Share this article:
Predicting the Volcano
We have recently made the first successful prediction of the end of an eruption, but we are not yet good at predicting the beginning of an eruption. This first successful prediction of the Bardarbunga volcano in Iceland has an important bearing on the history of science.
Throughout history of the human race, volcanic eruptions have been a source of wonder. Our association with volcanoes is ancient, as our ancestors evolved over two million years ago in the volcanic rift valley of East Africa, where their oldest remains are found today, often buried in volcanic ash. During the current century, the study of volcanoes has changed from a descriptive field of natural history to a truly quantitative science. As a measure of this trend, we now welcome the awarding of the highest recognition of excellence in earth science, the Vetlesen Prize, for the first time to a volcanologist, Stephen Sparks of Bristol University in 2015. We have indeed come a long way, from the initial description and recognition of volcanic rocks and processes, to the development of modern theories about how the earth works and what makes volcanoes erupt.
The basic method of science involves a number of steps that lead towards developing an understanding of how natural systems work. The first step in this methodology is the accumulation of data and observations about a natural process. The second is the development of a model, a hypothesis or a theory to explain the accumulated data. The third step is ideally to test the hypothesis by predicting the outcome of a process, on basis of the working model. If the prediction is accurate, then the credibility of the hypothesis is greatly strenghtened. Otherwise we can reject it. Prediction is really one of the ultimate goals of science and without the power of prediction, our scientific activities would really be of limited use. We are excellent at predicting a solar eclipe and we are getting very good at predicting the weather for the next couple of days, because we have great observations and excellent working models or theories that have been thoroughly tested. But when it comes to volcano prediction, we have not been doing as well. In fact, we have generally failed. In part this stems from the fact that we are dealing with a process (movement of magma or molten rock) that takes place largely deep within the earth and is therefore difficult to observe.
In late August 2014 a volcanic eruption broke out in north Iceland, which lasted for six months, until 27. February 2015. The eruption produced a large lava field known as Holuhraun. Simultaneously, the caldera of Bardarbunga volcano began to subside and there was clearly a connection between these events, 50 km apart. Working with my grandson, Gabriel Sölvi Windels, engineering student at the University of Reykjavik, I developed a mathematical prediction in October 2014 of the eruption behavior and we predicted that it would come to an end at the beginning of March 2015. We were off by only five days in this prediction of the end of the eruption.
The prediction was based on a model or a theory of the structure of Icelandic volcanoes and on observed changes in the Bardarbunga volcano during the eruption.
Volcanologist Haraldur Sigurðsson was off by five days when predicting the end of the eruption in Holuhraun. Starting on August 31, it lasted almost six months, or 181 days, concluding on February 27. Using a simple math formula with the help of his grandson, Gabríel Sölvi Windels, an engineering student at Reykjavík University, Haraldur calculated in October that judging by the decreasing subsidence of the Bárðarbunga caldera—the volcano which lies under Vatnajökull fed the eruption north of the glacier by an underground channel—and hence reduced pressure of magma, the eruption would end on March 4.
In 1978, my colleague Stephen Sparks and I put forth a theory about lateral magma flow in volcanoes, like Bárdarbunga, Askja, Grímsvötn and Krafla, but we didn’t have the opportunity to test it until now. The theory of lateral magma flow proposes a magma channel or a dike within the crust, connecting the large magma reservoir under the volcano to the eruption site at the surface. When the magma flow began in Bardarbunga on August 16. 2014, with magma reaching the surface on August 29, there was a question whether the theory could be put to the test. We had data about the subsidence of the caldera of the volcano and it followed a certain process with unbelievable accuracy. This is data posted on the web site of the Icelandic Meteorological Office. It was not a linear subsidence, but a smooth curve, which with time became flatter. And so a formula could be made: once the curve would become horizontal, the eruption would stop, as the curve indicates the outflow of magma.
We calculated when the curve would become horizontal and came up with the first week of March in 2015 as the end of the eruption. And we stood by this production, which we issued in October 2014. The curve proved remarkably even, there were almost no daily fluctuations, it was almost perfect. It was simple in this case because the magma chamber is probably very large, so that outside disturbances have no effect on the model. It is a method which can no doubt be used on other eruptions in the future.
Bárðarbunga is one of the biggest volcanoes in Iceland and has produced massive eruptions in the past. In fact, it has produced the largest known lava flow on Earth in the past ten thousand years. What lies ahead in this huge volcano is unknown. Will the magma chamber be replenished, and will that result in inflation of the volcano? When will the chamber again reach a critical pressure, that can lead to an eruption?
The Encyclopedia of Volcanoes, 2nd Edition is available for purchase on the Elsevier Store. Use discount code “STC215” at checkout and save up to 30% on your very own copy!
About the Author
Haraldur Sigurdsson is emeritus professor at the Graduate School of Oceanography, University of Rhode Island in the United States of America. He has worked on volcanic processes and the geochemistry of volcanic rocks for over fifty years. His studies have in part been focused on the impact of volcanic activity on human populations, especially his work on Vesuvius in Italy, Tambora in Indonesia, El Chichon in Mexico and studies of deadly gas bursts from Cameroon crater lakes.
He has also studied the global environmental effects of meteorite impacts, such as the one that marks the Cretaceous/Tertiary boundary. He is Editor in Chief of The Encyclopedia of Volanoes, 2nd Edition which published in March 2015. Haraldur is currently director of the Volcano Museum in Stykkisholmur, Iceland.
Earth & Environmental Science
The fields of Earth science, planetary sciences, and environmental science encompass disciplines critical to the future of our world and its inhabitants. Our well-being depends on a thorough understanding of air and water resources, soil chemistry, atmospheric dynamics, geology, and geochemistry, along with a myriad of other aspects of the environment we live in. Elsevier supports the efforts of researchers and scholars in these areas with content that meets their cross-disciplinary needs: journals, books, eBooks, and online tools that span computer science, chemistry, energy, engineering, biology, agronomy, ecology, environmental impact and many other topics fundamental to the study of our world. Learn more about our Earth and Environmental Science books here.