Remnants of the world-famous Pink and White Terraces, once thought to have been completely destroyed by the eruption of Mt Tarawera, may still exist and a manned submersible is about to go hunting for them.
Educationalist Sir Toby Curtis, chair of the Te Arawa Lakes Trust, was born in Rotoehu, north of Lake Tarawera, in 1939, just over half a century after the eruption of Mt Tarawera on June 10, 1886. He recalls the black ash still covering the roads before they were tar sealed. “So, it was part of my existence. I grew up as a child hearing people talk about the terraces as though they were still in existence – they spoke about them so much that I grew up believing I had seen them. It wasn’t until I went to see the terraces in Pamukkale, in Turkey, that I suddenly realised I had never seen a terrace. I had seen all the lovely pictures, but I had never seen a real terrace. That had such an effect on me personally – I was able to relate to the voyage of my self-history by looking at these real terraces.”
All going according to plan, Curtis may yet get to glimpse at least a fragment of the real terraces that haunt his muddled childhood memories.
Climbing above Lake Rotomahana in fern-framed tiers of fine-grained silica, or sinter, the famed Pink (Te Otukapuarangi) and White (Te Tarata) Terraces attracted local Māori, early settlers, photographers, writers and a throng of 19th-century tourists, and were a contested commercial opportunity for local Te Arawa hapū, now represented by the Tūhourangi Tribal Authority and Ngāti Rangitihi.
Then they were lost, left, like the phantom canoe seen days before the eruption, to a future of conjecture and debate.
Were they buried? Drowned? Or blasted into oblivion? Can they be recovered, dug up or revealed in a spectacular lake drain? Would they even be recognisable? GNS geologist Cornel de Ronde is hopeful a manned underwater expedition into the depths of Lake Rotomahana, due to be launched in November, will settle the matter.
“We have done high-resolution bathymetric and side-scan sonar mapping and sent down a remotely operated vehicle. Now it’s time to see them up close and personal,” he says.
Is this our Titanic? “I think it’s bigger than that,” says de Ronde. “The Titanic was a modern wonder of engineering and the world was very excited about it at the time, but this is a natural phenomenon of a sort there has never been before or since – there’s a lesser example in Turkey, but nowhere near the same size as the Pink and White Terraces. Rediscovery of some parts of the terraces has great importance for iwi and New Zealand.”
If, that is, such parts even exist. In the wake of the eruption, local tourist guide Alfred Warbrick claimed the terraces survived under 15-30m of mud and ash. Painter Charles Blomfield, whose 1882 paintings of the terraces are still considered definitive representations of the spectacular geothermal formations, said he was convinced of the impossibility “of the least fragment of the terraces ever being seen again”.
Looking in the lake
Over the past decade, debate about the Pink and White Terraces has resurfaced, sparked by two sketch maps prepared 100 years apart. The earlier of these, a recently discovered watercolour sketch of Lake Rotomahana by German-Austrian geologist Ferdinand von Hochstetter, dating back to a three-day visit he made to the lake in 1859, has been used to suggest whatever remains of the terraces would be buried wholly or partly on land close to the lake.
The other, a photogrammetric map made by physicist Ronald Keam in 1959, using pre-eruption photographs by Alfred Burton, George Valentine and Charles Spencer, shows any extant terrace features would be submerged in the lake, close to the shore.
In 2011, de Ronde led a combined US and New Zealand exploration of Lake Rotomahana, using two autonomous underwater vehicles (AUVs), not to find the terraces, but to understand what happens when a land-based geothermal system is drowned. Images from a side-scan sonar attached to one of the vehicles revealed what appeared to be crescent-shaped, stepped structures, their centres filled with brownish sediment, under about 60m of water.
“Where we thought the Pink Terraces might have been, we noticed these incredible structures that were hard and circular,” says de Ronde. “As we wrote in one of a tranche of published papers, it is conceivable parts of the Pink Terraces survived.”
Later inspection of the data revealed a rock pillar or dyke similar to that seen in a number of post-eruption photographs but since drowned by the rising lake waters. Called The Pinnacle, this is thought to be all that remains of the ridge on which the White Terraces were perched. In line with Keam’s placement of The Pinnacle on his map, this would place the sites of both terraces within the lake.
“We’ve always said it looks like the White Terraces have been demolished, because the ridge on which they were sitting, except for The Pinnacle, has been completely destroyed – we have evidence that shows the magma came up right underneath this ridge,” says de Ronde. “But never say never. We took a photograph a few years ago that could be a remnant of a terrace lying down there, but in terms of complete preservation, terrace upon terrace, I would say it’s impossible. Maybe there are pieces, but nothing like what we believe survived of the Pinks.”
A bunn fight
In the meantime, in 2010, in a private collection in Basel, Switzerland, research librarian Sascha Nolden discovered a field diary containing Hochstetter’s recorded observations and two sketch maps. Analysis of one of these, an unpublished 1859 watercolour map and accompanying field notes, by independent researcher Rex Bunn, led him to conclude that the later 1864 map, based on Hochstetter’s drawings and published under his name by German cartographer August Petermann, included errors made by Hochstetter. These mistakes, says Bunn, are probably the result of the use of steam plumes, “which were the only way of finding many of the geothermal features around the old lake”. After reconstructing the German geologist’s surveying methodology, Bunn announced Keam’s placement of The Pinnacle was wrong and that any surviving remnants of the terraces, including the less significant Black Terrace (Te Ngāwhā a Te Tuhi), documented by Hochstetter but left off 20th-century maps, would be wholly or partly on land, rather than in the lake itself.
De Ronde challenged this claim, arguing that data collected by GNS Science, in partnership with Woods Hole Oceanographic Institution in the US and the University of Waikato between 2011 and 2014, shows the terraces could not be buried on land. He and Keam re-examined pre-eruption photographs showing where the terraces were in respect to certain landmarks.
“If you were in a court of law and presented this evidence, it would be case closed,” says de Ronde.
Bunn thinks not. In an article published last month, he argues commercial and amateur photographers, including Spencer, “embellished, staged and sometimes faked” their photographs to boost sales or compensate for primitive 19th-century camera technology, so undermining the research based on these early images.
“I’m surprised it hasn’t been noticed before,” says Bunn, who is working on an updated edition of his 2016 book on the terraces, “because they have been blindingly obvious. Spencer was a bit of a rascal and a bit of a good businessman and capitalised on whatever he did to make the most of his income.”
To set the record straight, the Tūhourangi Tribal Authority, kaitiaki/guardians of the lake, and the Te Arawa Lakes Trust asked the National Institute of Water and Atmospheric Research (Niwa) to review Hochstetter’s survey notes and maps. Using digital elevation models and the historic documentation discovered by Nolden, Niwa scientists Andrew Lorrey and John-Mark Woolley confirmed Hochstetter and Petermann’s published map is more reliable than the topographical sketch map and that the terrace locations are entirely within the confines of modern Lake Rotomahana, close to where the GNS research team had determined them to be. Any remnants of the Black Terrace, they concluded, would lie on land west of the lake’s shoreline.
Further evidence was found last year when de Ronde teamed up with Boxfish Research to drop a remotely operated vehicle (ROV) into the lake where his previous research suggests the Pink Terraces would have stood. The images, shown in Prime TV documentary Beneath New Zealand in June, identify eroded terrace-like structures, sediment-filled pools and strange organ pipe-like features that can’t form underwater and are “consistent with them being remnants of the Pink Terraces”, says de Ronde.
Battle of the maps
Further validating these findings, science writer and field art historian George Hook has been working on the so-called “battle of the maps” by ignoring the maps altogether. Instead, he turned to the photographic collections of Te Papa and the Alexander Turnbull Library. In previous research, Hook had used satellite imagery, GPS and digital elevation models to locate the vantage point of artist Eugène von Guérard’s Australian landscapes. When he tried to apply the same processes to Blomfield’s pre-eruption paintings, however, he found the artist had ratcheted up the dramatic effect by making the distant mountains appear higher and foreshortening the middle ground, distorting the perspective. In frustration, Hook turned to early photographs of the lake, applying the same spatial technology to identify the spot where the photographers set up their tripods.
Working with geologist Stephen Carey and photographer Andrew Thomas, Hook retrieved five dry-plate negatives of the Pink and White Terraces by Dunedin photographer Burton, thought to be exposed in October 1885. These were digitally stitched together to form a 235° pre-eruption panorama of the lake’s surroundings. As described in the latest issue of Te Papa’s research journal Tuhinga, Hook used the PeakFinder app, commonly used by trampers to identify surrounding mountains, in conjunction with Google Earth’s satellite view of the landscape, to match the profiles and positions of distant peaks and hills with those in the panorama to determine the spot where Burton placed his camera.
This turned out to be well within the present lake, a site given further weight when the Tūhourangi Tribal Authority took Hook’s team on a boat trip across Lake Rotomahana. Hook recalls the strong sense of déjà vu as the skipper rotated the launch above the spot where Burton had once stood on solid ground to take his photos and “the more distant landscape features in the photographic panorama that I was familiar with so strongly resembled the actual modern post-eruption landscape”.
This did not prove the terrace locations, but by intersecting these bearings with those in one of Spencer’s photographs, Hook was able to concur with GNS Science and Niwa that any extant terrace features would not be under land, but under sediment on the lake bed close to the present shoreline “and therefore potentially accessible”.
Potentially, possibly, probably – these words litter the many reports related to the terraces.
For some, they’re a mystery best left alone. “Accuracy and knowledge have a scientific importance,” says chair of Te Mana o Ngāti Rangitihi Trust, Leith Comer. “But I don’t think an exact knowledge of where they are is that important. It is like the Loch Ness monster – you don’t have to see it or find it. The story is there with this big eruption and these wonderful things were buried – that in itself is the story and I wouldn’t mind if that was the story.”
For those wanting more answers, draining the lake is not an option – that would risk triggering a hydrothermal eruption, potentially blowing apart any terrace remains. De Ronde is pinning his hopes on the planned manned exploration of the lake. “We have gone some way in proving remnants of the Pink Terraces survived the eruption, then followed that up with the images obtained by the ROV, but that’s not hardcore proof,” he says. “Hardcore proof is to get a piece of the terrace itself, or blow off the floc [fine particles] sitting on the terrace surfaces and see some of the delicate detail you see in the photographs – then you’ve nailed it.”
Whatever they do find won’t look anything like the pastel flamboyance of a Blomfield or JC Hoyte painting – any vestiges of the terraces would have had tens of metres of debris showered on top, then sloughed off onto the lake floor. But, de Ronde says, “If we can recover a piece of sinter and show how that location fits in a photograph, then we have proved it.”
This article was first published in the August 31, 2019 issue of the New Zealand Listener.