JSEJournal of Seismic Exploration0963-0651AccScience Publishing10.36922/JSE025290036ArticleMicroseismic event picking and classification for hot dry rock hydraulic fracturing monitoring using SeisFormerhttps://geophysical-press.com/journal/JSE/34/6/10.36922/JSE025290036OuyangMingjun,LengZenan,HuHaotian,ChenZubin,ZhaoFa,SunFeng20253462025-12-02Accurate seismic monitoring is vital for the safe operation of enhanced geothermal systems in hot dry rock (HDR) reservoirs; however, robust P- and S-wave classification and precise first-arrival picking remain difficult under low signal-to-noise ratios and complex noise conditions. Hence, in this study, we present SeisFormer, a Transformer-based framework that couples adaptive multi-scale windowing with joint time–frequency analysis. It allocates time–frequency resolution on the fly to overcome the limitations of a fixed-window short-time Fourier transform and slowly extracts varying trends and dominant periodicities from waveform sequences. To stabilize the modeling of long-range dependencies, we introduce regularized pseudoinverse attention, which retains the speedups of low-rank approximations while damping amplification in directions associated with small singular values. We evaluated SeisFormer on a unified, multi-site dataset with data from HDR operations in the Qinghai Gonghe Basin and from an unconventional hydraulic-fracturing field in North China. Compared with baselines (EQTransformer, PhaseNet), it exhibited better performance across real-world data, noise-augmented data with non-stationary composite noise, and overlapping multi-event scenarios. On real-world data, it attained 98.30% classification accuracy, with mean arrival-time errors of 1.42 ms (P) and 2.29 ms (S). Ablations show that each component contributes substantially, indicating robustness for near-real-time monitoring and deployment.Microseismic monitoring, Hot dry rock hydraulic fracturing, Picking and classification, Transformer, Adaptive multi-scale windowing, Time–frequency domain

Dong S, Jiao J, Zhou S, Lu P, Zeng Z. 3-D gravity data inversion based on enhanced dual U-Net framework. IEEE Trans Geosci Remote Sens. 2023;61:1-11. doi: 10.1109/TGRS.2023.3306980   Cloetingh S, Sternai P, Koptev A, et al. Coupled surface to deep Earth processes: Perspectives from TOPO-EUROPE with an emphasis on climate- and energy-related societal challenges. Global Planet Change. 2023;226:104140. doi: 10.1016/j.gloplacha.2023.104140   Ma W, Wang Y, Wu X, Liu G. Hot dry rock (HDR) hydraulic fracturing propagation and impact factors assessment via sensitivity indicator. Renew Energy. 2020;146:2716-2723. doi: 10.1016/j.renene.2019.08.097   Xie J, Cao H, Wang D, Peng S, Fu G, Zhu Z. A comparative study on the hydraulic fracture propagation behaviors in hot dry rock and shale formation with different structural discontinuities. J Energy Eng. 2022;148(6):04022040. doi: 10.1061/(asce)ey.1943-7897.0000856   Cheng Y, Zhang Y, Yu Z, Hu Z, Yang Y. An investigation on hydraulic fracturing characteristics in granite geothermal reservoir. Eng Fract Mech. 2020;237:107252. doi: 10.1016/j.engfracmech.2020.107252   Cheng Y, Zhang Y, Yu Z, Hu Z. Investigation on reservoir stimulation characteristics in hot dry rock geothermal formations of China during hydraulic fracturing. Rock Mech Rock Eng. 2021;54(8):3817-3845. doi: 10.1007/s00603-021-02506-y   Cheng Y, Zhang Y. Experimental study of fracture propagation: The application in energy mining. Energies. 2020;13(6):1411. doi: 10.3390/en13061411   Wang H, Alkhalifah T, Waheed UB, Birnie C. Data-driven microseismic event localization: An application to the Oklahoma Arkoma Basin hydraulic fracturing data. IEEE Trans Geosci Remote Sens. 2022;60:1-12. doi: 10.1109/TGRS.2021.3120546   Warpinski N. Microseismic monitoring: Inside and out. J Petrol Technol. 2009;61(11):80-85. doi: 10.2118/118537-JPT   Allen RV. Automatic earthquake recognition and timing from single traces. Bull Seismol Soc Am. 1978;68(5):1521-1532. doi: 10.1785/bssa0680051521   Akaike H. A new look at the statistical model identification. IEEE Trans Autom Control. 1974;19(6):716-723. doi: 10.1109/tac.1974.1100705   Allen R. Automatic phase pickers: Their present use and future prospects. Bull Seismol Soc Am. 1982;72(6B):S225-S242. doi: 10.1785/BSSA07206B0225   Álvarez I, Garcia L, Mota S, et al. An automatic P-phase picking algorithm based on adaptive multiband processing. IEEE Geosci Remote Sens Lett. 2013;10(6):1488-1492. doi: 10.1109/lgrs.2013.2260720   Earle PS, Shearer PM. Characterization of global seismograms using an automatic-picking algorithm. Bull Seismol Soc Am. 1994;84(2):366-376. doi: 10.1785/bssa0840020366   Akazawa TA. Technique for Automatic Detection of Onset Time of P- and S-Phases in Strong Motion Records. In: Proceedings 13th Conference on Earthquake Engineering. Vancouver, Canada. Paper No. 786; 2004.   Kurz JH, Grosse CU, Reinhardt HW. Strategies for reliable automatic onset time picking of acoustic emissions and of ultrasound signals in concrete. Ultrasonics. 2005;43(7):538-546. doi: 10.1016/j.ultras.2004.12.005   Anikiev D, Birnie C, Waheed UB, et al. Machine learning in microseismic monitoring. Earth Sci Rev. 2023;239:104371. doi: 10.1016/j.earscirev.2023.104371   LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015;521:436-444. doi: 10.1038/nature14539   Mousavi SM, Beroza GC. Deep-learning seismology. Science. 2022;377:725-729. doi: 10.1126/science.abm4470   Zhu W, Beroza GC. PhaseNet: A deep-neural-network-based seismic arrival-time picking method. Geophys J Int. 2019;216(1):261-273. doi: 10.1093/gji/ggy423   He Z, Peng P, Wang L, Jiang Y. PickCapsNet: Capsule network for automatic P-wave arrival picking. IEEE Geosci Remote Sens Lett. 2021;18(4):617-621. doi: 10.1109/lgrs.2020.2983196   Zhao, Y, Xu, H, Yang, T, Wang D, Sun D. A hybrid recognition model of microseismic signals for underground mining based on CNN and LSTM networks. Geomat Nat Hazard Risk. 2021;12(1):2803-2834. doi: 10.1080/19475705.2021.1968043   Mousavi SM, Ellsworth WL, Zhu W, Chuang LY, Beroza GC. Earthquake transformer: An attentive deep-learning model for simultaneous earthquake detection and phase picking. Nat Commun. 2020;11:3952. doi: 10.1038/s41467-020-17591-w   Saad OM, Chen Y, Siervo D, et al. EQCCT: A production-ready earthquake detection and phase-picking method using the compact convolutional transformer. IEEE Trans Geosci Remote Sens. 2023;61:1-15. doi: 10.1109/TGRS.2023.3319440   Li S, Yang X, Cao A, et al. SeisT: A foundational deep-learning model for earthquake monitoring tasks. IEEE Trans Geosci Remote Sens. 2024;62:1-15. doi: 10.1109/tgrs.2024.3371503   Li XN, Chen FJ, Lai YP, Tang P, Liang X. ICAT-net: A lightweight neural network with optimized coordinate attention and transformer mechanisms for earthquake detection and phase picking. J Supercomput. 2025;81:191. doi: 10.1007/s11227-024-06664-y   Peng P, Lei R, Wang JM. Enhancing microseismic signal classification in metal mines using transformer-based deep learning. Sustainability. 2023;15(20):14959.doi: 10.3390/su152014959   Zhang X, Wang X, Zhang Z, Wang Z. CNN-transformer for microseismic signal classification. Electronics. 2023;12(11):2468. doi: 10.3390/electronics12112468   Zhu L, Chuang L, McClellan JH, Liu E, Peng Z. A Multi- Channel Approach for Automatic Microseismic Event Association using RANSAC-Based Arrival Time Event Clustering (RATEC). [arXiv Preprint]; 2017. doi: 10.48550/arXiv.1702.01856   Li Z, Gou X, Jin W, Qin N. Frequency features of microseismic signals. Chin J Geotech Eng. 2008;30(6):830-834.   Aki K, Richards PG. Quantitative Seismology. 2nd ed. Sausalito, CA: University Science Books; 2002.   Ma C, Ran X, Xu W, et al. Fine classification method for massive microseismic signals based on short-time Fourier transform and deep learning. Remote Sens. 2023;15(2):502. doi: 10.3390/rs15020502   Xiong Y, Zeng Z, Chakraborty R, et al. Nyströmformer: A Nyström-based algorithm for approximating self-attention. Proc AAAI Conf Artif Intell. 2021;35(16):14138-14148. doi: 10.48550/arXiv.2102.03902   Piao X, Chen Z, Murayama T, et al. Fredformer: Frequency Debiased Transformer for Time Series Forecasting. In: Proceedings 30th ACM SIGKDD Conference Knowledge Discovery and Data Mining (KDD 2024). Long Beach, CA, USA; 2024. p. 1234-1245. doi: 10.1145/3637528.3671928   Soto-Quiros P. A fast method to estimate the Moore- Penrose inverse for well-determined numerical rank matrices based on Tikhonov regularization. J Math Comput Sci. 2024;37(1):59-81. doi: 10.22436/jmcs.037.01.05   Longhas PRA, Abdul AM. Application of Tikhonov regularization in generalized inverse of adjacency matrix of undirected graph. Int J Math Trends Technol. 2022;68(2):1-6. doi: 10.14445/22315373/ijmtt-v68i2p501   Gulliksson ME, Wedin PÅ, Wei Y. Perturbation identities for regularized Tikhonov inverses and weighted pseudoinverses. BIT Numer Math. 2000;40:513-523. doi: 10.1023/A:1022319830134   Ross ZE, Meier MA, Hauksson E, Heaton TH. Generalized seismic phase detection with deep learning. Bull Seismol Soc Am. 2018;108(5A):2894-2901. doi: 10.1785/0120180080