Feature

History Never Repeats?

Researchers study the effects of the 1855 Wairarapa Earthquake to find clues about what may happen next time.

Rodney Grapes and Gaye Downes

The earthquake of January 23 1855 is New Zealand's largest earthquake since organised European settlement. Surprisingly there is no comprehensive account other than that of the great British geologist, Sir Charles Lyell, based on information given to him by three eyewitnesses. Naturally, Lyell's account is mainly concerned with the geological aspects of the earthquake -- faulting, uplift and subsidence -- and New Zealand featured in his work as a prime example of the enormous forces accompanying a great earthquake which could lift and crack a huge tract of land in just a few moments, the most extensive such movement that was then known.

Over the past two years, with the aid of grants from the Earthquake Commission and the Foundation for Research, Science and Technology, we have collected a large amount of data concerning the earthquake. Much of this information consists of first-hand accounts from diaries, letters, newspaper reports, contemporary Government papers, together with maps and charts, that are housed in the National Archives of New Zealand, the Alexander Turnbull Library, the National Library of New Zealand and in many other libraries and archival facilities in New Zealand and overseas.

The picture that has emerged has all the drama, implications and repercussions of a national disaster and provides the basis for a "scenario event" model of future large earthquakes occurring in the Wellington region, and New Zealand in general.

The settlement of Wellington, just fifteen years old, seems to have been the closest town to the epicentre of the earthquake. On January 23, residents were enjoying the second of two days of festivities celebrating the founding of the settlement -- not a particularly nice day, with some rain and a strong northwesterly wind. At about 9.11pm, the earthquake occurred, suddenly and without warning.

Accounts from Wellington and the Hutt Valley indicate that it lasted at least 50 seconds and possibly strong shaking went on for up to three minutes. Some people thought they could identify two or three separate shocks within that time. Aftershocks followed immediately, were frequent in the first two weeks, and even six weeks later at least one or two shocks were being felt daily.

Too Many to Record

In the first 24 hours after the main shock, people as far away as New Plymouth, Wairoa and Christchurch felt several distinct shocks with some trembling in between. In Wellington, there were so many aftershocks that identifying the times of individual events seems to have been too much for the diarists of the time. The Reverend Stock at Otaki had counted 250 shocks by 8am on the morning after the first shock and another 100 more in the next 24 hours and it is clear that he, like many others, was unable to sleep that night. Many moved out of their houses into hastily erected tents.

Our assessment from the contemporary evidence indicates that at least five aftershocks had magnitudes of about ML6.5 and many had magnitudes greater than 5.0. Over a year later, aftershocks were still being felt and Frederick Trolove, at Kekerengu on the coast north of Kaikoura, wrote in his diary in March 1856, "earthquakes as usual!"

The main shock was felt over about 135,000 square kilometres -- almost the entire land area of New Zealand. People on ships at sea also felt it, thinking that the ship had struck a reef or grated over rocks. The effects on people, buildings and the landscape have been assessed using the Modified Mercalli scale of felt intensity (adapted for New Zealand) to produce an isoseismal map in which contours enclose areas that have experienced approximately the same MM intensity.

The intensity scale has twelve steps, numbered MM1-MM12. At intensity MM9 many brick buildings are heavily damaged and some collapse, nearly all suffer some damage, many chimneys collapse, wooden houses can fall off their foundations, household goods are thrown about and many people find it almost impossible to stand unsupported. At the other end of the scale, at MM3, only a few people at rest can feel any shaking. Comparison of the isoseismal map of the 1855 earthquake with those of the 1931 Hawke's Bay and the 1929 Buller earthquakes (both with magnitudes of 7.8) show the 1855 event must have been greater with a magnitude between 8 and 8.2.

Demolition Damage

Auckland and Dunedin felt the main earthquake at intensity MM3, but in Wellington, the Wairarapa and in northern Marlborough, the damage was greatest and resulted in at least five, and possibly nine, recorded deaths, one in Wellington, at least four in the Wairarapa and possibly two in the Manawatu. Wellington and the Hutt Valley were badly damaged, to intensity MM9, but few buildings were totally demolished. Most well-constructed wooden buildings suffered little structural damage other than from chimneys falling through the roof.

The Government Offices, though wooden, collapsed because the supports, constantly wet, had rotted. Many important papers were lost. In Te Aro, a suburb of Wellington, Mr Hort lost his prized glasshouse and its contents, writing in his diary; "my greenhouse has blown up, bodily, as tho' shot with a Cannon and Shatter'd to pieces". In the Awatere Valley, Marlborough, most sod and cob houses were badly cracked and many completely destroyed. All chimneys in Wanganui fell and the little church at Putiki collapsed. Several chimneys as far away as New Plymouth and North Canterbury were cracked.

In Napier, over two hundred kilometres from Wellington, the Reverend William Colenso, the well known missionary and traveller, ran from his house and clung to the ground. There, he says in a letter to his friend Joseph Hooker at Kew Gardens in England, "the tall weeping willows with which I was surrounded threw their long drooping branches about in an imploring, frantic way -- now lashing the earth, and now sweeping the sky..the post and rail fences too, which were very dry, joined in with their unnatural notes and creaked and clattered prodigiously".

Unfortunately many settlers, like Colenso, chose to build their houses close to rivers or the sea, areas frequently underlain by alluvium, and it has been found that these areas are often subject to stronger shaking, resulting in greater building damage. In addition, the ground in these areas when subject to very strong shaking can liquefy, that is, lose its strength to support structures and act as a liquid. Sand fountains often occur and the ground can settle differentially.

Ground Deformation

Reports of superficial ground deformation in the form of extensive cracking and fissuring, sand fountaining, subsidence from compaction, and landsliding come from Waitotara, Manawatu and Horowhenua, the Heretaunga Plain, Wairarapa Valley and Rimutaka Range, the Hutt Valley and Wellington, and, in the South Island, from the Wairau Plain, Awatere Valley and Kaikoura coast - in all, an area covering about 52,000 square kilometres.

A graphic description of how parts of the Wairarapa appeared after the earthquake comes from William Bennet, a civil engineer, and serves to illustrate the typical nature of the ground deformation:

the ground [had] opened in many places 8 or 9 feet [2.4-2.7m], and sunk in one place for 300 yards [270 metres] square to a depth of 8 or 9 feet [2.4-2.7 metres]. The cracks were very frequent, and at first were of considerable depth (deemed unfathomable because people could not see their depth), perhaps 15 or 20 feet [4.5-6.0 metres] in depth, and extending for many hundred yards. Ploughed ground and mud, dry river or pond-beds were thrown into all sorts of undulations like a short cross sea, the ridges in some cases 2 feet [0.6m] in height, the prevailing direction of the cracks and ridges being generally at right angles to the apparent line of force, NE-SW.

The Rimutaka Range in the area of Palliser Bay was described as being "torn to pieces" by extensive landslides that carried away almost a third of the vegetation on both the eastern and western flanks.

Larger scale deformation of the Earth's crust caused by the earthquake were even more dramatic and were described by Sir Charles Lyell as, "the formation of a great fault and of an upheaval which is greatest in vertical height and horizontal extent than all dislocations of this kind that I am aware of to date". Surface rupture occurred on what is now known as the Wairarapa Fault and according to Lyell's informants the fissure ran for an "amazing distance of 90 miles [150 km]" along the western side of the Wairarapa Plain close to the base of the Rimutaka Range.

Uplift of the Rimutaka Range side of the fault at Palliser Bay was 9 ft (2.7m) and was measured by Edward Roberts of the Royal Engineers from the elevation of a white band of coralline encrustation that grew on the rocks up to low tide level. All along the western side of Palliser Bay, around Turakirae Head and along a large part of the south Wellington coast, the uplift stranded a wide gravel beach together with rock platforms both inside and outside Wellington Harbour.

Surprisingly, despite being an obvious feature, we have found no more than brief recognition with little description of this stranded beach except for the vicinity of Mukamuka near the northwest corner of Palliser Bay, where the uplift removed a tidal obstacle on the coastal route to and from the Wairarapa.

For many years the coastal route was important because there was only a narrow and difficult path over the Rimutaka Hill and it was the only way of driving stock to the Wairarapa. The earthquake made the Rimutaka Hill route nearly impossible, even for travellers on foot because of fallen trees and boulders and numerous landslides taking away the path altogether in places. It was to be nearly a year later before the Rimutaka Hill road was repaired and upgraded to take drays and carts.

Horizontal Movement

Horizontal displacement also occurred on the Wairarapa Fault during the earthquake and amounted to 12m, with the Rimutaka side of the fault moving northeast relative to the Wairarapa side. However, such movement was not recognised at the time. This is not surprising as disruption of man-made features such as a road or fence line are the most obvious reference lines for indicating horizontal movement on faults. It took another 100 years before the offset of various features such as streams and river terraces extending across the Wairarapa Fault were recognised as indicating that the land on either side of the fault could move horizontally as well as vertically.

Contemporary evidence that horizontal movement did occur on the Wairarapa Fault in 1855 is possibly implied by reports of sudden rise in water level by 2-3m along the southwestern side of Wellington Harbour immediately after the first shock of the earthquake that flooded the beachfront shops and houses along Lambton Quay. This rise in water level could be explained by the effect of a sudden northeast shift of the land west of the Wairarapa Fault and is analogous to the effect produced when a water-filled basin is jerked horizontally.

In addition to fault rupture, some 5,000 square kilometres of the southern part of the North Island west of the Wairarapa Fault was elevated, one report stating as far north as Manawatu Gorge. Coastal measurements of this uplift indicate that it gradually decreased in a northwest direction from the greatest elevation along the west side of Palliser Bay. At Wellington the uplift was about 1.5m and at Porirua about 0.4m. Across Cook Strait in the lower part of the Wairau Valley subsidence occurred. Reports of the amount of subsidence vary between 1.5 and 0.6 metres and can best be explained by differential compaction of soft alluvium rather than by regional subsidence of the northeastern part of Marlborough.

Tsunamis and Seiches

Another notable feature associated with the 1855 earthquake was the tsunami and seiche effects. A seiche is a regular side-to-side oscillation of an enclosed body of water which may be induced by earthquake, but by far the most common cause is wind. Seiching in Wellington Harbour following the main shock is described in many accounts but there is also evidence for seiching in lakes in Wanganui, Lake Rotorua and Rotoiti, rivers in the Hawke's Bay area, the Waikato, the Waimakariri, Kaiapoi and Avon rivers and Lyttelton Harbour in Canterbury and probably in Otago Harbour.

About 10-15 minutes after the first shock of the earthquake a tsunami swept the coasts on both sides of Cook Strait and water levels rose and fell for some hours. Reports of the effect of the wave extended as far north as Otaki, along the eastern side of Palliser Bay and along the Kaikoura coast. In Palliser Bay, the wave was estimated to have been about 9-10 metres high, and three successive waves were recorded, the last being the highest, the waves sweeping away several sheds used for storing wool.

The water in Wellington Harbour continued to ebb and flow for about twelve hours after the first shock, the movement being a combination of at least three effects -- the sudden lateral displacement of the land west the Wairarapa Fault, regional uplift and the tsunami generated in Cook Strait entering the harbour. At the same time because of the uplift there was an "excess" of water in the harbour which had to drain out, not to mention the normal tidal oscillation. It is fairly certain that the isthmus between Evans and Lyall bays was invaded by waves, possibly as high as four metres, from within and from outside the harbour.

The morning after the earthquake an enormous number of dead fish were seen floating on the surface in the middle of Cook Strait. It was remarked that these were mainly a bottom-dwelling variety (ling) that were presumably killed by shock waves propagated through the sea floor. Fish had also been stranded by the tsunami on the Rongotai Isthmus, along the south Wellington beaches and on sand dunes as far north as Otaki.

The Lesson for Today

Researching this earthquake has been very rewarding, furthering our knowledge of its effects on the landscape, buildings and lives of those who experienced it. Information on the distribution of building damage and ground damage of this earthquake and other historical and present day earthquakes is not just interesting but provides valuable information needed to understand the hazards that we are likely face in a large earthquake. Using such information, earthquake engineers can design structures and services to resist the effects of strong shaking and geologists and seismologists can identify areas susceptible to enhancement of shaking and faulting. Our emergency response organisations can better prepare for the time when we may have a large earthquake.

Even in 1855 the damage and casualties in Wellington were less than they might have been because seven years earlier in 1848 another earthquake, of lesser magnitude and centred in Marlborough, had taught the settlers that wooden structures survived where bricks and mortar fell. The early settlers were resilient, they had come to a new country far away from their homeland and were not easily daunted. Repairs were made, and a study of the best methods of building to resist earthquakes was carried out. The structural damage suffered by Wellington provided the data. Charles Rooking Carter, a builder and one of the authors of the damage report, and who also wrote on the subject after the 1848 earthquake, was perhaps first in the world to suggest the adoption of a building code. Unfortunately it was to be another 20 years before building regulations were adopted.

New Zealand sits astride the boundary between two converging tectonic plates, the Australian and the Pacific. Stresses and strains created by this convergence are relieved by many small and moderate magnitude earthquakes, and less frequently, a large earthquake. In the last 155 years, New Zealand has experienced three large earthquakes of magnitude 7.8 or more: the 1855 Wairarapa earthquake, the 1929 Buller earthquake and the 1931 Hawke's Bay earthquake.

What happened during a large earthquake a hundred years ago is very relevant to understanding what might happen in a future earthquake of comparable magnitude. Information from past and present earthquakes is required to build a more complete picture of their causes and occurrence in New Zealand and the hazards they pose.

Gaye Downes is a seismologist with the Institute of Geological and Nuclear Sciences in Wellington.
Rodney Grapes is an Associate Professor in the Research School of Earth Sciences, Victoria University of Wellington.