Machine Learning Discovers South American Subduction Zone Hotter than previously Predicted

Abstract

Geothermal heat flow (GHF) is crucial for characterizing the Earth’s thermal state. Compared to other regions worldwide, GHF measurements of South America are relatively sparse for mapping GHF over the continent based on traditional models. Here we apply the machine learning (ML) techniques to predict the GHF in South America. By comparing the global model, ML finds that South American subduction zones are hotter than the global model due to large-scale magmatism, which leads to the higher shallow arc temperatures than canonical thermomechanical and global models. Combining ML model with the local singularity analysis of heat flows, active volcanoes, and igneous rock samples, it is suggested that geothermal anomalies along the Andean Mountain Range are spatially correlated with magmatic activity in the subduction zone. It is concluded that the ML methods may provide reliable GHF prediction in regions like South America, where GHF measurements are limited and uneven.

Keywords

geothermal heat flow / machine learning / local singularity analysis / big data / tectonics

References

[1] AstortA, ColavittoB, SagripantiL, et al.. Crustal and Mantle Structure beneath the Southern Payenia Volcanic Province Using Gravity and Magnetic Data. Tectonics, 2019, 38(1): 144-158

[2] ÁvilaP, DávilaF M. Lithospheric Thinning and Dynamic Uplift Effects during Slab Window Formation, Southern Patagonia (45°–55°S). Journal of Geodynamics, 2020, 133: 101689

[3] BishopB T, BeckS L, ZandtG, et al.. Causes and Consequences of Flat-Slab Subduction in Southern Peru. Geosphere, 2017, 13(5): 1392-1407

[4] BodriL, BodriB. On the Correlation between Heat Flow and Crustal Thickness. Tectonophysics, 1985, 120(1/2): 69-81

[5] BreimanL. Random Forests. Machine Learning, 2001, 45: 5-32

[6] BruneJ N, HenyeyT L, RoyR F. Heat Flow, Stress, and Rate of Slip along the San Andreas Fault, California. Journal of Geophysical Research, 1969, 74(15): 3821-3827

[7] ChenG X, ChengQ M, LyonsT W, et al.. Reconstructing Earth’s Atmospheric Oxygenation History Using Machine Learning. Nature Communications, 2022, 13: 5862

[8] ChenG X, ChengQ M, PetersS E, et al.. Feedback between Surface and Deep processes: Insight from Time Series Analysis of Sedimentary Record. Earth and Planetary Science Letters, 2022, 579: 117352

[9] ChenT, HeT, BenestyM, et al.. Xgboost: Extreme Gradient Boosting. R Package Version 0.4-2, 2015, 1(4): 1-4

[10] ChengQ M. Mapping Singularities with Stream Sediment Geochemical Data for Prediction of Undiscovered Mineral Deposits in Gejiu, Yunnan Province, China. Ore Geology Reviews, 2007, 32(1/2): 314-324

[11] ChengQ M, AgterbergF P. Singularity Analysis of Ore-Mineral and Toxic Trace Elements in Stream Sediments. Computers & Geosciences, 2009, 35(2): 234-244

[12] ChengQ M. Singularity Analysis of Global Zircon U-Pb Age Series and Implication of Continental Crust Evolution. Gondwana Research, 2017, 51: 51-63

[13] ChengQ M. Singularity Analysis of Magmatic Flare-ups Caused by India-Asia Collisions. Journal of Geochemical Exploration, 2018, 189: 25-31

[14] ChengQ M. Extrapolations of Secular Trends in Magmatic Intensity and Mantle cooling: Implications for Future Evolution of Plate Tectonics. Gondwana Research, 2018, 63: 268-273

[15] ChristiansenR O, GianniG M, Ballivián JustinianoC A, et al.. The Role of Geotectonic Setting on the Heat Flow Distribution of Southern South America. Geophysical Journal International, 2022, 230(3): 1911-1927

[16] CordaniU G, RamosV A, FragaL M, et al. Tectonic Map of South America. Second Edition 1: 5 000 000, 2016 Paris CCGM/CGMW Commission for the Geological Map of the World

[17] CortesC, VapnikV. Support-Vector Networks. Machine Learning, 1995, 20(3): 273-297

[18] DaviesJ H, DaviesD R. Earth’s Surface Heat Flux. Solid Earth, 2010, 1(1): 5-24

[19] DaviesJ H. Global Map of Solid Earth Surface Heat Flow. Geochemistry, Geophysics, Geosystems, 2013, 14(10): 4608-4622

[20] FialkoY, PearseJ. Sombrero Uplift above the Altiplano-Puna Magma Body: Evidence of a Ballooning Mid-Crustal Diapir. Science, 2012, 338(6104): 250-252

[21] FriedmanJ, TibshiraniR, HastieT. Additive Logistic Regression: A Statistical View of Boosting (with Discussion and a Rejoinder by the Authors). The Annals of Statistics, 2000, 28(2): 337-407

[22] Friedman, J. H., 2001. Greedy Function Approximation: A Gradient Boosting Machine. The Annals of Statistics, 29(5). https://doi.org/10.1214/aos/1013203451

[23] FriedmanJ H. Stochastic Gradient Boosting. Computational Statistics & Data Analysis, 2002, 38(4): 367-378

[24] FurlongK P, ChapmanD S. Heat Flow, Heat Generation, and the Thermal State of the Lithosphere. Annual Review of Earth and Planetary Sciences, 2013, 41: 385-410

[25] FurukawaY. Depth of the Decoupling Plate Interface and Thermal Structure under Arcs. Journal of Geophysical Research: Solid Earth, 1993, 98(B11): 20005-20013

[26] GianniG M, GarcíaH P A, LupariM, et al.. Plume Overriding Triggers Shallow Subduction and Orogeny in the Southern Central Andes. Gondwana Research, 2017, 49: 387-395

[27] Global Heat Flow Data Assessment Group FuchsS, NeumannF, et al. The Global Heat Flow Database: Release 2024, 2024 [2024-11-9]

[28] Global Volcanism Program [Database] Volcanoes of the World (v. 5.0.4; 17 Apr 2023), 2023 Distributed by Smithsonian Institution, Compiled by Venzke, E

[29] GoutorbeB, PoortJ, LucazeauF, et al.. Global Heat Flow Trends Resolved from Multiple Geological and Geophysical Proxies. Geophysical Journal International, 2011, 187(3): 1405-1419

[30] GroomeW G, ThorkelsonD J. The Three-Dimensional Thermo-Mechanical Signature of Ridge Subduction and Slab Window Migration. Tectonophysics, 2009, 464(1/2/3/4): 70-83

[31] HamzaV M, MuñozM. Heat Flow Map of South America. Geothermics, 1996, 25(6): 599-646

[32] HamzaV M, GomesA J F, FerreiraL E T. Status Report on Geothermal Energy Developments in Brazil. Proceedings World Geothermal Congress 2005, Apr. 24–29, 2005, Antalya, 2005

[33] HeJ F, LiK W, WangX W, et al.. A Machine Learning Methodology for Predicting Geothermal Heat Flow in the Bohai Bay Basin, China. Natural Resources Research, 2022, 31(1): 237-260

[34] HuP Y, ZhaiQ G, CawoodP A, et al.. Detrital Zircon REE and Tectonic Settings. Lithos, 2024, 480: 107661

[35] JonesR D W, KatzR F, TianM, et al.. Thermal Impact of Magmatism in Subduction Zones. Earth and Planetary Science Letters, 2018, 481: 73-79

[36] KelemenP B, RillingJ L, ParmentierE M, et al. EilerJ, et al.. Thermal Structure Due to Solid-State Flow in the Mantle Wedge beneath Arcs. Inside the Subduction Factory, 2003 Washington, D. C. American Geophysical Union 293-311

[37] Lehnert, K., Su, Y., Langmuir, C. H., et al., 2000. A Global Geochemical Database Structure for Rocks. Geochemistry, Geophysics, Geosystems, 1(5). https://doi.org/10.1029/1999gc000026

[38] LiC F, LuY, WangJ. A Global Reference Model of Curie-Point Depths Based on EMAG2. Scientific Reports, 2017, 7: 45129

[39] LiM, HuangS, DongM, et al.. Prediction of Marine Heat Flow Based on the Random Forest Method and Geological and Geophysical Features. Marine Geophysical Research, 2021, 42(3): 30

[40] LösingM, EbbingJ. Predicting Geothermal Heat Flow in Antarctica with a Machine Learning Approach. Journal of Geophysical Research: Solid Earth, 2021, 126(6): e2020JB021499

[41] LucazeauF. Analysis and Mapping of an Updated Terrestrial Heat Flow Data Set. Geochemistry, Geophysics, Geosystems, 2019, 20(8): 4001-4024

[42] MajorowiczJ, GrasbyS E. Heat Flow, Depth-Temperature Variations and Stored Thermal Energy for Enhanced Geothermal Systems in Canada. Journal of Geophysics and Engineering, 2010, 7(3): 232-241

[43] McGaryR S, EvansR L, WannamakerP E, et al.. Pathway from Subducting Slab to Surface for Melt and Fluids beneath Mount Rainier. Nature, 2014, 511(7509): 338-340

[44] NavarreteC, GianniG, MassaferroG, et al.. The Fate of the Farallon Slab beneath Patagonia and Its Links to Cenozoic Intraplate Magmatism, Marine Transgressions and Topographic Uplift. Earth-Science Reviews, 2020, 210: 103379

[45] PedregosaF, VaroquauxG, GramfortA, et al.. Scikit-Learn: Machine Learning in Python. The Journal of Machine Learning Research, 2011, 12: 2825-2830

[46] PerkinsJ P, WardK M, de SilvaS L, et al.. Surface Uplift in the Central Andes Driven by Growth of the Altiplano Puna Magma Body. Nature Communications, 2016, 7: 13185

[47] PerrinA, GoesS, PrytulakJ, et al.. Reconciling Mantle Wedge Thermal Structure with Arc Lava Thermobarometric Determinations in Oceanic Subduction Zones. Geochemistry, Geophysics, Geosystems, 2016, 17(10): 4105-4127

[48] RezvanbehbahaniS, StearnsL A, KadivarA, et al.. Predicting the Geothermal Heat Flux in Greenland: A Machine Learning Approach. Geophysical Research Letters, 2017, 44(24): 12271-12279

[49] RosenbaumG, CaulfieldJ T, UbideT, et al.. Spatially and Geochemically Anomalous Arc Magmatism: Insights from the Andean Arc. Geochemistry, Geophysics, Geosystems, 2021, 22(6): e2021GC009688

[50] Russo, R. M., VanDecar, J. C., Comte, D., et al., 2010. Subduction of the Chile Ridge: Upper Mantle Structure and Flow. GSA Today: 4–10. https://doi.org/10.1130/gsatg61a.1

[51] Rychert, C. A., Fischer, K. M., Abers, G. A., et al., 2008. Strong Along-Arc Variations in Attenuation in the Mantle Wedge beneath Costa Rica and Nicaragua. Geochemistry, Geophysics, Geosystems, 9(10). https://doi.org/10.1029/2008gc002040

[52] SternC R. Active Andean Volcanism: Its Geologic and Tectonic Setting. Revista Geológica de Chile, 2004, 31(2): 161-206

[53] Syracuse, E. M., Abers, G. A., Fischer, K., et al., 2008. Seismic Tomography and Earthquake Locations in the Nicaraguan and Costa Rican Upper Mantle. Geochemistry, Geophysics, Geosystems, 9(7). https://doi.org/10.1029/2008gc001963

[54] ThorkelsonD J. Subduction of Diverging Plates and the Principles of Slab Window Formation. Tectonophysics, 1996, 255(1/2): 47-63

[55] ValdenegroP, MuñozM, YáñezG, et al.. A Model for Thermal Gradient and Heat Flow in Central Chile: The Role of Thermal Properties. Journal of South American Earth Sciences, 2019, 91: 88-101

[56] VapnikV The Nature of Statistical Learning Theory, 1999

[57] VölkerD, KutterolfS, WehrmannH. Comparative Mass Balance of Volcanic Edifices at the Southern Volcanic Zone of the Andes between 33°S and 46°S. Journal of Volcanology and Geothermal Research, 2011, 205(3/4): 114-129

[58] XuS, NiC, HuX Y. Predicting Terrestrial Heat Flow in North China Using Multiple Geological and Geophysical Datasets Based on Machine Learning Method. Energies, 2023, 16(4): 1620

[59] ZhaoD P, WangZ, UminoN, et al.. Tomographic Imaging Outside a Seismic Network: Application to the Northeast Japan Arc. Bulletin of the Seismological Society of America, 2007, 97(4): 1121-1132

[60] ZhouC H, WangH, WangC S, et al.. Geoscience Knowledge Graph in the Big Data Era. Science China Earth Sciences, 2021, 64(7): 1105-1114

[61] ZhouS Y, ChuX, CaoS B, et al.. Prediction of the Ground Temperature with ANN, LS-SVM and Fuzzy LS-SVM for GSHP Application. Geothermics, 2020, 84: 101757

[62] ZhouY Z, ChengQ M, LiuY, et al.. Singularity Analysis of Igneous Zircon U-Pb Age and Hf Isotopic Record in the Zhongdian Arc, Northwest Yunnan, China: Implications for Indosinian Magmatic Flare-up and the Formation of Porphyry Copper Deposits. Ore Geology Reviews, 2021, 139: 104476

[63] ZhuZ Y, CampbellI H, AllenC M, et al.. The Temporal Distribution of Earth’s Supermountains and Their Potential Link to the Rise of Atmospheric Oxygen and Biological Evolution. Earth and Planetary Science Letters, 2022, 580: 117391

RIGHTS & PERMISSIONS

China University of Geosciences (Wuhan) and Springer-Verlag GmbH Germany, Part of Springer Nature