Developing numerical modeling can be very interesting to test theory about the propagation of earthquakes. However so far, due to computational intensive calculation, numerical models of earthquakes remains simple. Some of the most used approximations are: planar fault, quasi-dynamic simulation, purely elastic medium, single fault, single cycle.

Because of my interests, I mainly focus on the long term effect of geometry and interaction of fault systems. I mainly tackle the planar approximation together with the single cycle approximation.

For this purpose, I develop a boundary element code, that is accelerated by the use of state-of-the-art mathematical technics such as Fast Multipole Method or Hierarchical Matrices.

Related articles:

• Romanet, P., Bhat, H.S., Jolivet, R. and Madariaga, R., 2018. Fast and slow slip events emerge due to fault geometrical complexity. Geophysical Research Letters, 45(10), pp.4809-4819.