shadow

Why tsunamis are so difficult to predict

Why We Wrote This

In the wake of Indonesia's devastating tsunami, questions have arisen about what could more have been done to save lives. But natural disaster detection and preparedness takes more than technological know-how.

Dita Alangkara/AP
Indonesian women stand on the rubble as they look for what's left from a relative's house at Balaroa neighborhood in Palu, Central Sulawesi, Indonesia, Oct. 11, 2018.

The first 20-foot wave struck just seven minutes after the earthquake. Two more waves arrived in the next eight minutes, devastating Palu, the provincial capital of Sulawesi, Indonesia, on Sept. 28. Some 2,100 lives have been reported lost and more than 270,000 people have been displaced.

In the aftermath of the 2004 Indian Ocean tsunami, which claimed 240,000 lives across 14 countries, scientists and officials in Southeast Asia and around the world have been collaborating to improve early warning capabilities. But outfitting an entire region with the sensors necessary for such a system takes more than technological know-how or political will. That challenge is keenly felt in Indonesia.

The archipelago nation is made up of more than 17,000 islands home to 261 million people, a heterogeneity that presents unique challenges in predicting tsunamis and alerting the public. In developing warning systems with limited resources, tough decisions need to be made about where to place sensors.

“We have limited resources so we have to do the best we can,” says Diego Melgar, an assistant professor of geophysics at the University of Oregon in Eugene, Ore., who designs tsunami early warning systems but was not involved in the Indonesian warning system. “And our knowledge of the world is still evolving and still imperfect. So unfortunately there are still surprises in store for us.”

According to UNESCO, InaTEWS, Indonesia’s tsunami early warning system, operates 170 seismic broadband stations, 238 accelerometer stations, and 137 tidal gauges. The agency also receives seismic data from 134 countries. But, as extensive as this infrastructure may sound, the system still has large gaps.

An unexpected disaster

Sitting at the end of a long, finger-like bay, the city of Palu is protected from the ocean in every direction except due north. Unfortunately for Palu’s residents, that’s where the earthquake’s epicenter was located, about 50 miles away and just over 6 miles beneath the surface. Like a funnel, the bay redirected the ocean waves directly into the city.

The quake came from a jolt on the Palu-Koro fault, a horizontal “strike-slip” fault in the Earth’s crust. It’s unusual, say geologists, that a quake on such a fault would generate a tsunami, which usually arise as a result of sudden vertical displacement of the water.

“People would not have been expecting a big tsunami from such an earthquake,” says Jane Cunneen, a geology research fellow at Curtin University in Bentley, Australia.

She notes that the wave could have been generated by a landslide triggered by the quake, and not the quake itself.

“The current tsunami warning systems, in any country no matter how high-tech, are not really set up to warn for that kind of event,” she says.

Professor Melgar points out that the Sulawesi tsunami arrived very soon after the earthquake, leaving little time for much warning. “It was just right there, and that’s why they had much less time,” he says “It was a very tragic situation.”

Detection is only half the battle. Authorities also need to find appropriate ways to educate the public and to convince them to heed warnings sent via cellphone alerts, TV messaging, or sirens. In this case, the earthquake knocked out the cellular network, which made it even more important for people to know to evacuate, says Dr. Cunneen.

Unlike the islands of Java and Sumatra, which experienced the the 2004 tsunami, Sulawesi hasn’t had a major tsunami since 1968.

“Part of the problem is keeping communities aware of things that happen so very infrequently,” says Cunneen, who worked with the Intergovernmental Oceanographic Commission of UNESCO to coordinate warning system development after the 2004 tsunami.

Alerts must also fit in with the local culture. In Bangladesh, where tsunami warnings are typically successful “people ride around on bicycles with bullhorns and sound the horns, and everybody knows I need to evacuate to higher ground,” says Cunneen. “I can say in downtown Perth, where I live, if someone rode around on a bicycle with a bullhorn you’d get a completely different response.”

Since September’s devastating tsunami, many in Sulawesi have raised concerns about the lack of warning sirens and elevated shelters. In any case, residents living near the ocean need to be trained to treat any earthquake lasting longer than 20 seconds as a warning, Cunneen says.

“Although the high tech warning system has its place,” she says, “really nothing will replace people knowing what they need to do when there’s a big earthquake like that.”

Jacob Turcotte and Eoin O'Carroll/Staff
of 5 stories this month > Get unlimited stories
You've read 5 of 5 free stories

Only $1 for your first month.

Get unlimited Monitor journalism.