GNGTS 2022 - Atti del 40° Convegno Nazionale

514 GNGTS 2022 Sessione 3.3 • Deep Attribute E: this attribute highlights very clearly the area where Forte et al. (2016) interpreted the salt brine overflow. An interesting feature is the area marked with “5?”, which seems to have a signature quite similar to the “5” one, which is the area with brines from the cited literature. From attribute E, we can check for other confirmations in previously descripted Deep Attributes, and we can see that the area marked with “5?” Is actually very coherent in all the other attributes. Conclusion. In this work we propose a new approach to GPR attribute estimation. The methodology is fully data-driven and fast. This method can provide further support in data classification / analysis by exploiting features that cannot be extracted from classical approaches and that represent a new class of geophysical attributes. The new attributes lend themselves to extensive use in the interpretation of 2-D profiles or 3-D data cubes not limited to the GPR case. In fact, the proposed procedure is totally suitable for application to any type of reflection seismic data. The way forward for this approach will be: • Check and guarantee the method stability. • Unlink the methodology from a specific task, e.g. from horizon extraction: this would produce more general Deep Attributes. • Try to fully understand the link between a set of attributes and its actual physical meaning. Acknowledgements. This research was partially supported by PNRA projects IPECA (PNRA18_00186) and CRIOVEG (PNRA18_00288) and by the project “Dipartimento di Eccellenza” of the Department ofMathematics andGeosciences of the University of Trieste. We gratefully acknowledge the support of Halliburton Landmark through the University Grants Program (UGP). References Anstey N., Bahorich M.S., Bridges S. R., Farmer S.L., et al. , 2007, Overview of Seismic Attributes, Geophysical Developments Series: 1-24, https://doi.org/10.1190/1.9781560801900.ch1. Chopra S., Marfurt K.J., 2005, Seismic attributes — A historical perspective, Geophysics, 70, 5, 3SO-28SO, https://doi. org/10.1190/1.2098670. Forte E.,Dalle Fratte M., Azzaro M., and Guglielmin M., 2016, Pressurized brines in continental Antarctica as a possible analogue of Mars, Scientific Reports, vol. 6, no. 1, p. 33158, Sep. 2016, doi: 10.1038/srep33158. Hochreiter, S., and Schmidhuber, J., 1997, Long short-term memory, Neural computation, 9(8), 1735-1780, doi: 10.1162/neco.1997.9.8.1735 Roncoroni G., Forte E., Bortolussi L., Pipan M., 2021, Efficient automatic extraction of seismic horizons with deep learning, proceedings of GNGTS 2021.