Wellingtonians are being warned to expect more vigorous shaking in the next few days.
Scientists last night said they were surprised by yesterday's 6.6 quake, as it was unusual to see a doublet of similar-sized jolts.
But while they can't say if or when a third plus-6 magnitude quake will strike, they are warning that decent-sized aftershocks will continue in the short term.
"We are definitely expecting it to be quite vigorous in the next couple of days, and then it will die off," GNS seismologist Dr Caroline Little said last night.
They also say there is no certain way to tell that The Big One won't strike the region.
Yesterday's 2.30pm quake struck near Lake Grassmere, Marlborough, centred 8km under the northeast of the South Island.
Scientists have identified it to be a "strike-slip earthquake", where each side slides past the other without uplift and down-thrust, and the same kind of quake as the magnitude 6.5 quake that struck nearby on July 21.
Dr Little said scientists had expected that the initial quake would have been followed by weaker and weaker aftershocks. "To have two at the same magnitude is not the most common occurrence," she said.
"This wasn't outside the realms of possibility - but it definitely wasn't the most likely scenario."
The latest quake would lead scientists studying the sequence of quakes in the region, where more than 3,000 have been recorded over the past month, to "refine" their investigations.
Dr Little could not say for how much longer the pattern would continue, but despite yesterday's event she thought the size of the aftershocks would continue to decrease.
Professor Euan Smith, of Victoria University's School of Geography, Environment and Earth Sciences, said there were many faults surrounding the area where yesterday's quake hit.
Professor Smith said it appeared it struck on a subsidiary fault southwest of the Awatere Fault, which runs through the small town of Seddon and produced a large quake in 1848.
He said it could take a long time for the earth to readjust itself, and the time delay since the July 21 quake was "by no means unprecedented".
"There's absolutely no doubt these quakes have been pushed along by what happened in July. The reality must be that this fault was loaded and just about ready to go, and why we had three to four weeks between that Sunday in July and today, we can't really explain."
Probability models used by GeoNet were based on statistical data collected from quakes already recorded, rather than physical tests.
"An extremely good example came from the Darfield (September 2010) quake and the February 22 quake. Three days after Darfield, there was a magnitude 5 quake under the CBD of Christchurch, very close to where the February 22 quake happened 170 days later," Professor Smith said. "Why we had to wait 170 days, no one can say."
But scientists hoped the field could be advanced by a forecast testing centre set up under GNS Science, allowing researchers freedom to submit their models to testing centres and have them rigorously and transparently tested.
As for "the big one" striking the capital, there was as yet no sign of major quakes on the Wairarapa Fault, which produced New Zealand's biggest ever recorded quake when it ruptured in 1855 and heaved Wellington more than a metre out of the ocean.