, were map-
ped under Kalahari cover to the eastern border with Bots-
wana by Corner (2000,2008), based on published mapping
and stratigraphy, known bounding regional structures, and
the overall internal magnetic signature of the zones. The
aeromagnetic data used at that time was the national regional
data set compiled from surveys own at line spacings
varying between 1 and 4 km. The interpretation has since
been rened and updated, as shown in Fig. 2.6, being more
accurate in local detail, based on the higher-resolution
aeromagnetic data (200 m ight line spacing) and mapped
geology at a 1:250 000 scale (all data from the Geological
Survey of Namibia). Ongoing extension of this work into
Botswana was facilitated by the publication of the 1998
edition of the National geological map of Botswana, in both
digital and hard copy form; the work of Key and Ayers
(2000), Singletary et al. (2003) and Rankin (2015); and the
availability of the medium-resolution National aeromagnetic
data of Botswana. These products have been reviewed,
revised in places, and integrated with the Namibian inter-
pretation shown in Fig. 2.6. Brief lithological, stratigraphic
and structural summary descriptions of each tectonostrati-
graphic zone shown in the interpretation map of Fig. 2.6
follow below. For the Damara Belt in Namibia, these are
based on detailed descriptions by Corner (2008) and Miller
(2008a). Furthermore, an aeromagnetic survey covering the
continental shelf of Namibia, own under contract to
NAMCOR (Fig. 2.2), although relatively coarse, has facili-
tated mapping of the extension of some of the tectonos-
tratigraphic zones offshore, up to the seismically interpreted
hinge zone (Sect. 2.2.9). It should be noted that the sys-
tematics of stratigraphic classication generally do not take
cognizance of geophysical responses. Stratigraphic bound-
aries may thus differ from geophysically apparent bound-
aries. The Damara Sequence does, however, show a strong
correlation in that the lower units, which include diamictites,
psammitic rocks and volcanics of the Nosib and lower
Swakop Groups, are often strongly magnetic, while the
overlying pelitic and carbonate sequences of the Swakop
Group tend to be relatively subdued magnetically.
2.2.6.2 Mesoproterozoic Sinclair-Rehoboth Zone
The Mesoproterozoic Sinclair-Rehoboth Zone (SRZ;
Fig. 2.6) comprises the southern Damara Basement and is
dominated by volcano-sedimentary cycles of the Rehoboth
Group and the Sinclair Supergroup, as well as by granitic
and mac rock suites. The overall magnetic signature of this
terrane is visibly different from that of the Gordonia Sub-
province to the south, being dominated by relatively
high-amplitude, curvilinear magnetic anomalies arising from
the volcanic sequences. Some of the granites, granodiorites
and orthogneisses are also magnetic. In contrast, the Gor-
donia Subprovince displays highly variable magnetic
responses owing to a range of rock types, which include
ortho- and para-gneisses, granites, granodiorites, ultramacs,
charnockites, metasediments and metavolcanics of the
Namaqua Metamorphic Complex, the Orange River Group
and the Vioolsdrif Intrusive Suite. Both the Gordonia Sub-
province (the northern extension of the Namaqua Province;
GSP; Fig. 2.6) and Sinclair Supergroup are characterized by
numerous remanent magnetic anomalies, which are dis-
tinctly different from the induced anomalies arising from the
Sinclair basalts. These result from gabbroic, ultramac and
charnockite bodies, although their individual signatures are
indistinguishable. Their magnetization is of uncertain age
but is expected to be post-Sinclair, possibly being set at the
time of emplacement during the c. 1000 Ma Namaqua
metamorphic event.
2.2.6.3 Pan-African Tectonostratigraphic Zones
Gariep Group
The current geophysical interpretation has not as yet been
extended to include the Gariep Belt, thus what is shown in
Fig. 2.6 is based entirely on published mapping (e.g., Frimmel
2008). However, the offshore geophysical data sets, although
of low resolution, have facilitated mapping of the offshore
extensions of the Port Nolloth Zone, and the Marmora Terrane
with its associated Schakalsberge volcanics (Fig. 2.6).
Tsumis Group of Namibia and the Ghanzi Group of
Botswana
The Tsumis Group comprises sediments of the Doornpoort,
Eskadron and Klein Aub Formations, which post-date the
period of large-scale Sinclair-Rehoboth Mesoproterozoic
igneous activity. The group has been considered as either
encompassing both the lower Nosib Group and the Klein
Aub Formation (e.g., Schalk 1988), or as occurring within
the Sinclair Supergroup (e.g., Miller 2008c). However,
Hoffman (1989a) and Becker et al. (2005), based on eld
observations and new geochronological evidence, place the
Tsumis Group in unconformable contact with, and thus
younger than, the Sinclair Supergroup. Although overlaid
para- to dis-conformably by the Damaran Nosib Group, the
Tsumis Group is considered to constitute the lowermost part
of the Damara Sequence (Hoffmann 1989a). The latter
interpretation, although equivocal, is favoured in this
review and is shown as such in Fig. 2.6.
2 An Integrated Geophysical and Geological Interpretation 33
Intra-Tsumis stratigraphic units display clear magnetic
signatures, which allow mapping of these units under cover,
eastward in Namibia and into Botswana. The main strati-
graphic units, with the Namibia nomenclature given rst,
followed by the Botswana nomenclature and general mag-
netic signature, are: Doornpoort Formation = Ngwako Pan
Formation (low, quiet magnetic response); Klein Aub For-
mation = DKar Formation (strongly magnetic fabric);
Nosib Group = Mamuno Formation (intermediate to low
magnetic fabric). These units, mapped with outcrop control
where available, are grouped together in Fig. 2.6 as the
T-G-C-B (Fig. 2.6). This belt is also commonly referred to
as the Kalahari Copperbelt in view of its numerous
copper-silver occurrences (e.g., Maiden and Borg
2011). The T-G-C-B thus constitutes the southern ank of
the Damara-Chobe Orogenic Belt.
Southern Margin Zone of Namibia and Its Extension
into Botswana
Miller (1983) describes the Southern Margin Zone (SMZ) as
comprising two subzones, a southern, less intensely
deformed subzone containing mainly thrust slices of
pre-Damaran rocks, and a northern subzone consisting of
complex thrust sheets containing both Damaran and
pre-Damaran rocks. Stratigraphic units present in the SMZ
thus include pre-Damara basement, gneissic basement, the
Nosib Group and the lithologically variable passive margin
succession of the Hakos Group (Miller 2008a). The northern
subzone of the SMZ contains relatively little-deformed cover
sequences of the early Damaran Nosib Group along the
northeastern margin (Hoffman 1983). The SMZ is bounded in
the south by the Frontal Thrust and in the north by the Gomab
River Line (Corner 2008; Miller 2008a). High-amplitude
magnetic fabric is associated with the Chuos diamictites
within the SMZ. Magnetic fabric, characterized by lower
levels of magnetization, is associated with the Nosib
Group. This is particularly evident in the higher-resolution
data.
The extension of the SMZ under cover eastwards to the
Botswana border, as interpreted by Corner (2008) working
from the western and central and delineation thereof by
Miller (1983,2008a), shows that it continues into Botswana
as what has been mapped as the Roibok Formation by Key
and Ayres (2000). The interpretation of Rankin (2015) does
not appear to support continuity of the SMZ throughout the
southern Damaran boundary, but does support the continu-
ation of a portion thereof in eastern Namibia, associated with
high-amplitude magnetic anomalies. His interpretation
favours mapping the extension of the Botswana Roibok
Formation into Namibia, where it also inter-ngers with
these high amplitude magnetic anomalies. This package is
considered here to be an integral part of the SMZ, extending
into Botswana as mapped in Fig. 2.6, recognizing that the
Roibok Formation may only be part of a much more com-
plex SMZ extension in Botswana.
Southern Zone or Khomas Zone
The Southern Zone (SZ), bounded by the SMZ in the south
and the Okahandja Lineament in the north, comprises a
thick, deformed succession of Kuiseb schists arising from
both active and passive margins of the Khomas Sea, which
are placed into separate formationsthat is, the Khomas and
Hureb Formations (Miller 2008a). The SZ is also often
referred to as the Khomas Zone or Khomas trough. This
succession mostly shows a low-order relatively quiet mag-
netic fabric in the higher-resolution data, but it is signi-
cantly more magnetic in the lower stratigraphic sequence,
particularly north and northeast of Windhoek, and also due
to the magnetic quartzites associated with the Matchless
Amphibolite Belt.
The eastward extension of the SZ into Botswana, based
on the magnetically quiet signature, follows on north of the
SMZ (Roibok Formation) in Botswana, where it is seen to
pinch out (Fig. 2.6). The offshore extension of the SMZ and
SZ, southwards toward the Gariep Belt, has been interpreted
as a package, from both the offshore aeromagnetic and Free
Air gravity data sets. The location of the boundary between
these two zones may not be exact as a result of the poor data
quality, but the trend of the package is clear, also abutting
against the hinge zone (Fig. 2.6).
Deep-Level Southern Zone
The magnetic signature of the Southern Zone (SZ) and
southern Central Zone (SCZ, described below) changes
dramatically in the Steinhausen area and northeast thereof
(note the Steinhausen anomaly, SMA, for location in
Fig. 2.6). Here, the Kuiseb Formation schists display
high-amplitude magnetic anomalies within the SZ. Magnetic
rock types in the area also include epidosite and gabbro,
uncommon for the Kuiseb schist sequence elsewhere. To the
north, a belt of very high-amplitude magnetic anomalies
striking roughly eastwest and cutting across the SZ and
SCZ fabric correlate with FeMn reefs, and both magnetic
and glassy quartzites. Hoffman (1989b) and K Kasch (pers.
comm. 2006) have questioned the inclusion of these strata in
the Damara Supergroup and have suggested that they form
part of the pre-Damara basement. They are thus tentatively
mapped in Fig. 2.6 as the Deep-level Southern Zone (Corner
2008), characterized by intense thrusting and deep strati-
graphic levels (Kasch 1986). The Deep-level Southern Zone
is largely centred between the Okahandja and Kudu Linea-
ments (Ok-L and Ku-L; Fig. 2.6), continuing east of the
latter with a signicant change in strike, suggesting a pos-
sible fault throw controlled by the latter lineament.
34 B. Corner and R. J. Durrheim
Southern and Northern Central Zones
The Central Zone is subdivided into northern and southern
partsthat is, the Northern Central Zone (NCZ) and the
Southern Central Zone (SCZ), as shown in Fig. 2.6both of
which are characterized by dome structures with an overall
northeast elongation (Miller 2008a). Voluminous syn- to
post-tectonic granite plutons occur in both. The SCZ is a
regional horst, bounded by the Okahandja and Omaruru
Lineaments (Ok-L and Om-L; Fig. 2.6), and characterized
by high-amplitude magnetic anomalies arising from the
exposure of deeper-level sequences, primarily of the Nosib
Group, lower Swakop Group (Chuos Formation diamictites)
and the Meso- to Palaeo-Proterozoic basement. These units
are exposed in dome and anticlinal structures with a pro-
nounced northeasterly trend. The relatively magnetically
quiet lower Swakop Group carbonates and schists are pre-
served in the intervening synclines. Some of the granite
phases are magnetic, particularly if derived in part from the
basement and Nosib Group. In the western portion of the
SCZ, where metamorphic grades are higher, the Etusis and
Khan Formations of the Nosib Group, as well as the granitic
derivatives therefrom (and also from basement), are strongly
magnetic with high-amplitude anomalies that have retained
the Damaran remanence (Corner 1983). This magnetic sig-
nature of the Khan Formation constitutes an important
geophysical marker horizon for the uraniferous granites
(e.g., at the Rössing and Husab mines), which mostly occur
immediately above it in the Rössing Formation.
Much higher stratigraphic levels are exposed in the NCZ
which largely comprises rocks of the Swakop Group. It is
bounded by the Omaruru and Autseib Lineaments (Om-L
and Au-L; Fig. 2.6). Regional-scale downthrow, or rapid
deepening, to the north of the Omaruru Lineament, as
also modelled magnetically by Corner (1983), has preserved
this thick succession of higher stratigraphic-level Karibib
Formation carbonates and Kuiseb Formation schists,
deformed in numerous basin and dome structures. The dome
structures are generally not cored by granitic basement but
rather by Karibib Formation marbles and Usakos Subgroup
schists, calcsilicates and marbles. The Swakop Group rocks
are relatively magnetically inert, giving an overall quiet
magnetic signature to the NCZ. Some low-order fabric is
nevertheless seen in the higher-resolution data. Extensive
syn- and post-tectonic granite emplacement has taken place,
although less so in the areas where the highest levels of the
Kuiseb Formation are exposed (Miller 2008a).
The offshore continuation of the SCZ and NCZ, westward
up to the hinge zone where they are dramatically truncated,
is clearly visible in the aeromagnetic image of Fig. 2.2.
Eastwards, the northeast trend of the Damaran Belt is of
particular interest as it changes dramatically east of the Kudu
Lineament (Ku-L; Fig. 2.6) in the following respects:
The SCZ terminates against the Ku-L, whereas the NCZ
changes strike east of the Om-L trending east-southeastwards
up to the Ku-L, thereafter regaining a northeasterly trend,
pinching out in Botswana (as does the SZ). An alternative
interpretation could be considered, given the relatively quiet,
thus ambiguous, magnetic signature of both the SZ and the
NCZthat is, that the NCZ also terminates against the Ku-L.
ThuswhatisshowninFig.2.6 as NCZ (blue) east of the
Ku-L may in fact constitute part of the SZ.
Recent interpretations of an area, focused on mineral
exploration, which straddles the Ku-L in the vicinity of the
Steinhausen Magnetic Anomaly (SMA; Fig. 2.6)reveala
signicant change in the Damaran stratigraphy eastwards as
the Ku-L is crossedthat is, with a much deeper level of
erosion to the west thereof, exposing basement domes such
as the Ekuja dome and deeper-level southern zone stratig-
raphy (e.g., Corner 2008;Naudé2012; K. Hartmann, pers.
comm.). Thus, not surprisingly, the Matchless Amphibolite
Belt (MAB; Fig. 2.6) terminates against the Ku-L, and is not
expected to continue on-strike further to the east.
The width of the Damara-Chobe Mobile Belt taken from
the SMZ to NMZ is approximately 350 km, west of the
Ku-L. The NMZ, NZ, NCZ and SCZ appear to truncate
against, or are in disconformable contact with, the Groot-
fontein Inlier (GI; Fig. 2.6). East of the Ku-L, this width
decreases dramatically to less than 150 km, striking roughly
parallel to the contact with the GI, pinching out altogether as
it progresses further northeastwards into Botswana towards
the TsodiloLineament. Similarly, the OkahandjaLineament
(Ok-L) appears to terminate against the Ku-L and cannot be
denitively followed east thereof in the aeromagnetic data.
Northern Zone
The Northern Zone (NZ; Fig. 2.6), comprising rocks of both
the Nosib and Swakop Groups, has been thrust northwards
onto Otavi, Mulden and pre-Damara rocks along the
Khorixas-Gaseneirob Thrust (KG-L; Fig. 2.6; Miller 2008a).
The Autseib Lineament (Au-L; Fig. 2.6), in part including
the Autseib Thrust, forms the southern boundary of the
highly and complexly deformed NZ, which shows an overall
strongly magnetic signature owing largely to the highly
magnetic diamictites of the Chuos and Ghaub Formations
and the mac volcanics of the Askevold Formation. The NZ,
together with the northern part of the NCZ, formed the oor
of the Outjo Sea during spreading (Miller 2008a).
Northern Margin Zone
The NMZ, constituting in-part the Khorixas-Gaseneirob
Lineament (the inferred Congo Craton boundary), is a nar-
row transition zone which Miller (2008a) describes as the
northern platform foreslope region where the rather uniform
2 An Integrated Geophysical and Geological Interpretation 35
facies of the Otavi Group to the north become more variable
and include deep-water-facies carbonates of the Otavi
Group. Mulden Group rocks are preserved in tight synclines
within this zone, which, although generally magnetically
quiet, display high-amplitude remanent magnetic anomalies.
Corner (2008) ascribes these to secondary magnetization
arising from the development of pyrrhotite in the phyllites
through the passage of uids along the lineament.
Northern Otavi Platform, and the Tsodilo Lineament
Damaran rocks are only exposed along the southern and
western edges of the Northern Platform (NP) but continue
northwards and eastwards below Karoo and Kalahari cover.
The limits of the NP in Namibia, as shown in Fig. 2.6, are
determined by the distribution of the shallow-water facies of
the Otavi Group, extending westwards to the Sesfontein
thrust (ST; Fig. 2.6; Miller 2008a). Predominant east
west-trending anticlinal structures are seen in the magnetic
data in the southern sector of the NP. These separate
expansive magnetically quiet areas that comprise higher
stratigraphic levels, including Mulden Formation strata, in
the synclines. The antiforms have an anomalous magnetic
signature largely due to the exposure, or shallow suboutcrop,
of the diamictites and possibly volcanics associated with the
lower Damara strata. The tight folding of the antiforms close
to the NMZ becomes progressively more open to the north.
Geophysical mapping of the eastward extension of the NP
under cover in Namibia, and into Botswana, is complicated by
the largely highly magnetic basement rocks of the Grootfontein
Inlier. Outcrops of the NP do nevertheless occur at the Aha Hills
close to the Botswana border, which allows continuity to be
inferred. The intervening area is hatched in Fig. 2.6 to indicate
the ambiguity of continuity owing to the low-magnetic signature
of the overlying NP carbonates. On strike to the northeast of the
Aha hills, in Botswana, is the Xaudum Group, which correlates
with the Aha Hills Formation (e.g., Rankin 2015), and hence
with the NP in Namibia. Extension of the NP into Botswana is
indicatedassuchinFig.2.6. The Tsodilo Hills Group (Figs.
2.6a, b), however, is of uncertain correlation. Miller (1983),
Breitkopf (1988)andBűhn et al. (1992) correlate it with the
Chuos Formation in Namibia in view of the abundant iron for-
mations and ferruginous quartzites. Rankin (2015), based on
further interpretation using potential eld data, concluded that the
Tsodilo Hills Group cannot be condently correlated with a
single tectonostratigraphic zone of the Damaran Belt, and that it
may even be a correlative of units in Angola or Zambia. What
appears to be singularly unique in the area is the dramatic change
in structural style displayed by the Tsodilo Hills Group, from the
expected northeast trend of the Damara-Ghanzi-Chobe Mobile
Belt to tight folding and faulting striking north-northwest (e.g.,
Rankin 2015). The Tsodilo Hills Group is thus specically
separately colour coded in Fig. 2.6.
The question may thus be asked as to the reason for this
dramatic change in structural style. In the current inter-
pretation, we single out the regional Bouguer Gravity low
in the Botswana Geosciences Institute data, with which the
Tsodilo Hills Group is associated (Fig. 2.4; see also Rankin
2015), as providing the clue. We have the further benetof
the WGM Bouguer Gravity data (Fig. 2.5), which shows
that the gravity low trends further north-northwestwards
and northwards into Angola. The gravity low thus appears
to hallmark a major, roughly northsouth lineament which
we have termed the Tsodilo Lineament (Ts-L; Fig. 2.6).
That it constitutes a major regional structural zone is not
only evidenced by the structural style displayed by the
Tsodilo Hills Group and environs, but also provides an
explanation as to why the regional east-northeast-
trending magnetic low in northern Namibia, correlated
with the southern Congo Craton margin, abruptly termi-
nates in northwestern Botswana at the Tsodilo Hills gravity
low. Viewed in a regional context, it is possible that the
Ts-L was the northward continuation of the Kalahari
Lineament (Kal-L) in pre-Pan-African times, as they are
virtually on strike with each other, given some local
north-northwest deformation in the Tsodilo Hills area. The
continuity between the two may thus have been inter-
rupted by the Damaran Orogeny (see also Sect. 2.2.8).
Kaoko Belt
The Kaoko Belt is subdivided into four sub-
zones (Goscombe et al. 2003,2005; Miller 2008a): the
Southern, Eastern, Central and Western Kaoko Zones:
The pelitic sequences of the Southern Kaoko Zone (SKZ),
which are tightly folded in northsouth-striking chevron
folds, display a low-amplitude magnetic fabric, giving an
overall quiet appearance to this zone which nevertheless
has enabled better denition of this subzones boundaries.
In contrast, the strong, roughly north-northwest magnetic
fabric of the Western, Central and Eastern Kaoko Zones
results from complex deformation and variable grades of
metamorphism (Goscombe et al. 2003).
The Eastern Kaoko Zone (EKZ) comprises upright folds
of subgreenschist-facies shelf carbonates (Goscombe
et al. 2003). These rocks are relatively non-magnetic, but
where lower stratigraphic levels are exposed or are in
shallow suboutcrop, a strong magnetic fabric is seen.
Miller (2008a) points out that the stratigraphy of the EKZ
is almost identical to that of the Northern Platform (NP),
but that the two regions are structurally distinct. This is
also observed in the magnetic data, which shows a dis-
tinct magnetic, and hence tectonic, eastern boundary
between the EKZ and the NP. The Sesfontein Thrust
(ST) marks the western boundary of the EKZ.
36 B. Corner and R. J. Durrheim
The Central Kaoko Zone (CKZ) comprises east-vergent
nappes of Swakop Group passive margin rocks (Goscombe
et al. 2003;Miller2008a). Within the CKZ, the Damara
stratigraphic units, the Okapuka Formation and the
Palaeoproterozoic crystalline basement of the Epupa Inlier
show variable magnetic responses, in some places subdued
and in other places with a slight magnetic fabric.
Higher-amplitude anomalies arise from the lower Swakop
Group diamictites and amphibolites, and from amphibolitic
and gabbroic units within the pre-Damara basement.
The CKZ has no clear eastern boundary with the EKZ as
evidenced in the magnetic data. Based on geological pre-
mises, the boundary is thus set, as above, at the Sesfontein
Thrust (ST; Fig. 2.6;Miller2008a). The western boundary
of the CKZ is the Purros Shear Zone (PuSZ; Fig. 2.6),
which comprises a series of closely spaced, steeply
westward-dipping ultramylonites (Miller 2008a).
The Western Kaoko Zone (WKZ) is predominantly a
deep basin facies sequence of high metamorphic grade,
intruded by numerous granites, which has experienced
intense wrench-style deformation along steep,
crustal-scale shear zones (Goscombe et al. 2003). Much
of the magnetic fabric results from the mylonitic shear
zones. The Purros Shear Zone constitutes the eastern
boundary of the WKZ, which is further subdivided into a
Coastal Terrane and an Orogen Core. The latter com-
prises a number of internal subdomains, which are sep-
arated by the Three Palms Mylonite Zone (Goscombe
et al. 2005; Miller 2008a).
2.2.7 Rehoboth Terrane
2.2.7.1 Sedimentary Sequences
The Rehoboth Terrane (RT) has a unique expression in the
aeromagnetic image of southern Africa (Figs. 2.2 and 2.6). It
is hallmarked by long-wavelength anomalies that reect
deep magnetic basement sources, an overlying thick package
of magnetically-inert sedimentary rocks including
pre-Nama, Nama, Karoo and Kalahari sequences, and some
shallow, high-frequency magnetic anomalies arising from
Karoo basalts and sills. The interpreted deep-source mag-
netic and gravity anomalies which underlie the above sedi-
mentary package (Corner 2008), are shown in Fig. 2.6.
The RT, which forms the western Palaeoproterozoic core
of the Kalahari Craton (Jacobs et al. 2008; Fig. 2.1), is
bounded in the east by the Kal-L. The change in depth to
magnetic basement on either side of the Kal-L is dramatic,
from *500 m east thereof to 810 km west thereof (Corner
2008). Isopach contour levels, derived by Corner (2008)
from a depth-to-basement analysis, of >4000 m (i.e.,
pre-Karoo sediments) and >6000 m, are shown in Fig. 2.8.
This analysis is in agreement with the earlier, more regional,
or localized, interpretation of Reeves (1978) and of an
interpretation by Petro-Canada (a hydrocarbon exploration
group which conducted seismic surveys in western Bots-
wana and an aeromagnetic interpretation; Botswana Geo-
science Institute archives, Wright and Hall 1990). In the
southwest, the RT is bounded by the northwestward exten-
sion of the Namaqua Lineament (Naq-L; Fig. 2.6;an
extension of the Namaqua-Natal Front), which also shows a
dramatic increase in depth to basement across it into the RT,
in excess of 4000 m to the northeast (Fig. 2.8). A consider-
able package of sediments, up to 10 km in total thickness, is
thus preserved within the RT. Karoo and Nama strata, or
Kalahari cover, are evident at surface. However, the deepest
section of Nama rocks yet drilled was in the Masethleng Pan
borehole in western Botswana, passing through the base of
the Nama at 3844 m. Underlying red beds were intersected
in both this and the Tses borehole in Namibia. Known red
beds, outcropping to the northwest and west of the Nama
Group, belong to the Doornpoort and Aubures Formations,
respectively, which have been variably correlated with either
the Tsumis or Sinclair Groups (Becker et al. 2005; Miller
2008a). The implication is thus that a red bed sequence of
Mesoproterozoic age, and possibly older strata, may form a
remarkably thick succession of up to approximately 6 km
beneath the Nama Group.
2.2.7.2 Interpreted Basement Geology, the Karas
Impact Structure
The only southern area proximal to the RT where basement
is exposed is in the 130 km-wide north-northeast-trending
Karas Horst (KH; Fig. 2.6), with which a dyke swarm is also
associated. The Karas Horst forms a beautiful mountain land
in southern Namibia, indicating relatively recent reactivation
along the north-northeast bounding faults. A focus of
attention arising from the deep magnetic and gravity sources
mentioned in Sect. 2.2.7.1 is a roughly circular set of mag-
netic anomalies in the southern portion of the RT proximal
to the Namaqua Lineament, approximately on strike with the
Karas Horst (Fig. 2.6; Corner 2008). Concentric gravity
highs collar the magnetic anomalies in the south and east but
centrally the circular magnetic anomalies and their core
correlate with a strong gravity low. This group of magnetic
and gravity anomalies is also the centre of focus of a number
of possible far-eld ring features which were mapped by
Corner (2008) from a number of the ltered magnetic data
sets (Karas Ring Structure (KRS); Fig. 2.7). Despite the
coarseness of gravity data, this combination of features,
together with numerous apparently associated radial faults,
has been interpreted to arise from a major meteorite impact
named the Karas Impact Structure (KIS; Fig. 2.6; Corner
2008). The depth solutions, discussed above, of *6kmin
this area suggest a pre-Aubures (Sinclair Group) age.
2 An Integrated Geophysical and Geological Interpretation 37
Furthermore, the deep magnetic anomalies appear to be
largely terminated against the Namaqua Lineament, possibly
also suggesting a Mesoproterozoic or earlier age.
2.2.8 Deep Neo- and Meso-Proterozoic Basins
in Namibia and Angola
The extent, and interpreted thickness of up to 10 km, of the
magnetically inert sedimentary succession which overlies the
basement of the Rehoboth Terrane is described in
Sect. 2.2.7.1 (Fig. 2.8). Note that the depth-to-basement
contours, which commence at 4000 m and extend to greater
than 6000 m, largely reect the thickness of a pre-Nama
Mesoproterozoic succession. Similar sedimentary succes-
sions have been interpreted in northern Namibia and
southern Angola from, rstly, a preliminary interpretation of
Angolan airborne gravity and magnetic data (Sonangol &
NAMCOR AAPG presentation 2009) and, secondly, a
hydrocarbon exploration interpretation of the Owambo
Basin in northern Namibia by Hoak et al. (2014). The
4000 m and 6000 m isopach contours are approximately
Fig. 2.8 Deep sedimentary basins in Namibia and Angola are
superimposed on the interpretation map of Fig. 2.6. The basins are
considered to comprise Meso- to Neo-Proterozoic sequences but are
overlaid here on younger sequences to illustrate their extent.
Depth-to-basement ranges of greater than 4000 and 6000 m are shown
38 B. Corner and R. J. Durrheim
merged in Fig. 2.8 to show the full extent of a combined
northern sedimentary succession. The former (Angolan)
interpretation stopped at the eastern border with Zambia. An
eastward thinning of the succession was nevertheless evi-
dent, thus the extent of the basin in Zambia, as shown in
Fig. 2.8, is schematic. Structural controls of the basin are
clearly evidenced from structures mapped in Namibia and
extended into Angola using the World Bouguer Gravity data
shown in Fig. 2.5. The basin is largely constrained by the
Opuwo and Kudu Lineaments (Op-L and Ku-L; Fig. 2.6)in
the northwest and southeast, respectively. The latter linea-
ment has a more complex, broader swathe as indicated in
Figs. 2.6 and 2.7. Whether these lineaments were active
during sedimentation, or subsequently controlled the inter-
preted distribution of the sedimentary succession through
faulting, is unknown.
Viewed in a regional context, there is a remarkable
similarity in the extent, thickness of sedimentary pile, and
location of the two basins with respect to the Kalahari
(Ka-L) and Tsodilo (Ts-L) Lineaments which bound them in
the east (Fig. 2.8). As noted in Sect. 2.2.6.3, it is possible
that the Ts-L was the northward continuation of the Kal-L in
pre-Pan-African times, and that the continuity between the
two was interrupted by the Pan-African Damaran Orogeny.
2.2.9 West Coast Offshore Domain
and the Namibian Passive Volcanic Margin
Corner and Swart (1997) and Corner et al. (2002) interpreted
the Namibian offshore airborne and ship-track magnetic
data, satellite Free Air gravity data, as well as numerous
seismic lines that traverse the continental shelf. The offshore
magnetic and gravity data sets are merged with the onshore
data in Figs. 2.2 and 2.4. Signicant contributions to the
interpretation of the offshore domain have been made by
Clemson et al. (1997) and Bauer et al. (2000). Corner and
Swart (1997) used more recent reection seismic surveys
covering the Hinge Zone (HZ; Fig. 2.6) to update the
mapping by Clemson et al. (1997) of the onset of the
basalt-related seaward-dipping reectors (SDR). Two
boundaries were mappedthat is, the HZ itself, and the
feather edge of the SDRs to the east of the Hinge Zone
(Fig. 2.6). The continental crust occurring offshore east of
the Hinge Zone was found not to have undergone any
observable extension during the break-up of Gondwana. The
extent of this block to the south of the Khorixas-Gaseneirob
Lineament is clearly evident in the offshore aeromagnetic
data. East of the HZ, the offshore extension of the Damara
Orogen was clearly evidenced by the magnetic data for the
rst time, allowing the offshore extensions of the NCZ, SCZ,
SZ and SMZ to be mapped (Fig. 2.6). A further signicant
result is that the mapped Mesozoic gravity and magnetic
anomalies have been clearly offset along the offshore con-
tinuations of a number of Pan-African lineamentsfor
example, the Autseib (Au-L), Omaruru (Om-L) and Wel-
witschia (We-L) Lineaments (Corner et al. 2002, Fig. 2.6).
The abrupt termination of Damaran rocks at the Au-L off-
shore is particularly dramatic. This observation suggests that
these Late Proterozoic to Early Palaeozoic structures not
only determined the architecture of the extended crust but
were likely reactivated during the Late Mesozoic, having
directions favourable for the initiation of transform faults.
An important implication arising from this interpretation is
that these structures may have provided potential pathways
for the larger drainage systems and hence the focus of major
offshore sedimentation, controlling the evolution of the off-
shore basins.
The M-type apparent spreading-related magnetic
anomalies of previous workers (e.g., Rabinowitz and
LeBrecque 1979; Gladczenko 1994) are readily evident in
the recent offshore datafor example, M2 and M4 in
Fig. 2.6. These are now more denitively delineated, and are
seen to converge towards the northern Namibian and
Angolan coastlines, in contrast to their previous placement
much further out to sea. This has major implications for
relative rates of extension from the southern to northern
offshore areas, and for the relative ages of the offshore
sedimentary sequences. The interpretation of Corner et al.
(2002) fundamentally claries, through forward modelling
of the magnetic and gravity data, the origin of the classical
M-type anomalies. These high-low pairs are shown to arise
from shallow westerly dipping wedges of lava (SDRs) at
their suboutcrop feather edge, as opposed to the classical
steep-dyke seaoor-spreading model. Such dykes are not
expected to occur senso stricto on the continental shelf or
within the extended continental crust, thus negating the M
nomenclature.
At least four coast-parallel gravity highs are seen in the
offshore data, the highest amplitude being closest to shore
(shallowest). These are also structurally disrupted along
offshore extensions of the major onshore Pan-African lin-
eaments, similarly signifying either later reactivation of these
lineaments or the control by these early lineaments of the
architecture of the extended crust. Corner et al. (2002)
showed, through modelling, that the eastern, and most
prominent, gravity high is caused primarily by the onset of a
major package of SDRs that dene the HZ. The amplitude of
this gravity high increases as the lava sequence starts to
thicken, but disappears completely further seaward as the
overlying low-density sedimentary pile progressively neu-
tralizes the effect of the deeper lava-related gravity high. The
gravity response to the resultant centres of mass may thus be
offset from the associated magnetic anomalies, which
delineate the shallower feather edge of the basalts. The
interpretation does not take cognizance of the
2 An Integrated Geophysical and Geological Interpretation 39
long-wavelength component, which is likely to be associated
with deep magmatic underplating, but does explain the
anomalies very well in the rst order. This interpretation is
in contrast with that of Watts and Fairhead (1999), who
applied a process-orientated approach to modelling edge
effectanomalies such as the HZ gravity high. The processes
which they considered included rifting, sedimentation and
magmatic underplating. By quantifying these, they were able
to model the edge effectgravity anomalies. However, they
did not consider the effect of the basaltic seaward-dipping
reectors and associated magnetic anomalies.
A number of probable offshore Cretaceous igneous com-
plexes were interpreted, the largest being the Walvis Bay
Complex (Wb; Fig. 2.6), situated close to the Om-L, and
hallmarked by signicant gravity and magnetic anomalies
(Corner and Swart 1997; Corner et al. 2002). The inland
Cretaceous Erongo Complex (Er) also occurs along the
Om-L. The offshore Cape Seal Complex (Cs) and the offshore
extension of the Cape Cross Complex (Cc) were also newly
identied. The offshore magnetic and gravity data adjacent to
South Africa have not been rigorously interpreted, but the
signicant anomalies and possible faults have been mapped
from the EMAG2 and World Gravity Map data, as shown in
Figs. 2.6 and 2.7. Noteworthy is the presence of the only
Cretaceous igneous complex associated with the South
African west coast regionthat is, the Koegelfontein Com-
plex (Kf; Fig. 2.6; Whitehead et al. 2016).
2.2.10 Namaqua-Natal Belt and Its Extension
as the Maud Belt in Antarctica
2.2.10.1 Namaqua-Natal Belt
The Namaqua-Natal Belt (Namaqua-Natal Metamorphic Belt
(NNB)) is an arcuate orogenic belt roughly 1400 km in length
which bounds the Kalahari Craton to the south (Figs. 2.2,2.6
and 2.7). It comprises igneous and metamorphic rocks formed
or metamorphosed during the Proterozoic, with four periods
of activity being evident at *1.82.00 Ga, *1.2
Ga, *1.15 Ga and *1.06 Ga, (Cornell et al. 2006; Egling-
ton 2006)that is, essentially two phases of activity of
Palaeo- and Meso-Proterozoic age. The extensive outcrops in
the west and east constitute the Namaqua and Natal Provinces,
respectively (NamqP and NatP; Fig. 2.6), the former
extending into Namibia as the Gordonia Subprovince (GSP;
Fig. 2.6). The region between the outcrop areas is covered by
Phanerozoic Karoo sediments, which hide the boundary
between the NNB and the younger Palaeozoic rocks of the
Cape Supergroup. The boundary is expected to be complex in
view of the late Permian to early Triassic Cape Orogeny. It is
thus only possible to map the southern boundary of the NNB
using geophysical techniques.
The NNB comprises a number of tectonostratigraphic ter-
ranes (i.e., areas of common lithostratigraphy and structural
fabric bounded by shear zones), which were assembled during
the Namaqua Orogeny (Cornell et al. 2006). However, the
structural fabric of the Namaqua and Natal Province differs
considerably (e.g., Cornell et al. 2006), as is also evidenced in
the magnetic images of Figs. 2.2 and 2.3. The Natal Province
may be subdivided into three distinct, structurally complex,
discontinuity-bound terranes: the northern Tugela, central
Mzumbe and southern Margate Terranes. These are bounded
by major shears and thrusts: the Tugela (thrust) Front,
Lilani-Matigulu Shear Zone, the Lovat Shear Zone and the
Melville Thrust (Thomas 1989;Jacobsetal.1993; Jacobs and
Thomas 2004; Cornell et al. 2006). Thrust tectonics are
extensively preserved within the Tugela Terrane, where rocks
were obductednorthwards onto the Kaapvaal Craton during the
Namaqua Orogeny, whereas the Mzumbe and Margate Ter-
ranes are dominated by major wrench faults (Matthews 1972;
Jacobs and Thomas 2004). The thrust and shear zones are
clearly evidenced by high-amplitude, linear magnetic anoma-
lies, as seen in Figs. 2.2,2.6 and 2.7 (Corner 1989;Corneretal.
1991;Hunteretal.1991;Thomasetal.1992). This structural
fabric, as seen in the aeromagnetic data, is in contrast to that of
the approximately coeval Namaqua Province, which has an
irregular, unstructured or curvilinear fabric. The magnetically
anomalous zone correlating with the mapped portion of the
Lovat Shear Zone (LSZ; Fig. 2.6) in the Natal Province can be
followed west-southwestwards in the aeromagnetic data, and
hasbeenmappedassuchinFigs. 2.6 and 2.7.
The continuity of the NNB from the Namaqua to the
Natal Provinces has commonly been accepted since the early
recognition thereof by Nicolaysen and Burger (1965). More
recently, Eglington and Armstrong (2003) conducted
geochronological and isotope studies on cores from four
deep boreholes drilled by SOEKOR (Ka, Kc, Qu and We;
Fig. 2.6), comparing these with known signatures from the
Namaqua and Natal Provinces. In particular, this study
showed that the four boreholes reect a Palaeoproterozoic
crustal history similar to that of the Namaqua Province, and
that reworked crust of this age is not evident in the Natal
Province nor in the Cape Meredith Complex of the Falkland
Islands. They thus concluded that there must be a signicant
terrane boundary between the exposures of the Natal Pro-
vince and the easternmost borehole studied (We; Fig. 2.6;
Eglington and Armstrong 2003). They further concluded that
this clear evidence for isotopically distinct terranes during
the Mesoproterozoic needs to be taken into account in
reconstructions of the NNB, and its easterly extensions
within Gondwana.
In the present study, a variety of regional-residual sepa-
ration and upward continuation lters were used to enhance
the sub-Karoo magnetic anomalies, including the Beattie
40 B. Corner and R. J. Durrheim
Magnetic Anomaly (Sect. 2.2.10.2), to assess the relation-
ship between the Namaqua and Natal Provinces (Figs. 2.2
and 2.6). It is clear that the central area within the NNB has a
distinctly different structural fabric to either the Namaqua or
Natal Provinces. This central area, here named the Khoisan
Province (KP; Fig. 2.6), is hallmarked by a disrupted but
nevertheless roughly continuous belt of magnetic anomalies,
of longer wavelength in the south as a result of the thicker
Karoo cover. The KP shows apparent closure, through
possible folding or thrusting in the west, where it adjoins the
Namaqua Province; whereas in the southeast it appears to be
discordantly truncated against the westward extension of the
Lovat Shear Zone. The distribution of residual gravity
anomalies in Figs. 2.4 and 2.6 also suggests apparent closure
in the west, despite the coarseness of the data. In the east, the
KP appears to terminate against a north-northeast-striking
fault just west of the SOEKOR borehole We (Fig. 2.6).
A further major north-northeast-striking fault occurs east of
borehole We, which appears to bound the Natal Province in
the west. This fault is here named the Elliot fault (EF;
Fig. 2.6; after the nearby town of Elliot). In part, the
northern sector of the Elliot fault coincides with the
escarpment, and it could be argued that it is simply the
manifestation of a dramatic increase in terrain clearance
during the airborne survey east of the fault. However, the
fault is also associated with a long-wavelength anomaly at its
northern end, and is interpreted to continue southwards away
from the escarpment. The intermediate area between these
two faults is hallmarked by a zone of high-frequency
anomalies arising from Karoo-age sills. The isotopic results
of borehole We discussed above (Eglington and Armstrong
2003) suggests a correlation of this area with the Namaqua
Province. No clear deeper sources, from either the KP or the
Natal Province, were evident in this intermediate area in any
of the lter products applied to the magnetic data. None of
the four SOEKOR boreholes was sited within the KP, and its
origin thus remains unknown. Given the location of the KP,
and its offset with respect to the Kaapvaal Craton, it could be
speculated that it may contain relicts of the southern portion
of a proto-Kaapvaal Craton, and associated Archaean basins,
assimilated during the Namaqua Orogeny. In the rst
instance, it is thus proposed here that, rstly, the Khoisan
Province constitutes a subterrane of the Namaqua Province
and, secondly, the Elliot fault is the western boundary of the
Natal Province.
2.2.10.2 Beattie Magnetic Anomaly and the
Southern Cape Conductivity Belt
Two remarkable, broadly spatially coincident, continental-
scale geophysical anomalies are evident within the NNB: the
Beattie Magnetic Anomaly (BMA; Beattie 1909) and the
Southern Cape Conductivity Belt (SCCB; Gough et al. 1973;
De Beer and Gough 1980; De Beer et al. 1982a; De Beer and
Meyer 1984). The BMA extends for more than 950 km, has
an apparent suboutcrop width in the order of 50 km, and
correlates spatially with the northern margin of the broader
SCCB, which extends from the western to the eastern coastal
margins for more than 1100 km (Figs. 2.6 and 2.12). Their
origin is enigmatic, with some earlier authors suggesting a
common source owing to their apparent correlation in
approximate location and scale, while more recent studies
indicate separate sources, but with a common evolutionary
origin. The SCCB is discussed in more detail in Sect. 2.3.2,
but reference is made to it here where relevant. Many geo-
physical studies have been directed at mapping and under-
standing the SCCB and the BMA, including interpretations
of aeromagnetic data, gravity data, magnetotelluric data,
reection seismic data and wide-angle seismic refraction
studies (e.g., Gough et al. 1973; De Beer and Gough 1980;
De Beer et al. 1982a; De Beer and Meyer 1984; De Beer and
Stettler 1988; Corner 1989; Pitts et al. 1992; Thomas et al.
1992; Lindeque et al. 2007,2011; Weckmann et al. 2007a,
2007b; Stankiewicz et al. 2008; Quesnel et al. 2009;
Scheiber-Enslin et al. 2014).
The BMA is not a single magnetic anomaly but rather
comprises a series of subparallel anomalies that may be
separated with spectral ltering into residual anomalies
reecting the shallower (in a relative sense, but nevertheless
deep) suboutcrop of magnetic sources beneath Phanerozoic
Karoo strata, and deeper sources within the crystalline crust,
which together give the appearance of a broad magnetically
anomalous belt. There is no gravity anomaly associated with
the BMA. Scheiber-Enslin et al. (2014) have mapped the
extent of the residual magnetic anomalies, extending into the
Natal Province. In parallel studies, Corner (e.g., 2015)
conducted more rigorous mapping of the residual anomalous
sources, an interpretation which is further rened and pre-
sented in Figs. 2.6 and 2.10. Forward models of magnetic
proles over the BMA, conducted by a number of
researchers (e.g., Maher and Pitts 1989; Du Plessis and
Thomas 1991; De Beer and Stettler, unpublished) show
similar resultsthat is, the BMA is readily modelled with
prism-shaped sources dipping at a shallow angle to the
south. These models are indicative of a possible
northward-directed thrust zone, as discussed further below.
An early model for the origin of the BMA was that of a
dipping slab of oceanic crust, representing a Pan-African
suture zone, with serpentinization being invoked to explain
the absence of an associated gravity high (De Beer et al.
1982a; De Beer and Meyer 1984).
Other forward models, conducted more recently, do not
necessarily agree with southward dips of the BMA but are
based on more simplistic source geometry models such as
horizontal prisms and spheres (Quesnel et al. 2009) or a very
coarse polygonal body interpreted to mimic a deeper-level
resistive crust (Weckmann et al. 2008). On the strength of
2 An Integrated Geophysical and Geological Interpretation 41
the model of Maher and Pitts (1989), Corner et al. (1991)
proposed that the linear magnetic anomalies in the Natal
Province associated with the known major craton-directed
thrust zones may share a common origin with the BMA.
Palaeomagnetic studies in northern Natal (Corner and
Maccelari unpublished) conrmed the presence of highly
magnetic mylonites associated with the Lilani-Matigulu
Thrust Zone (LMSZ; Fig. 2.6). Thomas et al. (1992) further
conrmed this, also indicating that the thrusts and shears of
the Natal Province are associated with linear magnetic
anomalies, giving a magnetically striped appearanceto the
province, and named the anomaly near Amanzimtoti the
Williston anomaly (ASZ/WA; Figs. 2.6 and 2.10). They also
supported the possibility that these may be associated with
the BMA, which is present south of the Williston anomaly.
Based on this model, Corner and Groenewald (1991), Corner
et al. (1991) and Hunter et al. (1991) proposed that the
regional-scale, high-amplitude magnetic anomalies in
Dronning Maud Land (DML), Antarctica, were the exten-
sion of the BMA within a Gondwana framework. Fieldwork
in DML, including palaeomagnetic sampling (Corner,
unpublished), showed strong similarities with northern
Natal: highly magnetic mylonites associated with
craton-directed thrusts were observed in the Kirwanveggen,
where the DML anomaly was observed in outcrop (see also
Sect. 2.2.10.3). Large euhedral magnetite grains were also
observed to be pervasive in the gneisses adjacent to the
thrusts. The similarity of signature between the magnetized
thrust zones of the Natal Province and the BMA was con-
rmed in a comprehensive study by Scheiber-Enslin et al.
(2014). The nding that high-amplitude magnetization is
pervasively associated with thrust zones and associated
gneisses in both the Natal Province and DML, as well as the
observation of the similar strike and linear magnetic signa-
ture of units within the BMA and in DML, is the best evi-
dence we have to date on the origin of the BMA. A reection
seismic section crossing the BMA (Figs. 2.11 and 2.13,
discussed in Sect. 2.3.3) shows a bean-shaped zone of strong
reections at the sub-Karoo location of the BMA. No
diagnostic dips or depth extent could, however, be
determined.
2.2.10.3 Extensions of the Namaqua-Natal Belt
Within Gondwana: The Maud Belt,
Antarctica
Early regional-scale aeromagnetic surveys were carried out
in Antarctica over Dronning Maud Land (DML) and the
Weddell Sea by the Polar Marine Geological Research
Expedition (PMGRE, VNIIOkeangeologia, USSR) during
the 1970s and 1980s, at a ight-line spacing of 20 km, with
a higher-resolution 5 km ight-line spacing survey being
own over the Kirwanveggen region (Golynsky et al.
2000a). Regional aeromagnetic data sets had also been
acquired by the German Federal Institute for Geosciences
and Natural Resources (BGR) and Alfred Wegener Institute
(AWI). A collaborative interpretation of these combined data
sets showed that the major magnetic anomalies in DML were
likely to be the continuation of counterparts in southern
Africa and the Falkland Plateau (Corner 1989;Fűtterer
1989). In addition, the geophysical components of the South
African National Antarctic Research Programme (SANARP)
included limited, but nevertheless diagnostic, regional sur-
face gravity proles traversing the Grunehogna and Maud
Provinces, limited helicopter-borne magnetic surveys, and
palaeomagnetic studies (Hodgkinson 1989; Hunter et al.
1991; Corner 1994b; Jones et al. 2003). More recently,
higher-resolution surveys, with a mostly 10 km ight-line
spacing, were undertaken in specic areas by the British
Antarctic Survey (BAS) in the Jutulstraumen region of
DML, by the BGR and AWI (Johnson et al. 1992; Riedel
et al. 2013). The area of early aeromagnetic coverage in
DML was partly reown, and extended, as part of the
AWI VISA project during the period 20012005, at
ight-line spacings of 10 km (mostly) and 20 km
(Riedel et al. 2013).
A zone of regional-scale linear or semi-linear en-echelon
magnetic lineaments (c. 40 km wide and extending over
700 km), revealed by the early and subsequent aeromagnetic
surveys, was termed the H.U. Sverdrupfjella-Kirwanveggen
Anomaly (SKA; Golynsky and Aleshkova 2000; Golynsky
et al. 2000a). Golynsky et al. (2000a,2000b) did not conduct
a detailed analysis of the anomalies owing to the regional
scale of their interpretation, but they attributed the anomalies
to high-grade meta-igneous gneisses of 10001200 Ma
(Grenvillian/Kibaran age), thrust or shear zones, and other
larger crustal structures and zones of weakness. The asso-
ciated magnetic anomalies were seen to be interrupted by
changes in strike and inferred faulting. Later deformation
and magmatism associated with the Ross orogeny
(Pan-African) is described as being considerably less
intense, with contrastingly weaker manifestation in the
magnetic data (Moyes et al. 1993; Golynsky et al. 2000b).
Importantly, Golynsky et al. (2000b) also recognized the
additional complexity of interpretation, in that many mag-
netic anomalies could arise from Mesozoic magmatism
associated with the break-up of Gondwana. The recent
interpretation of Riedel et al. (2013) is more rigourous, given
the larger area of coverage, the acquisition of data at a closer
ight-line spacing (other than over the Kirwanveggen), the
development and use of new data-ltering techniques, and
the benet of more eldwork having been conducted in the
interim.
The digital PMGRE aeromagnetic data set covering a
portion of DML was passed on to SANARP by PMGRE in
1989 as part of a collaborative research programme, and
interpreted by Corner (1989), Corner and Groenewald
42 B. Corner and R. J. Durrheim
(1991), Corner et al. (1991), Hunter et al. (1991) and Corner
(1994b). The aeromagnetic data is shown in Fig. 2.9 and the
interpretation is presented in Fig. 2.10, juxtaposed with
southern Africa within a Gondwana framework. That early
interpretation does not differ substantively from the more
recent interpretation of Riedel et al. (2013), but it benets in
detail locally from the higher-resolution 5 km PMGRE
ight-line coverage over the Kirwanveggen, where outcrop
occurs and ice cover is relatively thin in places. By com-
parison, the interpretation of Riedel et al. (2013) benets
from the much larger area of coverage of the VISA Project in
areas outside the higher-resolution PMGRE survey.
The continuation of the Namaqua-Natal Belt of southern
Africa into DML, as the Maud Belt (MB; Fig. 2.10) within
Gondwana, thus comprising the larger Namaqua-Natal-Maud
Belt (NNMB), has been established geologically for some time
(e.g., Groenewald et al. 1991;Hunteretal.1991;Jacobsetal.
1993;Riedeletal.2013). The interpretation of Corner (1989)
above was based, rstly, on the similarity of the Beattie and SKA
anomalies, in terms of scale and signature; and secondly, on the
fact that the former is clearly truncated by the Agulhas Fracture
Zone, implying possible eastward extension in Gondwana. The
Gondwana reconstruction (Martin and Hartnady 1986), initially
used by Corner (1989), aligned the SKA with the extension of the
Fig. 2.9 Total magnetic intensity image of southern Africa, the
Falkland Plateau (ship-track data) and Dronning Maud Land in
Antarctica. The closer Gondwana reconstructions, after Martin and
Hartnady (1986) and Grantham et al. (1988), have been relaxed here so
as to accommodate the interpreted Mesozoic basalt anomalies
(Fig. 2.10). The t is thus indicative, rather than rigorous, for the
purposes of this discussion
2 An Integrated Geophysical and Geological Interpretation 43
Beattie into DML. The intervening gap, known as the Natal
Embayment (Jacobs and Thomas 2004), was occupied by the
Maurice Ewing Bank (MEB) microplate of the Falkland Plateau,
over which marine magnetic data revealed a linear magnetic
anomaly which, if the MEB microplate is rotated slightly as
suggested in Figs. 2.9 and 2.10, would also constitute a possible
continuation of the Beattie anomaly, and a link between the
BMAandSKA(Figs.2.9 and 2.10). In the interpretation of
Corner (in Hunter et al. 1991), rened here, magnetic units and
faulting within the larger SKA zone were mapped using
higher-resolution residual-ltered aeromagnetic data. One aspect
that complicated this interpretation was the presence of numer-
ous magnetic anomalies owing to the Mesozoic basalts associ-
ated with the break-up of Gondwana, progressively dominating
westwards from the SKA towards the Explora Escarpment
(EE; Fig. 2.10). However, these appeared to have a strike
direction different from that of the SKA anomalous units, and
were thus subjectively separated, as shown in Fig. 2.10.This
interpretation may be equivocal in the zone where both of these
anomalous sources are present. The closer t of DML to southern
Fig. 2.10 Interpreted extension of the Namaqua-Natal Belt and
Kaapvaal Craton in Dronning Maud Land, Antarctica, based on the
correlation of the BMA (Beattie Magnetic Anomaly) and SKA (HU
Sverdrupfjella-Kirwanveggen Anomaly) magnetic anomalies. As with
Fig. 2.9, the closer Gondwana reconstructions, after Martin and
Hartnady (1986) and Grantham et al. (1988), have been relaxed here
so as to accommodate the interpreted Mesozoic basalt anomalies.
AAnnandagstoppane; AR Ahlmanryggen; BBorgmassivet; DML
Dronning Maud Land; EE Explora Escarpment; GC Grunehogna
Craton; HF Heimefrontfjella; KKirwanveggen; MB Maud Belt; MEB
Maurice Ewing Bank microplate; SKA Sverdrupfjella-Kirwanveggen
Anomaly; SV H.U. Sverdrupfjella. All other text abbreviations are
given in Fig. 2.6
44 B. Corner and R. J. Durrheim
Africa of Grantham et al. (1988) has been relaxed in Fig. 2.10 so
as to accommodate the interpreted Mesozoic basalt anomalies. It
is thus an indicative, rather than rigorous, t for the purposes of
this discussion. The continuation of the Namaqua-Natal Belt and
BMA into DML in Antarctica has been further supported by the
more recent work of Golynsky and Jacobs (2001); Jacobs and
Thomas (2004) and Riedel et al. (2013).
The postulated preservation of an Archaean cratonic
microplate in DML (the Grunehogna Craton), Antarctica, is
supported by Archaean ages of c. 3000 Ma (Halpern 1970),
and 3067 Ma (Marschall et al. 2010), determined on isolated
exposures of basement at the Juletoppane and
Annandagstoppane nunataks, respectively (red star;
Fig. 2.10; see also Barton et al. 1987; Hunter et al. 1991).
Corner (1994b), in interpreting the DML aeromagnetic data
and ground gravity proles, outlined the extent in DML of
this microplate (Figs. 2.1 and 2.10) and juxtaposed it with
the eastern margin of the Kaapvaal craton within a Gond-
wana reconstruction, suggesting the possibility that the cra-
tonic microplate constituted the eastward extension of an
early Kaapvaal craton. However, the location of the Kaap-
vaal Craton within Rodinia, and subsequently Gondwana,
has been the topic of much controversy. An early view was
that the DML Archaean microplate (Grunehogna Craton) is
not related to the Kaapvaal-Zimbabwe Province (Barton
et al. 1987; Barton and Moyes 1990). Dalziel (1991,2000)
and Moores (1991) hypothesized that Laurentia and East
Antarctica were juxtaposed in an early Neoproterozoic
supercontinent. This was named the SWEAT hypothesis
(Southwest United StatesEast Antarctica). Moyes et al.
(1993) examined the hypothesis by comparing coeval
magmatism, regional isotopic resetting and structural
deformation styles, and they concluded that this data neither
supports nor contradicts the SWEAT hypothesis. Storey
et al. (1994) compared geochronological, isotopic and
aeromagnetic data between Coats Land and DML, con-
rming rocks of Grenvillian age in both, and concluded that
this data supports the SWEAT hypothesis. On the other
hand, Golynsky et al. (2000b) investigated the tectonic
development of Coats Land and western DML using aero-
magnetic data and correlative outcrop data, and concluded
that Coats Land was never part of the Kaapvaal-Zimbabwe
craton. Overall support for or against the SWEAT hypoth-
esis is thus equivocal. It has also been suggested that Wes-
tern Australia was a collision partner for the Kalahari Craton.
This hypothesis was investigated by Ksienzyk and Jacobs
(2015), who could not nd support from a geochronological
point of view. In summary, much recent work supports the
extension of the Kaapvaal-Zimbabwe Cratons into DML,
and the existence of a continuous Grenvillian/Kibaran belt
bounding this proto-Archaean craton to the south and east.
2.3 Electrical Resistivity, Magnetotelluric
and Regional Seismic Investigations
2.3.1 Introduction
Over the past half-century, a series of passive and active
seismic experiments, and resistivity, geomagnetic induction
and magnetotelluric (MT) surveys, have been conducted in
southern Africa. Most of the early work was stimulated by
curiosity, with the objective of learning more about the
structure and evolution of the African continent. Much of the
later work was done with commercial intent, with the aim of
discovering new gold, platinum, diamond and hydrocarbon
resources. In particular, reection seismology surveys, while
much more expensive to carry out than gravity and magnetic
surveys, are able to produce far better images of the sub-
surface. The locations of MT and broadband seismic sta-
tions, Schlumberger resistivity soundings, seismic refraction,
seismic reection and MT proles are superimposed on the
geological interpretation map in Fig. 2.11.
In Sect. 2.2, magnetic and gravity data was used to map
features of the southern African crust (e.g., Fig. 2.6). These
regional surveys were mostly conducted by geological sur-
veys, geoscience councils and other government agencies
with the aim of stimulating exploration for metals and
minerals. Exploration and mining companies use these sur-
veys to identify target areas, secure prospecting licences,
conduct higher-resolution geophysical and geochemical
surveys, and, depending on the outcome, drill boreholes. In
this section we use the electrical resistivity, MT and seismic
data and their interpretations to validate and extend the
interpretation of the potential eld data. We recognize that
the data sets have not been fully integrated. The prime rea-
son for this is that the various geophysical studies addressed
different key questions. Consequently, the survey footprints
and depths of investigation (ranging from the near surface to
the mantle) do not always overlap. Nevertheless, we believe
that there is merit in providing the reader with a brief but
comprehensive review of all the major geophysical surveys
that have been conducted in southern Africa.
2.3.2 Electrical Resistivity and Magnetotelluric
Studies
2.3.2.1 Deep Electrical Resistivity
and Geomagnetic Induction Soundings
Between 1967 and 1986 the Council for Scientic and
Industrial Research in South Africa (CSIR) conducted 11
ultradeep electrical resistivity soundings using the Schlum-
berger array with electrode spacings of up to 1200 km,
2 An Integrated Geophysical and Geological Interpretation 45
probing the crust and upper mantle in the Kaapvaal, Zim-
babwe and Congo cratons, the Bushveld Complex, and the
Limpopo, Namaqua-Natal, Gariep and Damara Belts (De
Beer 2015b; Fig. 2.11). In general, ignoring the surcial
weathered layer, a ve-layer electrical structure was found in
both cratons and mobile belts:
Fig. 2.11 Southern African seismic and magnetotelluric surveys.
Blue, recent Magnetotelluric (MT) traverses conducted by the
SAMTEX and Inkaba ye Africa programmes; green circles, MT
soundings; green diamonds, vertical electrical soundings; red dia-
monds, SASE Broadband seismometer stations; red lines, national
reection seismic proles; orange lines, AAC reection seismic
proles; brown lines, refraction seismic proles; red boxes, geotran-
sects. KTA Kimberley Telemetered Array. The interpretation backdrop
is from Fig. 2.6
46 B. Corner and R. J. Durrheim
A high-resistivity layer (q> 30,000 Xm) extending to
about 10 km was found in massive terrains’—that is, in
the granitic, Archaean cratonic nuclei.
A moderate-resistivity zone (2000 Xm<q< 20,000
m), indicating the presence of water-lled fractures
associated with deformed metamorphic rocks in mobile
belts, was observed at a depth range of 030 km and in
the middle crust in Archaean cratons.
A highly conductive zone (q< 100 Xm), observed in the
depth range 2530 km, was speculated to consist of
serpentinized ultramac mantle rock.
The uppermost mantle was found to be highly resistive
(q> 20,000 Xm).
Thereafter the resistivity was found to decrease gradually
as mantle temperatures increase with depth.
The CSIR supplemented the resistivity surveys with a
series of geomagnetic induction campaigns (e.g., De Beer
2015c). Some 26 three-component magnetometers were
deployed in 1971 in a triangular array over central South
Africa, which straddled the boundary between the Kaapvaal
Craton and the Namaqua-Natal Belt (Gough et al. 1973;
Fig. 2.11). No signicant resistivity difference was found
between these domains, but a signicant induction anomaly
was evident under the southern edge of the array. In 1977 an
array of 52 magnetometers was used to map this electrical
conductivity anomaly, named the Southern Cape Conductive
Belt (SCCB; Fig. 2.12). This coincides in part with the
Beattie Magnetic Anomaly (De Beer and Gough 1980; see
also Sect. 2.2.10.2 for further discussion).
A magnetometer array was deployed in northeastern
Namibia, northern Botswana and northeastern Zimbabwe in
1971/1972 (De Beer et al. 1976; Fig. 2.11). A zone of low
resistivity was discovered that runs from the Zambezi Valley
to south of the Okavango Delta, and into the Damara Oro-
genic Belt (Fig. 2.12). Further surveys were conducted in the
Damara Belt in 1977 (De Beer et al. 1982b), which tracked
the Damara conductor to the Atlantic coast. The geomag-
netic surveys were complemented by more than 40 Sch-
lumberger soundings with maximum electrode spacings of
40 km, which showed that the northeast-striking conductor
has steep sides, is 310 km deep and at least 20 km thick,
and has a resistivity of less than 20 Xm (Van Zijl and De
Beer 1983; Fig. 2.12).
Magnetotelluric (MT) surveys conducted in northern
Zimbabwe (Losecke et al. 1988, cited by Weckman 2012)
detected highly conductive layers in the Lower Zambezi
Valley situated in the Zambezi Mobile Belt (lying between
the Congo and Kalahari cratons), which were tentatively
linked with the conductive structures of the Damara Belt. In
the late 1990s a series of MT surveys were carried out in
shallow Karoo-age basins in the Zambezi Valley (Whaler
and Zengeni 1993; Bailey et al. 2000a,b). The underlying
cratonic rocks had quite high resistivities, which Bailey et al.
(2000a,b) interpreted to indicate that cratons were not
affected by the basin-forming processes in the adjacent
mobile belts.
2.3.2.2 SAMTEX and Inkaba Ye Africa
Magnetotelluric Experiments
Between 2003 and 2008 the Southern African Magnetotel-
luric Experiment (SAMTEX) deployed more than 740 sta-
tions at a nominal spacing of 20 km on lines, with a total
length of some 15,000 km, straddling the major tectonic
provinces (Evans et al. 2011; Fig. 2.11). SAMTEX was led
by Dr A Jones of the Dublin Institute for Advanced Studies
and involved many African and international scientists and
institutions. MT arrays were also deployed under the aus-
pices of the Inkaba ye Africa programme (Weckmann 2012).
The principal ndings of these investigations are reviewed
below.
Kaapvaal Craton and Rehoboth Terrane (see also
Sect. 2.2.7): The MT observations were integrated with
various geophysical and petrological observables (viz. ele-
vation, surface heat ow, xenoliths) to derive an electrical
conductivity model (Fullea et al. 2011). The depth of the
present-day thermal lithosphereasthenosphere boundary
(LAB) was estimated to be at depths of 230260 km and
150 km for the western block of the Kaapvaal Craton and
Rehoboth Terrane, respectively. (It is important to note that
the thermal LAB may differ from the chemical and
mechanical LABs.)
Kaapvaal CratonLimpopo TerraneZimbabwe Cra-
ton (see also Sect. 2.2.3.3): Three proles were used to
investigate the electrical structure of this region (Khoza et al.
2013b; Fig. 2.11). The 30 km-wide
Sunnyside-Palala-Tshipise shear zone (PaSZ; Fig. 2.6) was
found to be a subvertical conductive feature that was inter-
preted to be the collisional suture between the Kaapvaal and
Zimbabwe cratons.
Congo CratonDamara Belt (see also Sects. 2.2.3.4
and 2.2.6.3): The boundary between the Congo Craton and
the Damara Belt is largely concealed by younger sediments.
Four semiparallel MT proles were used to investigate the
electrical structure (Khoza et al. 2013a; Fig. 2.11). The
Damara Belt lithosphere was found to be considerably
thinner and more conductive than the Congo Craton litho-
sphere. Resistive features in the upper crust are interpreted as
igneous intrusions emplaced during the Pan-African oro-
genic event, while highly conductive zones within the
Central Zone of the Damara Belt are believed to be related to
graphite- and possible sulphide-bearing stratigraphic units.
The boundary between the Congo Craton and Damara Belt
was shifted southwards, compared to prior models. A local
2 An Integrated Geophysical and Geological Interpretation 47
study across the Waterberg Thrust/Omaruru Lineament
(WT/Om-L; Fig. 2.6) found a 10 km-wide and at least
14 km-deep zone of anisotropic conductivity in the shallow
crust parallel to the WT/Om-L (Ritter et al. 2003; Weck-
mann et al. 2003) and was interpreted to be the exhumed
deep roots of ancient active shear zones.
Zimbabwe CratonMagondi BeltGhanzi-Chobe Belt
(see also Sect. 2.2.3.3): Kalahari sands cover most of
northeastern Botswana and little was known about
lithospheric structure and thickness prior to the study by
Miensopust et al. (2011). A 600 km-long prole (Fig. 2.11)
was interpreted. The Zimbabwe Craton is characterized by
thick (*220 km) resistive lithosphere; the
Tsumis-Ghanzi-Chobe Belt (TGCB; Fig. 2.6) by a some-
what thinner (*180 km) resistive lithosphere; while two
lower- to mid-crustal conductors were found in the inter-
vening Magondi Belt (MB; Fig. 2.6). The terrane boundary
between the Magondi and Ghanzi-Chobe Belts was
Fig. 2.12 Regional-scale conductivity anomalies overlaid on the
interpretation backdrop of Fig. 2.6. The blue-speckled polygons are
the conductivity anomalies mapped from the Magnetotelluric surveys
described in Sects. 2.2.10.2 and 2.3.2 (SCCB Southern Cape Conduc-
tivity Belt; DCCB Damara-Chobe Conductivity Belt). The local
anomalies shown are mapped from individual traverse data, with
inferred extensions shown as dashed lines. The bold solid-blue lines
show the DCCB of van Zijl et al. (1983) and the dashed lines show the
inferred extensions into Botswana
48 B. Corner and R. J. Durrheim
interpreted to be further north than previously inferred from
regional potential eld data.
Kaapvaal CratonNamaqua Natal Belt (see also
Sect. 2.2.10.1 ): Karoo Supergroup strata conceal the con-
tact between the Kaapvaal Craton and the Namaqua Natal
Belt, but a large contrast in resistivity (>5000 Xm and *30
Xm, respectively) makes it possible to map the contact
(Weckmann 2012).
Southern Cape Conductivity Belt (see also
Sect. 2.2.10.2): The source of the SCCB, discovered by De
Beer et al. (1982a; Fig. 2.12), was studied in detail at
speci