Seismo-tectonic analysis of the ongoing Mayotte crisis suggesting reactivation of a caldera structure below the Moho

Abstract:

Since May 2018, thirty kilometers to the SE off the island of Mayotte (Indian Ocean, Comoros archipelago), an exceptional seismic-volcanic activity began with a swarm of M5 + earthquakes (Lemoine et al., 2020), which migrated towards the new volcano discovered in May 2019 (Feuillet et al., in review). In summer 2018, a new earthquake swarm took place about 15 kilometers SE of Mayotte (Mercury et al., Abstract AGU 2020). Since February 2019, the seismicity of these two still active swarms has been recorded using Ocean bottom seismometers. Using their recordings we showed that this seismicity is lying between 20 and 50 km deep, at unusual depths under the Moho (Jacques et al., 2019).

According to our hypotheses, the first swarm allowed the opening of a dyke supplying the new volcano with a significant flow of lava, which caused a significant deflation of a reservoir located deep under the second swarm. This reservoir was connected to another at the level of the first swarm, and fed the new volcano. The deflation likely triggered the second swarm.

Within the framework the MAYOBS/ REVOSIMA seismological group, we carried out relative relocations (Hoste Colomer et al., Abstract AGU 2020) from new accurate absolute locations (Lavayssière et al., Abstract AGU 2020), and specified the geometry of the two swarms. The first swarm is organized in a narrow band of seismicity about 20 km long, trending N125°E towards the new volcano. The second swarm, in map view, has a hollow donut shape and its 3D geometry suggests an hourglass shape. Based on the new relative locations, we determined focal mechanisms using P wave polarity. The mechanisms related to the first swarm are compatible with a transtensive tectonic regime, while those of the second one show very different mechanisms (compressive, strike-slip, and extensional) with various directions of P and T axis. At the place of the second swarm, these mechanisms associated with the hourglass structure suggest the reactivation of a caldera-type structure. This is the first time that the activity of such a structure has been imaged under the Moho.