Scientists are still gaining rich lessons from the Kaikōura Earthquake, two years after the 7.8 jolt shook the country with the equivalent energy release of 400 atomic bombs.
GNS Science earthquake geologist Dr Rob Langridge said the event, which struck just after midnight on November 14, 2016, was one of the most complex earthquakes observed anywhere in the world.
That was largely due to the sheer number of faults that ruptured in just a single event.
More than 20 faults were activated - 14 of which ruptured violently enough to displace land by more than a metre.
Nowhere was this effect perhaps more dramatic than along Marlborough's Kekerengu Fault, where the land was offset by as much as 12m.
In some places the fault was visible with raised-up folds of earth stretching across the countryside.
"Another lesson was that this complex earthquake propagated across from one region to the other – the earthquake began on faults in North Canterbury and jumped across to the north, triggering the major faults in the Marlborough region," Langridge said.
"The faults that ruptured formed an intricate network of northeast-striking faults that were interlaced with north-striking faults, which had a different mode of movement."
More broadly, the event had prompted scientists to consider the role of the subducted Pacific Plate beneath Marlborough.
Previously, this part of the plate was thought to have been fully "locked", or attached to the crust, but now it appeared that it was playing a big seismic role in this part of the South Island.
Soon after the event, Langridge and colleagues began gathering scientific data on fault ruptures, landslides, landslide dams, tsunami and liquefaction to help the community and to provide advice to the Government and councils.
"After a lot of data gathering, we set about publishing a series of scientific papers in several overseas and local journals, so that the knowledge gained has a lasting place," he said.
"Currently, we are seeking to understand the past history of faulting on some of these faults.
"So we have funding to go back to the Papatea, Kekerengu, Hundalee, Humps and Leader faults and the coastal story, to understand how these faults worked in previous earthquake cycles."
There was still much work to be done across a range of scientific areas, spanning from seismology to GPS-based geodesy.
This summer, his team would be excavating trenches across the Papatea Fault to understand when it moved in the past and how much it moved over many thousands of years.
This fault released an incredible amount of slip in the 2016 earthquake – scientists had since recorded up to 9m vertical movement and 6m of horizontal movement.
"This is really up there on the world stage as a huge on-land displacement," Langridge said.
"Understanding how long it takes to accumulate this amount of strain is the key to unlocking these faults."
What remained less clear was when the next big earthquake might strike.
GeoNet recorded more than 20,000 aftershocks in the year following the earthquake, with several thousand more recorded since.
However, just a fraction of those were greater than 3.0 in magnitude, and the number of aftershocks continued to decrease.
GeoNet's most recent earthquake shaking forecast – a statistics-based measure that calculated probability – gave a 15 per cent chance of an aftershock between 6.0 and 6.9 striking within the next three months, and a 46 per cent chance of this happening at some point over a 12-month period.
Langridge viewed the potential for future quakes as a case of human timescales being compared with geological ones.