Correction to ‘Near Offset Reconstruction for Marine Seismic Data Using a Convolutional Neural Network’

Abstract

Huff, O.R., Thorkildsen, V.S., Greiner, T.L., Lie, J.E., Evensen, A.K., Bugge, A.J. et al. (2024) Near offset reconstruction for marine seismic data using a convolutional neural network. Geophysical Prospecting, 72, 2164–2185. https://doi.org/10.1111/1365-2478.13505

In Figure 12, the scaling factor applied to the CNN-predicted zero-offset section in Figure 12c and its spectrum in Figure 12f was incorrectly reported as 1.2 instead of 1.3. Accordingly, in Figure 12c, the label ‘1.2× amplitude’ should read: ‘1.3× amplitude’, and in Figure 12f, the label ‘CNN prediction (1.2×)’ should read: ‘CNN prediction (1.3×)’. In the caption of Figure 12, the text ‘(note that the amplitude is scaled by 1.2)’ and ‘The CNN prediction is scaled by 1.2’ should read: ‘(note that the amplitude is scaled by 1.3)’ and ‘The CNN prediction is scaled by 1.3’, respectively. The corrected Figure 12 and caption are provided below.

In addition, the corresponding text in the ‘Field Data Results’ section should be corrected as follows:

In the ‘Source-over-cable acquisition data (Barents Sea)’ subsection, the text ‘we scale the output of the CNN prediction by 1.2’ in paragraph 3 should read: ‘we scale the output of the CNN prediction by 1.3’, and the text ‘(previously scaled by 1.2 as explained in Figure 12)’ in paragraph 6 should read: ‘(previously scaled by 1.3 as explained in Figure 12)’.

In the ‘Conventional data (northern North Sea)’ subsection, the text ‘The CNN-predicted near offsets are not scaled by 1.2’ in paragraph 3 should read: ‘The CNN-predicted near offsets are not scaled by 1.3’, and the text ‘note the 1.2 scaling that was necessary to best match the spectra’ in paragraph 4 should read: ‘note the 1.3 scaling that was necessary to best match the spectra’.

We apologize for these errors.