An artificial intelligence study has recently mapped over 54,000 earthquakes at Campi Flegrei, a caldera located near Naples, Italy. This research revealed a distinct ring-shaped fault that was previously unnoticed.
Campi Flegrei spans roughly 7 miles and contains neighborhoods built on its floor and rim. The project, led by Xing Tan, a doctoral researcher at Stanford University, used machine learning techniques to enhance our understanding of seismic activity in this region.
The ring fault, a circular fracture marking a collapsed volcanic basin, plays a crucial role in guiding earthquake locations. As described by the USGS, such faults can channel stress and affect volcanic activity. The newly created map outlines a thin but well-defined ring, linking recent seismic events to this fault structure. Tan noted, “Our Italian colleagues were surprised to see the ring so clearly,” demonstrating how this advance reshapes scientific interpretation of the volcano’s inner workings.
Traditionally, seismologists relied on phase picking, which involves manually identifying the first signs of earthquakes on seismograms. Now, machine learning can automate this process, enabling faster and more accurate recognition of seismic signals. This technology allows researchers to observe previously unrecognized patterns in underground activity.
The updated catalog shows two significant faults converging under Pozzuoli, raising alarms about the potential risk of stronger earthquakes in densely populated areas. A quake in the magnitude 5 range is possible. This region has a history of seismic risk; for instance, part of Pozzuoli experienced evacuations in the 1980s due to geological concerns. The current study has made visible the geological structures behind these risks for the first time.
The caldera experiences uplifts called bradyseism, with the average rise currently around 15 millimeters per month. This rising land adds stress to local faults and buildings. Italy’s Civil Protection agency has developed an emergency plan to prepare for potential eruptions. They’ve categorized areas into red and yellow zones to guide local officials in case conditions change.
Understanding the fault system helps engineers assess how strong potential earthquakes could be, which aids in building codes and safety drills. Improved knowledge about earthquake risks allows planners to allocate resources effectively, like placing evacuation shelters and emergency services.
The AI-driven model’s success depends on extensive data collection and ongoing adjustments. With each new earthquake tracked and analyzed, scientists can refine their predictions. While this technology doesn’t predict when eruptions might occur, it reveals where stress might accumulate, which is vital for communities living near the caldera.
For more in-depth insights, you can read the original study published in Science. Understanding these developments not only enhances academic knowledge but also empowers communities to prepare for seismic events that could impact their lives.
