The Nov. 8, 1997 Earthquake in Tibet:
Fault Map, CMT Solutions, and Stress Change Modeling

This earthquake occurred within the Tibetan Plateau. At first, we thought that perhaps it had ruptured part of either the Altyn Tagh or Kunlun fault. Upon plotting up the location, however, it immediately became clear that this event was not on either of these major faults. The following is a preliminary investigation of this previously mapped, yet un-named fault, and our considerations of how this earthquake has likely influenced stresses on the neighboring major faults that are capable of even larger events.

Figure 1 shows the Centroid Moment Tensor (CMT) solutions from Harvard and NEIC for the recent earthquake in Tibet. The faults shown (in blue) were all digitized from Tapponier & Molnar (1977), who interpreted LANDSAT images to construct the map. The Altyn Tagh fault strikes NE across the upper left corner of the map and lies 200 km away from this rupture; the Kunlun fault also lies 200 km east of the mapped eastern end of this fault. The epicenter and left-lateral, ENE striking planes of the CMT mechanisms for this earthquake are consistent with rupture of the previously un-named fault that is colored red in this map view. The fault is long enough to sustain a rupture of this magnitude. A small lake at the west end of this un-named fault is named Yueh Qi Tai (from the ONC map). The lake is not on this map we put together (GMT full resolution does not contain the lake that is labelled on the ONC map).

Figure 2 shows the results of elastic half-space dislocation modeling to find the maximum stress changes on other faults in the vicinity of this earthquake. Several smaller faults that were mapped by Tapponier and Molnar (1977) are within the region where the stress was perturbed by more than one bar (indicated by the contour line). For the purpose of this simple model, we assumed a 100 km long fault centered on the Harvard CMT location. We also assumed a 20 km down-dip width and used the total moment from the Harvard CMT (so we assumed uniform slip of 470 cm in this model).

Figure 3 shows an overview of the region; the white lines are approximate locations of the nearest sections of the Altyn Tagh and Kunlun faults, onto which stress changes were projected.

Figure 4 shows the Coulomb stress changes resolved onto the Altyn Tagh and Kunlun faults, such that positive values would make the fault more likely to rupture with left-lateral slip. Note that because of the position of the Kunlun fault with respect to the Nov. 8 earthquake, this model suggests that stress changes on the Kunlun fault were negligible and in a sense that would not favor left-lateral slip there (it is nearly nodal). On the Altyn Tagh fault, the section of that fault that lies more than 150 km southwest of 87.5-E, 38.0-N (the center point of the white line shown on this map) experienced increased Coulomb stress favoring left-lateral rupture there. The amount of stress change, however, is quite small ( < 0.15 bars). The x-axis on this plot is in kilometers from the center point of the line segment shown in Figure 3: for the Altyn Tagh fault, southwest is to the left; for the Kunlun fault, west is to the left.

This modeling was performed using Shawn Larsen's DISN & DISC programs, and all graphics were prepared with GMT.

Click on any of the small images on this page to display enlarged and much clearer images of these figures. To obtain a color postscript version of any of these files, click on the following: Figure 1, Figure 2, Figure 3, or Figure 4. To obtain a color PDF (Portable Document Format) version of any of these files, click on the following: Figure 1, Figure 2, Figure 3, or Figure 4.

References:

Additional Sources of Information:

About this earthquake

About Stress Change Modeling

This web page was prepared by
Ken Hudnut and Gilles Peltzer
Updated 11/12/97 5:05 pm PST

Here is the CNN news report on the event.