The spatio-temporal behavior of the Mantle Wedge Seismicity and its relationship with the interface in Chile.

The serpentinized mantle wedge corner above subducting slabs has been long considered as mostly aseismic. However, mantle wedge seismicity (MWS) has been observed in different subduction zones such as Japan, New Zealand, Lesser Antilles, South America, Colombiaand the Hellenic subduction zone. Several hypothesis have been made to explain such seismicity: (i) a graveyard of piled seamounts that are detached from the subducting plate and underplates the forearc lithosphere; (ii) plume underplating leading to higher viscosity parts in the mantle corner; (iii) serpentine dehydration embrittlement and (iv) pulses of fluids released from the plate interface. Several subduction zones exhibit seismicity gaps on the interface that might indicate a diversion of the seismicity through vents. The rheological and mechanical behaviors of the mantle wedge and its possible interactions between its seismicity and the interface still remain largely unclear. In this study, we take advantage of the recent catalogue obtained with machine learning on Chile from November 2020 to February 2024 and analyse the spatio-temporal distribution and the statistics of the mantle wedge seismicity in this area.

We find that the MWS is mostly active between -27°S and 31°S latitude and shows a Gutenberg-Richter b-value of 1.4 which is higher than the interface seismicity (around 1). It presents a magnitude of completeness of 1.6 and is gathered as clusters of events that behave as swarms rather than mainshock-aftershocks sequences. Some clusters are triggered after a large event (magnitude > 5) occurring on the interface. The detailed analysis of the distribution of the MWS compared to the interface shows that the MWS is mainly located in a band between 130 and 160km away from the trench while the interface seismicity is mainly located in a band of 60km to 100km away from the trench. While the interface seismicity gap present just above the MWS might confirm the presence of vents that would deviate the seismicity, the difficulty to track potential fluid paths from the intraplate seismicity at depth to the MWS might rule out fluids as the origin of the MWS in Chile. Rather, this gap might indicate the importance of the mineralogical contact between the interface and the mantle wedge.