Seismic sources- characterisation: detection, location and characterisation of seismic sources, in particular of slow earthquakes.
Determination of the state of stress of an area: focal mechanism inversion
Geomechanical modeling: at the regional scale (i.e. lithospheric scale) or fault scale (i.e. slow slip modeling)
I am focusing on modelisation of the stress field using DFN (Discrete Fracture Network) at the regional or local scale. The aim is to apply these models to geothermal analyses (i.e. identifying areas for exploration or better characterize the behavior of zones already selected)
I am also interested in looking for the best friction laws to apply to fractures to identify seismic events as well as modeling aseismic slip.
I am interested in tectonic tremors detections and locations methods applied to the Mexican subduction zone. Tectonic tremors have been well characterized in Cascadia and Japan through several location methods. Previous studies using temporary networks in Mexico revealed several area along strike where tremors occur. I am interested in better characterising these areas and checking whether or not these tremors zone have differences between them.
The presence of two temporary network, in Guerrero, with different configurations present an occasion to study the impact of the seismic network on tremor location. Another aspect I am interested in, is the comparison of our results in tremor location with results obtained by different methods. To compare tremor location methods with different time-windows, detections threshold... common parameter must be found to realize a quantitative comparison.
It has been recently shown (Ide and Yabe, 2014) that by stacking signal in the VLF band at the time of occurrence of tremor, it is possible to detect VLF signal and determine their moment tensor. I am applying this technique to determine focal mechanisms. The network orientation play a key role on the focal mechanism reliability. With a lot of aligned station a seemingly reliable focal mechanism present in fact error in the fault plane orientation.
Another aspect of tremor charcatrisation is done by studying the energy of tremor. I am attempting to study this in Mexico and looking at the energy to moment ratio compared to regular earthquakes values. The lower value of this ratio compared to regular earthquakes seem a specific feature of slow earthquakes and is similar to the values obtained in Japan and Cascadia.
The north-west of the Alps is an intraplate domain with very slow deformations. So, it seems difficult to determine the probability of occurrence of a lithospheric earthquake (magnitude greater than 7) from microseismic observations. Such observations are superficial processes with little relation to deeper and bigger ones.
The aim is to determine the seismic potential of a lithospheric sector north-west of the Alps, studying the stress field generated by a gravity driven model. This model is 360 km by 400 km by 230 km deep, centered on the west alpine fossil subduction and going up to the north of Strasbourg.
The study of the north-west alpine structures shows the importance of the alpine orogen which generates variations in depth of the lithospheric interfaces. A study of the stress field in the basement shows a variation of principal stress directions along the strike of the Alpine chain. Even if the absolute magnitude of stresses could not be determined a relative magnitude ratio is calculated.
Results underline the importance of rheology for a gravity driven model. If an elastic rheology is modeled, calculated stress directions do not match observations. However, using an elasto-plastic rheology with a realistic geometry of the lithospheric interfaces, we can obtain stress directions coherent with the data.
Key-words : Stress ; Geomechanical model ; Lithosphere ; Rheology
The seismicity in the region north-west of the Alps is moderate but significant earthquakes can occur (1356 A.D. Basel earthquake, 6
Few deep stress measurements exist so I study focal mechanism inversions methods. They have the advantage to provide stress orientations at seismogenic depths. Several methods of focal mechanism inversion have been developed over the years to determine the principal stress orientations. I compare three of the most common ones to determine the method better adapted to my problem. Moreover before applying these methods, the dataset must be checked to see wether it verifies the physical assumptions made by each method.
from Maury et al., 2012
Principal stress directions determined with the three methods using a dataset composed of independent events. The solution in blue is determined with Angelier's method. The solution in red is determined with Michael's method. The solutions in shades of grey are determined with Gephart and Forsyth's method. Two minimas are found. The minimum principal stress direction calculated in Basel borehole is shown in reen hatches.
Finally I have to take into account the Moho which is a major density contrast and the lithosphere-asthenosphere boundary (LAB) which is the base of my model.
The main features of my model. The bigger slope is due to the dipping of the LAB under the Alps. The other important structure is the cenozoic rift system.
The variations of the geometry of the Moho are negligible before the variations of the LAB.
Taking into account these geometries and using an elasto-plastic rheology for the lithospher the principal stress directions can be reproduced.
Moreover the plastic zone identified with this model roughly correspond to the area of the most seismically active domain.
