**Changes in Apparent Resistivity in the Late Preparation Stages of Strong Earthquakes**

Du Xuebin et al.\*

*Lanzhou Base of Institute of Earthquake Science, CEA, Lanzhou China*

#### **1. Introduction**

196 Earthquake Research and Analysis – Statistical Studies, Observations and Planning

Zmazek, B., Živčić, M., Todorovski, L., Džeroski, S., Vaupotič, J., & Kobal, I. (2005). Radon in

Zmazek, B., Živčić, M., Vaupotič, J., Bidovec, M., Poljak, M., & Kobal, I. (2002b). Soil radon

Vol. 20, No. 6, pp. 1106–1119

4, pp. 649–657

soil gas: How to identify anomalies caused by earthquakes. *Applied Geochemistry,*

monitoring in the Krško Basin, Slovenia. *Applied Radiation and Isotopes,* Vol. 56, No.

In China, a large-scale observation network that is comprised of a number of apparent resistivity (for short, AR) stations has been established for the purpose of earthquake (EQ) monitoring and prediction since the 1966 Ms7.2 Xingtai EQ in Hebei Province. Presently, over 70 AR stations are in observation in seismically active belts in densely populated areas and nearby some of large, medium-sized cities. The 2008 Ms8.0 Wenchuan great EQ in Sichuan province occurred in the AR station network that was located in the border area of both Sichuan and Gansu provinces. In the 1970s-1980s, more than 110 AR stations had been in observation, and in those years, several great earthquakes (EQs) occurred nearby AR stations, such as the 1976 Ms7.8 Tangshan EQ in Hebei province, the 1976 Ms7.2 Songpan-Pingwu EQ in Sichuan province, the 1976 Ms7.4 Longling EQ in Yunnan province, and the 1988 Ms7.6 Lancang-Gengma EQ in Yunnan Province. At an AR station, two horizontally perpendicular observation channels or three channels, more one horizontally skewed channel (a NE or NW channel, as illustrated in Fig.1a), are employed, and for each channel, an AR observation configuration with a symmetry four-electrode array is installed (Fig.1b). For most stations, the current electrode spacing AB 1000 ~ 2000m .

Fig. 1. The observation channels (a) and symmetric four-electrode resistivity array for a channel (b) at a geo-electrical station

<sup>\*</sup> An Zhanghui1, Yan Rui3, Ye Qing2, Fan Yingying1, Liu Jun1, Chen Junying1 and Tan Dacheng1 <sup>1</sup>*Lanzhou Base of Institute of Earthquake Science, CEA, Lanzhou, China* 

<sup>2</sup>*China Earthquake Networks Center, China Earthquake Administration, Beijing, China* 

<sup>3</sup>*Earthquake Administration of Beijing Municipality, Beijing, China*

Changes in Apparent Resistivity in the Late Preparation Stages of Strong Earthquakes 199

great EQs and their recovery AR changes immediately after the two events had a similar changeable pattern in appearance. Therefore, the anomalous AR changes were believed to

Fig. 2. AR changes observed at CLH station before the 1976, Ms7.8 Tangshan EQ (a), and at

The second problem is on the drop pattern of the two AR anomalies. It is obvious from figure 2 that the AR changes have a notable drop pattern during the imminent stage of the two impending great EQs. In general, the underground medium is abundant with water and the medium resistivity is susceptible to water, therefore, we believed that the changes of underground water resulted in these drop AR changes before the two impending EQs. Figure 3 are the raw AR daily-mean curve at CLH station and the raw curve of water level

Fig. 3. The AR change observed at CLH station and the water level change observed at

Longjiadian station

WDH station before the 1976, Ms7.2 Songpan-Pingwu EQ (b)

be the anomalies related to the two great EQs.

Some of the AR stations have continuously observed for more than 40 years, and a lot of AR data are observed. Using these data we can understand whether the earthquake-related AR changes are recorded or not, which is an issue that should be seriously discussed because the precursory anomalies before EQs have been strongly debated. In this chapter we try to study the issue from two respects: (1) the EQ case research on AR changes recorded before EQs, and (2) the theoretical analysis on anisotropic AR changes related to the maximum principal compression stress ( P axis) azimuth of focal mechanism solution, nearby an EQ focal region in the late preparation stages of a strong EQ. This chapter will help objectively evaluate and comprehend the AR precursory changes related to EQs.

### **2. EQ case research**

Before several great EQs with magnitude of Ms≥7.0 and some moderate EQs, obvious AR anomalies, which include the medium-term anomalies that start to appear about 2-3 years to several months before EQs and the imminent anomalies that start to appear about 3 months to several days preceding EQs, are recorded at geo-electrical stations nearby the epicentral areas, in China. Some of the anomalies are ascertained after EQs, whereas some are discerned before EQs. More interestingly, two groups of EQs with magnitude of Ms 6 nearby AR stations are actually forecasted on a one-year time scale using medium-term AR anomalies, for which the expected EQ magnitudes and forecasted locations are all right.

#### **2.1 AR changes ascertained after EQs**

#### **2.1.1 Reappearing AR anomalies before two great EQs**

In 1976, three great EQs with magnitude of Ms 7.2 occurred in the mainland Chinese, such as the July 28 Ms7.8 Tangshan EQ in Hebei province, the Aug. 16 Ms7.2 Songpan-Pingwu EQs in Sichuan province, and the May 29 Ms7.4/7.3 Longling EQ in Yunnan province. Of the EQs, the Tangshan EQ and Songpan-Pingwu EQ occurred in a local AR station network, and significant AR changes were ascertained after the EQs. As shown in figure 2[1], obviously drop AR changes were recorded at station Changli-Houtuqiao (CLH) in Hebei province, before the Ms7.8 Tangshan EQ (80 km from CLH station) and at station Wudu-Hanwang(WDH) in Gansu province, before the Songpan-Pingwu Ms7.2 EQ (105 km from WDH station), respectively. On those days, the two stations were in normal operation. It can be seen from raw AR daily mean curves of the two stations that AR changes fell all during about 40 days before the occurrence of the two great EQs, which were notable short-term anomalies proceeding the two EQs. Especially, during about 20 days before the two EQs AR changes started to fall by a larger margin, which were imminent anomalies before the impending EQs. Then, immediately after the occurrence of the two EQs, the drop changes started to rise. Based on the two AR anomalies corresponding to the two EQs, we can discuss two problems as follows:

The first problem is on the repeatability of AR anomalies before the two EQs. The Ms7.8 Tangshan EQ occurred in Hebei Province, in east China, and the M7.2 Songpan-Pingwu EQ occurred in north Sichuan province, in western China. The distance between the two EQ epicenters were beyond 1500 km; and the two stations were located in different tectonic units (station CLH was nearby the Cangdong fault belt in the Beijing-Tianjin-Tangshan area, in east China, and station WDH was nearby the Bailong river fault belt in South Gansu province, in western China). The underground geo-electrical structure of the two stations was very different. Nevertheless, the drop AR changes of the two stations before the two

Some of the AR stations have continuously observed for more than 40 years, and a lot of AR data are observed. Using these data we can understand whether the earthquake-related AR changes are recorded or not, which is an issue that should be seriously discussed because the precursory anomalies before EQs have been strongly debated. In this chapter we try to study the issue from two respects: (1) the EQ case research on AR changes recorded before EQs, and (2) the theoretical analysis on anisotropic AR changes related to the maximum principal compression stress ( P axis) azimuth of focal mechanism solution, nearby an EQ focal region in the late preparation stages of a strong EQ. This chapter will help objectively

Before several great EQs with magnitude of Ms≥7.0 and some moderate EQs, obvious AR anomalies, which include the medium-term anomalies that start to appear about 2-3 years to several months before EQs and the imminent anomalies that start to appear about 3 months to several days preceding EQs, are recorded at geo-electrical stations nearby the epicentral areas, in China. Some of the anomalies are ascertained after EQs, whereas some are discerned before EQs. More interestingly, two groups of EQs with magnitude of Ms 6 nearby AR stations are actually forecasted on a one-year time scale using medium-term AR anomalies, for which the expected EQ magnitudes and forecasted locations are all right.

In 1976, three great EQs with magnitude of Ms 7.2 occurred in the mainland Chinese, such as the July 28 Ms7.8 Tangshan EQ in Hebei province, the Aug. 16 Ms7.2 Songpan-Pingwu EQs in Sichuan province, and the May 29 Ms7.4/7.3 Longling EQ in Yunnan province. Of the EQs, the Tangshan EQ and Songpan-Pingwu EQ occurred in a local AR station network, and significant AR changes were ascertained after the EQs. As shown in figure 2[1], obviously drop AR changes were recorded at station Changli-Houtuqiao (CLH) in Hebei province, before the Ms7.8 Tangshan EQ (80 km from CLH station) and at station Wudu-Hanwang(WDH) in Gansu province, before the Songpan-Pingwu Ms7.2 EQ (105 km from WDH station), respectively. On those days, the two stations were in normal operation. It can be seen from raw AR daily mean curves of the two stations that AR changes fell all during about 40 days before the occurrence of the two great EQs, which were notable short-term anomalies proceeding the two EQs. Especially, during about 20 days before the two EQs AR changes started to fall by a larger margin, which were imminent anomalies before the impending EQs. Then, immediately after the occurrence of the two EQs, the drop changes started to rise. Based on the two AR anomalies corresponding to the two EQs, we can

The first problem is on the repeatability of AR anomalies before the two EQs. The Ms7.8 Tangshan EQ occurred in Hebei Province, in east China, and the M7.2 Songpan-Pingwu EQ occurred in north Sichuan province, in western China. The distance between the two EQ epicenters were beyond 1500 km; and the two stations were located in different tectonic units (station CLH was nearby the Cangdong fault belt in the Beijing-Tianjin-Tangshan area, in east China, and station WDH was nearby the Bailong river fault belt in South Gansu province, in western China). The underground geo-electrical structure of the two stations was very different. Nevertheless, the drop AR changes of the two stations before the two

evaluate and comprehend the AR precursory changes related to EQs.

**2. EQ case research** 

**2.1 AR changes ascertained after EQs** 

discuss two problems as follows:

**2.1.1 Reappearing AR anomalies before two great EQs** 

great EQs and their recovery AR changes immediately after the two events had a similar changeable pattern in appearance. Therefore, the anomalous AR changes were believed to be the anomalies related to the two great EQs.

Fig. 2. AR changes observed at CLH station before the 1976, Ms7.8 Tangshan EQ (a), and at WDH station before the 1976, Ms7.2 Songpan-Pingwu EQ (b)

The second problem is on the drop pattern of the two AR anomalies. It is obvious from figure 2 that the AR changes have a notable drop pattern during the imminent stage of the two impending great EQs. In general, the underground medium is abundant with water and the medium resistivity is susceptible to water, therefore, we believed that the changes of underground water resulted in these drop AR changes before the two impending EQs. Figure 3 are the raw AR daily-mean curve at CLH station and the raw curve of water level

Fig. 3. The AR change observed at CLH station and the water level change observed at Longjiadian station

Changes in Apparent Resistivity in the Late Preparation Stages of Strong Earthquakes 201

aftershocks occurred along the NE-strike fault belt beyond 300 km long, from the Wenchuan county to the Ningqiang county in Shaanxi province. There were six AR stations within 400 around the main epicenter, which were Chengdu station (CDU, 35km), Jiangyou station (JYO, 150 km), Ganzi station (GAZ, 331 km), Mianning station (MNI, 260) and Xichang station (XCM, 360 km) in Sichuan province, and WDH station (300 km) in Gansu province (Fig.4).

Significant AR anomalies were recorded at four stations in the medium-term stage before

The anomalies were recorded at CDU, JYO, GAZ and WDH during the medium-term stage before the EQ, which were in the range of 400km from the main epicenter (Fig.5, Fig.6).

Fig. 5. AR monthly mean changes observed at four stations that were located along the

Songpan-Ganzi active block before the M8.0 Wenchuan EQ

**2.1.2.1 Medium-term AR anomalies before the EQ** 

the Ms 8.0 Wenchuan EQ as follows. 1. Locally concentrated anomalies

observed at underground water station[2], Longjiadian station which was about 20 km from station CLH and nearby the Cangdong rupture zone. We can notice from figure 3 that immediately before and after the occurrence date of the EQ, the drop AR change at station CLH was well corresponding to the rise change in water level at station Longjiadian. The opposite changeable patterns between electric and water are quite significant, which indicated that the water in the underground medium nearby the focal region played an important role in AR changes.

It can be seen from figure 3 that nothing was recorded before the Ms6.9 Ninghe aftershock (Nov. 15, 1976) at the geo-electrical and water stations, a possible reason for which was explained by associating with the mainly active faults in/nearby the focal region and the focal mechanism of the aftershock by Du *et al*.[3-4].
