Articles | Volume 16, issue 6
https://doi.org/10.5194/os-16-1367-2020
https://doi.org/10.5194/os-16-1367-2020
Research article
 | 
11 Nov 2020
Research article |  | 11 Nov 2020

Random noise attenuation of sparker seismic oceanography data with machine learning

Hyunggu Jun, Hyeong-Tae Jou, Chung-Ho Kim, Sang Hoon Lee, and Han-Joon Kim

Cited articles

Araya-Polo, M., Farris, S., and Florez, M.: Deep learning-driven velocity model building workflow, The Leading Edge, 38, 872a1–872a9, https://doi.org/10.1190/tle38110872a1.1, 2019. 
Blacic, T. M., Jun, H., Rosado, H., and Shin, C.: Smooth 2-D ocean sound speed from Laplace and Laplace-Fourier domain inversion of seismic oceanography data, Geophys. Res. Lett., 43, 1211–1218, https://doi.org/10.1002/2015GL067421, 2016. 
Dagnino, D., Sallares, V., Biescas, B., and Ranero, C. R.: Fine-scale thermohaline ocean structure retrieved with 2-D prestack full-waveform inversion of multichannel seismic data: Application to the Gulf of Cadiz (SW Iberia), J. Geophys. Res.-Ocean., 121, 5452–5469, https://doi.org/10.1002/2016JC011844, 2016. 
Den Bok, H.: SMAART publicly released data sets, available at: http://www.delphi.tudelft.nl/SMAART/ (last access: 6 November 2020), 2002. 
Den Bok, H.: SMARRT publicly released data sets, available at: http://www.delphi.tudelft.nl/SMAART/, last access: 06 November 2020. 
Download
Short summary
Seismic oceanography acquires water column reflections by seismic exploration. The use of a high-frequency seismic source such as sparker can enhance the resolution of reflection image but suffers from a low signal-to-noise ratio problem. In this study, we applied a machine learning to remove the random noise in water column seismic section. We constructed appropriate training data and showed that the machine learning can successfully remove the random noise in the water column seismic section.