LETMA - CeLyA Spring School 2022: Waves & geosciences: Infrasound and beyond

 

 

Context:

 

The spring school was proposed in the framework of LETMA, which is a joint laboratory between CEA, Sorbonne University and Centrale Lyon. LETMA studies the propagation of low frequency waves in the atmosphere. In order to raise awareness of the subjects studied in relation to environmental issues, LETMA's partners proposed a spring school Waves & geosciences: Infrasound and beyond for students and researchers.

 

The spring school took place from 28 March 2022 to 1 April 2022 at the Valpré seminar centre in Ecully near Lyon, France. The number of participants was near 70.

 

Program:

 

Topics have included low frequency monitoring and application aspects, physics of the atmosphere and infrasound propagation, and signal processing techniques. Specific lectures on uncertainty and machine learning methods have been given by international experts in these disciplines. Two practical sessions have introduced participants to the algorithms used for infrasound detection, available data and methods for numerical simulation of infrasound propagation. One half-day was devoted to review the work done in the community, in the form of seminars and poster sessions. The timetable of the spring school is given below.

 

 

Detailed program:

 

The book of abstracts can be downloaded  here.

 

For each lecture, the abstract and the short bio of the speaker can be found below. If available, a link is provided to download the PDF of the presentations:

 

Scientific organizing committee:

 

Philippe Blanc-Benon, CNRS Director of Research - Centrale Lyon
François Coulouvrat, CNRS Director of Research - Sorbonne University
Didier Dragna, Assistant Professor - Centrale Lyon
Thomas Farges, Research Engineer - CEA
Regis Marchiano, Professor - Sorbonne University
Christophe Millet, CEA Director of Research - CEA

 

Local organizing committee:

 

Damien Bestard, Sorbonne-Université - CEA
Philippe Blanc-Benon, Centrale Lyon
Jules Colas, Centrale Lyon
Agnès Delebassée-Nabet, Labex CeLyA
Didier Dragna, Centrale Lyon
Arthur Guibard, Centrale Lyon
Codor Khodr, Centrale Lyon
Roman Leconte, Sorbonne-Université
Marie-Gabrielle Perriaux, Centrale Lyon
Laurent Pouilloux, Centrale Lyon
Carine Zambardi, Labex CeLyA
Annie Zelias, Sorbonne-Université - CEA

haut/top

 

LETMA - CeLyA Spring School 2022: Abstracts of the presentations

CTBTO and advances in infrasound monitoring - Pierrick Mialle (CTBTO, Austria)

 

For 25 years, the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO) operates the International Monitoring System (IMS), which comprises four networks with different sensor technologies: seismic, hydroacoustic, infrasound and radionuclides (Kalinowski and Mialle, 2021). The purpose of the IMS sensor network is to detect, identify and locate nuclear explosions underground, underwater or in the atmosphere. Upon completion, the infrasound component of the IMS network will consist of 60 infrasound stations to survey the atmosphere of which 53 are operational as of March 2022. In 2001, when the first data from an IMS infrasound station started to arrive in near real-time at the International Data Centre (IDC), its infrasound processing system was in a premature state. The IDC embarked for a multi-year design and development of its dedicated processing system, which led to operational IDC automatic processing and interactive analysis systems in 2010. In the next ten years the IDC produced over 40,000 infrasound events reviewed by expert analysts.

 

In an effort to continue advancing its methods, improving its automatic system and providing software packages to CTBTO users, the IDC focuses on several projects. First, the automatic system for the identification of valid signals that has been redesigned with the development of DTK-(G)PMCC (Progressive Multi-Channel Correlation) in collaboration with the Commissariat à l'Energie Atomique et aux Energies Alternatives (CEA), and which is made available to CTBTO users within NDC-in-a-Box. And second in order to replace its legacy network processing system, an infrasound model was developed for the automatic waveform network processing software NET-VISA with an emphasis on the optimization of the network detection threshold by identifying ways to refine signal characterization methodology and association criteria.

 

Ongoing and future improvements of the IDC processing system are planned to further reduce IDC analyst workload and offer better capabilities to CTBTO users that includes atmospheric propagation modeling and enhancements of the automatic pipeline components.

 

About the speaker:

 

Pierrick Mialle is seismic-acoustic officer at the Comprehensive Nuclear-Test-Ban Treaty Organization in Vienna. Since 2008, he is a scientist and project manager at the International Data Centre (IDC) with a focus on infrasound technology. He leads the IDC infrasound programme. His area of expertise includes waveform data processing, the study of natural and man-made events using infrasound technology. He also has a strong interest for civil and scientific applications using CTBT data and products. Prior to that Pierrick Mialle was a researcher at the Commissariat à l'Energie Atomique et aux Energies Alternatives (Atomic Energy and Alternative Energies Commission) in France.

haut/top

Atmosphere, structure and dynamics - Pietro Salizzoni (LMFA, Centrale Lyon, France)

 

The Earth's atmosphere is a unique environment, within which take place a huge variety of physical phenomena. Among these, our focus is on those that produce the climatic and meteorological conditions having a direct impact on our every day life. Radiative properties of the atmosphere, as determined by its chemical composition, play an essential role in mitigating the climatic condition on Earth. Simple mathematical model of these transfers enlighten the role of the atmosphere in warming the Earth surface, through the well known green-house effect, and allow for a relatively accurate estimate for the height of the troposphere, the region within which most of meteorological phenomena occur. To obtain realistic estimate of the vertical temperature gradient within the atmosphere, we have however to include another phenomenon, notably the heat (sensible and latent) transfer between the earth surface and the lower atmosphere. This forcing gives rise to a complex motion of masses of air on a wide variety of temporal and length scale, which is in turn deeply affected by the rotation of Earth. The latter is responsible for the characteristic dynamics that can be observed at the synoptic scales: the recirculation of cyclones and anticyclones, the rising of high altitude air streams (characterised by very intense velocities), and the occurrence of large scale waves surrounding the polar regions. The modelling of the dynamics of the atmosphere represent nowadays several challenges. Most of these deal with dynamical and thermodyanmical phenomena related to the changing state of water, that further amplify the complexity of the turbulent atmospheric motions, and call into question our ability in making prediction both for short term weather forecasts and long term climate scenarios.

 

About the speaker:

 

Pietro Salizzoni holds a Bachelor degree in Environmental Engineering from Politecnico di Torino and a PhD in Environmental Fluid Mechanics. He is currently Professor at École Centrale de Lyon. His teaching and research activities address physical phenomena associated with natural and technological risks, both accidental and persistent, with a focus on the dynamics of environmental flows. In particular, his studies deal with pollutant dispersion in the urban atmosphere, the propagation of hot smokes in free and confined environments, the ventilation of road tunnel in case of fire, the characterization of atmospheric dispersion of toxic and flammable substances, the transfer within the environment and food chain of dioxin and other persistent organic pollutants.

 

 presentation

haut/top

Gravity waves - Alvaro de la Cámara (Universidad Complutense de Madrid, Spain)

 

Gravity waves in the atmosphere are oscillations maintained by the vertical stratification of density in the atmosphere. They have spatial scales that range from tens to hundreds of kilometers, which makes their observation a challenge. Observational platforms typically have to decide between spatial coverage (regional versus global) and resolution (capturing the full spectrum versus just the larger scales). Their effects on atmospheric dynamics are multiple; they interact with turbulence in the planetary boundary layer, with cirrus cloud formation in the upper troposphere, modify the paths of infrasound propagation, or maintain a much warmer wintertime mesopause than the summer counterpart, just to mention a few.

 

This presentation will provide an overview of the general characteristics of atmospheric gravity waves, some of their impacts on the circulation of the atmosphere, and the treatment they require in general circulation models.

 

About the speaker:

 

Alvaro de la Cámara is an associate professor with the Department of Earth Physics and Astrophysics at Universidad Complutense de Madrid, Spain. Before, he worked as a postdoctoral researcher at LMD and CMLA (CNRS) in France, and NCAR (National Center for Atmospheric Research) in the USA. His research interests and experience are on the topics of middle atmospheric dynamics and modeling, including stratosphere-troposphere interactions, climate variability and gravity wave parameterizations for climate models.

 

 presentation

haut/top

Physics of infrasound propagation - Roger Waxler (NCPA, USA)

 

The physics of infrasound propagation in the atmosphere will be discussed. Fluid Mechanics, in which sound arises as compression/rarifaction waves, will be taken as the fundamental, underlying physical model. The linear approximation to the equations of Fluid Mechanics will be introduced. It will be seen that superposition and spectrum preservation are features of the linear approximation, and thus approximations themselves. Non-linear corrections will be touched upon. We will then focus on the linear approximation. Acoustic attenuation will be introduced and its physical basis described. Sound propagation models will then be developed. Refraction by temperature gradients and by wind shear will be illustrated using propagation models. This will require specification of the atmospheric state at the time of propagation. The prominent features of the Earth atmosphere, and their role in infrasound propagation will be discussed as will the available atmospheric specifications. Finally, we will discuss the role of non-linearity in atmospheric infrasound.

 

About the speaker:

 

Roger Waxler is a Mathematical Physicist by training who started his scientific career proving theorems about mathematical problems that arise in Quantum Mechanics. He drifted into Atmospheric Acoustics a bit more than twenty years ago and was delighted to see that much of the mathematics needed was similar to that used in Quantum Mechanics. Since then he has been involved in the development of the NCPAprop infrasound propagation sofware package. He has taken part in the design and execution of numerous field experiments, and has learned just enough signal analysis to deal with the resulting data.

 

 presentation

haut/top

Uncertainty Quantification and Bayesian Calibration in Computational Geosciences - Olivier Le Maitre (CMAP, France)

 

Uncertainty quantification (UQ) is crucial to simulations in geo-science. The last decade has witnessed the fast emergence of large-scale simulation infrastructures (HPC) and experimental capabilities (monitoring, imaging). This progress has raised the expectation of effectively improving numerical predictions by relying on complex computational models calibrated on massive measurements. Unfortunately, these promises remain out of reach and do not transform into reality. This presentation will show several advanced UQ techniques and calibration approaches applied to geoscience problems. These examples are selected to illustrate the need for holistic approaches in predictive simulations and the risk of focusing on one specific aspect (physical modeling, numerical modeling, or data collection) rather than balancing their respective contributions.

 

About the speaker:

 

Olivier Le Maître is a CNRS senior researcher at the Center for Applied Mathematics of the Ecole Polytechnique (France). Before joining CNRS in 2009, he was a Professor Assistant in the Mechanical Engineering Department of the University of Evry Val d'Essonne (France). Affiliated to LIMSI from 2009 to 2019, he held multiple visiting research and professor positions at Johns Hopkins University and Sandia National Laboratories. From 2013 to 2015, he was Visiting Professor at Duke University and Visiting Faculty at the King Abdullah University of Science and Technology (KAUST). He is currently a permanent member of the joint Inria-CNRS-Polytechnique team PLATON. After a Ph.D. (1998, Université du Havre, France) on non-linear fluid-structure interaction problems, he started a long term collaboration with Pr. Omar Knio (Johns Hopkins University) on Uncertainty Quantification (UQ) with the development of stochastic spectral methods and their applications to a broad range of non-linear models. Over the years, his research has evolved to encompass sensitivity analysis, Bayesian inference, robust optimization, and other thematics related to uncertainty management. He has published more than 100 paper and his contributions range from the formulation of UQ problems to their efficient numerical solution and HPC implementations. They have been applied in several domains, including environmental and geo-sciences, fluid engineering, nuclear engineering, combustion and reactive systems, aerospace (reentry, launchers), and naval engineering.

 

 presentation

haut/top

Algorithms for infrasound detection and wave parameter estimation: state-of-the art and new approaches - Maurice Charbit (Institut Mines-Telecom, France), Alexis Le Pichon (CEA, France)

 

Infrasound are low-frequency acoustic waves with frequencies less than 20 Hz. An important property of these waves is their capability to travel through the atmospheric waveguide structure over thousands of kilometers. The infrasound technology is one component of the International Monitoring System (IMS) being used to monitor compliance with the Comprehensive Nuclear Test-Ban Treaty (CTBT). The performance of IMS infrasound network strongly depends on station-specific environmental noise, which exhibits significant spatial-temporal variation. Its detection capability is a key concern since ubiquitous coherent noise generated by repetitive natural or anthropogenic made processes interfere with the detection and identification of explosive events. The two main issues are the detection of signals of interest embedded in the background noise and the estimation of the direction of arrivals of the coherent waves. The need to distinguish between incoherent wind noise and real coherent infrasonic wave signals has motivated the development of different approaches.

 

We first review state-of-the-art methods which include the Time Difference Of Arrival (TDOA) method on which the Progressive Multi-Channel Correlation (PMCC) is based, beamformer and focusing techniques, as well as algorithms based on subspace decomposition developed initially for multi-source narrowband signals (e.g. MUSIC, CLEAN). We also present new results derived from the likelihood function of a parametric stochastic model. We establish the expressions of both the Generalized Likelihood Ratio Test (GLRT) and the Maximum Likelihood Estimator (MLE) of the wavefront parameters (e.g. back-azimuth, trace velocity). The major results are on one hand the expression of the asymptotic distribution of the GLRT under null hypothesis (absence of spatially coherent signals), leading to the p-value computation, and on the other hand the expression of the asymptotic covariance of the MLE.

 

The Multi-Channel Maximum-Likelihood (MCML) detection and estimation method is implemented in the time-frequency domain in order to avoid the presence of interfering signals. Extensive simulations with synthetic signals show that MCML outperforms the state-of-the-art multi-channel correlation detector algorithms in terms of detection probability and false alarm rate in poor signal-to-noise ratio scenarios. We illustrate the use of the MCML on real data from the International Monitoring System (IMS) and show how the improved performances of this new method lead to a refined analysis of events in accordance with expert knowledge.

 

About the speakers:

 

Maurice Charbit is a Professor Emeritus at Telecom-Paris since 2013. He is also Research and Development Officer at OZE-energies, since October 2016 for energy/comfort management of office buildings. He has been a consultant with CEA since 2010 and was a consultant for CTBT in Vienna in 2016, 2017 and 2018. His domains of interest cover statistics for signal processing, array processing and data sciences.

 

Alexis Le Pichon is geophysicist at CEA, in charge of Infrasound research activities on topics relevant to Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO), including operation of low-frequency microphone arrays, automated signal processing for detection and source location or atmospheric remote sensing methods. He received in 2003 the award Science and Defence, delivered by Ministry of Defence for the development and optimization of infrasound monitoring techniques at CEA DAM.

 

 presentation - M. Charbit  presentation - A. Le Pichon

haut/top

Infrasound sources in geosciences - Jelle Assink (KNMI, Netherlands)

 

Infrasound waves are low-frequency acoustic waves with frequencies between ~3 mHz and 20 Hz, which are inaudible to humans. Because infrasound waves undergo very little absorption, infrasound can propagate over thousands of kilometers through atmospheric waveguides. The atmospheric ’soundscape’ consists of contributions from both geophysical and man-made infrasound sources, that can be detected across the infrasonic frequency band as small-amplitude pressure waves, even over very large distances. Sensitive barometers, or microbarometers, are used for the detection of infrasound waves. At longer ranges, infrasound arrays are in place to detect coherent infrasound with amplitudes below the noise level. Over the past decades, infrasound has evolved as a monitoring technique for various sources of interest, including volcanoes, meteors, earthquakes and calving glaciers. The International Monitoring System (IMS) that has been installed for the verification of the Comprehensive-Nuclear-Test-Ban-Treaty (CTBT) has been the backbone for much of the infrasound research around the world. In this lecture, we will review several sources that are routinely detected on the global infrasound network. The elegance of infrasound is that it is a passive monitoring technique that allows one to learn something about a geophysical source by simply listening to it. The available datasets allow for studies with temporal resolutions ranging from seconds to decades, which begin to allow for climatological studies.

 

About the speaker:

 

Jelle Assink is a Senior Geophysicist with the Seismology and Acoustics department at the Royal Netherlands Meteorological Institute (KNMI). He makes use of seismic and acoustic data for the study of these waves in the Earth, Ocean & Atmosphere. His efforts have contributed to the use of infrasound for atmospheric remote sensing, the monitoring of natural hazards and the verification of the Comprehensive Nuclear-Test-Ban Treaty (CTBT). For his research, he has developed algorithms for data processing and infrasound propagation modelling. He is currently responsible for the operations of KNMI's microbarometer array network and he is interested in synergies with other measurement techniques. Before joining KNMI in 2015, he worked at the National Center for Physical Acoustics in Oxford, MS, USA and the Alternative Energies and Atomic Energy Commission (CEA) near Paris, France

 

 presentation

haut/top

Machine learning in geosciences - Scott Hosking (BAS, UK)

 

Machine learning (ML) methods have the potential to advance our understanding within geoscience and unlock new scientific discoveries. However, not all geoscience problems can or should be tackled with ML. In this presentation I will highlight some use-cases where ML has worked well and provide some insight into how we got there - the challenges and decisions that needed to be made along the way. I will comment on where I see ML is going within the geosciences, and what is needed to get us there over the next 10 years. This will include a discussion on nurturing and contributing to open-source communities, building networks of collaborators and partners spanning the traditional science disciplines, and I will address the current buzz around the development of Digital Twins of the natural environment.

 

About the speaker:

 

Dr Scott Hosking is the leader of the AI lab at the British Antarctic Survey (BAS). He is also a co-Director of the UKRI Centre for Doctoral Training in the Application of AI for the study of Environmental Risks at the University of Cambridge and BAS, and a Senior Research Fellow at The Alan Turing Institute, UK.

haut/top

Lightning - Thomas Farges (CEA, France), François Coulouvrat (Sorbonne-Université, France)

 

Lightning generates infrasound and sound waves (thunder). Acoustic measurements in networks since the early 2000s allow to revisit the mechanisms of thunder formation. After a brief review of the physics of lightning and a description of the main characteristics of thunder, we present what acoustic networks can tell us about the tracking of lightning at short and long distances. Principles of complex sources localisation and reconstruction will first be recalled and illustrated by two measurement campaigns performed in Southern France in 2012 (in Cévennes region) and 2018 (in Corsica). At long range, the tracking of thunderstorm cells is possible up to at least 500 km, allowing to probe the middle atmosphere. At short range (less than 20 km), the 3D reconstruction of the lightning channel provides information on the structure of discharges, both within the clouds or between the clouds and the ground. These measurements lead us to propose a model of thunder generation taking into account the observed spectra, especially at low frequencies. This model identifies a near field (typically at less than 3 km) where the detailed source geometry must be considered, behaving mostly as a cylindrical one but with a high variability due to the tortuosity. In the far field, this variability is reduced and the lightning source can be seen as spherical. The comparison of the results of the model and the measurements allow for the first time to quantify the energy deposited by the discharge in the lightning channel, which turns out to be more intense than intracloud discharges. Recent analysis show however that the power distribution is not uniform along the lightning channel, with strong observed heterogeneities.

 

In addition, the electric field of particularly strong lightning can produce transient light phenomena occuring in the upper atmosphere. They are called sprites. These sprites last a few milliseconds and have a vertical spatial extension of more than 40 km and a horizontal extension of 1 to 100 km. The impacted space can be heated by a few degrees Kelvin. When sprites are very extended, this heating can generate a pressure wave in the infrasound range. These signals have a different acoustic signature than thunder that will be discussed.

 

About the speakers:

 

Thomas Farges is a researcher at CEA and member of LETMA. He is interested in thunderstorms and transient luminous events. He studies their electric field, their optical emissions and their acoustic waves (thunder). He is particularly interested in showing how thunder can help to characterize lightning. He led several acoustic measurement campaigns in this perspective. Recent topics of study are: 3D reconstruction of lightning from microphone arrays, variation of amplitude with distance and detection, monitoring of thunderstorms with acoustic measurements, and modeling of the source, ...

 

François Coulouvrat is a CNRS researcher at Institut Jean Le Rond d'Alembert (Sorbonne Université, Paris, France) and member of LETMA. He is interested in the theoretical modelling and numerical simulation of atmospheric propagation of sound and weak shock waves, including the physical phenomena of refraction, diffraction, absorption, scattering and non linearities. He contributed to the development of various numerical solvers, from ray tracing to one-way approximations. He studied various types of low frequency sound sources, namely sonic boom from supersonic aircraft and meteoroids, and lightning. In the recent years, he contributed with Thomas Farges to the 3D reconstruction of lightning from microphone arrays, and to the analysis of lightning as sound source of thunder, in particular in terms of energy.

 

 presentation

haut/top

Volcano acoustic - Maurizio Ripepe (University of Firenze, Italy)

 

Volcanoes are a very efficient source of pressure waves both in the acoustic, acousto-gravity and gravity frequency range. Gas rapidly expanding in the atmosphere produce pressure perturbation which contains valuable insights into eruption dynamics. Linear theory of sound is currently used to explain acoustic pressure below 20 Hz (infrasound) in terms of mass flow rate with strong implications for explosive dynamics and risk assessment. According to the explosive mechanisms acoustic waves can be modeled as a compact point source with wave field directivity strongly dependent on the crater rim geometry, ground topography and atmospheric conditions. Violent explosive eruptions can propel a plume of 1-10 km3 of hot ash and scoriae at tens of kilometers high into the atmosphere. This large mass flux induces acousto-gravity oscillation of the atmosphere which can be recorded at thousands of kilometers away at a global scale. But volcanoes can also generate acoustic waves by non-explosive non-point sources such as pyroclastic flows, which can be tracked in real-time using array techniques methods.

 

About the speaker:

 

Maurizio Ripepe is lecturer in Geophysics and Physical Volcanology at the Department of Earth Sciences of the University of Firenze. He was teaching Geophysics and Seismology at the University of Camerino and University of Siena. External Professor at the Open University, Milton Keans (UK) and Member of the Graduate School of Rio Negro University (Argentina). He was Visiting Professor at the University of Southern California (Los Angeles, USA), Earthquake Research Institute (Tokyo, Japan), The Leeds University (Leeds, UK) and Ecole Normal Superioure (Lion, France). Responsible for the Double PhD Degree in geophysics with the Tohoku University (Sendai, Japan).

 

He is responsible for the Laboratory of Experimental Geophysics (LGS) of Florence University and for the Center of Excellence in Volcano Monitoring, of the Italian Department of Civil Protection. He was scientific consultant during the 2003 Stromboli eruption for the Ministry of Interiors, Department of Civil Protection. He is member of the Seismic Microzoning Group of the Seismic Service of the Italian Civil Protection, member of the Great Risk Commission of the Italian Prime Ministry Council and member of the INFN (National Institute of Nuclear Physics) for VIRGO interferometer (gravitational waves). He is the coordinator of Project on Seismic Vulnerability of the City of Florence and scientific consultant on seismic risk for the Cultural Heritage of the Opera del Duomo of Florence. He was team Expert for the European Civil Protection Mechanism in Guatemala during the 2018 eruption of El Fuego volcano.

 

He cooperates with several research centers for the infrasonic monitoring of volcanoes and snow avalanches in Iceland (IMO), Japan (ERI), Montserrat (MVO), Ecuador (IGPN), Guatemala (INSIVUMEH), Costa Rica (OVSICORI), Peru (INGEMMET), Argentina (SEGEMAR), Chile (OVDAS) and Swiss (SLF).

haut/top

Data assimilation - Olivier Marsden (ECMWF, UK)

 

Data assimilation can be described as the blending of observations of a system with a model of the system's evolution, in order to obtain an estimation of the state of the system at a point in time. It is essential to the field of Numerical Weather Prediction, as observational data of the earth system is vastly insufficent to fully constrain the set of initial condition required for a numerical weather forecast. An overview of different facets of data assimilation will be provided, with examples taken from the field of NWP. A brief description of some of the main data assimilation techniques will be given, along with how these techniques can be implemented in NWP systems.

 

About the speaker:

 

Olivier Marsden is a computational scientist working at the European Centre for Medium-range Weather Forecasts (ECMWF), focusing on the computational efficiency of the IFS weather prediction suite. Before working at ECMWF he was a lecturer in fluid mechanics and aeroacoustics at the Ecole Centrale de Lyon.

 

 presentation

haut/top

Inverse problems for atmospheric dispersion - Lionel Soulhac (LMFA, INSA Lyon, France)

 

Atmospheric turbulent dispersion of pollutant material is an important subject in different contexts: air quality, industrial safety, transport of hazardous material, terrorist attacks. Pollutants are transported by the turbulent flow field induced in the atmospheric boundary layer by the interaction between the large scale geostrophic wind and the complexity of the Earth surface, with exchanges of mass, momentum, energy and humidity. If the characteristics of the source of pollutants are known, direct numerical models are useful to predict the concentration field and its time evolution. But when the characteristics of the source (position, emission rate, etc.) are unknown, an inversion modelling approach can be applied to estimate the source parameters starting from measurements of concentration in the atmosphere. The inversion strategy has to take into account the number of unknown parameters and the type of measurements used (uncertainty, fixed or mobile, etc.). This presentation will overview some recent works done in the AIR group of the LMFA on inversion modelling of atmospheric dispersion for different operational applications.

 

About the speaker:

 

Lionel Soulhac is professor and deputy head of the Fluid Mechanics and Acoustics Laboratory, at INSA Lyon. For the last 25 years, his research has been dedicated to the modelling and simulation of atmospheric flow and dispersion at local scale, in the atmospheric boundary layer. Lionel Soulhac has developed several dispersion software (SIRANE, SLAM, etc.) that are operationally used by engineers in national or local authorities and in industrial or consulting companies.

 

 presentation

haut/top

Contributed presentations

 

Numerical modeling of infrasonic waves in the Earth’s atmosphere - Roberto Sabatini (Embry-Riddle Aeronautical University, USA)

 

Infrasonic waves are inaudible acoustic waves with frequencies between 3 mHz and 20 Hz. They are generated by natural events, such as volcanic eruptions or earthquakes, and anthropogenic sources, like nuclear or chemical explosions and supersonic booms. They propagate through the layers of the atmosphere up to considerable distances (from hundreds to thousands of kilometers) and potentially carry relevant information about their sources. For this reason, direct and indirect measurements of infrasound-induced disturbances constitute a unique tool for the remote sensing of natural hazards and artificial sources.

 

The propagation of infrasonic waves is primarily driven by the vertical gradients of the speed of sound and the winds. Nevertheless, several complex phenomena may also affect the infrasonic signals traveling through the different atmospheric layers. A non-exhaustive list includes nonlinear steepening and lengthening, diffraction at caustics, scattering or partial reflections by small-scale atmospheric turbulent fluctuations, energy absorption by viscous stresses, thermal conduction, and vibrational relaxation.

 

The numerical modeling of infrasound propagation has been historically based on various approximations of the Navier-Stokes or Euler equations. Among the most widely employed techniques are ray tracing, normal modes, and one-way methods. Unfortunately, albeit computationally efficient, these approaches are generally unable to account for all the aforesaid physical phenomena. For this reason, over the past decade, significant efforts have been made to carry out infrasound studies via the direct numerical resolution of the complete set of equations of fluid mechanics.

 

In this talk, I will summarize some of my recent investigations on the propagation of infrasonic waves based upon the direct numerical simulations of the Navier-Stokes equations. I will particularly focus on the numerical aspects of the computations as well as on selected physical phenomena not accurately predicted by classically employed approaches.

 

Observation and modeling of the acoustic response of the ionosphere to earthquakes and volcano eruptions - Lucie Rolland (Geoazur, France)

 

Natural Hazards such as earthquakes, tsunamis and volcano eruptions can strongly affect the upper atmosphere. This presentation will show how sudden coseismic motions and explosive eruptions trigger acoustic waves that propagate upwards and impact the upper atmosphere. Observations of upper atmospheric motions can be conducted routinely using ionospheric monitoring methods, mostly using Ground Positioning System techniques. After reviewing state-of-the-art numerical methods used to model these phenomena, we will present simulations of observations conducted after the Hunga Tonga eruption

 

 presentation

haut/top