Samples from the crystallising boundary layer of a zoned magma chamber

Abstract

The spatial and chemical relationships between the melt occupying the reservoir and the mineral assemblages crystallising at the margins are reconstructed for the magma chamber which produced the 11000 yr.B.P. tephra deposit of Laacher See Volcano. The melt showed vertical chemical zonation immediately prior to eruption, and throughout most of the magma volume only a small fraction of crystals were present. The eruption also ejected crystal-rich “nodules”, ranging from mafic to felsic in composition, which are samples of the materials crystallising at the boundaries of the chamber. New data on nodule petrography and chemical compositions of whole-rocks, minerals and interstitial glasses are presented. Volume fraction of interstitial glass is not systematically related to mineral assemblage and varies typically between 1 and 20 vol%, i.e. the crystals interlock. One exception is a group of mafic nodules with glass volume fractions between 25 and 40 vol%. Bulk compositions of mafic nodules show strong enrichments or depletions in all major elements relative to the mafic phonolite interstitial melt. Felsic nodules show much less pronounced differences with their interstitial melt. Felsic nodules contain interstitial glasses with a range of compositions similar to that in the zoned bulk of the chamber and were probably derived from different heights on the walls. Mafic nodules have glass compositions similar to those at the base of the zoned liquid column and were probably derived from the floor. Modal mineralogy, glass composition and mineral composition are systematically related in the nodules whereas in individual pumices samples derived from the main body of the chamber, a broader range of mineral compositions are often found. Mineral assemblages were especially diverse in the upper part of the chamber. It is deduced that the whole of the essentially liquid part of the chamber was emptied by the eruption, that strongly contrasting mineral assemblages were forming simultaneously on the walls and floor, that the gradient in crystal content in the crystallisation boundary layer was more gradual at the floor than at the walls, and that the pumice mineralogy is not a simple phenocryst assemblage but is a mixture of crystals which grew from melts separated in space and/or time.