The quest to understand future earthquakes — where, when and with what magnitude they might occur — is one that’s been testing seismologists and civil authorities for many, many years.
Recent events such as the devastating series of earthquakes in Nepal have reminded us, as if we needed reminding, of just how disastrous for human populations these events can be.
Nepal was a area where large earthquakes (around magnitude 8) are rare but not unexpected; the last was over 80 years ago.
The gaps between major earthquakes can be much longer and so any information which helps us to piece together a pattern of past seismic events is helpful.
One such study, published in the Bulletin of the Seismological Society of America (BSSA), looks at a part of New Zealand and provides just such and additional piece of the jigsaw.
Earthquake Archives and Earthquake History
Modern seismology is a relatively young science, and seismometers as we know them today have been around only since the beginning of the twentieth century. This is potentially problematic for those who want to use the record of past events in an effort to identify (if not predict) patterns of future earthquake behaviour or seismic hazard, because the cycle of continual breakup and convergence, which is the cause of the vast majority of the planet’s large earthquakes, takes place over millions of years.
Even taking into account written and oral histories, in most places the record goes back not much longer than hundreds of years — a particular problem when the time between major earthquakes on some stretches of plate boundary may be much longer.
One example is the Cascadia subduction zone, off western North America, which is known to have ruptured in a major tremor in 1700 but has not moved (significantly) since then. A second example, the focus of the recent study, is the Hikurangi Trench off New Zealand, which displays all the characteristics for a potential large — very large — tremor but along which no such tremor has been recorded. And the study proves that such large earthquakes have taken place in the past.
Identifying Prehistoric Earthquakes
The study, led by Dr Kate Clark of GNS Science, focussed on an area of the margin between the Pacific and Australian plates just off the southern part of New Zealand’s North Island and extending towards the South Island. Here, the Pacific plate subducts beneath the Australian plate creating the conditions in which large, so-called ‘megathrust’ earthquakes occur.
The world, and the Pacific Ocean in particular, has many of these zones and a map produced from the United States Geological Survey’s real time earthquake map and archive shows that earthquakes of M8.0 or larger are regular features of the Pacific margin. But the Hikurangi Trench has remained quiet.
By studying microfossils in deposits along the margin at different locations in the Big Lagoon in the southeastern Wairau River valley on South Island, Dr Clark and her team were able to identify soil horizons where significant and sudden movement had taken place in the past and, from these data, were able to identify not one but two major earth tremors. The earliest of these occurred between 880 and 800 years ago and the later one between 520 and 470 years ago.
Although it comes as no surprise to learn that the Hikurangi margin is capable of such large earth movements, the results of the study are significant. “We can now confirm that the southern Hikurangi margin does rupture in large to great earthquakes,” Dr Clark told Decoded Science. “This is something we have always assumed is a hazard to the central part of New Zealand and it is very important that it has now been verified.”
Significantly, the evidence from the deposits in the lagoon suggest that the earlier of the two tremors was a companied by a large tsunami which reached as far as 360m inland.
Subduction Earthquake — or Not?
The area of the study is at the south end of the subduction zone and other earthquakes, resulting from shallow deformation and crustal stresses, can also occur.
Subduction earthquakes, which invoke a (sometimes large) component of vertical movement are more likely to cause tsunamis. So how can we know, looking so far back, that the earthquakes identified are the result of movement at the subduction zone?
“The southern Hikurangi margin has many active upper plate (crustal) faults which does make searching for and identifying evidence of subduction earthquakes difficult,” explained Dr Clark.
“However, most of the upper plate faults have remarkably good paleoearthquake records on them, so, in the case of our stud,y we were able to compare the timing of our subsidence events with records of earthquakes on nearby faults to determine if the subsidence could have been caused by a crustal fault, and if not, it was more likely to have been caused by subduction interface rupture.”
Ancient Quakes: Lessons Learned
In some ways the findings of the study will be reassuring for seismologists and disaster planners alike, confirming as they do that the existing assumptions relating to earthquake activity are reasonably sound.
“The current National Seismic Hazard Model actually reconciles very well with our findings of two earthquakes in the past 1000 years so there is not a pressing need to revise our estimates of the southern Hikurangi margin seismic hazard at this stage,” said Dr Clark.