Researchers are embarking on an ambitious experiment to test real-time eruption forecasts for Axial Seamount, an active submarine volcano situated off the coast of Oregon. This volcano has been erupting often, with previous eruptions carried out in 1998, 2011 and 2015. Axial Seamount is one of the most actively monitored volcanoes in the world. It has one of the largest, most sophisticated monitoring networks in the world that continuously monitors and evaluates seismic activity and the surrounding geology.
The monitoring network comprises four cabled bottom pressure recorders that measure seafloor uplift and a series of seismometers that detect earthquake activity. These instruments, shown in Figure 1, are in use in the Ocean Observatories Initiative (OOI) Regional Cabled Array (RCA). The information they generate is incredibly important to that ongoing experiment. This experiment is to produce long-range (monthly) forecasts for future eruptions at Axial Seamount.
Prior eruptions, such as the 2015 eruption, were forecast on the order of seven months in advance. We made these predictions solely based on the patterns that we saw in geodetic data. Most recently, scientists have found that inflation at Axial Seamount has exceeded maximum inflation levels displayed in the 2015 eruption. This finding has critical implications for when the next eruption might occur. The modeling work had previously forecasted an eruption as early as 2025. Recent congressional analyses indicate a significant likelihood of this event happening no later than mid-to late-2026. This is driven in part by a dramatic slowdown in the detected seafloor uplift.
This experiment has an ambition to test those two big hypotheses. The first proposes that volcanic eruptions may be forecasted in real time through careful evaluation of specific patterns that indicate an impending catastrophic failure of the volcano’s system. Our second hypothesis is that we can better predict the timing of these eruptions. With probabilistic approaches, we can still be very precise in our predictions.
Bill Chadwick from Oregon State University and Scott Nooner from the University of North Carolina at Wilmington lead this study. Their rigorous approach lays the groundwork for advancing our understanding of volcanic unrest and eruption initiation. They intend to improve public safety and scientific understanding by improving forecasting techniques.
Progress and developments from this experiment are documented on the blog dedicated to Axial Seamount, where researchers share insights and updates regarding their findings.