Compositional variation in and among small-volume primitive basaltic eruptions (Invited)
Abstract
Although conventional sampling practices rarely allow it to be seen, the extent of compositional variation within and among basaltic eruptions is typically inversely proportional to erupted volume(s). This is true at a range of scales: in general large shields are more homogeneous than surrounding monogenetic vents, and large eruptions are more homogeneous than smaller ones. In addition, in many cases compositional variation within individual basaltic eruptions is systematic with respect to time; commonly, intraplate eruptions of relatively primitive basalts show decreasing concentrations of incompatible elements and alkalinity with time, often correlated with trends in radiogenic isotope composition. Much or most of the variation expressed in these trends is consistent with an origin by mixing of two endmember melts with high and low incompatible element concentrations. In at least some cases this variation occurs in relatively primitive magmas (MgO >8-10 wt.%, Ni >150 ppm) with little or no evidence for differentiation, making it difficult to explain by fractional crystallization or crustal contamination. If the temporal-compositional trends observed in many primitive basaltic eruptions are due to differentiation or crustal contamination, it requires a mechanism capable of producing large variation in highly incompatible elements with very little change in differentiation indices. If it exists, there is no reason why it should not completely obscure mantle geochemical signatures in relatively primitive basalts in general. An alternative explanation is that small-volume primitive basalt eruptions are incomplete mixtures of mantle-derived melts from distinct sources, and 1) smaller volume eruptions contain a larger proportion of a low-degree melt endmember, and 2) temporal-compositional trends reflect a decreasing proportion of this low-degree melt endmember in successively erupted melt batches. The Big Pine volcanic field in southern California contains several such examples of temporal-compositional trends (Blondes et al., 2008) consistent with systematic shifts in the mixing proportions of low- and high-degree mantle melts. Morphometric analyses and compilation of geochemical data shows that volumes of individual eruptions within the field (from ~10-3 to ~1 km3) are also inversely correlated with incompatible element concentrations and the inferred proportions of these melt endmembers (from ~50-10% of the low-degree melt component, respectively). We interpret these observations to reflect mixing of small-degree melts produced at shallower levels with varying volumes of large-degree melts produced at depth. Small-volume eruptions contain a larger proportion of low-degree melt, which may be produced in lithospheric roofs or conduits above asthenospheric melting columns, caused by low melt productivity over a wide temperature range adjacent to higher-degree melting regions. Volume-composition relationships and detailed sampling of individual eruptions in other primitive basaltic volcanic fields and could be used to test this hypothesis.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2009
- Bibcode:
- 2009AGUFM.V24C..01R
- Keywords:
-
- 1037 GEOCHEMISTRY / Magma genesis and partial melting