October 23, 2003
A weekly feature provided by scientists at the Hawaiian Volcano Observatory.
Can we predict the next "big" one?
On November 29, 1975, the latest magnitude-7 earthquake in Hawai`i struck near Kalapana, causing a tsunami that killed two campers at Halape and resulting in $14 million (2003 dollars) in damage. It was the only magnitude-7 earthquake to hit the Big Island during the 20th century.
The earth experiences an average of one magnitude-8 and 18 magnitude-7 earthquakes annually. On September 25, a magnitude-8.1 earthquake off Hokkaido, Japan, generated widespread media attention, injuring nearly 600 people and causing extensive damage, landslides, and a brief tsunami alert here in Hawai`i.
Many more earthquakes, about 150 per year, occur at the magnitude-6 level; these are violent enough to cause substantial damage in populated areas. An example is the 1994 Northridge (M6.7) earthquake, which killed 60 people and injured more than 7,000 others in the Los Angeles area.
Scientists for some time have understood that most large earthquakes occur on long stretches of faults that ring the Pacific Ocean, where ocean floor is subducted (or pulled beneath) continental plates. On the Big Island, however, the occasional large earthquake is triggered by tectonic settling of the island's volcanoes.
An educated guess as to when and how large the next "big" earthquake will be on a given fault depends on a complex combination of the history of previous earthquakes, the size of the fault, mechanical properties of the rocks, and amount of strain accumulation. Often, adequate knowledge about one or more of these factors is lacking. In places where a dense network of monitoring instruments exists, such as in the Los Angeles basin, we are still discovering active faults which may interact with, and affect behavior on better-known fault systems.
Many psychics, self-proclaimed "scientists", and other well-meaning citizens claim success at predicting significant earthquakes. They occasionally substantiate their methods by quoting high success rates at previous predictions. Many of their predictions are actually made for high-probability events. They encompass a region or time frame so wide, or include events of such low magnitude, that the chance of an earthquake happening can exceed the chance of one not happening. Lack of rigid criteria for evaluating predictions can make it easy to claim success. In seismically active regions, earthquakes occur daily. Here on the Big Island, we usually experience at least one magnitude-2 earthquake a day. Most of these are posted in near-real time on our web page http://hvo.wr.usgs.gov/earthquakes/new.
To be declared scientific, an earthquake prediction must include the specific location, time, magnitude, probability, and method used to make the prediction. The methodology is critical, because it must be testable and give reproducible results in order to be deemed scientific. Vague and unscientific predictions cannot be argued or taken seriously, as there is always a possibility of the event's happening by chance.
At present, scientists have found no reliable method for making useful and accurate earthquake predictions for specific locations. Obviously, there are tremendous societal ramifications for issuing general or vague earthquake predictions.
Research continues towards a better understanding of how earthquakes work. This has been possible only with recent improvements in computers, instrument design, and station coverage. We can now obtain detailed information about ground shaking from earthquakes at many locations, which will improve earthquake hazard assessment and result in better probability estimates of large earthquakes on many of the most dangerous fault systems. Such estimates are already calculated for a number of fault segments in California. This falls short of earthquake prediction, but it delineates high-hazard areas where mitigation planning is needed.
The behavior of a fault is so complex that earthquake predictions to the exact hour or even day, may never be possible. Even if prediction were possible, we cannot stop the earthquake from happening. As Big Island residents living in earthquake country, we must always be prepared for the possibility of a destructive earthquake, much as we do for natural disasters such as hurricanes or flash flooding.
Eruptive activity at the Pu`u `O`o vent of Kilauea Volcano continued unabated during the past week. Surface flows are mainly confined to the upper areas of the Mother's Day flow above Pulama pali where perched lava ponds in rootless shields have formed. Three distinct clusters of breakouts occur on Pulama pali along the western Kohola arm of the Mother's Day flow. The distal end of activity is down to the 600-ft elevation, a short distance above Holei pali. The east side of the Mother's Day flow, and the coastal flat below Paliuli, continue to be dark. No lava is entering the ocean.
Three minor earthquakes were reported felt in the week ending on October 23. A resident of Hilo felt the earth move at 10:21 p.m. on Saturday, October 18. The earthquake responsible was a magnitude-2.8 event located 10 km (6 mi) northwest of Ka`ena Point at a depth of 10 km (6 mi). A magnitude-2.4 earthquake originating 5.5 km (3.3 mi) beneath Kilauea Caldera was felt in Mauna Loa Estates at 6:41 a.m. on Sunday, October 19. At 3:53 p.m. on October 21, a magnitude-2.6 earthquake was felt in Volcano. The temblor was located 9 km (5.4 mi) northwest of Volcano at a depth of 24 km (14.4 mi).
Mauna Loa is not erupting. The summit region continues to inflate. Seismic activity remains low, with three earthquakes located in the summit area during the last seven days. Visit our website (hvo.wr.usgs.gov) for daily volcano updates and nearly real-time earthquake information.
Updated: October 31, 2003 (pnf)