This vibroseismic survey was conducted on Thwaites Glacier in West Antarctica in 2022/23, as part of the International Thwaites Glacier Collaboration's project GHOST (Geophysical Habitat of Subglacial Thwaites). The 210 km long seismic profile (section 20230551) was measured parallel to the average flow direction. The reflection coefficients at the ice base were derived from the amplitude ratio of the incident to the reflected waves at the basal return. We used an algorithm to determine the amplitude of the prime basal reflection for small offsets from unfiltered common midpoint (CMP) gathers. We determined the average source amplitude from the multiple bounce method (Holland and Anandakrishnan, 2009). To assess the consistency of the source power, we calculate the amplitude of the direct wave for all shots and offsets. For each offset, we determined the relative deviation from the mean across all shots. We then smoothed the amplitudes based on a running mean with a window of 3 km and averaged the relative deviation across all offsets. For further processing we use a relative power dependent source amplitude. The attenuation of the seismic wave in ice was derived from the estimated quality factor Q of 451±23. To estimate the attenuation, we used the centroid frequency of the power spectrum of 146±25 Hz and the interval velocity of ice 3770 m/s, giving an estimated attenuation of (2.7±0.5) x10^-4 1/m. Next, we derived the reflection coefficient for each CMP. We eliminated outliers by filtering for polarities. To determine the dominant polarity, we averaged polarities using a moving window of 16 CMPs (200 m along the profile), approximately matching the Fresnel zone width (220 m). If 80% or more of the CMPs had the same polarity, we removed the conflicting ones and averaged the remaining amplitudes and reflection coefficients within that group (200 m window). Next, we manually eliminated groups with ambiguous or inconsistent basal returns, and those intersected by diffraction hyperbola that interfere with the reflection amplitude. The uncertainty of the reflection coefficient was derived from the uncertainties of the source amplitude, and of the attenuation. The acoustic impedance of the subglacial material was derived from the reflection coefficient of the ice base and the acoustic impedance of the basal ice of (3.33±0.04)x10^6 kg/(m^2 s). The uncertainty of the acoustic impedance was derived from error propagation including the uncertainty of the reflection coefficient and of the acoustic impedance of ice. This dataset contains the seismic attributes (reflection coefficient and acoustic impedance) of the bed.

If you have questions, please contact Ole Zeising (ole.zeising@awi.de) or co-authors.Data were acquired in the field by: Ole Zeising, Coen Hofstede and Olaf Eisen.Data processing was done by Ole Zeising.This survey has been implemented by Alex Brisbourne and Sridhar Anandakrishnan as part of the ITGC-GHOST Project.The authors would like to thank Aspen Technology, Inc. for providing software licenses and support.