Seismologists at Caltech developed a method for the detection of natural disasters — namely earthquakes — that leverages the capabilities of underwater telecommunication cables already in operation. The method, which does not require additional equipment or instrumentation, analyzes light that travels through these so-called lit fibers to monitor earthquakes and ocean waves, converting the fibers into geophysical sensors. Previous efforts based on optical fiber infrastructure for the study of seismicity have used “dark” fibers, meaning those that are out of active use. Scientists have also relied on other scientific instruments beyond those that classify as optical. Using optical fiber infrastructure already in place to gauge seismicity (the occurrence/frequency of earthquakes in a region) is cost-effective and resourceful because more than 70% of Earth is covered by water. It is both expensive and onerous to install and actively monitor submerged cables, though they serve as a backbone of international communication. In the new method, lasers send pulses of information through glass fibers bundled within host cables. That action delivers data to receivers at rates that exceed 200,000 km/s. To monitor the polarization of the laser light that travels through the fibers, the scientists operating the transmission process intentionally controlled the direction of the electric field to allow multiple signals to travel through the same fiber at once. At the receiving end of the system, devices check the polarization state of each signal to determine how, and how much, it changed along the path of the cable. This ensures that the signals do not mix as they are in motion. Though factors such as changes in temperature and weather are apt to influence the polarization of light traveling through land-based fiber optic cable networks, the deep-ocean environment remains nearly constant, without disturbances. In testing, the scientists focused on a submarine fiber optic cable that stretches from California to Chile, the Curie Cable. They found that although the change in polarization from one end of the cable to the other remained stable over time, polarization changed suddenly and significantly during earthquakes and storms that produced large ocean waves, yielding acquirable data. The researchers ultimately measured polarization as often as 20 times per second — meaning that the system could deliver a warning to those in a potentially earthquake-affected area within seconds of an earthquake striking near that area. Currently, when earthquakes occur offshore, it can take minutes for seismic waves to reach land-based seismometers. It takes even longer to verify tsunami waves. In a study introducing the underwater telecommunication approach, the researchers reported testing for a nine-month period, in which they detected about “20 moderate-to-large” earthquakes along the Curie Cable. They did not detect a tsunami in that period, though they did detect polarization caused by ocean swells. These changes, the team believes, stemmed from pressure changes along the seafloor as high-powered waves moved past the cable. If correct, it would mean the method successfully showed its ability to detect ocean waves, and that it is plausible to think the approach could also apply to the detection of tsunami waves. The Caltech team is now developing a machine learning algorithm aimed at determining whether earthquakes or ocean waves (as opposed to some other change to the system, such as a ship or crab moving the cable) produced the detected changes in polarization. The researchers expect they will be able to automate the entire detection and notification process to provide additional information, beyond the data already collected by the global network of land-based seismometers and the buoys in the Deep-Ocean Assessment and Reporting of Tsunamis (DART) system, operated by the National Oceanic and Atmospheric Administration’s National Data Buoy Center. The research was funded by the Gordon and Betty Moore Foundation; was led by Zhongwen Zhan, assistant professor of geophysics; and was published in Science (www.doi.10.1126/science.abe6648).