Predictive analysis of total organic carbon (TOC) in shale targets: example from the Lower Cretaceous of the Austral Basin (Patagonia, Argentina) using machine learning on outcrop data
Palabras clave:
reservorios no convencionales, reducción dimensional, machine learning, predicción de COT, Cuenca AustralResumen
La Formación Río Mayer (Cretácico Inferior) de la Cuenca Austral, Patagonia, es la principal roca generadora y fuente de reservorios no convencionales. Este estudio analiza el potencial del uso de técnicas de machine learning (ML) para predecir el contenido de carbono orgánico total (COT) utilizando datos de afloramientos, un enfoque novedoso en comparación con las aplicaciones tradicionales de datos del subsuelo. Mediante técnicas de reducción dimensional (PCA, T-SNE, UMAP), el análisis mostró una clara agrupación de valores altos de COT en el feature space, lo que respalda la viabilidad del modelado predictivo. Se probaron tres modelos ML: regresión logística, clasificador de vectores de soporte (Support Vector Classifier-SVC) y Vecinos más cercanos (K-Nearest Neighbors –KNN-), utilizando un conjunto de características derivado de las clasificaciones F-Score del ANOVA. La reducción dimensional mejoró el rendimiento del modelado; el modelo SVC obtuvo los resultados más robustos. A pesar del bajo número de muestras con análisis de COT, las predicciones entre los modelos fueron consistentes, y se identificó una región prometedora para un alto contenido de carbono orgánico total (COT). Este estudio destaca la importancia de integrar variables geológicas de campo con datos de difracción de rayos X (DRX) en el modelado de distribución COT y enfatiza la necesidad de ampliar los conjuntos de datos y ampliar las secciones sedimentarias para mejorar las interpretaciones regionales.
Referencias
Aguirre Urreta, M.B. (2002). Invertebrados del cretácico inferior. In: Haller, M.J. (Ed.), Geología y Recursos Naturales de Santa Cruz. Relatorio del XV Congreso Geológico Argentino, 925 pp.
Arbe, H.A. (2002). Análisis estratigráfico del Cretácico de la Cuenca Austral. In: Haller, M.J. (Ed.), Geología y Recursos Naturales de Santa Cruz. Relatorio del XV Congreso Geológico Argentino, pp. 103–128.
Arbe, H.A. and Hechem, J. (1984). Estratigrafía y facies de depósitos continentales, litorales y marinos del Cretácico superior, lago Argentino. IX Congreso Geológico Argentino Actas 7:124-158.
Belotti, H., Pagan, F., Perez Mazas, A., Agüera, M., Rodriguez, J., Porras, J., Köhler, G., Weiner, G., Conforto, G., and Cagnolatti, M. (2013). Geologic Interpretation and Assessment of Early Cretaceous Shale Oil and Gas Potential in Austral Basin, Santa Cruz, Argentina, In: Unconventional Resources Technology Conference, Denver, Colorado, USA, August 2013.
Belotti, H., Rodriguez, J., Conforto, G., Pagan, F., Pérez Mazas, A., Agüera, M., Porras, J., Köhler, G., Cagnolatti, M., Weiner, G., Nigro, E., and Cangini, A. (2014). La Formación Palermo Aike como Reservorio No Convencional en la Cuenca Austral, Provincia de Santa Cruz, Argentina. IX Congreso de Exploración y Desarrollo, 1–17.
Brown, G., and Brindley, G.W. (1980). X-ray diffraction procedures for clay mineral identification. In: Brindley, G.W., Brown, G. (Eds.), Crystal structures of Clay Minerals and Their X-ray Identification. Mineralogical Society, London, pp. 305–359.
Busch, B., Spitzner, A., Adelmann, D., and Hilgers, C. (2022). The significance of outcrop analog data for reservoir quality assessment: A comparative case study of Lower Triassic Buntsandstein sandstones in the Upper Rhine Graben. Marine and Petroleum Geology 141, 105701. https://doi.org/10.1016/j.marpetgeo.2022.105701
Cuitiño, J., Varela, A., Ghiglione M., Richiano, S., and Poire, D.G. (2019). The Austral-Magallanes Basin (Southern Patagonia): A synthesis of its stratigraphy and evolution. Latin American Journal of Sedimentology and Basin Analysis 26, 155–166.
Féraud, G., Alric, V., Fornari, M., Bertrand, H., and Haller, M. (1999). 40Ar/39Ar dating of the Jurassic volcanic province of Patagonia: migrating magmatism related to Gondwana break-up and subduction. Earth and Planetary Science Letter 172, 83–98. https://doi.org/10.1016/S0012-821X(99)00190-9
Guo, J., Wang, S., Chang, L., Li, Y., Lyu, Q., and Fan, J. (2022). A Novel Method for Conventional Logging Prediction of Brittleness Index: A Case of Tight Sandstone in Western Ordos Basin. Frontiers in Energy Research 10.3389/fenrg.2022.819078.
Handhal, A., Al-Abadi, A., Chafeet, H., and Ismail, M. (2020). Prediction of total organic carbon at Rumaila oil field, Southern Iraq using conventional well logs and machine learning algorithms. Marine and Petroleum Geology 116, 104347. https://doi.org/10.1016/j.marpetgeo.2020.104347
Kraemer, P.E., and Riccardi, A.C. (1997). Estratigrafía de la región comprendida entre los lagos Argentino y Viedma (49º40’ - 50º10’ LS), Provincia de Santa Cruz. Revista de la Asociación Geológica Argentina 52, 333–360.
Melendo, F., Gargiulo, C., Garcia, N., and Jait, N. (2023). Description and Analysis of the First Two Vertical Pilot Wells Completions to the Palermo Aike Shale: A Highly Promising Unconventional Reservoir in Austral Basin, Argentina. Latin America Unconventional Resources Technology Conference (LA URTeC) https://doi.org/10.15530/urtec-2023-3969079
Milad, B., Slatt, R., and Fuge, Z. (2020). Lithology, stratigraphy, chemostratigraphy, and depositional environment of the Mississippian Sycamore rock in the SCOOP and STACK area, Oklahoma, USA: Field, lab, and machine learning studies on outcrops and subsurface wells. Marine and Petroleum Geology 115, 104278. https://doi.org/10.1016/j.marpetgeo.2020.104278
Moore, D.M., and Reynolds Jr., R.C. (1997). X-ray Diffraction and the Identification and Analysis of Clay Minerals. Oxford University Press, Oxford.
Mustafa, A., Tariq, Z., Abdulraheem, A., Mahmoud, M., Kalam, S., and Khan, R. (2022). Shale brittleness prediction using machine learning—A Middle East basin case study. AAPG Bulletin, 106 (11), 2275–2296. https://doi.org/10.1306/12162120181
Niu, D., Li, Y., Zhang, Y., Sun, P., Wu, H., Fu, H., and Wang, Z. (2022). Multi-scale classification and evaluation of shale reservoirs and ‘sweet spot’ prediction of the second and third members of the Qingshankou Formation in the Songliao Basin based on machine learning. Journal of Petroleum Science and Engineering 216, 110678. https://doi.org/10.1016/j.petrol.2022.110678
Ore, T. and Gao, D. (2023). Prediction of reservoir brittleness from geophysical logs using machine learning algorithms. Computers & Geosciences 171, 105266. https://doi.org/10.1016/j.cageo.2022.105266
Pankhurst, R.J., Riley, T.R., Fanning, C.M., and Kelley, S.P. (2000). Episodic silicic volcanism in Patagonia and Antarctic Peninsula: chronology of magmatism associated with the break-up of Gondwana. Journal of Petrology 41, 605–625.
Passey, Q., Creaney, S., Kulla, J., Moretti, F., and Stroud, J. (1990). A practical model for organic richness from porosity and resistivity logs. AAPG Bulletin 74, 1777–1794. https://doi.org/10.1306/0C9B25C9-1710-11D7-8645000102C1865D
Poiré, D., Richiano, S., Varela, A., and Sandoval, P.N. (2007). Descripción detallada de coronas, interpretación paleoambiental y logs numéricos de facies sedimentarias, Proyecto Lomitas. Informe confidencial preparado para Larriestra-Geotecnologías SA y Repsol-YPF SA. Parte I-II.
Prochnow, S., Raterman, N., Swenberg, M., Reddy, L., Smith, I., Romanyuk, M., and Fernadez, T. (2022). A subsurface machine learning approach at hydrocarbon production recovery & resource estimates for unconventional reservoir systems: Making subsurface predictions from multimensional data analysis. Journal of Petroleum Science and Engineering 215, 110598. https://doi.org/10.1016/j.petrol.2022.110598
Richiano, S. (2014). Lower cretaceous anoxic conditions in the Austral basin, south-western Gondwana, Patagonia Argentina. Journal of South American Earth Sciences 54, 37–46.
Richiano, S. (2015). Environmental factors affecting the development of the Zoophycos ichnofacies in the lower cretaceous Río Mayer Formation (Austral basin, Patagonia). Palaeogeography Palaeoclimatology, Palaeoecology https://doi.org/10.1016/j.palaeo.2015.03.029.
Richiano, S., Varela, A.N., Cereceda, A., and Poiré, D.G. (2012). Evolución paleoambiental de la Formación Río Mayer, cretácico inferior, cuenca austral, Patagonia Argentina. Latin American Journal of Sedimentology and Basin Analysis 19, 3–26.
Richiano, S., Poiré, D.G., and Varela, A.N. (2013). Icnología de la Formación Río Mayer, cretácico inferior, SO Gondwana, Patagonia, Argentina. Ameghiniana 50, 273–286.
Richiano, S., Varela, A.N., Gómez-Peral, L.E., Cereceda, A., and Poiré, D.G. (2015). Composition of the lower cretaceous black shales from the Austral basin (Patagonia, Argentina): implication for unconventional reservoirs in the southern Andes. Marine and Petroleum Geology 66, 764–790. http://dx.doi.org/10.1016/j.marpetgeo.2015.07.018
Richiano, S., Varela, A.N., and Poiré, D.G. (2016). Heterogeneous distribution of trace fossils across initial transgressive deposits in rift basins: an example from the Springhill Formation, Argentina. Lethaia 49, 524–539. https://doi.org/10.1111/let.12163
Richiano S., Gómez-Peral, L.E., Varela, A.N., Gómez Dacal, A., Cavarozzi, C., and Poiré, D.G. (2019). Geochemical characterization of black shales from the Río Mayer Formation (Early Cretaceous), Austral-Magallanes Basin, Argentina: Provenance response during Gondwana break-up. Journal of South American Earth Sciences 93, 67–83. https://doi.org/10.1016/j.jsames.2019.04.009
Rojas, C., Jait, D., Aimar, E., Rivero, M., Cevallos, M., Gargiulo, C., and Melendo, F. (2022). Formación Palermo Aike, Caracterización y potencial productivo como reservorio no convencional, Cuenca Austral, Provincia De Santa Cruz, Argentina. 11º Congreso de Exploración y Desarrollo de Hidrocarburos Exploración y Sistemas Petroleros, 73–96.
Rousseeuw, P.J. (1987). Silhouettes: a Graphical Aid to the Interpretation and Validation of Cluster Analysis”. Computational and Applied Mathematics 20, 53–65. https://doi.org/10.1016/0377-0427(87)90125-7
Saporetti, C.M., Fonseca, D.L., Olivera, L.C., Pereira, E., and Goliatt, L. (2022). Hybrid machine learning models for estimating total organic carbon from mineral constituents in core samples of shale gas fields. Marine and Petroleum Geology 143, 105783. https://doi.org/10.1016/j.marpetgeo.2022.105783
Schultz, L.G. (1964). Quantitative interpretation of mineralogical composition from X-ray and chemical data for Pierra Shale. In: U.S. Geological Survey Professional Paper, vol. 391, pp. 1–31.
Tognoli, F. M. W., Spaniol, A., de Mello, M., and de Souza, L. (2024). A machine-learning based approach to predict facies associations and improve local and regional stratigraphic correlations. Marine and Petroleum Geology 160, 106636. https://doi.org/10.1016/j.marpetgeo.2023.106636
Descargas
Publicado
Número
Sección
Licencia
Derechos de autor 2025 Sebastián M. Richiano, Federico Ares
Esta obra está bajo una licencia internacional Creative Commons Atribución-NoComercial 4.0.
