Earthquake forecasts

Looking at all the seismic activity in the aftershock area of November’s Kaikōura Earthquake, the expected numbers of earthquakes contine to drop. We are now updating the forecast every three months. There is now a 19% chance of one or more M6.0-6.9 earthquakes occurring within the next three months; the annual forecast for one or more M6.0-6.9 earthquakes has decreased to 53% from our last forecast (19 October 2017).

What has happened so far

There have been over 19,000 earthquakes since the M7.8 Kaikōura earthquake stopped shaking our islands (we ran the numbers on the 13th of November 2017).

Aftershock Forecast starting on 14 November 2017

For the anniversary edition of the forecast, we have revised the forecast model slightly to make it consistent with the Kaikoura hazard model (the details can be found here). we will now provide future forecast up-dates every three months.

While no one can yet scientifically predict earthquakes, we can provide forecasts of future aftershocks (based on probabilities), as well as some scenarios from what is most likely to happen to what is very unlikely, but still possible. Most earthquake aftershock sequences decay (i.e. the number of earthquakes generally decreases) over time, with spikes of activity that can include larger earthquakes.

Average number of M5.0-5.9 Range* of M5.0-5.9 Probability of 1 or more M5.0-5.9 Average number of M6.0-6.9 Range* of M6.0-6.9 Probability of 1 or more M6.0-6.9 Average number of M≥7 Range* of M≥7 Probability of 1 or more M≥7
within three months 2.4 0-8 81% 0.2 0-1 19% 0.02 0-1 2%
within one year 8.5 2-18 >99% 0.7 0-3 53% 0.06 0-1 6%

Forecast for rectangular box (see map below) with the coordinates -40.7, 171.7, -43.5, 171.7, -43.5, 175.5, -40.7, 175.5 at 12 noon, Tuesday, 14 November; 95% confidence bounds.

The aftershocks of the magnitude 7.8 Kaikoura earthquake are mostly occurring throughout a broad area from North Canterbury through to Cook Strait that surrounds the faults that ruptured in that earthquake, although a few have occurred in the lower North Island. We forecast aftershock probabilities for the area in the red box on the map below. The area near the centre of the box (around Kaikoura) is more likely to experience felt aftershocks than areas towards the edge of the box. See the MMI map below for more information on the forecast shaking for the Wellington area. Earthquakes can and do happen outside this box but the box represents the most likely area for aftershocks in this sequence.

For example, there is a 19% chance of one or more M6.0-6.9 earthquakes occurring within the next three months. We estimate there will be between 0 and 1 earthquakes in this magnitude range within the next three months.

The current rate of magnitude 6 and above earthquakes for the next three months is about 2 times larger than what we would normally expect for long term seismicity represented in our National Seismic Hazard model. As the aftershock rates decrease, this difference will decrease as well.

Aftershock shaking forecasts

We have also calculated the probability of strong earthquake shaking from aftershocks over the next year (starting 14 November 2017). On the Modified Mercalli Intensity (MMI) scale Strong shaking is classed as MM6, and severe as MM7. The MMI scale is different to earthquake magnitude – it describes the intensity and impacts of the shaking, which depend on the magnitude of the earthquake, how far away the earthquake was and the type of ground you are on. At MM7 intensity shaking levels it is difficult to stand, furniture and appliances move, contents are damaged, there is minor building damage and liquefaction can occur in susceptible sediments.

The maps show the probability of MM6 and MM7 shaking within the aftershock region, which includes Wellington. Over the next year the probability of MM7 shaking around the wider Kaikoura/northern area is around 20%. In comparison, the probability of MM7 shaking in the Wellington area is around 3% (dark blue) in the next year. While this probability is considerably lower in Wellington than in the areas around Kaikoura, it is possible for shaking similar to what occurred during the mainshock to happen again in Wellington.

Probability of damaging shaking MM7 in the next year

Probability of damaging shaking MM7 in the next year

Probability of damaging shaking MM6 in the next year

Probability of damaging shaking MM6 in the next year


The scenarios specifically address the probabilities of what we might see happen within the first year following the mainshock, they were estimated on 5 December 2016 and reflect the probabilities calculated at the time. Comparing these probabilities with the ones in the table you will find that the annual forecast has dropped significantly, e.g. from a probability of a 98% to now 53% that there will be at least one aftershock of magnitude 6.0-6.9 in the next year. The scenarios cover a wider geographic area than the aftershock probability forecast area. The probability numbers in the table above differ to the scenarios. This is because they were estimated at a different time in the aftershock sequence, and we have used new information we have gathered from the slow-slip events, and their potential impact on the plate interface and other faults, to help define our probabilities in scenario three.

On the anniversary of the workshop GNS Science is holding an international workshop to reassess the impact of the slow slip events on future large earthquakes.

Scenario One: Likely (approximately 70% within the year to December 2017)

The most likely scenario is that aftershocks will continue to decrease in frequency (and in line with forecasts) over the next year and no aftershocks of magnitude 7 or larger will occur. Felt aftershocks (e.g. over magnitude 5) can occur in the area from North Canterbury to Cape Palliser/Wellington. It's very likely (98% within the year to 5 December 2017) that there will be at least one aftershock of magnitude 6.0-6.9 regardless of there being a larger (magnitude 7.0+) earthquake.

Scenario Two: Unlikely (approximately 25% within the year to December 2017)

An earthquake smaller than the mainshock and between magnitude 7.0 to magnitude 7.8 will occur. There are numerous mapped faults in the North Canterbury, Marlborough, Cook Strait and Southern North Island areas capable of such an earthquake. It may also occur on an unmapped fault. This earthquake may be onshore or offshore but close enough to cause severe shaking on land. This scenario includes the possibility of an earthquake in the Hikurangi Subduction Zone. Earthquakes originating from here or in the Cook Strait have the potential to generate localised tsunami. The Hawke’s Bay earthquake sequence in 1931 provides an analogy to scenario two, as a magnitude 7.3 aftershock occurred approximately 2 weeks after the initial magnitude 7.8 earthquake.

Scenario Three: Very unlikely (5% within the year to December 2017)

A much less likely scenario than the previous two scenarios is that recent earthquake activity will trigger an earthquake larger than the magnitude 7.8 mainshock. This includes the possibility for an earthquake of greater than magnitude 8.0, which could be on the plate interface (where the Pacific Plate meets the Australian Plate). Although it is still very unlikely, the chances of this occurring have increased since before the magnitude 7.8 earthquake, and have also been also been slightly increased by the slow-slip events.

Initially our scenarios covered what might happen over the first 30 days, but were shifted to cover what might happen over the year (to December 2017). This is because the aftershocks were generally becoming smaller and less frequent (decaying) over time, and this lower aftershock rate increases the uncertainty of what might happen over shorter time periods. The change in forecast does not hugely affect the scenarios and we will continue to review them. While we will continue to update the aftershock probabilities regularly, we will not update the scenarios as often.

Can't get enough technical information? Here's the fine print on how we model aftershock probabilities.