Crustal structure beneath Hudson Bay from ambient-noise tomography: implications for basin formation

• Published in: Geophysical Journal International
• Main Authors: Pawlak, Agnieszka, Eaton, David W., Bastow, Ian D., Kendall, J-Michael, Helffrich, George, Wookey, James, Snyder, David
• Format: Text
• Language: English
• Published: Oxford University Press 2011
• Subjects: Geodynamics and Tectonics; Hudson; Hudson Bay; Pacific; North Atlantic; Alaska
• Online Access: http://gji.oxfordjournals.org/cgi/content/short/184/1/65; https://doi.org/10.1111/j.1365-246X.2010.04828.x

The Hudson Bay basin is the least studied of four major Phanerozoic intracratonic basins in North America and the mechanism by which it formed remains ambiguous. We investigate the crustal structure of Hudson Bay based on ambient-noise tomography, using 21 months of continuous recordings from 37 broad-band seismograph stations that encircle the Bay. Green's functions that emerge from the cross correlation of these ambient noise recordings are dominated by fundamental-mode Rayleigh waves. In the microseismic period band (5–20 s), these signals are most prominently expressed in certain preferred azimuths indicative of stationary coastal source regions in southern Alaska and Labrador. Seasonal variations are subtle but consistent with more energetic noise sources during winter months, when wave heights in the Pacific and north Atlantic are larger than in the summer. Noise emanating from Hudson Bay does not appear to contribute significantly to the cross correlograms. Group-velocity dispersion curves are obtained by time-frequency analysis of cross-correlation functions. We test and implement a signal-to-noise ratio (SNR) selection method for producing one-sided cross correlograms, which yields better-defined dispersion ridges than the standard two-sided averaging approach. Tomographic maps and cross sections obtained in the 5–40 s period range reveal significantly lower crustal velocities beneath Hudson Bay than in the bounding Archean Superior craton. The lowest mid-crustal velocities correspond to a previously determined region of maximum lithospheric stretching near the centre of the basin. Pseudosections extracted from the tomographic inversions along profiles across Hudson Bay provide the first compelling direct evidence for crustal thinning beneath the basin. Our results are consistent with a recent estimate of 3 km of crustal thinning, but not consistent with a proposed model for basin subsidence triggered by eclogitization of a remnant crustal root.