New Zealands volcanoes: Mayor Island

VOLCANIC HAZARDS AT MAYOR ISLAND

By, B.F. Houghton. Institute of Geological & Nuclear Sciences, C.J.N. Wilson. Institute of Geological & Nuclear Sciences, S.D. Weaver. University of Canterbury, M.A. Lanphere. U.S. Geological Survey, J. Barclay. University of Bristol

Mayor Island is perhaps the most unusual of New Zealand’s volcanoes. Despite its strikingly uniform composition of magma, Mayor Island’s history includes virtually the full range of known eruption styles over a wide range of eruption sizes. Mayor Island is the visible portion of a 700m high, 15km-wide shield volcano. The island is dominated by the 3 km-wide caldera collapse crater and yet contains numerous vents active in 3 cycles of eruptions over the last 130 000 years (see figure 1 & 2). Over this period at least 52 eruptions have been recorded from the island. The diversity of eruption styles and sizes causes problems in defining comprehensive eruption scenarios.

The majority of Mayor Island eruptions are small in size, by world standards and would pose few threats beyond the island. Only renewed volcanism equivalent to the largest known pre-historic eruption would cause risk to the mainland in the form of tsunami (tidal waves), and pumice and ash fall.

Figure 1: Map of Mayor Island volcano showing the outline of the calderas and the products from 3 major cycles of eruption. Insert shows location of the island. Note how Caldera C ‘overprints’ the southern half of the older Caldera A.

Figure 2: Aerial view of the Mayor Island caldera, showing products of cycles 1,2, and 3 of the growth of the island. The deposits are separated by the topographic walls of calderas A and C.

INTRODUCTION

Mayor Island is a strikingly beautiful and peaceful island volcano, treasured for the quality of the obsidian associated with its young lava flows. Mayor Island sits quietly in the shadow of the active volcano centres of the Taupo Volcanic Zone to the southeast, and its history as a volcano is much less well known than its human history. Like many infrequently active volcanoes, Mayor Island is deceptively benign but is best treated as an active volcano in a state of quiescence. It has erupted on average at least once every 3000 years for the last 130 000 years. (Many other smaller eruptions may have left no permanent record of deposits). This past history offers clues to the form of the next volcanic episode in the future.

This booklet is written to acquaint you with the styles of volcanic eruptions that have been recognised on the island, and the hazards that would result from a resumption of activity. Simplified hazard maps for three types of eruption style are based on the range of prehistoric activity at Mayor Island. More detail is contained in publications listed in the bibliography at the end of the booklet. For locations referred to in the text please see Figure 3.

The most striking feature of Mayor Island’s volcanic history is the diversity of eruption types. Virtually every known style of volcanic eruption is known from this small volcano that has erupted only a single uniform composition of magma (molten rock). These eruptions have included: Hawaiian fire-fountaining, less continuous Strombolian explosions, spatter-fed lava flows, viscous lava domes, phreatomagmatic explosions involving inter-action of magma and external water, Plinian falls and ignimbrite. Each of these styles is described briefly below.

Hawaiian eruptions involve rapid streaming of gas bubbles through very fluid magma, disrupting the liquid, and generating a near-continuous ‘fountain’ of gas and lava droplets often from an elongate fissure vent. The immediate product at Mayor Island is a steep-sided pumice cone (e.g. at Tutaretare), but the lava particles may re-coalesce to form liquid spatter-fed lava flows on landing. Several of the Mayor Island lava flows are clearly spatter-fed, e.g. the 8000 year old lava above Te Paritu and the top lava flow at Taumou.

Strombolian eruptions involve numerous individual explosions at intervals of a few seconds to a few hours caused by the release of larger individual bubbles through fluid magma. At Mayor Island they produce steep-sided pumice cones like those forming the sea cliffs at Oira and Oturu bays (se Figure 4).

Figure 4: Pumice cone at Oturu Bay, formed by Strombolian fire fountaining. Beds in the cone dip gently to right from the source vent to the left of the picture.

Plinian eruptions produce a rapid discharge of large volumes (0.1-10km³) of relatively viscous magma generating a 10-5- km high eruption column which rains pumice over a wide area downwind from the vent(s). In some eruptions the high Plinian column may collapse or be replaced by a lower fountain generating pyroclastic flows, hot mixtures of pumice, ash and gas that sweep outwards across the ground surface to deposit ignimbrites (see Figure 5). Only the 6300 year-old caldera-forming eruption at Mayor Island contained Plinian phases, but several earlier eruptions were close to Plinian in scale (called ‘subplinian’).

Figure 5: Deposits of the 6300 year-old caldera-forming eruption exposed at Te Ananui. ‘Normal’ pumice-rich ignimbrite (i) alternates with coarse breccias (c) rich in rock fragments derived from the walls of the vent. Note the presence of ballistic blocks (b) that impacted down into the underlying ash causing "bedding sags".

Phreatomagmatic eruptions involve hot magma coming into contact with surface or ground water. Heat from the magma transforms the water into steam with an enormous volume increase, which tears the magma and surrounding wall rocks apart. Phreatomagmatic eruptions are difficult to model and predict because there are numerous factors involved. These eruptions are important at Mayor Island because many vents lie at or near sea-level, providing a source of ample external water.

Figure 6: Front edges of several lobes of lava dome (d) inside Mayor Island caldera. Vents are to the left of the picture in the vicinity of Tarewakoura.

 
Growth of lava cones occurs when gas-poor and viscous magma is erupted quietly and passively from a vent. The viscous lava forms steep-sided overlapping lobes that build up a round steep-sided dome (see Figure 6). The principal hazard associated with the growth of lava domes is sporadic collapse of the steep dome walls to generate small pyroclastic flows and surges. Other eruptions from Mayor Island have generated lava flows which form the main lava shield(s). The lava flows have rubbly tops and zones of obsidian at their tops and bases (see Figure 7).

Figure 7: Sequence of lava flows at Cathedral Bay with contacts marked by zones of black obsidian (o).

ERUPTION HISTORY

The history of Mayor Island can be divided into three cycles (see Fig.1).

Figure 8: View of Te Ananui Flats, which are a fan of ignimbrite (i) from the 6340 year-old eruption, banked against the remains of the cycle 2 lava shield (2), which in turn is banked against the remains of the older cycle 1 lava shield (1). The contact between the ignimbrite and (2) is a fossil sea cliff, the contact between (1) and (2) is the wall of Caldera A.

Figure 9: Oblique aerial view of the northwestern coast of Mayor Island showing the presence of cycle 2 lava domes outside the caldera at Wharenui (W) and Paretao (P). T is an 8000 year-old spatter-fed lava flow draped by tephra from the 6340 year-old eruption. 1 are portions of the old lava shield that pre-dates collapse of the caldera.

Cycle One. In the first cycle, between 130 000 and 36 000 years ago, the eruption of at least 9 lava flows and 12 explosive eruptions built up a lava shield or shields now represented by the western and north-eastern caldera walls (Oira Bay to Cathedral Bay and Nohangatorea to Taumou). The explosive eruptions included Strombolian, subplinian or Plinian and preatomagmatic events. The cycle was terminated by collapse forming a caldera (Caldera A), possibly in piecemeal fashion during several small eruptions.

Cycle Two. A second smaller lava shield grew within Caldera A, on top of the remains of the Pre-A shield, between 33 000 and 8 000 years ago (see Figure 8). At the same time two lava domes and two pumice cones and lava ponds formed outside the caldera (see Figure 9). At least one subplinian eruption from this time deposited minor amounts of ash on the mainland and minor caldera collapse occurred. Cycle 2 ended with the second major caldera collapse (Caldera C), 6340 years ago accompanied by the eruption of the Tuhua Tephra. This eruption was the only definite Plinian eruption during the island’s history (see Figures 10,11). Fall deposits from this eruption are up to 70cm thick on the mainland and define a downwind westward lobe of ash and lapilli crossing the North Island. Six phases of the eruption can be recognised on the island (see Table 1). The major phases of the eruption deposited large fans of ignimbrite with coarse breccias (see Figure 5). The pyroclastic flows entered the sea, and must have caused tsunami.

Cycle Three. The third cycle, from 6340 years ago to the present day, has built a cluster of lava domes and flows within the caldera, with minor explosive activity. These lava are chemically distinct from earlier deposits at Mayor Island and so mark the start of eruption of a new batch of magma. The age of these laves remains uncertain, although they are very youthful in appearance, and lack any cover of ash from the mainland or soil development. Most authors have concluded that they may be as young as 500 to 1 000 years old. All available evidence indicates that we are in the early stages of the third cycle at present, and this may culminate at some stage in the future in another caldera-forming episode.

FUTURE HAZARDS

Only the largest eruptions from Mayor Island constitute a threat to the North Island. Most Mayor Island eruptions would not have a significant impact on the mainland but would produce widespread devastation on the island, due to tephra fall and fires. The largest events, on the scale of the 6340 year-old eruption, would cause major problems associated with fall of ash over large areas (particularly with an easterly to southeasterly wind) and the development of tsunami associated with pyroclastic flows and caldera collapse. Aviation and shipping in and around the port of Tauranga would be disrupted for an extended period.

The most significant problem at Mayor Island is the wide range of eruption sizes and styles recorded in the recent past. There is also a global absence of detailed case studies of the precursors associated with eruptions of this composition. These factors make it almost impossible to predict the exact path and size of the next event, even if a forecast of the timing of the eruption began with weak phreatomagmatic and Hawaiian fire-fountaining phases, giving no indications of the violent Plinian and ignimbrite-forming phases to follow.

The most likely products of future eruptions are described below.

Pyroclastic fall (tephra)

During explosive eruptions, magma and wall rock are torn apart and the resulting particles are carried up by the hot gas to form an eruption column. The rain of ash, lapilli (cm-sized particles) and blocks down wind from the column forms pyroclastic (or tephra) fall deposits. The distribution of the fall deposits reflects the height of the eruption column (i.e. the power of the eruption), and the wind velocity. The largest particles (blocks bigger than 20cm) do rise in the eruption column but follow independent flight paths like projectiles, and land from within a few hundred meters to 4km from the vent. These large, highly destructive particles are known as ballistic blocks.

The problems associated with fall deposits are caused both by the particles and by the presence of toxic or acidic sublimates and condensates adhering to the particles.

Fall deposits are likely to be of damaging thickness only on the island, and the high temperature of the Mayor Island magma means that particles are likely to fall hot and ignite vegetation. The typical small eruptions of Hawaiian, Strombolian or phreatomagmatic type are likely to build a 10-50cm high cone at the vent, and deposit millimetres to decimetres of ash and lapilli over the rest of the island. Little or no ash fall will occur on the mainland. Even these small events are likely to force evacuation of the island until the eruption is over, and water and power supplies can be assured.

Pyroclastic fall deposits associated with an event of the magnitude of the eruption 6340 years ago will be much more widely dispersed. An unfavourable wind velocity (northerly or easterly) would lead to direct problems for human beings, animals, horticulture and population infr-structures in the Coromandel, South Auckland, Waikato and particularly Bay of Plenty. The Mayor Island magma is also exceptionally rich in chlorine and fluorine. Poisoning of stock and water pollution may be problems even in areas where only trivial amounts of ash have fallen.

The largest prehistoric eruptions at Mayor Island have generated pyroclastic flows (mixtures of pumice, ash and hot gas) that swept outwards from the vent, hugging the topography to become channelled within valleys. Pyroclastic surges are more dilute, turbulent mixtures of gas and particles. Surges tend to be radial in their distribution and less affected by topography than pyroclastic flows.

The pumice-rich pyroclastic flows from future events comparable with the 6340 year-old eruption would be very hot and travel at high velocity. Most valleys on the island would be filled by ignimbrite and thin veneers of ignimbrite would cover most of the rest of the island. The pyroclastic flows would reach the sea and probably trigger secondary explosions or blasts as the hot pyroclastic material mixed with sea water. Theses localised but unpredictable secondary blasts would be as significant a hazard as the primary pyroclastic flows.

During the growth of lava domes and flows, the margins of the lavas will be unstable, and will periodically collapse to generate smaller pyroclastic flows. These would be of very limited extent.

Pyroclastic surges have accompanied explosive eruptions at Mayor Island, on a wide range of scales. Even smaller preatomagmatic eruptions at Mayor Island will be accompanied by surges that define a radial zone of devastation up to 1km wide. Larger examples would affect the entire island.

Many eruptions at Mayor Island have produced lava flows and domes. Some have formed as a result of slow, non-explosive extrusion of gas-poor lava from a vent, while other examples formed from rapid Hawaiian fire-fountaining with hot liquid lava droplets falling and coalescing to form "spatter-fed" lava. The growth of lava domes or flows presents few direct threats but these eruptions will be accompanied by or follow explosive activity, and potentially the formation of small pyroclastic flows associated with collapse of the margins of the lava.

The abundance of sea-, lake-, and ground-water at Mayor Island means virtually all future eruptions will include steam explosions, as magma or magmatic gas mixes with water. Phreatomagmatic explosions typically involve production of dense water-rich pyroclastic surges, or blasts, which outline circular zones of total devastation and rain-out of wet, often muddy, fall deposits over large down wind sectors, generally with less destructive effect.

All volcanic eruptions are accompanied by the release of a range of gases, principally steam and carbon dioxide (CO₂), but also toxic gases including sulphur gases (SO₂ and H₂S), chlorine, and fluorine. Mayor Island magma is unusually rich in chlorine and fluorine which are both highly toxic. The gases will accumulate in any depressions close to the vents, particularly inside the caldera. The gases are also carried in the eruption plume, often condensated on ash particles, and will produce acid rain and mist many kilometres down wind from the vent. The gases affect eyes and respiratory systems of humans and animals and damage vegetation and metal surfaces. Gas masks with acid filters offer some limited protection against volcanic gases. The gases can continue to be released from the vent area after eruptions have ceased; thus a gas hazard may continue long after other potential hazards.

Tsunami are waves caused by disturbances of the sea floor. Tsunami originating at Mayor Island could result from collapse of portions of the volcano into the sea, large submarine explosions, or entry of ignimbrites into the sea. They would require a large displacement of water and hence would probably be associated with either a very major eruption (on the scale of the 6340 year-old caldera-forming event) or a major collapse of a sector of the island.

The sheer diversity of eruption styles in the recent history of the volcano makes it difficult to come up with a selection of comprehensive eruption scenarios for Mayor Island. We have chosen three examples to present here, based on prehistoric activity, but any single future eruption may incorporate elements of all three scenarios.

Scenario 1: caldera-forming eruption

A repeat of the caldera-forming eruption 6340 years ago would have a major short term effect on Bay of Plenty and the Coromandel and potentially Waikato and/or Poverty Bay. In the maps below (Figures 10 and 11) we have taken a worse case scenario (which does coincide with the events of 6340 years ago) in which tephra fall is dispersed over South Auckland-Waikato by an easterly wind. Wind shifts during eruption to the north, east or south could easily place fresh sectors at risk. Tephra fall would be the most widespread problem associated with such a large volume event. However pyroclastic flows would enter the sea at a number of points (Figure 10) and secondary explosions would create additional hazards. There remains a possibility that the collapse would trigger a tsunami although it is not possible to prove that a tidal wave accompanied the 6340 year-old event. The pyroclastic flows, surges and ballistic blocks would devastate the entire island, which would remain uninhabitable for some years.

Figure 10: Map showing the distribution of pyroclastic (Scenario 1) flow and surge deposits associated with a repeat of the 6340 year-old eruption. This map is based on the pre-6340 year-old topography and the distribution of thick pyroclastic fans would vary from those shown and be influenced by the location of the vents within the caldera.

The principal difficulty with a future small phreatomagmatic eruption would be to forecast the precise location of the future vent. Earlier eruptions have occurred from numerous locations both on and outside the margins of the caldera. Impact of ballistic blocks, pyroclastic surges and blasts would devastate an approximately circular area of radius less than 2km (see Figure 12). Fall of tephra would create problems downwind and defoliate vegetation in the downwind sector but would probably not ignite vegetation and cause fires. Unless this event was centred in Opo Bay there would be few long term consequences of the eruption for habitation of the island.

Figure 12: Hazard map for Scenario 2.

Distinctly different scenarios apply to the extrusion of lava domes within and outside of the caldera. Eruption of a new dome within the caldera (see Figure 13) poses little threat because of the low fluidity (high viscosity) of the magma and the flat floor and high walls of the caldera. Pre-historic domes have been accompanied by minor tephra fall associated with the initial phases of the eruptions and this is likely to reach the coastline, but nit the mainland. Pre-historic domes have not been accompanied by block and ash flows, largely because the lava has been extruded on to the flat floor of the steep-sided caldera. Steam explosions could result if the flows enter the caldera lakes. Prolonged eruptions could fill the caldera leading to gravitational collapse of lava onto the outer flanks of the caldera. However the lead-in time will be considerable (weeks to years) and these events could be forecast well in advance.

Eruption of extra-caldera domes are potentially more hazardous. Dome extrusion could occur on relatively steep slopes, and at or below sea level. These circumstances would create a difficult situation, in which dome failure could occur randomly at almost any stage. The difficulties could be confined to a radius of a few kilometres from the eruption site, but could only be dealt with by implementing a rigorous exclusion zone.

This requirement for on-going forecasting during this scenario and scenario 1 above may require disposition of ocean floor seismometers, a resource not currently available in New Zealand.

Figure 13: Hazard map for Scenario 3.

More detailed information about Mayor Island volcano and volcanic activity in general is contained in the following publications:
Blong, R.J. 1984: Volcanic hazards: A source book on the effects of volcanic eruptions. Academic Press, Australia, North Ryde, New South Wales.

Buck, M.D., Briggs, R.M., Nelson, C.S. 1981: Pyroclastic deposits and volcanic history of Mayor Island.

New Zealand Journal of Geology & Geophysics 24: 449-467.

Houghton, B.F., Weaver, S.D., Wilson, C.J.N., Lanphere, M.A. 1992: Evolution of a Quaternary peralkaline volcano: Mayor Island, New Zealand. Journal of Volcanology & Geothermal Research 51: 217-236.

This text is taken from one of a series booklets which cover volcanic hazards at each active or potentially active volcanic centre in New Zealand.

The series was produced by the Volcanic Hazards Working Group of the Civil Defence Scientific Advisory Committee, which includes scientists from the Institute of Geological and Nuclear Sciences and the Universities.

• Booklets published in the series so far are:
• Number One ‘Egmont Volcano’.
• Number Two ‘Okataina Volcanic Centre’.
• Number Three ‘White Island’.
• Number Four ‘Kermadec Islands’
• Number Five ‘Auckland Volcanic Field’
• Number Six ‘Mayor Island’
• Number Seven "Taupo Volcanic Centre’