Satellite image of Anak Krakatau, Indonesia, with one slope covered in sediment.
The landslide and tsunami associated with the 2018 eruption of Anak Krakatau, Indonesia, were responsible for more than 400 deaths. Credit: ESA

On 22 December 2018, an eruption rocked Indonesia. Anak Krakatau, a volcano that sits in the Sunda Strait between the islands of Java and Sumatra, let loose. The eruption was accompanied by a massive landslide that triggered a tsunami and inundated nearby coastlines. A rock concert was interrupted when a wave crashed through the tent and swept the band off the stage. All told, more than 400 people died, around 14,000 were injured, and 33,000 were displaced.

Eruptions are nothing unusual in this country with more than 100 volcanoes. “Every day we heard that, hey, another volcano is erupting,” said Edo Nurshal, a research assistant at the Bandung Institute of Technology in Indonesia. But the 2018 eruption of Anak Krakatau was among the most damaging eruptions in recent history, and scientists soon began to ask how this catastrophe occurred.

Had the eruption triggered the landslide that caused the tsunami? Or had the landslide come first, triggering both the eruption and the tsunami? Oxford University Ph.D. student Kyra Cutler and her colleagues analyzed geological samples and remote images to tease apart the sequence of events. Experts say their results point to a type of volcanic hazard that’s potentially dangerous and difficult to predict.

History Repeats Itself

A penchant for destruction runs in Anak Krakatau’s family. The volcano’s name translates to “child of Krakatau,” denoting its relationship with one of the most deadly volcanoes in human history. When Krakatau famously erupted in 1883, it triggered a tsunami about eight stories tall that killed around 36,000 people. After the eruption subsided, magma continued to push its way out of Krakatau’s crumbled remains, forming a new volcano—Anak Krakatau—on the edge of the basin carved out by the eruption.

Anak Krakatau had been undergoing a slow trickle of eruptions for about 6 months prior to its collapse in 2018. “On the day of the collapse, it was still kind of erupting as usual,” Cutler said. “There wasn’t really anything unusual about the volcanic activity.” This changed abruptly just before 9:00 p.m., when about a quarter of a square kilometer of land slipped off the side of the volcano and into the ocean while magma spewed from the volcano’s vent.

“At first we thought that, OK, the eruption triggered the collapse. Actually, the condition is vice versa.”

After the eruption, Cutler’s colleagues quickly headed to Indonesia, where they collaborated with local scientists to collect samples of the rubble. The researchers drew on previous knowledge of the area’s geology to find layers that had accumulated before, during, and after the collapse. The team noticed that the layer of rubble that formed just after the landslide contained crystals of magma that were much smaller than the crystals dating from before the landslide. Combining this information with other observations, they concluded that the landslide acted like a pressure release valve that allowed magma to spew forth rapidly, escaping the volcano before crystals had had a chance to grow.

“At first we thought that, OK, the eruption triggered the collapse,” said Nurshal, who analyzed remote images that complemented the geological results. “Actually, the condition is vice versa.”

Monitoring for Future Collapses

“They did some very careful work,” said geophysicist Emilie Hooft, a professor at the University of Oregon and an expert in the transport of magma from mantle to surface. “I think it’s a very plausible sequence of events.” She added that other volcanic eruptions provide evidence that landslides can trigger eruptions. For example, when Washington’s Mount St. Helens erupted in 1980, an initial small eruption triggered a landslide that then led to a larger eruption.

Some volcanologists had theorized that Anak Krakatau might be headed for a similar fate. In 2012, Thomas Giachetti, now a professor at the University of Oregon, and his colleagues published a paper outlining how such a collapse might occur and how big the ensuing tsunami would be. Their model ended up being very close to the actual events of December 2018.

When Giachetti learned that his model had become reality, he said he felt “extremely weird.” In the years since the eruption, he’s put some thought into what kind of warning system could have saved lives when Anak Krakatau collapsed. Unfortunately, the answer is not straightforward. “There was no easy, cheap instrument that we could have put there” to monitor for the collapse, he said. To gain a full understanding of when similar events might occur, he thinks health officials will need to combine data from many instruments, including satellites and seismometers.

Indonesians are likely to need such a system. Nurshal and his colleagues are monitoring the remnants of Anak Krakatau as the volcano regrows. Already, it’s developed a dangerously steep slope. He hopes that public health officials will heed warnings from him and other researchers. “If we don’t do anything,” he said, “a similar thing will happen again.”

—Saima Sidik (@saimamaysidik), Science Writer

Citation: Sidik, S. (2022), Which came first, the eruption or the landslide?, Eos, 103, https://doi.org/10.1029/2022EO220094. Published on 25 February 2022.
Text © 2022. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.