Characterization of stratigraphic interfaces and embedded limestone cavities using integrated geophysical and geological data

Abstract

Accurate identification of stratigraphic interfaces and embedded cavities is critical for geological modeling, engineering design, and infrastructure maintenance. However, integrating geophysical and geological data remains challenging due to discrepancies in spatial resolution, signal-to-noise ratio, and geological representation. This study presents a stratigraphic U-Net model that combines geophysical and geological information to improve subsurface interpretation. An initial U-Net model is trained using a preliminary P-wave velocity model derived from conventional seismic inversion. The predicted interfaces, however, show noticeable discrepancies with borehole observations because of manual velocity picking and limited borehole data. To address these limitations, the stratigraphic U-Net model is retrained using the initial predictions together with both drilled and synthetic boreholes. This iterative strategy significantly improves the identification of stratigraphic interfaces, particularly at greater depths. The stratigraphic U-Net model is further integrated with an opening detection model to identify embedded cavities within geological layers. Model evaluation shows that geological diversity captured by borehole data contributes more to prediction accuracy than simply increasing the borehole number. A field case study demonstrates that the proposed framework accurately identifies stratigraphic interfaces and subsurface cavities, providing a robust workflow for reducing geological uncertainty and improving the spatial coverage and reliability of subsurface characterization.

Keywords

Stratigraphic interfaces / Embedded cavities / Borehole data / Machine learning

References

[1] Ghosh R, Gustafson A, Schunnesson H. Development of a geological model for chargeability assessment of borehole using drill monitoring technique. Int J Rock Mech Min Sci, 2018, 109: 9-18

[2] Day-Lewis FD, Slater LD, Robinson J, Johnson CD, Terry N, Werkema D. An overview of geophysical technologies appropriate for characterization and monitoring at fractured-rock sites. J Environ Manag, 2017, 204: 709-720

[3] Rafat G, Lehmann B, Toumani A, Rueter H. Characterisation of rock ahead and around tunnels and boreholes by use of geophysical and geological methods. Int J Rock Mech Min Sci, 2001, 38: 903-908

[4] Gallardo L, Fontes S, Meju M, Buonora M, De Lugao P. Robust geophysical integration through structure-coupled joint inversion and multispectral fusion of seismic reflection, magnetotelluric, magnetic, and gravity images: example from Santos Basin, offshore. Brazil. J Geophys, 2012, 77: B237-B251

[5] Fehler MC, Huang L. Modern imaging using seismic reflection data. Annu Rev Earth Planet Sci, 2002, 30: 259-284

[6] Zakarewicz G, Maciel STR, da Cunha LS. Systematic literature review on seismic diffraction imaging. Earth-Sci Rev, 2024, 254 104792

[7] Zhang C, Chu J, Wu W, Poh TY, Lim ZL, Veeresh C. Smart detection of subsurface anomalies: concept, validation and applications. Tunn Undergr Space Technol, 2024, 154 106107

[8] Yu S, Ma J. Deep learning for geophysics: current and future trends. Rev Geophys, 2021, 59 e2021RG000742

[9] Lin L, Zhong Z, Li C, Gorman A, Wei H, Kuang Y, Wen S, Cai Z, Hao F. Machine learning for subsurface geological feature identification from seismic data: methods, datasets, challenges, and opportunities. Earth-Sci Rev, 2024, 257 104887

[10] Xiong W, Ji X, Ma Y, Wang Y, AlBinHassan NM. Seismic fault detection with convolutional neural network. Geophysics, 2018, 83 1SO-Z29

[11] Meng W, Wu W, Poh TY, Lim ZL. Intelligent detection of subsurface openings and surrounding disturbed zones. Tunn Undergr Space Technol, 2024, 154 106122

[12] Keiswetter D, Steeples D. Practical modifications to improve the sledgehammer seismic source. Geophys Res Lett, 1994, 21: 2203-2206

[13] Keiswetter DA, Steeples DW. A field investigation of source parameters for the sledgehammer. Geophysics, 1995, 60: 1051-1057

[14] Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation. Lect Notes Comput Sci, 2015, 9351: 234-241

[15] LeCun Y, Boser B, Denker J, Henderson D, Howard R, Hubbard W, Jackel L. Handwritten digit recognition with a back-propagation network. Adv Neural Inf Process Syst, 1989, 2: 1-9

[16] LeCun Y, Bottou L, Bengio Y, Haffner P. Gradient-based learning applied to document recognition. Proc Inst Electr Electron Eng, 1998, 86: 2278-2324

[17] Boureau YL, Ponce J, LeCun Y (2010) A theoretical analysis of feature pooling in visual recognition. In: Proceedings of the 27th International Conference on Machine Learning (ICML 2010), pp 111–118

[18] Nair V, Hinton GE (2010) Rectified linear units improve restricted boltzmann machines. In: Proceedings of the 27th International Conference on Machine Learning, pp 807–814

[19] Phaisangittisagul E (2016) An analysis of the regularization between L2 and dropout in single hidden layer neural network. In: 2016 7th International Conference on Intelligent Systems, Modelling and Simulation, pp 174–179

[20] Wager S, Wang S, Liang PS. Dropout training as adaptive regularization. Adv Neural Inf Process Syst, 2013, 26: 1-11

[21] Almarzoug AM, Ahmed FY (2012) Automatic seismic velocity picking. In: Proceedings of the SEG International Exposition and Annual Meeting 2012

[22] Yilmaz Ö, Doherty SM. Seismic data analysis: processing, inversion, and interpretation of seismic data, 2001: 231-233

[23] Pan SJ, Yang Q. A survey on transfer learning. IEEE Trans Knowl Data Eng, 2009, 22: 1345-1359

[24] Siahkoohi A, Louboutin M, Herrmann FJ. The importance of transfer learning in seismic modeling and imaging. Geophysics, 2019, 84: A47-A52

[25] Jackson DD, Anderson DL. Physical mechanisms of seismic wave attenuation. Rev Geophys, 1970, 8: 1-63

[26] Wu W, Li HB, Zhao J. Dynamic responses of non-welded and welded rock fractures and implications for P-wave attenuation in a rock mass. Int J Rock Mech Min Sci, 2015, 77: 174-181

[27] Qi X, Wang H, Chu J, Chiam K. Two-dimensional prediction of the interface of geological formations: a comparative study. Tunnel Undergr Space Technol, 2022, 121 104329

[28] Meng WZ, Chong JQ, Wu W. High-resolution seismic scattering imaging for urban underground infrastructure mapping. Tunnel Undergr Space Technol, 2025, 164 106811

Funding

National Research Foundation Singapore(COT-V1-2020-5)

RIGHTS & PERMISSIONS

The Author(s)