The Vaalbara Hypotheses Reviewed 1Evans, D.A.D
Total Page:16
File Type:pdf, Size:1020Kb
THE VAALBARA HYPOTHESES REVIEWED 1EVANS, D.A.D., 2MARTIN, D.McB., 2NELSON, D.R., 1POWELL, C.McA., and 1WINGATE, M.T.D. 1Tectonics Special Research Centre, The University of Western Australia, Nedlands, WA, 6907, Australia; 2Geological Survey of Western Australia, Mineral House, 100 Plain St., East Perth, WA, 6004, Australia. Summary present northern Pilbara with eastern Kaapvaal (figure 2b). The The present outlines of Archean cratons commonly show Zegers et al. (1998) Vaalbara model proposes amalgamation by truncation of tectonostratigraphic features, so that wider original 3.1 Ga but perhaps as early as 3.6 Ga, and fragmentation before extents can be inferred. The Kaapvaal and Pilbara cratons share 2.05 Ga. According to isotopic ages alone, a shorter-duration similar 3.6–1.7 Ga geological histories and may have been joined “Zimvaalbara” is suggested by Aspler and Chiaranzelli (1998), for all or part of that interval. Asymmetry of Paleoproterozoic who attributed the widespread 2.8–2.7-Ga igneous activity on foldbelts and coeval foreland basins on both cratons provide Kaapvaal and Pilbara to incipient breakup (also considered by additional, qualitative constraints upon possible reconstructions. Zegers et al., 1998). Although the most reliable paleomagnetic data from ca. 2.8 Ga appear to rule out a direct or even indirect connection at that time, the succeeding billion years’ history lacks pairs of simultaneous and reliable paleomagnetic poles from both blocks, leaving the Paleoproterozoic existence of Vaalbara open to speculation. Introduction Neoarchean and Paleoproterozoic stratigraphic similarities between the Kaapvaal craton in southern Africa, and the Pilbara craton in Western Australia, are so striking that a direct paleogeographic connection during that interval has been proposed and coined “Vaalbara” (Cheney, 1996). Alternative reconstructions have been proposed; this paper reviews geological constraints on the various “Vaalbara” hypotheses, as well as other proposed Neoarchean and Paleoproterozoic supercontinental configurations that include the Kaapvaal and Pilbara cratons. For each of the Kaapvaal and Pilbara cratons, common supracrustal tectonostratigraphic elements (figure 1) include: voluminous mafic magmatism at 2.8–2.7 Ga (reviewed by Nelson et al., 1999), perhaps indicating distension and plate breakup; and passive-margin or submerged-plateau sedimentation at 2.6–2.45 Ga (reviewed by Martin et al., 1998a). On the southern margin of the Pilbara craton, an episode of 2.4–2.2-Ga deformation is documented by folding and foreland-basin development (Martin et al., 2000) that affects the 2.21-Ga Cheela Springs Basalt (Martin et al., 1998b). In the southwestern corner of the Kaapvaal craton, earliest Paleoproterozoic deformation entirely precedes the Makganyene glaciation and Ongeluk flood volcanic episode at 2.22 Ga (Altermann and Hälbich, 1991; Cornell et al., 1996), but mild warping of these strata continued until peneplanation prior to 1.93 Ga (Beukes and Smit, 1987; Cornell et al., 1998). Subsequent common tectonic features include: rifting at 1.93– 1.84 Ga (Pidgeon and Horwitz, 1991; Cornell et al., 1998) followed by passive-margin sedimentation (Ashburton “trough” Figure 1. Schematic tectonostratigraphy of the Kaapvaal and and upper Olifantshoek Group) and basin inversion at 1.8–1.7 Ga Pilbara cratons. Triangles = glaciogenic deposits; ovals = (Thorne and Seymour, 1991, Cornell et al., 1998). conglomerate; stipple = sandstone; dash = shale; Fe, Mn = Banded iron and manganese formation; brick = carbonate; v, ∧ = Vaalbara variations mafic and silicic volcanic rock; ⊕ = granitoid. According to the sequence-stratigraphic comparisons of Cheney (1996), Vaalbara existed between 3.1 and about 1.4 Ga, and the Zimbabwe craton belonged to this assemblage after 2.47 Ga. The timing of Zimbabwe’s accretion via the Limpopo belt is debated, but it has certainly been connected since 2.0 Ga (Jaeckel et al., 1997). Cheney’s (1996) reconstruction places the southern margin of Kaapvaal against the northern margin of Pilbara (figure 2a). Using paleomagnetic data from ~2.87 Ga, Zegers et al. (1998) preferred a reconstruction of Archean lineaments joining the Figure 2. Hypothesized reconstructions of Vaalbara. (a) after Cheney (1996), proposed for the interval 3.1–1.4 Ga. Earliest Paleoproterozoic sedimentary records are emphasized; tectonic vergence in 1.8–1.7 Ga foldbelts added by us. (b) after Zegers et al. (1998), for 3.1–2.75 Ga, perhaps forming as early as 3.6 Ga and disaggregating as late as 2.1 Ga. Archean lineaments are emphasized. Present north-seeking arrows for each of the cratons are shown rotated to their relative reconstructed orientations. (c) after Wingate (1998), reconstructing the cratons’ paleolatitudes according to 2.78 Ga paleomagnetic poles; a direct juxtaposition is not permitted. “Indirect” reconstructions of the Kaapvaal and Pilbara blocks— Spatial constraints arise from geographical asymmetries of the two whereby they are not joined but form part of a larger cratons and their Paleoproterozoic marginal foldbelts. In the supercontinent whose intervening blocks have since dispersed— southern Pilbara region, a thick wedge of dominantly clastic include the snakelike “Ur” configuration (Rogers, 1996) with sediment (as thick as 8 km) was deposited between ca. 2.4 and 2.2 intervening Indian and Madagascan cratons already in their Ga, marginal to WNW folding of the NNE-verging Ophthalmian Gondwanaland configuration by 3.0 Ga, a similar “Capricornia” orogeny (figure 1; Powell and Horwitz, 1994; Martin et al., 2000). formed at 1.7 Ga (Krapez, 1999), and the model of Wingate Subsequent deformation at 1.8–1.7 Ga, termed the Ashburton (1998), whose paleomagnetic results at 2.78 Ga required >700 km orogeny, shows NE vergence and extends farther north along the of separation between the two blocks according to the 95% level craton’s western margin (figure 2a). On the Kaapvaal craton, ca. of uncertainty, with Pilbara lying north of Kaapvaal in the latter 2.4–2.2 Ga deformation is spottily exposed at its block’s present coordinates (figure 2c). Martin et al. (1998a) also southwesternmost (Altermann and Hälbich, 1991) and alluded to an indirect connection between the Kaapvaal and northeasternmost (McCourt, 1995) corners (figure 1). The 1.8–1.7 Pilbara blocks. Finally, there is the “trivial” solution that the Ga Kheis belt (extending toward the Magondi belt in Zimbabwe) Kaapvaal and Pilbara cratons were paleogeographically separated lies along the western margin of Kaapvaal, verging eastward throughout Neoarchean and Paleoproterozoic time, and that their (Beukes and Smit, 1987; Altermann and Hälbich, 1991; Cornell et similar tectonostratigraphic histories manifest global tectonic and al., 1998). Tectonic strike and vergence of the 1.8–1.7 Ga Kheis paleoclimatic regimes. This is the hypothesis of Nelson et al. and Ashburton belts are consistent with Cheney’s (1996) (1999) and an implicit possibility in the cyclic- supercontinental reconstruction of Vaalbara (figure 2a), although alternative interpretation of Krapez (1997) and Barley et al. (1998). configurations could also be consistent with these constraints. Geological constraints Paleomagnetic tests Geochronological syntheses can demonstrate subtle differences The Zegers et al. (1998) Vaalbara reconstruction (figure 2b) is between two cratonic blocks, but they may have been adjacent consistent with the most reliable paleomagnetic data from 2.87 even if events are separated by 100 Myr or more (for example, Ga, albeit pushing the uncertainties to their limits. This contrasts diachronous opening of the northern Atlantic Ocean, or closure of with Wingate’s (1998) results for 2.78 Ga (figure 2c). If the the Neotethysides). Although Wingate (1998) pointed out paleomagnetic data quoted by these studies are correct, then the previously unrecognized freedom in assigning an age to discordant two cratons were either (i) widely separated throughout the U-Pb data from the Ventersdorp Supergroup in Kaapvaal, which Neoarchean, (ii) adjacent at 2.87 Ga but separated by 2.78 Ga, or could assist in intercratonic correlation with the Fortescue Group (iii) indirectly connected through intervening blocks throughout in Pilbara, Nelson et al. (1999) noted enough differences between that interval. Possibility (i) is quite plausible but not a useful null the two cratons’ Neoarchean isotopic datasets to propose only a hypothesis for paleomagnetism. Possibility (ii) would appear global tectonic association. On the other hand, we note that the contrary to the tectonic record of voluminous flood basalt “indirect” supercontinental conjunction postulated by Wingate magmatism indicating extension during or after, not before, 2.78 (1998) also permits subtle distinction in rift-related and orogenic Ga. ages, due to zipper-like opening or closure of oceanic basins (Martin et al., 1998a). In the “indirect” model (iii), the two cratons could lie at some distance within the same supercontinent; if so, they should share a common apparent polar wander (APW) path for the interval in question. In other words, the APW path segments should have the Whereas low paleolatitudes of 10–20° are suggested for the same magnitude and timing. In the case of Vaalbara at 2.87–2.78 Pilbara during 1.8–1.7 Ga orogenesis (Schmidt and Embleton, Ga, we may use the compilations of Wingate (1998) and Zegers et 1985; Li et al., 1993; Schmidt and Clark, 1994), the Kaapvaal al. (1998) to construct rudimentary APW paths to search for