Sub-surface magma movement inferred from low-frequency seismic events in the off-Nicobar region, Andaman Sea

The study investigates whether recent earthquake swarms off Nicobar Island are linked to subsurface magma movement beneath submarine volcanoes. Volcanic seismicity comprises volcano-tectonic (VT) events, tremors, and low-frequency events including long-period and very long-period events (VLPEs), with hybrids showing high-frequency onsets followed by long-period codas. Underwater VT events can generate hydro-acoustic waves detectable in deep ocean via the SOFAR channel. The Andaman–Nicobar–Sumatra subduction zone hosts an inner volcanic arc and a network of sliver strike-slip faults (including the Seulimeum Fault) that accommodate trench-parallel motion. Following the 2004–2005 megathrust earthquakes, Coulomb stress changes modulated regional seismicity, with multiple swarms occurring off Nicobar. Here, OBS data from March 21–22, 2014 are analyzed to report, for the first time in this region, hybrid VLPEs and associated hydro-acoustic phases as indicators of subsurface magmatism and its interaction with local strike-slip faulting.

Prior work classifies volcanic earthquakes and documents their association with magmatic systems, including VT events due to shear failure from magma-induced stress, LPEs/VLPEs with dominant low frequencies, and hybrid events with high-frequency onsets activating long-period signals. Hydro-acoustic (T-wave) monitoring has been widely used for submarine volcanic activity. Earthquake swarms linked to subsurface magma movement have been reported at Loihi Seamount, the Okinawa Trough, Axial Volcano (Juan de Fuca Ridge), and mid-ocean ridges. In the western Indian Ocean near Mayotte, abundant VT events and VLPEs were associated with drainage of a deep magma reservoir and dike propagation. In the Andaman–Nicobar region, the January 2005 off Nicobar swarm was among the most energetic globally and has been attributed to stress changes from the 2004 megathrust event, local strike-slip faulting, and submarine volcanism; bathymetry shows cratered submarine volcanoes, but direct geophysical evidence of recent magma movement had been lacking. The regional sliver fault system (SGF, ATF, ABSC, ANF, WAF, GSF including Seulimeum) accommodates oblique convergence and can facilitate magma ascent; stress changes after the megathrusts promoted failure along parts of this system, making the region susceptible to swarms.

A passive OBS experiment deployed 12 broadband, three-component OBS units (K.U.M. Umwelt- und Meerestechnik) across the Andaman Sea from 22 December 2013 to 15 May 2014. For the off Nicobar swarm of 21–22 March 2014, data from OBS (notably OBS02 in the source region), the Andaman ISLANDS network (CBY, BAKU, BARA, SBY, PBA, HUTB), and additional land stations (IRIS DMC) were integrated. Event identification across frequency bands used SEISAN. P- and S-wave onsets and first-motion polarities were picked from high-frequency impulsive waveforms; hybrid character was assessed via time–frequency analysis showing high-frequency onsets (1–10 Hz) followed by low-frequency (0.01–0.5 Hz) oscillations of long duration (300–600 s), and prominent hydro-acoustic energy (10–40 Hz) in pressure channels. Hypocenters were computed with Hypocent v3.2 using a minimum 1D velocity model for the region (Singha et al., 2019), with location error ellipses from covariance matrices. Local magnitudes were calculated using equivalent Wood–Anderson amplitudes. Additional locations were derived from hydro-acoustic phases assuming a water sound speed of 1.5 km/s. T-wave rise times were computed following Schreiner et al. (1995) to infer source depth evolution. Focal mechanisms for four representative M>4 events were determined using HASH from first-motion polarities picked on >30 stations. Spectral properties were analyzed (e.g., peak frequencies, duration). Spatial distributions were compared to high-resolution multibeam bathymetry highlighting volcanoes and major faults (ANF, WAF, Seulimeum).

- A mainshock of Mw 6.5 occurred on 21 March 2014 at 13:41 UTC, followed by an earthquake swarm through 22 March 2014 recorded by all OBS and ISLAND stations. - A total of 141 local high-frequency onset events (detected on ≥3 stations) were identified on 21–22 March 2014; their magnitudes ML ranged from 2.0 to 5.2. The sequence did not follow a typical mainshock–aftershock decay, consistent with a swarm. - Seventy-one of the 141 events exhibited very long-period oscillations (VLPEs) with dominant frequencies 0.01–0.5 Hz and durations of 300–600 s; VLPE magnitudes ML ranged from 2.6 to 4.2. - Spectra showed a characteristic low-frequency peak (e.g., vertical component spectral peak at ~0.13 Hz). Event durations were much longer than typical tectonic earthquakes. - Hypocentral depths for located events spanned ~30 to 1 km, though depth accuracy was limited by station azimuthal coverage (<180°). Epicenters cluster near OBS02, forming a NW–SE linear trend consistent with the Seulimeum Fault and inner volcanic arc alignment. - Focal mechanisms indicate vertical right-lateral strike-slip faulting. The main event solution: strike 334°, dip 89°, rake 171°, with P-axis oriented SSW and T-axis ESE; three other larger events showed similar mechanisms. - Hydro-acoustic phases (10–40 Hz) were recorded on OBS hydrophones located in deep waters (>1500 m) within the Andaman Sea. A total of 181 hydro-acoustic detections on ≥4 OBS were documented; all 141 seismic events generated hydro-acoustic phases, and an additional ~40 hydro-acoustic-only detections indicated smaller events observable in pressure but not on enough seismometers. - Hydro-acoustic locations also trend NW–SE along the volcanic arc. The hydro-acoustic waveforms are emergent, long duration (>100 s). - T-wave rise time analysis showed an initial rapid decrease from ~35 s to ~5 s during the early swarm, indicating hypocenter migration from deeper to shallower depths; the average rise time was ~20 s, consistent with relatively deeper (but shallow crustal) sources; no very small rise times (<2 s) were observed, arguing against seafloor eruption. - A b-value analysis of the March 2014 swarm yielded an average b ≈ 1.39, consistent with a magmatic or volcano-tectonic influence rather than purely tectonic. - Recurring swarms (October 2014, November 2015, March 2019) recorded at nearby stations show similar low-frequency spectral peaks (~0.1 Hz). GCMT mechanisms for 2014–2019 larger events indicate strike-slip faulting, reinforcing the association with the sliver fault system.

The detection of hybrid VLPEs with clear high-frequency onsets followed by long-duration low-frequency oscillations, together with pervasive hydro-acoustic phases, indicates shallow sources and links the off Nicobar swarm to subsurface magmatic processes. The NW–SE alignment of epicenters along the inner volcanic arc and the Seulimeum strand of the Great Sumatra Fault, combined with vertical right-lateral strike-slip focal mechanisms, suggests that the regional sliver strike-slip fault system provides pathways facilitating magma ascent via dike intrusions aligned with the principal stress field. The observed decrease in T-wave rise time during the early swarm implies upward migration of hypocenters, consistent with ascending magma pulses from a deeper reservoir to a shallow chamber. The elevated b-value further supports a magmatic or mixed volcano-tectonic regime. These characteristics parallel observations at other volcanic systems (e.g., Redoubt, Soufrière Hills, Mayotte, mid-ocean ridges) where hybrid events and T-wave-rich swarms accompany dike intrusions and magma movement. While causality between fault slip and magma migration cannot be definitively established here, the co-occurrence of strike-slip seismicity and VLPEs indicates tight tectono-magmatic coupling in the off Nicobar segment since the 2004–2005 megathrust stress perturbations. The findings provide direct geophysical evidence for subsurface magmatism beneath submarine volcanoes in this region and offer a framework for monitoring via OBS hydrophones and broadband seismology.

- Hybrid VLPEs with high-frequency onsets followed by up to ~600 s of low-frequency (0.01–0.5 Hz) oscillations were identified for the first time off Nicobar, indicating subsurface magma movement near the inner volcanic arc. - Onset earthquakes are shallow, with vertical right-lateral strike-slip focal mechanisms; epicenters align along the Seulimeum Fault. - Hydro-acoustic phases (10–40 Hz) are closely associated with the swarm and corroborate shallow source depths. T-wave rise time evolution indicates hypocenter migration from deeper to shallower levels, consistent with upward magma movement. - The observations suggest that upward migration of magma pulses from a deeper reservoir to a shallow chamber reactivated sliver strike-slip faults, inducing swarms along pre-existing structures in the off Nicobar seismogenic zone. - These results contribute critical geophysical evidence of ongoing tectono-magmatic processes beneath submarine volcanoes in the Andaman Sea and motivate continued dense local monitoring to resolve small-magnitude events and volumetric components, and to clarify the temporal interplay between magmatic intrusions and fault slip.

- Depth estimates are limited by the sparse station coverage in the source area and inadequate azimuthal coverage (<180°), leading to uncertainty in hypocentral depths. - Hydro-acoustic T-wave rise time provides approximate depth indicators and may be influenced by path and site effects. - Low-magnitude events lack focal mechanism solutions; a denser local network is required to resolve small events and potential volumetric (non–double-couple) components expected during dike intrusion. - Confirmation of VLPEs on distant continental stations is limited by inelastic attenuation that can mimic low-frequency content. - The study cannot unambiguously determine whether strike-slip faulting triggered magma ascent or vice versa; causal direction remains unresolved. - No direct seafloor eruption evidence was observed (e.g., no very small T-wave rise times), so conclusions are restricted to subsurface magmatism rather than eruptive activity.