ALBERT

Description: Visco-acoustic waveform tomography was applied to marine seismic reflection data across the Seattle fault zone in Puget Sound in the northwestern USA. Using the recovered velocity and attenuation models, we performed a set of synthetic visco-acoustic and viscoelastic checkerboard tests, and compared the results to verify the effectiveness of applying visco-acoustic waveform tomography to viscoelastic field data. Visco-acoustic waveform tomography produces higher resolution velocity and attenuation models than ray-based tomography, but artefacts due to elastic effects such as mode conversion are present at layer interfaces where the velocity contrast is high. Elastic effects also affect attenuation values, which can be too high or too low in places because visco-acoustic inversion compensates the loss of amplitude due to mode conversion by inadequately estimating the attenuation. A comparison of the attenuation models inverted from viscoelastic and visco-acoustic synthetic data suggests that inverted attenuation values can be reliable when the velocity gradient is low, and the quality of the inversion improves in a highly attenuating medium or in a medium with high attenuation contrasts. Joint interpretation of the derived velocity and attenuation models enables us to identify Quaternary (glacial and postglacial Pleistocene) sedimentary, Tertiary sedimentary and Eocene volcanic rocks. Several shallow faults, anticlines and a syncline are identified across the Seattle uplift and the Seattle fault zone. Our interpretation of faults using the velocity model, attenuation model and migrated seismic section is consistent with two possible published models of the Seattle Fault Zone: either a thrust fault that accommodates north–south shortening by forming a fault-propagation fold with a forelimb breakthrough, or part of a passive roof duplex in which the Seattle Fault Zone is located at the leading edge of a triangle zone that is propagating into the Seattle basin.