Final Report Summary - WHISPER (Towards continuous monitoring of the continuously changing Earth)
For years, the image of the Earth crust was one of a stable mass of rock, unperturbed for long time periods while stress or magma were slowly and silently developing. The crust would then suddenly come to life in dramatic, rapid instabilities in the form of earthquakes, eruptions or landslides. The expression of continuous changes of the crust is actually very subtle, and has been generally ignored. The goal of the Advanced Grant Whisper was therefore to detect and image the way the rocks are silently changing, and sometime evolve towards catastrophic events. Indeed these changes are not emitting waves and it has therefore been assumed that seismology cannot study these slight variations in the mechanical properties of the crust. It has been attempted in the past to measure these changes in very specific conditions, namely by relying on rare repeating earthquakes or to use costly controlled sources.
A fundamental change occurred in seismology with the development of methods using the permanent motion of the ground. This ambient ‘noise’ is produced continuously by natural and anthropogenic sources. It was shown by our group that these records could be used to build virtual seismograms. In essence, with records of the ambient noise at two stations, one can use a mathematical operation called correlation to retrieve the signal that would be produced at a station if an impulsive source were acting at the other. Because the ambient noise is always available, the virtual seismograms can be constructed at every date and can then be used to infer the physical changes of the medium through the seismic wave velocity of the rocks. This principle was applied broadly and related techniques were developed and improved during this project. We studied different regions of the crust, from the shallow layers to the deep crust, and various scales, from the thin damaged zone produced by earthquakes on active faults to the entire area of Japan.
When viewed through the lens of our refined methods, the crust is living and changing at all time scales. It reacts to rain fall, or changes in surface temperature, as to tectonic transient deformations. Even the subtle seismic velocity changes associated to the tidal deformation induced by the relative motion of our planet with the Sun and the Moon can be detected.
Volcanic eruptions and landslides seem to be sudden and hardly predictable. Indeed these dangerous phenomena are the results of an evolution that is difficult to observe with the resolution of our standard techniques. With the new monitoring tools provided with Whisper, we can observe slight changes that predate such catastrophic events. For earthquakes, only the posterior response of the crust has been revealed so far, indicating a non-elastic response of the crust at time scales of weeks to months. It also indicates that the sensitivity of the local seismic velocity to a sudden deformation is highly variable and is characteristic of the medium in terms of its damage and/or its internal fluid pressure. We propose to term this new parameter; ‘seismic susceptibility’. This could be regarded as an indicator of the proximity of the rock to a critical state of instability. A systematic measure of this parameter requires correcting for the perturbation associated with hydro-meteorological processes. This is why this last point was an important part of our work.
Seismic monitoring also has applications to underground industrial activity. Continuous recordings of ambient industrial noise allow for imaging of shallow structures and for monitoring the changes produced by industrial processes such as ore extraction or fluid injection. These activities have to be managed carefully to avoid wasting limited resources, to limit their impact on the environment, and to ensure public and employee safety when dangerous natural processes like earthquakes are triggered. Passive monitoring is therefore a new tool that contributes to a more effective and environmentally safer exploitation of the resources.
A fundamental change occurred in seismology with the development of methods using the permanent motion of the ground. This ambient ‘noise’ is produced continuously by natural and anthropogenic sources. It was shown by our group that these records could be used to build virtual seismograms. In essence, with records of the ambient noise at two stations, one can use a mathematical operation called correlation to retrieve the signal that would be produced at a station if an impulsive source were acting at the other. Because the ambient noise is always available, the virtual seismograms can be constructed at every date and can then be used to infer the physical changes of the medium through the seismic wave velocity of the rocks. This principle was applied broadly and related techniques were developed and improved during this project. We studied different regions of the crust, from the shallow layers to the deep crust, and various scales, from the thin damaged zone produced by earthquakes on active faults to the entire area of Japan.
When viewed through the lens of our refined methods, the crust is living and changing at all time scales. It reacts to rain fall, or changes in surface temperature, as to tectonic transient deformations. Even the subtle seismic velocity changes associated to the tidal deformation induced by the relative motion of our planet with the Sun and the Moon can be detected.
Volcanic eruptions and landslides seem to be sudden and hardly predictable. Indeed these dangerous phenomena are the results of an evolution that is difficult to observe with the resolution of our standard techniques. With the new monitoring tools provided with Whisper, we can observe slight changes that predate such catastrophic events. For earthquakes, only the posterior response of the crust has been revealed so far, indicating a non-elastic response of the crust at time scales of weeks to months. It also indicates that the sensitivity of the local seismic velocity to a sudden deformation is highly variable and is characteristic of the medium in terms of its damage and/or its internal fluid pressure. We propose to term this new parameter; ‘seismic susceptibility’. This could be regarded as an indicator of the proximity of the rock to a critical state of instability. A systematic measure of this parameter requires correcting for the perturbation associated with hydro-meteorological processes. This is why this last point was an important part of our work.
Seismic monitoring also has applications to underground industrial activity. Continuous recordings of ambient industrial noise allow for imaging of shallow structures and for monitoring the changes produced by industrial processes such as ore extraction or fluid injection. These activities have to be managed carefully to avoid wasting limited resources, to limit their impact on the environment, and to ensure public and employee safety when dangerous natural processes like earthquakes are triggered. Passive monitoring is therefore a new tool that contributes to a more effective and environmentally safer exploitation of the resources.