sauce: Geophysical Research Journal: Machine Learning and Computation
A single seismometer is often insufficient to reliably detect human activities such as earthquakes or underground nuclear tests. Rather, researchers combine measurements from seismometers distributed over a small geographic area to increase the reliability of their analysis. Artificial intelligence (AI) can combine measurements from multiple sensors more efficiently than traditional techniques, allowing weak seismic signals to be detected more reliably, a new study says. Köhler et al. is shown.
The researchers leveraged 30 years of readings from seismic arrays run by the Norwegian research foundation NORSAR and other operators to train an AI model in three different ways to detect seismic signals. We first trained on data from one individual station at a time and then applied the model to combine the results from each station. We then used classical techniques to combine signals from multiple sensors in the same array and trained a model based on these combined signals from multiple stations. And third, we fed the model all the data from all the array stations and let it decide how to combine them.
The second method (combining signals before training) amplified weak signals and provided the most accurate signal detection of all three methods. On the other hand, the third model (let the model decide how to combine the station data) was the most computationally efficient strategy and fell between the other two methods in terms of accuracy.
Given the need to balance accuracy and speed, researchers recommend letting the model decide how to combine data when doing real-time monitoring, but combining data before and after model application when a slower approach is acceptable.
However, this model does not generalize well to regions other than the one it was trained on. The reason is that a regionally limited training dataset was used. Training on global data is expected to improve results. The problem mainly occurred in S waves, on the other hand P Generalization of wave detection was not an issue.
Overall, the results show that AI can improve earthquake monitoring by allowing researchers to detect weak signals from earthquakes, underground nuclear tests, and other seismic activity that are difficult to identify using other methods. (Geophysical Research Journal: Machine Learning and Computationhttps://doi.org/10.1029/2026JH001249, 2026)
—Saima May Siddique (@saimamay.bsky.social), science writer
