Task 3: Modelling of near-field ground motions in catastrophic earthquakes in Iceland

The M=5.1 earthquake on June 4, 1998, at Hellisheiði in the Hengill area (Figure 2) provided excellent geodetic and seismic data for modelling of near-field displacements of the largest earthquake in the area since 1955.

Results from modelling GPS data spanning the M=5.1 earthquake, June 4, 1998, were described in the first PRENLAB-2 annual report and by Árnadóttir et al. (1999). Preparations are ongoing for modelling near-field ground motions expected in large South Iceland seismic zone earthquakes on basis of GPS data, strong motion and records of the SIL network for the earthquakes on June 17 and June 21, and on basis of historical documentation of near-field destruction in historical South Iceland seismic zone earthquakes.

The modelling of ground motions expected in large South Iceland seismic zone earthquakes has been done for the M_S=7.1 earthquake of August 14, 1784, and written in a report (Árnadóttir and Olsen 2000). We use a finite-difference method to simulate the M_S=7.1 earthquake of August 14, 1784, which occurred in the South Iceland seismic zone, believed to have been the largest historical earthquake in Iceland. The August 14 earthquake was followed two days later by a M_S=6.7 event located approximately 30 km to the west.

We simulate a rupture on a N-S, vertical, right-lateral, strike-slip fault, vary the fault geometry, slip distribution and rupture velocity, and compare the peak velocities calculated at the surface obtained for the different models to a reference model. We find that the simulated peak velocities depend significantly on the depth to the top of the fault. The fault-parallel and vertical peak velocities decrease significantly if the fault does not break the surface, while the fault-perpendicular component is less affected. A model with a heterogeneous slip distribution yields a very different pattern and lower magnitude of surface peak velocities than uniform moment models (Figure 5). This is partly due to the variable slip at shallow depth in the distributed slip model.

We calculate the static Coulomb stress change for two models of the August 14, 1784, earthquake to examine if it is likely to have triggered the second large earthquake on August 16, 1784. We find that the stress change caused by a 20 km long fault is larger in the hypocentral region of the second earthquake than that for a 50 km long fault. This indicates that if the M_S=7.1 earthquake occurred on a short rather than a long fault, it is likely to have triggered the second large earthquake.

  

Figure 5: Peak velocities for a model with variable slip distribution. The top panel shows the E-W component, the center panel shows the N-S component, and the bottom panel shows the vertical component. The color scale extends from 0 m/s (dark blue) to 1.0 m/s (red). The N-S line shows the surface trace of the fault model. The fault is 50 km long and extends from the surface down to 15 km depth. The rupture starts at the center of the fault at 10 km depth, and propagates bilaterally with a constant rupture velocity of 2.7 km/s. In general, the largest peak velocities correspond to areas of large slip in the model. The squares depict current locations of towns (Árnadóttir and Olsen 2000).