Giselle F. Marriner

Geochemistry of Cenezoic volcanic rocks, Baja California, Mexico: Implications for the petrogenesis of post-subduction magmas

Abstract Late Cenozoic volcanism in Baja California records the effects of cessation of subduction at a previously convergent, plate margin. Prior to 12.5 m.y., when subduction along the margin of Baja ceased, the predominant volcanic activity had a calc-alkaline signature, ranging in composition from basalt to rhyolite. Acidic pyroclastic activity was common, and possibly represented the westermost, distal edge of the Sierra Madre Occidental province. After 12.5 m.y., however, the style and composition of the magmatic products changed dramatically. The dominant rock type within the Jaraguay and San Borja volcanic fields is a magnesian andesite, with up to 8% MgO at 57% SiO2, low Fe/Mg ratios, and high Na/K ratios. These rocks have unusual trace-element characteristics, with high abundances of Sr (up to 3000 ppm), low contents of Rb; K/Rb ratios are very high (usually over 1000, and up to 2500), and Rb/Sr ratios are low (less than 0.01). Furthermore, Lan/Ybn ratios are high, consistent with derivation from a mantle source with fractionated REE patterns. 87Sr/86Sr ratios are less than 0.7048, and usually less than 0.7040, whereas the pre-12.5 m.y. lavas have 87Sr/86Sr ratios between 0.7038 and 0.7063. We have previously termed these rocks bajaites, in order to distinguish them from other magnesian andesites. Bajaites also occur in southernmost Chile and the Aleutian Islands, areas which also have histories of attempted or successful ridge subduction. It is proposed that the bajaite series is produced during the unusual physico-chemical conditions operating during the subduction of young oceanic lithosphere, or subduction of a spreading centre. During normal subduction, the oceanic crust dehydrates, releasing volatiles (water, Rb and other large-ion lithophile elements) into the overlying wedge. Subduction of younger crust will result in a progressive decrease, and eventual cessation of the transfer of volatiles when subduction stops. Thermal rebound of the mantle may cause the slab to melt, perhaps under eclogitestable conditions. The resulting melt will be heavy-REE-depleted, perhaps dacitic, but will otherwise inherit MORB-like Rb/Sr and K/Rb ratios. The ascending melt will react with the mantle to form the source of the bajaitic rocks. Furthermore, any amphibole in the mantle, stabilised during the higher PH2O conditions of earlier subduction, will break down and contribute a high-K/Rb ratio component. The implications of this study are that firstly, the subducted slab does not contribute a highly fractionated REE component in most modern arcs (i.e. the slab does not melt); secondly, Rb has a very short residence time in the mantle, and its abundance in arc rocks is a direct reflection of the input from the dehydrating slab; and thirdly, bajaitelike rocks may provide recognition of attempted or successful ridge subduction in the geologic past.

The petrogenesis of Gorgona komatiites, picrites and basalts: new field, petrographic and geochemical constraints

Gorgona Island, Colombia is remarkable not only because it contains the only Phanerozoic komatiites, but also because it has mafic to ultramafic lavas with a wide range of compositions, from moderately enriched to extremely depleted (relative to Bulk Earth). The komatiite flows are, in many respects similar to Archaean komatiites; they formed from MgO-rich (18%) liquids and have upper spinifex zones and lower cumulate zones. The cumulate zones of Archaean komatiites contain many solid grains, in contrast more than 90% of the olivine in the Gorgona cumulates is highly skeletal. This combined with the fact that the Gorgona cumulate zones are thinner than those in Archaean komatiites, suggests that the komatiite magma became strongly superheated en route to the surface. The komatiites have trace element contents intermediate between those of the basalts and the ultramafic tuffs. Some basalts have isotope compositions indicative of long-term enrichment in incompatible elements, whereas other basalts and ultramafic volcanics have isotopic signatures that imply corresponding depletion. It is apparent that the plume source region of the Gorgona magmas was markedly heterogeneous, with at least two source components contributing to the observed variation in composition. This heterogeneity may have resulted from the incorporation of different components into the plume source, or it may be the result of complex melting and melt extraction processes during the ascent of a heterogeneous plume. Despite earlier suggestions that there may have been a significant age gap between depleted komatiite and basalt flows and the enriched basal& new 4oAr-39Ar dating of basalts and gabbros are more consistent with all being generated at 87 Ma during formation of the Caribbean/Colombian plateau, possibly at the Galapagos hotspot.

Geochemical variation in peridotite xenoliths and their constituent clinopyroxenes from Ray Pic (French Massif Central): implications for the composition of the shallow lithospheric mantle

Abstract Anhydrous mantle peridotite xenoliths from a single volcanic vent in the French Massif Central are compositionally varied, ranging from relatively fertile lherzolites to refractory harzburgites. Fertile lherzolites closely resemble previous estimates of undepleted mantle compositions but the average of the Ray Pic xenoliths is much less enriched in LILE and LREE than McDonoughs (1990) average mantle [McDonough, W.F., 1990. Constraints on the composition of the continental lithospheric mantle. Earth Planet. Sci. Lett., 101, 1–18]. The wide geochemical variation in the bulk rocks reflects significant heterogeneities that can be attributed to two major processes within the shallow lithospheric mantle. The first process is depletion, related to variable degrees of partial melting and melt extraction from an originally near-chondritic mantle. This process has largely controlled the major elements and much of the trace element variation between fertile lherzolites and refractory peridotites. LREE-depleted compositions are also produced by this process. During partial melting, HREE behaved coherently with the major oxides and the moderately incompatible trace elements (Y, V and Sc). A subsequent process of enrichment is indicated by high concentrations of incompatible trace elements in many of the xenoliths. Sr, Ba, K, Th, U, Nb and LREE abundance are independent of major oxide variations and reflect enrichment related to infiltration by alkaline silicate melts/fluids. Both fertile and refractory mantle were enriched but harzburgites were particularly affected. Modal metasomatism occurred only rarely and is indicated by Cr-diopside-rich veins and patches in a few samples. Their chemistry suggests that they were also formed by migration of similar magmas/fluids from the asthenospheric mantle, although the presence of wehrlitic patches may indicate interaction with carbonate melts. In both depleted and enriched xenoliths, trace element patterns for separated clinopyroxenes closely reflect those of the bulk rock, except for Rb, Ba and Nb, which are probably hosted by other phases.

The internal structure of oceanic plateaus: inferences from obducted Cretaceous terranes in western Colombia and the Caribbean

Although the structure of mantle plume-derived oceanic plateaus has recently been assessed using remote geophysical techniques combined with petrological modelling, it is nevertheless desirable to test whether these (theoretical) rock types (dunites, gabbros and basalts) actually exist, and to establish their geochemical nature. Oceanic plateaus may have initially formed above or near sea level during a short vigorous pulse, and thereafter commonly subside to abyssal depths as the lithosphere cools, thus making sampling of their deeper levels extremely difficult. However, the Cretaceous-age Colombian–Caribbean oceanic plateau was partially accreted against the South American continent so making the imbricated segments available for study. During the process of plateau accretion and imbrication it is predominantly the basaltic layers which are obducted, but parts of the sequence down to layered and banded gabbros with associated pyroxenites and dunites (sometimes foliated) can be exposed where the imbricate thrusting brings up deeper levels. Most of the upper crustal sequence in western Colombia is composed of basaltic pillowed and massive flows and sills that are chemically uniform and ‘undepleted’ relative to normal mid-ocean ridge basalts. Komatiites and (more abundant) picrites are found at intervals, and appear to occur near the base of the sequence. In these zones both ‘depleted’ and moderately ‘enriched’ basalt and komatiite compositions occur, and may result from dynamic partial melting and mixing processes associated with the high-temperature part of the plume. It is possible to integrate these compositional characteristics into a general model for oceanic plateau structure where the rate of magma supply is in excess of that which can be accommodated by normal spreading processes, thus leading to extrusion of flows and the emplacement of sills and high-level magma chambers (but relatively few dykes). The dense and chemically heterogeneous ultramafic magmas intrude the base of the pile and undergo fractionation to form ultramafic cumulates, whereas the well-mixed basaltic magmas are erupted to form the homogeneous plateau basalts. Based on these observations a possible structure of the Caribbean–Colombian plateau is proposed which is compatible with geophysical models for other less well exposed oceanic plateaus.

Pervasive mantle plume head heterogeneity: Evidence from the late Cretaceous Caribbean‐Colombian oceanic plateau

[1] In SW Colombia picritic pillow lavas and tuffs, as well as breccias composed of picritic clasts, occur interspersed with basalts of the Central Cordillera and represent accreted portions of the 90 Ma Colombian/Caribbean oceanic plateau (CCOP). We present new geochemical data for these picrites and high-MgO basalts from SW Colombia, along with new data from Deep Sea Drilling Project Leg 15 drill sites. The 40 Ar/ 39 Ar ages for the CCOP in the Central Colombian Cordillera range from 87 to 93 Ma. Both SW Colombia picrites and Leg 15 basalts are compositionally diverse and range from reasonably enriched ((La/Nd)n > 1 and (eNd)i +8.0). Nb/Yand Zr/Y systematics suggest that the depleted component is not depleted MORB mantle, but is an intrinsic part of the plume. The bulk of the CCOP compositions can be explained by mixing between this depleted mantle and a HIMU component. However, radiogenic isotope systematics indicate the presence of an EM2 (or possibly EM1) component within the plume. Mantle melt modeling suggests that the enriched magma types are the product of deeper, small degree melting of a pervasively heterogeneous plume comprising a refractory matrix with enriched streaks/blobs, whereas shallower, more extensive melting, results in the formation of relatively depleted magmas. INDEX TERMS: 8121 Tectonophysics: Dynamics, convection currents and mantle plumes; 1040 Geochemistry: Isotopic composition/chemistry; 3640 Mineralogy and Petrology: Igneous petrology; 1025 Geochemistry: Composition of the mantle; 3670 Mineralogy and Petrology: Minor and trace element composition; KEYWORDS: mantle plume, mantle melting, oceanic plateau, picrite, komatiite, Colombia

The geochemistry and tectonic setting of late Cretaceous Caribbean and Colombian volcanism

Abstract Late Cretaceous mafic volcanic sequences in Western Colombia and in the southern Caribbean have a striking coherence in their chemistry and compositional range which suggests they are part of the same magmatic province. The chemical characteristics of the majority of the mafic lavas are totally unlike those of island arc or marginal basin basalts, so the sequences cannot represent accreted arc terranes. On the other hand their trace element characteristics closely resemble those of Icelandic/Reykjanes Ridge basalts that represent an oceanic plateau formed by extensive decompression melting of an uprising deep mantle plume. The occurrence of komatiites on Gorgona and high-MgO picritic lavas in S.E. Colombia and on Curacao, representing high temperature melts of the plume tail, confirms this analogy. Likewise, late stage rhyolites within the Colombian mafic volcanics may well be the equivalent of the extensive silicic magmas on Iceland and at Galapagos, possibly formed by remelting of the deep parts of the overthickened basaltic crust above the plume head. These volcanics, plus others around the Caribbean, including the floor of the Central Caribbean, probably all represent part of an oceanic plateau that formed rapidly at the Galapagos hotspot at 88 Ma, and that was too hot and buoyant to subduct beneath the margin of S. America as it migrated westwards with the opening of the South Atlantic, and so was imbricated along the continental margin. Minor arc-like volcanics, tonalites and hornblende leucogabbro veins may represent the products of subduction-flip of normal ocean crust against the buoyant plateau, or hydrous melts developed during imbrication/obduction.