B.V.Platov, E.R. Yusupova, A.S. Borisov
A method for effective thickness maps calculating at the seismic exploration resolution limit using machine learning: the example of an oil field in the Republic of Tatarstan
DOI 10.47148/0016-7894-2026-3-103-112
Keywords: Seismic exploration; thin formation; seismic attribute; machine learning.
For citation: Platov B.V., Yusupova E.R., Borisov A.S. A method for effective thickness maps calculating at the seismic exploration resolution limit using machine learning: the example of an oil field in the Republic of Tatarstan. Geologiya nefti i gaza. 2026;(3):103–112. DOI: 10.47148/0016-7894-2026-3-103-112. In Russ.
Funding: This paper has been supported by the Kazan Federal University Strategic Academic Leadership Program (PRIORITY-2030).
Seismic exploration has significant limitations in terms of resolution. The limit of vertical resolution is considered to be one-quarter of the wavelength. For this reason, identifying thin beds (at the limit of seismic resolution) is difficult. Mapping thickness for such low-thickness beds can be performed using seismic attributes combined with machine learning techniques. In this work, the authors propose a new method for thickness estimation at the resolution limit: tracing the target reflective horizon (the top of the interval for which thickness is being calculated); calculating frequency-based seismic attributes and finding correlations between “effective bed thickness – seismic attribute”; extracting seismic attribute cubes into sedimentation slices parallel to the interval top. Sampling seismic attribute values along the sedimentation slices. Selecting attribute slices with a linear correlation value exceeding 0.75; using the selected slices in machine learning algorithms. Running 100 realizations while excluding 30 % of the input data; selecting the algorithm with the lowest root mean square error. To assess the quality of the predictions, maps of the standard deviation of the realizations and the root mean square error at well locations were used. Areas on the map with the highest standard deviation of realizations have the greatest uncertainty.
Boris V. Platov ORCiD Scopus
Senior Lecturer
Kazan (Volga Region) Federal University,
4/5, ul. Kremlyovskaya, 420111, Russia
e-mail: swborispl@mail.ru
SPIN: 4939-4807
Enzhe R. Yusupova ORCiD
Engineer
Kazan (Volga Region) Federal University,
4/5, ul. Kremlyovskaya, 420111, Russia
e-mail: enzhe.yusupova.03@mail.ru
Borisov A. Sergeevich
Doctor of Geological and Mineralogical Sciences,
Professor
Kazan (Volga Region) Federal University,
4/5, ul. Kremlyovskaya, 420111, Russia
e-mail: ABorisov@kpfu.ru
1. Widess M.B. How thin is a thin bed? Geophysics. 1973;38(6):1176–1180.
2. Vilmaz O. Obrabotka seismicheskikh dannykh [Processing of seismic data]: v 3 t. Moscow: RGU NIU im. Gubkina, 1986. V. 3. 212 p.
3. Kunin N.Ya. Issledovanie litosfery dokriticheskimi otrazhennymi volnami [Investigation of the Lithosphere by Subcritical Reflected Waves]. Moscow: Nauka, 1993. 223 p.
4. Neidell N.S. Stratigraphic modeling and interpretation — Geophysical principles and techniques: Section 3. Stratigraphic models from seismic data. Seismic Stratigraphy Applications to Hydrocarbon Exploration. Tulsa, 1977. pp. 389–416.
5. Saeed W. An Integrated Petrophysical-Based Wedge Modeling and Thin Bed AVO Analysis for Improved Reservoir Characterization of Zhujiang Formation, Huizhou Sub-Basin, China: A Case Study. Open Geosciences. 2020;12(1):256–274.
6. Gurvich I.I. Seismicheskaya razvedka [Seismic exploration]. Moscow: Nedra, 1980. 551 p.
7. Naidel’ N.S. Geofizicheskoe obosnovanie i metodika seismostratigraficheskogo modelirovaniya i interpretatsii [Geophysical basis and methodology of seismic stratigraphic modeling and interpretation]. Seismicheskaya stratigrafiya. Moscow: Mir, 1982. pp. 646–692.
8. Thickness prediction for high-resolution stratigraphic interpretation by fusing seismic attributes of target and neighboring zones with an SVR algorithm. Marine and Petroleum Geology. 2020;(113):1–14.
9. Shoemaker M.L., Robinson J.B., Trumbly C.N., Brennan B.A. Prediction of thin bed reservoirs below 1/4 wavelength tuning thickness using full bandwidth inverted seismic impedance. International Petroleum Technology Conference (IPTC) 2007. European Association of Geoscientists & Engineers, 2007. DOI: 10.2523/IPTC-11359-MS.
10. Bitrus P.R. Defining the 3D geometry of thin shale units in the Sleipner reservoir using seismic attributes. Marine and Petroleum Geology. 2016;(78):405–425.
11. Geologiya Respubliki Tatarstan [Geology of the Republic of Tatarstan] : ucheb.-metod. Posobie. R.Kh. Sungatullin, G.M. Sungatullina, ed. Kazan: K(P)FU, 2014. 72 p.
12. Geologiya Tatarstana. Stratigrafiya i tektonika [Geology of Tatarstan. Stratigraphy and Tectonics]. B.V. Burov (ed.). Moscow: GEOS, 2003. 403 p.
13. Priezzhev I.I., Dan’ko D.A., Onishchenko A.N. Hierarchical neural networks in predicting the properties of oil and gas reservoirs based on well and seismic data. Russian Geology and Geophysics. 2025;6(1):118–126. DOI: 10.2113/RGG20244760.
14. Kolmogorov A.N. O predstavlenii nepreryvnykh funktsii neskol’kikh peremennykh superpozitsiyami nepreryvnykh funktsii men’shego chisla peremennykh [On the representation of continuous functions of several variables as superpositions of continuous functions of a smaller number of variables]. Doklady Akademii Nauk. 1957;14(5):953–956.
15. Priezzhev I. New Age: Kolmogorov Full Functional Neural Network Usage for Nonlinear Predictive Seismic Inversion. EAGE Saint Petersburg 2020. 2020. pp. 1–5.