112
J. Eur. Opt. Society-Rapid Publ. 22, 11( 2026)
Fig
. 2. Spectrogram comparison results of an earthquake with a magnitude of 2.7 read in a DAS( top) and on the two OBS( bottom) in LOC1 and MAD01( a) and LOC2 and MAD02( b).
Fig
. 3. a) Earthquake magnitude versus distance for events detected by land stations during the HDAS – OBS overlap. The black line shows the best-fit logarithmic curve, and the red line indicates the sensitivity threshold. b) Magnitude versus distance for the 229 earthquakes examined for HDAS recordings. Colors represent waveform quality( see text). The red line shows the proposed logarithmic reference threshold law M *. In blue we show M *+ 1 in magnitude, which is close to the threshold for good-quality P-waves recorded by HDAS.
other energy signals that have not yet been identified. By comparing the two OBS locations, we see that their responses are similar, with a time delay that is expected due to the relative positioning of the OBS units. During the five-month interval in which the HDAS and the OBS were recording simultaneously, the IPMA network – comprising land-based stations on the neighboring islands( see Fig. 1 for locations) – detected 229 local and regional earthquakes. To quantitatively assess the network’ s sensitivity and detection threshold in this area, we examined the relationship between event magnitude and the minimum distance to the submarine cable. Specifically, by plotting magnitude as a function of the closest distance to the cable( see Fig. 3), we can evaluate how sensitivity decreases with increasing source – receiver separation. This allowed us to delineate the magnitude – distance envelope below which events are no longer consistently recorded. This provides a useful baseline for comparing the performance of DAS / HDAS and OBS observations with that of the land network, particularly for offshore seismicity where terrestrial instrumental coverage is inherently limited.
A sensitivity-threshold function for the land network was derived( red curve in Fig. 3a), empirically outlining the magnitude – distance range in which onshore stations reliably detect earthquakes; events falling below this curve are increasingly likely to go undetected. In parallel, HDAS recordings were examined for 229 events near the OBS sites( LOC1 near MAD01 and LOC2 near MAD02), where crossmethod comparison is strongest. Each event’ s HDAS signal quality was classified into five categories: Very good( clear P and S waves on both cable sections), Good( at least one identifiable P wave plus clear S waves), Fair( welldefined S waves but no clear P wave), Poor( weak, poorly defined S waves), and None( no recognizable seismic