Earthquake Reaearch in China  2018, Vol. 32 Issue (3): 412-424
A Preliminary Analysis of the Earthquake Disaster of Buildings during Two Destructive Earthquakes in Xinjiang
Wen Heping, Hu Weihua, Tan Ming, Sun Jianing, Tang Lihua     
Earthquake Agency of Xinjiang Uygur Autonomous Region, Urumqi 830011, China
Abstract: This paper introduces the characteristics of earthquake disasters in the 2013 Urumqi MS5.1 and the 2015 Pishan MS6.5 earthquakes, which are directly beneath the cities. Based on the discussion on regional tectonic background, site conditions, seismic fortification level of buildings, we preliminarily analyzed the causes of earthquake disaster for buildings caused by the two earthquakes.Finally, we give some advice on earthquake resistance for residential buildings and earthquake damage prevention in Xinjiang.
Key words: Urban earthquake     Earthquake disaster of buildings     Earthquake resistance for residential buildings     Seismic fortification level     Urumqi earthquake     Pishan earthquake    

INTRODUCTION

There are 4 major earthquake belts in Xinjiang, namely the Tianshan, Altay, west Kunlun and Altun earthquake belt, regions of strong earthquakes in the western part of the Chinese mainland, which suffers the most severe earthquake disasters, and 70% of cities and towns in Xinjiang have been hit by earthquakes with seismic intensity above Ⅶ. The Atushi M8.0 earthquake in 1902 destroyed more than 30, 000 buildings and killed more than 1, 000 people, with an intensity greater than Ⅹ in the meizoseismal area (Editorial Board of "Xinjiang Annals: Seismological Chronicle", 2002). The Wuqia M7.4 earthquake in 1985 caused 64 people dead and more than 10, 000 buildings destroyed (Wang Xiaorong et al., 1999), and counties and towns were destroyed and forced to relocate. The Bachu-Jiashi MS6.8 in 2003 caused 268 people dead and 19, 899 buildings collapsed, which turns out to be the most severe earthquake disaster in Xinjiang since the founding of the People's Republic of China (Jiang Haikun et al., 2014). In the 1970s, Xinjiang earthquake departments organized personnel to conduct scientific investigation on important historical earthquakes, including the above-mentioned historical earthquakes, especially to investigate various kinds of earthquake damage (Editorial Board of"Xinjiang Annals: Seismological Chronicle", 2002). Since the 1990s, earthquake field work has been carried out by the Earthquake Agency of Xinjiang Uygur Autonomous Region to investigate seismic damage of buildings caused by destructive earthquakes.

Since 2000, with the development of the national economy, Xinjiang has begun the renovation projects of shanty towns and dilapidated houses in cities and towns, and the projects of prospering husbandry by settlement and comfortable housing in rural areas, especially the construction of earthquake-resistant houses in poor and high-seismic intensity areas, which have effectively improve the ability of Xinjiang residents to withstand earthquake disasters. After the Bachu-Jiashi MS6.8 earthquake in 2003 to the end of June 2015, more than 50 earthquakes with magnitude over 5.0 occurred in Xinjiang, including 2 earthquakes with MS7.0 and 8 earthquakes with MS6.0. The highest intensity of these earthquakes reached Ⅷ in the densely populated areas, however earthquake disaster loss and casualties are relatively minor.

The Urumqi MS5.1 earthquake on August 30, 2013 and the Pishan MS6.5 earthquake in Hotan on July 3, 2015 are two typical earthquakes in Xinjiang in recent years. The meizoseismal areas of both earthquakes cover the main urban areas. The events belong to the right-beneath-city type, where there are large population density and complex structures of buildings within the influence field. Analytical summary of earthquake damage to the dwellings can serve as a guideline for scientific implementation of earthquake damage prevention in post-earthquake reconstruction. Especially with the work in vulnerable spots, such as dilapidated houses in cities and old houses in rural areas badly damaged by earthquakes, is of positive significance in dealing with future major earthquakes and minimizing earthquake casualties.

1 AN OVERVIEW OF EARTHQUAKE DAMAGE CAUSED BY THE URUMQI MS5.1 EARTHQUAKE ON AUGUST 30, 2013 1.1 Field Survey of Earthquake Damage

According to the China Earthquake Networks Center, a MS5.1 earthquake hit Urumqi, Xinjiang (43.75°N, 87.63°E) at 13:27 p.m. on August 30, 2013, at a depth of 12km. After the earthquake, the Earthquake Agency of Xinjiang Uygur Autonomous Region quickly learned about the disaster through the Earthquake Disaster Rapid Report Network, and sent 3 working groups to carry out a field survey. Intensity survey points are all in the area with high ground motion acceleration recorded by the Intensity Rapid Report Network of Urumqi, and the survey of fundamental situation of 13 sampling points were completed2.

2 Earthquake Agency of Xinjiang Uygur Autonomous Region. The Urumqi MS5.1 Earthquake on August 30, 2013 and Recent Earthquake Tendency in Xinjiang. Reports to the Standing Committee of Xinjiang Autonomous Region, 2013.

Urumqi is the capital of the autonomous region and enjoys important political and economic status. With the development of the economy, urban construction has developed rapidly in recent years; therefore, buildings in the city are mainly constructed with frame structure (including shear wall structure) and brick-concrete structure of a standard design, with high earthquake fortification levels. However, in suburban and urban village areas, there are still some multistage self-built brick-concrete structures and brick-wood structure buildings, as well as a small number of civil structure dwellings, with relatively lower seismic capacity.

According to later information, the quake was felt strongly in the downtown area of Urumqi, with a small number of indoor objects that toppled and fell, with majority of residents in central downtown fleeing their homes to roads and open spaces to seek shelter. The survey found that the worst affected areas include Yamalik Management Committee Community, areas along south Huanwei Road, the 9th Geological Team of Geological Exploration Bureau of Xinjiang and the Liudaowan area (Fig. 1), the comprehensive judgment of the earthquake damage is level Ⅵ, and in general, self-built and old houses were more seriously damaged by the earthquake (Table 1). At other survey points, earthquake damage was mostly found as slight cracks in walls, the ceilings and wall decorations falling off, but building structures were basically intact (Fig. 2). Spatially, survey points with intensity of Ⅵ are mainly distributed near the Yamalik fault, however, surface rupture has not been found in the disaster survey.

Fig. 1 The isoline of PGA and the seismic damage distribution in the Urumqi MS5.1 earthquake (the aftershock distribution by the Xinjiang Digital Seismic Network)

Table 1 Seismic damage of buildings at survey points with intensity Ⅵ based on field sampling

Fig. 2 The seismic damage of buildings in the Urumqi MS5.1 earthquake
1.2 Relevant Data

The Intensity Rapid Report Network of Urumqi of the Xinjiang Digital Seismic Network incorporates 50 sub-stations. Observation stations are densely distributed in the urban areas near the epicenter of the MS5.1 earthquake (Fig. 1), and according to PGA records for strong earthquakes, the maximum horizontal peak ground acceleration was recorded by the Geological Team No.156 substation, with a peak value of 186gal and instrumental intensity reaching Ⅶ. As recorded by the intensity rapid report network, there are 5 strong earthquakes with instrumental intensity reaching Ⅶ and 11 with instrumental intensity reaching Ⅵ. On the whole, the Ⅵ-Ⅶ region of isolines of ground motion accelerations (instrumental intensity) show rapid attenuation in the NW and EW direction.

According to the Earthquake Catalogue of the Xinjiang Digital Seismic Network, up to 12:00 p.m. on September 3, 2013, ML≥2.0 aftershocks of the MS5.1 earthquake showed NWW distribution (Fig. 1). The meizoseismal area in seismograms from the Institute of Geophysics, China Earthquake Administration, shows an intensity of Ⅶ, with a major axis in the NNW direction. The distribution of both is close to the results of instrumental intensity obtained from the intensity rapid report network of Urumqi.

According to the published results on official websites of the China Earthquake Networks Center, Institute of Geophysics, China Earthquake Administration, Harvard University and Earthquake Agency of Xinjiang Uygur Autonomous Region, the focal mechanism of this earthquake shows the nature of the thrust fault with a certain amount of strike-slip movement and a nearly EW-striking fault plane, and the fault plane dips to the south with an angle of 20°-80°, which, in general, is similar to the nature of the nearby Yamalik fault.

2 AN OVERVIEW OF EARTHQUAKE DAMAGE CAUSED BY THE PISHAN MS6.5 EARTHQUAKE ON JULY 3, 2015 2.1 A General Review of the Earthquake

According to the China Earthquake Networks Center, a MS6.5 earthquake hit Pishan County of Hotan, Xinjiang (43.75°N, 87.63°E) at 09:07 a.m., on July 3, 2015, at a depth of 10km. This earthquake caused 3 dead, 260 injured, and caused a direct economic loss of 5.43 billion yuan (RMB). 6.71 billion yuan (RMB) was needed for reconstruction of the earthquake disaster area, which was a major destructive earthquake 3. According to the intensity isoseismal maps released by the China Earthquake Administration (Fig. 3), this earthquake had the highest intensity of Ⅷ. The meizoseismal area with an intensity of Ⅷ is about 1, 110km2, covering 5 villages and towns of Pishan County, the Ⅶ-degree area is about 3, 410km2, covering 10 villages and towns of Pishan County, and the Ⅵ-degree area is about 10, 060km2, covering Pishan County and Yecheng in Kashi.

3 Earthquake Agency of Xinjiang Uygur Autonomous Region. Review of Earthquake Damage Losses Caused by the Pishan MS6.5 Earthquake on July 3, 2015.

Fig. 3 The Seismic Intensity Map of the MS6.5 Pishan earthquake (by China Earthquake Administration)

Fig. 4 The seismic damage of buildings in the MS6.5 Pishan earthquake (a)Yangtake village of Pixina township; (b)Ayageazigan village of Pixina township; (c)Office building of Pixina township; (d)Office building of the National Tax Bureau of Pishan County
2.2 An Overview of Earthquake Damage

The epicenter of this earthquake is 7km from Guma Town of Pishan County, Hotan Prefecture in Xinjiang, and the whole county seat is in the sphere of influence of earthquake intensity Ⅷ. In accordance with the Code for Investigation of Post-earthquake Field Works, the disaster area is divided into 3 subareas for statistical analysis according to different degrees of damage of buildings and different structural types of houses in the area, which respectively correspond to the intensity Ⅷ area, the Ⅶ-degree area and the Ⅵ-degree area.

Residential buildings in cities, towns and villages affected by the MS6.5 earthquake are typical urban-rural dual structures. Guma Town is the seat of Pishan County government, buildings are mainly frame structures and brick-concrete structures in the downtown area and in the suburbs, they are mainly brick-wood structures and a few civil structure. In fact, frame structure and brick-concrete structure in the quake zone are mainly public, government and new residential buildings which are concentrated in counties and villages (towns) with standard fortification, and houses in rural areas are mainly brick-wood structure and civil structure. Local civil structure houses mainly adopt typical type of wooden frame, board filling and adobe structure as in rural areas of southern Xinjiang, and rural dwellings built with local fortification standards following the comfortable housing project are mostly brick-wood structure type, and a small number are brick-concrete structures.

Therefore, the types of building structures involved in the earthquake influence field include civil structures, brick-wood structures, brick-concrete structures and frame structures. Except buildings in Pishan and Yecheng County, structure types of buildings within the influence field of the MS6.5 earthquake are relatively simple, the characteristics of earthquake damage conform to regional site features and structural characteristics of buildings, therefore, only the characteristics of earthquake damage of various types of buildings in the meizoseismal area with intensity of Ⅷ are summarized in this paper (Table 2).

Table 2 Main characteristics of earthquake damage of various types of buildings in the meizoseismal area

In terms of the degree of damage to buildings caused by the earthquake (Table 3), collapsed buildings in the disaster area are mostly civil structures but there were also some old brick-wood structures. Damage to most civil structure houses is above medium level, with little rehabilitation value. More than 50% brick-wood structure houses are slightly damaged, suffering less loss. Many old brick-concrete multi-storey buildings in counties were damaged, 15% of which were damaged at a medium level. Very few frame structure buildings suffer structural failure, which is often slight damage and large area of shear cracks on infilled walls. Only several comfortable houses suffer medium damage at a few spots in meizoseismal area, and more than 95% are intact or slightly damaged in areas with intensity below Ⅶ, which have basically survived the earthquake disaster.

Table 3 Summary sheet of damage ratio of buildings

At the sampling survey site of Ayageazigan village in Pixina township (37.545°N, 78.059°E), one of the 10 brick-wood structure houses with local standards of seismic design collapsed, one seriously damaged, 6 moderately damaged and 2 slightly damaged. All wood-fence-wall houses(civil structure) in the village collapsed, which is the most severe building damage in the disaster area, with integrated intensity reaching Ⅸ. It's an anomalous point with high intensity.

On the whole, it is believed that the main reason for earthquake damage to buildings in the disaster area is the lack of seismic fortification for old houses and poor construction quality.

A large number of ground surface tensile fissures appeared in some parts of Pixina township and Guma town in the Ⅷ-degree area, with widths of 10cm-20cm and extending length of 10m-120m for a single fissure, showing NW parallel oblique fissure distribution. Near the outcrop of springs, river courses and some farm fields, appeared the phenomenon of water spraying and sand emitting, covering some farm fields in large-scale area. Part of Pixina river suffered bank avalanche.

The Pishan MS6.5 earthquake caused 3 deaths in the Ⅷ-degree meizoseismal area. Among them, two people in Guma Town died when brick walls in suburban village collapsed, and one person in Pixina township was killed when an old brick-wood structure houses collapsed. We can therefore say that the collapse of old houses or lack of seismic fortification standards for buildings led to the deaths. Meanwhile, severe destruction of old buildings in the disaster area was also the cause of a large number of people that need to be relocated.

2.3 Other Information

Stations of the Xinjiang Strong Earthquake Network near the epicenter are mainly concentrated in the west section of the southern Tianshan seismic belt and the Yarkant River Basin, far away from the earthquake zone. Some stations with epicentral distances between 190-300km have an instrumental intensity of Ⅴ-Ⅵ. The nearest Selibuya station with an epicentral distance of 191km and horizontal peak ground acceleration of 30.4gal(Ⅴ level). Qiongku, erqiake station, 192km from the epicenter, has the maximum horizontal peak ground acceleration, which is recorded as 62.5gal (Ⅵ level).

According to earthquake catalogue of the Xinjiang Digital Seismic Network, up to 12:00 p.m. on July 13, 2015, most of ML≥3.0 aftershocks of the Pishan MS6.5 earthquake were to the west of the main shock, showing NW distribution in the south of the Zepu buried fault. The intensity of the meizoseismal area in seismograms from the Institute of Geophysics, China Earthquake Administration, is Ⅷ, with major axis in NW direction, which is close to the planar spreading direction of aftershocks and the strike of the Zepu fault.

According to published results on official websites of the Institute of Geophysics, China Earthquake Administration and Harvard University, the focal mechanism of this earthquake shows the nature of a source rupture with mainly NWW low-angle south-dipping thrust motion and a certain right-lateral strike-slip movement, which is similar to the nature of the Zepu fault.

3 DISCUSSION

Building damage caused by earthquakes is actually the result of ground motion effect on buildings caused by the transmission of seismic waves produced by source rupture to the site. In a specific area, regional geological conditions and tectonic settings determine the earthquake rupture process and site factors, and seismic fortification level and construction quality of buildings are the guarantee of shock resistance of buildings. Therefore, it is necessary to discuss possible causes for the differences of building damage caused by the two earthquakes based on these factors.

3.1 Regional Tectonics

Urumqi is located in the middle section of the northern Tianshan fold belt where the Urumqi piedmont depression meets the Bogda uplift. The Urumqi Cenozoic depression where the Urumqi MS5.1 earthquake is located consists of root reverse fault, central detachment surface and front compressive-fold uplift, and on the upper part of the fold zone, Holocene active faults are developed, which is a typical thrust (nappe) thin-skinned tectonic. According to relevant studies, future M7.0 earthquakes may occur near the root fault, while the Xishan fault and Yamalik fault concomitant with the front compressive-fold uplift have the tectonic conditions for M5.0-6.0 earthquakes (Song Heping et al., 2009). Since no direct evidence of surface rupture was found during field investigations, the seismogenic structure for the MS5.1 earthquake is only for speculation, and more scientific evidence requires further studies.

Tectonically, the epicenter area and its vicinity of the Pishan MS6.5 earthquake belongs to the Yecheng-Hotan depression of the southwestern depression of Tarim Basin, and the Pishan area is manifested as the Pishan anticline. The southwestern depression of the Tarim Basin is a large-scale sag formed by superimposition of multiple types of basins in the geological history of the Tarim block, which has displayed characteristics of foreland thin-skinned thrust belt since the Meso-Cenozoic era. Tectonic activity has been intense particularly since Quaternary, and large-scale overthrusting occurred towards basins, forming three lines of nearly EW and NW-trending fault-related folds in front of the Kunlun Mountains. The Zepu fault near the epicenter is a late Pleistocene fault, which has the tectonic conditions for generating M5.0-6.0 earthquakes (Hu Wangshui et al., 1996; Hu Jianzhong et al., 2008; Zhang Wei et al., 2011; Qu Guosheng et al., 2005; Pan Jiawei et al., 2007). The site working group found a group of NW-striking fault scarps between Guma Town and Pixina township in the southwest of the epicenter of the Pishan MS6.5 earthquake, which is considered to be the latest evidence of the activities of the thin-skinned nappe fault fold system. The above-mentioned tensile ground fissures appearing in parts of the meizoseismal area are mainly caused by tensile stress environments along the axis of the Pishan anticline and its two wings during the napping process of the fold system, rather than the manifestation of the non-thrusting source rupture to the surface.

Due to the interaction between basins and mountains, there developed a series of thrust nappe fault folds in the piedmont of both the Tianshan and Kunlun Mountains. In terms of focal depth and regional tectonic background, source rupture of these two earthquakes may occur in an active fault zone formed by compression within the thin-skinned nappe tectonic system. Relevant studies suggest that these active fold faults are capable of generating M5.0-6.0 earthquakes. The Urumqi MS6.5 earthquake on November 13, 1965 and the Pishan MS6.2 earthquake on May 29, 1998 may be a similar type of source rupture (Wang Jian et al., 1999; Qu Yanjun et al., 2003). Moreover, the so-called root faults, which are closer to the cores of fold systems, are the boundaries between tectonic units, which are potential source areas for M≥7.0 earthquakes.

3.2 Site Conditions in the Earthquake Influence Field

The influence field of the Urumqi MS5.1 earthquake is limited to the downtown area of Urumqi. Downtown Urumqi, located in Urumqi valley area, is embraced on three sides by mountains in the east, south and west, with a sloping plain in the north, which on the whole shows two types of geomorphic units, low mountains and hills and alluvial-proluvial plains. Regional strata mainly consists of Quaternary alluvial-proluvial sandy soil and gravel layers, fluvial-lacustrine sediments, slop wash as loess, silt, sandy soil and gravel soil and sedimentary strata of Mesozoic carboniferous-Jurassic system mudstone, sandstone and shale, and the Quaternary deposits are gradually thickened from south to north. In some areas, there are subsidence areas formed by coal mining, groundwater exploitation and wet collapsible loess. Affected by local topography, backfill is often used at engineering sites, and site engineering conditions for buildings are relatively complex, but on the whole, sites of classⅡor above take the majority (Shen Jun et al., 2003; Li Li, 1999; Chen Hongjun et al., 2004). Ⅵ-level points of actual earthquake damage survey for the MS5.1 earthquake are distributed along the Yashan fault. However, it is mostly the results of comprehensive evaluation of old houses and self-built houses with non-standard seismic fortification and comprehensive assessment of site factors, building structures and quality and other factors. The results of the contour line of the horizontal PGA of the MS5.1 earthquake given by the Intensity Rapid Report Network of Urumqi may be closer to the real regional ground motion at a macro level. Although some studies have shown that seismic intensity and ground motion parameters are not closely correlated (Lin Lin, 2011), in the case of the Xinjiang Strong Earthquake Network with a large density of network stations and reasonable layout, in combination with the NWW distribution features of aftershocks, the results recorded by the network may be more convincing.

The Yecheng-Hotan depression in the influence field of the Pishan MS6.5 earthquake is featured by extremely thick deposits since the Tertiary period, presenting landform of piedmont alluvial-proluvial sloping plain. In Pishan County, there are rivers such as the Kangazi river, the Sulegazi river, the Sangzhu river, the Duwa river and the Bosika river, and the main residential area is distributed along river valleys. Strata of sites in most of the disaster area are silty soil, mealy sand soil, clayey silt and sandy soil of Quaternary Holocene alluvial-proluvial and aeolian types, gravel layers can be seen at river terraces. In general, soil in sites is of the medium-soft type. In addition, in piedmont alluvial-proluvial areas, due to surface inflow and seepage flow, underground water level is relatively shallow, which obviously amplifies the effect of ground motions, and together with the above-mentioned regional napping and extensional forces, causing more severe earthquake damage to some dwellings and resulting in some intensity anomaly points. Therefore, site factors in this region are particularly important, and special treatments should be implemented on building structure and foundation at sites of class Ⅲ or above to accommodate the change of characteristic periods of acceleration response spectrum.

3.3 Seismic Fortification for Buildings

Along with social economic development, Code for Seismic Design of Buildings, Code for Design of Masonry Structures and Code for Design of Concrete Structures have been revised three times respectively in 1989, 2001 and 2010. With the development of scientific research, in 1990 and 2001, Seismic Ground Motion Parameters Zonation Map of China was also revised, and the fifth-generation zonation map was issued and implemented in 2015. Therefore, even in accordance with the standards for fortification and structural design back then, within the urban fortification area, restricted by economic conditions, shock resistance capability and fortification level of various kinds of buildings are also different at different times.

According to the third-generation seismic intensity zonation map, seismic fortification intensity is Ⅶ in Urumqi. According to the fourth-generation zonation map, peak ground acceleration is 0.20g in downtown Urumqi, and seismic fortification intensity for buildings is Ⅷ. In the fourth-generation zonation map, ground motion parameters in Pishan County are 0.05-0.15g, among which, Guma Town of Pishan County has the value of 0.10g, with fortification intensity of Ⅶ. In the fifth-generation zonation map, peak ground acceleration is also 0.20g in downtown Urumqi, still 0.10g in Guma Town of Pishan County and nearby Keketiereke village and 0.15g or higher in other villages and towns (General Administration of Quality Supervision, Inspection and Quarantine, the People's Republic of China et al., 2015).

Combined with basic data in previous statements, it can be seen that in terms of basic fortification intensities in both places, the Urumqi MS5.1 earthquake is quite common, while the Pishan MS6.5 earthquake is a rare earthquake with magnitude exceeding local fortification intensity level. Therefore, there are significant differences in building damage caused by these two earthquakes. Studies have shown that the probability of M7.0 earthquakes within 50 years has met the standard for rare earthquakes in a number of densely populated cities in Xinjiang (Zhu Gang et al., 2007), which from another perspective also reflects the limitation of earthquake intensity zonation map of China and the need for seismic microzonation in densely populated cities and towns.

Due to imbalanced economic development in Xinjiang, areas inhabited by ethnic minorities remain relatively underdeveloped, thus the differences in seismic fortification level of residential buildings are more prominent. Targeted researches are needed in particular for old houses with poor shock resistance capability in earthquake zones, as revealed by these two earthquakes. Old houses in urban areas and rural-urban fringe zones can be demolished as far as possible to speed up the transformation of shanty towns and dilapidated houses. Old houses should also be properly disposed of after the construction of comfortable houses in rural areas, which can be demolished as far as possible or used as assistance buildings such as warehouses to reduce the casualties caused by earthquakes.

3.4 Catastrophic Earthquakes and Fortification Level

So far since this century, rural areas in Xinjiang have undergone changes from houses with basically no seismic fortification criteria to the implementation of comfortable housing projects, and earthquake resistance and fortification level of buildings in urban areas has also been improving. The content of the comfortable housing project that has been implemented since 2004 is to take all kinds of construction measures based on local conditions, so as to ensure that rural dwellings have the ability to resist earthquakes of a magnitude equivalent to local seismic fortification intensity. Building types include civil structures with sandwich wood frame, brick-wood and brick-concrete structures with earthquake-resistant constructional measures. Since 2011, the Xinjiang government has made decision to improve construction standards for rural dwellings. There are no special requirements for roofs of new houses, while walls are built with ring beams and constructional columns. The comfortable houses are brick-wood or brick-concrete structures with local basic seismic fortification criteria. At present, the proportion of comfortable houses of brick-wood structure or better in all rural dwellings in Xinjiang is more than 50%, and has reached 100% in high-intensity areas4.

4 Lyu Hongshan et al. Quick Assessment of Earthquake Damage Along the Tianshan Seismic Belt Based on Technology of Rapid Seismogram. A Research Report on Special Project of Seismology, 2015.

Since 2004, more than 60 earthquakes with M≥5.0 have occurred in Xinjiang, including 5 earthquakes with M6.0 and 2 earthquakes with M7.0. The implementation of the comfortable housing project has effectively reduced the casualties and earthquake damage (Tang Lihua et al., 2013; Liu Jun et al., 2014; Tan Ming et al., 2014). All previous earthquakes before the Pishan MS6.5 earthquake created a miracle of zero death. But the intensity of these earthquakes in relatively densely populated areas reached only Ⅶ, lower than or equal to local fortification intensity. From the perspective of actual earthquake damage caused by these two right-beneath-city earthquakes as mentioned in this paper, ground motion of the Urumqi MS5.1 earthquake is obviously lower than the fortification intensity, and only a small number of self-built houses or old buildings with no seismic design are damaged. The Ⅷ-degree meizoseismal zone of the Pishan MS6.5 earthquake exceeds lower fortification intensities in local areas, a large number of urban and rural buildings with no seismic design are damaged, but for buildings with standard seismic fortification, including comfortable houses in rural areas, it shows good performance of no collapse under strong earthquakes.Therefore, it can be concluded that the difference in building structures under different fortification standards is the fundamental cause of the difference of earthquake damage caused by these two earthquakes.

Earthquake damage caused by the Pishan MS6.5 earthquake indicates the possibility of rare ground motion in the piedmont thrust nappe fault fold zone. From this point of view, the residential buildings in most cities of Xinjiang have not been tested by a real catastrophic earthquake. Therefore, in order to minimize earthquake damage, in key areas such as river valleys, oasis and urban densely-populated areas in the tectonic zones similar to the Tarim Basin Ring and the Junggar Basin, the comprehensive fortification level for all kinds of buildings should, at a minimum, be able to withstand the test of right-beneath-city earthquakes under conditions of fortification intensity.

4 CONCLUSIONS

Conclusions are gained by comparative analysis of hazards of the two recent destructive earthquakes in Xinjiang.

(1) Seismic fortification level and site condition for dwellings and buildings are fundamental causes for the obvious difference in earthquake damage caused by these two earthquakes. Old houses in urban and rural areas in Xinjiang are with relatively poor earthquake resistance capability, which need to be properly disposed.

(2) There is the possibility of rare earthquakes in key fortification areas in Xinjiang, such as river valleys, oasis and urban densely-populated areas. Seismic microzonation is essential in order to more accurately determine regional fortification level.

(3) The comfortable housing project has effectively improved seismic fortification level for rural dwellings in Xinjiang.Because Xinjiang is geographically vast, regional geological conditions of residential buildings are quite different, thus elementary work of engineering site survey should be solid and effective.

This paper has been published in Chinese in the journal of Inland Earthquake, Volume 31, Number 4, 2017.

REFERENCES
Chen Hongjun, Hu You'an, Tang Zhongyi, Wang Yujun, Zhang Fuyun, Hou Youqing [J]. West-China Exploration Engineering, 2004, 16(8): 160-162 (in Chinese).
Editorial Board of "Xinjiang Annals: Seismological Chronicle". Xinjiang Annals: Seismological Chronicle[M]. Urumqi: Xinjiang People's Publishing House, 2002 (in Chinese).
General Administration of Quality Supervision, Inspection and Quarantine, the People's Republic of China, China National Standardization Management Committee. Seismic Ground Motion Parameters Zonation Map of China (GB18306-2015) [S]. Beijing: China Standard Press, 2015 (in Chinese).
Hu Jianzhong, Tan Yingjia, Zhang Ping, Zhang Yanqiu. Structural features of Cenozoic thrust-fault belts in the piedmont of southwestern Tarim basin[J]. Earth Science Frontiers, 2008, 15(2): 222–231 (in Chinese with English abstract).
Hu Wangshui, Liu Xuefeng, Chen Yusui, Xiao Ancheng, Liu Shengguo. The Cenozoic structural evolution and its relationship with oil and gas accumulation in the southwestern Tarim Depression[J]. Experimental Petroleum Geology, 1996, 18(3): 244–251 (in Chinese with English abstract).
Jiang Haikun, Fu Hong, Yang Maling, Ma Hongsheng, et al. Earthquake Cases in China (2003-2006)[M]. Beijing: Seismological Press, 2014 (in Chinese).
Li Li. The application of cushion replacement method for treating the soft foundation in Nanhu area of Urumqi City[J]. Geotechnical Engineer, 1999, 11(2): 43–45 (in Chinese with English abstract).
Lin Lin. The Estimation of Wenchuan Earthquake Ground Motion and the Correlation between Strong Ground Motion Parameters and Seismic Intensity [D]. Doctoral thesis. Beijing: Institute of Engineering Mechanics, China Earthquake Administration, 2011 (in Chinese with English abstract). https://www.sciencedirect.com/science/article/pii/S0267726114002425
Liu Jun, Hu Weihua, Li Zhiqiang, Song Lijun, Tan Ming, Guli Zipa. Research on the earthquake resistance project in the disaster reduction of the Yutian M7.3 earthquake on February 12, 2014[J]. Earthquake Research in China, 2014, 30(2): 243–248 (in Chinese with English abstract).
Pan Jiawei, Li Haibing, Van Der Woerd J., Sun Zhiming, Pei Junling, Barrier L., Si Jialiang. Late Cenozoic morphotectonic features of the thrust belt in the front of the West Kunlun Mountains[J]. Geological Bulletin of China, 2007, 26(10): 1368–1379 (in Chinese with English abstract).
Qu Guosheng, Li Yigang, Li Yanfeng, Canerot J., Chen Xinfa, Yin Junping, Chen Xinan, Zhang Ning, Deramond J. Segmentations of foreland belts and their tectonic mechanism in the Southwest Tarim Basin[J]. Science in China (Series D: Earth Sciences), 2005, 48(10): 1585–1598 . DOI:10.1360/02YD0071.
Qu Yanjun, Wang Haitao, Yang Youling. The Pishan May 29, 1998 Earthquake with MS6.2 in Xinjiang Uygur Autonomous Region[C]. In: Chen Qifu, Zheng Dalin, Gao Rongsheng (Editors). Earthquake Cases in China (1997-1999)[M]. Beijing: Seismological Press, 2003 (in Chinese).
Shen Jun, Li Jun, Zhao Chunqing, Ge Min. Detecting and seismic risk estimation of the active fault at the site of proposed Hualing underground parking lots in Urumqi, Xinjiang, China[J]. Inland Earthquake, 2003, 17(1): 66–73 (in Chinese with English abstract).
Song Heping, Shen Jun, Xiang Zhiyong, Li Jun, Rou Jie. Exploration of Active Faults and Evaluation of Earthquake Risk in Urumqi City[M]. Beijing: Seismological Pres, 2009 (in Chinese).
Tan Ming, Chang Xiangde, Sun Jing, Zhang Tao, Hu Weihua. Anti-seismic building damage ratio preliminary analysis of Yutian MS7.3 earthquake on Feb 12th, 2014[J]. Inland Earthquake, 2014, 28(2): 98–103 (in Chinese with English abstract).
Tang Lihua, Sun Jing, Wu Guodong, Wang Wei, Song Lijun, Zhang Yong. Damage characteristics of Atushi-Jiashi MS5.8 earthquake on Aug. 11 in 2011[J]. Journal of Seismological Research, 2013, 36(1): 93–97 (in Chinese with English abstract).
Wang Jian, Kou Dabing, Dai Xiaomin, Zhang Yong, Li Jun, Ai Maiti. The intensity distribution features and loss assessment of Pishan MS6.2 earthquake on May 29, 1998[J]. Inland Earthquake, 1999, 13(1): 50–57 (in Chinese with English abstract).
Wang Xiaorong, Wang Haitao, Yang Youling. The August 23, 1985 Wuqia Earthquake with MS7.4 in Xinjaing Uygur Autonomous Region[C]. In: Zhang Zhaocheng, Zheng Dalin, Xu Jinghua (Editors), Earthquake Cases in China (1986-1988)[M]. Beijing: Seismological Press, 1999 (in Chinese).
Zhang Wei, Qi Jiafu, Li Yong. Structural styles in south-western margin of Tarim Basin and their dominate factors[J]. Chinese Journal of Geology, 2011, 46(3): 723–732 (in Chinese with English abstract).
Zhu Gang, Gao Mengtan, Ren Xuemei. Analysis of seismic hazard in the cities of China[J]. Earthquake Research in China, 2007, 23(2): 124–130 (in Chinese with English abstract).
新疆两次破坏性地震的房屋震害初步分析
温和平, 胡伟华, 谭明, 孙甲宁, 唐丽华     
新疆维吾尔自治区地震局,乌鲁木齐 830011
摘要:介绍2013年新疆乌鲁木齐Ms5.1、2015年皮山Ms6.5两次城市直下型地震震害特征基本情况。从区域构造、场地条件、建筑物设防等方面,对产生房屋震害的原因进行了初步分析。在此基础上对今后新疆民居抗震、震害防御等工作提出一些建议。
关键词新疆乌鲁木齐地震    城市直下型地震    房屋震害    民居抗震