Читать книгу Why do buildings collapse in earthquakes? Building for safety in seismic areas - Robin Spence - Страница 17
2.2 Significant Earthquakes Since 2000 2.2.1 The 26.1.2001 Bhuj Earthquake: Mw7.7, 13 481 Deaths
ОглавлениеAt 8.46 a.m. on 26 January 2001, India's 52nd Republic Day, one of the most devastating earthquakes ever to strike India occurred in the Kutch Region of Gujarat State. The earthquake of moment magnitude Mw7.7 and focal depth 23 km was located approximately 70 km east of the historic city of Bhuj. Heavy ground shaking affected an area of tens of thousands of square kilometres, although there was no surface fault rupture observed. The isoseismal map prepared by the EERI team indicates that the area subject to shaking at a level exceeding MM Intensity VIII (‘heavily damaging’) was over 30 000 km2 (Jain et al. 2002). The area has experienced a previous large earthquake (Mw about 8.0) in 1819, and a moderate Mw6.1 one in 1956, and is in the zone with the highest earthquake loading requirements in the Indian code of practice for the design of buildings.
Table 2.1 Significant earthquakes worldwide since 2000, ordered by number of deaths.
Sources: CRED (2020); Pomonis, A., 2020. Personal communication.
Date | Country | World Bank Income Group | Event | Magnitude (Mw) | Total deaths | Total damage (US$bn) | Insured losses (US$bn) | Percent insured |
---|---|---|---|---|---|---|---|---|
26/12/2004 | Indonesia, Thailand, Sri Lanka, India | UM, LM | Indian Ocean earthquake and tsunami | 9.1 | 225 841 | 7.8 | 0.48 | 6.2 |
12/01/2010 | Haiti | L | Haiti | 7.0 | 222 570 | 8 | 0.2 | 2.5 |
12/05/2008 | China | UM | Wenchuan | 7.9 | 87 476 | 85 | 0.37 | 0.4 |
08/10/2005 | Pakistan | LM | Kashmir | 7.6 | 73 338 | 5.2 | 0 | 0 |
26/12/2003 | Iran | UM | Bam | 6.6 | 26 796 | 0.5 | 0 | 0 |
26/01/2001 | India | LM | Bhuj | 7.7 | 13 481 | 2.6 | 0.1 | 3.8 |
11/03/2011 | Japan | H | Great Tohoku a | 9.1 | >18000 | 210 | 37.5 | 18 |
25/04/2015 | Nepal | L | Gorkha | 7.8 | 8 831 | 7.1 | 0.1 | 1.4 |
26/05/2006 | Indonesia | UM | Yogyakarta | 6.3 | 5 778 | 3.1 | 0.04 | 1.3 |
28/09/2018 | Indonesia | UM | Sulawesi a | 7.5 | 4 340 | 1.5 | 0 | 0 |
21/05/2003 | Algeria | UM | Boumerdes | 6.8 | 2 266 | 5 | 0 | 0 |
03/08/2014 | China | UM | Yunnan | 6.2 | 731 | 5 | 0 | 0 |
27/02/2010 | Chile | H | Maule a | 8.8 | 562 | 22 | 8 | 36 |
19/09/2017 | Mexico | UM | Puebla | 7.1 | 369 | 2.9 | 1.3 | 45 |
24/08/2016 | Italy | H | Amatrice | 6.2 | 296 | 7.9 | 0.12 | 2 |
20/04/2013 | China | UM | Lushan | 6.6 | 198 | 6.8 | 0.023 | 0 |
22/02/2011 | New Zealand | H | Christchurch | 6.1 | 181 | 15 | 12 | 80 |
16/04/2016 | Japan | H | Kumamoto | 7.0 | 49 | 20 | 5 | 25 |
23/10/2004 | Japan | H | Niigata | 6.6 | 40 | 28 | 0.76 | 3 |
16/07/2007 | Japan | H | Niigata | 6.6 | 9 | 12.5 | 0.34 | 3 |
20/05/2012 | Italy | H | Emilia‐Romagna | 6.0 | 7 | 15.8 | 1.3 | 8 |
14/11/2016 | New Zealand | H | Kaikoura a | 7.8 | 2 | 3.9 | 2.1 | 54 |
04/09/2010 | New Zealand | H | Darfield | 7.0 | 0 | 6.5 | 5 | 77% |
Income groups are from World Bank data (High, H; Upper‐middle, UM; Lower‐middle, LM; Low, L). See also Table 2.2. Dates are given in DDMMYYYY format. Some casualty and loss data are amended based on more recent estimates (Pomonis 2020).
a Events with significant tsunami impacts are shown.
Load‐bearing masonry is the predominant way of building throughout the affected area, but methods have changed over time. The most common masonry technique is a single‐storey house with walls of random rubble stone masonry set in a mud mortar, with a clay tile roof: these buildings are found everywhere, both in the main towns and in the villages (Figure 2.1). More substantial dwellings use dressed or semi‐dressed stone or sometimes clay brick walls; these are commonly two‐storey buildings. In recent years, the use of reinforced concrete (RC) slabs for floors and roofs, with coursed masonry walls, has become common in the wealthier parts of Kutch (Figure 2.2). The main towns have also significant numbers of multistorey apartment blocks in RC (Figure 2.3). None of these forms of building were spared by the intense and widespread ground shaking.
The major city of Gandhidham, and four large towns Bhuj, Anjar, Bhachau and Rapar, all in the Kutch district, were devastated, as was every village within a wide area. Over 230 000 one‐ and two‐storey masonry buildings and several hundred concrete frame buildings collapsed. However, as pointed out by Sudhir Jain (2016), the collapse rate of buildings in the zone of highest intensity was much lower in this earthquake than in the 1993 Mw6.2 Latur earthquake in India's Maharashtra province where rubble stone walls with heavy mud roof are typical.
In Ahmedabad, about 200 km from the epicentre, severe shaking was experienced and over 100 multistorey RC frame buildings collapsed. A survey of damaged buildings in Bhuj and neighbouring villages by EEFIT (2005), including the author's team, showed that the rubble masonry buildings performed worst (over 30% collapse rate) while masonry with RC slabs and RC frame apartment buildings performed better (7 and 3% collapse rates). The collapse of buildings in Ahmedabad, all of which were of multistorey RC frames, can be attributed to amplification of the ground motion through the deep alluvial deposits on which Ahmedabad stands, coupled with poor design and construction – soft‐storey apartment blocks were common. The Indian earthquake design code in use at the time is well‐written and comprehensive, but it was not binding on private builders, and was largely ignored (see Chapter 8).
Figure 2.1 Stone masonry building in the Kutch district damaged in the Bhuj earthquake.
Figure 2.2 Brick masonry building with reinforced concrete floors damaged in the Bhuj earthquake.
Figure 2.3 Reinforced concrete building in the Kutch district and typical damage patterns. Notice ground floor failure.
Figure 2.4 Damage to a reinforced concrete building in Bhuj, and view of the same building in satellite image, arrow showing viewing direction.
Source: Saito et al. (2004).
This earthquake was one of the first in which available high‐resolution satellite imagery could be used to identify the damage to individual buildings, and this resulted in several studies to develop this technology (Saito et al. 2004) (Figure 2.4).
The final toll of dead and injured shows that altogether 13 481 people were killed in the earthquake (Jain et al. 2002). There were more than 166 000 injured, 20 000 of them seriously. From medical reports, it is clear that both death and injury were mainly the result of traumas associated with the collapse of buildings. Over 1000 school students and teachers were killed, though because it was a public holiday many schools were closed. There were also more adult female than male deaths (Murty 2005). There can be little doubt, though, that failure of weak masonry walls and the resulting collapse of dwellings was the main cause of death, and the magnitude of the death toll is a reflection of the very wide area over which heavy ground shaking was observed, combined with the weakness of the typical masonry buildings.