Feature

Watching White Island

Long-term monitoring of White Island eruptions is providing a better understanding of geological processes.

By Dr Jim Cole

White Island, in the Bay of Plenty, is New Zealand's most active volcano. It was first recorded and named by Captain Cook in 1769, but he made no mention of volcanic activity and the name seems likely to have come from a similarity between the outline of some of the cliffs on the island to those of the Isle of Wight in southern England.

The island is itself the emergent summit of a much larger volcanic structure, the White Island Massif, which is about the same size as Tongariro.

Ash eruptions were reported by early settlers in the 1830s and 1885-6. In the late 1880s, a sulphur mining operation began. This continued intermittently until 1914 when a landslide from the back wall of the crater covered the main steam vents. The steam fluidised the debris and created a hot avalanche which flowed down the gently sloping floor of the crater, killing 11 sulphur workers.

Explosions created new craters in 1933, 1947, 1962 and 1966. In 1967, a party of geologists from Victoria University visited the island and recognised the potential for establishing a volcanic surveillance programme. This programme has continued for 25 years, involving Victoria and Canterbury Universities, and the DSIR geology division, now operating as the Institute of Geological and Nuclear Sciences (IGNS) Ltd.

Monitoring

The first surveillance visit to the island was made in July 1967, when a series of level survey pegs was installed around the floor of the crater to form a closed loop about two kilometres in circumference. By comparison of their relative heights, the pegs provided a means of checking the rise and fall of the crater floor over time.

In later years, other pegs were added until at the peak of the level survey programme in 1984 there were 35 stations. One peg, at Level Station XI, proved particularly useful, and became a reliable indicator of activity. The crater floor in this area would rise several centimetres prior to eruption -- a significant amount -- and then slowly deflate afterwards.

Another monitoring technique used in the programme is magnetic field variation. When rock is heated above a certain point, known as the Curie temperature, the Earth's magnetic field becomes disrupted. This disruption can be identified by a proton magnetometer. By comparing it to the results of the level survey, researchers can gain an idea of how deep the hot rock is below the surface.

Volcanic gases have been collected from the island for the past 22 years. Corrosion-resistant tubes of pure titanium are inserted in fumarolic vents and flushed for a sufficient time to remove air. Special sampling bottles containing appropriate solutions are then attached and gas condensed within them. After sealing, the bottles are taken back to the laboratory for analysis.

A permanent seismograph was installed on the island in December 1976. A seismometer was cemented in place on a lava outcrop part way up the crater wall on the south side. This is linked by a cable to a transmitter at the top of the crater wall, and the signal telemetered to Mt Ngongataha and then to a recording site in Rotorua. In future, the signal will be sent on to IGNS's Wairakei Research Centre via a telephone line.

Throughout the period, logistic support has been provided by No.3 (helicopter) Squadron of the RNZAF, without whose help the programme would almost certainly have folded years ago.

Eruptive Cycles

The island has experienced three eruptive cycles during the monitoring programme, and the combination of monitoring techniques has allowed prediction of the cycles' activities.

The first cycle was under way when the programme began, and continued until 1971. Two new vents, Rudolf and 1971 crater, formed. Both were small, but accompanying eruptions were predicted by monitoring information.

In 1972, the crater was quiet, but throughout 1973, 1974 and early 1975 parts of the crater floor adjacent to the active vents began to bulge, the magnetic field decreased and fumaroles became more active. On the morning of 21 October 1975, a thick plume of white steam rose from the island and two scallop-shaped depressions were spotted on the crater floor. These soon filled with water.

Activity was low in 1976 but fumaroles remained hot at around 630^oC. On 18 December 1976, a new major eruptive cycle began that heralded the island's most intensive activity for the last 100 years, including the eruption of new lava for the first time since records began.

The rising body of magma which created this eruption interacted with the acidic and saline hydrothermal system surrounding the vent, leading to highly explosive eruptions. The eruptions proceeded through vent-clearing phases which simply erupted crater fill material, to strombolian eruptions when new lava bombs and blocks were scattered around the crater, and finally a period of vent collapse in which the crater floor surrounding the active vent was engulfed to form large vertical-walled collapse craters or maars. This cycle ended in 1982.

From 1983 to 1985, the island was relatively quiet before entering a third phase of activity in 1986. This was dominated by collapse. New craters were formed by explosive eruptions, only to be consumed in later episodes of collapse. The result is that the active crater area has now dramatically subsided, the subsidence destroying many of the key level-survey pegs in the western part of the crater.

Prediction and
Modelling

The 25 years of observations have indicated that a combination of monitoring techniques can be successful in predicting larger eruptions in this type of volcano. As the data pool increases, it may be possible to predict changes in eruptive style as well.

The data also allows us to build up an integrated model of what must be happening under White Island. The existence of substantial volumes of magma at depth is indicated by the long term output of sulphur dioxide gas and convective heat from the island. It was a small volume of this magma which rose to within a kilometre of the surface in 1976, causing a series of highly explosive eruptions. These eruptions left gaps above the magma intrusion, into which crater fill material collapsed to form the maars. This process is continuing today.

The longer the period over which data is collected, the more accurate predictions are likely to be. An exciting possibility is to drill into White Island to confirm conditions beneath the surface. An international group including IGNS, Geological Survey of Japan and US scientists hope to be able to do this within the next two years, if funding can be secured. This would be a marvelous opportunity to confirm the suggested structure of the island and learn something about an environment where many hydrothermal ore deposits are thought to form.

Dr Jim Cole is professor of geology at Canterbury University.