Perished in Style – Ritter Island, P.N.G.

Ritter Island, 9/2010 © John B. Dikaung (via Panoramio, Peter John Tate)

William Dampier, 1651-1715

A PERFECT VOLCANO

“The Island all Night vomited Fire and Smoak very amazingly; and at every Belch we heard a dreadful noise like Thunder, and saw a Flame of Fire after it, the most terrifying that ever I saw. … From the Furrows made by this descending Fire, we could in the Day Time see great Smoaks arise, which probably were made by the sulphureous Matter thrown out of the Funnel at the Top, which tumbling down to the bottom, and there lying in a Heap, burn’d till either consumed or extinguished; and as long as it burn’d and kept its Heat, so long the Smoak ascended from it; which we perceived to increase or decrease, according to the Quantity of Matter discharged from the Funnel.” (W. Dampier, March 25th, 1700)

Dampier’s voyage with the Roebuck was the British Navy’s first expedition devoted to science and exploration. If it were not for William Dampier, a former buccaneer or pirate turned explorer, navigator, naturalist and writer, we would not have this early description and drawing of the volcano of Ritter Island. We don’t even have the island any longer because 188 years later it unceremoniously slid into the ocean leaving no more behind than a narrow crescent of its eastern flank – a mere spec on the maps of less than 2 km length! The first I heard of it was in Wally Johnson’s book Fire Mountains of the Islands – a very commendable read by the way, on the eruptive histories of the volcanoes in Papua New Guinea and the Solomon islands; online link is at the bottom.
Ritter Island, with its steep-sided conical shape rising straight from the ocean and its fuming top must have seemed the prototype of a volcano to the seafaring explorers of the 17-19th centuries, perhaps this is why it was more often mentioned in their reports than any other of the volcanoes along the Bismarck Arc. The Island was reportedly a very regular volcanic cone about “400 toises high and 600 toises in diameter” – that is about 780 m high and 1170 m across (1 Toise was 6 pieds [ft.] = 1.949 m).

More historical sketches of Ritter Island

 

DISASTER STRIKES

In the early morning of 1888 March 13, roughly 5 km^3 of Ritter Island Volcano fell violently into the sea northeast of New Guinea and – caused a huge tsunami. Some scientists think the collapse may not at all have been related to an eruption but other papers, as well as the GVP database, mention a VEI 2 eruption for that date. Anyway, residents did note sounds of explosions and reported slight ash fall at Finschafen. Johnson (1987) associated the accounts of explosions at Ritter Island with phreatic activity caused by explosive decompression of a hydrothermal system or a small magma body as the summit of the volcano slid away. Such explosions could not have contributed much to the ensuing tsunami. Ward and Day suggested that much of the western submarine flank of the volcano, an estimated 4–5 km^3 of the Island’s debris, forms a deposit that extends down a curved northwest path through the channel between Umboi and Sakar Islands. From the marks it appears that it travelled at speeds between 40 and 50 m/s, at least 70 km from Ritter Island.

Sketch of Ritter Island in 1835 looking from the south, modified from Jacobs (1844).  Arrows show how the modern island was used to estimate the scale of the sketch. (From “Submarine landslide deposits of the historical lateral collapse of Ritter Island”)

This event, the largest lateral collapse of an island volcano to be recorded in historical time, flung devastating tsunami tens of metres high on to the adjacent shores. Several hundred kilometres away, observers on New Guinea noted 3 min period waves up to 8 m high, that lasted for as long as 3 h. It first hit adjacent coasts where hundreds of people may have died, then spread further, inflicting significant damage on islands much further out. Destructive tsunami waves may have wreaked havoc up to 500 km away.
The table below shows eyewitness observations of the tsunami waves and the damage on surrounding coasts. However, the heights (~15 m) of the zones of complete vegetation stripping on Sakar and Umboi as seen from offshore probably understate the true limits because palms and other trees along tropical coasts can withstand several metres of inundation. Therefore, the true inundation heights on Sakar and Umboi Islands may have exceeded 20 m.

Summary of direct eyewitness reports and subsequent observations of damage. (Ward and Day, 2003) – Click!

 

IS THE RITTER STILL ALIVE?
(Playing here on the German word Ritter, which means a knight, although the Island has been named after the German geographer Carl Ritter).

Oblique aerial view of Ritter Island photographed in 2006 by John Holder (the originator of Oceanic Expeditions) from SW looking NE, with some of the location names used in the report by Saunders and Kuduon (2009).

Since 1888, an active submarine vent has grown in the throat of the collapsed volcano. Due to the collapse the position of the original central conduit and an unstable steeply sloping remnant of the volcano‘s core (with some residual heat) are now below sea level.

This highly interesting report of a 4-day cruise to Ritter Island and its environs, including dives and seismological data collection, The June 2009 Investigation Of Ritter Volcano by Steve Saunders & Jonathan Kuduon of the Rabaul Volcano Observatory, shows that there are lots of questions open for future research.

Questions upon questions

? – It was found that the Southern Islet is only the surface part of a massive lava body at least 70 m thick. The origin of this very thick lava body with an apparent subaerial top surface, on such a steep sided volcano as Ritter with a quite fluid magma type, is uncertain. It could have been lava ponded in a valley or behind scoria cones, or a lava lake in an open vent. Alternatively, a degassed intrusion with an intersection with the flank or the remnant of an effusive eruption point. It may be in-situ, or perhaps, a mega block that has slipped from higher up.
? – Although specifically looked for, there was no obvious sign of pillow lavas or hyaloclastites. The chemistry and very fluid appearance of the preserved flows indicates that pillows probably would have been produced at Ritter. It is possible pillows were present but not seen, however if they are absent this may indicate that the highest parts of the zone where lava and seawater interacted are now below the area investigated, due to subsidence.
? – The first sighting of a thermal spring on Ritter was made at about 9 m depth in the north part of the South Cove. It was a hole emitting lukewarm water; a dead oyster was in-situ within the surrounding red/brown deposits. This probably indicates that the spring was formed or reactivated after the oyster grew. The lukewarm nature of the spring may imply that it is a waning feature.
? – The most obvious fact from the geological perspective is that the island appears to be extremely unstable and is undergoing rapid erosion on all sides. Even the eastern slope, which is considered a remnant of the pre-1888 cone, shows signs of frequent and extensive mass movements. Recent photos suggest substantial E-W thinning since 1940. Has this erosion been constant since 1888, or has the island been destabilised by periodic tsunamigenic-seismic and eruptive episodes?
? – Sonar images from 2004-2007 revealed what appears to be a fresh cone with its rim at 200-250 m depth. The morphology and size of this feature indicates it is almost certainly a cone formed by fresh magmatic activity, and probably not water interaction with remnant geothermal heat.
? – Another so far unresolved problem was the question Why did Ritter not stir between its collapse in 1888 and 1972 when it was reportedly constantly active before? A number of possible answers are given in the report, as well as as another number of observations that raise still more questions…

Ritter Island, 9/2010, with Central Cusp and offshore stacks. © John B. Dikaung (via Panoramio, Peter John Tate)

 

RECENT VOLCANIC ACTIVITY

A side scan sonar image of the area around Ritter, dated 2007 courtesy of R.Arculus. A fresh cone can be seen located in the centre of the concave side of the island at about 200-250 m depth. (Saunders and Kuduon, 2009)

Ritter had not been known to be active since 1888, and no thermal areas existed on the island, therefore it came as a surprise when on 8 Oct. 1972 an unusual earthquake swarm registered on remote seismographs. The main quakes were detected by acoustic sensors at the US Pacific Missile Range in the central Pacific. On local inspection it turned out that black eruption clouds from submarine activity had been seen just off the west side of Ritter, witnessed by residents on neighboring Umboi and Sakar Islands. Explosive ejections, thrown to 120 m high, occurred at the rate of one or two per minute and rumbling noises were heard at a distance of 35 km. White vapor continued to rise from the eruption site for two days. Accompanying the eruptive activity was a strong seismic disturbance. Numerous tremors were felt at neighboring islands during the disturbance and the vibrations from explosions caused landslides on Ritter Island. The macroseismic data and the fact that there were relatively long-period waves in the seismic wave trains indicated a shallow source for the seismic waves, i.e. they were probably a direct result of some type of submarine explosion or caldera collapse.

Further small eruptions were observed during 1974, 1997, 2002 and 2006.

2007: No further activity was reported until 19 May, when sea surges destroyed a boat and four houses on Umboi following an eruption, according to media reports. These also noted that 1,500-2,000 people from two villages had moved to higher ground for fear of tsunamis. As of 21 May residents could still hear rumbling noises, see “smoke” rising from Ritter Island, and feel tremors. The Rabaul Volcanological Observatory (RVO) and the Geophysical Observatory in Port Moresby did not record seismic activity from the eruption on their equipment. Ten days later, on 30 May, reports came in of high waves (4-10 m high) around Ritter Island, with a maximum of 10 m on the S part of the island. Scorched vegetation was seen in the same area, and dead marine animals, mainly reef fish, were seen around the coastline. There was no evidence of fresh volcanic material, but a new landslide scar was visible on the S tip of the island, an area that extended from the uppermost part (~100 m elevation) down to sea level. (GVP Bulletins).

The 2009 RVO report states that “Ritter is active, both volcanically and geomorphologically. More volcanic activity can be expected. Seismically this will probably not be as intense as in the early 1970’s, as the conduit seems now to be open. Volcanic phenomena may however increase in importance if the cone continues to grow towards the surface and magma is erupted into an environment of lower hydrostatic pressures. There seems to be two causes of eruptive activity, one is the rise of fresh magma at the site of the submarine cone and the other is slope instability causing water to come into contact with residual hot rocks leading to small hydrovolcanian events close inshore of Ritter“.

The 2007 episode seems to have been the last activity so far, but, to be sure, from now on I will take notice of every little earthquake reported from the area!

 

GEOLOGICAL SETTING

First, a location map for Ritter Island. The upper map shows the region of Papua New Guinea containing Ritter Island and other volcanoes, and the lower map is an enlargement of the center of the upper map, focused on Ritter Island. Via GVP, from Saunders and Kuduon (2009):

The Bismarck volcanic arc overlies the northward subducting Solomon Sea, a small ocean basin that is rapidly disappearing. The rate of subduction is highest in the eastern part of the arc and slowest in the west. In addition, the Solomon sea oceanic crust disappears from view at the surface to the west of New Britain island, as it is overrun by the collision of the South Bismarck Sea plate with the Australian plate. Nonetheless, subduction of the Solomon Sea continues westward (as seen by both seismicity data and the presence of 10Be in the lavas from volcanoes both east and west of West New Britain, s. Gill et al., 1993). Both the western and eastern ends of the Bismarck volcanic arc pass close to or are cut by transform faults of the Bismarck Sea Seismic Lineation (BSSL).

Geological Setting of Ritter Volcano, West Bismarck-New Britain Arc, after Woodhead et al., 2009

 

WHY IS THE STUDY OF A MINOR ERUPTION AND A MAJOR SUBMARINE LANDSLIDE SO IMPORTANT?

The year 1888 fell in the first decade of colonization of the New Guinea archipelago by Germany. German colonists in Papua New Guinea provided perhaps the most detailed tsunami account of the March 13th, 1888 collapse of Ritter Island volcano. They used their watches to document the incoming waves like videocam-wielding tourists would do in modern scenarios. The colonists’ Watches enabled them to mark time intervals as short as minutes, so eyewitness accounts of the Ritter Island tsunami include estimates of the wave period and wave arrival time, as well as wave amplitude. This fortunate circumstance provides now the pivotal data for scientific studies of events that happened almost 130 years ago.

Geometry of the present-day collapse scar at Ritter Island (after Johnson 1987) showing the toreva blocks in the mouth of the horseshoe-shaped collapse scar and the elongated conical shape of the remaining flanks of the Ritter Island volcano.

Nearly a century after the Ritter Island collapse, in 1987 R.W. Johnson surveyed the scar and the proximal part of the landslide using a commercial single-beam echo sounder mounted on a schooner. His survey identified a submerged WNW-facing amphitheatre 3.5 km wide and more than 4 km from mouth to headwall, the arcuate crest of which forms the western side of Ritter Island itself. At the mouth of the collapse scar, the survey located flat-topped mounds up to 2 km long, rising hundreds of metres above the surrounding sea floor, interpreted as slide blocks or torevas.

“Deposits from the 1888 Ritter Island landslide and tsunami represent a living laboratory for investigating volcano-collapse-generated tsunami.

“We had found that waves produced by a 20 m/s model landslide were too small compared with the actual data. Waves produced by a 80 m/s landslide were too big; but waves from a 40 m/s landslide were just right. It looks as though our models simultaneously satisfy data from both the landslide deposit and the tsunami that the landslide created. This sort of agreement gives us confidence in using similar, but scaled-up models to understand past giant landslides and to predict what might happen the next time the flank of a volcano heads for the ocean floor.” (From: Tsunami thoughts, III. Ritter Island) – Click!

Information discovered in this laboratory can be applied to tsunami hazards associated with similar size stratovolcanoes elsewhere“. From the data found locally, a kinematic history of the collapse can be inferred – that enables scientists to create a landslide model and estimate landslide energy – wich gives the basis for a tsunami model and its energy – and from a model of estimated damage compared to the historical facts it is possible to calculate damage predictions in similar cases and, through scaling, to the much greater hazards posed by collapses of oceanic island volcanoes. So, all this is pure math and can be found in detail in the paper by Steven N. Ward and Simon Day 2003 (link below).

More research has been done since with later tsunamis, more modern instruments and newer scientific knowledge. All compared and combined will lead to even better models of the possible ways tsunamis could take and so prevent losses and destruction.

N. B.
When researching this post I have come across a lot of material and tried to glean the most interesting bits out of each. Later I realized that in some of the newer articles/papers updated figures appeared for measurements, amounts, or calculations. To my best of knowledge I have used the latest figures but still it could be possible that I didn’t find the most recent ones. If you see something contradictory or outdated, please let me know. And, here my usual disclaimer: I am not a scientist, just a person interested in everything “volcanic”.

Enjoy! – GRANYIA

SOURCES & FURTHER READING

– GVP: Ritter Island
– Submarine landslide deposits of the historical lateral collapse of Ritter Isl., P.N.G. (2015)
– Volcano collapse and tsunami generation in the Bismark Volcanic Arc, P.N.G. (2009)
– Tsunami thoughts, III. Ritter Island: Year 1888 (Scroll down)
– Large-scale volcanic cone collapse: The 1888 slope failure of Ritter … (paywalled)
– Fire Mountains of the Islands (Book, online)
– Extracts from Cooke-Ravian…, GS of P.N.G (1981)
– 2 Books by W. Dampier (Voyage 1699) (download)
– Ritter Island Volcano – lateral collapse and the tsunami of 1888 (Ward and Day, 2003)
– The June 2009 Investigation Of Ritter Volcano (PDF)
– Oceanic Expeditions Key Target-Ritter Active Volcanic Island
– Island Arc Debris Avalanches and Tsunami Generation (2005, PDF)
– Mit Muskete und Malstift (W. Dampier’s life, 3 parts, in German)