Local earthquakes detection: A benchmark dataset of 3-component seismograms built on a global scale
Authors:
Fabrizio Magrini,
Dario Jozinović,
Fabio Cammarano,
Alberto Michelini,
Lapo Boschi
Abstract:
Machine learning is becoming increasingly important in scientific and technological progress, due to its ability to create models that describe complex data and generalize well. The wealth of publicly-available seismic data nowadays requires automated, fast, and reliable tools to carry out a multitude of tasks, such as the detection of small, local earthquakes in areas characterized by sparsity of…
▽ More
Machine learning is becoming increasingly important in scientific and technological progress, due to its ability to create models that describe complex data and generalize well. The wealth of publicly-available seismic data nowadays requires automated, fast, and reliable tools to carry out a multitude of tasks, such as the detection of small, local earthquakes in areas characterized by sparsity of receivers. A similar application of machine learning, however, should be built on a large amount of labeled seismograms, which is neither immediate to obtain nor to compile. In this study we present a large dataset of seismograms recorded along the vertical, north, and east components of 1487 broad-band or very broad-band receivers distributed worldwide; this includes 629,095 3-component seismograms generated by 304,878 local earthquakes and labeled as EQ, and 615,847 ones labeled as noise (AN). Application of machine learning to this dataset shows that a simple Convolutional Neural Network of 67,939 parameters allows discriminating between earthquakes and noise single-station recordings, even if applied in regions not represented in the training set. Achieving an accuracy of 96.7, 95.3, and 93.2% on training, validation, and test set, respectively, we prove that the large variety of geological and tectonic settings covered by our data supports the generalization capabilities of the algorithm, and makes it applicable to real-time detection of local events. We make the database publicly available, intending to provide the seismological and broader scientific community with a benchmark for time-series to be used as a testing ground in signal processing.
△ Less
Submitted 6 August, 2020;
originally announced August 2020.
Geophysical tests for habitability in ice-covered ocean worlds
Authors:
Steven D. Vance,
Mark P. Panning,
Simon Stähler,
Fabio Cammarano,
Bruce G. Bills,
Sharon Kedar,
Christophe Sotin,
William T. Pike,
Ralph Lorenz,
Victor Tsai,
Hsin-Hua Huang,
Jennifer M. Jackson,
Bruce Banerdt
Abstract:
Geophysical measurements can reveal the structure of icy ocean worlds and cycling of volatiles. The associated density, temperature, sound speed, and electrical conductivity of such worlds thus characterizes their habitability. To explore the variability and correlation of these parameters, and to provide tools for planning and data analyses, we develop 1-D calculations of internal structure, whic…
▽ More
Geophysical measurements can reveal the structure of icy ocean worlds and cycling of volatiles. The associated density, temperature, sound speed, and electrical conductivity of such worlds thus characterizes their habitability. To explore the variability and correlation of these parameters, and to provide tools for planning and data analyses, we develop 1-D calculations of internal structure, which use available constraints on the thermodynamics of aqueous MgSO$_4$, NaCl (as seawater), and NH$_3$, water ices, and silicate content. Limits in available thermodynamic data narrow the parameter space that can be explored: insufficient coverage in pressure, temperature, and composition for end-member salinities of MgSO$_4$ and NaCl, and for relevant water ices; and a dearth of suitable data for aqueous mixtures of Na-Mg-Cl-SO$_4$-NH$_3$. For Europa, ocean compositions that are oxidized and dominated by MgSO$_4$, vs reduced (NaCl), illustrate these gaps, but also show the potential for diagnostic and measurable combinations of geophysical parameters. The low-density rocky core of Enceladus may comprise hydrated minerals, or anydrous minerals with high porosity comparable to Earth's upper mantle. Titan's ocean must be dense, but not necessarily saline, as previously noted, and may have little or no high-pressure ice at its base. Ganymede's silicious interior is deepest among all known ocean worlds, and may contain multiple phases of high-pressure ice, which will become buoyant if the ocean is sufficiently salty. Callisto's likely near-eutectic ocean cannot be adequately modeled using available data. Callisto may also lack high-pressure ices, but this cannot be confirmed due to uncertainty in its moment of inertia.
△ Less
Submitted 10 May, 2017;
originally announced May 2017.
