We conduct two-dimensional (2D) and two-and-one-half dimensional (2.5D) visco-acoustic full-waveform inversion in the frequency domain using new long-offset data from the northeast lobe of the Sudbury structure acquired in 2017. We implement a multiscale inversion strategy based on frequency continuation, and the progressive inclusion of later arrivals, in an effort to mitigate the nonlinearity of the inverse problem. This strategy is equally implemented in both 2D and 2.5D schemes, enabling proper comparisons between their respective results. We start by minimizing logarithmic phase-only residuals, and continue with the minimization of conventional phase-amplitude residuals at later stages, addressing large dynamic variations within our dataset. We demonstrate that the 2.5D modeling technique, which requires more computational resources than its 2D counterpart, is not necessary at this location because the acquisition geometry is only mildly crooked. We illustrate this by analyzing inverted source signatures, time-domain synthetic waveforms, and by performing visual comparisons of the inverted 2D and 2.5D velocity models. We successfully retrieve the velocity structure in the first 1.5 km of the subsurface, and the internal stratigraphic character of the Sudbury Igneous Complex (SIC) is identified within this velocity model. Different velocity domains within our model closely correlate with known geology. This allows us to proceed with a joint analysis of the inverted velocity model and the migrated seismic section of the reflection survey that reveals important structural characteristics of prominent SIC layers, such as their inclination degree and thicknesses, as well as their continuation at depth.

Mostafa Naghizadeh

and 10 more

Passive seismic methods are considered as cost-effective and environmental-friendly alternatives to active (reflection) seismic methods. We have acquired co-located active and passive seismic surveys over a metal-endowed Archean granite-greenstone terrane in the Larder Lake area to investigate the reliability of the estimated elastic properties using the passive seismic methods. The passive seismic data was processed using two different data processing approaches, the ambient noise surface wave tomography (ANSWT) and receiver function analysis methods to generate shear-wave velocity and P- to S-wave (P-S) convertibility profiles of the subsurface, respectively. The Cadillac-Larder Lake Fault (CLLF) was imaged as a south-dipping sub-vertical zone of weak reflectivity in the reflection seismic profile. To the north of the CLLF, a package of north-dipping reflections in the upper-crust (at depths of 5-10 km) resides on the boundary of high (on the top) and low (on the bottom) shear-wave velocity zones estimated using the ANSWT method. This package of reflections is most likely caused by overlaying mafic volcanic and underlying felsic intrusive rocks. The P-S convertibility profile imaged the Moho boundary at ~40 km depth as well as a south-dipping slab that penetrates into the mantel which was interpreted to be either caused by the delamination of the lower crust or a possible deeper extension of the Porcupine-Destor Fault. Overall, the reflectivity, shear-wave velocity, and P-S convertibility profiles exhibited a good correlation and provided a detailed image of the subsurface lithological structure to a depth of 10 km.