RESOLVING CRUSTAL STRUCTURE IN THE SUPERDEEP BENGAL BASIN
Here we model crustal structure by waveform matching P receiver functions and teleseismic P autocorrelograms for 33 seismic stations across the Bengal basin. Our P receiver functions use teleseismic Ps conversions generated at local impedance contrasts (including the Moho, the sediment-basement interface, and several intrasedimentary contacts). Teleseismic P autocorrelograms recover P reverberations generated at these same impedance contrasts. Jointly, they constrain crustal structure that is parameterized in terms of interface depth, compressional wave velocity (VP), and the ratio of compressional to shear wave velocity (VP/VS).
Based on observable phases and solution stability, our models are comprised of either three sedimentary units overlying the crystalline crust, for 21 stations, or two sedimentary units overlying crystalline crust, for 12 stations (i.e., the two deeper sedimentary units from the 4-layer models are combined). The uppermost sedimentary unit is ~3 km thick with an average VP of 3.3 km/s. The second sedimentary unit averages ~8 km thick with an average VP of 5.9 km/s. In the deepest portion of the basin, we model a low velocity zone with an average VP of 4.4 km/s for the four-layer stations. We interpret this low velocity zone as an indication of fluids trapped in the deep basin. Potential origins of fluids include a) rapid sedimentation that led to incomplete dewatering of buried sediments, b) fluid release from diagenetic alteration of clay minerals at the pressures and temperatures present in the deep basin, or c) some combination of these factors. Below the sedimentary layers we observe a 14-25 km thick layer with an average VP of 6.9 km/s, which we interpret to be extended continental crust that was likely altered during the Cretaceous by intrusions from the Kerguelen plume.