The Geology of the Belingwe Greenstone Belt, Zimbabwe

A study of Archaean continental crust

M.J. Bickle, A. Martin, E.G. Nisbet, J.L. Orpen

• Publisher: CRC Press
• Year: 2020
• Print ISBN: 9789054101208
• eBook ISBN: 9781000142686
• Edition: 1
• Topic: Biological Sciences
• Subtopic: Geology & Earth Sciences
• Index: Biological Sciences

A study on the Bellingwe Greenstone Belt, presenting the stratigraphy of the belt, its structure and tectonic setting, the sedimentology of what may be a rift basin, the geochemistry of the freshest Archaean komatiites yet found, and models of the evolution of the region.

CHAPTER 1

Introduction: The Zimbabwe Craton and controversies over Archaean granite-greenstone terrains

M.J. BICKLE, E.G. NISBET, A. MARTIN & J.L. ORPEN

ABSTRACT

The Belingwe Greenstone Belt contains a wide variety of Archaean rocks which provide evidence of major significance in interpreting the strati graphic history of the Zimbabwe Craton, and, more generally, in understanding the nature of the Archaean Earth. The stratigraphic sequences in the belt are well preserved, and the interpretation of the Belingwe rocks provides the underpinning of the present stratigraphic evolution of the Zimbabwe Craton. The sequence includes old basement, consisting of granitoid gneisses ca. 3.5 Ga old, overlain by three older greenstone sequences and a well developed younger (2.7 Ga) greenstone sequence. At the base of the 2.7 Ga sequence is a well exposed unconformity, demonstrating that the lavas and sediments of the younger greenstones were laid down on a mature terrain of eroded granitoid and greenstones.

The belt contains a wide variety of unusual rocks, including an extensive and locally very fresh suite of komatiites and komatiitic basalts, as well as carbonates which contain stromatolites. These rocks demonstrate that some greenstone belts formed on continental crust. The variety of sedimentary and volcanic environments probably reflects interaction of a range of tectonic processes as diverse as those operating today.

1.1 PURPOSE AND HISTORY OF PRESENT STUDY

The Belingwe Belt contains one of the most complete and best preserved greenstones sequences yet described. This study is a report of the results of several years (1973–1977, 1981–1983, 1986) of field investigation of the Archaean rocks within and around the Belingwe Greenstone Belt in Zimbabwe. The work was initiated as a field mapping program both to resolve basic geological relationships in the area and to search for insight into the more general problems of understanding the Archaean Earth. The mapping has been published by the Geological Survey of Zimbabwe (Martin 1978;Orpenetal. 1986). The stratigraphic relationships inferred in the area, which include a well-preserved basal unconformity to part of the supracrustal greenstone sequence, have led to a major revision of Archaean stratigraphy in Zimbabwe. The well preserved volcanic, sedimentary and fossiliferous rocks have provided much important petrological, geochemical and isotopic information. Here we report the field, petrological and geochemical data on which interpretations of stratigraphy and field relations are based. The data are integrated into a tectonic model for the development of this block of early continental crust.

The work described here has its roots in the revolution in the Earth Sciences which took place in the late 1960’s. Plate tectonics provided a coherent explanation of both the processes and fundamental causes of tectonic activity on the modern earth. Its success as an explanation of the workings of the modern Earth promoted enquiry into the evolution of tectonic processes over the earlier part of the Earth’s history. Could detailed study of the earliest part of the geological record, the Archaean, provide a similar understanding of the young Earth? Two other factors added to the especial interest in Archaean geology at this time: 1) Improvements in radiometric methods of geochronology during the 1960’s demonstrated the antiquity of Archaean rocks and provided a relative chronology for the first time; 2) magnesium-rich lavas or ‘komatiites’ were discovered in the Barberton Mountain Land by Viljoen & Viljoen (1969) and in Western Australia during the Nickel boom (e.g. Nesbitt 1971). The realisation that these ultramafic lavas must have erupted at high temperatures suggested that the thermal structure of the earth had changed substantially during earth history. This was the basis for numerous tectonic interpretations of Archaean geology. However the lack of consensus on these tectonic interpretations reflected fundamental problems which arose in part from uncertainty over the basic geological relationships within Archaean terrains. This study was initiated in order to gather much needed basic field information which could then serve as a foundation for interpretation.

Plate tectonic theory had illustrated the significant role of oceanic crust in modern tectonic processes and by the early 1970’s it was realised how difficult it could be to distinguish allochthonous from autochthonous supracrustal sequences. A fundamental problem of Archaean geology was whether the greenstone sequences dominated by mafic volcanics were formed in oceanic environments and tectonically emplaced to their present crustal settings, or whether they had been laid down directly onto continental crust. Many authors in the early 1970’s believed greenstone belts to be analogous to modern ophiolites: Many still do. The report that in the Belingwe Belt a basal unconformity existed between a greenstone belt succession and a granitoid-gneiss crust provided an unusual chance to evaluate the stratigraphic setting of some greenstone belt rocks. In addition to this basic aim, a study of well-preserved Archaean supracrustal rocks was likely to provide important information on Archaean igneous rocks, Archaean sedimentary environments and the geological setting of early life.

This study began in the early 1970’s. In 1970 and 1971 M.J. Bickle and E.G. Nisbet became interested in the contrasts and similarities between young ophiolite complexes and Archaean ultramafic bodies. Separately, A. Martin mapped the northern half of the Belingwe Greenstone Belt for the Zimbabwe (formerly Rhodesian) Geological Survey during 1973, 1974 and 1975 and concurrently began doctoral research under the supervision of J.F. Wilson. In 1974 M.J. Bickle, on a postdoctoral research fellowship at the University of Zimbabwe (formerly University of Rhodesia) commenced work in the south-eastern part of the greenstone belt, and was joined by E.G. Nisbet on a NERC postdoctoral fellowship held at Oxford. This field work continued through 1975 with a short field seasons in 1976(M.J.B.), 1982(E.G.N.), 1983 and 1984 (A.M.,J.L.O.). and further mapping and drilling in 1986 (E.G.N., MJ.B., A.M.). J.L. Open studied the south-west part of the greenstone belt in an undergraduate project in 1974 and undertook doctoral work in the same area from 1975 to 1977 supervised by J.F. Wilson. The field studies were curtailed by escalation of hostilities in Zimbabwe in 1976. For this reason the study is not as complete in some areas as the authors originally intended. The field studies were complemented by a number of laboratory studies. Major and trace element concentrations of volcanic rocks were analysed in Oxford, Zurich and Leeds and more recently in Mainz and Saskatoon. The Rb-Sr isotopic studies on volcanic and granitoid rocks from the area were undertaken in Oxford and Leeds in conjunction with CJ. Hawkesworth, S. Moorbath, P. N. Taylor, P.J. Hamilton and R.K.O’Nions. Experimental studies on komatiites were undertaken initially at Edinburgh in conjunction with C.Ford and later at Lamont with D. Walker. Stable and radiogenic isotope and rare gas studies have been carried out at Saskatoon (by T.K. Kyser and E. Hegner) and at the University of Rhode Island (with P. Abell). Many of the laboratory studies on rocks from the Belingwe Greenstone Belt are still continuing especially in Rennes, Cambridge and Saskatoon and samples have been provided to a wide variety of other laboratories. A bibliography of work on Belingwe is listed at the end of the references. It should be noted that much of the initial inspiration and encouragement to work in the area for all the authors came from J.F. Wilson of the University of Zimbabwe who took a continuing interest in the field studies and their regional significance to the Zimbabwe Archaean. We hope that this work will be seen as initially envisaged: As a continuation of the tradition of field investigation established in Zimbabwe by EP. Mennell and A.M. Macgregor.

1.2 ARCHAEAN GEOLOGY — SOME OF THE PROBLEMS

The Archaean record extends over one third of earth history from ca. 4.3 or 4.2 Ga ago until approximately 2.5 Ga, when the Great Dyke was emplaced in Zimbabwe. The Archaean thus provides most of the available data about the critical stages in the early evolution of the earth. Our ability to interpret this geological record is hindered by some very special problems. Tectonic and surface geological processes, may have differed substantially from modern processes, making uniformitarianism suspect Preservation of terrains is likely to have been selective and the preserved sample may be nonrepresentative. The size of most Archaean terrains is relatively small. Some are comparable to modern orogenic provinces, others are mere fragments: This makes resolution of tectonic provinces difficult. Fossils are absent or not stratigraphically useful, and chronostratigraphy is possible only where U-Pb zircon geochronology is available. In most Archaean terrains it is difficult to resolve even relatively major geological events. For these reasons it is perhaps not surprising that there is little consensus in current interpretations of Archaean geology and the interpretations are invariably simplistic. An additional problem is that many Archaean terrains are in relatively remote or inaccessible and less populated parts of the world and have not been mapped in the detail to which more recent orogenic belts are known.

Archaean terrains are conveniently grouped into the high-grade gneiss terrains, granite-greenstone terrains and late Archaean sediment-dominated cratonic basins. The relationships of these three environments are not properly known. It should be noted that cratonic sequences such as the 3000 Ma Pongola and 2700 Ma (?) Witwatersrand successions in S. Africa and 2700 Ma Fortescue Group in the Pilbara, W. Australia were being deposited on relatively rigid cratons over the same period as much of the Belingwe Greenstone Belt was laid down. Similarly, the Limpopo Belt, a diverse orogenic belt including high grade gneiss, which lies only 50 km south-east of Belingwe, comprises rocks ranging from early to late Archaean in age which underwent major tectono-metamorphic events at 3300 and 2900 Ma (Hickman & Wakefield 1975). These events to some extent were synchronous with the development of the granite greenstone terrain of the Zimbabwe Craton.

The granite-greenstone terrains are the most distinctive type of Archaean outcrop and the mafic-volcanic dominated supracrustal greenstone sequence disposed around ovoid or more linear granitoid and gneiss batholiths exhibit a tectonic style not often seen in younger provinces. Since Macgregor (1951) drew attention to the distinctive pattern of ‘gregarious’ batholiths there has been much speculation on the origin of granite-greenstone terrains. This speculation has drawn attention to a number of unresolved geological relationships within the terrains as discussed by Macgregor. First, there is the problem of the nature of the contact between the basement and the supracrustal greenstone sequences. Within Archaean terrains most granite-greenstone contacts are either intrusive or tectonic. Macgregor described important within-greenstone unconformities in Zimbabwe and speculated (1947) that the Belingwe Greenstone Belt was unconformable on a granite-gneiss basement. This unconformity was later proved by Laubscher (1963) but not described in detail until the present work (Bickle et al. 1975). From an understanding of the regional field geology Macgregor inferred an older granitoid gneiss and greenstone basement to the main greenstone sequences.

Outside Zimbabwe, the lack of any clear unconformities in many granite-greenstone areas and the advent of plate-tectonic interpretations — with the recognition of the significance of allochthonous oceanic rocks in orogenic terrains (ophiolites) — led to alternative interpretations of greenstone sequences as disrupted allochthonous oceanic crust (e.g. Anhaeusser 1973; Burke et al. 1976; Goodwin & Ridler 1970). Distinction between the ‘autochthonous continental ‘ or the ‘ allochthonous oceanic ‘ models is a crucial initial step in tectonic interpretation of granite-greenstone terrains. Are greenstone sequences typically disordered, allochthonous and bounded by thrusts or are they stratigraphically intact autochthonous sequences? This problem (which may have different answers in different places) can only be resolved by direct observation of the evidence for tectonic, or stratigraphic and unconformable relations, observation of local sedimentological environments or larger scale stratigraphic correlation and sedimentological facies variation. This was an important aim of our study. In addition, there are the related questions about the original extent of greenstone sequences and the possibility of regional stratigraphic correlation. The presently preserved greenstone belts may be remnants from a once continuous cover or may have been individual basins, filled with sediment and volcanic rocks, with dimensions similar to the present size of the belts.

It is perhaps a measure of the primitive state of Archaean geological interpretations that few if any general subdivisions of the tectonic environments in which greenstone belts formed are widely recognised or accepted. Greenstone belts contain most of the rock types found in Phanerozoic successions and exhibit significant variations in rock content, tectonic style and metamorphic grade. It might be expected that these supracrustal sequences were formed in as diverse a range of tectonic settings as is seen in Phanerozoic sediment-volcanic associations. The few attempts to establish either chronological variations in greenstone belt style (e.g. Glikson 1979) or spatial variations in greenstone beltenvironments(e.g. Groves& Batt 1984) remain unconvincing because the environmental interpretations on which they are based are poorly documented.

The sedimentary and volcanic rocks of the greenstone sequences also contain information on two other aspects of Earth evolution. Volcanic rocks provide a ‘window’ to mantle processes and mantle composition. Komatiites, which are magnesium-rich lavas erupted at high temperature, are mostly restricted to the Archaean. They form a minor but distinctive part of many greenstone sequences and have attracted much attention because study of these rocks has produced geochemical and thermal constraints on the Archaean m...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Authors
• 1 Introduction: The Zimbabwe Craton and controversies over Archaean granite-greenstone terrains
• 2 Rock units and stratigraphy of the Belingwe Greenstone Belt: The complexity of the tectonic setting
• 3 Structure and metamorphism of the Belingwe Greenstone Belt and adjacent granite-gneiss terrain: The tectonic evolution of an Archaean craton
• 4 The Mtshingwe Group in the west: Andesites, basalts, komatiites and sediments of the Hokonui, Bend and Koodoovale Formations
• 5 Sedimentology of the Brooklands Formation, Zimbabwe: Development of an Archaean greenstone belt in a rifted graben
• 6 The Ngezi Group: Komatiites, basalts and stromatolites on continental crust
• 7 Pb and Nd isotopic correlation in Belingwe komatiites and basalts
• 8 Geochemistry of the igneous rocks of the Belingwe Greenstone Belt: Alteration, contamination and pedogenesis
• 9 Controls on the formation of the Belingwe Greenstone Belt, Zimbabwe
• References
• Bibliography of the Archaean crustal study project
• Subject index