Deep learning for denoising High-Rate Global Navigation Satellite System data


  • Amanda Thomas Department of Earth Sciences, University of Oregon
  • Diego Melgar Department of Earth Sciences, University of Oregon
  • Sydney N. Dybing Department of Earth Sciences, University of Oregon
  • Jacob R. Searcy Data Science Initiative, University of Oregon



GNSS, deep learning, denoising


High-rate global navigation satellite system (HR-GNSS) data records ground displacements and can be used to identify earthquakes and slow slip events.  One limitation of such data is the high amplitude, cm-level noise which make it difficult to identify processes that produce surface displacements smaller than these values.  Deep learning has proven adept at performing many useful tasks in seismology and geophysics.  Here we explore using deep learning to denoise HR-GNSS data.  We develop three different convolutional neural networks with similar architectures but different targets.  Training data are synthetic HR-GNSS records and actual noise recordings that are superimposed to generate noisy signals.  We train each of the three models to output masks that can be used to reconstruct the true signal.  We use a set of performance metrics that quantify the models’ ability to denoise the testing data and find that denoising significantly improves the signal-to-noise ratio and the ability to identify first arrivals.  Finally, we test the models on HR-GNSS records from the Ridgecrest earthquakes recorded at stations that have nearly colocated strong-motion sites used ground-truth the denoising results.  We find that the models greatly improve the signal-to-noise ratios in these records and make the P-wave onset clearly identifiable.


Abadi, M., et al. “Tensorflow: Large-Scale Machine Learning on Heterogeneous Distributed Systems.” Preprint, 2016, doi:10.48550/arXiv.1603.04467.

Agarap, A. F. “Deep Learning Using Rectified Linear Units (ReLU).” Preprint, 2018, doi:10.48550/arXiv.1803.08375.

Beyreuther, M., et al. “ObsPy: A Python Toolbox for Seismology.” Seismological Research Letters, vol. 81, no. 3, 2010, pp. 530–33, doi:10.1785/gssrl.81.3.530.

Bock, Y., and D. Melgar. “Physical Applications of GPS Geodesy: A Review.” Reports on Progress in Physics, vol. 79, no. 10, 2016, p. 10 1088 0034-4885 79 10 106801, doi:10.1088/0034-4885/79/10/106801.

Choi, K., et al. “Modified Sidereal Filtering: Implications for High‐rate GPS Positioning.” Geophysical Research Letters, vol. 31, no. 22, 2004, doi:10.1029/2004GL021621.

Chollet, F. Keras. 2015,

Crowell, B. W., et al. “Earthquake Magnitude Scaling Using Seismogeodetic Data.” Geophysical Research Letters, vol. 40, no. 23, 2013, pp. 6089–94, doi:10.1002/2013GL058391.

Dong, D., et al. “Spatiotemporal Filtering Using Principal Component Analysis and Karhunen‐Loeve Expansion Approaches for Regional GPS Network Analysis.” Journal of Geophysical Research: Solid Earth, vol. 111, no. B3, 2006, doi:10.1029/2005JB003806.

Ende, M., et al. “A Self-Supervised Deep Learning Approach for Blind Denoising and Waveform Coherence Enhancement in Distributed Acoustic Sensing Data.” IEEE Transactions on Neural Networks and Learning Systems, 2021, doi:10.1109/TNNLS.2021.3132832.

Frank, W. B. “Slow Slip Hidden in the Noise: The Intermittence of Tectonic Release.” Geophysical Research Letters, vol. 43, no. 19, 2016, pp. 10–125, doi:10.1002/2016GL069537.

Geng, J., P. Jiang, et al. “Integrating GPS with GLONASS for High‐rate Seismogeodesy.” Geophysical Research Letters, vol. 44, no. 7, 2017, pp. 3139–46, doi:10.1002/2017GL072808.

Geng, J., Y. Pan, et al. “Noise Characteristics of High‐rate Multi‐GNSS for Subdaily Crustal Deformation Monitoring.” Journal of Geophysical Research: Solid Earth, vol. 123, no. 2, 2018, pp. 1987–2002, doi:10.1002/2018JB015527.

Goldberg, D. E., et al. “Complex Rupture of an Immature Fault Zone: A Simultaneous Kinematic Model of the 2019 Ridgecrest, CA Earthquakes.” Geophysical Research Letters, vol. 47, no. 3, 2020, p. 2019 086382, doi:10.1029/2019GL086382.

Goulet, C. A., et al. “The SCEC Broadband Platform Validation Exercise: Methodology for Code Validation in the Context of Seismic‐hazard Analyses.” Seismological Research Letters, vol. 86, no. 1, 2015, pp. 17–26, doi:10.1785/0220140104.

Graves, R., and A. Pitarka. “Refinements to the Graves and Pitarka (2010) Broadband Ground‐motion Simulation Method.” Seismological Research Letters, vol. 86, no. 1, 2015, pp. 75–80, doi:10.1785/0220140101.

Graves, R. W., and A. Pitarka. “Broadband Ground-Motion Simulation Using a Hybrid Approach.” Bulletin of the Seismological Society of America, vol. 100, no. 5A, 2010, pp. 2095–123, doi:10.1785/0120100057.

He, X., et al. “Accuracy Enhancement of GPS Time Series Using Principal Component Analysis and Block Spatial Filtering.” Advances in Space Research, vol. 55, no. 5, 2015, pp. 1316–27, doi:10.1016/j.asr.2014.12.016.

Hodgkinson, K. M., et al. “Evaluation of Earthquake Magnitude Estimation and Event Detection Thresholds for Real‐time GNSS Networks: Examples from Recent Events Captured by the Network of the Americas.” Seismological Research Letters, vol. 91, no. 3, 2020, pp. 1628–45, doi:10.1785/0220190269.

Hunter, J. D. “Matplotlib: A 2D Graphics Environment.” Computing in Science and Engineering, vol. 9, no. 3, May 2007, pp. 90-95, doi:10.1109/MCSE.2007.55.

Kingma, D. P., and J. Ba. “Adam: A Method for Stochastic Optimization.” Preprint, 2014, doi:10.48550/arXiv.1412.6980.

Krischer, L., et al. “ObsPy: A Bridge for Seismology into the Scientific Python Ecosystem.” Computational Science and Discovery, vol. 8, no. 1, 2015, p. 14003, doi:10.1088/1749-4699/8/1/014003.

Larson, K. M., et al. “Improving the Precision of High‐rate GPS.” Journal of Geophysical Research: Solid Earth, vol. 112, no. B5, 2007, doi:10.1029/2006JB004367.

---. “Unanticipated Uses of the Global Positioning System.” Annual Review of Earth and Planetary Sciences, vol. 47, 2019, pp. 19–40, doi:10.1146/annurev-earth-053018-060203.

Lay, T. “A Review of the Rupture Characteristics of the 2011 Tohoku-Oki Mw 9.1 Earthquake.” Tectonophysics, vol. 733, 2018, pp. 4–36, doi:10.1016/j.tecto.2017.09.022.

Li, Y., et al. “A Data-Driven Approach for Denoising GNSS Position Time Series.” Journal of Geodesy, vol. 92, 2018, pp. 905–22, doi:10.1007/s00190-017-1102-2.

Lin, J. T., et al. “Early Warning for Great Earthquakes from Characterization of Crustal Deformation Patterns with Deep Learning.” Journal of Geophysical Research: Solid Earth, vol. 126, no. 10, 2021, doi:10.1029/2021JB022703.

Mai, P. M., and G. C. Beroza. “A Spatial Random Field Model to Characterize Complexity in Earthquake Slip.” Journal of Geophysical Research: Solid Earth, vol. 107, no. B11, 2002, p. 10, doi:10.1029/2001JB000588.

Melbourne, T. I., et al. “Global Navigational Satellite System Seismic Monitoring.” Bulletin of the Seismological Society of America, vol. 111, no. 3, 2021, pp. 1248–62, doi:10.1785/0120200356.

Melgar, D., B. W. Crowell, J. Geng, et al. “Earthquake Magnitude Calculation without Saturation from the Scaling of Peak Ground Displacement.” Geophysical Research Letters, vol. 42, no. 13, 2015, pp. 5197–205, doi:10.1002/2015GL064278.

Melgar, D., R. J. LeVeque, et al. “Kinematic Rupture Scenarios and Synthetic Displacement Data: An Example Application to the Cascadia Subduction Zone.” Journal of Geophysical Research: Solid Earth, vol. 121, no. 9, 2016, pp. 6658–74, doi:10.1002/2016JB013314.

Melgar, D., B. W. Crowell, T. I. Melbourne, et al. “Noise Characteristics of Operational Real‐time High‐rate GNSS Positions in a Large Aperture Network.” Journal of Geophysical Research: Solid Earth, vol. 125, no. 7, 2020, p. 2019 019197, doi:10.1029/2019JB019197.

Melgar, D., Y. Bock, et al. “On Robust and Reliable Automated Baseline Corrections for Strong Motion Seismology.” Journal of Geophysical Research: Solid Earth, vol. 118, no. 3, 2013, pp. 1177–87, doi:10.1002/jgrb.50135.

Melgar, D., and G. P. Hayes. “The Correlation Lengths and Hypocentral Positions of Great EarthquakesThe Correlation Lengths and Hypocentral Positions of Great Earthquakes.” Bulletin of the Seismological Society of America, vol. 109, no. 6, 2019, pp. 2582–93, doi:10.1002/2016JB013314.

Murray, J. R., et al. “Development of a Geodetic Component for the US West Coast Earthquake Early Warning System.” Seismological Research Letters, vol. 89, no. 6, 2018, pp. 2322–36, doi:10.1785/0220180162.

Ronneberger, O., et al. “U-Net: Convolutional Networks for Biomedical Image Segmentation.” International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer, 2015, pp. 234–41, doi:10.1007/978-3-319-24574-4_28.

Ross, Z. E., et al. “Hierarchical Interlocked Orthogonal Faulting in the 2019 Ridgecrest Earthquake Sequence.” Science, vol. 366, no. 6463, 2019, pp. 346–51, doi:10.1126/science.aaz0109.

Rousset, B., et al. “Slow Slip Events in the Roots of the San Andreas Fault.” Science Advances, vol. 5, no. 2, 2019, doi:10.1126/sciadv.aav3274.

Satake, K., and M. Heidarzadeh. “A Review of Source Models of the 2015 Illapel, Chile Earthquake and Insights from Tsunami Data.” Pure and Applied Geophysics, vol. 174, no. 1, 2017, pp. 1–9, doi:10.1007/978-3-319-57822-4_1.

Shelly, D. R. “A High‐resolution Seismic Catalog for the Initial 2019 Ridgecrest Earthquake Sequence: Foreshocks, Aftershocks, and Faulting Complexity.” Seismological Research Letters, vol. 91, no. 4, 2020, pp. 1971–78, doi:10.1785/0220190309.

Thomas, A. M., G. C. Beroza, et al. “Constraints on the Source Parameters of Low‐frequency Earthquakes on the San Andreas Fault.” Geophysical Research Letters, vol. 43, no. 4, 2016, pp. 1464–71, doi:10.1002/2015GL067173.

Thomas, A. M., A. Inbal, et al. “Identification of Low‐Frequency Earthquakes on the San Andreas Fault With Deep Learning.” Geophysical Research Letters, vol. 48, no. 13, 2021, p. 2021 093157, doi:10.1029/2021GL093157.

Tibi, R., et al. “Deep Learning Denoising Applied to Regional Distance Seismic Data in Utah.” Bulletin of the Seismological Society of America, vol. 111, no. 2, 2021, pp. 775–90, doi:10.1785/0120200292.

Trugman, D. T., et al. “Peak Ground Displacement Saturates Exactly When Expected: Implications for Earthquake Early Warning.” Journal of Geophysical Research: Solid Earth, vol. 124, no. 5, 2019, pp. 4642–53, doi:10.1029/2018JB017093.

Van Rossum, G., and F. L. Drake Jr. Python Reference Manual. Centrum voor Wiskunde en Informatica Amsterdam, 1995,

Wdowinski, S., et al. “Southern California Permanent GPS Geodetic Array: Spatial Filtering of Daily Positions for Estimating Coseismic and Postseismic Displacements Induced by the 1992 Landers Earthquake.” Journal of Geophysical Research: Solid Earth, vol. 102, no. B8, 1997, pp. 18057–70, doi:10.1029/97JB01378.

Wessel, P., et al. “The Generic Mapping Tools Version 6.” Geochemistry, Geophysics, Geosystems, vol. 20, 2019, pp. 5556–64, doi:10.1029/2019GC008515.

Williamson, A. L., et al. “Toward Near‐field Tsunami Forecasting along the Cascadia Subduction Zone Using Rapid GNSS Source Models.” Journal of Geophysical Research: Solid Earth, vol. 125, no. 8, 2020, p. 2020 019636, doi:10.1029/2020JB019636.

Zhu, L., and L. A. Rivera. “A Note on the Dynamic and Static Displacements from a Point Source in Multilayered Media.” Geophysical Journal International, vol. 148, no. 3, 2002, pp. 619–27, doi:10.1046/j.1365-246X.2002.01610.x.

Zhu, W., et al. “Seismic Signal Denoising and Decomposition Using Deep Neural Networks.” IEEE Transactions on Geoscience and Remote Sensing, vol. 57, no. 11, 2019, pp. 9476–88, doi:10.1109/TGRS.2019.2926772.

Zhu, W., and G. C. Beroza. “PhaseNet: A Deep-Neural-Network-Based Seismic Arrival-Time Picking Method.” Geophysical Journal International, vol. 216, no. 1, 2019, pp. 261–73, doi:10.1093/gji/ggy423.




How to Cite

Thomas, A., Melgar, D., Dybing, S. N., & Searcy, J. R. (2023). Deep learning for denoising High-Rate Global Navigation Satellite System data. Seismica, 2(1).