关键词: 汶川M_S8.0地震, 格点搜索法, b值下降幅度, 孕震区

Abstract:

The b-value in the G-R relation proposed by Gutenberg-Richter is often used to evaluate the regional stress level. A lot of studies on the b-value before the M_S8.0 earthquake occurring in Wenchuan, Sichuan on May 12, 2008 have been carried out, and the conclusions are different. There was a big difference in the distribution of low b-value area before the Wenchuan earthquake through space scanning. Most studies on temporal variations believe that there was a significant drop in b-value before the Wenchuan earthquake, but the decrease patterns are different, some results show that the b-value decline was not obvious before the earthquake.

The selection of study area is important in analyzing the temporal variations of b-value. The existing researches are mainly based on the temporal variations of b-value in the surrounding area of the seismogenic zone or rupture area of Wenchuan earthquake. The earthquake samples in the selected area may contain the ones irrelevant with the seismogenic process of the main shock, but these samples will affect the statistical characteristic, leading to the real seismogenic information to be ignored. The low b-value area obtained by spatial scanning is the absolute b-value during the study period, while this low b-value may be caused by the difference in the medium strength or geological structure, and has nothing to do with the dynamic change process of stress. Earthquakes may occur in area with low b-value, and there is no obvious low-value abnormality in absolute b-value before some big earthquakes, while there is a significant decline process before these big earthquakes. The decrease of b-value indicates that the stress level is rising, which reflects the dynamic evolution process of seismogenicity. Therefore, in this study, attention is paid to the decrease process of b-value and the spatial distribution of the decrease range. Based on the theoretical basis that the b-value decreases with the increase of stress, this paper attempts to determine the seismogenic zone before the Wenchuan earthquake by the range of decrease in b-value(Δb). Taking the epicenter of the Wenchuan earthquake as the center, the earthquake catalog from January 2000 to April 2008 in the range of 10°×10° is selected, and the abnormal grid points with a significant decrease in b-value before the Wenchuan earthquake are searched by grid point search method, and the temporal variations of b-value in the region where the abnormal grid points locate are analyzed.

The results show that the abnormal grid points are distributed around the Longmenshan fault zone, Anninghe fault zone and Ma’erkang, when Δb≥5%, and the largest anomalous area is located in the Longmenshan fault zone. The anomaly range decreased gradually with the increase of Δb, which converged to the Longmenshan fault zone and the Ma’erkang area when Δb≥20%, and concentrated in the Dujiangyan-Beichuan section of the Longmenshan fault zone when Δb≥30%, which is located in the southern section of the main rupture zone of the Wenchuan earthquake. The initial rupture point of the Wenchuan earthquake is located in the southeast margin of the anomaly zone. The Longmenshan fault zone is not the only area that experienced the decline in b-value before the Wenchuan earthquake in the study region, but it experienced the largest decline in b-value. The main shock finally occurred in the area with the largest Δb, indicating that the loading of stress before a large earthquake was not limited to the focal area and the stress increased in a large range. The Longmenshan fault zone was subjected to the maximum stress loading before the Wenchuan earthquake, which led to its final rupture. The maximum slip, main rupture range and seismogenic zone of the Wenchuan earthquake obtained by previous studies are similar to the area with the largest Δb obtained in this paper, indicating that the abnormal area with a large Δb can reflect the spatial range of preparation of strong earthquakes. This provides a way for judging the location of strong earthquakes, that is, the region with the largest Δb is more likely to generate strong earthquakes in the future.

We selected an area greater than the distribution of grid points with Δb≥25% as the anomaly area. Earthquakes in this area were calculated. The b-value was calculated with 500 earthquake samples as a time window and the temporal variations of b-value were obtained by sliding with 10 earthquakes. The temporal variations of b-value in the abnormal area are shown in Fig.5b. It can be seen that before the Wenchuan M_S8.0 earthquake, the temporal variations of b-value in the anomaly area experienced three stages of change. In the first stage, the b-value declined continuously from 1.5 to 1.1 from May 2002 to the end of 2005, with a decrease rate of about 27%in three and a half years. In the second stage, the b-value was relatively stable for 2 years, remaining at about 1.1 from 2006 to the end of 2007. In the third stage, the b-value decreased rapidly. In early 2008, the b-value decreased from 1.15 to 1.05 just before the Wenchuan earthquake, reaching a decrease rate of about 9% in just 4 months. Then, the Wenchuan M_S8.0 earthquake occurred. The temporal variations of b-value are similar to the acoustic emission test results of rock, which may reflect the three stages of the development of rock crack underground. The b-value continues to decline while new cracks are constantly formed under the compression of tectonic stress in the first stage. The b-value basically remains stable in the second stage. The relative relationship among the rupture events of different sizes basically remains unchanged, the tectonic stress continues to increase, and the rock internal crack presents a progressive steady extension. The b-value rapidly decreases in the third stage just before the main shock. The rapid decline of b-value before the Wenchuan earthquake may provide a reference for short-term prediction.

The grid point search method has the advantage of searching for the area with Δb in a wide range before strong earthquake, and then, the area with the strongest stress accumulation can be identified according to the amplitude of Δb, thus the seismogenic zone of the future strong earthquakes can be determined. In the actual prediction research, the combination of b-value and other disciplines can provide a useful reference for judging the seismogenic zone.

Key words: the M_S8.0 Wenchuan earthquake, grid point search method, the decreasing amplitude of b-value, the seismogenic zone