Abstract Scattered occurrences of Miocene–Recent volcanic rocks of the alkaline intraplate association represent one of the last expressions of magmatism along the Antarctic Peninsula. The volcanic rocks were erupted after the cessation of subduction which stopped following a series of northward-younging ridge crest–trench collisions. Volcanism has been linked to the development of a growing slab window beneath the extinct convergent margin. Geochemically, lavas range from olivine tholeiite through to basanite and tephrite. Previous studies have emphasized the slab-window tectonic setting as key to allowing melting of peridotite in the asthenospheric void caused by the passage of the slab beneath the locus of volcanism. This hypothesis is revisited in the light of more recent petrological research, and an origin from melting of subducted slab-hosted pyroxenite is considered here to be a more viable alternative for their petrogenesis. Because of the simple geometry of ridge subduction, and the well-established chronology of ridge crest–trench collisions, the Antarctic Peninsula remains a key region for understanding the transition from active to passive margin resulting from cessation of subduction. However, there are still some key issues relating to their tectonomagmatic association, and, principally, the poor geochronological control on the volcanic rocks requires urgent attention.
A suite of lavas from northwest Mull, British Palaeocene Igneous Province, exhibit major and trace element characteristics that are best explained by very high pressure (>1.8 GPa) crystallization of an assemblage comprising aluminous clinopyroxene (Al-Cpx) plus garnet. The resulting series of consanguineous magmas are mildly Si-undersaturated. The trace element effects of this crystallizing assemblage are manifested in increasing light rare earth element (LREE) enrichment and heavy REE (HREE) depletion with decreasing whole-rock MgO, a predictable consequence of the bulk distribution coefficient (D) of this assemblage being > > 1 for HREE but <<1 for the LREE. Early crystallization of Al-Cpx and replacement of plagioclase feldspar by garnet in the crystallizing assemblage also results in increasing or near constant Al2O3 abundances in cotectic compositions with MgO < ~7 wt%. Garnet as opposed to plagioclase crystallization is also reflected in the incompatibility of Sr and compatibility of Y resulting in very high Sr/Y lavas (up to 100) with high Sr (~1200 ppm) and low Y contents (~12 ppm). Lavas also have high Zr/Y (up to 30). Major element constraints suggest the crystallizing assemblage comprised ~40% garnet and 60% Al-Cpx. Magmas that fractionated at ~1.8 GPa rose to the surface without interacting with the continental crust. Immediately underlying basalts and picrites show evidence of crustal contamination and assimilated fusible portions of the crust, therefore effectively lining the plumbing system and allowing later magmas to rise to the surface without crustal contamination. A local change in tectonic regime from extension to passive appears to be linked to a change from low pressure to very high-pressure crystallization of magmas. Whilst evidence for garnet fractionation in continental flood basalts is very rare, this paper provides a characterization of its geochemical consenquences.
Investigations into alkali basalts in the Antarctic Peninsula were made when MJH was an employee of the British Antarctic Survey. The number of people who made the fieldwork possible are too numerous to mention, but to anyone involved in the 1983-84, 85-86 and 87-88 Austral Summer field seasons, I owe a debt of gratitude. The manuscript was improved by the thoughtful reviews of Kurt Panter and John Gamble. The late Peter Barker first introduced me to the complexities of ridge-crest trench interactions and for that I will also be grateful.
The Middle Jurassic Rattray Volcanic Province is located at the triple junction of the North Sea continental rift system. It has previously been thought to be sourced from three large central volcanoes: the Glenn, Fisher Bank and Ivanhoe volcanic centres. Re-interpretation using 3D seismic and well data shows that no volcanic centres are present and the Rattray Volcanics were instead sourced in fissure eruptions from linear vents, including the Buchan–Glenn Fissure System, a c. 25 km long zone of WSW–ENE-striking linear fissure vents and associated small volcanic edifices across the Buchan–Glenn Horst. The orientation of the fissures is broadly parallel to the Highland Boundary Fault, which intersects the Rattray Volcanics at the Buchan–Glenn Fissure System, implying that Mid-Jurassic magmatism exploited pre-existing crustal structural anisotropies established during the Caledonian Orogeny. The lack of large intrusive complexes beneath the Rattray Volcanics indicates that the pre-Middle Jurassic sedimentary sequences (e.g. the Devonian–Carboniferous Old Red Sandstone Group, the Permian Rotliegend and Zechstein groups and the Triassic Skagerrak Formation) extend further than previously supposed and therefore the presence of possible subvolcanic reservoir and source rock units within the triple junction of the Central North Sea may have been overlooked.
Abstract The Paleocene to Early Eocene Faroe Islands Basalt Group (FIBG) comprises a c. 6.5-km-thick lava flow–dominated sequence located within the centre of the North Atlantic Igneous Province (NAIP). The currently defined pre-breakup and syn-breakup sequences of the FIBG are separated by a significant volcanic hiatus, during which time the coal-bearing Prestfjall Formation was deposited. This major volcanic hiatus is identified across large parts of the NAIP and was preceded on the Faroe Islands by a reduction in eruption rate evidenced by an increased number and thickness of inter-lava sedimentary beds between the simple lava flows of the pre-breakup Beinisvørd Formation. High tempo eruptions resumed after this hiatus with the development of the compound lava flow fields of the Malinstindur Formation which reveal limited evidence for inter-lava breaks. In order to investigate this key transition, flow by flow geochemical sampling of a composite c.1.1-km-thick lava flow sequence spanning this transition were collected and analysed. Three chemically distinct groups are defined based on rare earth elements (REEs) and incompatible trace element signatures. Two high-Ti groups (TiO 2 > 2 wt%), B2 and B3, dominate the sampled Beinisvørd Formation and display light REE-enriched signatures (La/Yb N c. 2.9–5.9) and evidence for garnet in the source melting region (Dy/Yb N c. 1.5–1.6). At the very top of the Beinisvørd Formation, a distinct group of lava flows, B1, displaying lower TiO 2 for a given MgO wt% (TiO 2 c. 1–2 wt%), weakly light REE-enriched profiles (La/Yb N c. 1.7–2.4) and a spinel-dominated mantle melting signature (Dy/Yb N c. 1.1–1.2) is identified. Sr, Nd and Pb isotopic signatures for the three groups overlap, revealing limited evidence of crustal contamination, and therefore supporting a mantle melting origin for inter-group variations, rather than source composition or contamination. The group B1 lava flows form a unique stratigraphic occurrence on the islands and provide clear evidence for both a reduction in the initial pressure of melting, alongside an increase in the overall degree of partial melting relative to groups B2 and B3. Increased partial melting is interpreted as evidence for the early onset of rifting and lithospheric thinning to the north of the Faroe Islands. The accompanying reduction in initial pressure of melting provides the first petrological evidence that a transient reduction in mantle temperature leads to the province-wide volcanic hiatus. Our study demonstrates an intimate linkage between rifting history and fluctuations in mantle temperature highlighting that any over-arching model for the evolution of the NAIP must take both into equal account.
A geochemical study of the heavy-mineral suite in Triassic strata from two wells in the Beryl Embayment was undertaken in order to identify changes in provenance through the Lewis Formation. Detailed provenance studies also allow the reconstruction of paleodrainage patterns and aid in predicting facies distributions. The heavy-mineral assemblage in the Lewis Formation is dominated by almandine-pyrope garnets, which are typically derived from pelitic metasediments, together with zircon, apatite, tourmaline, and rutile, some of which are rich in niobium (Nb). These latter phases could have been sourced either from metasediments or granitic protoliths, although Nb-rich rutiles have a more restricted paragenesis, being associated with alkaline plutonic rocks and granite pegmatites. In the uppermost members of the Lewis Formation influxes of more exotic provenance-specific heavy-minerals such as Cr-spinel, staurolite, and chloritoid become more common. The chemistry of the detrital Cr-spinel implies that it was derived from an ophiolitic source. Gross grain size, and facies changes in the Embayment suggest a NW to SE dispersal of detritus. Although hydraulic sorting of the heavy-mineral suite between relatively distal and proximal locations has resulted in subtle variations in whole-rock trace-element geochemical signatures, the complete overlap in the geochemical data from distal and proximal wells requires that the sediments were essentially derived from a single source area. The combination of mineralogical, sedimentological, and geochemical data implies that the dominant source for these sediments was the Devonian Old Red Sandstone (ORS), the Shetland ophiolite and the underlying Dalradian metasedimentary rocks of the Shetland Isles and the East Shetland Platform. Provenance studies such as this are central to the reconstruction of paleo-drainage patterns, which in turn lead to improved prediction of potential reservoir quality, distribution, and connectivity.