 |
| Titel |
Numerical Simulation of Stress evolution and earthquake sequence of the Tibetan Plateau |
| VerfasserIn |
Peiyu Dong, Caibo Hu, Yaolin Shi |
| Konferenz |
EGU General Assembly 2015
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 17 (2015) |
| Datensatznummer |
250101473
|
| Publikation (Nr.) |
 EGU/EGU2015-624.pdf |
|
|
|
|
|
| Zusammenfassung |
| The India-Eurasia’s collision produces N-S compression and results in large thrust fault in the
southern edge of the Tibetan Plateau. Differential eastern flow of the lower crust of the
plateau leads to large strike-slip faults and normal faults within the plateau. From 1904 to
2014, more than 30 earthquakes of Mw > 6.5 occurred sequentially in this distinctive
tectonic environment. How did the stresses evolve during the last 110 years, how did
the earthquakes interact with each other? Can this knowledge help us to forecast
the future seismic hazards? In this essay, we tried to simulate the evolution of the
stress field and the earthquake sequence in the Tibetan plateau within the last 110
years with a 2-D finite element model. Given an initial state of stress, the boundary
condition was constrained by the present-day GPS observation, which was assumed as a
constant rate during the 110 years. We calculated stress evolution year by year, and
earthquake would occur if stress exceed the crustal strength. Stress changes due to
each large earthquake in the sequence was calculated and contributed to the stress
evolution.
A key issue is the choice of initial stress state of the modeling, which is actually
unknown. Usually, in the study of earthquake triggering, people assume the initial stress is
zero, and only calculate the stress changes by large earthquakes – the Coulomb failure stress
changes (δ CFS). To some extent, this simplified method is a powerful tool because it can
reveal which fault or which part of a fault becomes more risky or safer relatively.
Nonetheless, it has not utilized all information available to us. The earthquake sequence
reveals, though far from complete, some information about the stress state in the region. If the
entire region is close to a self-organized critical or subcritical state, earthquake stress drop
provides an estimate of lower limit of initial state. For locations no earthquakes occurred
during the period, initial stress has to be lower than certain value. For locations where large
earthquakes occurred during the 110 years, the initial stresses can be inverted if
the strength is estimated and the tectonic loading is assumed constant. Therefore,
although initial stress state is unknown, we can try to make estimate of a range of
it.
In this study, we estimated a reasonable range of initial stress, and then based on
Coulomb-Mohr criterion to regenerate the earthquake sequence, starting from the
Daofu earthquake of 1904. We calculated the stress field evolution of the sequence,
considering both the tectonic loading and interaction between the earthquakes.
Ultimately we got a sketch of the present stress. Of course, a single model with certain
initial stress is just one possible model. Consequently the potential seismic hazards
distribution based on a single model is not convincing. We made test on hundreds
of possible initial stress state, all of them can produce the historical earthquake
sequence occurred, and summarized all kinds of calculated probabilities of the
future seismic activity. Although we cannot provide the exact state in the future, but
we can narrow the estimate of regions where is in high probability of risk. Our
primary results indicate that the Xianshuihe fault and adjacent area is one of such
zones with higher risk than other regions in the future. During 2014, there were 6
earthquakes (M > 5.0) happened in this region, which correspond with our result in some
degree.
We emphasized the importance of the initial stress field for the earthquake sequence, and
provided a probabilistic assessment for future seismic hazards. This study may bring some
new insights to estimate the initial stress, earthquake triggering, and the stress field evolution
. |
|
|
|
|
|
|