Earthquakes frequently claim hundreds of lives and cause major damage to cities and infrastructures. Recent large (M 8.8 in Chile, 2010; M 9.1 in Japan, 2011) as well as moderate-size devastating events (M 6.3 in New Zealand, 2010) are forceful reminders that earthquakes cannot be predicted. However, we can prepare for the expected shaking levels and potential secondary effects (tsunamis, landslides, liquefaction) by investigating the physics of earthquake rupture, by studying seismic wave propagation in the Earth crust, and by finding innovative methods to  quantify the seismic hazard.

The CES-group at KAUST conducts research to study earthquake source physics and ground-motion generation, with the goal to gain insight into  earthquake properties and to create new tools for seismic-shaking estimation for earthquake-engineering applications. We use seismic data to image the kinematic rupture process during earthquakes, and perform forward simulations to understand the dynamcis of the rupture process under various initial conditions. We calculate the radiated seismic wavefield emitted by the space-time varying rupture process, and investigate seismic wave scattering in heterogeneous Earth crust. We are also interested in retrieving accurate information about Earth structure in Saudi Arabia in order to understand better the seismo-tectonics and geo-dynamics of the Arabian Plate, and to improve earthquake locations and thus seismic monitoring capabilities and seismic hazard calculations in the region.


Current Research

​The geometric roughness of exposed fault scarps may be used to serve as a proxy for surface exposure age and by that help to constrain slip amount as well as timing of successive, surface-rupturing earthquakes.
We utilize existing earthquake location catalogs of well-located events to investigate the temporal variability of earthquake depth (mainly aftershocks that follow large earthquakes) and transient changes in the seismic-aseismic transition zone depth.
Multi-cycle earthquake rupture simulation an allow insights into long-term earthquake recurrence characteristics and fault system interaction. With MC-Qsim, we developed a tool enabling us to create long sequences of seismic records along predefined faults and fault systems of arbitrary geometric complexity for statistical analysis to investigate the system’s behavior.​​
​As part of an interdisciplinary research project with geologists, palaeontologists and anthropologists from USA and UK, we installed a dense seismic network in the Olduvai Gorge and Laetoli basins. These sites are located within the Ngorogoro Conservation Area (NCA) and are two paleo-antropological excavation sites of global importance for understanding the evolution of early humans. Our seismic network covers a surface of approximately 30 x 40 km and is operating with ten 120 s Trillium compact sensors since June 2016. Five more instruments were installed in June 2017.​​​
We develop an online collaboration platform to disseminate resources related to earthquake research: eQuake-RC. This site serves as an access-point to SRCMOD, a finite-fault rupture model database, the Source Inversion Validation(SIV) site with its wiki and lists of benchmarks, and computational tools for earthquake rupture model generationand related software.
The Source Inversion Validation project (SIV) constitutes an international collaboration, embedded in the research activities of the Southern California Earthquake Center (SCEC), and gathers earthquake scientists to verify, validate, and improve current strategies for earthquake source inversion. The goal is to also develop rigorous uncertainty quantification for earthquake source imaging. The CES-team leads the SIV-efforts at SCEC, and is responsible for the corresponding online cooperation platform. ​​
​Complex finite faulting source processes have important consequences for near source ground motions. Using 3D kinematic numerical modeling codes we simulate an ensemble of planar and listric faults (curved faults in which dip decreases with depth) with a variety of source parametrisations and discover that listricity causes systematic changes in ground motions, making it an important factor when estimating seismic hazard in the near field region.
​How do earthquake-rupture dimensions grow with increasing magnitude? Understanding earthquake source-scaling behavior provides important input for reliable seismic hazard analysis, studies on earthquake mechanics and for simulating rupture dynamics. The surge in earthquake-source imaging studies during recent years- captured by the SRCMOD database - allows investigating more deeply this fundamental topic on a global scale.  We examine source-scaling properties in different faulting regimes, illuminating their variability and control by tectonic settings such as seismogenic depth, fault-dip, and faulting mechanisms.
Earthquake source imaging is the key to better understand the kinematics of the space-time evolution of the earthquake rupture process. The resulting kinematic rupture models serve to study earthquake mechanics, to model earthquake rupture dynamics, to compute Coulomb-stress variations after significant earthquakes, and to build realistic rupture models for ground-motion simulation.
Dynamic rupture modeling entails a physics-based characterization of the earthquake rupture process based on first-order principles. Using 2D and 3D numerical modeling codes (e.g. finite-difference, spectral element methods), parameterizing the bulk medium properties, the stresses acting on the fault, and the frictional breakdown process at the propagating fracture, we solve the equations of motion to compute how earthquake ruptures behave under various initial and boundary conditions. Most of these simulations are computationally very demanding and require large-scale computing facilities.
We develop methods to generate complete broadband ground-motions (i.e. time histories of seismic shaking at the Earth surface) for earthquake engineering applications. The term "broadband" here refers to the frequency range of interest for earthquake engineering, and covers frequencies from 0 Hz (static ground displacement) up to ~10 Hz (to which stiff structures are sensitive to).
In this work we use broadband seismic data collected by the Saudi Geological Survey to study Earth structure in Saudi Arabia. Initial work deploys a receiver-function technique to map Moho thickness and potential deeper interfaces in the Earth. We also focus on the western regions in Saudi Arabia that are characterized by volcanic provinces and a rapid transition from oceanic crust to more continental crust at the margins of the Red Sea. This area is also more prone to earthquakes, and detailed knowledge of Earth structure will help to better locate earthquake, a key step in improved seismic hazard analysis.​
While the Earth, to first order, can be described as a stratified layered medium, it does contain considerable lateral variations in seismic wave speeds. In particular the Earth' upper layers (crust and upper mantle) contain heterogeneities at all scales that lead to intricate wave propagation effects and seismic scattering. Our interests are focused on modeling the Earth crust as a random field to study seismic scattering, the generation and properties of coda waves, and how seismic scattering affects near-source ground motions and their variability.