A MULTIPLE-PARAMETER APPROACH TO ANDESITIC TEPHRA CORRELATION, RUAPEHU VOLCANO, NEW-ZEALAND

Citation
Sj. Cronin et al., A MULTIPLE-PARAMETER APPROACH TO ANDESITIC TEPHRA CORRELATION, RUAPEHU VOLCANO, NEW-ZEALAND, Journal of volcanology and geothermal research, 72(3-4), 1996, pp. 199-215
Citations number
38
Categorie Soggetti
Geosciences, Interdisciplinary
ISSN journal
03770273
Volume
72
Issue
3-4
Year of publication
1996
Pages
199 - 215
Database
ISI
SICI code
0377-0273(1996)72:3-4<199:AMATAT>2.0.ZU;2-K
Abstract
A multi-parameter approach was used to correlate andesitic tephras in a complex tephra sequence ranging in age from ca. 23 to ca. 75 ka on t he eastern ring plain of Ruapehu volcano, North Island. Field properti es, combined with ferromagnesian mineral assemblages and mineral compo sitions, were required to map and correlate this sequence. Three tephr a units could be identified based on their unique physical appearance, but other tephras could not be correlated on this basis alone. Homble nde and olivine proved to be valuable marker minerals enabling further distinction of two of the marker units recognised by field properties , as well as defining two further marker tephras. Unweathered titanoma gnetite crystals, present in all of the tephras, were subjected to maj or-element analysis by electron microprobe. Canonical discriminant fun ction analysis (DFA) of these analyses enabled the grouping and discri mination of tephra units, further aiding the identification of defined marker units, as well as defining new marker units, The titanomagneti te chemistry showed a strong relationship to the ferromagnesian minera logy, showing that the ferromagnesian phenocrysts formed from the same melt or under the same melt conditions prior to eruption of each teph ra, Canonical DFA was also applied to homblende and olivine mineral an alyses to identify further marker beds and to confirm identifications of previously defined units. This statistical analysis was found to be invaluable in reducing the large amount of compositional data from th is study into a useable form for andesitic tephra correlation and mapp ing.