Magma Origin and Evolution of White Island (Whakaari) Volcano, Bay of Plenty, New Zealand

Abstract

White Island is an active composite stratovolcano in the Bay of Plenty, New Zealand, that comprises many small volume (<0·1 km3) andesite–dacite lava flows and pyroclastic deposits with phenocryst contents of ∼15–44%. Minor high-Mg basaltic andesite explosive eruptions, such as those of 1976–1992, may have occurred at intervals throughout the history of White Island, but are rarely preserved. These alternate with major episodes of andesite–dacite lava extrusion. The high-Mg magmas form by hydrous melting of mantle, metasomatized by fluids from the dehydrating slab at the slab–mantle wedge interface, that rise rapidly to shallow magma chambers (2–7 km?) where limited mixing and contamination occurs before eruption. Some of this magma remains in the magma chamber where it interacts with the crystal mush, from which it inherits phenocrysts, to form so-called ‘dirty’ lavas. Total phenocryst content of these lavas is correspondingly higher. As more magma is intruded into the chamber, the heat flux will increase and melt fraction will eventually rise to the surface to form high-silica andesite–dacite magma (‘clean’ lavas) with fewer inherited phenocrysts. Similar multi-magma chamber plumbing systems, with complex evolution involving fractionation and contamination, probably occur in most andesite–dacite arc volcanoes.

Geomorphic development of White Island Volcano based on slope stability modelling

“…Cole et al (2000) suggest instead that the breaches in the crater rim represent hydrothermal explosion craters formed by "long-lived, semicontinuous, weak, hydrothermal explosions" like those that have been documented in the western sub-crater over the last few decades.…”

“…Clark and Cole (1986) and Graham and Cole (1991) support this idea, with the latter presenting geochemical evidence that distinguishes Troup Head lavas (silicic andesite) from Central Cone lavas (dacite). Cole et al (2000), from analysis of a large amount of geochemical data from White Island rocks compiled from various sources, also note that lavas from Troup Head differ from those in Central Cone in terms of mineral assemblage and chemistry, and show similarities with some Ngatoro Cone eruptives. However, they interpret this in terms of eruptive cycles and mixing history as they are emphatic that the lavas of Troup Head, Pinnacle Head and the southern main crater wall comprise a single flow.…”

“…However, they interpret this in terms of eruptive cycles and mixing history as they are emphatic that the lavas of Troup Head, Pinnacle Head and the southern main crater wall comprise a single flow. This is based on interpretation of the stratigraphy in this area, with "good evidence that Troup Head lavas … are stratigraphically part of Central Cone" (Cole et al, 2000); unfortunately, this stratigraphic evidence is not presented in the paper. While the origin of Troup and Pinnacle Heads is debatable, there is enough evidence to question an unequivocal interpretation of these as in situ components of the Central Cone rim.…”

“…In particular, there is a long period with little history of deposition between the youngest lava flows identified by Black (1970) on the upper rim of Central Cone, and the 1977 eruptive material described by Cole and Graham (1989), Graham and Cole (1991) and Cole et al (2000). Historical activity at White Island has been characterized by continuous, intense fumerolic activity interspersed with small phreatic and phreatomagmatic eruptions (Houghton and Nairn, 1989a); the largest eruption sequence occurred during December 1976-January 1982 with sequences of phreatomagmatic and strombolian eruptions (Houghton and Nairn, 1989b).…”

“…The timing of the development of Central Cone with respect to erosion of Ngatoro Cone is unknown, and Cole et al (2000) suggest that the stratigraphic sequence is at least semi-continuous through time. Central Cone developed as a small, but "typical" stratovolcano something like that shown in the reconstruction.…”