The AIST, the National Institute of Advanced Industrial Science and Technology (Hiroyuki Yoshikawa, Director), in collaboration with Hiroshima Institute of Technology, Tsuru Gakuen (Kazuhiro Mori, President), Kyoto University (Kazuo Oike, President), the Geological Survey of Pakistan, and others, conducted an on-site survey of affected area by the magnitude 7.6 earthquake on 8 October 2005 in northern Pakistan. As a result of their work, an outline of the surface fault having a length of approximately 65 km and a maximum vertical displacement of 5.5m was identified for the first time.

As part of its research on active faults and earthquakes, the AIST conducts urgent surveys of major earthquakes, both in Japan and overseas. Recently, it conducted emergency surveys of the 2004 Sumatra earthquake and tsunami and the 2003 Bam earthquake in Iran. Furthermore, the predecessor of part of AIST, the Geological Survey of Japan, the former Agency of Industrial Science and Technology of the Ministry of International Trade and Industry conducted emergency surveys of the 1999 Izmit earthquake in Turkey and the 1999 Chi-Chi earthquake in Taiwan.

On 18-19 January 2006, the Pakistan Geological Survey held the "International Symposium on the 8 October 2005 Pakistan Earthquake: Its Significance and Disaster Mitigation" at Islamabad. Representing AIST were 4 staff members and 2 visiting scholars (one each from Hiroshima Institute of Technology and Kyoto University). Four of the members of the AIST delegation also worked in collaboration with the Geological Survey of Pakistan and Oregon State University to conduct on-site surveys of the surface fault associated with the earthquake, on 20-22 January and again on 24-28 January, for a total of 8 working days.

Figure 1 The surface fault rupture of the 2005 Pakistan Earthquake which was confirmed by on-site surveys. The red circles show survey locations with the amount of vertical displacement. The distribution of the active fault (shown by the red line) is based on Nakata and Kumahara (2006) The topographical shading map was produced using DEM based on ASTER satellite images.

As a result of the surveys, a 65 km-long fault was confirmed to have appeared on the ground surface (Figure 1) and researchers were able for the first time to get a picture of the entire fault that caused the earthquake.

The main part of the fault, an approximate 50 km-long section running from the NW to the central part of the fault strand, showed major displacement that was dominated by reverse fault components. Their measurements show maximum vertical displacement to be approximately 5.5 m (the NE side-up), and maximum net displacement to be approximately 9 m including a horizontal component. In addition, at several survey sites minor right-lateral slip components were identified. It is also noted that in the main fault section between the NW and central strands, there was discontinuity and bending of the fault line.

Along the SE strand, the continuity of the surface faulting was unclear. Only at two locations in a mountains fault ruptures were found to be occurring with small amount of right-lateral displacement of not more than a few dozen centimeters.

Judging from the distribution of the newly discovered surface fault, it appears that the earthquake occurred when a 20-30 km-long fault segment in each of the NW and central areas ruptured with major displacement. Then, somewhat small-scale displacement in the SW area may also have led to some rupturing.

The main part of the surface fault responsible for the earthquake appeared along an active fault that had previously been known to exist (Nakata et al., 1991; Tanaka and Kumahara, 2005 and 2006). In addition, the distribution and amount of displacement of the identified surface fault roughly corresponded to the those of crustal movements during the earthquake analyzed from satellite observations, as well as to the under-ground rupture process that was analyzed from seismic waves.

Areas where damage was especially severe, such as in the town of Balakot and the northern hamlets of Muzaffarabad, were either just on the surface fault, or very close to it. The rate of house collapse tended to be greater near the fault than in surrounding areas. These findings suggest that ground deformation resulting from fault displacement caused particularly severe damage, and that ground shaking was especially intense near the fault.

Figure 2 Vertical displacement of 5.5 m of a current stream bed that occurred at a point (Locality 3 in Figure 1) about 5 km SE of Muzaffarabad.

Photographs A: Surface faulting in the town of Balakot. The fields in the foreground have deformed by flexure. The vertical displacement is 2.6-2.8m. (Locality 1 in Figure 1). B: Surface faulting at the north of Muzaffarabad. The vertical displacement is 4.0 meters or greater. (Locality 2 in Figure 1). C: Offset streambed where the maximum vertical displacement was measured. The vertical displacement is 5.5m. (Locality 3 in Figure 1). D: Displacement of the floodplain that was seen near the SE end of the main section of the surface fault. The vertical displacement is 2.0m. (Point 4 in Figure 1).