Forensic Geotechnical Analyses on the 2009 Building-Overturning Accident in Shanghai, China: Beyond Common Recognitions

Tan, Yong; Jiang, Wei-Zhen; Rui, Hai-Sheng; Lu, Ye; Wang, Da-Long

doi:10.1061/(ASCE)GT.1943-5606.0002264

Open access

Case Studies

Apr 29, 2020

Forensic Geotechnical Analyses on the 2009 Building-Overturning Accident in Shanghai, China: Beyond Common Recognitions

Authors: Yong Tan, M.ASCE https://orcid.org/0000-0003-3107-5454 [email protected], Wei-Zhen Jiang [email protected], Hai-Sheng Rui [email protected], Ye Lu, A.M.ASCE [email protected], and Da-Long Wang [email protected]Author Affiliations

Publication: Journal of Geotechnical and Geoenvironmental Engineering

Volume 146, Issue 7

https://doi.org/10.1061/(ASCE)GT.1943-5606.0002264

PDF

Abstract

To date, why a 13-story building in Shanghai fell abruptly to the ground in 2009 remains uncertain. By review of physical evidence, limit-equilibrium analyses, extensive calculations, and three-dimensional numerical simulations, this study reveals that the failure arose from a rainfall-triggered undrained general shear failure of the subgrade underneath the adjacent stockpile rather than previously identified lateral ground movement shearing off piles below the building or failure of piles along the excavation side incurring progressive tilting failure of the building. Deep-seated slip failure of the stockpile applied an impact load on the building and immediately produced an unallowable overturning moment around the south building edge; consequently, the building suddenly fell over southward. Furthermore, the following vital phenomena, which received no attention in the previous investigations, are also explored: (1) how this building fell over but an adjacent building subjected to similar conditions did not; (2) how this building supported by piles fell over but a flood wall supported by shallow foundation remained intact, although both were near a 10-m-high stockpile; and (3) how the flood wall near the 10-m-high stockpile remained intact but the flood wall near a 6-m-high stockpile underwent catastrophic failure. Finally, some design and construction issues involved in this failure case are discussed.

Introduction

Descriptions of Events

Development of the Lotus Riverside neighborhood south of the Dingpu River in Shanghai, China included 11 buildings and an underground garage (Figs. 1 and S1 in the supplementary data). Each building consisted of a 13-story superstructure and a basement buried at a depth of 2 m below ground surface (BGS), which was founded on 118 lightly reinforced prestressed high-strength concrete (PHC) pipe piles (400 mm in diameter, 80 mm in wall thickness, and 33 m in length) toed into dense to very dense silty sand; refer to Fig. 2. The uppermost 3.3–3.6 m subsurface soils at the site were composed of fill and firm to stiff silty clay and clay, locally called hard crust, which were underlain by very soft muddy silty clay and clay to 13.0–13.5 m BGS. In December 2008, Buildings 1–10 were completed. Excavation work on the basement of Building 11 started and the removed soils were temporarily dumped on the landscaping area between the flood wall along the south bank of the Dingpu River (Fig. 1) and Buildings 5–7 forming a 3–4 m-high stockpile (Phase 1 of Stockpile 1 in Fig. 2). Over the next six months, Building 11 was completed and excavation work for the underground garage about 7 m to the south of Buildings 6–7 commenced. The uppermost 1.5 m soils were removed on June 1–19, 2009, using a sloped open-cut method. The soils below 1.5 m were excavated on June 20–25, 2009, and the excavation was supported by a soil-mixed wall (0.7 m thick and 9.2 m high) anchored by two rows of 6–9-m-long inclined soil nails. The excavated soils within Section 1 of the garage continued to be dumped on Stockpile 1 to form a 10-m-high stockpile (Phase 2) behind Buildings 6–7, and the excavated soils within Section 2 were dumped on the landscaping area between Buildings 10–11 and the Dingpu River to form a 6-m-high stockpile (Stockpile 2); refer to Fig. 1. At noon on June 26, 2009, the flood wall to the north of Stockpile 2 abruptly failed (Fig. 3), threatening the adjacent two-story temporary shed for construction workers. Four workers in charge of installing windows on Building 7 were relocated into the basement of Building 7 and the other workers were relocated into Buildings 8–9 for living space. To prevent potential failure of other flood walls, Stockpile 2 was quickly flattened from 6 m to 1–2 m high in the afternoon on June 26, 2009. At about 5:37 a.m. Beijing local time on June 27, 2009, Building 7 abruptly fell southward to the ground on the excavation side with little warning. One worker living in the basement escaped southward and was killed by the falling building while the other three survived by running northward. After falling to the ground, the superstructure of Building 7 still remained intact; see Fig. 4. On the same day of the building failure, a forensic diagnosis committee (FDC) including 14 experts was appointed by Shanghai Government and began a forensic investigation on the site. A timeline briefly describing the major events at this site is presented in Fig. 5.

Fig. 1. Plan layout of Lotus neighborhood, where the accident took place.

Fig. 2. Cross section of the site in the south–north direction.

Fig. 3. View of the site after failure of the flood wall near Stockpile 2 on June 26, 2009: (a) view from the east (image by Yong Tan); (b) view from the north (image by Da-Long Wang); (c) failed flood wall and heaved riverbed (image by Yong Tan); and (d) paved road near soil Stockpile 2 (image by Da-Long Wang.)

Fig. 4. View of the site after toppling failure of Building 7 near Stockpile 1 on June 27, 2009: (a) right after falling down of Building 7; (b) several days after falling down of Building 7 (image by Yong Tan); (c) view of Building 7 after failure; (d) view of ground and stockpile after the accident; (e) view of the foundation and the ground below; and (f) view of the pieces of broken PHC pipe piles. [Images (a, c–f) by Da-Long Wang.]

Fig. 5. Timeline describing the major events at the site.

Previous Investigations and Analyses

The field inspection and survey performed by the assigned FDC indicated that neither the other buildings nor the flood wall to the north of Stockpile 1 appeared to be in an immediate danger although southward lateral displacement up to 2.9 cm was surveyed at the footing of Building 6. To prevent potential failure of Building 6, the FDC decided to remove Stockpile 1 and backfill the uncompleted garage immediately. Physical evidence collected postfailure demonstrated no structural deficiency in either the superstructure or the underlying piles for Building 7 (Netease 2009). Preliminary conclusions from the FDC on July 2, 2009, were that the southward horizontal ground movement deriving from unequal lateral earth pressures on the two sides of Building 7 (a 4.6-m-deep excavation 7 m to its south and a 10-m-high stockpile 1 m to its north) sheared off the underlying PHC piles, and then the building fell southward to the ground (Netease 2009; Yin and Xu 2009; Khudeira 2010).

Via two-dimensional (2D) numerical analyses, Chai et al. (2014) postulated that initial tensile failure of the piles on the excavation side led to further building settlement along the south side, which ultimately caused rotational failure of Building 7. Wang et al. (2017, 2019) presented the results of field investigation, three-dimensional (3D) numerical analysis, and centrifuge test for this case. The postfailure physical evidence indicated that the piles along the south building edge underwent eccentric compression failure while the distant piles experienced tensile failure. However, based on their numerical simulation results Wang et al. (2017, 2019) reached the conclusion that bending failure of the piles along the excavation side took place first, which eventually caused tilting failure of Building 7. Despite the discrepancy of their postulates for the cause of the building failure, both Chai et al. (2014) and Wang et al. (2017, 2019) concluded a progressive failure of Building 7.

Other Considerations

To date, the geotechnical community is still not fully aware of exactly how Building 7 abruptly fell southward to the excavation side and the inherent failure mechanism of the building still remains unclear. From taking a close look at the on-site failure photos (Figs. 3–4), an evident feature of a general shear failure of the subgrade similar to those cases of bearing-capacity failure in literature (Peck and Byrant 1953; Norlund and Deere 1970) can be observed, which received no attention from the previous studies. More importantly, none of the previous investigations paid attention to or tried to explore the following vital phenomena: (1) what role did the 5-h intense rainfall just before the accident play in the tilting failure of Building 7; (2) both being subjected to similar conditions, why did Building 7 fall but Building 6 nearby did not; (3) why did Building 7 founded on 33-m-long piles to the south of Stockpile 1 fall to the ground but the flood wall situated on shallow foundation to the north of Stockpile 1 maintain its stability; and (4) why did the flood wall adjacent to 6-m-high Stockpile 2 fail while the flood wall adjacent to 10-m-high Stockpile 1 remained intact. To answer these questions, comprehensive in-depth investigations were conducted in this study, which included field observations, theoretical calculations, global stability assessments by limit-equilibrium method (LEM), extensive comparisons of the design and construction adopted for this project with local design and construction codes and standards, and 3D numerical simulations.

Subsurface Condition

Field geotechnical exploration and laboratory soil tests were carried out twice for this site. The first one was conducted for design from October 19, 2005, to April 11, 2006 (SXGEI 2006). The second one was conducted from September 11 to October 20, 2009 (SGIDI 2009) to verify the reliability of the first one, two and a half months after the building failure. As shown in Figs. 6(a–b), the postfailure cone penetration test (CPT) data and soil parameters (water contents, void ratio, and constrained modulus) were in fairly good agreement with the preconstruction data. Generally, the soil conditions across the site were relatively uniform (Fig. 6 and Table 1). In the upper 60 m, soils were comprised of four distinct horizons. Horizon I was approximately 3.3–3.6 m thick and primarily composed of desiccated fill (Layer 1) and a layer of firm to stiff silty clay and clay (Layer 2). Horizon II consisted of very soft sensitive clayey strata to 13.0–13.5 m BGS, which was subdivided into two layers (Layer 3—muddy silty clay and Layer 4—muddy clay). Horizon III was comprised of four sublayers (Layer

5_{1}

—firm to stiff silty clay interbedded with sand seams, Layer

5_{2}

—medium dense clayey silt, Layer

5_{3}

—firm silty clay, and Layer 6—stiff to very stiff silty clay), extending to 27.5–29.5 m BGS. Horizon IV consisted of dense to very dense silty and sandy soils, which was subdivided into three layers, i.e., Layer

7_{1 - 1}

—medium dense to dense clayey silt, Layer

7_{1 - 2}

—dense to very dense silty sand, and Layer

7_{2}

—dense to very dense silty fine sand extending to considerable depth.

Fig. 6. Field and laboratory test results: (a) field exploration data; (b) typical laboratory data; and (c) relation between water content and shear strength of the top crust.

Table 1. Summary of soil parameters at the site from SXGEI (2006)

Stratum number	Borehole number	Depth (m, BGS)	$γ$ ( $kN / m^{3}$ )	$ω$ (%)	$e$	$c_{c u}$ (kPa)	$φ_{c u}$ (8)	$E_{s}$ (MPa)
1 (Crust)	B2	0–1.4	18.5	28.5	0.87	29	18.5	4.17
	B3	0–1.3	—	—	—	—	—	—
	B6	0–1.1	—	—	—	—	—	—
2 (Crust)	B2	1.4–3.6	17.9	37.8	1.06	18	12.0	4.13
	B3	1.3–3.5	19.7	25.6	0.71	22	22.0	4.00
	B6	1.1–3.3	16.7–19.2	25.1–53.2	0.74–1.01	13–16	16.0–18.0	3.70–4.20
3	B2	3.6–7.4	15.8	65.1	1.82	8	14.0	1.40
	B3	3.5–7.5	15.6	67.9	1.90	7	14.0	1.30
	B6	3.3–7.0	16.5	62.7	1.78	8	12.0	1.53
4	B2	7.4–13.8	16.6	48.5	1.40	9	13.0	2.10
	B3	7.5–13.5	16.7–17.5	38.9–53.6	1.12–1.47	7	17.0	2.20
	B6	7.0–13.0	16.6–17.4	43.3–59.3	1.20–1.58	7–9	10.5–17.5	2.14–3.41
$5_{1}$	B2	13.8–21.3	17.5–18.1	31.7–39.3	0.93–1.12	13	19.5	4.68
	B3	13.5–19.2	17.8–18	36.8–37.8	1.03–1.07	13–16	19.5–22.5	4.40
	B6	13.0–17.5	17.3–17.7	35–38.8	1.08–1.09	13	18.0	3.16
$5_{2}$	B2	—	—	—	—	—	—	—
	B3	19.2–21.5	—	—	—	—	—	7.25
	B6	17.5–20.4	18.4–18.5	27.4–27.6	0.83–0.84	10	28.5–29.0	7.30
$5_{3}$	B2	21.3–26.6	17.7–17.9	33.6–35.6	0.99–1.03	15–16	16.0–19.0	4.53–5.18
	B3	21.5–28.0	17.7–20.3	19.4–37.6	0.57–1.07	16–47	16.0–18.5	4.51–9.14
	B6	20.4–25.9	18–18.1	31.7–33.6	0.95–0.97	15–16	17.0–20.0	4.36–5.19
6	B2	26.6–28.8	19.3–19.6	23.5–24.0	0.68–0.73	31–40	15.0–19.5	7.18–11.14
	B3	28.0–29.7	—	—	—	—	—	—
	B6	25.9–27.6	19.8	21.5	0.64	31	23.5	9.46
$7_{1 - 1}$	B2	28.8–32.6	19.0	24.0	0.73	10	27.5	8.12
	B3	29.7–32.8	19.4	24.5	0.70	12	27.5	9.83
	B6	27.6–31.3	18.5–19.2	23.2–34	0.70–0.92	10–15	25.0–26.5	9.13–9.61
$7_{1 - 2}$	B2	32.6–42.0	18.0–19.0	25.8–35	0.75–0.99	7–10	27.0–31.5	10.31–11.58
	B3	32.8–40.0	18.6–18.7	29.1–31.4	0.82–0.87	7–10	27.5–29.5	8.71–11.02
	B6	31.3–40.0	18.3–18.8	27.1–31	0.79–0.89	6–9	28.0–31.0	9.46–11.92
$7_{2}$	B2	Below 42.0	18.8–19.1	25.6–26.6	0.73–0.77	6–8	30.0–33.0	12.88–13.04
	B3	Below 40.0	18.8–19.4	23.5–26.9	0.67–0.78	6–7	31.0–31.5	14.39–17.77
	B6	Below 40.0	18.6–19.5	22.7–27.4	0.65–0.76	5–7	30.5–33.0	12.30–19.17

Note: $γ$ = soil unit weight; $ω$ = water content; $e$ = void ratio; $c_{u}$ = cohesion measured by CUDSST; $φ_{c u}$ = friction angle measured by CUDSST; CUDSST = consolidated undrained direct simple-shear test; $E_{s}$ = constrained modulus measured by odometer test; and BGS = below ground surface.

In general, long-term phreatic water level in Shanghai was located at about 1.0–1.5 m BGS, which fluctuated with weather changes. At this site, the measured phreatic levels were around 2.5–3.2 m BGS by the initial field exploration program (SXGEI 2006) and 2.86–3.10 m by the postfailure field exploration program (SGIDI 2009).

Postfailure Investigations

Simple Theoretical Analyses

Until now, the abrupt building failure has been widely attributed to the unequal lateral earth pressures on the two sides of Building 7 (an excavation to its south and a stockpile dumped to its north), which caused lateral southward ground movement below Building 7 and consequent shear-off of the piles. However, this opinion is questionable. From the perspective of structural mechanics, the pile heads cast within the building footing can be treated as fixed supports and the pile tips toed into the dense to very dense silty sand can be treated as pinned supports, i.e., each pile below the building footing can be simplified as a beam fixed at one end and pinned at the other end (a statically indeterminate beam) in theoretical analyses. If the lateral southward ground movement did shear off the piles, the piles on the north building side should have fractured earlier than those on the south side. Thus, Building 7 would have fallen northward to the stockpile instead of southward to the excavation. On the other hand, if the lateral ground movement (lateral spreading of the ground) was not strong enough to shear off the piles, the piles on the north side would have developed greater southward lateral deflection than the piles on the south side as a result of much greater lateral earth pressures on the north side; refer to the pressure bulb (Lambe and Whitman 1969) in Fig. 2. This would have incurred sway of elevated building gravity center and produced northward overturning moment. As a consequence, the building should have tilted northward rather than southward; refer to the schematic illustration in Fig. 7(a). This deduction can be also validated by the soil-stockpiling cases in the literature. As reported in Fellenius (1972), Sy et al. (2011), and Moffitt and Shelly (2015), lateral spreading of the ground due to the soil stockpiles had adjacent pile-supported building/bridge piers displaced laterally opposite to the stockpiles and meanwhile the superstructures tilted to the stockpiles. Evidently, the sudden southward fall of Building 7 [Fig. 7(b)] was not caused by the lateral ground movement arising from the inequality of lateral earth pressures on its two sides. In addition, the stockpile would have induced much greater ground compression surrounding the piles on the north side, accompanied by much greater downdrag force and settlement for the piles; refer to Fig. 7(c), Fig. 8 and “Theoretical Calculation of Stockpiling Effect on Adjacent Building 7” in the Supplemental Data. Apparently, either the lateral or the vertical additional earth pressures in the subgrade resulting from Stockpile 1 would have caused Building 7 to tilt northward rather than southward [Fig. 7(d)].

Fig. 7. Scenarios of the building failure in case of excessive lateral or vertical additional stresses on piles generated by the adjacent stockpile: (a) scenario in case of lateral ground spreading only; (b) actual scenario; (c) scenario in case of ground compression only; and (d) scenario in case of combined results of $Δ σ_{h}$ and $f_{v}$ .

Fig. 8. Preliminary evaluation of long-term stockpiling effect on Building 7: (a) conceptual calculation model; (b) plan layout of strip footing atop PHC pipe piles; and (c) estimated development of downdrag forces and pile settlements due to Stockpile 1.

Another popular opinion on the building failure was that initial failure of the piles on the south building side led to progressive tilting failure of Building 7. Excavation-induced ground movement might impose adverse effects on piles underneath superstructures in close proximity (Finno et al. 1991; Poulos and Chen 1997; Korff et al. 2016). However, the excavation was only 4.6 m deep and 7 m distant from Building 7; its ground settlement influence zone, in terms of either the classic wedge slide by Rankine theory or the empirical criterion, where the influence zone is related to

2 H_{e}

(

H_{e} = 4.6 m

, excavation depth) (Clough and O’Rourke 1990; Hsieh and Ou 1998; Tan and Wang 2013a, b), was not likely to have extended below Building 7 (Fig. 2). As reported by Tan et al. (2016), a much deeper excavation (

H_{e} = 24.8 - 25.2 m

) in Shanghai barely caused settlements of adjacent high-rise buildings founded on long piles toed into dense to very dense silty sand like this site. Furthermore, the theoretical calculation results in Fig. 8(c) show that the piles on the north (stockpile) side would have settled much more than the piles on the south (excavation) side did. All of these analyses revealed that the possibility of the piles on the excavation side failing/settling in advance of the piles on the stockpile side was slim. Additionally, since the buildings had RC frame-shear wall superstructures, excessive local pile settlements would impair effective wall stiffness first; then, cracked walls would gradually conform to underlying ground settlement trough. If Building 7 had undergone a progressive failure, its superstructure would have already cracked severely before it was about to reach leaning instability. According to the eyewitnesses, Building 7 fell to the ground in 5–10 s without warning; no crack was detected on the building facade before the failure. All of these indicated that a progressive failure did not occur for Building 7.

Preliminary Global Stability Assessment

It is clearly shown in Figs. 3–4 that both the sites of Stockpiles 1 and 2 exhibited typical features of global stability failure. To verify this observation, some preliminary stability analyses were performed using the 2D LEM version 2012 software Slope/W (2012), which is based on the method of slices. Although LEM is unable to calculate stress in soil mass, it has already been proven in practice to be reliable for evaluating slope stability (Duncan 1996; Duncan et al. 2008; Han et al. 2010; Wright 2013; Leshchinsky and Ambauen 2015; Stark et al. 2018; Tan et al. 2018). The method of Morgenstern and Price (1965) was used in the analysis, because it satisfies both force and moment equilibrium rather than just satisfying either moment or force equilibrium (Fellenius 1936; Janbu 1954; Bishop 1955). Regarding strength parameters for stability analysis, consolidated undrained (CU) strengths of fine-grained soils are recommended by Vanden Berge and Wright (2016) for undrained slope stability analysis in practice. Because (1) Shanghai soft clay is a kind of normally consolidated soil, and (2) both excavation and stockpiling were short-term activities at this site, CU strength parameters (Table 1) measured by consolidated undrained direct simple-shear tests (CUDSST) were adopted for analysis, which is also recommended by both the state and local design specifications in China [GB50007 (Ministry of Construction of China 2002); DG/TJ08-61 (Shanghai Urban and Rural Construction and Transportation Committee 2010)]. In Slope/W (2012), the structural elements (flood wall, earth retaining wall, building basement, and PHC piles) were treated as shear reinforcements and their displacements were unable to be simulated.

Soil Stockpile 1

Since multiple cases have been reported for slope failures triggered by excavations or cuts at or near the toes of earthen embankments or waste landfills (Stark et al. 2010; Bonaparte 2018), it seemed to be plausible that the excavation might have reduced resisting forces in a global stability failure extending from the stockpile crest through the very soft clays. As marked in Fig. 2, the excavation was 22 m (

4.8 H_{e}

) and 32 m (

7 H_{e}

) to the southern toe and the southern crest of Stockpile 1, respectively, which was far beyond the potential excavation influence zone within

2 H_{e}

behind the retaining wall. Moreover, as schematically illustrated in Fig. 2, the excavation influence zone did not overlap with the additional stress zone of Stockpile 1. Most importantly, the buried building basement and its underlying piles located between the excavation and Stockpile 1 would have functioned as barriers mitigating potential interaction between the excavation and the stockpile. Hence, it can be inferred that the existence of the building basement and underlying piles would have enhanced the global stability of Stockpile 1 to some extent (e.g., Kourkoulis et al. 2011; Tan et al. 2018). To verify this barrier effect, global stability analyses were conducted using Slope/W (2012) for two cases with and without the building basement and its underlying piles, respectively. Figs. 9(a and b) present the global stability analysis results for these two cases.

Fig. 9. Global stability analysis result for the site of Stockpile 1 with respect to: (a) without basement and piles at Building 7; (b) with basement and its underlying piles at Building 7; (c) with basement and its underlying piles and accounting for rainfall effect at Building 7; and (d) with basement and its underlying piles and accounting for rainfall effect at Building 6.

As indicated in Fig. 9(a), in case of no building basement and piles, Stockpile 1 behind Building 7 had a minimum factor of safety,

{FOS}_{\min}

, of 1.017 against a shallow toe slip failure and a factor of safety (FOS) of 1.4–1.6 against a deep-seated slip failure (bearing-capacity failure). If both the building basement and its underlying piles were considered in analysis, its

{FOS}_{\min}

slightly increased to 1.027, but its FOS against deep-seated slip failure increased to 1.6–2.0. Because the stability of the stockpile above the ground level was dominated by slope gradient, slope height, soil unit weight, and soil shear strength, the existence of the buried basement and its underlying piles hardly increased the

{FOS}_{\min}

against a shallow toe slip failure. However, FOS of the stockpile against deep-seated slip failure increased significantly with the presence of the buried basement and its underlying piles. These comparisons proved the previously postulated barrier effect.

Based on these analyses, the 4.6-m-deep excavation can be precluded as a major factor associated with the global stability failure of Stockpile 1 behind Building 7. As recorded by the local weather station (Fig. 10), about half an hour before Building 7 fell over, there was a heavy rainfall lasting from 0:00 a.m. to 5:00 a.m. on June 27, 2009. As recognized in literature (Stark and Duncan 1991; Collins and Znidarcic 2004; Gamez and Stark 2014; Stark et al. 2017), shear strength of desiccated stiff clay decreased very rapidly to fully softened strength once the clay was soaked. Using filter paper method, osmotic method, and vapor phase technique, Ye et al. (2006) explored the suction in the desiccated top crust at 1.5–1.8 m BGS in Shanghai. Consistent with the findings in the literature, the measured suction in the crust reduced quickly owing to saturation. Based on direct shear tests on the soil samples from the top crust in Shanghai, Hu and Fu (2001) investigated the relationship between soil shear strength,

τ_{f}

, and water content,

ω

, with respect to four different vertical stresses,

q

, of 100, 200, 300, and 400 kPa, respectively. As plotted in Fig. 6(c),

τ_{f}

decreased rapidly as

ω

increased.

Fig. 10. Summary of the weather information before overturning of Building 7.

As pointed out by Houston (2019), rainfall that results in reduction of suction is the most common triggering mechanism of slope failure. Based upon the preceding analyses, it can be postulated that the abrupt deep-seated slip failure of Stockpile 1 behind Building 7 might have been triggered by the heavy rainfall. Intense water infiltration into the ground caused rapid degradation of shear strength of the desiccated crust and increased weight of the stockpile due to saturation; thus, an undrained general shear failure of the subgrade below Stockpile 1 took place, featuring a deep-seated slip failure surface. To verify this postulate, another LEM analysis was conducted for Stockpile 1 behind Building 7, in which reduced strength parameters were used for the top crust. As plotted in Fig. 6(b),

ω

was around 20%–30% for the top crust;

q

resulting from the weight of Stockpile 1 on the crust was estimated by

q = γ_{s} \times h

, where

γ_{s}

was unit weight of the soil stockpile, equal to

16.5 kN / m^{3}

, and

h

was the height of stockpile. As plotted in Fig. 2, the 10-m-high Stockpile 1 with a trapezoid cross section can be simplified as a 5.75-m-high uniform strip load for analysis. Then,

q = 16.5 kN / m^{3} \times 5.75 m = 94.9 kPa

. Because the rainfall lasted for 5 h with precipitation of 23.7 mm (Fig. 10), it can be postulated that the crust was already saturated before the building failure. Therefore, it was reasonable to assume that the crust before the building failure had a similar water content,

ω

, as its underlying saturated soft muddy clay below the phreatic water level, whose

ω

was 40%–60% [Fig. 6(b)]. As plotted in Fig. 6(c), in the case of

q = 100 kPa

(close to

q = 94.9 kPa

of this case),

τ_{f}

reduced from about 90 kPa at

ω = 20 % - 30 %

to about 40 kPa at

ω = 40 %

. In light of this, the strength parameters of the top crust were reduced by 50% as previously introduced in the section entitled “Preliminary Global Stability Assessment” and the saturated unit weights of the stockpile and the crust were adopted in the LEM analysis to account for the rainfall effect. As shown in Fig. 9(c), Stockpile 1 would undergo a deep-seated slip failure and the estimated conceptual failure surface by LEM was in reasonable agreement with the field observation.

As shown in Fig. 4(d), Stockpile 1 behind Building 6 did not undergo slip failure like Stockpile 1 behind Building 7. This inconsistency largely derived from the following two facts: (1) the 10-m-high stockpile behind Building 6 had a smaller size in plane than that behind Building 7 (Fig. 1), and (2) the stockpile behind Building 6 featured a more gentle slope angle (38°) than that (45°) behind Building 7 (Fig. 2). Thus, the stockpile behind Building 7 was more susceptible to slip failure than that behind Building 6. To verify this postulate, a new LEM analysis was conducted for the stockpile behind Building 6, in which the rainfall effect was considered. As presented in Fig. 9(d), the stockpile behind Building 6 would not undergo a deep-seated slip failure after the rainfall, although its

{FOS}_{\min}

against a shallow toe slip failure was only 1.014.

Flood Walls Adjacent to Stockpiles 1 and 2

Stability analyses were performed for the flood walls (Fig. 11) to the north of both Stockpiles 1 and 2, using Slope/W (2012). Fig. 12(a) presents the global stability analysis result for the site of Stockpile 2. In case of no displacement of the flood wall (flood wall only provided shear reinforcement in the LEM analysis), Stockpile 2 would have a

{FOS}_{\min} = 1.134

against a global slip failure. Despite the flood wall adjacent to Stockpile 1 did not fail, global stability analyses were done for it as well. As shown in Fig. 12(b), Stockpile 1 had

{FOS}_{\min} = 1.376

in case of no rainfall. If the rainfall effect was considered in analysis, its

{FOS}_{\min} = 0.975

, already approaching deep-seated slip failure [Fig. 12(c)].

Fig. 11. Cross section of flood wall to the north of soil Stockpiles 1 and 2.

Fig. 12. Global stability analysis results: (a) the flood wall behind Stockpile 2; (b) flood wall behind Stockpile 1 without rainfall; and (c) flood wall behind Stockpile 1 accounting for rainfall.

Comprehensive Calculations and Analyses

The preceding LEM analyses revealed that the global stability failure of Stockpile 1 behind Building 7 could have been triggered by the heavy rainfall. However, the LEM analysis results for the flood walls adjacent to Stockpiles 1 and 2 were opposite to the field observations. In light of these, comprehensive in-depth calculations and analyses were carried out in the following sections.

Flood Wall along the Dingpu River

The cantilever concrete flood wall along the south bank of the Dingpu River, which was about 6 m to the north of Stockpile 2, failed abruptly at noon on June 26, 2009 (Fig. 3), but the flood wall about 10 m to the north of Stockpile 1 remained intact even after failure of Building 7 one day later [Fig. 4(b)]. No sign of damage to either the flood wall or the paved road behind Stockpile 1 was observed. Field inspection revealed that the damaged flood wall (about 83-m-long) displaced up to 4 m into the Dingpu River and the riverbed in front of the damaged flood wall heaved dramatically above the river water. The visually detected heave zone extended around 15–20 m northward into the river [Figs. 3(a–c)]. The paved road directly behind the damaged flood wall underwent conspicuous differential subsidence and the concrete pavement crumbled severely. Pronounced gaps up to 0.7 m wide were measured on the paved road near the zone directly behind the failed flood wall [Fig. 3(d)]. Evidently, the flood wall had undergone both a global stability failure featuring a deep-seated slip surface of the subgrade and a sliding failure; refer to the schematic illustrations in Figs. 13(a–b). Meanwhile, the flood wall head tilted northward instead of southward [Fig. 3(c)], an obvious sign of overturning failure [Fig. 13(c)]. If the failure of the flood wall was caused by a bearing-capacity failure of the subgrade below the flood wall footing, the subgrade would have featured a shallow shear failure surface rather than the observed deep-seated shear failure surface for a global stability failure [Fig. 13(a)]. This postulate can be validated by the following bearing-capacity analyses for the subgrade below the flood wall footings.

Fig. 13. Schematic illustration of failure of the flood wall behind Stockpile 2.

As schematically illustrated in Fig. 11, the flood wall was a cantilever wall and the vertical stress on its subgrade was roughly equal to the weight of the soils above its footing. Hence, the flood wall did not undergo a bearing-capacity failure in the case of no stockpile nearby. With the presence of the stockpiles nearby, excessive additional vertical stress,

Δ σ_{v}

, on the footing base of the flood walls due to the stockpiles might incur a bearing-capacity failure of the flood walls. Based on this consideration, bearing capacity,

q_{ult}

, of the subgrade below the flood wall footing was estimated by the bearing-capacity equation of Terzaghi (1943). For a strip footing with an undrained condition,

q_{ult}

is equal to

q_{ult} = 5.7 S_{u} + q_{s} = 224.5 kPa

(1)

where

S_{u}

= undrained shear strength of subgrade; and

q_{s}

= weight of soil above the base of the footing. If the equation proposed in Das and Sobhan (2017) for strip footing under an undrained condition was used,

q_{ult}

is equal to

q_{ult} = 5.14 S_{u} [1 + 0.2 (\frac{b}{L_{f}})] [1 + 0.4 (\frac{D_{f}}{b})] + q_{s} = 367.8 kPa

(2)

where

b

= width of a strip footing, equal to 2.3 m in this case; and

L_{f}

= length of strip footing. Because of the large

L_{f}

in this case,

(b / L_{f}) \approx 0

. Apparently, the equation of Das and Sobhan (2017) accounting for both footing shape and depth factors yielded a much greater

q_{ult}

than the equation of Terzaghi (1943). Using Boussinesq solution (Terzaghi 1943), the estimated

Δ σ_{v} = 2.5 - 4.1 kPa

for the flood wall adjacent to Stockpile 1 and

Δ σ_{v} = 7.9 - 14.4 kPa

for the flood wall adjacent to Stockpile 2. Then, the factor of safety,

{F S}_{1}

, against a bearing-capacity failure of the flood walls can be estimated by

{F S}_{1} = \frac{q_{ult}}{Δ σ_{v} + q_{s}}

(3)

In the case of

q_{ult} = 224.5 kPa

, the calculated

{F S}_{1} = 2.61 - 2.66

for the flood wall adjacent to Stockpile 1 and

{F S}_{1} = 2.32 - 2.50

for the flood wall adjacent to Stockpile 2; in the case of

q_{ult} = 367.8 kPa

,

{F S}_{1} = 4.27 - 4.35

for the flood wall adjacent to Stockpile 1 and

{F S}_{1} = 3.82 - 4.09

for the flood wall adjacent to Stockpile 2. The preceding calculations indicated that the subgrade below the footings of the flood walls would not have undergone a bearing-capacity failure due to the presence of Stockpiles 1 and 2 in the proximity. For detailed calculations of

q_{ult}

and

{F S}_{1}

, refer to “Calculation of Factor of Safety against Bearing-Capacity Failure for Both Flood Walls and Stockpiles” in the Supplemental Data.

Based on the previous analyses, the likely failure model of the flood wall adjacent to Stockpile 2 can be schematically depicted by Fig. 13(d). Undoubtedly, the failure of the flood wall is inherently correlated with Stockpile 2 nearby. However, it was unknown which type of failure occurred first and then triggered the others, or whether the three failures happened simultaneously. To identify this and find out why the flood wall behind Stockpile 1 remained intact while being subjected to a higher surcharge nearby, calculations were carried out to evaluate FS against bearing-capacity failure underneath Stockpiles 1 and 2 and against sliding and overturning of the adjacent flood walls.

Evaluation of FS against Bearing-Capacity Failure of the Subgrade below Stockpiles

As shown in Fig. 6(a), the subgrade below the stockpiles was composed of a crust underlain by the soft strata. The top crust functioned as a stiff working platform intermediate between the stockpile and the underlying soft strata; thus, surface load from the stockpile was spread over a wider area, having a load-spread or stress-diffusion effect as described by Michalowski (1992, 2004), Burd and Frydman (1997), Wei et al. (2012), and Liu and Yu (2017). Consequently, additional pressure transferred from the surcharge to the underlying soft strata was smaller than that in the absence of the crust, i.e., bearing capacity of the soft strata (Layers 3–4) was enhanced by the upper crust. Most low-rise buildings in Shanghai have been founded on shallow foundations for a long time because of the existence of this top crust; refer to (Tan et al. 2016). In light of this, Wang et al. (2017) underestimated

q_{ult}

of the subgrade below Stockpile 1, by disregarding the aforementioned load-spread effect and treating the top crust as part of the surcharge on the underlying soft strata. Because the traditional methods for calculating

q_{ult}

(Prandtl 1920; Hansen 1970; Vesic 1973; Meyerhoff and Hanna 1978) were unable to account for the load-spread effect of the top crust,

q_{ult}

was estimated by the method of Liu and Yu (2017) in this study. This method was developed for large-scale surcharge on thin stiff crust underlain by weak clayey strata like this site

q_{ult} = q_{b} (1 + \frac{H}{B} \tan θ) + \frac{H}{B} \cos θ (γ H K_{p} + 4 c_{c u} \sqrt{K_{p}}) \cdot (\tan φ_{c u} - \tan θ) + 2 c_{c u} \frac{H}{B} - γ H (1 + \frac{H}{B} \tan θ)

(4)

q_{b} = (π + 2) S_{u} + γ H

(5)

θ = \tan^{- 1} (\frac{σ_{mss} - q_{b} + (σ_{mss} + \frac{c_{c u}}{\tan φ_{c u}}) \sin^{2} φ_{c u}}{(σ_{mss} + \frac{c_{c u}}{\tan φ_{c u}}) \sin φ_{c u} \cos φ_{c u} + S_{u}})

(6)

σ_{mss} = (q_{b} + c_{c u} \cos φ_{c u} \sin φ_{c u}) + \sqrt{{(c_{c u} \cos φ_{c u} + q_{b} \sin φ_{c u})}^{2} - {(S_{u} \cos φ_{c u})}^{2}} / \cos^{2} φ_{c u}

(7)

where

H

= crust thickness;

B

= surcharge width;

θ

= load-spread angle;

γ

= unit weight of crust;

K_{p} = \tan^{2} [45 ° + (φ_{c u} / 2)]

, Rankine passive earth pressure coefficient of crust;

φ_{c u}

= friction angle of crust measured by CUDSST;

c_{c u}

= cohesion of crust measured by CUDSST; and

S_{u}

= undrained shear strength of the underlying soft clayey strata measured by the field vane shear tests. Detailed calculations of

q_{ult}

using the method of Liu and Yu (2017) are presented in “Calculation of Factor of Safety against Bearing-Capacity Failure for Both Flood Walls and Stockpiles” in the Supplemental Data. The soil parameters adopted in calculations are summarized in Table 1, which came from the geotechnical exploration report for design of this project (SXGEI 2006). Then, the factor of safety,

{F S}_{2}

, of subgrade against bearing-capacity failure can be estimated by

{F S}_{2} = q_{ult} / q

(8)

where

q

= surcharge load per unit area. The estimated

{F S}_{2} = 1.10 - 1.20

for the subgrade below Stockpile 1 and

{F S}_{2} = 2.20

for the subgrade below Stockpile 2. It indicated that failure of the flood wall to the north of Buildings 10–11 was not initiated by a bearing-capacity failure of the subgrade below Stockpile 2.

As discussed previously, a 4.6-m-deep excavation was unlikely to have affected the stability of Stockpile 1; more importantly, the buried building basement and the underlying pile groups located between the excavation and the stockpile would have enhanced global stability of Stockpile 1. From this perspective, Eqs. (4)–(8) underestimated

{F S}_{2}

for Stockpile 1. Regarding Stockpile 2 about 6 m north to the flood wall, Eqs. (4)–(8) only estimated its

{F S}_{2}

before the adjacent flood wall failed. Because sliding and overturning displacements of the adjacent flood wall rapidly reduced horizontal confining pressure of the subgrade below Stockpile 2 and then impaired the bearing capacity of the subgrade,

{F S}_{2}

for Stockpile 2 was not calculated after the adjacent flood wall failed. As shown in Fig. 12(a) (the flood wall was treated as shear reinforcement and its displacement was unable to be simulated in the LEM analysis), without northward sliding and overturning displacements of the flood wall, deep-seated slip failure would not occur for Stockpile 2. To verify this deduction, the safety factors of the flood walls against sliding and overturning were assessed in the following paragraphs.

Evaluation of FS for Flood Walls against Sliding and Overturning Failures

The safety factors of flood wall against sliding (

{F S}_{3}

) and overturning (

{F S}_{4}

) were estimated by Eqs. (9) and (10) as

{F S}_{3} = \frac{F_{τ} + F_{u} + F_{p}}{F_{a}}

(9)

{F S}_{4} = \frac{M_{W} + M_{u} + M_{p}}{M_{a}}

(10)

where

F_{τ}

= shear resistance on the bottom of the flood wall footing;

F_{u}

= thrust of the water pressure above the riverbed against the front side of the flood wall;

F_{p}

= thrust of lateral passive earth pressure below riverbed against the front side of the flood wall;

F_{a}

= thrust of lateral active earth pressure against the back side of the flood wall;

M_{W}

= resisting moment about flood wall footing due to the weights of both the flood wall and soil mass above flood wall footing;

M_{u}

= resisting moment about the flood wall footing due to

F_{u}

;

M_{p}

= resisting moment about the flood wall footing due to

F_{p}

;

M_{a}

= driving moment about the flood wall footing due to

F_{a}

. Detailed calculations of

{F S}_{3}

and

{F S}_{4}

are presented in “Calculation of Factors of Safety against Sliding and Overturning Failures of Flood Wall” in the Supplemental Data.

For the flood wall behind Buildings 10–11, the estimated

{F S}_{3} = 1.23

and

{F S}_{4} = 1.63

in case of no stockpile nearby and

{F S}_{3} = 0.80

and

{F S}_{4} = 0.94

with the proximity of 6-m-high soil in Stockpile 2. This indicates that the flood wall would undergo both sliding and overturning failures due to Stockpile 2. The calculation results (

{F S}_{2} = 2.20

;

{F S}_{3} = 0.80

;

{F S}_{4} = 0.94

) imply that sliding and overturning failures of the flood wall took place first and the corresponding ground movement undermined (softened) the subgrade below Stockpile 2 as a result of rapid reduction in the horizontal confining pressure of the subgrade. Then, the top thin crust was punched through by Stockpile 2 owing to rapid strength degradation, evidenced by no obvious ground heave outside the loaded (stockpile) area [Fig. 3(d)]. Subsequently, the underlying soft strata experienced an undrained general shear failure, evidenced by significant riverbed heave and flood wall tilting [Figs. 3(a–c)]. This failure model, a typical punching failure of the top thin crust followed by a general shear failure of the underlying soft strata, is consistent with the case studies involving surcharge on thin crust overlying soft clay as described by Wei et al. (2012) and Liu and Yu (2017). As schematically illustrated in Fig. 14, the riverbed heave zone due to the general shear failure of the soft clayey strata underneath Stockpile 2 conforms to the failure zones associated with a general bearing-capacity failure as described by Terzaghi (1943).

Fig. 14. Graphical estimation of bearing-capacity failure of the subgrade below Stockpile 2.

The calculated

{F S}_{3} = 0.80

and

{F S}_{4} = 0.94

for the flood wall adjacent to Stockpile 2 hinted that the flood wall would have already approached sliding and overturning failures at

FS \leq 1.0

before Stockpile 2 was dumped to a height of 6 m on June 25, 2009. However, the flood wall did not fail until noon on June 26, 2009. The delayed sliding and overturning failures of the flood wall implied that the development of stockpiling-induced additional lateral active earth pressures against the flood wall was not immediate. The occurrence of shear failure of the subgrade below Stockpile 2 concurrent with sliding and overturning failures of the flood wall disclosed that undrained shear failure in the subgrade was immediate upon loading.

Regarding the flood wall behind Buildings 6–7, the estimated

{F S}_{3} = 1.46 - 1.51

and

{F S}_{4} = 1.80 - 2.17

without stockpile nearby,

{F S}_{3} = 1.05 - 1.07

and

{F S}_{4} = 1.08 - 1.21

with the 4-m-high stockpile (Phase 1) nearby, and

{F S}_{3} = 0.90

and

{F S}_{4} = 0.92 - 1.06

after the stockpile was dumped to 10-m height (Phase 2). These results indicated both sliding and overturning failures of the flood wall would have occurred once Stockpile 1 was dumped to 10-m height (Phase 2), whereas the flood wall remained intact in the field. It was postulated that because Phase 1 of Stockpile 1 was formed approximately six months before the accident, consolidation of the underlying soft strata under the weight of Stockpile 1 (Phase 1) for six months should have enhanced soil shear strength,

τ_{f}

. Therefore, the degree of consolidation,

U_{t}

, of the subgrade due to preloading by the Phase 1 of Stockpile 1 was estimated using the one-dimensional consolidation theory of Terzaghi (1943), and the associated increment,

Δ τ_{f}

, in

τ_{f}

due to consolidation was estimated by the method specified in Chinese Building Foundation Code—JGJ79 (Chinese Standard 2002)

U_{t} = 1 - \frac{8}{π^{2}} \exp (- \frac{π^{2}}{4} T_{v})

(11)

T_{v} = \frac{c_{v} t}{H_{1}^{2}}

(12)

c_{v} = \frac{k \cdot (1 + e_{0})}{γ_{ω} a_{v}} = \frac{k \cdot E_{s}}{γ_{ω}}

(13)

Δ_{τ_{f}} = q \cdot U_{t} \cdot \tan φ

(14)

where

γ_{ω}

= unit weight of water;

a_{v}

= compressibility coefficient of subgrade;

e_{0}

= initial void ratio of the subgrade;

E_{s}

= constrained modulus of subgrade;

k

= permeability coefficient of subgrade;

q = 47.5 kPa

, equivalent surcharge of Phase 1 of Stockpile 1 (refer to “Calculation of Factor of Safety against Bearing-Capacity Failure for Both Flood Walls and Stockpiles” in the Supplemental Data); and

φ

= effective friction angle measured by consolidated undrained triaxial tests. Because the failures of this site were inherently due to short-term undrained loading (stockpiling),

φ_{c u}

instead of

φ

was used in Eq. (14) for estimation of

Δ_{τ_{f}}

. If

φ

was used for estimation, a greater

Δ_{τ_{f}}

would be obtained because

φ > φ_{c u}

. Detailed calculations are described in “Evaluation of Preloading Effect on the Soft Subgrade by the 4-m-High Stockpile” in the Supplemental Data. The estimated

U_{t} = 0.38

and

Δ_{τ_{f}} = 6.76 kPa

.

U_{t} = 0.38

under

q = 47.5 kPa

for six months at this site matched reasonably with observed preloading at a project near this site (Yang 2010), whose measured

U_{t} = 0.42

under

q = 60 kPa

for six months. If

Δ_{τ_{f}}

was accounted for in analysis, the estimated

{F S}_{2} = 1.40 - 1.50

,

{F S}_{3} = 1.12 - 1.18

, and

{F S}_{4} = 1.19 - 1.23

for the flood wall when the 10-m-high stockpile was dumped. Evidently, the preloading effect on the soft subgrade, resulting from Phase 1 of Stockpile 1, unexpectedly saved the flood wall when the height of Stockpile 1 increased to 10 m six months later. The calculated magnitudes of

{F S}_{1}

,

{F S}_{2}

,

{F S}_{3}

, and

{F S}_{4}

are summarized in Table 2.

Table 2. Summary of the calculated

{F S}_{1}

,

{F S}_{2}

,

{F S}_{3}

, and

{F S}_{4}

for this project

Scenarios for the site of Stockpile 1	${F S}_{1}$	${F S}_{2}$	${F S}_{3}$	${F S}_{4}$
1. No stockpile	—	—	1.46–1.51	1.80–2.17
2. After stockpiling to 4 m high (Phase 1)	—	3.50	1.05–1.07	1.08–1.21
3. After stockpiling to 10 m high (Phase 2) and disregarding preloading effect of Phase 1	2.61–2.66	1.10–1.20	0.90	0.92–1.06
4. After stockpiling to 10 m high (Phase 2) and taking into account the preloading effect of Phase 1	—	1.40–1.50	1.12–1.18	1.19–1.23
Scenarios for the site of Stockpile 2	${F S}_{1}$	${F S}_{2}$	${F S}_{3}$	${F S}_{4}$
1. No stockpile	—	—	1.23	1.63
2. After stockpiling to 6 m high	2.32–2.50	2.20	0.80	0.94

In addition to the previous calculations, stability analyses using Slope/W (2012) were carried out to verify the preloading effect on enhancing the global stability of the flood wall near Stockpile 1. If the preloading effect was considered in analysis,

{FOS}_{\min}

can be increased from 1.376 [Fig. 12(b)] to 1.425 [Fig. 15(a)]. Even though the rainfall effect was considered, its

{FOS}_{\min} = 1.039

was still above 1.0 [Figs. 15(b) versus 12(c)]. Once again, these LEM analysis results verified that the preloading effects from Phase 1 of Stockpile 1 saved the flood wall adjacent to Stockpile 1.

Fig. 15. Global stability analysis results: (a) flood wall behind Stockpile 1 considering the preloading effect; (b) flood wall behind Stockpile 1 considering both preloading effect and rainfall; and (c) the site of Stockpile 1 after the rainfall in case of an embedment depth of 6.6 m for the basement of Building 7.

Buildings Founded on Piles

Examination on Pile Capacities

The buildings were founded on 33-m-long PHC piles and each pile was designed with an allowable axial compression capacity of 1,300 kN with a safety factor of 2.0 (an ultimate axial compression bearing capacity,

Q_{u k}

, of 2,600 kN); see Wang et al. (2017). To examine the pile capacities, extensive calculations were carried out to evaluate pile cracking moment (

M_{r}

), ultimate bending moment (

M_{u}

), ultimate shear resistance (

Q

), ultimate compression (

R_{p}

) and tension (

T_{Q}

) strengths, and ultimate axial compression (

Q_{u k}

) and tension (

T_{u k}

) bearing capacities, according to Chinese Technical Code for Prestressed Concrete Pipe Piles—DGJ32/TJ109 (Jiangsu Department of Housing and Rural Development 2010). The structural ultimate capacities of PHC piles in terms of

M_{r}

,

M_{u}

,

Q

,

R_{p}

, and

T_{Q}

were calculated as follows:

M_{r} = \frac{I_{p}}{r_{o}} (σ_{p c} + σ_{cbt}) = 76.1 kN \cdot m

(15)

σ_{p c} = 0.6 n_{s} A_{a} \cdot \frac{F_{ptk}}{A_{j}} = 6.1 kPa

(16)

M_{u} = α \cdot M_{r} = 125.6 kN \cdot m

(17)

Q = \frac{2 t \cdot I_{p}}{S_{0}} \cdot \frac{1}{2} \sqrt{{(σ_{p c} + 2 \cdot Ø \cdot σ_{t})}^{2} - σ_{p c}^{2}} = 182.9 kN

(18)

R_{p} = 0.3 (f_{c e} - σ_{p c}) A_{j} = 1,783.7 kN

(19)

T_{Q} = σ_{p c} \cdot A_{c} = σ_{p c} \cdot (A_{j} - A_{p}) = 484.7 kN

(20)

The geotechnical ultimate capacities of PHC piles in terms of

Q_{u k}

and

T_{u k}

were calculated as follows:

Q_{u k} = Q_{s k} + Q_{p k} = u \sum q_{sik} l_{i} + q_{p k} (A_{j} + λ_{p} A_{p 1}) = 2,056.6 kN

(21)

T_{u k} = {\begin{cases} \sum λ_{i} q_{sik} u l_{i} \\ \frac{1}{n} u_{l} \sum λ_{i} q_{sik} l_{i} \end{cases} = {\begin{cases} 1,068.8 kN (local failure of pile group) \\ 821.4 kN (general failure of pile group) \end{cases}

(22)

where

I_{p}

= moment of inertia of pile cross section;

r_{o}

, outer radius of pipe pile;

σ_{p c}

= effective prestress of concrete pile;

σ_{c b t}

= tension strength of concrete subjected to bending;

n_{s}

= number of rebar in pile;

A_{a}

= cross section area of one rebar;

F_{ptk}

= tensile strength of rebar;

A_{j} = π \times (r_{o}^{2} - r_{i}^{2})

, area of pile rim;

r_{i}

= inner radius of pile;

α

= ultimate coefficient proposed for PHC pile;

t

= wall thickness of pipe pile;

S_{0}

= statical moment of pile cross section;

Ø

= dimensionless coefficient equal to

d_{o}

;

d_{o}

= outer diameter of pipe pile in meter;

σ_{t}

= tensile strength of prestressed concrete;

f_{c e}

= compression strength of concrete;

A_{c}

= concrete area on pile cross section;

A_{p}

= rebar area on pile cross section;

Q_{s k}

= pile side resistance;

Q_{p k}

= pile tip resistance;

u = π d_{o}

is the pile perimeter;

q_{sik}

= unit exterior side friction measured by CPT;

l_{i}

= depth interval for each soil layer;

q_{p k}

= tip resistance measured by CPT at the elevation of pile toe;

A_{p 1} = π r_{1}^{2}

is the open area of pile cross section;

λ_{p} = 0.8

is the empirical coefficient accounting for soil-plugging effect of open-ended pipe pile at

h_{b} / d_{0} \geq 5

(

h_{b}

= pile embedment length in the bearing stratum—Layer

7_{1 - 2}

, equal to 2.2 m;

d_{0}

, pile diameter, equal to 0.4 m);

λ_{i}

= empirical pullout coefficient;

u_{l}

= exterior perimeter of pile group; and

n

= number of piles, equal to 118 in this case; refer to “Comprehensive Evaluations on Pile Capacities” in the Supplemental Data for the detailed calculations. The calculated

Q_{u k} = 2,056.6 kN

is somewhat smaller than

Q_{u k} = 2,600 kN

adopted in the design.

As specified in AASHTO (2014), for a group of piles closely spaced, its ultimate bearing capacity could be greater or smaller than the number of piles times the ultimate bearing capacity per pile because of complex interactions of group piles. However, for a group of piles far apart (pile spacing,

S_{c}

, is greater than three-pile diameter) or end-bearing in hard strata (e.g., rock and compact sand or gravel), its ultimate bearing capacity will be equal to the number of piles times the ultimate bearing capacity per pile; refer to the schematic illustrations in Fig. S24. As marked in Fig. 8(b),

S_{c}

was around 1.6–4.0 m at this site, which was 4–10 times the pile diameter (0.4 m). Therefore, the factor of safety,

{F S}_{5}

, against axial compression capacity failure for the pile can be reasonably estimated by

{F S}_{5} = Q_{u k} / Q_{d} = 2.88

(23)

where

Q_{d} = 714 kN

denotes the axial load carried by per pile for the 13-story building. The magnitude of

Q_{d}

was calculated by assuming that the building weight was evenly shared by the 118 PHC piles underneath the footing. The open-ended PHC piles at this site were designed to be tipped into the dense to very dense silty sand (Layer

7_{1 - 2}

) and installed by a traditional driving method. During pile installation, a fully plugged condition could be formed. As recognized in literature (Paikowsky et al. 1989; Paikowsky and Whitman 1990; Raines et al. 1992; Paik and Salgado 2003), a driven open-ended pipe pile in a fully plugged mode behaved like a closed-ended pile. By full-scale field load testing, McVay et al. (2004) and Tan and Lin (2013) found that open-ended pipe piles toed in dense sand still behaved like fully plugged piles even if the piles were partially plugged yet close to a fully coring mode during pile driving. Based on this consideration, if a fully plugged condition (

λ_{p} = 1.0

) was assumed in calculation,

Q_{u k} = 2,092 kN

and

{F S}_{5} = 2.92

can be obtained for the PHC piles.

The calculations above verify that the adopted PHC piles had sufficient axial compression bearing capacities. Moreover, as interpreted previously, the 4.6-m-deep excavation about 7 m to the south of the building barely affected the piles below the building, which can also be demonstrated by the 3D numerical simulation to be presented in a later section entitled “3D Numerical Analyses.” Thus, the shallow excavation to the south of Building 7 can be excluded as a major factor for the southward building-overturning failure.

Examination on Subgrade

As shown in Figs. 4(d–e), the residual lengths of broken piles connected to the building footing on the north side were longer than those on the south side, an apparent sign of pullout tension-fracture failure of the piles as a result of southward rotational failure of Building 7. The subgrade below Stockpile 1 exhibited a conventional general shear-failure mechanism, i.e., (1) the subsoils were pushed significantly outward on the south side, (2) the subgrade below the building footing adjacent to the stockpile heaved dramatically from 2 m below ground level before the accident to above ground level after the accident, (3) a scarp was visible on the stockpile, and (4) a longitudinal crack occurred on the stockpile [Fig. 4(d)]. According to the interview with on-site workers, there existed no sign of slip failure on the stockpile before June 27, 2009. The previous calculations and LEM analyses [“Calculation of Factor of Safety against Bearing-Capacity Failure for Both Flood Walls and Stockpiles” in the Supplemental Data and Fig. 9(b)] also indicated that the top crust overlying the soft strata would not undergo failure under the stockpile weight. As revealed by the LEM analysis result in Fig. 9(c), the 5-h intense rainfall before the building failure could have triggered the undrained shear failure of the subgrade underneath Stockpile 1 behind Building 7 and led to a deep-seated slip failure of the stockpile.

Considering that (1) the 4.6-m-deep excavation barely affected the building, (2) additional earth pressures generated by the stockpile 1 m to the north of Building 7 could only produce northward building tilting, and (3) the building sustained southeast wind (Fig. 10) before its failure (the southeast wind slightly increased resisting moment of Building 7 against southward tilting), the abrupt southward toppling failure of Building 7 should be attributed to the general shear failure of the subgrade below the 10-m-high stockpile. The subsoil below Building 7 was thrust up as the stockpile underwent a deep-seated slip failure. The rapid ground heave imposed violent impact force,

F_{i}

, against the building basement and the sliding soil mass imposed massive impact force,

F_{n}

, on the north side of the superstructure. Consequently, Building 7 rapidly tilted southward causing southward sway of its elevated gravity center. Southward sway of the building gravity center further increased the southward overturning moment of Building 7. As a result of tensile fracturing failure of its underlying piles [Figs. 4(e–f)], Building 7 fell southward to the ground within seconds. The associated failure mechanism is schematically illustrated in Fig. 16, in which a passive (ground heave) zone almost extending to the south edge of Building 7 can be estimated with the graphical solution of Terzaghi (1943). The conceptual failure surface in Fig. 16 was in reasonable agreement with that of Fig. 9(c) estimated by the LEM analysis. Overturning moment,

M_{d}

, about the south footing edge due to a general shear failure of the subgrade can be roughly estimated by

M_{d} = \sum F_{i} h + \sum F_{n} y

(24)

M_{d} > M_{r}

(25)

M_{r} = G \frac{w}{2} + \sum (T_{Q} + G_{p}) x + \sum f_{w} y + \sum σ_{p} y

(26)

where

F_{i}

= driving force on building basement resulting from a general shear failure of the subgrade below Stockpile 1;

F_{n}

= impact force against superstructure resulting from sliding of Stockpile 1;

h

= lever arm of

F_{i}

around the south edge of building at Point O;

M_{r}

= resisting moment around Point O due to weight of building

G

, buoyant weight of pile

G_{p}

, ultimate pile tensile strength

T_{Q}

, southeast wind load

f_{w}

, and lateral passive earth pressure

σ_{p}

against the south side of the building basement;

w

= building width;

x

= lever arms of

T_{Q}

and

G_{p}

around Point O; and

y

= lever arms of the thrusts of

F_{n}

,

f_{w}

, and

σ_{p}

around Point O. Because the piles underwent tensile fracturing failure near the pile heads rather than a pullout failure of the piles during overturning failure of Building 7 (Fig. 4),

T_{Q}

rather than

T_{u k}

was adopted in Eq. (26) for estimation of

M_{r}

. The estimated

M_{d} > M_{r} = 1,013 MN \cdot m

. Detailed calculations are presented in “Evaluation of Overturning Moment Leading to Southward Rotational Failure of Building 7” in the Supplemental Data.

Fig. 16. Schematic illustration of the mechanism of toppling failure of Building 7.

3D Numerical Analyses

The previous calculations and analyses indicated that (1) the abrupt toppling failure of Building 7 could have resulted from a rainfall-triggered general shear failure of the subgrade below Stockpile 1, and (2) because of a more gentle slope angle and smaller plan size, the subgrade below Stockpile 1 behind Building 6 did not fail after the intense rainfall and Building 6 survived. To verify these postulates, numerical simulations were carried out using the commercial FE software version 2014 Midas GTS NX (GTS NX 2014).

FE Model

If a 2D numerical model was built for analysis, the problem investigated had to be treated as a plane–strain model in simulation. Thus, each row of piles below the building footing would be modeled as a continuous plate in a 2D model; consequently, the capability of piles for restraining ground movements would be inappropriately exaggerated. Considering this, a 3D model about

207 m long \times 168 m wide \times 50 m high

was built for simulation, which included Buildings 6–7, Stockpile 1, and the 4.6-m-deep excavation; refer to Fig. 17. Since the flood wall to the north of Stockpile 1 remained intact, it was excluded in the numerical model to improve computational efficiency.

Fig. 17. 3D numerical model built for FE simulation: (a) 3D view; and (b) side view.

Constitutive Models and Parameters

Because of the complex stress–strain behaviors of soft clays (Duncan and Chang 1970; Koutsoftas and Ladd 1985; Whittle et al. 1994; Schanz et al. 1999; Finno and Kim 2012), the appropriateness of specific soil models and parameters adopted in software as well as the knowledge and experience of personnel using software are essential for yielding reliable simulation results; otherwise, software would be “a dangerous robot” (Lacasse and Fuleihan 2018). Elastic–plastic Mohr-Coulomb (MC) model allows for a fast and simple calculation, but it does not take into account typical characteristics of soils, e.g., stress dependency of stiffness moduli and irreversible strain due to primary isotropic compression. Hence, the MC model is too simplistic for the problem considered. A Cam-clay (CC) model was developed for primary compression of normally consolidated clay, but it was not suitable for an unloading problem (e.g., excavation and tunneling). Based on these considerations, an advanced constitutive model in the software similar to hardening soil model with small-strain stiffness (HSS model) was used for the simulation, which can account for stress dependency of stiffness moduli and increased stiffness of soils at small strains. Thereby, it is appropriate for both loading (stockpiling) and unloading (excavation) problems in soft strata.

Because (1) Shanghai clay is a kind of normally consolidated soil, and (2) both excavation and stockpiling in this study were short-term activities, CU parameters of the clayey strata measured by CUDSST were adopted in analysis (Table 1). Using CUDSST data to analyze undrained problems in Shanghai soft clay is also recommended by the local design specification [DG/TJ08-61 (Shanghai Urban and Rural Construction and Transportation Committee 2010)]. The structural elements of Buildings 6–7 and the earth retaining wall of the excavation were simulated by plate elements, the soil nails of the earth retaining wall were simulated by anchor elements, and the piles below the buildings were simulated by beam elements. To account for soil–structure interaction, interface elements were applied to the outer perimeter of piles and a spring element was applied to the tip of each pile. Detailed information about the material parameters used in the FE simulation is presented in “3D Numerical Simulations—Material Parameters and Typical Results” in the Supplemental Data.

Simulation Procedures

The simulated construction procedures duplicated the following on-site construction activities: (1) construction of Buildings 6–7, (2) formation of Phase 1 of Stockpile 1 (4-m high) behind Buildings 6–7, (3) consolidation of subgrade underneath the stockpile for six months, (4) excavation of the 4.6-m-deep underground garage 7 m to the south of Buildings 6–7 and dumping the excavated soils above the existing 4-m-high stockpile (Phase 1) to form a 10-m-high stockpile (Phase 2) about 1 m to the north of Buildings 6–7, and (5) 5-h intense rainfall just before the failure of Building 7.

Analysis of Simulation Results

Fig. 18 presents the simulated lateral displacements of the ground and structures and Fig. 19(a) shows the simulated lateral displacements of the building footings at the completion of excavation and stockpiling behind Buildings 6–7 before the heavy rainfall. Because of the uneven distribution of Stockpile 1 behind Building 6 (Fig. 1), the east section of Building 6 underwent greater lateral displacement than its west section. The upper stories of Building 6 had small lateral displacement northward to the stockpile while its lower stories had small lateral displacement southward to the excavation (northward inclination of Building 6). The simulated maximum southward lateral displacement at the footing of Building 6 was around 3.3–3.5 cm, which matched reasonably with the field-surveyed 2.9 cm on June 27, 2009. As for Building 7, the entire superstructure moved southward to the excavation side simultaneously, and its lower stories featured much greater southward lateral displacement than its upper stories (northward inclination of Building 7). Compared with Building 6, Building 7 had greater southward lateral displacement. Fig. 19(b) shows the simulated vertical displacements at the building footings, and Fig. 20 presents the simulated vertical displacements of the ground and structures at the completion of excavation and stockpiling behind Buildings 6–7. As shown in these figures, the north sides of Buildings 6–7 settled up to 12.6–12.9 mm while the south sides of Buildings 6–7 heaved slightly up to 0.87 mm, i.e., northward inclination of Buildings 6–7.

Fig. 18. Simulated lateral displacements of the ground and the structures at the completion of excavation and stockpiling: (a) 3D view of the entire site; and (b) 3D and side views of Building 7 and its underlying piles. (Negative displacement, southward to excavation; positive displacement, northward to stockpile.)

Fig. 19. Simulated displacements of the building footings at the completion of excavation and stockpiling: (a) lateral footing displacement; and (b) vertical footing displacement.

Fig. 20. Simulated vertical displacements of the ground and the structures at the completion of excavation and stockpiling: (a) 3D view of the entire site; and (b) Buildings 6 and 7.

The numerical simulation results in Figs. 18–20 clearly indicated a northward rather than southward inclination of Buildings 6–7 at the completion of excavation and stockpiling, which matched with the previous postulate schematically illustrated in Fig. 7. The simulated northward inclinations were about 0.091% for Building 6 and 0.10% for Building 7, both of which were much less than the specified 0.2% by Chinese Technical Code for Highrise Concrete Superstructure—JGJ3 (Chinese Standard 2010). Moreover, both the simulated maximum bending moment (

M = 56.4 kN \cdot m

) and shear force (

Q = 49.5 kN

) in the piles below Building 7 (“3D Numerical Simulations—Material Parameters and Typical Results” in the Supplemental Data) were much smaller than their ultimate magnitudes (

M_{u} = 125.57 kN \cdot m

and

Q = 182.85 kN

in “Comprehensive Evaluations on Pile Capacities” in the Supplemental Data). There did not exist a possibility of shearing or bending failure of the piles at the completion of excavation and stockpiling. These simulation results were consistent with the following two facts: (1) no obvious sign of inclination was reported for either Building 7 or Building 6 before the abrupt toppling failure of Building 7, and (2) the post-failure examination on the piles left in place showed that the piles along the south footing edge of building 7 had undergone eccentric compression failure and the other piles had undergone tensile fracturing failures (Wang et al. 2017); no physical evidence indicated shearing or bending failure of the piles.

Figs. 21(a and b) show side views of the simulated total displacements and shear strains in the ground at the completion of excavation and stockpiling behind Building 7. Consistent with the field observation before the building failure, the stockpile was in a stable state. Both the simulated total displacements and shear strains in the ground reduced significantly in a plunging manner at the border of the pile groups and the subgrade below the stockpile. These simulation results verified the previously mentioned barrier effect of the building basement and its underlying piles.

Fig. 21. Side views of the simulated barrier effect of the buried basement and its underlying piles at the completion of excavation and stockpiling: (a) total displacements; and (b) shear strains in ground. (Structural elements not shown.)

To verify the preceding postulate that the 5-h intense rainfall could have played a key role in the overturning failure of Building 7, the strength parameters of the top crust were reduced by 50% as previously introduced in the section entitled “Preliminary Global Stability Assessment” and the saturated unit weights of the stockpile and the crust were adopted in the 3D numerical modeling. Because the FE method is based on continuum mechanics and has to satisfy deformation compatibility at nodes of adjacent elements, it is hard to simulate a large deformation problem. As a result of deformation incompatibility, the numerical simulation accounting for the rainfall automatically terminated at an early calculation stage. Thus, simulation of a general shear failure of the subgrade below the stockpile like that depicted in Fig. 16 could not be achieved. In spite of this, as shown in Fig. 22, Stockpile 1 behind Building 7 underwent a base failure with significant ground heave extending below the footing of Building 7. The simulated stockpile behind Building 6 was still in a safe condition, evidenced by no substantial ground heave below Building 6. This analysis demonstrated that the 5-h intense rainfall before the building failure was a decisive factor, which rapidly triggered an undrained shear failure of the subgrade below Stockpile 1 and led to a deep-seated slip failure of the stockpile and abrupt overturning failure of Building 7.

Fig. 22. FE simulation analysis results accounting for the intense rainfall: (a) side view of the FE simulated total displacement field; and (b) top view of the FE simulated vertical displacement field. (Positive magnitude, heave; negative magnitude, settlement.)

Discussion

Construction Sequence

As pointed out by Terzaghi (1958), Peck (1973), Lambe et al. (1981), Sowers (1993), and Marr (2013), failures of constructed facilities could arise from either technical or nontechnical causes of shortcomings originating in the attitudes and actions of owner, designer, constructor, and technical consultant. The preceding analyses disclosed that stockpiling soils between the buildings and the flood walls and the 5-h intense rainfall were two decisive factors contributing to the accidents. Additionally, some design and construction deficiencies existed in this project. As shown in Fig. 1, the development included an underground garage in the initial design. According to Chinese Building Foundation Code—GB50007 (Ministry of Construction of China 2002), excavation of an underground garage should be completed before construction of its adjacent superstructure. However, excavation of the garage was executed after completion of the superstructures in 2008. If the construction of the underground garage and the superstructures had strictly followed the design specification, overturning failure of building 7 would have been avoided.

Embedment Depth of Building Footing

Moreover, in accordance with Chinese Design Code for Highrise Building [JGJ3-2010 (Chinese Standard 2010)], embedment depths of building footings should be no less than

(1 / 18) H_{b} = 2.1 m

(

H_{b}

is superstructure height, equal to 38 m at this site) in case of shallow foundation and

(1 / 15) H_{b} = 2.6 m

in case of the pile foundation. The buildings of this site had an inadequate embedment depth of 2.0 m. Moreover, GB50007 (Ministry of Construction of China 2002) specifies that if a building has a podium or an annex (e.g., underground garage), its footing should have an embedment depth no less than 2 m below the footing of the podium or annex. According to this specification, building footings should have a minimum embedment depth of 6.6 m instead of 2.0 m at this site. In case the basement had an embedment depth of 6.6 m (

H_{e} = 6.6 m

) and meanwhile the underground garage was completed before excavation of the building basement, the maximum influence zone of the basement excavation (

2 H_{e} = 13.4 m

) could reach the completed garage at 7 m distance from the building. In such case, the adverse influence of the basement excavation on the completed garage could be effectively mitigated by installation of cost-effective barrier piles (e.g., recyclable steel sheet-pile or H-pile wall) or mixed-in-place piles between them as introduced in Tan and Li (2011) and Tan et al. (2019).

If building footings had an embedment depth of 6.6-m BGS, then using the graphical approach of Terzaghi (1943), overturning-failure possibility of Building 7 due to a general shear failure of the adjacent stockpile subgrade would be mitigated significantly (Fig. S33) because (1) the 6.6-m-thick soils against the south side of the building basement would provide much greater

σ_{p}

against southward tilting of the building, (2) the deeply buried building basement would block extension of southward subsoil flow and hence much smaller

F_{i}

would be generated once the adjacent stockpile subgrade suffered a general shear failure, and (3) more side resistance in restraining upward movement of the building basement would be provided by its surrounding strata. To verify this postulate, global stability analyses using Slope/W (2012) were carried out for a case with a basement embedment depth of 6.6 m, in which the rainfall effect was considered. As shown in Fig. 15(c), because of the blocking effect of the 6.6-m-deep basement, only a deep toe slip failure would take place for the stockpile behind Building 7. Although the sliding soil mass due to a toe slip failure of the stockpile might have caused Building 7 to tilt southward, the risk of an abrupt overturning failure of Building 7 resulting from a general shear failure of the subgrade below Stockpile 1 (deep-seated slip failure of Stockpile 1) would have been mitigated significantly.

Pile Types

Because of their cost-effectiveness and relatively high axial compression bearing capacity, PHC pipe piles have been widely adopted in Shanghai to support superstructures since the 1990s (Tan and Lan 2012). As shown in Fig. 4(f), steel reinforcement was not observed at the exposed cross section of some broken piles, although manually cutting off the residual piles buried in the ground verified the presence of the designed steel reinforcements (Wang et al. 2017). Steel reinforcements retracted into concrete owing to loss of prestress, which resulted from the breakage of piles. The prestressed rebar inside the piles had high strength and low ductility, featuring a stress–strain curve without a well-defined yield point or plateau (Fig. S34). If nonprestressed concrete piles reinforced with regular mild steel (e.g., Tan et al. 2015; Lu et al. 2019) instead of PHC pipe piles reinforced with prestressed steel had been adopted for the building, the steel rebar might have not undergone brittle tensile failure as observed. Conversely, it might have taken time to undergo yielding, strain hardening, necking, and ultimate failure. During this period, the building might have tilted progressively rather than abruptly fallen to the ground with little warning. Then, appropriate countermeasures or remedial measures could have been adopted to mitigate the tilting.

Temporary Soil Stockpiling

Apart from the previously discussed technical and nontechnical causes, one lesson learned from this failure case is that the potential adverse impact of the stockpiles on the preexisting buildings and flood wall had never been considered in the design. If a plan for the stockpiles had been reviewed/analyzed by project geotechnical engineers as part of design services during construction, the failures of the building and the flood wall should have been preventable with geotechnical input into planning of the earthwork staging and stockpile heights and locations on the site. For a site like Shanghai featuring a top crust underlain by thick soft clayey deposits, softening of the upper crust in intense rainfall events should be accounted for in routine geotechnical design services, e.g., plans for temporary stockpiling of soils on site.

Summary and Conclusion

Until now, the abrupt overturning failure of the 13-story high-rise building in 2009 in Shanghai, China was attributed to the following: (1) lateral ground movement below the building arising from unbalanced lateral earth pressures on the two sides of the building shearing off its underlying piles, or (2) first settlement/failure of the piles underneath the south building side caused progressive tilting failure of the building. However, the comprehensive investigations presented in this study disclosed a radically different failure mechanism. Moreover, this study has explored the following important phenomena, which had received no attention from previous investigations: (1) what role the 5-h intense rainfall just before overturning failure of Building 7 played in this accident; (2) why Building 6 did not fall to the ground like Building 7 did, although this building was subjected to the similar conditions; (3) why Building 7, founded on 33-m-long piles to the south of Stockpile 1, underwent failure but the flood wall founded on shallow foundation to the north of Stockpile 1 remained motionless; and (4) why the flood wall adjacent to 6-m-high Stockpile 2 experienced a catastrophic failure but the flood wall adjacent to 10-m-high Stockpile 1 remained intact.

Because of its shallow depth (4.6 m), the excavation of the underground garage about 7 m to the south of Building 7 had little effect on the performance of the building. All the buildings were founded on 33-m-long PHC pipe piles with

{F S}_{5} = 2.88

. In addition, the stockpile induced much greater downdrag forces and settlements for the piles on the north building side than for the piles on the south side. Thus, there was no possibility that settlement or failure of the piles on the south building side progressively incurred a tilting failure of the building. The piles closer to the stockpile sustained much larger lateral earth pressures and hence the piles on the north building side would deflect or fail in advance of the piles on the south side. If the lateral ground movement below the building could have sheared off the piles and caused building-tilting failure, the building would have fallen northward to the stockpile rather than southward to the excavation. The abrupt southward overturning failure of Building 7 most likely resulted from an undrained general shear failure of the subgrade below the 10-m stockpile followed by a deep-seated slip failure of the stockpile about 1 m to the north of Building 7. Rapid ground heave against the building basement within the passive zone of bearing-capacity failure for the stockpile subgrade and sliding of the soil mass instantaneously produced a giant overturning moment around the south building edge, which outweighed the resisting moments from the weights of the building and its underlying piles, tensile forces from the piles, resistance from the southeast wind, and the lateral passive earth pressure against the south side of the building basement. As a result, the building fell southward to the ground without warning. The abrupt undrained shear failure of the stockpile subgrade was triggered by intense infiltration of the 5-h heavy rainfall into the desiccated top crust, which degraded soil strengths significantly. Benefiting from a gentler slope angle and smaller plan size, the subgrade of Stockpile 1 behind Building 6 did not approach bearing-capacity failure and Building 6 survived.

Although the subgrade below Stockpile 2 had

{F S}_{2} = 2.2

against a bearing-capacity failure, the stockpile-induced lateral earth pressures on the adjacent flood wall resulted in sliding and overturning failures of the flood wall. The sliding and overturning displacements of the flood wall caused rapid reduction in the horizontal confining pressure of the subgrade below Stockpile 2 and the bearing capacity of the subgrade was impaired immediately. Then, Stockpile 2 punched into the top crust and the underlying soft strata underwent an undrained general shear failure. As a result, the flood wall displaced significantly northward into the river with the wall head tilting northward and the riverbed in front of the displaced flood wall heaving remarkably. The placement of the first 3–4-m-high stockpile about six months before the accident unexpectedly saved the flood wall to the north of Stockpile 1. Consolidation of the soft subgrade under the weight of Stockpile 1 (Phase 1) for six months enhanced the soil shear strength substantially, and then the capabilities of the subgrade against an undrained shear failure and the flood wall against sliding and overturning failures were increased. Thus, the flood wall adjacent to Stockpile 1 still remained intact and did not show a sign of displacing after Stockpile 1 was dumped to a height of 10 m (Phase 2).

Undoubtedly, stockpiling the excavated soils between the buildings and the flood wall was the primary factor contributing to the accidents. There was a chance to prevent rotational failure of Building 7, if Stockpile 1 could have been timely flattened after failure of the flood wall adjacent to Stockpile 2 one day before. Unfortunately, only Stockpile 2 had been timely flattened and no action was adopted for Stockpile 1. The 5-h intense rainfall just before the building failure rapidly deteriorated the situation and thus overturning failure took place for Building 7 as a consequence of an abrupt general shear failure of the subgrade below Stockpile 1 in close proximity. Apart from these, the inappropriate construction procedure (excavation of the underground garage was executed after the completion of the buildings) and the insufficient embedment depth (2.0 m) of the building basements (it should be no less than 6.6 m according to the design code) further increased the risk of building-tilting failure. If nonprestressed concrete piles rather than PHC pipe piles had been adopted for the buildings, the piles might have not undergone an abrupt tensile fracturing failure with rapid rotational failure of Building 7.

Supplemental Data

Figs. S1–S34 and Tables S1–S30 are available online in the ASCE Library (www.ascelibrary.org).

Supplemental Materials

File (supplemental_data_gt.1943-5606.0002264_tan.pdf)

Download
1.75 MB

Acknowledgments

Financial support from the National Natural Science Foundation of China (Grant No. 41877286) and the National Basic Research Program (973 Program) (Grant No. 2015CB057800) are gratefully acknowledged. The comprehensive, insightful, and valuable comments and suggestions from the two kind and knowledgeable anonymous reviewers, and the Associate Editor, are sincerely appreciated, because they enhanced the presentation of this paper remarkably.

References

AASHTO. 2014. AASHTO LRFD bridge design specifications. 7th ed. Washington, DC: AASHTO.

Google Scholar

Bishop, A. W. 1955. “The use of the slip circle in the stability analysis of slopes.” Géotechnique 5 (1): 7–17. https://doi.org/10.1680/geot.1955.5.1.7.

Google Scholar

Bonaparte, R. 2018. “Geotechnical stability of waste fills—lessons learned and continuing challenges.” In Proc., 54th ASCE Karl Terzaghi Lecture, IFCEE 2018. Reston, VA: ASCE.

Google Scholar

Burd, H. J., and S. Frydman. 1997. “Bearing capacity of plane-strain footings on layered soils.” Can. Geotech. J. 34 (2): 241–253. https://doi.org/10.1139/t96-106.

Crossref

Google Scholar

Chai, J., S. Shen, W. Ding, H. Zhu, and J. Carter. 2014. “Numerical investigation of the failure of a building in Shanghai, China.” Comput. Geotech. 55 (Jan): 482–493. https://doi.org/10.1016/j.compgeo.2013.10.001.

Google Scholar

Chinese Standard. 2002. Technical code for ground treatment of buildings. [In Chinese.] JGJ79-2002. Beijing: China Academy of Building Research.

Google Scholar

Chinese Standard. 2010. Technical code for highrise concrete superstructure. [In Chinese.] JGJ3-2010. Beijing: Ministry of Construction of China.

Google Scholar

Clough, G. W., and T. D. O’Rourke. 1990. “Construction induced movements of in-situ walls.” In Design and performance of earth retaining structures, Geotechnical Special Publication 25, 439–470. Reston, VA: ASCE.

Google Scholar

Collins, B. D., and D. Znidarcic. 2004. “Stability analyses of rainfall induced landslides.” J. Geotech. Geoenviron. Eng. 130 (4): 362–372. https://doi.org/10.1061/(ASCE)1090-0241(2004)130:4(362).

Google Scholar

Das, B. M., and K. Sobhan. 2017. Principals of geotechnical engineering. 9th ed. Boston: Cengage Learning.

Google Scholar

Duncan, J. M. 1996. “State of the art: Limit equilibrium and finite-element analysis of slopes.” J. Geotech. Eng. 122 (7): 577–596. https://doi.org/10.1061/(ASCE)0733-9410(1996)122:7(577).

Google Scholar

Duncan, J. M., T. L. Brandon, S. G. Wright, and N. Vroman. 2008. “Stability of I-walls in New Orleans during Hurricane Katrina.” J. Geotech. Geoenviron. Eng. 134 (5): 681–691. https://doi.org/10.1061/(ASCE)1090-0241(2008)134:5(681).

Google Scholar

Duncan, J. M., and C. Y. Chang. 1970. “Nonlinear analysis of stress and strain in soil.” J. Soil Mech. Found. Div. 96 (5): 1629–1653.

Crossref

Google Scholar

Fellenius, B. H. 1972. “Buckling of piles due to lateral soil movements.” In Vol. 2 of Proc., 5th European Conf. on Soil Mechanics and Foundation Engineering, 282–284. Madrid, Spain.

Google Scholar

Fellenius, W. 1936. “Calculation of the stability of earth dams.” In Vol. 4 of Proc., 2nd Congress of Large Dams, 445–463. Washington, DC: International Commission on Large Dams.

Google Scholar

Finno, R. J., and T. Kim. 2012. “Effects of stress path rotation angle on small strain responses.” J. Geotech. Geoenviron. Eng. 138 (4): 526–534. https://doi.org/10.1061/(ASCE)GT.1943-5606.0000612.

Google Scholar

Finno, R. J., S. A. Lawrence, N. F. Allawh, and I. S. Harahop. 1991. “Analysis of performance of pile groups adjacent to deep excavation.” J. Geotech. Eng. 117 (6): 934–955. https://doi.org/10.1061/(ASCE)0733-9410(1991)117:6(934).

Google Scholar

Gamez, J. A., and T. D. Stark. 2014. “Fully softened shear strength at low stresses for levee and embankment design.” J. Geotech. Geoenviron. Eng. 140 (9): 06014010. https://doi.org/10.1061/(ASCE)GT.1943-5606.0001151.

Google Scholar

GTS NX (Geo-Technical Analysis System New eXperience). 2014. Midas GTS NX user manual. Republic of Korea: GTS NX.

Google Scholar

Han, J., J. F. Chen, Z. S. Hong, and S. L. Shen. 2010. “Mitigation of levee failure using deep mixed columns and geosynthetics.” Geomech. Geoeng. Int. J. 5 (1): 49–55. https://doi.org/10.1080/17486020903452733.

Google Scholar

Hansen, J. B. 1970. A revised and extended formula for bearing capacity. Copenhagen, Denmark: Danish Geotechnical Institute.

Google Scholar

Houston, S. L. 2019. “It is time to use unsaturated soil mechanics in routine geotechnical engineering practice.” J. Geotech. Geoenviron. Eng. 145 (5): 02519001. https://doi.org/10.1061/(ASCE)GT.1943-5606.0002044.

Google Scholar

Hsieh, P. G., and C. Y. Ou. 1998. “Shape of ground surface settlement profiles caused by excavation.” Can. Geotech. J. 35 (6): 1004–1017. https://doi.org/10.1139/t98-056.

Google Scholar

Hu, Z. F., and Y. R. Fu. 2001. “Laboratory study on the shear strengths of Shanghai clay with different water contents.” [In Chinese.] Shanghai Geol. 77 (1): 38–42.

Google Scholar

Janbu, N. 1954. “Applications of composite slip surfaces for stability analysis.” In Vol. 3 of Proc., European Conf. on the Stability of Earth Slopes, 39–43. London: International Society of Soil Mechanics and Geotechnical Engineering.

Google Scholar

Jiangsu Department of Housing and Rural Development. 2010. Technical specification for prestressed concrete pipe pile foundation. [In Chinese.] DGJ32/TJ109. Nanjing, China: Jiangsu Dept. of Housing and Rural Development.

Google Scholar

Khudeira, S. 2010. “Building collapse during construction.” Pract. Period. Struct. Des. Constr. 15 (2): 99–100. https://doi.org/10.1061/(ASCE)SC.1943-5576.0000058.

Google Scholar

Korff, M., R. J. Mair, and F. A. F. Van Tol. 2016. “Pile-soil interaction and settlement effects induced by deep excavations.” J. Geotech. Geoenviron. Eng. 142 (8): 04016034. https://doi.org/10.1061/(ASCE)GT.1943-5606.0001434.

Google Scholar

Kourkoulis, R., F. Gelagoti, I. Anastasopoulos, and G. Gazetas. 2011. “Slope stabilizing piles and pile-groups: Parametric study and design insights.” J. Geotech. Geoenviron. Eng. 137 (7): 663–677. https://doi.org/10.1061/(ASCE)GT.1943-5606.0000479.

Google Scholar

Koutsoftas, D. C., and C. C. Ladd. 1985. “Design strengths for an offshore clay.” J. Geotech. Eng. 111 (3): 337–355. https://doi.org/10.1061/(ASCE)0733-9410(1985)111:3(337).

Google Scholar

Lacasse, S., and N. F. Fuleihan. 2018. “Is understanding the stress history and behavior of soft soils at risk of taking a back seat?” Geo-Strata 22 (2): 14–22.

Google Scholar

Lambe, T. W., W. A. Marr, and F. Silva. 1981. “Safety of a constructed facility: Geotechnical aspects.” J. Geotech. Eng. Div. 107 (3): 339–352.

Google Scholar

Lambe, T. W., and R. V. Whitman. 1969. Soil mechanics. New York: Wiley.

Google Scholar

Leshchinsky, B., and S. Ambauen. 2015. “Limit equilibrium and limit analysis: Comparison of benchmark slope stability problems.” J. Geotech. Geoenviron. Eng. 141 (10): 04015043. https://doi.org/10.1061/(ASCE)GT.1943-5606.0001347.

Google Scholar

Liu, Q. S., and J. Yu. 2017. “Calculation of ultimate bearing capacity of the man-made crust over dredged waste dump sites.” [In Chinese with English Abstract.] China Harbour Eng. 37 (9): 32–37.

Google Scholar

Lu, Y., Y. Tan, and H. Lan. 2019. “Full-scale load testing of 75-90-m-long post-grouted drilled shafts in Suzhou stiff clay.” J. Test. Eval. 47 (1): 284–309. https://doi.org/10.1520/JTE20170442.

Google Scholar

Marr, W. A. 2013. “Instrumentation and monitoring of slope stability.” In Proc., Geo-Congress 2013: Stability and Performance of Slopes and Embankments III, GSP 231, 2231–2252. Reston, VA: ASCE.

Google Scholar

McVay, M. C., D. Bardri, and Z. Hu. 2004. Determination of axial pile capacity of prestressed concrete cylinder piles. Tallahassee, FL: Florida Dept. of Transportation.

Google Scholar

Meyerhoff, G. G., and A. M. Hanna. 1978. “Ultimate bearing capacity of foundations on layered soils under inclined load.” Can. Geotech. J. 15 (4): 565–572. https://doi.org/10.1139/t78-060.

Google Scholar

Michalowski, R. L. 1992. “Bearing capacity of nonhomogeneous cohesive soils under embankments.” J. Geotech. Eng. 118 (7): 1098–1118. https://doi.org/10.1061/(ASCE)0733-9410(1992)118:7(1098).

Google Scholar

Michalowski, R. L. 2004. “Limit loads on reinforced foundation soils.” J. Geotech. Geoenviron. Eng. 130 (4): 381–390. https://doi.org/10.1061/(ASCE)1090-0241(2004)130:4(381).

Google Scholar

Ministry of Construction of China. 2002. Code for design of building foundation. [In Chinese.] GB50007. Beijing: Ministry of Construction of China.

Google Scholar

Moffitt, P. H., and B. A. Shelly. 2015. “The leaning towers of Wilmington, Delaware: Emergency response and retrofit of the I-495 viaduct to address lateral squeeze.” Geo-Strata 19 (5): 50–55.

Google Scholar

Morgenstern, N. R., and V. E. Price. 1965. “The analysis of the stability of general slip surfaces.” Géotechnique 15 (1): 79–93. https://doi.org/10.1680/geot.1965.15.1.79.

Google Scholar

Netease. 2009. Primary factor triggering falling over of the building within Lotus Riverside neighborhood: Lateral ground movement arising from unequal lateral earth pressure conditions. Shanghai, China: Netease.

Google Scholar

Norlund, R. L., and D. U. Deere. 1970. “Collapse of Fargo grain elevator.” J. Soil Mech. Found. Div. 96 (2): 585–607.

Google Scholar

Paik, K. H., and R. Salgado. 2003. “Determination of the bearing capacity of open-ended piles in sand.” J. Geotech. Geoenviron. Eng. 129 (1): 46–57. https://doi.org/10.1061/(ASCE)1090-0241(2003)129:1(46).

Google Scholar

Paikowsky, S., R. V. Whitman, and M. M. Baligh. 1989. “A new look at the phenomenon of offshore pile plugging.” Mar. Geotechnol. 8 (3): 213–230. https://doi.org/10.1080/10641198909379869.

Google Scholar

Paikowsky, S. G., and R. V. Whitman. 1990. “The effect of plugging on pile performance and design.” Can. Geotech. J. 27 (4): 429–440. https://doi.org/10.1139/t90-059.

Google Scholar

Peck, R. B. 1973. Influence of nontechnical factors on the quality of embankment dams. New York: Wiley.

Google Scholar

Peck, R. B., and F. G. Byrant. 1953. “The bearing-capacity failure of the Transcona elevator.” Géotechnique 3 (5): 201–208. https://doi.org/10.1680/geot.1953.3.5.201.

Google Scholar

Poulos, H. G., and L. T. Chen. 1997. “Pile response due to excavation induced lateral soil movement.” J. Geotech. Geoenviron. Eng. 123 (2): 94–99. https://doi.org/10.1061/(ASCE)1090-0241(1997)123:2(94).

Google Scholar

Prandtl, L. 1920. “Uber die harte plastischer korper.” Nachr. K. Gesell. Wiss. Gottingen 1920: 74–85. [In German.].

Google Scholar

Raines, R. D., O. Ugaz, and M. O’Neill. 1992. “Driving characteristics of open toe piles in sand.” J. Geotech. Eng. 118 (1): 72–88. https://doi.org/10.1061/(ASCE)0733-9410(1992)118:1(72).

Google Scholar

Schanz, T., P. A. Vermeer, and P. G. Bonnier. 1999. “Formulation and verification of the hardening-soil model.” In Beyond 2000 in computational geotechnics, edited by R. B. J. Brinkgreve, 281–290. Rotterdam, Netherlands: Balkema.

Google Scholar

SGIDI (Shanghai Geotechnical Investigation and Design Institute). 2009. Post-failure geotechnical exploration report for Lotus Riverside resident neighborhood. [In Chinese.] 2009-GZ-208. Shanghai, China: SGIDI Engineering Consulting Group.

Google Scholar

Shanghai Urban and Rural Construction and Transportation Committee. 2010. Technical code for excavation engineering. [In Chinese.] DG/TJ08-61. Shanghai, China: Tongji Univ. Press.

Google Scholar

Slope/W. 2012. Slope/W for slope stability analysis—user’s guide. Calgary, AB, Canada: Geoslope International.

Google Scholar

Sowers, G. F. 1993. “Human factors in civil and geotechnical engineering failures.” J. Geotech. Eng. 119 (2): 238–256. https://doi.org/10.1061/(ASCE)0733-9410(1993)119:2(238).

Google Scholar

Stark, T. D., A. K. Baghdady, O. Hungr, and J. Aaron. 2017. “Case study: Oso, Washington, landslide of March 22, 2014—Material properties and failure mechanism.” J. Geotech. Geoenviron. Eng. 143 (5): 05017001. https://doi.org/10.1061/(ASCE)GT.1943-5606.0001615.

Google Scholar

Stark, T. D., and J. M. Duncan. 1991. “Mechanisms of strength loss in stiff clays.” J. Geotech. Eng. 117 (1): 139–154. https://doi.org/10.1061/(ASCE)0733-9410(1991)117:1(139).

Google Scholar

Stark, T. D., H. T. Eid, W. D. Evans, and P. E. Sherry. 2010. “Municipal solid waste slope failure. II. Stability analyses.” J. Geotech. Geoenviron. Eng. 126 (5): 408–419. https://doi.org/10.1061/(ASCE)1090-0241(2000)126:5(408).

Google Scholar

Stark, T. D., P. J. Ricciardi, and R. D. Sisk. 2018. “Case study: Vertical drain and stability analyses for a compacted embankment on soft soils.” J. Geotech. Geoenviron. Eng. 144 (2): 05017007. https://doi.org/10.1061/(ASCE)GT.1943-5606.0001786.

Google Scholar

SXGEI. 2006. Geotechnical exploration report for lotus riverside resident neighborhood. [In Chinese.] 2005-03-22. Shanghai, China: Shanghai Xie-Li Geotechnical Engineering Investigation.

Google Scholar

Sy, A., T. Thavaraj, B. Hamersley, and D. Gillespie. 2011. “Retrofit of a pile-supported bridge in soft clay damaged by failure of adjacent soil stockpile.” In Proc., 2011 Pan-Am CGS Geotechnical Conf. Toronto, ON, Canada: Canadian Geotechnical Society.

Google Scholar

Tan, Y., R. Huang, Z. Kang, and W. Bin. 2016. “Covered semi-top-down excavation of subway station surrounded by closely spaced buildings in downtown Shanghai: Building response.” J. Perform. Constr. Facil. 30 (6): 04016040. https://doi.org/10.1061/(ASCE)CF.1943-5509.0000892.

Google Scholar

Tan, Y., W. Jiang, W. Luo, Y. Lu, and C. Xu. 2018. “Longitudinal sliding event during excavation of Feng-Qi Station of Hangzhou Metro Line 1: Postfailure investigation.” J. Perform. Constr. Facil. 32 (4): 04018039. https://doi.org/10.1061/(ASCE)CF.1943-5509.0001181.

Google Scholar

Tan, Y., and H. Lan. 2012. “Vibration effects attributable to driving of PHC pipe piles.” J. Perform. Constr. Facil. 26 (5): 679–690. https://doi.org/10.1061/(ASCE)CF.1943-5509.0000278.

Google Scholar

Tan, Y., and M. Li. 2011. “Measured performance of a 26 m deep top-down excavation in downtown Shanghai.” Can. Geotech. J. 48 (5): 704–719. https://doi.org/10.1139/T10-100.

Google Scholar

Tan, Y., and G. Lin. 2013. “Full-scale testing of open-ended steel pipe piles in thick varved clayey silt deposits along the Delaware River in New Jersey.” J. Geotech. Geoenviron. Eng. 139 (3): 518–524. https://doi.org/10.1061/(ASCE)GT.1943-5606.0000777.

Google Scholar

Tan, Y., and Y. Lin, and X. Li. 2015. “Closure to ‘Comprehensive load test on prestressed concrete piles in alluvial clays and marl in Savannah, Georgia’ by Yong Tan and Guoming Lin.” J. Perform. Constr. Facil. 29 (1): 07014002. https://doi.org/10.1061/(ASCE)CF.1943-5509.0000712.

Crossref

Google Scholar

Tan, Y., and Y. Lu, and D. Wang. 2019. “Practical solutions for concurrent excavation of neighboring mega basements closely surrounded by utility tunnels in Shanghai Hongqiao CBD.” Pract. Period. Struct. Des. Constr. 24 (4): 05019005. https://doi.org/10.1061/(ASCE)SC.1943-5576.0000437.

Google Scholar

Tan, Y., and D. Wang. 2013a. “Characteristics of a large-scale deep foundation pit excavated by the central-island technique in Shanghai soft clay. I: Bottom-up construction of the central cylindrical shaft.” J. Geotech. Geoenviron. Eng. 139 (11): 1875–1893. https://doi.org/10.1061/(ASCE)GT.1943-5606.0000928.

Google Scholar

Tan, Y., and D. Wang. 2013b. “Characteristics of a large-scale deep foundation pit excavated by the central-island technique in Shanghai soft clay. II: Top-down construction of the peripheral rectangular pit.” J. Geotech. Geoenviron. Eng. 139 (11): 1894–1910. https://doi.org/10.1061/(ASCE)GT.1943-5606.0000929.

Google Scholar

Terzaghi, K. 1943. Theoretical soil mechanics. New York: Wiley.

Crossref

Google Scholar

Terzaghi, K. 1958. Consultants, clients and contractors: Harvard soil mechanics series, 1–67. Boston: Harvard University.

Google Scholar

Vanden Berge, D. R., and S. G. Wright. 2016. “Improved undrained strength interpolation scheme for rapid drawdown.” J. Geotech. Geoenviron. Eng. 142 (6): 06016002. https://doi.org/10.1061/(ASCE)GT.1943-5606.0001471.

Google Scholar

Vesic, A. S. 1973. “Analysis of ultimate loads of shallow foundations.” J. Soil Mech. Found. Div. 99 (1): 45–76.

Crossref

Google Scholar

Wang, W. D., Q. Li, Y. Hu, J. W. Shi, and C. W. W. Ng. 2017. “Field investigation of collapse of a 13-story high-rise residential building in Shanghai.” J. Perform. Constr. Facil. 31 (4): 04017012. https://doi.org/10.1061/(ASCE)CF.1943-5509.0001005.

Google Scholar

Wang, W. D., C. W. W. Ng, Y. Hong, Y. Hu, and Q. Li. 2019. “Forensic study on the collapse of a high-rise building in Shanghai: 3D centrifuge and numerical modelling.” Géotechnique 69 (10): 847–862. https://doi.org/10.1680/jgeot.16.P.315.

Google Scholar

Wei, X., J. Zheng, and L. Wu. 2012. “Failure model of soft clay hard shell layer foundation.” [In Chinese with English Abstract.] J. Highway Transp. Res. Dev. 29 (8): 31–37. https://doi.org/103969/j.issn.1002-0268.2012.08.006.

Google Scholar

Whittle, A. J., D. J. DeGroot, C. C. Ladd, and T. H. Seah. 1994. “Model prediction of anisotropic behavior of Boston Blue Clay.” J. Geotech. Eng. 120 (1): 199–224. https://doi.org/10.1061/(ASCE)0733-9410(1994)120:1(199).

Google Scholar

Wright, S. G. 2013. “2013 H. Bolton Seed lecture: Slope stability computations.” In Proc., Geo-Congress 2013. Reston, VA: ASCE.

Google Scholar

Yang, S. F. 2010. “Disciplines of foundation deformation under large area earth loading.” [In Chinese with English Abstract.] Shanghai Geol. 1: 16–20. https://doi.org/CNKI:SUN:SHAD.0.2010-01-006.

Google Scholar

Ye, W. M., Y. Bai, Q. Jin, B. Chen, and Y. J. Cui. 2006. “Lab experimental study on soil-water characteristics of Shanghai soft clay.” [In Chinese with English Abstract.] Chin. J. Geotech. Eng. 28 (2): 260–263.

Google Scholar

Yin, J., and F. Xu. 2009. “3D numerical analysis of the overturning of a residential under construction in Shanghai.” [In Chinese with English Abstract.] In Proc., Annual Meeting of Geotechnical Engineering in Shanghai, 152–156. Shanghai, China: Shanghai Society of Civil Engineers.

Google Scholar

Information & Authors

Information

Published In

Journal of Geotechnical and Geoenvironmental Engineering

Volume 146 • Issue 7 • July 2020

Copyright

This work is made available under the terms of the Creative Commons Attribution 4.0 International license, https://creativecommons.org/licenses/by/4.0/.

History

Received: Nov 7, 2018

Accepted: Jan 14, 2020

Published online: Apr 29, 2020

Published in print: Jul 1, 2020

Discussion open until: Sep 29, 2020

Authors

Affiliations

Yong Tan, M.ASCE https://orcid.org/0000-0003-3107-5454 [email protected]

Professor, Dept. of Geotechnical Engineering, College of Civil Engineering, Tongji Univ., 1239 Sipping Rd., Shanghai 200092, China. ORCID: https://orcid.org/0000-0003-3107-5454. Email: [email protected]

View all articles by this author

Wei-Zhen Jiang [email protected]

Graduate Student, Dept. of Geotechnical Engineering, College of Civil Engineering, Tongji Univ., 1239 Sipping Rd., Shanghai 200092, China. Email: [email protected]

View all articles by this author

Hai-Sheng Rui [email protected]

Graduate Student, Dept. of Geotechnical Engineering, College of Civil Engineering, Tongji Univ., 1239 Sipping Rd., Shanghai 200092, China. Email: [email protected]

View all articles by this author

Ye Lu, A.M.ASCE [email protected]

Associate Professor, Dept. of Civil Engineering, Shanghai Univ., 99 Shang-Da Rd., Shanghai 200444, China (corresponding author). Email: [email protected]

View all articles by this author

Da-Long Wang [email protected]

Senior Engineer, Shanghai Geotechnical Investigation and Design Institute Engineering Consulting (Group) Co., Ltd., 180 Shu-Feng Rd., Shanghai 200032, China. Email: [email protected]

View all articles by this author

Metrics & Citations

Metrics

Citations

Download citation

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

Abstract

Introduction

Descriptions of Events

Previous Investigations and Analyses

Other Considerations

Subsurface Condition

Postfailure Investigations

Simple Theoretical Analyses

Preliminary Global Stability Assessment

Soil Stockpile 1

Flood Walls Adjacent to Stockpiles 1 and 2

Comprehensive Calculations and Analyses

Flood Wall along the Dingpu River

Evaluation of FS against Bearing-Capacity Failure of the Subgrade below Stockpiles

Evaluation of FS for Flood Walls against Sliding and Overturning Failures

Buildings Founded on Piles

Examination on Pile Capacities

Examination on Subgrade

3D Numerical Analyses

FE Model

Constitutive Models and Parameters

Simulation Procedures

Analysis of Simulation Results

Discussion

Construction Sequence

Embedment Depth of Building Footing

Pile Types

Temporary Soil Stockpiling

Summary and Conclusion

Supplemental Data

Supplemental Materials

Acknowledgments

References

Information

Published In

Copyright

History

Authors

Affiliations

Metrics

Citations

Download citation

Cited by

Figures

Other

Share

Copy the content Link

Share with email

Share

Request Username

Create a new account

Change Password

Password Changed Successfully

Verify Phone

Congrats!