Last month's eruption at White Island likely would have killed anyone standing close to the floor of its Crater Lake, scientists say.
Fortunately, the eruption happened late at night and no one was on the island when a surge of ash was thrown across the crater floor late on April 27.
In a blog post, GNS Science volcanologist Brad Scott said the steam and gas driven eruption from the offshore volcano created a new crater, caused landslides and excavated some of the lake.
Much of the crater floor was covered with a green-tinged ash - but Mr Scott and his colleagues have been able to confirm that none of the material contained magma from deep within the active volcano, which rises 1.6km from the ocean floor off the coast of Bay of Plenty.
The scientists recorded a "complex pulse of seismic data" that lasted about one and a half hours, starting at about 9.31pm.
The largest and most energetic portion was at about 9.53pm, suggesting the eruption climaxed about then.
But the size of the signal was also small compared with activity between 2011 and 2013 and the acoustic signals were not clear, and were difficult to interpret.
Detecting the eruption was also hampered by the fact the island's gas sensors work only in daylight, while the web cameras were also able to pick little up in the darkness.
"What we didn't know on the evening was that the blast deposit obscured our web camera on the north rim, so even as the sun came up we still couldn't see to confirm the activity," Mr Scott said.
Since then, GNS has found images from the MetService's rain radar, showing an eruption plume reaching to 3km by 9.52pm and over 4km by 10pm.
The next morning, scientists checked the web cams and noticed a potential new deposit of ash and later found two solar panels weren't working properly because they may have been coated in ash.
"We knew something had happened but it wasn't until GNS volcanologists flew over the volcano in the afternoon they were able to fully appreciate the impacts of the previous night's eruption," he said.
"A dark yellow-green ash deposit from the eruption covered about 80 per cent of the floor of the main crater, blasted high up the sides of the crater wall on both the north and south sides."
While GNS didn't consider it safe to visit the crater floor, staff did service the equipment outside that area to ensure valuable monitoring data continued to reach volcanologists.
The eruption deposit at the north rim site was typical of what volcanologists referred to as a "hot volcanic surge deposit".
It was plastered on sides facing the vent, but was absent on shaded surfaces, suggesting the surge travelled horizontally, across the crater floor, and didn't fall from a plume high above.
Mr Scott said the ash also had a dry and hot texture, which he likened to apple crumble.
"Survivability during the eruption would have been low on much of the crater floor."
Q&A: GNS Science volcanologist Brad Scott answers a few question about the eruption.
Q. What's up with that green ash? Is that mineral makeup?
A. No, thanks Twitter. It is not mineral makeup. That amazing green colour was created from a unique combination of minerals on White Island. The ash contained quite a bit of pale yellow sulphur which appeared as irregular-shaped lumps, and some translucent golden yellow grains. This is consistent with the colour of the deposit.
Q. Was there any juvenile material found in the ash?
A. No. The reason we tested for juvenile material was to determine whether there was new magma (lava) coming up or whether this was older material.
New material could indicate that this would have been the start of a whole new phase of activity for White Island; however, in this eruption it only tossed out old lake sediments and crater wall material.
For people who want more science, here it is: the volcanic blast deposit collected from the north rim site solar panel was found to have no new juvenile material.
Most of the ash is strongly hydrothermally altered old rock fragments, with some fresher fragments of volcanic glass and crystal.
However, these are attached to or form a partial coating of white to pale grey hydrothermal altered material.
From our past experience of examining White Island ash we are confident that this is not April 2016 magma (new stuff).
When no juvenile material is found, this means that new magma was not involved in the eruption.
Q. If people were hanging around the crater during the eruption, would they have survived?
A. We are really happy the eruption happened at night.
The blast would have been travelling fast and was hot, hence survivability during the eruption would have been low in many areas of the crater.
Q. What can you tell us about this new vent?
A. During the visit on Thursday afternoon we obtained some thermal images of the crater area, and during the aerial inspection on Friday to measure the gas output we had better views of the 1978/90 Crater area.
From the photography and thermal images we were able to identify the location of a new vent that had erupted on Wednesday evening.
Significant collapse had also occurred in this area, formally known as Donald Duck Crater.
Several metres of Donald Mound have collapsed into the 1978/90 Crater area, taking away our target that was used to calibrate measurements from our web camera images. It had been repainted only six days before.
Q. Are we going to call this new vent Venty McVentface?
A. No. Not even once. Stop asking us that, Reddit.
We are much more boring, we go for month/year so we can keep track of them ... once you have 20 to 30 vents with cool names it becomes a challenge. We have had some great ones; Orca, Gilliver, Rudolf, Donald Mound, Christmas, Noisy Nellie, RF, PJ, TV1, Blue Duck, Donald Duck (when Donald mound and Blue Duck joined), Big John, GIBRUS, just to name few along with the boring ones like 1931, 1971, 1978/90.
Q. Why did the Volcanic Alert Level get lowered so quickly?
A. Following the eruption the volcano did not start to produce volcanic ash from any of the active vents.
The seismic activity remained at relatively low levels and gas emission data indicated SO2 output was at levels similar to those prior to the eruption.
As a consequence of the continued lack of activity after the eruption, the Volcanic Alert Level was lowered to Level 2 on Thursday evening.
This indicated that the volcano was not erupting but remained in a state of moderate to heightened volcanic unrest.
Q. What's going to happen next with White Island?
A. That's the $64 million question.
White Island is by far one of our most active volcanoes. Heightened volcanic unrest continues at White Island and no eruptions have occurred since April 27.
Volcanic gas output and seismic activity, which is dominated by volcanic tremor, increased in 2011 when volcanic unrest developed.
That led to the eruptions in 2012 and 2013.
These two unrest parameters remain elevated and have defined a "new background".
The possibility of further eruptions remains high and the Volcanic Alert Level remains at Level 2 (moderate to heightened volcanic unrest).
White Island can and does continue to intrigue us, providing us rich scientific data that increases our understanding of volcanoes in New Zealand and around the world.
We aren't entirely sure what the volcano will do next.