Supplementary material to "Machine Learning can predict the timing and size of analog earthquakes"

Abstract

This data set includes the results of digital image correlation of one experiment on subduction megathrust earthquakes with interacting asperities performed at the Laboratory of Experimental Tectonics (LET) Univ. Roma Tre in the framework of AspSync, the Marie Curie project (grant agreement 658034) lead by F. Corbi in 2016-2017. Detailed descriptions of the experiments and monitoring techniques can be found in Corbi et al. (2017 and 2019) to which this data set is supplementary material.


We here provide Digital Image Correlation (DIC) data relative to a 7 min long interval during which the experiment 
produces 40 seismic cycles with average duration of about 10.5 s (see Figure S1 in Corbi et al., 2019). The DIC analysis yields quantitative about the velocity field characterizing two consecutive frames, measured in this case at the model surface. For a detailed description of the experimental procedure, set-up and materials used, please refer to the article of Corbi et al. (2017) paragraph 2. This data set has been used for: a) studying the correlation between apparent slip-deficit maps and earthquake slip pattern (see Corbi et al., 2019; paragraph 4); and b) as input for the Machine Learning investigation (see Corbi et al., 2019; paragraph 5).


Further technical information about the methods, data products and matlab scripts is proviced in the data description file. The list of files explains the file and folder structure of the data set.

Authors

• Corbi, Fabio;Università degli Studi Roma Tre, Rome, Italy
• Sandri, Laura;INGV Bologna
• Bedford, Jonathan;GFZ German Research Centre for Geosciences, Potsdam, Germany
• Funiciello, Francesca;Università degli Studi Roma Tre, Rome, Italy
• Brizzi, Silvia;Università degli Studi Roma Tre, Rome, Italy
• Rosenau, Matthias;GFZ German Research Centre for Geosciences, Potsdam, Germany
• Lallemand, Serge;Géosciences Montpellier, CNRS, Montpellier, France;Montpellier University, Montpellier, France

Contact

• Corbi, Fabio (post-doc) ; Univ. Roma Tre; ➦

Contributors

Laboratory of Experimental Tectonics (University of Roma TRE, Italy)

Keywords

machine Learning, analogue models of geologic processes, subduction megathrust earthquakes, asperities, multi-scale laboratories, EPOS, Analog modelling results, Software tools, deformation, geologic process, tectonic process, tectonic process > subduction, Digital Image Correlation (DIC) / Particle Image Velocimetry (PIV) > MatPIV, Earthquake simulator, Wedge simulator, Gelatine, Gelatine > Pig skin, Surface image, tectonic and structural features, tectonic setting > plate margin setting, tectonic setting > plate margin setting > subduction zone setting, thrust fault, Videocamera, Digital Image Correlation (DIC) / Particle Image Velocimetry (PIV) > MatPIV, Earthquake simulator, Gelatine, Gelatine > Pig skin, Surface image, Videocamera, Wedge simulator, deformation, geologic process, tectonic and structural features, tectonic process, tectonic process > subduction, tectonic setting > plate margin setting, tectonic setting > plate margin setting > subduction zone setting, thrust fault

GCMD Science Keywords

• EARTH SCIENCE > SOLID EARTH > TECTONICS > EARTHQUAKES
• EARTH SCIENCE > SOLID EARTH > TECTONICS > EARTHQUAKES > EARTHQUAKE PREDICTIONS
• EARTH SCIENCE > SOLID EARTH > TECTONICS > PLATE TECTONICS > FAULT MOVEMENT