Constraints on the formation of cyclic units in ultramafic zones of large basaltic chambers |
| |
| Authors: | Geneviève Brandeis |
| |
| Institution: | (1) Laboratoire de Dynamique des Systèmes Géologiques, Institut de Physique du Globe de Paris, 4. Place Jussieu, F-75252 Paris Cedex 05, France |
| |
| Abstract: | Petrological models for the formation of cyclic units in ultramafic zones at the base of large mafic layered intrusions are still few and simple. In this study, we develop simple physical constraints, such as the volume balance and density relationships between the various liquids involved. We consider the formation of an entire ultramafic zone, made of N cyclic units due to N successive reinjections. We tackle the following problems. Are all the injections of the same chemical composition? Are the N injections responsible for the formation of the whole magma chamber? Two end-member models are examined. In the first one, the chamber grows with each injection and does not lose any magma (variable volume model, VV). In the second one, the chamber initially forms with a large volume of magma and remains at a constant volume with subsequent reinjections of small volumes balanced by eruptions of equal volume (fixed volume model, FV). Several scenarios for the formation of a cyclic unit are envisioned. Petrological data such as thicknesses of the cumulate layers and their compositions, together with the fractionation density of the cumulate phases, allow the calculation of the density evolution, as a function of the initial density only. Thus, the model yields constraints on the density of the initial liquid, and hence on its chemical composition. The volumes of the injected magmas are calculated as well as the evolution of the densities of the different involved liquids. We apply the calculation to the specific case of the Ultramafic Series of the Stillwater complex. The multiple reinjection models require that the initial liquids densities vary between 2.71 and 2.72 g/cm3, corresponding to MgO-rich liquids for which there is no field evidence, either in dikes or in chilled margins. They also require that the source produces liquids of different chemical composition. Thus, the evolution of the igneous system requires another, larger reservoir where differentiation takes place prior to injection into the shallower one. Contrary to the multiple reinjection models, a closed system model where crystallization proceeds in an isolated bottom layer does not require any specific value of density for the liquids, provided, of course, that it is of basaltic composition. On the negative side, such a model is not supported by any physical processes which would account for the formation of cyclic units. |
| |
| Keywords: | |
| 本文献已被 SpringerLink 等数据库收录！ |
|
