I am an observational earthquake seismologist currently working on earthquake signal characterization using artificial intelligence techniques. My goal is to gain insights from weak signals and big data. I obtained my Ph.D. degree in geophysics under the supervision of Prof. Charles Langston from the University of Memphis in 2017 with a thesis entitled “Microseismic Monitoring and Denoising.” I developed novel statistical signal processing tools for seismic signal denoising and decomposition, used machine learning for source characterization, studied crustal and upper mantel attenuation, and performed statistical analyses of spatio-temporal patterns of seismicity. In 2017 I joined Stanford's earthquake seismology group as a postdoctoral fellow to study the seismic hazard assessment of induced seismicity and machine-learning applications for earthquake monitoring under the supervision of Prof. Gregory Beroza. I enjoy exploring new approaches to problems and the surprises lying at the intersections of disciplines like geoscience, statistics, and computer science.
Phys Sci Res Assoc, Geophysics
Honors & Awards
The Reviewer of the Year Award, Seismological Research Letters (2019)
Top Reviewer for Geoscience, Publons (2018)
Outstanding Presentation Award, Seismological Society of America (2016)
1st Place Award in Physical and Applied Sciences, University of Memphis (2016)
ExxonMobil SEP/SEG Award, Society of Exploration Geophysics (2014)
Jeanne X. Kasperson Award, AAG Hazards and Risks Group (2014)
Fellow of National Talent Foundation, Ministry of Science, Research and Technology of Iran (2011)
Boards, Advisory Committees, Professional Organizations
Associate Editor, IEEE Transactions on Geoscience and Remote Sensing (2020 - Present)
Review Panelist, USGS Earthquake Early Warning Program. NSF Statistics Division of Mathematical Sciences. (2019 - 2020)
Journal Reviewer, Science, Nature Communication, Geophysical Journal International, Journal of Geophysical Research, Geophysical Research Letters, Scientific Report, Bulletin of Seismological Society of America, Seismological Research Letters, Tectonophysics, Pure and Applied Geophysics, Journal of Seismology, Journal of Asian Earth Science, Acta Geophysica, Communications Earth & Environment, Earth Sciences Research Journal, Earth and Space Science, Earth Planet and Space, Computers and Geosciences, Surveys in Geophysics, Geophysics, Geophysical Prospecting, Exploration Geophysics, Energies, Episodes, Sensors, Entropy, Applied Sciences, Measurement, Petroleum Science, Structures, Neural Computing and Applications, IEEE Geoscience and Remote Sensing Letters, IEEE Transaction on Geoscience and Remote Sensing, IEEE Access, IEEE Transaction of Image Processing. (2017 - Present)
Proposal Reviewer, NSF-EAR, Icelandic Research Fund. (2017 - Present)
Ph.D., University of Memphis, Geophysics (2017)
M.Sc., University of Memphis, Geophysics (2014)
M.Sc., University of Tehran, Risk Engineering (2010)
B.Sc., Tehran University, Civil Engineering (2006)
Current Research and Scholarly Interests
Machine Learning/Deep Learning (time-series analyses, representation learning, and pattern recognition),
Statistical Seismology (probabilistic seismic hazard, risk assessment, and seismicity analyses),
Signal Processing (time-frequency analyses, seismic denoising, and signal decomposition),
Earthquake Monitoring (earthquake detection and characterization).
Earthquake transformer-an attentive deep-learning model for simultaneous earthquake detection and phase picking.
2020; 11 (1): 3952
Earthquake signal detection and seismic phase picking are challenging tasks in the processing of noisy data and the monitoring of microearthquakes. Here we present a global deep-learning model for simultaneous earthquake detection and phase picking. Performing these two related tasks in tandem improves model performance in each individual task by combining information in phases and in the full waveform of earthquake signals by using a hierarchical attention mechanism. We show that our model outperforms previous deep-learning and traditional phase-picking and detection algorithms. Applying our model to 5 weeks of continuous data recorded during 2000 Tottori earthquakes in Japan, we were able to detect and locate two times more earthquakes using only a portion (less than 1/3) of seismic stations. Our model picks P and S phases with precision close to manual picks by human analysts; however, its high efficiency and higher sensitivity can result in detecting and characterizing more and smaller events.
View details for DOI 10.1038/s41467-020-17591-w
View details for PubMedID 32770023
- A Machine-Learning Approach for Earthquake Magnitude Estimation GEOPHYSICAL RESEARCH LETTERS 2020; 47 (1)
Bayesian-Deep-Learning Estimation of Local Earthquake Location from Single-Station Observations
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
View details for DOI 10.1109/TGRS.2020.2988770.
- Hierarchical Attentive Modeling of Earthquake Signals ICLR 2020
- Seismic Signal Denoising and Decomposition Using Deep Neural Networks IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2019; 57 (11): 9476–88
- Unsupervised Clustering of Seismic Signals Using Deep Convolutional Autoencoders IEEE GEOSCIENCE AND REMOTE SENSING LETTERS 2019; 16 (11): 1693–97
- Separating Signal from Noise and from Other Signal Using Nonlinear Thresholding and Scale-Time Windowing of Continuous Wavelet Transforms BULLETIN OF THE SEISMOLOGICAL SOCIETY OF AMERICA 2019; 109 (5): 1691–1700
CRED: A Deep Residual Network of Convolutional and Recurrent Units for Earthquake Signal Detection.
2019; 9 (1): 10267
Earthquake signal detection is at the core of observational seismology. A good detection algorithm should be sensitive to small and weak events with a variety of waveform shapes, robust to background noise and non-earthquake signals, and efficient for processing large data volumes. Here, we introduce the Cnn-Rnn Earthquake Detector (CRED), a detector based on deep neural networks. CRED uses a combination of convolutional layers and bi-directional long-short-term memory units in a residual structure. It learns the time-frequency characteristics of the dominant phases in an earthquake signal from three component data recorded on individual stations. We train the network using 500,000 seismograms (250k associated with tectonic earthquakes and 250k identified as noise) recorded in Northern California. The robustness of the trained model with respect to the noise level and non-earthquake signals is shown by applying it to a set of semi-synthetic signals. We also apply the model to one month of continuous data recorded at Central Arkansas to demonstrate its efficiency, generalization, and sensitivity. Our model is able to detect more than 800 microearthquakes as small as -1.3ML induced during hydraulic fracturing far away than the training region. We compare the performance of the model with the STA/LTA, template matching, and FAST algorithms. Our results indicate an efficient and reliable performance of CRED. This framework holds great promise for lowering the detection threshold while minimizing false positive detection rates.
View details for DOI 10.1038/s41598-019-45748-1
View details for PubMedID 31311942
STandord Earthquake Dataset (STEAD): A Global Data Set of Seismic Signal for AI
View details for DOI 10.1109/ACCESS.2019.2947848.
- Evaluating the 2016 One-Year Seismic Hazard Model for the Central and Eastern United States Using Instrumental Ground-Motion Data SEISMOLOGICAL RESEARCH LETTERS 2018; 89 (3): 1185–96
Variabilities in Probabilistic Seismic Hazard Maps for Natural and Induced Seismicity in the Central and Eastern United States
The Leading Edge
View details for DOI 10.1190/tle37020810.1
Lateral Variation of Crustal Lg Attenuation in Eastern North America.
2018; 8 (1): 7285
We perform Q Lg tomography for the northeastern part of North America. Vertical broadband seismograms of 473 crustal earthquakes recorded by 302 stations are processed to extract the Lg amplitude spectra. Tomographic inversions are independently conducted at 58 discrete frequencies distributed evenly in log space between 0.1 and 20.0 Hz. This relatively large dataset with good ray coverage allows us to image lateral variation of the crustal attenuation over the region. Obtained Q Lg maps at broadband and individual frequencies provide new insights into the crustal attenuation of the region and its relationship to geological structures and past tectonic activity in the area. The Q Lg shows more uniform values over the older, colder, and drier Canadian Shield, in contrast to higher variations in the younger margins. Results confirm the correlation of large-scale variations with crustal geological features in the area. Existence of low-velocity anomalies, thick sediments, volcanic rocks, and thin oceanic crust are potential sources of observed anomalies. The mean Q values are inversely correlated with average heat flow/generation for main geological provinces.
View details for DOI 10.1038/s41598-018-25649-5
View details for PubMedID 29740108
View details for PubMedCentralID PMC5940689
- Comment on "Recent developments of the Middle East catalog" by Zare et al. JOURNAL OF SEISMOLOGY 2017; 21 (1): 257-268
Mapping Seismic Moment and b-value within Continental Collision Orogenic Belt Region, the Iranian Plateau
JOURNAL of GEODYNAMICS
View details for DOI 10.1016/j.jog.2016.12.001
- Automatic Noise-Removal/Signal-Removal Based on the General-Cross-Validation Thresholding in Synchrosqueezed domains, and its application on earthquake data GEOPHYSICS 2017
Investigation of Spatial Variations of Seismic Energy Released and b-value Under Tectonics Framework of the Middle East Region
JOURNAL of ASIAN EARTH SCIENCE
View details for DOI 10.1016/j.jseaes.2017.07.040
- Seismic features and automatic discrimination of deep and shallow induced-microearthquakes using neural network and logistic regression GEOPHYSICAL JOURNAL INTERNATIONAL 2016; 207 (1): 29-46
- Adaptive noise estimation and suppression for improving microseismic event detection JOURNAL OF APPLIED GEOPHYSICS 2016; 132: 116-124
- Automatic microseismic denoising and onset detection using the synchrosqueezed continuous wavelet transform GEOPHYSICS 2016; 81 (4): V341-V355
Hybrid Seismic Denoising Using Wavelet Block Thresholding and Higher-Order Statistics
BULLETIN OF THE SEISMOLOGICAL SOCIETY OF AMERICA
View details for DOI 10.1785/0120150345
- Adaptive Microseismic Noise Estimation and Denoising SEG Technical Program Expanded Abstracts 2016
- Automatic Denoising and Detection of Microseismic Events using the Synchrosqueezing SEG Technical Program Expanded Abstracts 2016
Average QLg, QSn, and Observation of Lg Blockage in the Continental Margin of Nova Scotia
JOURNAL of GEOPHYSICAL RESEARCH: SOLID EARTH
View details for DOI 10.1002/2014JB011237
- Quantitative risk analysis for earthquake-induced landslides-Emamzadeh Ali, Iran ENGINEERING GEOLOGY 2011; 122 (3-4): 191-203