Top storiesNew ZealandPoliticsBusinessEntertainmentSportsWorld

Real earthquake far out into ocean sets off reports of North Island 'ghost' quakes

Tuesday, 7 March 2017

The real quake was centred 765km northeast of Whakatane at a depth of 446km with a magnitude of 5.7.
The real quake was centred 765km northeast of Whakatane at a depth of 446km with a magnitude of 5.7.

A deep, strong earthquake well to the northeast of the country has set off three reports of ghost quakes around the central North Island.

The real quake was centred 765km northeast of Whakatane at a depth of 446km and had a magnitude of 5.7. GeoNet received 169 reports from people who said they felt it. The ghost quakes, about 8.30am on Tuesday, had magnitudes from 4.4 to 4.7.

GNS Science seismology scientist John Ristau said measures had been taken to weed out ghost quakes from the GeoNet system. 'It does catch them sometimes but it doesn't always do it.'

When a person looked at the data they could immediately tell which quakes were real and which weren't, 'but it's a lot more difficult to teach a computer to do that,' Ristau said.

READ MORE: Large magnitude 6.6 shake triggered 'ghost quakes' in the North Island, but why?

He had seen the ghost quake reports coming in and, looking at the seismic data streaming into GeoNet, had been able to delete them quickly.

'With a bit of experience you look at it and you know what it's going to be just by looking at the data,' Ristau said.

Ghost quakes were particularly a problem with large, deep quakes to the north of the country, as happened on Tuesday.

As soon as seismic waves started arriving at monitoring stations at the north of the North Island, the system started trying to find exactly where the event happened.

'Because of the complexity of the tectonics to the north, our automated system sometimes has issues locating these earthquakes properly.'

If there were two quakes, close together in time but geographically far apart, algorithms in the system tried to distinguish between the pair, but at times it could mean the system thought one quake was more than one.

Another issue with the large, deep quakes was that they had sharp, well-defined 'S (secondary) waves', arriving after the 'P (primary) waves'. In some cases the system could get confused and think the S waves were from a separate earthquake, Ristau said.

The large, deep quakes, even though centred well offshore, could be felt over a wide area. 'Particularly if you're in a multi-storey building  and you're sitting quietly, the building might sway a little bit from it.'

Some people might think they had felt a quake because it was reported on GeoNet around the same time they felt some shaking from another source, such as a truck going by or strong wind shaking a building.

'We do have people after the fact … for large earthquakes who go through the felt reports and try to weed out the ones that are obviously inaccurate,' Ristau said. The number of inaccurate reports was 'pretty small'.

Information from felt reports was helpful in understanding how earthquakes affected different areas.

'I think for the most part the vast majority of people are accurately describing what they felt … and describing as best they can how strongly they felt the shaking and what damage occurred if there was any,' Ristau said.

There were some reports that obviously weren't accurate, either because people were mistaken, or maybe because they thought they might mess with the system. 'But we usually pick those up pretty quick.'