A comparison of forward modeling and inversion of seismic first arrivals over the Kapuskasing Uplift

Abstract

Abstract A comparison has been made between the results of trial-and-error forward modeling and a tomographic inversion of first-arrival travel times from a reversed seismic refraction profile across the Kapuskasing Uplift, Northern Ontario. The data were collected as part of a crustal scale experiment and were originally interpreted using forward modeling with synthetic seismograms. The original analysis, utilizing refracted arrivals, imaged a high-velocity anomaly (6.5 to 6.6 km / sec) beneath the Chapleau block of the Kapuskasing Uplift that extended to 17 to 20 km depth with a dip of 15 ± 2°. Based on the surface geological exposures, this dipping anomaly was interpreted as a zone of granulites up-thrust from the mid-crust in early Proterozoic times. To test the significance of this anomaly, a tomographic inversion was performed on the first arrivals from the profile. The inversion results corroborate the main features of the original velocity model; a high-velocity anomaly (6.5 to 6.6 km / sec) at the surface in the granulite zone flanked by lower velocities (6.0 to 6.1 km / sec) in the Wawa terrane to the west and (6.2 to 6.3 km / sec) in the Abitibi belt to the east. However, resolution analysis and test inversions indicate that the first-arrival travel times alone can only image the western edge of the dipping high velocity anomaly to less than 10 km depth. At depths greater than 7.5 to 10 km in the west and ∼ 2.5 km in the east, model resolution significantly deteriorates and the sensitivity of the final velocities to the starting model used for inversion increases. Resolution analysis also demonstrates how additional intermediate shot points or increased aperture would have improved resolution of the anomaly at greater depth, emphasizing the importance of performing resolution analysis during the planning stages of refraction experiments.