Bulk-rock geochemistry and plagioclase zoning in lavas exposed along the northern flank of the Western Blanco Depression (Northeast Pacific): Insight into open-system magma chamber processes

The present study is based on a set of lavas and crosscutting dikes collected by dives along detailed vertical transects on the northern flank of the western part of the Blanco Transform Fault, Northeast Pacific. The studied area consists of a small basin, the Western Blanco Depression (WBD), extending from the southern end of the Juan de Fuca ridge to a pseudofault trace 60 km eastward. The Northern Scarp of the WBD comprises a volcanic unit overlying a sheeted-dike complex. Major and trace element data, coupled with Sr–Nd isotope ratios, reveal a two-component mantle source, composed by an isotopically depleted matrix variably veined by more enriched material. One chemical group (NS2), indistinguishable from the other Northern Scarp samples on the basis of trace element data, has an unusually depleted isotopic composition typical of a nearly pure mantle end-member. Some cogenetic samples of the Northern Scarp have been used to constrain the differentiation modalities. Anorthite and MgO content profiles in plagioclase xenocrysts and phenocrysts reveal (i) the existence of H₂O-bearing evolved melts in the mushy zones and (ii) the occurrence of mixing process between these melts and anhydrous mafic liquids. The hydration is supported by other petrographic features such as high magmatic fO₂ values, calculated from Fe–Ti oxide pairs, and the presence of pyroxene inclusions in plagioclase phenocrysts. Mixing, consistent with the existence of Ni-rich ferrobasalts, is interpreted to be the consequence of the reservoir refilling by mafic liquids (Mg# = 70). These petrological and geochemical evidences are combined with the evolution of Mg# with depth to suggest a periodic open-system magma chamber evolution beneath the southern end of the Juan de Fuca ridge.     

Petrogenetic variability along the North–South Propagating Spreading Center of the North Fiji Basin

Summary Previous studies on propagating rifts suggested that segmentation of a spreading axis could represent the superficial mark of mantle behavior (Sinton et al., 1983; Nicolas, 1989; Gente et al., 1995). The study of North–South Propagating Spreading Center (NS-PSC) from the North Fiji Basin (NFB) brings new insights to this debate. Basalts from the central part of the propagator have more variable incompatible and isotopic ratios then those from its northern tip. A model of dynamic partial melting of a thermally and slightly geochemically and isotopically heterogeneous mantle is proposed. Beneath the central segment, the partial fusion starts deeper (ca. 30km) and reaches a higher rate (ca. 22%). Further open system differentiation occurs within shallow permanent magma reservoirs along most of the central segment. Below the segment closest to the tip of the propagator the partial fusion starts shallower (ca. 25km) and stays at a lower rate (ca. 16%). The maximum of differentiation occurs close to the propagator, in small, periodically disconnected, magma bodies resulting in the production of ferrobasalts close to the tip. In order to explain these variations, the presence of an asthenospheric diapir focused beneath the central part of the NS-PSC is proposed. The petrogenetic processes of propagating spreading centers of mature oceanic basins or back-arc basins are identical.     

Highly magnetised and differentiated basalts at the 18–19° S propagating spreading centre in the North Fiji Basin

Geochemical and magnetic properties of basalts dredged along a propagating spreading centre (central part of the North Fiji Basin) have been analysed. Two phenomena appear to be important in controlling variations of natural remanent magnetisation (NRM): 1) an extensive magma differentiation to Fe-Ti basalts increases with the intensity of NRM; and both increase towards the propagator tip, 2) low temperature oxidation (maghemitisation) seems to have already occurred along the ridge axis for the zero age basalts. This paper shows that despite the apparent lack of correlation between NRM and the maghemitisation process (masked by the effect of extensive magma differentiation), the latter involves a change of the domain state of magnetic carriers, from pseudo single-domain to single-domain. We separated the NRM into 4 partial NRMs (pNRM) depending on the coercivity of grains. This analysis showed that the contribution to the remanence of grains with high coercivity increases with maghemitisation, whereas the contribution of grains with weak coercivity decreases.Despite the relatively high degree of maghemitisation, the variations of natural remanent magnetisation intensity principally reflect the variations of magmatic processes. This joint work on magnetic and magmatic properties of basalts collected on the NS-propagating spreading centre from the North Fiji Basin reinforces the magnetic telechemistry hypothesis of Vogt (1979).     

The late Neoproterozoic Enganepe ophiolite, Polar Urals, Russia: An extension of the Cadomian arc?

The Enganepe ophiolite, Polar Urals was formed at 670 Ma and records a diverse geochemical association of tholeiite, arc-tholeiite, adakite, and OIB-like lithologies. This constrains the tectonic setting of the protolith of the ophiolite to an oceanic island-arc, with ridge-trench interaction most readily explaining the diverse compositions. The initiation of intra-ocean subduction and the development of the Enganepe island arc off the eastern margin of Baltica probably pre-dated the formation of the Enganepe ophiolite, i.e. prior to 670 Ma. The timing of island-arc magmatism is similar in age to that recorded off Avalon in the Cadomian arc. We propose that the active margin of Baltica in the Vendian is an extension of the Cadomian arc. This requires the northeast margin of Baltica (present-day coordinates) to have been in a southerly position in the Vendian, in agreement with proposed tectonic reconstructions. Consequently, the post-Rodinia continental amalgamation, Pannotia, had active ocean-continent convergence along its entire southerly (west Avalonia and Amazonian cratons) margin at the time of its break-up.     

The north–south propagating spreading center of the North Fiji Basin. Modeling of the geochemical evolution in periodically replenished and tapped magma chambers

Summary Two main numerical approaches have been previously used to model the behavior of replenished and tapped magma chambers from geochemical data: 1) iterative computations, in which the magma evolution within steady-state reservoirs is modeled cyclically (P-RTF models); each cycle involves adding recharge magma, mixing the remaining liquid together, crystallizing the mixed product, then expelling part of the residual liquid (Model A); the expulsion can also take place after the mixing event but before crystallization (Model B); 2) continuous models (C-RTF models): in the corresponding time-dependent equations, the magma undergoes fractional crystallization with simultaneous replenishment of fresh liquid (Models C and D). A pertinent test of these models requires a cogenetic magmatic series having geochemical data that are not consistent with closed-system fractional crystallization. The northern tip of the north–south propagating spreading center, located in the North Fiji Basin between 18° and 19°S (NS-PSC 18–19°S), responds to this requirement. The lava ages range from 0 to 1Ma. The dredged volcanic rocks studied are cogenetic in a broad sense (constant isotopic and incompatible trace elements ratios). While no petrographic indications of wall-rock assimilation have been found, evidence of magma mixing has been observed in one basaltic sample (ribbon structures). The lavas, which are normal mid-ocean ridge basalts (N-MORBs), are distributed between three homogeneous compositional groups spatially ordered. The most differentiated lavas have a Fe-Ti basalt composition. We find that one version of open-system fractionation in a periodically replenished reservoir (Model B) is consistent with both the petrologic and geochemical data in explaining the formation of the two most mafic lava groups (Group 1, 64mg#61; Group 2, 59mg#52). In our model, the liquids expelled from a first magma chamber at the end of each cycle (Group 1 magmas) feed a second reservoir, which in its turn expels cyclically Group 2 liquids. A part of these expelled liquids are then stocked in a third closed-system magma chamber, where the Fe-Ti basalts (Group 3 lavas: 50mg#46) are generated through additional crystallization. Thus, the NS-PSC 18–19°S lavas seem to have been produced by three magma chambers interconnected by a sill (and/or pipe) network, ending in the last 18km of the northern tip. Consequently, only a small fraction of magma expelled from each open-system magma chamber reaches the surface as lava flows, because a fraction of it migrates from one reservoir to another. The off-axis sampling provides evidence for the persistence of open-system fractionation over time.Received January 23, 2002; revised version accepted May 31, 2003