whole_AdamJohnDouglas1989_thesis.pdf (7.17 MB)
Experimental studies of alkaline basalt genesis
thesisposted on 2023-05-27, 06:42 authored by Adam, JD
This thesis documents an experimental study of the effects of H2O, CO2 and pressure on the compositions of basaltic liquids in equilibrium with garnet, clinopyroxene, orthopyroxene and olivine (garnet lherzolite). The results of this study have been combined with compositional data for natural basalts and used to examine some currently popular theories for the origin of alkaline basalts. For this purpose, the thesis has been divided into two parts. Part I is a study of liquidus phase relationships at 28 kb in the simplified basaltic system CMAS (CaO-MgO-Al2O3-SiO2). Phase relationships were studied for dry, hydrous and CO2-bearing systems. In the dry system garnet, diopside, enstatite and forsterite, were found to co-exist with a liquid of the approximate composition 50% SiO2, 16% Al2O3, 25% MgO, 9% CaO. Increasing pressure results in an expansion of the garnet liquidus volume and thereby causes liquids in equilibrium with garnet lherzolite to increase in CaO/Al203. The effect of pressure on the enstatite-forsterite cotectic is, however, relatively small. With the addition of 20 wt.% H20, the enstatite and forsterite liquidus volumes are greatly expanded relative to those of clinopyroxene and garnet. As a result of this, H20-rich liquids in equilibrium with garnet lherzolite are more CaAl204-rich, and less magnesian, than those produced under dry conditions. Addition of CO2 produces a large expansion of the garnet and enstatite liquidus volumes. This causes CO2-rich liquids in equilibrium with garnet lherzolite to be Si02-poor and have high Ca0/A1203. The simple-system liquid trends can be compared with the compositional trends of primitive alkaline basalts from southeastern Australia, central Europe, Hawaii, and the Canary Islands. Although the two sets of data cannot be directly superimposed, the primitive alkaline basalts show the same sense of variation as do liquids produced in the simple system by partial melting of garnet lherzolite in the presence of variable concentrations of CO2. When the effects of the additional components TiO2, FeO, P2O5, alkalis and H20 (found in natural systems) are considered, the peridotite-melting trends become more closely matched with the compositional range of the natural basalts. The effect of pressure on the orthopyroxene-olivine cotectic is also increased; with increasing pressure during partial melting of natural garnet lherzolites, liquids become more Si02-poor. Both variation in pressure and variation in CO2 concentrations during partial melting of garnet lherzolite appear to be capable of producing the compositional range of the natural alkaline basalts. Additional information is needed to decide in individual cases which of the two mechanisms, if either, offers the more likely explanation. Part II is a case study of a small group of basalts from the Oatlands district of Tasmania. The basalts are of Oligocene age and include olivine tholeiite, alklali olivine basalt, nepheline basanite, transitional nephelinite, and olivine melilitite compositions. Experimental data, combined with independent evididence for primary volatile concentrations in basalt magmas, indicate that the primary magmas of these basalts probably formed in equilibrium with lherzolite and garnet lherzolite residues at pressures between 10 and 50 kb. Calculated concentrations of minor and trace elements in the basalts source are consistent with a source that was enriched in H2O, CO2, K20, P2O5, FeO, Ba, Sr, Nb and LREE relative to the bulk composition of the mantle. Deep mantle plume and heterogeneous mantle models appear to be inconsistent with the distribution of Cainozoic basalts in Tasmania and other parts of eastern Australia. Experimental and petrological data are consistent with a relatively shallow (lower lithospheric/upper asthenospheric) location for the source prior to the beginning of volcanism; isotopic data indicate that this would need to have become enriched in LREE (and presumably in other incompatible elements as well) at some time following a previous and much longer history of LREE depletion. Because of similarities between the calculated source compositions and those of metasomatized Cr-diopside xenoliths from southeastern Australia, it is possible that the basalts'source region was enriched in incompatible elements by processes similar to those which affected the lherzolite xenoliths. However, differences between the calculated source compositions and those of the metasomatized xenoliths (particularly in TiO2) suggest that the basalts'source was unlikely to have been depleted and refractory lithosphere that was subsequently re-enriched in incompatible elements. This factor, in combination with the high pressures of origin (40-50 kb) inferred for some of the basalts, is consistent with an asthenospheric derivation for the source region.
Rights statementCopyright 1989 the Author - The University is continuing to endeavour to trace the copyright owner(s) and in the meantime this item has been reproduced here in good faith. We would be pleased to hear from the copyright owner(s). Thesis (Ph.D.)--University of Tasmania, 1990. Bibliography: p. 120-135