Journal of Petrology Volume 42 Number 11 Pages 2049-2081 2001
© Oxford University Press 2001
Early Cretaceous Basalt and Picrite Dykes of the Southern Etendeka Region, NW Namibia: Windows into the Role of the Tristan Mantle Plume in Paraná–Etendeka Magmatism
1DEPARTMENT OF GEOLOGICAL SCIENCES, UNIVERSITY OF DURHAM, SOUTH ROAD, DURHAM DH1 3LE, UK
2DEPARTMENT OF EARTH SCIENCES, UNIVERSITY OF CAMBRIDGE, DOWNING STREET, CAMBRIDGE CB2 3EQ, UK
3DEPARTMENT OF GEOLOGY, McMASTER UNIVERSITY, 1280 MAIN STREET WEST, HAMILTON, ONT., CANADA L8S 4M1
Received March 8, 2000; Revised typescript accepted April 5, 2001
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONFIELD RELATIONS AND SAMPLING...ANALYTICAL TECHNIQUESDYKE MAGMA TYPESHORINGBAAI MAGMA TYPEPETROGRAPHY OF OTHER DYKE...POST-GENESIS, PRE-EMPLACEMENT...HOW DID THE CRUSTAL...MAGMA GENESISROLE OF TRISTAN MANTLE...REFERENCES |
|---|
Abundant dykes in the southern Etendeka region, NW Namibia, mostly contain 8–20% MgO. Almost all can be allocated to previously described Early Cretaceous magma types. Horingbaai-type basalts–picrites occur up to 120 km inland. Some have superficially mid-ocean ridge basalt (MORB)-like compositions: (La/Nb)n
1·0; (Sm/Lu)n 1·5; 
Nd >8; initial 87Sr/86Sr 0·7032. Others show major- and trace-element, and Sr–Nd–Pb isotopic evidence of contamination during upwelling by up to a few per cent of K-feldspar-rich upper crust. Extremely magnesian olivine macrocrysts (Fo91–93·3) in some Horingbaai picrites indicate that komatiitic (MgO 24%) liquids were associated with this suite, although they were too dense to reach the crustal levels currently exposed. Ferropicrite dykes resemble closely the nearby Etendeka ferropicrite basal lavas, with total iron (as Fe2O3) >MgO; (La/Nb)n 1·0–1·5; (Sm/Lu)n 2·5–7; Nd 2–3, except in samples with geochemical evidence of contamination by Archaean–Proterozoic lower crust. A few low-Ti dykes resemble geochemically the lavas of the main southern Etendeka succession, with (La/Nb)n 2·0–2·5; Nd = 0 to -8; initial 87Sr/86Sr >0·707. Dykes north of the Huab river define a fourth magma type, Nil Desperandum (ND), that may have fed part of the Huab sill complex. The dyke ages are constrained by their field relationships. Ferropicrite and low-Ti dykes are consanguineous with lavas erupted at 133 and 132 Ma, respectively. Both Horingbaai and ND dykes cut 132 Ma Etendeka lavas, and the main swarm of Horingbaai picrites with forsteritic macrocrysts is cut by the 131 Ma Brandberg plutonic complex. Forward and inverse modelling of the genesis of the ferropicrites and the Horingbaai picrite–komatiites gives two temporal ‘windows’ into physicochemical conditions within the head of the impacting Tristan mantle plume. At 133 Ma the southern Etendeka lithosphere was >100 km thick and only incipient melting of predominantly Fe-rich peridotite streaks occurred in the rising plume head (Tp = 1500°C). By 2 my later, pre-Atlantic extension had reduced the lithosphere thickness beneath and adjacent to NW Namibia, and there was intense melting (Tp = 1500–1700°C) of even depleted peridotite in the plume head. KEY WORDS: dykes; crustal contamination; mantle plume; Namibia; picrites
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD RELATIONS AND SAMPLING...ANALYTICAL TECHNIQUESDYKE MAGMA TYPESHORINGBAAI MAGMA TYPEPETROGRAPHY OF OTHER DYKE...POST-GENESIS, PRE-EMPLACEMENT...HOW DID THE CRUSTAL...MAGMA GENESISROLE OF TRISTAN MANTLE...REFERENCES |
|---|
The extent to which decompression melts from upwelling mantle plumes contribute to the magmatism of Large Igneous Provinces (LIPs) is a continuing subject of intense debate (e.g. Mahoney & Coffin, 1997
). Recent accounts of LIPs, including those erupted on continents, envisage substantial or predominant contributions to the magmas from the convecting mantle (e.g. White & McKenzie, 1995; Lassiter & DePaolo, 1997; Saunders et al., 1997). Nevertheless, a group of LIPs that erupted on the former Gondwana supercontinent have been considered by some to originate from within the sub-continental lithospheric mantle (SCLM), with input from below confined to conductive heat transfer (Hawkesworth et al., 1999, 2000). Part of one of these, the Early Cretaceous Paraná–Etendeka LIP, is the subject of this study. Peate (1997) has reviewed the entire province in some detail and we shall focus here on only the southern Etendeka region of Namibia.
The chemical compositions of the lavas, dykes and sills of the Paraná–Etendeka province are bimodal, with predominant basic and subordinate acid groups. Although the minor-element and Sr–Nd–Pb isotopic compositions of the basic lavas divide them into several sub-types (Peate, 1997; Marsh et al., 2001), they are almost all Mg-poor basalts and tholeiitic andesites with MgO rarely exceeding 7 wt % in published analyses. Their compositions are largely controlled by low-pressure phase equilibria (see below) and they therefore provide restricted evidence about their genesis. Only in one area, the southern part of the Etendeka sub-province in NW Namibia, does a more varied assemblage of mafic lavas, dykes and sills accompany the Mg-poor lavas. Erlank et al. (1984) and Duncan et al. (1990) described basaltic dykes and inclined sheets cutting Etendeka lavas near the coast at Horingbaai (Fig. 1) and emphasized their mid-ocean ridge basalt (MORB)-like geochemistry. They called these the Horingbaai dolerites. Subsequently, a thin succession of Mg-rich basalt and picrite lavas was discovered at the base of the Etendeka volcanics in the Goboboseb Mts and Huab Outliers areas (Milner & le Roex, 1996; Gibson et al., 1997; Ewart et al., 1998a). These picrites are geochemically entirely different from the Horingbaai dolerites, and their unusual Fe-rich nature was emphasized and discussed by Gibson et al. (2000).
|
Abundant dykes occur throughout the southern Etendeka region. As most of them are emplaced in pre-Cretaceous rock-types, there has been doubt about their ages and, consequently, their relevance to Etendeka magmatism (Erlank et al., 1984). Erlank et al. and Marsh et al. (1991) sampled these dykes extensively and published limited geochemical data on them. Here we report more detailed studies of a new set of samples. We show how these throw further light on the nature of the initial head of the Tristan mantle plume and the processes that took place during its sub-continental impact.
|
FIELD RELATIONS AND SAMPLING STRATEGY |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD RELATIONS AND SAMPLING...ANALYTICAL TECHNIQUESDYKE MAGMA TYPESHORINGBAAI MAGMA TYPEPETROGRAPHY OF OTHER DYKE...POST-GENESIS, PRE-EMPLACEMENT...HOW DID THE CRUSTAL...MAGMA GENESISROLE OF TRISTAN MANTLE...REFERENCES |
|---|
Dyke swarms occur at intervals along the NW Namibian coast (Fig. 1). Similar near-coast dykes are widespread in SW Africa and thought to be related to Atlantic Ocean opening (Hunter & Reid, 1987
; Reid & Rex, 1994). The NW Namibian swarms show coast-parallel NNW–NW trends within 30 km of the sea but the swarms that penetrate further inland trend predominantly N–NNE (Erlank et al., 1984; Lord et al., 1996; Geological Survey of Namibia, 1997). Erlank et al. reported that the dykes are overwhelmingly basaltic (sensu lato) in composition and found neither olivine-rich variants nor intermediate latites. Marsh et al. (1991) subsequently noted that olivine-rich dykes are much more common than first supposed. Our samples come from five principal and several minor localities. These are numbered and marked by stars, respectively, in Fig. 1. At each principal locality (except 4; see below) we collected all dykes exposed along a traverse perpendicular to the swarm. It is clear from the preliminary geochemical data of Marsh et al. (1991) on a large number of Southern Etendeka dykes that our approach should have sampled them adequately.
Horingbaai coast
The original locality of the MORB-like Horingbaai magma type (Erlank et al., 1984, fig. 4) is between the Albin Ridge and the Atlantic Ocean (Fig. 1). Both Duncan et al. (1990, fig. 6) and Marsh et al. (1991, fig. 1) subsequently extended this coastal zone to the mouth of the Huab River (Fig. 1). Fine-grained dolerite dykes, inclined sheets and sills are no more than 2 m thick. As Erlank et al. (1984) emphasized, it is noticeable on the Albin Ridge outcrop of Etendeka lavas that the flows are hydrothermally altered but the sheets cutting them are fresh. The coastal dykes cut Damara basement granites. We sampled eight dykes and inclined sheets, trending 290–320°, along a traverse between S21°23'01'', E13°55'28'' and S21°33'09'', E13°52'50'' (roads C35 and D2303).
|
|
|
South of Brandberg
A dolerite dyke swarm runs inland from Henties Bay, 40 km SE of Cape Cross (Fig. 1), towards the Brandberg intrusive complex (Geological Survey of Namibia, 1997). We sampled all 21 dykes exposed along a traverse across this swarm 5–15 km south of Brandberg, between S21°14'59'', E14°26'31'' and S21°18'41'', E14°33'38'' along road D2342. Most of the dykes here are emplaced in 550 Ma post-tectonic granites of the Pan-African Damara mobile belt basement; a few cut Karoo sediments (Geological Survey of Namibia, 1997). Their strikes are extremely variable (290–90°) but most trend N–NE. Their thickness varies from about 20 cm to 4 m. Although many are fine grained, others have abundant <5 mm phenocrysts of olivine, plagioclase and sometimes clinopyroxene. Two similar dykes were sampled at S21°06', E14°52' beside road C35, east of Brandberg (Fig. 1).
Doros
A sparse dolerite dyke swarm immediately NW of the Doros intrusive complex cuts Damara mobile belt sediments (Fig. 1). The dykes all trend due north and vary from 0·5 to 3 m thick, and from feldspathic to olivine-rich in composition. We sampled all seven dykes exposed along a traverse between S20°43'26'', E14°18'37'' and S20°45'13'', E14°13'12''.
Huab Outliers
In the eastern part of the Huab Outliers (also known as Awahab) area (Fig. 1) we sampled five of the dolerite dykes that cut the Etendeka lavas and the underlying Early Cretaceous Etjo aeolian sediments (Duncan et al., 1998; Jerram et al., 1999). These dykes are 0·5–2 m wide, trend 285–05° and range from feldspathic to olivine rich. In this region we did not attempt a complete traverse across the swarm because these dykes have been sampled comprehensively by D. A. Jerram (personal communication, 1999).
Nil Desperandum–Twyfelfontein
A swarm of dykes runs NNW between Twyfelfontein and Nil Desperandum, a farm immediately south of the Tafelberg type-section of the southern Etendeka lavas (Erlank et al., 1984; Duncan et al., 1998; Jerram et al., 1999). Scattered sub-parallel dolerite dykes are found across a zone 80 km wide at S20°15' to 20°30' (Fig. 1). Erlank et al. (1984) sampled and analysed a few of these from the Tafelberg area. We collected eight samples along roads C39 and D2628, between S20°26'58'', E14°43'24'' and S20°17'46'', E13°57'39'', mostly around S20°28', E14°20'. The dolerites vary from feldspathic to olivine rich. Although some are thin (1 m), the remainder are notably thicker (5–50 m) than elsewhere in the region. A single 1·5 m vitreous acid dyke (PB11) was also sampled.
|
ANALYTICAL TECHNIQUES |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD RELATIONS AND SAMPLING...ANALYTICAL TECHNIQUESDYKE MAGMA TYPESHORINGBAAI MAGMA TYPEPETROGRAPHY OF OTHER DYKE...POST-GENESIS, PRE-EMPLACEMENT...HOW DID THE CRUSTAL...MAGMA GENESISROLE OF TRISTAN MANTLE...REFERENCES |
|---|
Our initial collection comprised 51 dolerite and associated dykes, distributed as above, eight lavas from the basal Etendeka flows and two micaceous mafic potassic–ultrapotassic dykes each from the Erongo and Okenyenya plutonic complexes (Fig. 1). After rejecting any with substantial hydrothermal alteration, we analysed them all by X-ray fluorescence (XRF) and most by inductively coupled plasma mass spectrometry (ICP-MS). Analyses of 43 samples are listed in Table 1. The analysed dykes excluded from this study are alkalic (three samples) or of uncertain geochemical affinities. Most of the latter have MgO below 8 wt % and cannot be allocated unequivocally to the magma types with which this study is concerned (see below).
The elemental analyses (XRF, ICP-MS) were made at Durham University and the isotopic analyses were made at McMaster University. All our current analytical techniques have been described by Gibson et al. (1999, appendix) and elemental analyses of blanks and standards used in this study have been given by Gibson et al. (2000, online background dataset).
|
DYKE MAGMA TYPES |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD RELATIONS AND SAMPLING...ANALYTICAL TECHNIQUESDYKE MAGMA TYPESHORINGBAAI MAGMA TYPEPETROGRAPHY OF OTHER DYKE...POST-GENESIS, PRE-EMPLACEMENT...HOW DID THE CRUSTAL...MAGMA GENESISROLE OF TRISTAN MANTLE...REFERENCES |
|---|
Because dykes lack stratigraphic evidence for their relative ages and it is impractical to date them all radiometrically, we have adopted the approach of grouping them chemically into magma types before describing the petrography and geochemistry of each group. The nomenclature in the following account uses the new IUGS definitions of the terms picrite and komatiite (Le Bas, 2000
). Two basaltic magma types have been recognized in the southern Etendeka lavas:
- olivine to quartz tholeiites, with MgO generally <5 wt %, form most of the volcanic successions. These were called Tafelberg and Albin types by Erlank et al. (1984) and LTZ-L (‘Low Ti and Zr lava type; sub-type with lower Ti/Zr ratios’) by Ewart et al. (1998a). Marsh et al. (2001) have introduced further subdivisions.
- In the Huab Outliers and Goboboseb Mts areas the LTZ-L flows are underlain by a thin basal suite of olivine tholeiite lavas that are distinctively rich in both total Fe and MgO. Ewart et al. (1998a) called these LTZ-H; other workers prefer the term Tafelkop (e.g. Milner & le Roex, 1996; Jerram et al., 1999; Marsh et al., 2001). Gibson et al. (2000) showed that this phenomenon of a brief episode of Fe–Mg magmatism at the outset of flood-basalt volcanism has happened worldwide since at least the Proterozoic. They suggested that ‘ferropicrite’ was the best name for such lavas. Ewart et al. (1998a) described the petrology and geochemistry of both lava magma types in detail and compared them with the low-Ti Paraná magma groups (Peate, 1997). Erlank et al. (1984) and Duncan et al. (1990) showed that olivine tholeiite dolerite dykes and sills along the Horingbaai coast (Fig. 1) included a third magma type (called Horingbaai by them) that showed strong trace-element and isotopic resemblances to MORB.
In Fig. 2a we plot TiO2 vs Zr, showing the previously defined fields for Etendeka magmatism (e.g. Peate, 1997; Duncan et al., 1998). This diagram clearly separates all the Etendeka lava suites and associated minor intrusions, except for the original Horingbaai dolerites and the ferropicrite field, which partially overlap. Figure 2b shows that a plot of chondrite-normalized La/Nb vs Sm/Lu separates these magma types more clearly (at least in southern Etendeka). Most of the dykes we have sampled are from the Horingbaai magma type (Table 1). Members of this suite dominate the Mg-rich basalt–picrite dykes at the Coastal, Brandberg and Huab localities but we have not found them north of the Huab River (Fig. 1). At Doros the Horingbaai-type dykes are accompanied by ferropicrites. The reasons for using two symbols (filled stars and circles) for sub-types of the Horingbaai dykes will be explained below, when their geochemistry is discussed in detail.
Five of our dyke samples (from the Brandberg and Huab swarms) have the high La/Nb ratios characteristic of the LTZ-L field in Fig. 2b and four of these dykes also plot in or adjacent to the LTZ-L field in Fig. 2a. Finally, it is clear from Fig. 2b that the magnesian dykes of the Nil Desperandum–Twyfelfontein swarm (excluding ferropicrite PB9) define a separate field that combines relatively high values of both (La/Nb)n and (Sm/Lu)n. We therefore recognize a fourth magma type, Nil Desperandum (ND) in the Southern Etendeka area. During this study we have also analysed one sample from the extensive Huab Sill complex (Fig. 1). It classifies with the ND magma type in Fig. 2. Table 2 summarizes the occurrence of the four magma types shown in Fig. 2b at the five sampled dyke localities (Fig. 1).
|
|
Ages of the dykes
Although it is, of course, impossible to be certain about the absolute ages of a suite of dykes without dating almost every sample, we consider that each of the magma types shown in Fig. 2b is sufficiently distinctive to have been probably associated with a single magmatic event. The ages of the dyke groups can be constrained by their field relationships to the well-dated lava pile (Renne et al., 1996). We assume that the distinctive ferropicrite dykes are coeval with the nearby ferropicrite basal lavas of the Huab Outliers and Goboboseb Mts (Fig. 1) at 133 Ma (Jerram et al., 1999). Likewise, the handful of LTZ-L dykes in our collection are probably related to the predominant Tafelberg/LTZ-L lavas of southern Etendeka (Erlank et al., 1984; Ewart et al., 1998a, 1998b), erupted at 132 Ma (±0·7 Ma; 2
) according to Renne et al. (1996). A single date of 130·3 Ma (±1·2 Ma; 1) by Stewart et al. (1996) for an Etendeka rhyolite is probably not significantly younger.
The Horingbaai dykes cut these lavas in both the Albin Ridge and Huab Outliers areas (Fig. 1). The predominantly Horingbaai dykes of the swarm south of Brandberg cut baked lavas adjacent to the Brandberg granite complex but both are intersected by the granites. Schmitt et al. (2000) recently published a detailed 40Ar/39Ar geochronological study of these granites. They showed that their two dates below 130 Ma had both been lowered by secondary loss of radiogenic Ar. They found concordant plateau and total-gas dates between 130·5 and 133·0 Ma, using biotite, arfvedsonite and astrophyllite separates from several plutons. We accept their inference that the youngest plausible age for the Brandberg complex is 131 Ma (±1 Ma); it could well be 1 my older. Nevertheless, Renne et al. (1998) have measured Ar/Ar ages several million years younger in coastal dykes and it is widely thought (e.g. Hawkesworth et al., 1999) that these are all of the Horingbaai magma type. Our own limited sampling of the Horingbaai–Albin Ridge coastal area (Fig. 1) found only one sheet with MgO >8 wt % that is fresh and clearly Horingbaai-type (Table 1, 97SB29). Another (unpublished) analysis of a fresh dolerite dyke with 10% MgO is alkalic, appears to be younger, and will not be considered further here (see Marsh et al., 1991). The remainder (Table 1) all have MgO <8% (Fig. 2) and might be fractionated Horingbaai melts that have undergone considerable crustal contamination (see below). The coastal dykes need more geochemical study, to clarify their affinities. Finally, the Nil Desperandum dykes also appear to have been emplaced just after 132 Ma because one of them cuts Etendeka lavas west of Tafelberg (Fig. 1) and, as a group, they closely resemble chemically some of the dolerites of the Huab Sill suite, dated at 132 Ma (Renne et al., 1996).
|
HORINGBAAI MAGMA TYPE |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD RELATIONS AND SAMPLING...ANALYTICAL TECHNIQUESDYKE MAGMA TYPESHORINGBAAI MAGMA TYPEPETROGRAPHY OF OTHER DYKE...POST-GENESIS, PRE-EMPLACEMENT...HOW DID THE CRUSTAL...MAGMA GENESISROLE OF TRISTAN MANTLE...REFERENCES |
|---|
Petrography
All the dykes that we are confident to identify as Horingbaai type (Fig. 2b) have olivine macrocrysts that vary in amount from tens of per cent in the picrites to sparse when MgO is <9%. These occur in two morphologies: (1) equant macrocrysts
1–6 mm in size—the smaller ones are euhedral–subhedral and the larger ones are subhedral to rounded; (2) sparse elongate crystals, 0·1–0·4 mm wide, showing up to 15:1 elongation and frequently displaying a central cavity (i.e. they are tubes). Some of the equant olivine macrocrysts are in clusters of a few crystals, with curved grain boundaries between individuals. Deformation lamellae are very rare in both olivine morphologies. All the olivine macrocrysts enclose small, equant, euhedral–subhedral, dark brown Cr–Al spinels.
Euhedral <2 mm plagioclase phenocrysts accompany the olivine in almost all Horingbaai dolerites. They are sparse in the picrites and become the dominant macrocryst phase when MgO falls below 8%. We found clinopyroxene macrocrysts that might be phenocrysts in only one (8·5% MgO) dolerite. Even in this sample, the provenance of the clinopyroxene is uncertain because the rock contains clinopyroxene-rich gabbro micro-xenoliths. The groundmass of all our samples is holocrystalline, granular to sub-ophitic and rarely variolitic. It is composed of plagioclase, pale green clinopyroxene, olivine in varying amounts and Fe–Ti oxides. Careful search of the coarsest groundmasses (1 mm) found no traces of magmatic hydrous minerals. Minor sub-solidus alteration to serpentine, chlorite and sericite affects some samples but many are extremely fresh.
Mineral compositions
A comprehensive mineral chemistry investigation of the Southern Etendeka dykes falls outside the aims of this research. Nevertheless, we have studied the olivine macrocrysts, and their included chromite, in the Horingbaai dykes, to discover how closely the Mg-rich basalts and picrites fit a model of olivine accumulation and fractionation.
Olivine
Representative analyses of olivine macrocrysts in the Horingbaai dykes are given in Table 3. In this paper we use the common practice (e.g. Nisbet et al., 1993; Garcia, 1996) of referring to olivine mg-number [100Mg/(Mg + Fe)] as per cent forsterite. Although some of the olivines are exceptionally Mg rich (Fo93·3), their CaO contents are all within the range appropriate to phenocrysts precipitated from basalt–picrite magmas and much higher than the CaO of olivines in mantle xenoliths (Thompson & Gibson, 2000, fig. 1). Both the textural relationships of the small chromites scattered within each olivine macrocryst, and the chromite compositions themselves (see below), support this crucial point. The average values of minor elements in olivines with Fo >92 are: CaO = 0·30% (2 = 0·02%); NiO = 0·42% (2 = 0·03%); Cr2O3 = 0·13% (2 = 0·03%).
|
Figure 3 shows our individual microprobe analyses as functions of the mg-number of the host dyke. We calculated rock mg-number with 10% of total Fe as Fe3+ (Thompson & Gibson, 2000). Two values of Kd for the distribution of Fe2+ and Mg between olivine and melt are plotted: 0·30 is appropriate for equilibria between picrites and their olivines at 1 atm, whereas 0·31 applies at 0·5 GPa pressure (Ulmer, 1989). Garcia et al. (1995) and Garcia (1996) showed how some Hawaiian basalts and picrites from Mauna Loa and Mauna Kea are in equilibrium with their olivines, whereas others are best explained as cumulates because their olivines plot to the right of the equilibrium Kd (i.e. the rock is more magnesian than appropriate to precipitate the olivines it contains). Only one Horingbaai sample, 97SB62, has olivine phenocrysts that could be in equilibrium with the bulk rock containing them. Several of the dykes (e.g. 97SB29, 33, 35, 37 and 53) have more than one macrocryst compositional population. Picrites 97SB34, 37 and 41 plot to the right of the equilibrium Kd lines and hence may be interpreted as olivine cumulates. The most Mg-rich olivine in both 97SB34 and 41 is Fo86–87, which corresponds to a melt mg-number of 66–67 (Fig. 3). These figures may be plotted on a graph of mg-number vs MgO for the Horingbaai dykes (not shown), to give a value of 12·0% MgO for the liquid that precipitated the olivines (magmatic Fe2O3 set at 10% of total Fe).
|
In terms of olivine–melt equilibria, all the olivines in Fig. 3 that plot to the left of the Kd = 0·30 curve are magnesian xenocrysts in the rocks that contain them, but we avoid the use of the term ‘xenocryst’ here because it causes confusion with accidental crustal and lithospheric mantle (SCLM) xenocrysts, and prefer the neutral term ‘macrocryst’. Using the same method as for picrites 97SB34 and 41 above, we calculate that the Fo90+ macrocrysts in samples 97SB29, 33, 35, 46 and 53 originated in melts with MgO between about 16 and 24%. We have considered the most magnesian olivine (Fo93·3) in detail elsewhere (Thompson & Gibson, 2000) and calculated that the most likely MgO content of its parent melt is 24·0%.
Spinel
Representative microprobe analyses of the opaque phase within Horingbaai olivine macrocrysts are given in Table 3. In this account we use the term spinel in a general way; in the IMA nomenclature (Nickel, 1992) they are chromite, spinel (sensu stricto) and rare magnetite (Fig. 4), with mg-number from 16·5 to 69·3. The Mg-rich spinels [MgO = 11·0–15·7%; (Fe2O3 + FeO) < 29%; TiO2 < 0·7%] are predominantly Al rich, with Al2O3 = 31·5–35·5% and Cr2O3 = 21·9–23·7%, but there is a smaller, more variable, population with Al2O3 = 18·0–22·6% and Cr2O3 = 34·8–46·2%. Analyses near some grain margins and cracks that cross the olivine macrocrysts show a trend towards low MgO, Al2O3 and Cr2O3, and higher (Fe2O3 + FeO) 66% and TiO2 7%; i.e. a chromite-to-magnetite trend.
Figure 4 compares these groups and trends with data for Kilauea Iki lava lake (Scowen et al., 1991) and komatiites (Barnes, 1998). The Horingbaai spinels show the compositional trends that Scowen et al. (1991) identified as caused by re-equilibration of primary chromite with interstitial melt at temperatures between liquidus and solidus. These are a rising ratio of Fe3+ to total trivalent cations, and rising TiO2, with falling mg-number (Fig. 4b and d). To investigate further, we analysed pairs of olivine macrocrysts and their enclosed spinels, at points 15 µm on each side of olivine–spinel grain boundaries (Barnes, 1998). Using the methods of Sack & Ghiorso (1991) and Wood (1991), we calculated T and fO2 values for each olivine–spinel pair. The two sets of results overlap, with T = 800–1020°C and fO2 = 4–9 log units below the fayalite = magnetite + quartz (FMQ) buffer.
|
PETROGRAPHY OF OTHER DYKE TYPES |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD RELATIONS AND SAMPLING...ANALYTICAL TECHNIQUESDYKE MAGMA TYPESHORINGBAAI MAGMA TYPEPETROGRAPHY OF OTHER DYKE...POST-GENESIS, PRE-EMPLACEMENT...HOW DID THE CRUSTAL...MAGMA GENESISROLE OF TRISTAN MANTLE...REFERENCES |
|---|
We have not studied the petrography and mineralogy of the other dyke types in as much detail as the Horingbaai suite because they are not so well represented in our collections. We have found no traces of hydrous minerals, except for late-stage alteration to serpentine and chlorite in some cases.
Ferropicrite dykes
Ewart et al. (1998a) and Gibson et al. (2000) have described the petrography and mineralogy of the ferropicrite (LTZ-H) lavas and dykes. There are two principal petrographic differences between the ferropicrite (and related) and Horingbaai samples: (1) equant euhedral–subhedral olivine phenocrysts (up to Fo86) in the ferropicrites are only 2 mm and crystals with the very elongate morphology are absent; (2) sparse <1·5 mm euhedral clinopyroxenes accompany the olivine in almost all samples, whereas plagioclase phenocrysts are absent until MgO falls below 7%.
LTZ-L dykes
Erlank et al. (1984) and Ewart et al. (1998a) have described the petrography and mineralogy of the Southern Etendeka Tafelberg–Albin–LTZ-L lavas. All of our samples are aphyric, with <1·5 mm grain size. They mostly comprise granular intergrowths of clinopyroxene, plagioclase and minor Fe–Ti oxide, with some olivine in all except the most MgO-poor sample (97SB69). A subtle difference between the LTZ-L and other dyke suites is that the former have a few per cent of isotropic interstitial glass.
Nil Desperandum dykes
This magma type has not been previously recognized in the southern Etendeka area and its dykes lack any distinctive petrographic features. ND dolerites with MgO >8% have euhedral–subhedral, equant olivine phenocrysts, mostly <1 mm (rarely up to 2 mm) in a groundmass indistinguishable from those of Horingbaai dolerites, except that clinopyroxene is slightly pinkish rather than greenish in colour. As MgO falls below 8%, the texture becomes very fine grained and aphyric, and eventually glassy (devitrified).
|
POST-GENESIS, PRE-EMPLACEMENT MAGMATIC PROCESSES |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD RELATIONS AND SAMPLING...ANALYTICAL TECHNIQUESDYKE MAGMA TYPESHORINGBAAI MAGMA TYPEPETROGRAPHY OF OTHER DYKE...POST-GENESIS, PRE-EMPLACEMENT...HOW DID THE CRUSTAL...MAGMA GENESISROLE OF TRISTAN MANTLE...REFERENCES |
|---|
In this section we consider two aspects of the post-genesis evolution of the southern Etendeka mafic magmas that form the dykes: (1) phase-equilibria evidence of their magmatic evolution; (2) elemental and isotopic evidence that some of them behaved as open systems during uprise, gaining contributions from the crust, SCLM or both.
Phase equilibria
Figure 5 is a CIPW-normative plot that is useful for showing the effects of cotectic crystallization and olivine accumulation. All the norms were calculated with Fe2O3 set at 10% of total Fe (Thompson & Gibson, 2000). Figure 5a supports the view that the final stage of evolution of the Horingbaai magmas took place at low pressures; samples with <9% MgO plot just below and along the 1 atm cotectic for basaltic liquids in equilibrium with olivine, Ca-rich clinopyroxene and plagioclase. We showed above (Fig. 3) that Horingbaai compositions with <12% MgO could have once been all liquid, judging from the compositions of their olivine phenocrysts, but that more Mg-rich magmas contained accumulative olivine.
|
In Fig. 5b we show the published analyses of southern Etendeka ferropicrite and Fe-rich basalt lavas as a field. The four dykes that we identified above as ferropicrites (e.g. Fig. 2) fall in this field. Gibson et al. (2000) discussed the lavas and showed that the ferropicrite magma type behaves in a similar way to the Horingbaai on this normative projection; i.e. the samples have a range of silica saturation and the more MgO-rich compositions plot closer to the Ol apex than those with less MgO. The ferropicrites carry additional evidence of high-pressure magmatic evolution, in the form of aluminous clinopyroxene phenocrysts (Gibson et al., 2000).
Only 10% of the very large number of published analyses of Paraná–Etendeka low-Ti flood-basalt lavas have >5% MgO. The fields for these are superimposed in Fig. 5b and lie slightly below the 1 atm cotectic. We agree with Peate (1997) and Ewart et al. (1998a) that these lavas are products of extensive fractional crystallization at low pressures. The depths at which this evolution occurred can be estimated because the cotectic moves downwards on the normative plot with increasing pressure up to 0·6 GPa and would pass through the axes of the Paraná–Etendeka low-Ti lava fields at 0·2–0·3 GPa pressure (Grove et al., 1992). One of the small number of LTZ-L dykes that we have identified above (Fig. 2) falls within the field of the Etendeka low-Ti (Tafelberg) lavas in Fig. 5b. The others have variable silica saturation and lower normative diopside. All we can say from our few LTZ-L samples (6·1–11·3% MgO; Table 1) is that there appear to be MgO-rich progenitors of the low-Ti lavas amongst the dykes but that the samples identified here are too few to construct a petrogenetic model. Marsh et al. (1997) have given a preliminary report that picritic examples of the LTZ-L magma type are abundant in a dyke swarm south of the area shown in Fig. 1.
The Nil Desperandum (ND) dykes show yet another distinctive pattern in Fig. 5b. With one exception, the analyses plot in the same field as the ferropicrites but the behaviour of MgO in the ND suite is completely different: it falls progressively with rising silica saturation. Such behaviour suggests magmatic evolution controlled by cotectic crystallization. There is indeed an experimentally determined cotectic in the natural basic magmatic system that falls close to the ND data points in Fig. 5b. This is for closed-system anhydrous evolution of basic liquids from olivine tholeiite to ne-normative hawaiite, by fractionation of olivine + plagioclase + (dominant) aluminous subcalcic clinopyroxene (Thompson, 1974, 1982; Grove et al., 1992) at continental base-of-crust pressures (0·9 GPa). But falling T and liquid MgO along this cotectic are in the opposite direction from that required to fit the ND dyke suite. We shall use geochemical data below to argue that the ND magmas show evidence of open-system evolution.
Crustal contamination
Despite the clearly defined chemical groups of Southern Etendeka magmas (Fig. 2), there is considerable elemental and Sr–Nd–Pb isotopic variation within each magma group. Figure 6a is a set of normalized incompatible-element plots, which demonstrates that a few examples of each of the Horingbaai, ferropicrite and Nil Desperandum magma types have patterns very similar to those of typical oceanic basalts. Ferropicrites 96SB48 and PB9 resemble Hawaiian ocean-island basalt (OIB) (Gibson et al., 2000), whereas Horingbaai sample 97SB29 is close to the average E-MORB, except for a slight relative depletion in Yb and Lu. All of our small ND dyke set have incompatible element patterns that are not quite so smooth as 96SB48 and 97SB29; they tend to have troughs at Nb and Ta, and peaks at Rb and K. The example with the smoothest patterns is shown in Fig. 6a.
Figure 6b shows, for contrast, some of the most ‘spiky’ Southern Etendeka dyke patterns. One representative >9% MgO analysis is given for each magma type, except for Horingbaai where 97SB34 is typical of the predominant spiky patterns and 97SB79 shows a sub-type found in the Huab Outliers area (Fig. 1). Diversity amongst the ratios of elements to the left of La is clearly substantial and we shall attempt to use these and isotopic data to try to identify the causes of this variation.
Horingbaai
The three Horingbaai dykes with MgO >9 wt % and incompatible element patterns that most closely resemble average E-MORB (Fig. 6a) are 97SB29 (Horingbaai coast swarm), 97SB53 (south of Brandberg) and 97SB75 (Huab Outliers). Sample 97SB29 is shown in Fig. 6a; the others are in Table 1. On a plot of initial 87Sr/86Sr vs Nd (Fig. 7a) these samples fall close together at 87Sr/86Sri 0·7032 and Nd = 7·8–9·0; this is an area of the diagram where the fields of MORB and OIB overlap. It is logical to suppose that the source of these three samples was largely, if not entirely, within the convecting mantle. In contrast, Horingbaai dykes with spiky incompatible-element patterns form an array in Fig. 7a, towards high 87Sr/86Sri. These are unlike all OIB and form a trend towards a very 87Sr-rich end-member composition that could, in theory, reside in either the crust or the SCLM.
Cretaceous Mg-rich potassic–ultrapotassic rock-types (e.g. minettes, kamafugites, lamproites) are scattered throughout the Paraná–Etendeka LIP. It is generally agreed that the source of these melts was within the most-fusible regions of the SCLM underlying the area (Gibson et al., 1995a, 1995b, 1999; Carlson et al., 1996; Milner & le Roex, 1996; le Roex & Lanyon, 1998; Thompson et al., 1998). These rock-types therefore allow assessment of the geochemical characteristics of any fusible SCLM fraction that may have contributed to the Paraná–Etendeka basaltic melts. The abundant Brazil–Paraguay examples define a trend in Fig. 7a towards very low Nd, with 87Sr/86Sri remaining below 0·708 (e.g. Gibson et al., 1995a, 1995b). Comparable samples from dykes in the Damaraland Cretaceous plutonic complexes (Fig. 1 inset) plot around the high-Nd, low-87Sr/86Sri end of this array (Table 1; 96SB30, 97SB94), as do published data from these complexes (Milner & le Roex, 1996; le Roex & Lanyon, 1998). Unless one postulates a high-87Sr/86Sr fusible fraction within the sub-Etendeka SCLM that escaped sampling by the Mg–K-rich magmatism, it is difficult to see why the SCLM should be considered as a significant contributor to the Horingbaai magmas. The incompatible-element patterns summarized in Fig. 6 reinforce this point. It should be noted that La/Nb is higher in the Horingbaai dykes with spiky incompatible-element patterns (Fig. 6b) than those with relatively smooth patterns (Fig. 6a). This is the opposite sense of change in La/Nb from that which would occur if Damaraland Cretaceous Mg-rich potassic–ultrapotassic melts (Fig. 6c) were added to the latter.
Turning to continental crust as the site of the high-87Sr/86Sr and relatively low-Nd end-member, we can consider this in general terms of sediments, low-grade metamorphics and granites overlying deep-crust, high-grade metamorphics. Acid rock-types are the ones potentially most fusible by upwelling basic–ultrabasic magmas (Thompson, 1981). Depending on their ages, these acid rocks have Nd values between about -5 and <-20, and 87Sr/86Sr varies greatly because high-grade metamorphism drives off Rb in fluids, melts or both (Dickin, 1997). Therefore, ancient granulite-facies acid gneisses have a distinctive combination of both low 87Sr/86Sr and Nd, whereas amphibolite-facies gneisses may show 87Sr/86Sr up to 0·72 (Dickin, 1997). McDermott et al. (1989) and McDermott & Hawkesworth (1990) showed that a combination of ancient sources, high Rb/Sr and light rare earth element (LREE) enrichment caused extreme isotopic ratios (up to 87Sr/86Sr = 0·87 and Nd = -24·3 at 132 Ma) in Damara mobile belt clastic sediments.
In Fig. 7a we show a mixing curve between Horingbaai sample 97SB75 and average Damara sediment composition (87Sr/86Sr = 0·7634; Nd = -12·6). It passes through the array formed by the Horingbaai suite, apart from two samples with higher 87Sr/86Sri. This diagram shows that amounts of added sediment up to only 7% are sufficient to account for the Sr–Nd isotopic range of the Horingbaai magmas. Despite the variability of individual samples of each rock-type, published analyses of the Damara belt granites have an average composition close to that of the sediments. Granite average values (at 132 Ma) are: 87Sr/86Sr = 0·7728 and Nd = -14·8; both granite and sediment have 125 ppm Sr (McDermott et al., 1989, 1996; McDermott & Hawkesworth, 1990). The two samples with very high 87Sr/86Sr at a given Nd value (Fig. 7a) may have selectively assimilated upper-crust alkali feldspar. We shall return to this point below, in connection with Pb abundances and isotopic ratios (see Fig. 10e).
Having eliminated Mg-rich potassic–ultrapotassic SCLM melts as a possible end-member to account for the variation in Horingbaai magma geochemistry, it is likely that any contribution to the melts from fusible crustal rock-types or alkali feldspar was relatively Nb poor (Rudnick & Fountain, 1995). Therefore Rb/Nb should monitor Rb addition to the Horingbaai magmas. Figure 7b uses Rb/Nb and initial 87Sr/86Sr to show that trace-element and radiogenic isotopic ratio variations are linked in the Horingbaai magmas.
Pb isotopic ratios are plotted in Fig. 8. The three high-Nd, low-87Sr/86Sri, E-MORB-like Horingbaai samples cluster at 208Pb/204Pb 37·63, 207Pb/204Pb 15·57 and 206Pb/204Pb 18·03. The samples with higher 87Sr/86Sri and spiky incompatible-element patterns form elongate arrays on this diagram, trending towards higher 208Pb/204Pb, 207Pb/204Pb and 206Pb/204Pb. The Horingbaai samples that plot furthest from 97SB29, 53 and 75 fall adjacent to the field of Southern Etendeka latite lavas, considered by Ewart et al. (1998b) to contain a high fraction of crustal melt. Both average Damara mobile belt fine-grained clastic sediment (McDermott & Hawkesworth, 1990) and some of the Mg-rich potassic–ultrapotassic dykes from Damaraland Cretaceous complexes (SCLM melts) also plot in this part of the diagram. Nevertheless, the Sr–Nd isotope relationships discussed above rule out the latter as significant contributors to the Horingbaai magmas. We therefore conclude that the ‘continental’ component in this basalt–picrite suite is upper crust. Pb abundances in the Horingbaai samples are concordant with this proposal. The three most MORB-like samples (97SB29, 53, 75) have Pb contents of 0·49–0·68 ppm (Table 1), whereas samples with spiky incompatible-element patterns have up to 8 ppm Pb. Damara belt sediments average 18 ppm Pb (McDermott et al., 1989; McDermott & Hawkesworth, 1990). We shall show below (Fig. 10e) that the process whereby crustal lead entered these magmas was more complex than either bulk assimilation of crust or linked assimilation and fractional crystallization (AFC).
|
.
|
Ferropicrites and Nil Desperandum
Our suites of ferropicrite and ND dykes are too small to consider in the same detail as the Horingbaai suite. In Fig. 7a they form an array with the samples that have smoothest incompatible-element patterns (Fig. 6a) plotting around 87Sr/86Sri = 0·704 and
Nd around +1 to +3, and those with spiky elemental patterns (Fig. 6b) having higher 87Sr/86Sri and lower Nd. As with the Horingbaai suite, it is hard to see how a fusible lithospheric mantle (SCLM) fraction could be responsible for the high-87Sr/86Sri and low-Nd ferropicrite and ND melts because the Mg-rich potassic magmas of the Cretaceous Damaraland intrusive complexes cluster tightly within the same part of Fig. 7a as the most OIB-like dyke samples, extending to lower 87Sr/86Sri ratios (Milner & le Roex, 1996; le Roex & Lanyon, 1998). Trace-element data make the same point because, as Fig. 6b and c shows, dykes with spiky trace-element patterns have higher La/Nb than the OIB-like ones, whereas the Mg-rich ultrapotassics also have relatively low (OIB-like) La/Nb.
The Pb isotope ratios of the ferropicrite and ND magmas (Fig. 8) form a single broad trend, dominated by variation in 206Pb/204Pb. The published data for Goboboseb LTZ-L lavas (Ewart et al., 1998a) fall at the high-206Pb/204Pb end of this array. The high-206Pb/204Pb ferropicrite–LTZ-L–ND samples plot close to the fields for Tristan and Gough basalts but at lower 208Pb/204Pb. The lowest 206Pb/204Pb samples plot to the right of the Geochron in Fig. 8b. One of the three continental flood basalt (CFB)-like dykes that we have identified (97SB36) shows the same feature.
The trend of Pb isotopes in these Southern Etendeka dyke suites towards low 206Pb/204Pb values could be related to an end-member that was either EMI-like convecting mantle, or fusible lithospheric mantle (SCLM) or lower crust (Rudnick & Goldstein, 1990; Dickin, 1997). Milner & le Roex (1996) favoured the convecting-mantle EMI hypothesis. At first sight the SCLM alternative is supported because mafic ultrapotassic magmas from both Damaraland and formerly adjacent parts of southern Brazil plot in the appropriate parts of the Pb isotope diagrams (Fig. 8). Nevertheless, our new elemental data can be used to show that crustal contamination appears to be a better hypothesis. A plot of U/Pb vs 206Pb/204Pb (Fig. 9) shows that all our Southern Etendeka samples have much lower U/Pb than OIB in general and the Tristan–Gough basalts in particular. The only slight overlap is between two OIB samples and the Horingbaai dykes that have MORB-like Sr–Nd isotopic ratios. The global MORB range of U/Pb overlaps that of the Southern Etendeka samples but it is clear from Table 1 and Figs 2, 6, 7 and 8 that the ferropicrite and ND parental magmas were definitely not MORB. It is also apparent from Fig. 9 that lithospheric partial melts resembling the Damara–Southern Brazil Cretaceous mafic potassic–ultrapotassic magmas could not lower U/Pb in the Southern Etendeka melts because they also have relatively high, OIB-like U/Pb.
|
Figure 9 implies that, although upper-crust contamination is plausible as a component in some Horingbaai magmas, it cannot be the cause of the low 206Pb/204Pb in some members of the ferropicrite and ND suites. The most likely component to have generated both low U/Pb and 206Pb/204Pb in the ferropicrite and ND suites (and one of the LTZ-L dykes) is granulite-facies crust. In Fig. 9 we illustrate this point with data for Archaean and Phanerozoic granulites (Rudnick & Goldstein, 1990; Rudnick & Presper, 1990; Rudnick & Fountain, 1995). Values of 206Pb/204Pb below 17·5 in some of the ferropicrite and ND magmas suggest that the granulite-facies crustal contaminant was geologically old. A likely source would be material from the Archaean–Proterozoic Congo craton (Fig. 1 inset), incorporated into the deep crust beneath Southern Etendeka during the development of the Pan-African Damara Orogen (Miller, 1983). Gibson et al. (2000) attempted to model this process for the ferropicrites, by using an Archaean leucogranite granulite-facies gneiss from NW Scotland. The model showed that assimilation of 15% of such crust could explain the observed isotopic variation.
|
HOW DID THE CRUSTAL CONTAMINATION TAKE PLACE? |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD RELATIONS AND SAMPLING...ANALYTICAL TECHNIQUESDYKE MAGMA TYPESHORINGBAAI MAGMA TYPEPETROGRAPHY OF OTHER DYKE...POST-GENESIS, PRE-EMPLACEMENT...HOW DID THE CRUSTAL...MAGMA GENESISROLE OF TRISTAN MANTLE...REFERENCES |
|---|
The chemical features that we discussed above, as signs of crustal inputs to the magmas, do not behave systematically within our relatively small ferropicrite and ND sets of samples. Nevertheless, the Horingbaai dykes in the Brandberg swarm provide an ideal set of data to consider more closely. Figure 10 shows the normalized REE patterns of the samples with analysed olivine macrocrysts (Fig. 3), divided into basalt and picrite groups by MgO contents above and below 12 wt %. MORB-like Horingbaai basalt 97SB29 (Horingbaai coast swarm) is plotted for comparison. The picrites (Fig. 10b) have distinctly steeper REE patterns than the basalts (Fig. 10a). Furthermore, a plot of REE pattern slope (La/Yb)n vs MgO (Fig. 10c) shows a scattered but clear positive correlation between these parameters. The most fusible upper-crust rock-types of the Damara orogenic belt have much higher (La/Yb)n than the Horingbaai dykes (Fig. 6d).
|
). 
, samples 97SB29 and 75 from other swarms (see text). (d) Normalized La/Yb vs modal volume per cent olivine (2000 points counted per thin section) for the same samples as in (c). (e) 1/Pb vs initial 206Pb/204Pb in all the Horingbaai dykes that we have analysed isotopically. MgO values are marked beside each point. Unfilled crosses are the three least contaminated samples (see text). Damara sediment average from McDermott & Hawkesworth (1990)
A positive correlation between amount of crustal contamination and MgO in consanguineous magmas could arise in two ways: (1) assuming all-liquid magmas, together with approximately similar dyke widths, crustal geology and melt flow rates, a hotter upwelling magma would melt a larger amount of fusible crust than a cooler one (Huppert & Sparks, 1985). Kerr et al. (1995) suggested the term ATA (assimilation during turbulent ascent) for this process. (2) If the MgO range was instead controlled by varying amounts of olivine phenocrysts in a liquid of approximately constant composition, then the picrites would have had lower REE contents than the basalts. Addition of a similar amount of LREE-enriched crust to all magma batches would then change (La/Yb)n more in the picrites than in the basalts. Figure 10d shows that (La/Yb)n is indeed higher in the olivine-rich than the olivine-poor dykes.
The second option fits the Brandberg Horingbaai data better because their olivine phenocryst assemblages are complex and characterized, in many cases, by magnesian macrocrysts that are too forsteritic to have precipitated from the magmas that carried them (Fig. 3). This situation can hardly have arisen at the crustal depth now exposed at the surface; the Horingbaai melts must have collected their forsteritic macrocrysts before their final uprise and freezing. Therefore, we attribute the upper-crust REE contamination of the Horingbaai magmas to fusion or dissolution of crust during the uprise of melts with variable macrocryst contents. This is similar, but not identical, to the ATA mechanism of Kerr et al. (1995).
In the case of relatively simple crustal contamination models, plots of 1/Sr, 1/Nd and 1/Pb versus the appropriate isotopic ratios should yield straight-line mixing trends that may, in turn, be used to clarify the composition of the contaminant. For this approach to work successfully, it is necessary: (1) for both the uncontaminated magma and contaminant to have homogeneous compositions; (2) for the abundances of Sr, Nd and Pb to be substantially different in the magma and contaminant; (3) that no significant fractional crystallization took place after the contamination event. Plots of 1/Sr vs 87Sr/86Sri and 1/Nd vs Nd (not shown) for the Brandberg Horingbaai dykes show discernible but very poor trends, with slopes suggesting that the crustal contaminant had lower Sr and Nd contents than the upwelling MORB-like magmas. This is surprising because it implies that the contaminant possessed Sr <170 ppm and Nd <8 ppm—far below average upper-crust values (Taylor & McLennan, 1985) or Damara belt granite and clastic sediment average compositions (McDermott et al., 1989, 1996; McDermott & Hawkesworth, 1990). A possible explanation is that the contaminant was dominated by alkali feldspar, a hypothesis for which we already noted some Sr–Nd isotopic evidence above (Fig. 7). If so, it would be likely to have a Pb content considerably higher than the upper-crust average of 20 ppm (Taylor & McLennan, 1985) and this should generate an excellent correlation between 1/Pb and 206Pb/204Pb in the Horingbaai suite because their MORB-like end-members have only 0·6 ppm Pb. Figure 10e shows that this correlation is indeed good (R2 = 0·87). The Pb isotopic contamination contrasts with REE behaviour, in that it is clearly not linked to the MgO contents of the magmas. This supports the possibility that some of the crustal Pb entering the upwelling basic melts may have done so by selective dissolution of or diffusion from alkali feldspar—within clastic sediments, granites or both—rather than via cotectic crustal melts.
|
MAGMA GENESIS |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD RELATIONS AND SAMPLING...ANALYTICAL TECHNIQUESDYKE MAGMA TYPESHORINGBAAI MAGMA TYPEPETROGRAPHY OF OTHER DYKE...POST-GENESIS, PRE-EMPLACEMENT...HOW DID THE CRUSTAL...MAGMA GENESISROLE OF TRISTAN MANTLE...REFERENCES |
|---|
Horingbaai magma type
Duncan et al. (1990)
put forward a view about the genesis of the Horingbaai magmas that has been generally accepted. They emphasized their MORB-like nature and that they cut the lava pile. They therefore proposed that their genesis was essentially the same as that of MORB; by shallow decompression melting of the asthenospheric (MORB-source) mantle, as it rose in response to thinning and eventual splitting of the Gondwana lithospheric plate. The presence of extremely magnesian cognate olivine macrocrysts in some of the picrites suggests that the genesis of the suite must be more complex. Thompson & Gibson (2000) concentrated on the most Mg-rich olivine (Fo93·3), to establish an end-member parent melt. They estimated its major-element composition by using the olivine–melt equilibria of Ulmer (1989) to calculate mg-number (81·3), then converting this to MgO (24·0 wt %) via a plot of Horingbaai suite mg-number vs MgO, then calculating other oxides from regression analysis of oxide vs MgO diagrams (Table 1). This reconstructed parental liquid (HPK) is a komatiite in the current IUGS nomenclature (Le Bas, 2000). Thompson & Gibson (2000) calculated its theoretical 1 atm liquidus as 1480°C, using the equation of Renner given by Nisbet et al. (1993). Then, assuming that the Horingbaai magmas were probably generated at 132 Ma (see above) beneath the axis of the pre-Atlantic rift—about 200 km west of the present NW Namibian coast (Gladczenko et al., 1997)—they used a P–T diagram for KLB-1 peridotite mantle to estimate that HKP was produced by 44% decompression melting of anhydrous mantle with potential temperature (Tp) 1700°C (McKenzie & Bickle, 1988). The melting was calculated to have started at 6 GPa (180 km depth) and ended at 50 km depth, when the melt entered lithosphere beneath the rift. The accumulated error for Tp is about ±100°C or slightly less. Details of the calculation method and assessment of all the errors have been given by Thompson & Gibson (2000). Gill et al. (1992) and Nisbet et al. (1993) used a similar approach and reached similar conclusions for West Greenland Palaeocene ‘MORB-like’ picrites and Archaean komatiites, respectively.
Forward and inverse modelling
For the Horingbaai suite as a whole we have used two independent modelling approaches, to see whether they can throw consistent light on the melt genesis. The forward modelling approach of Langmuir et al. (1992) calculates major-element compositions for decompression melting paths of plausible mantle peridotites and compares the results with natural magmatic rocks. The inverse modelling approach of McKenzie & O’Nions (1991, 1995) calculates a best-fit mantle peridotite decompression melting scenario to reproduce the REE patterns of magmatic suites. To use the Langmuir et al. approach, it is necessary to recalculate the major elements of Horingbaai suite samples (filtered to >9% MgO) to a constant value of MgO. We have chosen 15% MgO for the picrite-dominated southern Etendeka dykes, as others have done for similar suites (Langmuir et al., 1992; Gibson et al., 2000; Scarrow et al., 2000). Figure 11 shows SiO2 and FeO* (total Fe as FeO) in analyses of Horingbaai, ferropicrite and Nil Desperandum dykes (and ferropicrite lavas) corrected for fractionation to 15% MgO, using regression analysis of oxide–MgO plots for the dykes that contain only olivine macrocrysts (Langmuir et al., 1992). The fields for other lava suites (similarly corrected) are shown for comparison. Melts produced in experimental studies of both KLB-1 and two iron-rich peridotites, HK-66 and PHN1611, are also plotted but these data are not corrected to 15% MgO.
|
The Horingbaai data show a range of 2% Si15 (fractionation-corrected SiO2) and 1% Fe*15. As the three samples we have identified previously as minimally affected by crustal contamination plot at the low-Si end of the range in Fig. 11, it seems prudent to suspect that Horingbaai picrites with Si15 >47% have their SiO2 contents enhanced by a small crustal input. The Horingbaai Fe*15 range is above those of MORB and Iceland picrites. In simple picrite suites, where varying amounts of cumulus olivine occur in a liquid of approximately constant composition, the regression analysis we have used [following Langmuir et al. (1992)] yields erroneously high Fe*15 results because the cumulus olivine (Fo86–87) is, by definition, more Fe rich than the composition that would be in equilibrium with a liquid of the same bulk MgO content. Thus, the resulting Fe*15 values are higher than the true ones. In the Horingbaai picrite suite we do not encounter this problem because, as Fig. 3 shows clearly, the olivine macrocrysts in these rocks are, as often as not, more forsteritic than equilibrium values, potentially leading to underestimation of Fe*15. We therefore consider Fe*15 in our Horingbaai suite to be approximately correct. It is then apparent that Fe*15 in the Horingbaai magmas is too high for them to be isobaric melts of KLB-1 (or similar) mantle at 2–3 GPa pressure; a characteristic they share with Hawaii, West Greenland and Skye basalts–picrites. Scarrow & Cox (1995) and Scarrow et al. (2000) used this point to argue that the Skye lavas were melts of an iron-rich mantle, such as HK-66. Alternatively, the computations of Langmuir et al. (1992) offer another way to generate magmatic suites with enhanced Fe*15: decompression melting of KLB-1 or similar mantle beginning at high pressures, so that deep-source, Fe-rich melts contribute to the final magma.
Figure 12 is a plot of Na15 and Fe*15, showing the results of our anhydrous decompression melting calculations for peridotite KLB-1, using the method of Langmuir et al. (1992) and partition coefficients of Ulmer (1989). Following Langmuir et al., we have assumed that the amount of melting during decompression is 1·2% per kbar, within the regime appropriate to this study (P 3 GPa or more; melt fraction <0·2). We have also assumed a retained melt fraction of 1% in the residue during each decompression step. The Horingbaai data (Mg-basalts and picrites, fractionation corrected to MgO = 15%) can be modelled by dry decompression melting of KLB-1 mantle that began at 4 GPa (120–130 km, depending on assumed crustal thickness; Gibson et al., 2000; Thompson & Gibson, 2000). The melting ceased at 3 GPa (90–100 km depth), after 12% of liquid was produced.
|
, Horingbaai dykes; •, Nil Desperandum dykes. (b) Anhydrous P–T diagram for mantle lherzolite KLB-1 (the only peridotite for which sufficient experimental data are published); data for the closely similar composition MM3 are also incorporated (Thompson & Gibson, 2000
Figure 12b shows that the calculated initial melting pressure can be plotted in the KLB-1 phase diagram (Takahashi et al., 1993; Herzberg & Zhang, 1996; Gibson et al., 2000; Thompson & Gibson, 2000) to show that the potential temperature (Tp) of the upwelling mantle was 1560°C, a typical Phanerozoic steady-state mantle plume temperature (White & McKenzie, 1995; Ribe & Christensen, 1999; Thompson & Gibson, 2000). As we explained above, the extremely magnesian olivine macrocrysts in some of the Horingbaai picrites are evidence for much hotter mantle (Tp 1700°C) that produced komatiitic melts, which failed to rise to the crustal depths currently exposed (Thompson & Gibson, 2000).
An alternative way to model the genesis of the Horingbaai magmas is to use the method of McKenzie & O’Nions (1991): inversion of REE concentrations to give partial melt distributions. This is an independent approach from that of Langmuir et al. (1992) because it focuses on incompatible trace elements. Our inversion calculations for the Horingbaai suite used only the three samples (97SB29, 53 and 75) that have been shown above to be essentially free from crustal contamination. The ratio of ‘primitive’ to ‘depleted’ model mantle end-member (McKenzie & O’Nions, 1995) was calculated using Nd values.
The inversion results (Fig. 13) show that the Horingbaai REE are best modelled by melting mantle in the approximate proportion 1:9, primitive to depleted; fusion beginning at 105 km and completed at 55 km, with melt fraction 0·22. Comparison of the calculated melt distribution with that produced by theoretical isentropic decompression of mantle at various potential temperatures (Fig. 13c) shows that Tp between 1470 and 1500°C is appropriate for the genesis of the Horingbaai suite. This is of course a minimum potential temperature, calculated from the REE of the dykes. The note in Fig. 13c is a reminder that the cognate olivine macrocrysts in these melts were precipitated from a suite of much hotter liquids. Considering that their approaches are so different, it is encouraging to see that both modelling methods (forward using major elements; inverse using REE) indicate genesis of the Horingbaai magmas by decompression melting within a mantle plume (Tp 1470–1560°C), beginning at 105–125 km depth, beneath a ‘lid’ 55–95 km thick and involving 10–20% of fusion. The value calculated by inverse modelling (55 km; Fig. 13c) for the thickness of lithospheric ‘lid’ above the decompressing Horingbaai source-mantle agrees very well with the ‘lid’ value of 50 km postulated by Thompson & Gibson (2000), from geophysical evidence, to occur above the super-hot mantle source of the hypothetical komatiite HPK that precipitated Fo93·3 macrocrysts.
|
Ferropicrite magma type
Gibson et al. (2000) considered the origin of the ferropicrites in detail and showed that a mantle source more Fe rich than KLB-1 was necessary to achieve plausible results. Their preferred model, using the Langmuir et al. (1992) approach, required decompression melting of Fe-rich peridotite (PHN1611 or similar) between about 4·5 and 3·5 GPa. The calculated average degree of partial melting was 10% and the top of the melting column was 115 km.
Nil Desperandum magma type
Figure 11 shows clearly how the major-element compositions of the ND dykes are distinct from those of both the Horingbaai and ferropicrite suites. Successful inverse modelling of REE in this magma type is difficult because only a few MgO-rich dykes have yet been found (Table 1) and all of them appear to be affected to some extent by crustal contamination (see above). With this reservation in mind, the ND data in Fig. 12 (forward model) suggest that these magmas were produced along essentially the same dry decompression path as the Horingbaai suite but with 9% partial melting, beginning at 4 GPa (120–130 km) and ceasing at 3·25 GPa (just over 100 km). Until uncontaminated ND dykes are found, it seems sensible to apply Occam’s Razor and construct the inverse model using the same mantle source as for the Horingbaai suite (Fig. 13b). The calculated decompression melting path is the same as for the Horingbaai (Fig. 13c) but with a 70 km lithospheric ‘lid’ and melt fraction of 13%. The very small amounts of melting below 105 km, calculated by both Horingbaai and ND inversions, may possibly relate to incipient ‘wet’ melting deep in the upwelling mantle. Although the forward and inverse modelling of the Horingbaai and ND magmas are far from identical, there is agreement from these diverse approaches that the lithospheric ‘lid’ above the decompressing ND mantle sources was a few kilometres thicker (at 132 Ma) than that above the Horingbaai sources. Figure 1, inset, shows that this could be because the ND sources rose beneath the SW corner of the Congo craton.
|
ROLE OF TRISTAN MANTLE PLUME AT 133–131 Ma |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD RELATIONS AND SAMPLING...ANALYTICAL TECHNIQUESDYKE MAGMA TYPESHORINGBAAI MAGMA TYPEPETROGRAPHY OF OTHER DYKE...POST-GENESIS, PRE-EMPLACEMENT...HOW DID THE CRUSTAL...MAGMA GENESISROLE OF TRISTAN MANTLE...REFERENCES |
|---|
The structural geology and stratigraphy of both sides of the Atlantic at Paraná–Etendeka latitudes shows prolonged and widespread extension during the Phanerozoic, affecting both former Brasiliano and Pan-African mobile belts and, in Brazil, parts of adjacent cratons (Zalán et al., 1991
; Clemson et al., 1997; Gladczenko et al., 1997; Peate, 1997; Jerram et al., 1999). There is a difference in perspective between those studying the Paraná basin in southern Brazil and surrounding areas, who see Gondwana intracontinental basin formation proceeding stepwise since the Ordovician (Zalán et al., 1991), and those who focus on the South Atlantic margins of Brazil–Argentina and Angola–Namibia–South Africa, who see the Permian–Cretaceous part of this story as a prolonged pre-oceanic rifting event (e.g. Clemson et al., 1997; Gladczenko et al., 1997; Jerram et al., 1999).
In NW Namibia thin terrigenous sediments underlie the southern Etendeka lavas. These accumulated during a period of pre-magmatism extension variously estimated at from 10 my (Jerram et al., 1999) to 25 my (Gladczenko et al., 1997). Before that there was tectonic quiescence for 100 my (Jerram et al., 1999). This is long enough for thinned continental lithosphere to have re-thickened by conductive cooling to the oceanic maximum of 125 km (McKenzie & Bickle, 1988). We agree with Ewart et al. (1998a), Jerram et al. (1999) and Marsh et al. (2001) that the 133 Ma basal ferropicrites are so small in volume and geographically restricted in occurrence that they probably did not travel far laterally between their genesis and eruption. Thus, our first ‘window’ into the impact of the Tristan mantle plume shows an initial plume head rising at 133 Ma beneath NW Namibia where lithospheric thickness was 100 km, or slightly more, along the former Pan-African mobile belts and, of course, thicker beneath the SW margin of the Congo craton. The plume was still too deep for pervasive decompression melting to begin and only Fe-rich peridotite streaks within it contributed significantly to the earliest magmas (Gibson et al., 2000). The next ‘window’ into the Tristan plume is given by the Horingbaai magmas. Both they and the localized Nil Desperandum suite cut 132 Ma lavas. Coastal dykes may have been emplaced over a period of several million years (Renne et al., 1998) but, as we noted above, the NW Namibia coast-parallel dykes are geochemically complex and from more than one magma type. Dykes throughout southern Etendeka that can be unambiguously assigned to the Horingbaai suite (Fig. 2) are distinctive basalt–picrite, predominantly characterized by extremely forsteritic olivine macrocrysts (Fo91–93·3) and cut by the 131 Ma Brandberg complex (Schmitt et al., 2000). Although the Horingbaai liquids that reached the upper crust were basaltic, with MgO <10% (now aphyric dolerites and groundmasses), their cognate olivine macrocrysts show that the MgO contents of these melts originally ranged up to 24% (Thompson & Gibson, 2000).
Our study reveals a fundamental difference between the Horingbaai and the localized ferropicrite and ND suites. The dominant role of Horingbaai magmas in dense dyke swarms spread over at least 10 000 km2 (Fig. 1) suggests that this was a substantial igneous episode that may have fed a considerable lava pile (now eroded). The magmas of dense, laterally extensive dyke swarms, such as the one south of Brandberg (Fig. 1), have been shown in many examples worldwide to have migrated laterally within the crust for distances of hundreds of kilometres or more (Ernst et al., 1995). Thus, the restriction that the melts were generated within the mantle immediately beneath the dykes where they were sampled may not apply to the Horingbaai magmas.
Thompson & Gibson (2000) used this reasoning to postulate that the komatiite (Horingbaai Parental Komatiite; HPK) that they calculated as the melt that precipitated the Fo93+ macrocrysts found in some Horingbaai picrites was generated 200 km west of the present NW Namibian coast, beneath the axis of the pre-Atlantic rift at 132–131 Ma, where lithospheric thinning of 50% had occurred (Gladczenko et al., 1997). Thompson & Gibson (2000, fig. 2) showed that, if HPK had separated from its mantle residuum at 50 km depth, it would have been in equilibrium with Fo93.
Although the melting parameters for the genesis of the Horingbaai magmas calculated by forward and inverse modelling are not sufficiently consistent to discuss further, the variable REE (Table 1) of the three Horingbaai samples considered to be least affected by crustal contamination (see above) requires comment. Their diverse REE pattern slopes cause the poor fit between observed and modelled La abundances in Fig. 13a. The most widely used hypothesis to explain such variability in the ratios of more incompatible to less incompatible elements (e.g. La/Lu) in suites of uncontaminated mafic–ultramafic magmas is that individual magma batches separated at varying depths from a single upwelling mantle column (e.g. Elliott et al., 1991; Eggins, 1992; Gurenko & Chaussidon, 1995; Iwamori et al., 1995; Arndt et al., 1997; Scarrow et al., 2000). This may also apply to the Horingbaai-HPK suite.
Figure 14 shows how this study and those of Gibson et al. (2000) and Thompson & Gibson (2000) clarify and constrain concepts about southern Etendeka Early Cretaceous magmatism. The impression of rapid lithospheric thinning with time may be more apparent than real because, as explained above, we suspect that the site of magma genesis and escape may have moved 200 km westward during the 133–131 Ma interval, from beneath southern Etendeka to the axis of the pre-Atlantic offshore rift (Gladczenko et al., 1997). Additional local lithospheric thinning may well have occurred beneath major Southern Etendeka plutonic centres, such as Messum (Ewart et al., 1998a, 1998b), and this may explain why radiometric dates from plutons at both Messum and Okenyenya (Fig. 1, inset) extend for several million years younger than 130 Ma (Milner et al., 1993; Renne et al., 1996; Stewart et al., 1996).
|
The peak value of Tp (1700°C) required to generate the HPK with 24% MgO in the initial Tristan mantle plume head is high but comparable values of Tp can be calculated for two other Phanerozoic mantle plumes, Iceland and the plume responsible for the Gorgona suite, that also produced magmas bearing Fo93+ olivine macrocrysts (Francis, 1985; Gill et al., 1992; Kerr et al., 1996; Arndt et al., 1997), although estimation of magmatic Fe2O3/FeO greatly affects such calculations (e.g. Larsen & Pedersen, 2000). Thus Tristan appears to have been amongst the hottest Phanerozoic starting mantle plumes, when compared with the melting conditions for West Greenland and Gorgona picrites determined by Herzberg & O’Hara (1998). Nevertheless, we can be sure that the proportion of the Tristan plume head that was hot enough to produce melts like HPK was small. This is because an entire plume head with Tp 1700°C, decompressing beneath a rapidly extending continental margin, would generate an enormous thickness of melt; the precise amount would depend on the plate thickness (McKenzie & Bickle, 1988). The maximum melt thickness generated in NW Namibia at 132–131 Ma is best estimated from seismic studies summarized by Gladczenko et al. (1997). They showed that the dipping reflector sequences along this segment of the SW African coast are 100 km wide and reach a maximum thickness of 7 km near the landfall of the Walvis Ridge (Fig. 1, inset). From these data they calculated that the melting products of the Tristan plume were comparable in volume (2 x 106 km3) with those of the Iceland–North Atlantic plume. Thompson & Gibson (2000) used the reasoning outlined above to argue that the upwelling mechanism of the initial Tristan plume may have been non-Newtonian, as Larsen et al. (1999) have proposed for the initial Iceland plume.
Finally, we think that it is inappropriate for us to speculate in detail here about the genesis of the southern Etendeka subaerial flood basalts, despite their dominance in NW Namibia (Fig. 1) and throughout the Paraná–Etendeka Large Igneous Province. The remarkably few LTZ-L dykes that were found by our sampling add little to what is already known about this magma type and its sub-types (e.g. Peate, 1997; Ewart et al., 1998a; Marsh et al., 2001). Nevertheless, Fig. 14 gives a framework of factors that can be used in future studies to evaluate the different models currently proposed for CFB genesis. These are:
- that there was minimal magma generation within the initial Tristan plume head during the CFB episode; heat transfer upwards was only by conduction and, during this interval of 1–2 my, the SCLM underwent wholesale hydrous melting (Peate, 1997; Hawkesworth et al., 1999, 2000). Presumably other subsequent processes must be invoked to account for the lack of hydrous minerals, even interstitially, in both basic and acid lavas (e.g. Ewart et al., 1998a, 1998b).
- Upwelling mafic–ultramafic melts, generated during the pre-climactic and climactic stages of Tristan plume impact, carried heat upwards by advection and caused substantial melting of both SCLM and crust through which they passed and within which they fractionated (Gibson et al., 1995b; Ewart et al., 1998a, 1998b), leading to complex hybrid-source CFB lavas. We see no reason arising from the present study to change our preference for the second alternative.
|
ACKNOWLEDGEMENTS |
|---|
We are grateful to Ercan Aldanmaz, Mike Bickle, Affonso Brod, Vicky Hards, Ron Hardy, Dougal Jerram, Wendy McClean, Glen Milne, Simon Milner, Anna-Karren Nguno, Chris Ottley, Graham Pearson, Stephen Reed and Mathew Tucker for invaluable assistance and discussions. The perceptive comments of Chris Harris, Tony Ewart, Andrew Kerr, Linda Kirstein, Graham Pearson and Marjorie Wilson substantially improved the manuscript. The Royal Society, and Durham and Cambridge universities funded the research in part.
|
FOOTNOTES |
|---|
Extended dataset can be found at http://www.petrology.oupjournals.org
*Corresponding author. E-mail: r.n.thompson{at}durham.ac.uk
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONFIELD RELATIONS AND SAMPLING...ANALYTICAL TECHNIQUESDYKE MAGMA TYPESHORINGBAAI MAGMA TYPEPETROGRAPHY OF OTHER DYKE...POST-GENESIS, PRE-EMPLACEMENT...HOW DID THE CRUSTAL...MAGMA GENESISROLE OF TRISTAN MANTLE...REFERENCES |
|---|
Arndt, N. T., Kerr, A. C. & Tarney, J. (1997). Dynamic melting in plume heads: the formation of Gorgona komatiites and basalts. Earth and Planetary Science Letters 146, 289–301.
Barnes, S. J. (1998). Chromite in komatiites, I. Magmatic controls on crystallization and composition. Journal of Petrology 39, 1689–1720.
Carlson, R. W., Esperança, S. & Svisero, D. P. (1996). Chemical and Os isotopic study of Cretaceous potassic rocks from Southern Brazil. Contributions to Mineralogy and Petrology 125, 393–405.
Clemson, J., Cartwright, J. & Booth, J. (1997). Structural segmentation and the influence of basement structure on the Namibian passive margin. Journal of the Geological Society, London 154, 477–482.
Cliff, R. A., Baker, P. E. & Mateer, N. J. (1991). Geochemistry of Inaccessible Island volcanics. Chemical Geology 92, 251–260.
Dickin, A. P. (1981). Isotope geochemistry of Tertiary igneous rocks from the Isle of Skye, N. W. Scotland. Journal of Petrology 22, 155–189.
Dickin, A. P. (1997). Radiogenic Isotope Geology. Cambridge: Cambridge University Press.
Duncan, A. R., Armstrong, R. A., Erlank, A. J., Marsh, J. S. & Watkins, R. T. (1990). MORB-related dolerites associated with the final phases of Karoo flood basalt volcanism in Southern Africa. In: Parker, A. J., Rickwood, P. C. & Hunter, D. H. (eds) Mafic Dykes and Emplacement Mechanisms. Rotterdam: Balkema, pp. 119–129.
Duncan, A. R., Milner, S. C. & Jerram, D. A. (1998). Field Excursion C4: Southern Etendeka Volcanism, Northwestern Namibia. IAVCEI International Volcanological Congress. Cape Town: IAVCEI.
Eggins, S. M. (1992). Petrogenesis of Hawaiian tholeiites: 2, aspects of dynamic melt segregation. Contributions to Mineralogy and Petrology 110, 398–410.
Elliott, T. R., Hawkesworth, C. J. & Gronvold, K. (1991). Dynamic melting of the Iceland plume. Nature 351, 201–206.
Erlank, A. J., Marsh, J. S., Duncan, A. R., Miller, R. McG., Hawkesworth, C. J., Betton, P. J. & Rex, D. C. (1984). Geochemistry and petrogenesis of the Etendeka volcanic rocks from SWA/Namibia. Geological Society of South Africa, Special Publication 13, 195–245.
Ernst, R. E., Head, J. W., Parfitt, E., Grosfils, E. & Wilson, L. (1995). Giant radiating dyke swarms on Earth and Venus. Earth-Science Reviews 39, 1–58.
Ewart, A., Milner, S. C., Armstrong, R. A. & Duncan, A. R. (1998a). Etendeka volcanism of the Goboboseb Mountains and Messum igneous complex, Namibia. Part I: Geochemical evidence of Early Cretaceous Tristan plume melts and the role of crustal contamination in the Paraná–Etendeka CFB. Journal of Petrology 39, 191–225.
Ewart, A., Milner, S. C., Armstrong, R. A. & Duncan, A. R. (1998b). Etendeka volcanism of the Goboboseb Mountains and Messum igneous complex, Namibia. Part II: Voluminous quartz latite volcanism of the Awahab magma system. Journal of Petrology 39, 227–253.
Francis, D. (1985). The Baffin Bay lavas and the value of picrites as analogues of primary magmas. Contributions to Mineralogy and Petrology 89, 144–154.
Garcia, M. O. (1996). Petrography and olivine and glass chemistry of lavas from the Hawaii Scientific Drilling Project. Journal of Geophysical Research 101, 11701–11713.
Garcia, M. O., Hulsebosch, T. P. & Rhodes, J. M. (1995). Olivine-rich basalts from the southwest rift zone of Mauna Loa volcano: implications for magmatic processes and geochemical evolution. In: Rhodes, J. M. & Lockwood, J. P. (eds) Mauna Loa Revealed: Structure, Composition, History and Hazards. Washington, DC: American Geophysical Union, pp. 219–239.
Geological Survey of Namibia (1997). Sheet 2114—Omaruru, 1:250 000 Geological Series. Windhoek: Geological Survey of Namibia.
Gibson, S. A., Thompson, R. N., Dickin, A. P. & Leonardos, O. H. (1995a). High-Ti and low-Ti mafic potassic magmas: key to plume–lithosphere interactions and continental flood-basalt genesis. Earth and Planetary Science Letters 136, 149–165.
Gibson, S. A., Thompson, R. N., Leonardos, O. H., Dickin, A. P. & Mitchell, J. G. (1995b). The Late Cretaceous impact of the Trindade mantle plume: evidence from large-volume, mafic, potassic magmatism in SE Brazil. Journal of Petrology 36, 189–229.
Gibson, S. A., Thompson, R. N., Dickin, A. P., Mitchell, J. G. & Milner, S. C. (1997). Temporal variation in magma sources related to the impact of the Tristan mantle plume. Journal of Conference Abstracts 2, 32.
Gibson, S. A., Thompson, R. N., Leonardos, O. H., Dickin, A. P. & Mitchell, J. G. (1999). The limited extent of plume–lithosphere interactions during continental flood basalt genesis: geochemical evidence from Cretaceous magmatism in southern Brazil. Contributions to Mineralogy and Petrology 137, 147–169.
Gibson, S. A., Thompson, R. N. & Dickin, A. P. (2000). Ferropicrites: geochemical evidence for Fe-rich streaks in upwelling mantle plumes. Earth and Planetary Science Letters 174, 355–374.
Gill, R. C. O., Pedersen, A. K. & Larsen, J. G. (1992). Tertiary picrites in West Greenland: melting at the periphery of a plume? In: Storey, B. C., Alabaster, T. & Pankhurst, R. J. (eds) Magmatism and the Causes of Continental Break-up. Geological Society, London, Special Publications 68, 335–348.
Gladczenko, T. P., Hinz, K., Edholm, O., Meyer, H., Neben, S. & Skogseid, J. (1997). South Atlantic volcanic margins. Journal of the Geological Society, London 154, 465–470.
Gradstein, F. M., Agterberg, F. P., Ogg, J. G., Hardenbol, J., van Veen, P., Thierry, J. & Huang, Z. (1994). A Mesozoic timescale. Journal of Geophysical Research 99, 24051–24074.[Web of Science]
Grove, T. L., Kinzler, R. J. & Bryan, W. B. (1992). Fractionation of mid-ocean ridge basalt (MORB). In: Morgan, J. P., Blackman, D. K. & Sinton, J. M. (eds) Mantle Flow and Melt Generation at Mid-Ocean Ridges. Washington, DC: American Geophysical Union, pp. 281–310.
Gurenko, A. A. & Chaussidon, M. (1995). Enriched and depleted primitive melts included in olivine from Icelandic tholeiites: origin by continuous melting of a single mantle column. Geochimica et Cosmochimica Acta 59, 2905–2917.[Web of Science]
Hart, S. R. (1984). A large-scale isotope anomaly in the Southern Hemisphere mantle. Nature 309, 753–757.
Hawkesworth, C. J., Kelley, S., Turner, S., le Roex, A. & Storey, B. (1999). Mantle processes during Gondwana break-up and dispersal. Journal of African Earth Sciences 28, 239–261.
Hawkesworth, C. J., Gallagher, K., Kirstein, L., Mantovani, M. S. M., Peate, D. W. & Turner, S. P. (2000). Tectonic controls on magmatism associated with continental break-up: an example from the Paraná–Etendeka province. Earth and Planetary Science Letters 179, 335–349.
Herzberg, C. & O’Hara, M. J. (1998). Phase equilibria constraints on the origin of basalts, picrites and komatiites. Earth-Science Reviews 44, 39–79.
Herzberg, C. & Zhang, J. (1996). Melting experiments on anhydrous peridotite KLB-1: compositions of magmas in the upper mantle and transition zone. Journal of Geophysical Research 101, 8271–8295.
Hirose, K. & Kushiro, I. (1993). Partial melting of dry peridotites at high pressures: determination of compositions of melts segregated from peridotite using aggregates of diamond. Earth and Planetary Science Letters 114, 477–489.[Web of Science]
Hofmann, A. W., Jochum, K. P., Seufert, M. & White, W. M. (1986). Nb and Pb in oceanic basalts: new constraints on mantle evolution. Earth and Planetary Science Letters 79, 33–45.[Web of Science]
Hunter, D. R. & Reid, D. L. (1987). Mafic dyke swarms in southern Africa. In: Halls, H. C. & Fahrig, W. F. (eds) Mafic Dyke Swarms. Ottawa, Ont.: Geological Society of Canada, pp. 445–456.
Huppert, H. E. & Sparks, R. S. J. (1985). Cooling and contamination of mafic and ultramafic magmas during ascent through continental crust. Earth and Planetary Science Letters 74, 371–386.[Web of Science]
Iwamori, H., McKenzie, D. & Takahashi, E. (1995). Melt generation by isentropic upwelling. Earth and Planetary Science Letters 134, 253–266.
Jerram, D., Mountney, N., Holzförster, F. & Stollhofen, H. (1999). Internal stratigraphic relationships in the Etendeka Group in the Huab Basin, NW Namibia: understanding the onset of flood volcanism. Journal of Geodynamics 28, 393–418.
Kerr, A. C., Kempton, P. D. & Thompson, R. N. (1995). Crustal assimilation during turbulent magma ascent (ATA); new isotopic evidence from the Mull Tertiary lava succession, NW Scotland. Contributions to Mineralogy and Petrology 119, 142–154.[Web of Science]
Kerr, A. C., Mariner, G. F., Arndt, N. T., Tarney, J., Nivia, A., Saunders, A. D. & Duncan, R. A. (1996). The petrogenesis of Gorgona komatiites, picrites and basalts: new field, petrographic and geochemical constraints. Lithos 37, 245–260.
Klemme, S. & O’Neill, H. St. C. (2000). The near-solidus transition from garnet lherzolite to spinel lherzolite. Contributions to Mineralogy and Petrology 138, 237–248.
Kushiro, I. (1996). Partial melting of a fertile mantle peridotite at high pressures: an experimental study using aggregates of diamond. In: Basu, A. & Hart, S. R. (eds) Earth Processes: Reading the Isotopic Code. Washington, DC: American Geophysical Union, pp. 109–122.
Langmuir, C. H., Klein, E. M. & Plank, T. (1992). Petrological systematics of mid-ocean ridge basalts: constraints on melt generation beneath ocean ridges. In: Morgan, J. P., Blackman, D. K. & Sinton, J. M. (eds) Mantle Flow and Melt Generation at Mid-Ocean Ridges. Washington, DC: American Geophysical Union, pp. 183–280.
Larsen, L. M. & Pedersen, A. K. (2000). Processes in high-Mg, high-T magmas: evidence from olivine, chromite and glass in Palaeogene picrites from West Greenland. Journal of Petrology 41, 1071–1098.
Larsen, T. B., Yuen, D. A. & Storey, M. (1999). Ultrafast mantle plumes and implications for flood basalt volcanism in the North Atlantic region. Tectonophysics 311, 31–43.
Lassiter, J. C. & DePaolo, D. J. (1997). Plume/lithosphere interaction in the generation of continental and oceanic flood basalts: chemical and isotopic constraints. In: Mahoney, J. J. & Coffin, M. F. (eds) Large Igneous Provinces: Continental, Oceanic and Planetary Flood Volcanism. Washington, DC: American Geophysical Union, pp. 335–355.
Le Bas, M. J. (2000). IUGS reclassification of the high-Mg and picritic volcanic rocks. Journal of Petrology 41, 1467–1470.
le Roex, A. P. & Lanyon, R. (1998). Isotope and trace element geochemistry of Cretaceous Damaraland lamprophyres and carbonatites, northwestern Namibia: evidence for plume–lithosphere interactions. Journal of Petrology 39, 1117–1146.
le Roex, A. P., Cliff, R. A. & Adair, B. J. I. (1990). Tristan da Cunha, South Atlantic: geochemistry and petrogenesis of a basanite–phonolite lava series. Journal of Petrology 31, 779–812.
Lightfoot, P. C., Hawkesworth, C. J., Olshefsky, K., Green, T., Doherty, W. & Keays, R. R. (1997). Geochemistry of Tertiary tholeiites and picrites from Qeqertarssuaq (Disko Island) and Nuusuaq, West Greenland with implications for the mineral potential of comagmatic intrusions. Contributions to Mineralogy and Petrology 128, 139–163.
Lord, J., Oliver, G. J. H. & Soulsby, J. A. (1996). Landsat MSS imagery of a Lower Cretaceous regional dyke swarm, Damaraland: a precursor to the splitting of western Gondwana. International Journal of Remote Sensing 17, 2945–2954.[Web of Science]
Mahoney, J. J. & Coffin, M. F. (1997). Large igneous provinces: continental, oceanic and planetary flood volcanism. Geophysical Monograph, American Geophysical Union 100, 438 pp.
Marsh, J. S., Erlank, A. J. & Duncan, A. R. (1991). Preliminary geochemical data for dolerite dykes and sills of the southern part of the Etendeka igneous province. Communications of the Geological Survey of Namibia 7, 71–73.
Marsh, J. S., Harris, C. & Kirstein, L. (1997). Etendeka dykes, Namibia—relationship to Paraná–Etendeka lavas with implications. Abstracts: IAVCEI International Volcanological Congress, Puerto Vallarta, Mexico. Puerta Vallarta: IAVCEI, 161.
Marsh, J. S., Ewart, A., Milner, S. C., Duncan, A. R. & Miller, R. McG. (2001). The Etendeka igneous province: magma types and their stratigraphic distribution with implications for the evolution of the Paraná–Etendeka flood basalt province. Bulletin of Volcanology 62, 464–486.
McDermott, F. & Hawkesworth, C. J. (1990). Intracrustal recycling and upper-crust evolution: a case study from the Pan-African Damara mobile belt, central Namibia. Chemical Geology 83, 263–280.
McDermott, F., Harris, N. B. W. & Hawkesworth, C. J. (1989). Crustal reworking in southern Africa: constraints from Sr–Nd isotope studies in Archaean to Pan-African terrains. Tectonophysics 161, 257–270.
McDermott, F., Harris, N. B. W. & Hawkesworth, C. J. (1996). Geochemical constraints on crustal anatexis: a case study from the Pan-African Damara granitoids of Namibia. Contributions to Mineralogy and Petrology 123, 406–423.
McKenzie, D. & Bickle, M. J. (1988). The volume and composition of melt generated by extension of the lithosphere. Journal of Petrology 29, 625–679.
McKenzie, D. & O’Nions, R. K. (1991). Partial melt distributions from inversion of rare earth element concentrations. Journal of Petrology 32, 1021–1091.
McKenzie, D. & O’Nions, R. K. (1995). The source regions of ocean island basalts. Journal of Petrology 36, 133–159.
Miller, R. McG. (1983). Evolution of the Damara orogen of South West Africa/Namibia. Geological Society of South Africa, Special Publication 11, 515.
Milner, S. C. & le Roex, A. P. (1996). Isotope characteristics of the Okenyenya igneous complex, northwestern Namibia: constraints on the composition of the early Tristan plume and the origin of the EM1 mantle component. Earth and Planetary Science Letters 141, 277–291.
Milner, S. C., le Roex, A. P. & Watkins, R.T. (1993). Rb–Sr age determinations of rocks from the Okenyenya igneous complex, northwestern Namibia. Geological Magazine 130, 335–343.[Abstract]
Milner, S. C., le Roex, A. P. & O’Connor, J. M. (1995). Age of Mesozoic rocks in northwestern Namibia, and their relationship to continental breakup. Journal of the Geological Society, London 152, 97–104.
Nickel, E. H. (1992). Solid solutions in mineral nomenclature. Canadian Mineralogist 30, 231–234.
Nisbet, E. G., Cheadle, M. J., Arndt, N. T. & Bickle, M. J. (1993). Constraining the potential temperature of the Archaean mantle: a review of the evidence from komatiites. Lithos 30, 291–307.[Web of Science]
Norman, M. D. & Garcia, M. O. (1999). Primitive magma compositions and source characteristics of the Hawaiian plume: constraints from tholeiitic picrites. Earth and Planetary Science Letters 168, 27–44.
Peate, D. W. (1997). The Paraná–Etendeka province. In: Mahoney, J. J. & Coffin, M. F. (eds) Large Igneous Provinces: Continental, Oceanic and Planetary Flood Volcanism. Washington, DC: American Geophysical Union, pp. 217–245.
Reid, D. L. & Rex, D. C. (1994). Cretaceous dykes associated with the opening of the South Atlantic: the Melhberg dyke, northern Richtersveld. South African Journal of Geology 97, 135–145.[Abstract]
Renne, P. R., Glen, J. M., Milner, S. C. & Duncan, A. R. (1996). Age of Etendeka flood volcanism and associated intrusions in southwestern Africa. Geology 24, 659–662.
Renne, P. R., Ludwig, K. R., Ernesto, M., Milner, S. C., Glen, J. M. & Fanning, C. M. (1998). Geochronology of the Paraná–Etendeka magmatic province. Abstracts: IAVCEI International Volcanological Congress, Cape Town, South Africa. Cape Town: IAVCEI, 49.
Ribe, N. M. & Christensen, U. R. (1999). The dynamical origin of Hawaiian volcanism. Earth and Planetary Science Letters 171, 517–531.
Richardson, S. H., Erlank, A. J., Reid, D. L. & Duncan, A. R. (1984). Major and trace elements and Nd and Sr isotope geochemistry of basalts from the Deep Sea Drilling Project leg 74 Walvis Ridge transect. In: Blakeslee, J. H. & Lee, M. (eds) Initial Reports of the Deep Sea Drilling Project, 74. Washington, DC: US Government Printing Office, pp. 739–754.
Rudnick, R. L. & Fountain, D. M. (1995). Nature and composition of the continental crust: a lower crustal perspective. Reviews of Geophysics 33, 267–309.[Web of Science]
Rudnick, R. L. & Goldstein, S. L. (1990). The Pb isotopic composition of lower crustal xenoliths and the evolution of lower crustal Pb. Earth and Planetary Science Letters 98, 192–207.[Web of Science]
Rudnick, R. L. & Presper, T. (1990). Geochemistry of intermediate- to high-pressure granulites. In: Vielzeuf, D. & Vidal, Ph. (eds) Granulites and Crustal Evolution. Dordrecht: Kluwer Academic, pp. 523–550.
Sack, R. O. & Ghiorso, M. S. (1991). Chromian spinels as petrogenetic indicators: thermodynamic and petrological applications. American Mineralogist 76, 827–847.[Abstract]
Sack, R. O., Walker, D. & Carmichael, I. S. E. (1987). Experimental petrology of alkalic lavas: constraints on cotectics of multiple saturation in natural basic liquids. Contributions to Mineralogy and Petrology 96, 1–23.[Web of Science]
Saunders, A. D., Fitton, J. G., Kerr, A. C., Norry, M. J. & Kent, R. W. (1997). The North Atlantic igneous province. In: Mahoney, J. J. & Coffin, M. F. (eds) Large Igneous Provinces: Continental, Oceanic and Planetary Flood Volcanism. Washington, DC: American Geophysical Union, pp. 45–93.
Scarrow, J. H. & Cox, K. G. (1995). Basalts generated by decompressive adiabatic melting of a mantle plume: a case study from the Isle of Skye, NW Scotland. Journal of Petrology 36, 3–22.
Scarrow, J. H., Curran, J. M. & Kerr, A. C. (2000). Major element records of variable plume involvement in the North Atlantic province Tertiary flood basalts. Journal of Petrology 41, 1155–1176.
Schmitt, A. K., Emmermann, R., Trumbull, R. B., Bühn, B. & Henjes-Kunst, F. (2000). Petrogenesis and 40Ar/39Ar geochronology of the Brandberg complex, Namibia: evidence for a major mantle contribution in metaluminous and peralkaline granites. Journal of Petrology 41, 1207–1239.
Scowen, P. A. H., Roeder, P. L. & Helz, R. T. (1991). Reequilibration of chromite within Kilauea Iki lava lake, Hawaii. Contributions to Mineralogy and Petrology 107, 8–20.
Stewart, K., Turner, S., Kelley, S., Hawkesworth, C., Kirstein, L. & Mantovani, M. (1996). 3-D, 40Ar/39Ar geochronology in the Paraná continental flood basalt province. Earth and Planetary Science Letters 143, 95–109.[Web of Science]
Sun, S. S. & McDonough, W. F. (1989). Chemical and isotope systematics of oceanic basalts: implications for mantle compositions and processes. In: Saunders, A. D. & Norry, M. J. (eds) Magmatism in the Ocean Basins. Oxford: Blackwell, pp. 313–345.
Takahashi, E., Shimazaki, T., Tsuzaki, Y. & Yoshida, H. (1993). Melting study of a peridotite KLB-1 to 6·5 GPa and the origin of basaltic magmas. Philosophical Transactions of the Royal Society of London, Series A 342, 105–120.
Taylor, S. R. & McLennan, S. M. (1985). The Continental Crust: its Composition and Evolution. Cambridge, MA: Blackwell Scientific.
Thompson, R. N. (1974). Primary basalts and magma genesis. I. Skye, north-west Scotland. Contributions to Mineralogy and Petrology 45, 317–341.
Thompson, R. N. (1981). Thermal aspects of the origin of Hebridean Tertiary acid magmas. I. An experimental study of partial fusion of Lewisian gneisses and Torridonian sediments. Mineralogical Magazine 44, 161–170.
Thompson, R. N. (1982). Magmatism of the British Tertiary Volcanic Province. Scottish Journal of Geology 18, 49–107.[Web of Science]
Thompson, R. N. & Gibson, S. A. (2000). Transient high temperatures in mantle plume heads inferred from magnesian olivines in Phanerozoic picrites. Nature 407, 502–506.
Thompson, R. N., Morrison, M. A., Hendry, G. L. & Parry, S. J. (1984). An assessment of the relative roles of crust and mantle in magma genesis: an elemental approach. Philosophical Transactions of the Royal Society of London, Series A 310, 549–590.
Thompson, R. N., Gibson, S. A., Mitchell, J. G., Dickin, A. P., Leonardos, O. H., Brod, J. A. & Greenwood, J. C. (1998). Migrating Cretaceous–Eocene magmatism in the Serra do Mar alkaline province, SE Brazil: melts from the deflected Trindade mantle plume? Journal of Petrology 39, 1493–1526.
Ulmer, P. (1989). The dependence of the Fe2+–Mg cation-partitioning between olivine and basaltic liquid on pressure, temperature and composition. Contributions to Mineralogy and Petrology 101, 261–273.
White, R. & McKenzie, D. (1995). Mantle plumes and flood basalts. Journal of Geophysical Research 100, 17543–17585.
Wood, B. J. (1991). Oxygen barometry of spinel peridotites. In: Lindsley, D. H. (ed.) Oxide Minerals: Petrologic and Magnetic Significance. Mineralogical Society of America, Reviews in Mineralogy 25, 417–431.
Zalán, P. V., Wolfe, S., Astolfi, M. A. M., Viera, I. S., Conelcaõ, J. C. J., Appi, V. T., Neto, E. V. S., Cerqueira, J. R. & Marques, A. (1991). The Paraná Basin, Brazil. In: Leighton, M. W., Kolata, D. R., Oltz, D. F. A. & Eidel, J. J. (eds) Interior Cratonic Basins. American Association of Petroleum Geologists, Memoir 51, 681–708.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. B. Trumbull, D. L. Reid, C. de Beer, D. van Acken, and R. L. Romer Magmatism and continental breakup at the west margin of southern Africa: A geochemical comparison of dolerite dikes from northwestern Namibia and the Western Cape South African Journal of Geology, September 1, 2007; 110(2-3): 477 - 502. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Rehfeldt, D. E. Jacob, R. W. Carlson, and S. F. Foley Fe-rich Dunite Xenoliths from South African Kimberlites: Cumulates from Karoo Flood Basalts J. Petrology, July 1, 2007; 48(7): 1387 - 1409. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. N. Thompson, A. J. V. Riches, P. M. Antoshechkina, D. G. Pearson, G. M. Nowell, C. J. Ottley, A. P. Dickin, V. L. Hards, A.-K. Nguno, and V. Niku-Paavola Origin of CFB Magmatism: Multi-tiered Intracrustal Picrite-Rhyolite Magmatic Plumbing at Spitzkoppe, Western Namibia, during Early Cretaceous Etendeka Magmatism J. Petrology, June 1, 2007; 48(6): 1119 - 1154. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. TUFF, E. TAKAHASHI, and S. A. GIBSON Experimental Constraints on the Role of Garnet Pyroxenite in the Genesis of High-Fe Mantle Plume Derived Melts J. Petrology, October 1, 2005; 46(10): 2023 - 2058. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. N. THOMPSON, C. J. OTTLEY, P. M. SMITH, D. G. PEARSON, A. P. DICKIN, M. A. MORRISON, P. T. LEAT, and S. A. GIBSON Source of the Quaternary Alkalic Basalts, Picrites and Basanites of the Potrillo Volcanic Field, New Mexico, USA: Lithosphere or Convecting Mantle? J. Petrology, August 1, 2005; 46(8): 1603 - 1643. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. JOHNSON, S. A. GIBSON, R. N. THOMPSON, and G. M. NOWELL Volcanism in the Vitim Volcanic Field, Siberia: Geochemical Evidence for a Mantle Plume Beneath the Baikal Rift Zone J. Petrology, July 1, 2005; 46(7): 1309 - 1344. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. R. RILEY, P. T. LEAT, M. L. CURTIS, IAN. L. MILLAR, R. A. DUNCAN, and A. FAZEL Early-Middle Jurassic Dolerite Dykes from Western Dronning Maud Land (Antarctica): Identifying Mantle Sources in the Karoo Large Igneous Province J. Petrology, July 1, 2005; 46(7): 1489 - 1524. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Frindt, S. Frindt, I. Haapala, and L. Pakkanen Anorogenic Gross Spitzkoppe granite stock in central western Namibia: Part I. Petrology and geochemistry American Mineralogist, May 1, 2004; 89(5-6): 841 - 856. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. EWART, J. S. MARSH, S. C. MILNER, A. R. DUNCAN, B. S. KAMBER, and R. A. ARMSTRONG Petrology and Geochemistry of Early Cretaceous Bimodal Continental Flood Volcanism of the NW Etendeka, Namibia. Part 1: Introduction, Mafic Lavas and Re-evaluation of Mantle Source Components J. Petrology, January 1, 2004; 45(1): 59 - 105. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. T. MURPHY, B. S. KAMBER, and K. D. COLLERSON A Refined Solution to the First Terrestrial Pb-isotope Paradox J. Petrology, January 1, 2003; 44(1): 39 - 53. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. B. TRUMBULL, B. BUHN, R. L. ROMER, and F. VOLKER The Petrology of Basanite-Tephrite Intrusions in the Erongo Complex and Implications for a Plume Origin of Cretaceous Alkaline Complexes in Namibia J. Petrology, January 1, 2003; 44(1): 93 - 112. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||