A severe earthquake struck south of the Alaska Peninsula on Monday. The 7.6 magnitude earthquake is currently the third largest earthquake in 2020. The good news with this earthquake is that the impact has been minimal. But all great earthquakes have lessons to teach and this one was no exception.
At first glance, the earthquake was no surprise. It was located south of the Alaska Peninsula, 60 miles southeast of Sand Point, and only 52 miles southwest of the largest (to date) earthquake of 2020, July 22nd M7.8. The July 22nd earthquake caused 20 aftershocks of magnitude 5 and larger prior to last week, including two 6.1. Technically, the 7.6 is another aftershock – it is located within the aftershock zone and is clearly associated with voltage changes compared to the 7.8.
However, the 7.6 differs from most aftershocks in several ways. It's almost as big as the main quake. The rule of thumb for seismology is that the largest aftershock is about one unit less than the main earthquake. It was a different bug with a different bug slip than the bug that caused the 7.8. And almost three months have passed since 7.8, and the largest aftershocks are most common in the first few weeks.
Rules of thumb have little weight in seismology. It is not uncommon for multiple earthquakes of nearly similar size to occur in a short amount of time and with different but related faults. In 1992 we saw 7.2, 6.5 and 6.6 within 18 hours, all with different errors. The observation of earthquake activity in the western US this year shows that months can elapse between significant earthquakes that are likely to be regionally related. But I have to admit, the earthquake on Monday came as a real surprise to me.
The United States Geological Survey quickly analyzes any significant earthquakes and typically finds a location, time of origin, magnitude, and likely fault within 15 to 20 minutes of the occurrence. In another half an hour, a model of the probable failures is created and an estimate of the vibration resistance and impact is made. It is impressive how quickly so much detailed information is available. As early as 1992, it took hours until the 7.2 was reached.
Seismologists describe errors through a focus mechanism. Informally, we call them beach balls because they look a little like a four-colored beach ball, with dark areas representing areas around the earthquake source where the rock was compressed and light areas where they stretched. It doesn't take long to see the three main types of flaws in the focus mechanism, and on Monday 7/6 it showed the classic four-quadrant pattern of a strike-slip earthquake. USGS seismologists build models of possible fault slip and fault orientations and see how well the models match the observed seismic records. In this case, the best fit was a strike-slip bug facing north-northwest.
The surprise is that the north-northwest is perpendicular to the Alaska-Aleutian subduction zone. The July 22nd earthquake had a thrust fault that appears to coincide with the main subduction interface. This is what I would expect a well mannered upper M7 or M8 to look like. What was a major strike event that went against the grain here, so to speak? It is another lesson that subduction zones are not easy. There is evidence of an offset in the seismic trends just where the 7.6 was centered, a large crack in the plane of the subduction zone. The 7.8 added stress and the 7.6 released it.
As on July 22, the M7.6 triggered a tsunami warning. Five minutes after the earthquake, the National Tsunami Warning Center (NTWC) issued an initial bulletin placing the Alaska Peninsula, the Eastern Aleutian Islands and Cook Inlet in warning status. The tsunami threat to the west coast of the US and Canada is currently being analyzed. It took them more than two hours to explain that the tsunami was unlikely to hit us.
I assumed that each tsunami would be smaller than last July. The greater the strength, the larger the tsunami (usually), and thrust earthquakes are more likely to produce measurable tsunamis than strike-slip earthquakes. The pushing creates a vertical deformation of the seabed and the slippage is mainly horizontal. The obvious conclusion is that the 7.6 tsunami should be much smaller than the one caused by the larger quake.
My conclusion was wrong. The July M7.8 caused a 9-inch tsunami at Sand Point. It was not observed on any other tide meter on the coast. Monday 7.6 reached 2.3 feet at Sand Point and was recorded on 13 other tide gauges in Alaska, Hawaii, and Crescent City, where it reached 8 inches. I wasn't the only one who was surprised. The scientists at the Pacific Tsunami Warning Center issued an initial bulletin in which they stated that the earthquake posed no tsunami threat to Hawaii. Five hours and 40 minutes later, they changed the status to an advisory after waves were recorded on Hawaiian tide gauges.
I have no answer as to why the tsunami was so much bigger this week. The good thing is that it did no harm and again provides research fodder and a better understanding of tsunami generation. But it also provides a much-needed opportunity to assess the warning system and make sure it is working better when the next noxious tsunami hits us.
Monday's earthquake shows why tsunamis from Alaska are my least favorite far-field events. It only takes four to five hours to reach us, and if there is no deep sea sensor near the source, it will be difficult for the tsunami warning centers to make a definitive call. By the time NTWC called on Monday, we would have had less than two and a half hours to coordinate an evacuation if necessary.
It also highlighted my great anger with the U.S. tsunami warning system – two centers and five different types of bulletins. Each bulletin is for a different purpose, but it is difficult for most people to understand why one states a warning and another does not state a threat. We can do better.