I13: Geophysics and gravitational wave interferometric detectors

Collaborations with J.Harms (INFN Florence), B.Whiting (Florida University), J.-P.Ampuero (Caltech), M.Ando (Tokyo University), F.Sorrentino (INFN Genova).

This project started with a completely exploratory nature. The first success of the E3 WP is to have allowed exchanges between the geophysical and gravitational-wave communities at APC and IPGP in order to explore new ideas and methods. Moreover, around our APC-IPGP team, several international collaborations started, both on the  geophysics side (Prof. Ampuero, Caltech-Seismolab) and on the gravitational-wave side (Prof.Whiting, U.Florida and Dr.Harms, INFN-Urbino). During this first exploratory phase (2012-2014) we have analyzed a few research possibilities (use of Virgo as long-base tiltmeter and strainmeter, study of the gravity-gradient noise, study of the prompt gravity perturbations due to earthquakes). We have then identified an original research direction: the detection of the prompt gravity perturbation produced by the mass redistribution during earthquakes, and its potential application to the improvement of earthquake early-warning systems.

Main scientific results achieved:

1) Search of an instantaneous gravity signal from the Tohoku 2011 earthquake

The first results, demonstrating the detection of the prompt gravity signal at 99% using a superconducting gravimeter at Kamioka and a japanese network of F-NET seismometers are summarized in the publication: Prompt gravity signal induced by the 2011 Tohoku-Oki earthquake, J.-P.Montagner et al. – Nature Communications, DOI:10.1038/natcomms13349 (2016). See also the CNRS press release: http://www.insu.cnrs.fr/node/6165.

Since this first work, a lot of progress has been made in 2017 about the observation of the prompt gravity signal and its modeling. This is summarized in a publication published in Science: Observation and modeling of the elastogravity signal preceding the direct seismic waves, Vallée et al., Science 358, 1164–1168 (2017)

The results were also summarized in the PhD thesis of Kévin Juhel : Signaux gravitationnels transitoires générés par rupture sismique (defended in December 2017)

2) First analytical computation of the prompt gravity signal by an earthquake.

The prompt gravity signal has been computed by an original analytical model, validated with numerical simulations.

The results are described in the publication : Transient gravity perturbations induced by earthquake rupture, J.Harms, J.-P. Ampuero, M. Barsuglia, E. Chassande-Mottin, J.-P. Montagner, S. N. Somala and B. F. Whiting, Geophys.Journal International (2015) 201, 1416-1425

3) Simulation of the gravity signal with the method of Earth normal modes

The prompt gravity signal has been also simulated with the decomposition of the gravity signal on the Earth Normal modes. A strong effort in the last three years (during K.Juhel thesis) has been made to overcome several computational problems and now the method is robust. The corresponding publications has been recently accepted by Geophysical
Journal International: Normal mode simulation of prompt elastogravity signals induced by an earthquake rupture, K. Juhel, J.-P. Montagner, M. Vallée, J. P. Ampuero, M. Barsuglia, P. Bernard, E. Clévédé, J. Harms and B. F.Whiting, Geophys. Journal International (2018, in press).

4) Feasibility study of a gravity strainmeter

From points 2 and 3 and from the finding of the Tohoku-oki earthquake analysis, we have demonstrated that conventional instruments (seismometers, gravimeters) cannot be used to detect medium size earthquakes (M~7) in ~ 10 seconds. New instruments, measuring the gravity strain and immune to seismic noise are necessary. The sensitivity needed is h ~ 10−15 Hz−1/2 at 0.1 Hz. These instruments do not exist, but prototypes (torsion bar antennas, atom interferometers, superconducting gravimeters) are developed in the context of gravitational-wave detection.

To study the feasibility of a gravity strainmeter, we have so far : a) contributed to the TOBA prototype in Japan with visits to the Tokyo University, b) studied the impact of the local gravity noise fluctuations on the detectors. The study of the atmospheric local gravity noise is summarized in a publication in Physical Review D: Impact of infrasound atmospheric noise on gravity detectors used for astrophysical and geophysical applications, D. Fiorucci, J.Harms, M. Barsuglia, I. Fiori, and F. Paoletti, Phys. Rev. D 97, 062003 (2018)

5) A feasibility study for a gravity-based earthquake early warning system network

A preliminary study of the implementation of an early detection system of seismic ruptures based on gravity has been carried out. This study assumes the availability of a TOBA high precision instrument, and will seek to optimize the detection performance of a network comprising these instruments. First, a database of seismic ruptures is simulated to detect events of varying durations, either still in progress or whose rupture has already been completed.

The location of the different instruments is drawn, and the configuration leading to the optimal detection must be estimated.
Knowing the noise model of future gravity strainmeters, it is possible to generate time series of noise at the chosen locations, and thus to calibrate the network in a “quiet” period. New time series containing a seismic event can also be generated, adding to the instrumental noise the response of the instrument to a rupture, described in Harms
(2016). The rupture detection will be performed in “real time”, correlating each record of the network to the database, and comparing the value obtained to the reference coefficient during a calm period. If the value obtained exceeds a fixed threshold, imposed so that the rate of false alarms is acceptable, the detection is carried out.

Once the detection is performed by the network, the location and the amplitude of the source are searched for. The search for the location of the source point is performed on a grid grouping the risk areas, according to a least squares error minimization method. The magnitude is then obtained at the most probable location by estimating the scale factor between the database (whose amplitude of the sources is known) and the “recorded” data. An alert will then be issued on the estimated magnitude imposed.

These results are explained in the PhD thesis of K.Juhel and have been submitted to JGR and also published in the EarthArXiv pre-print archive at the following address: https://eartharxiv.org/yqb3a/ Kévin Juhel, Jean-Paul, Ampuero, Matteo Barsuglia, Eric Chassande-Mottin, Donatella Fiorucci, Jan Harms, Jean-Paul Montagner, Martin Vallée, Bernard Whiting, Earthquake early warning using future generation gravity strainmeters,
submitted to JGR, 2O18.

The objectives for 2019 are the following:

1) Continue to exploit current seismometer and gravimeter data to look for prompt gravity signals associated with large earthquakes (> M8.0). Develop data-analysis strategies to improve the signal detection for smaller earthquakes, using coherence or stacking techniques across a network of seismometers and gravimeters, and the a-priori knowledge of the gravity perturbations from the numerical simulations. Develop real-time monitoring strategies to detect prompt gravity signals with a network of seismometers, and complement existing techniques on early magnitude estimation.

2) Refine the simulation of the gravity perturbation and instrumental noise on gravity strainmeters. Improve the estimation of the capabilities of future instruments, and precise instrumental specifications needed for earthquake early detection. Improve the characteristics of a network of future instruments (number, spacing, geometry, distance to the source). Set the uncertainties on the source location and magnitude, and improve the
alert time and earthquake position/magnitude estimation through more sophisticated data analysis techniques.

3) Develop a conceptual design for a gravity gradiometer with seismic isolated test masses, and prepare an ERC synergy project with other EU partners, which will be submitted in fall 2019.

In addition, we would like to continue the use of Virgo data for geophysics. The aim is to find signatures of local earthquakes in new Virgo-LIGO data and to exploit the long baseline and very high sensitivity of the instrument network to bring additional information to existent detectors.

J. Harms, J.-P. Ampuero, M. Barsuglia, E.Chassande-Mottin, J.-P. Montagner, S. N. Somala and B. F.Whiting,
Transient gravity perturbations induced by earthquake rupture,
Geophys. Journal International (2015) 201, 1416-1425

Jean-Paul Montagner , Matteo Barsuglia , Kévin Juhel , Jean-Paul Ampuero, Eric Chassande-Mottin, Jan Harms, Bernard Whiting, Pascal Bernard, Eric Clévédé, Philippe Lognonné
Prompt gravity signal due to the 2011 Tohoku-oki earthquake
Nature Communications, 2016, DOI: 10.1038/ncomms13349

Vallée et al.,
Observation and modeling of the elasto-gravity signal preceeding the direct seismic waves,
Science 358, 1164–1168 (2017)

D. Fiorucci, J. Harms, M. Barsuglia, I. Fiori, and F. Paoletti,
Impact of infrasound atmospheric noise on gravity detectors used for astrophysical and geophysical applications
 Physical Review D., January 2018, DOI: 10.1103/PhysRevD.97.062003

Juhel K. et al.
Normal mode simulation of prompt elastogravity signals induced by an earthquake rupture
accepted for publication in GJI

Kévin Juhel, Jean-Paul, Ampuero Matteo Barsuglia, Eric Chassande-Mottin, Donatella Fiorucci, Jan Harms, Jean-Paul Montagner, Martin Vallée, Bernard Whiting
Earthquake early warning using future generation gravity strainmeters
Submitted for publication to JGR