{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T04:54:35Z","timestamp":1787028875022,"version":"3.56.0"},"reference-count":40,"publisher":"MDPI AG","issue":"4","license":[{"start":{"date-parts":[[2022,10,7]],"date-time":"2022-10-07T00:00:00Z","timestamp":1665100800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"UEFISCDI","award":["PN-III-P2-2.1-PED-2019-1693"],"award-info":[{"award-number":["PN-III-P2-2.1-PED-2019-1693"]}]},{"name":"UEFISCDI","award":["PN19080102"],"award-info":[{"award-number":["PN19080102"]}]},{"DOI":"10.13039\/501100005802","name":"MCI","doi-asserted-by":"publisher","award":["PN-III-P2-2.1-PED-2019-1693"],"award-info":[{"award-number":["PN-III-P2-2.1-PED-2019-1693"]}],"id":[{"id":"10.13039\/501100005802","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100005802","name":"MCI","doi-asserted-by":"publisher","award":["PN19080102"],"award-info":[{"award-number":["PN19080102"]}],"id":[{"id":"10.13039\/501100005802","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["MAKE"],"abstract":"<jats:p>We designed a convolutional neural network application to detect seismic precursors in geomagnetic field records. Earthquakes are among the most destructive natural hazards on Earth, yet their short-term forecasting has not been achieved. Stress loading in dry rocks can generate electric currents that cause short-term changes to the geomagnetic field, yielding theoretically detectable pre-earthquake electromagnetic emissions. We propose a CNN model that scans windows of geomagnetic data streams and self-updates using nearby earthquakes as labels, under strict detectability criteria. We show how this model can be applied in three key seismotectonic settings, where geomagnetic observatories are optimally located in high-seismicity-rate epicentral areas. CNNs require large datasets to be able to accurately label seismic precursors, so we expect the model to improve as more data become available with time. At present, there is no synthetic data generator for this kind of application, so artificial data augmentation is not yet possible. However, this deep learning model serves to illustrate its potential usage in earthquake forecasting in a systematic and unbiased way. Our method can be prospectively applied to any kind of three-component dataset that may be physically connected to seismogenic processes at a given depth.<\/jats:p>","DOI":"10.3390\/make4040046","type":"journal-article","created":{"date-parts":[[2022,10,9]],"date-time":"2022-10-09T01:43:11Z","timestamp":1665279791000},"page":"912-923","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":11,"title":["Prospective Neural Network Model for Seismic Precursory Signal Detection in Geomagnetic Field Records"],"prefix":"10.3390","volume":"4","author":[{"given":"Laura","family":"Petrescu","sequence":"first","affiliation":[{"name":"National Institute for Earth Physics, 077125 M\u0103gurele, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8199-8594","authenticated-orcid":false,"given":"Iren-Adelina","family":"Moldovan","sequence":"additional","affiliation":[{"name":"National Institute for Earth Physics, 077125 M\u0103gurele, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2022,10,7]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"679","DOI":"10.1785\/gssrl.71.6.679","article-title":"The GSHAP global seismic hazard map","volume":"71","author":"Shedlock","year":"2000","journal-title":"Seismol. Res. Lett."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"371","DOI":"10.1016\/j.tecto.2009.06.008","article-title":"A systematic compilation of earthquake precursors","volume":"476","author":"Cicerone","year":"2009","journal-title":"Tectonophysics"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"992","DOI":"10.1002\/2013JA019530","article-title":"On the use of geomagnetic indices and ULF waves for earthquake precursor signatures","volume":"119","author":"Currie","year":"2014","journal-title":"Geophys. Res. Space Phys."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Campbell, W.H. (2003). Introduction to Geomagnetic Fields, Cambridge University Press. [2nd ed.].","DOI":"10.1017\/CBO9781139165136"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"St\u0103nic\u0103, D.A., and St\u0103nic\u0103, D. (2021). Possible Correlations between the ULF Geomagnetic Signature and Mw6. 4 Coastal Earthquake, Albania, on 26 November 2019. Entropy, 23.","DOI":"10.3390\/e23020233"},{"key":"ref_6","first-page":"1","article-title":"Correlation of geomagnetic anomalies recorded at Muntele Rosu Seismic Observatory (Romania) with earthquake occurrence and solar magnetic storms","volume":"55","author":"Moldovan","year":"2012","journal-title":"Ann. Geophys."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"59","DOI":"10.1016\/j.jseaes.2013.08.011","article-title":"Anomalous behaviors of geomagnetic diurnal variations prior to the 2011 off the Pacific coast of Tohoku earthquake (Mw9. 0)","volume":"77","author":"Xu","year":"2013","journal-title":"J. Asian Earth Sci."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"25","DOI":"10.2183\/pjab.91.25","article-title":"Criticality features in ULF magnetic field prior to the 2011 Tohoku earthquake","volume":"91","author":"Hayakawa","year":"2015","journal-title":"Proc. Jpn. Acadademy Ser."},{"key":"ref_9","first-page":"84","article-title":"Imagenet classification with deep convolutional neural networks","volume":"25","author":"Krizhevsky","year":"2012","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"10289","DOI":"10.1002\/2015JA021336","article-title":"Are there new findings in the search for ULF magnetic precursors to earthquakes?","volume":"120","author":"Masci","year":"2015","journal-title":"J. Geophys. Res. Space Phys."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"389","DOI":"10.1016\/j.pce.2006.02.027","article-title":"Electric currents streaming out of stressed igneous rocks\u2014A step towards understanding preearthquake low frequency EM emissions","volume":"31","author":"Freund","year":"2006","journal-title":"Phys. Chem. Earth"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"2662","DOI":"10.1785\/0120140144","article-title":"Comparison of the stress-stimulated current of dry and fluid-saturated gabbro samples","volume":"104","author":"Dahlgren","year":"2014","journal-title":"Bull. Seism. Soc. Am."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"1127","DOI":"10.1029\/91GL01000","article-title":"ULF magnetic signatures at the Earth\u2019s surface due to ground water flow: A possible precursor to earthquakes","volume":"18","author":"Draganov","year":"1991","journal-title":"Geophys. Res. Lett."},{"key":"ref_14","first-page":"585","article-title":"Tectonomagnetic modeling on the basis of the linear piezomagnetic effect","volume":"66","author":"Sasai","year":"1991","journal-title":"Bull. Earthq. Res. Inst. Univ. Tokyo"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"771","DOI":"10.5194\/npg-20-771-2013","article-title":"Current challenges for pre-earthquake electromagnetic emissions: Shed- ding light from micro-scale plastic flow, granular packings, phase transitions and self-affinity notion of fracture process","volume":"20","author":"Eftaxias","year":"2013","journal-title":"Nonlinear Processes Geophys."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1016\/j.tecto.2006.05.027","article-title":"Surface oscillations: A possible source of fracture induced electromagnetic radiation","volume":"431","author":"Rabinovitch","year":"2007","journal-title":"Tectonophysics"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"045901","DOI":"10.1088\/0031-8949\/79\/04\/045901","article-title":"Discrimination between pre-seismic electromagnetic anomalies and solar activity effects","volume":"79","author":"Koulouras","year":"2009","journal-title":"Phys. Scr."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1126\/science.1194777","article-title":"The dynamic of the onset of frictional slip","volume":"330","author":"Cohen","year":"2010","journal-title":"Science"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"992","DOI":"10.1017\/S0016756817000954","article-title":"Use of electromagnetic radiation for potential forecast of earthquakes","volume":"155","author":"Rabinovitch","year":"2018","journal-title":"Geol. Mag."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"1691","DOI":"10.1029\/94JA02524","article-title":"Penetration characteristics of electromagnetic emissions from an underground seismic source into the atmosphere, ionosphere, and magnetosphere","volume":"100","author":"Molchanov","year":"1995","journal-title":"J. Geophys. Res. Space Phys."},{"key":"ref_21","first-page":"1","article-title":"Principal component analysis of ULF geomagnetic data for Izu islands earthquakes in July 2000","volume":"22","author":"Gotoh","year":"2002","journal-title":"J. Atmos. Electr."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"61","DOI":"10.1016\/j.tecto.2006.05.030","article-title":"A multiple fracture model of pre-seismic electromagnetic phenomena","volume":"431","author":"Morgunov","year":"2007","journal-title":"Tectonophysics"},{"key":"ref_23","unstructured":"Simonyan, K., and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"632","DOI":"10.1038\/s41586-018-0438-y","article-title":"Deep learning of aftershock patterns following large earthquakes","volume":"560","author":"DeVries","year":"2018","journal-title":"Nature"},{"key":"ref_25","first-page":"0328","article-title":"Neural network approach to the prediction of seismic events based on low-frequency signal monitoring of the Kuril-Kamchatka and Japanese regions","volume":"56","author":"Popova","year":"2013","journal-title":"Ann. Geophys."},{"key":"ref_26","first-page":"0543","article-title":"Detection of gravity changes before powerful earthquakes in GRACE satellite observations","volume":"57","author":"Shahrisvand","year":"2014","journal-title":"Ann. Geophys."},{"key":"ref_27","unstructured":"Chollet, F. (2022, March 01). Keras. Available online: https:\/\/github.com\/fchollet\/keras."},{"key":"ref_28","unstructured":"Goodfellow, I., Bengio, Y., and Courville, A. (2017). Deep Learning (Adaptive Computation and Machine Learning Series), The MIT Press."},{"key":"ref_29","unstructured":"Kingma, D.P., and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"251","DOI":"10.1016\/j.tecto.2004.09.013","article-title":"A new view of Italian seismicity using 20 years of instrumental recordings","volume":"395","author":"Chiarabba","year":"2005","journal-title":"Tectonophysics"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1146\/annurev.earth.32.101802.120415","article-title":"The North Anatolian fault: A new look","volume":"33","author":"Imren","year":"2005","journal-title":"Annu. Rev. Earth Planet. Sci."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Wenzel, F., Lorenz, F., Sperner, B., and Oncescu, M. (1999). Seismotectonics of the Romanian Vrancea area. Vrancea Earthquakes: Tectonics, Hazard and Risk Mitigation, Springer.","DOI":"10.1007\/978-94-011-4748-4"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Oncescu, M.C., Marza, V.I., Rizescu, M., and Popa, M. (1999). The Romanian Earthquake Catalogue Between 984\u20131997. Vrancea Earthquakes: Tectonics, Hazard and Risk Mitigation, Springer.","DOI":"10.1007\/978-94-011-4748-4_4"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"18","DOI":"10.1186\/s40562-020-00164-6","article-title":"Rebuild of the Bulletin of the International Seismological Centre (ISC)\u2014Part 2: 1980\u20132010","volume":"7","author":"Storchak","year":"2020","journal-title":"Geosci. Lett."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"e2020SW002641","DOI":"10.1029\/2020SW002641","article-title":"The geomagnetic Kp index and derived indices of geomagnetic activity","volume":"19","author":"Matzka","year":"2021","journal-title":"Space Weather"},{"key":"ref_36","unstructured":"Yu, T., and Zhu, H. (2020). Hyper-parameter optimization: A review of algorithms and applications. arXiv."},{"key":"ref_37","unstructured":"Cho, J., Lee, K., Shin, E., Choy, G., and Do, S. (2015). How much data is needed to train a medical image deep learning system to achieve necessary high accuracy?. arXiv."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1007\/s11263-015-0816-y","article-title":"ImageNet Large Scale Visual Recognition challenge","volume":"115","author":"Russakovsky","year":"2015","journal-title":"Int. J. Comput. Vis."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Figueroa, R.L., Zeng-Treitler, Q., Kandula, S., and Ngo, L.H. (2012). Predicting sample size required for classification performance. BMC Med. Inform. Decis. Mak., 12.","DOI":"10.1186\/1472-6947-12-8"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Jaipuria, N., Zhang, X., Bhasin, R., Arafa, M., Chakravarty, P., Shrivastava, S., Manglani, S., and Murali, V.N. (2020, January 14\u201319). Deflating Dataset Bias using Synthetic Data Augmentation. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition Workshops, Virtual.","DOI":"10.1109\/CVPRW50498.2020.00394"}],"container-title":["Machine Learning and Knowledge Extraction"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2504-4990\/4\/4\/46\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T00:47:48Z","timestamp":1760143668000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2504-4990\/4\/4\/46"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,10,7]]},"references-count":40,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2022,12]]}},"alternative-id":["make4040046"],"URL":"https:\/\/doi.org\/10.3390\/make4040046","relation":{},"ISSN":["2504-4990"],"issn-type":[{"value":"2504-4990","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,10,7]]}}}