Available online at www.sciencedirect.com ScienceDirect Procedia Engineering 173 (2017) 1739 – 1746 11th International Symposium on Plasticity and Impact Mechanics, IMPLAST 2016 Seismic Vulnerability of Urban Bridges due to Liquefaction using Nonlinear Pushover Analysis and Assessing Parameters for Damage Detection Kashif Quamar Inqualabia, Rajeev Kumar Gargb*, K Balaji Raoc a JRF and Phd Scholar (AcSIR), CSIR-CRRI, New Delhi, 110025, India Chief Scientist, Bridge Engineering and Structures Division, CSIR-CRRI, New Delhi, 110025, India c Chief Scientist, Risk & Reliability of Structures and Advisor (M), CSIR-SERC, Chennai, 700113, India b Abstract Pre-disaster planning strategies and strengthening to ensure certain degree of performance of deficient bridges can reduce the seismic risk to a great extent Codes have been steadily moving towards performance based designs This study aims at investigating the effect of liquefaction during earthquake on bridges founded in the soft alluvium For this purpose, nonlinear static analysis has been used to derive fragility and vulnerability curves A two span box girder bridge has been analysed The bridge foundation soil strata have layers of fine sand to sandy silt deposits The capacity and vulnerability of the bridge under normal soil and liquefied conditions have been studied It has been observed that the time period and damping of the structure have increased Fragility curves have been developed to assess the seismic performance of the bridge This study will help to ensure the performance criteria/ levels of the bridges and more informed pre-disaster planning strategies to avoid seismic risk and add to the existing knowledge of the concept of performance based design of RC bridges © 2017 The Authors Published by Elsevier Ltd This is an open access article under the CC BY-NC-ND license © 2016 The Authors Published by Elsevier Ltd (http://creativecommons.org/licenses/by-nc-nd/4.0/) Peer-reviewunder underresponsibility responsibility ofthe organizing committee of Implast Peer-review of the organizing committee of Implast 2016 2016 Keywords:Bridge; performance; earthquake; liquefaction; fragility, damage assessment, seismic risk 1.0 Introduction The frequent observations of liquefaction during earthquakes, and its associated damage, have made liquefaction a major concern in geotechnical earthquake engineering Hazen used the term "liquefied" first time in reference to the 1918 failure of the Calaveras Dam in California Soil liquefaction is a phenomenon describing a saturated or partially saturated soil substantially losing shear strength and stiffness due to sudden change in effective stress, usually seismic activity or other sudden changes in stress condition, causing the soil to behave like a liquid In other words when the effective stress becomes zero on development of excess pore water pressure the soil behaves like * Corresponding author Tel.: +91 9868111558 E-mail address: rkgcrri@gmail.com 1877-7058 © 2017 The Authors Published by Elsevier Ltd This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) Peer-review under responsibility of the organizing committee of Implast 2016 doi:10.1016/j.proeng.2016.12.211 1740 Kashif Quamar Inqualabi et al / Procedia Engineering 173 (2017) 1739 – 1746 liquid (having no shear strength) The pressures generated during large earthquakes with many cycles of shaking can cause the liquefied sand and excess water to force its way out to the ground surface from several metres below the ground Liquefaction was a major factor in the destruction of structures in 1989 Loma Prieta earthquake, 1995 great Han Shin earthquake, 2010 Canterbury earthquake and in 2001 Bhuj earthquake Now liquefaction is acknowledged as a crucial problem in seismically-active regions and research have been carried out to know the nature of cyclic soil behaviour, the composition of susceptible soils, and techniques to mitigate the associated problems Pile foundations is being used extensively for supporting bridge structures in seismic areas and have been used for this purpose in nonliquefying as well as liquefying soils In liquefiable soils, progressive build-up of pore water pressure may result in loss of strength and stiffness resulting in large bending moments and shear forces on the pile The mechanism of pile behaviour in liquefying soil has been investigated by several investigators in the recent years based on observations of pile performance during earthquakes The behaviour of piles in liquefying soil is discussed in this paper The response of pile in liquefied soil is influenced by non-linear behaviour of soil, which results in degradation of shear modulus and increased material damping with increase in displacement Soil displacements and lateral spreading associated with liquefaction may exert damaging lateral pressure on the piles resulting in the failure under buckling Especially for structures like bridges, which may be founded on spatially varying soil profile liquefaction effect can be devastating Bridges whose foundations bear directly on loose sand which may liquefy will experience a sudden loss of support resulting in drastic and irregular settlement of the bridge piers and the pile may buckle under kinematic loading causing structural damage, including cracking of foundations and damage to the bridges structure itself, or may leave the structure unserviceable afterwards, even without structural damage Where a thin crust of non-liquefied soil exists between building foundation and liquefied soil, punching shear failure may occur The failures occur in the form of settlement and deflection of the superstructure, rendering it either useless or very expensive to rehabilitate Several studies have been carried out to investigate the failure pattern of the piles due to liquefaction [1] These studies hinted at the location of the cracks, hinge formation and damage patterns for the piles Due to liquefaction the piles act as long columns with increased slenderness and its response is more flexible/elastic and it may experience more lateral deformation As the effective length (Leff) increases during liquefaction critical load also decreases If a constant static axial load is assumed, it can be observed that the ratio of critical load before and after liquefaction increases The pile load deflection and bending moment in pile also increases As the ratio increases the bending moment in the pile reaches its plastic moment capacity (Mp) at a much lower lateral load This indicates that the stresses in the pile during and after liquefaction have exceeded the yield stress of its material Hinges were found to have formed at various depths along the pile: at the pile head, at the middle of the liquefiable layer, and toward the interface of liquefiable/non-liquefiable layer This clearly demonstrates the important of consideration of liquefaction for design In this context it must be mentioned that piles are currently designed for geotechnical load-carrying capacity (maximum allowable load in the pile) and against bending failure however, influence of buckling due to lateral thrust caused by liquefaction is generally ignored Bhattacharya [1] observed that the surrounding soil liquefies in an earthquake axial load alone can cause a pile to fail by developing buckling mechanism The liquefaction will increase the time period of pile-supported structures along with an increase in the overall damping ratio (around 20%) (Lombardi and Bhattacharya [2].These parameters have important design implications All current design methods, such as IS 1893 (2002) [3], NEHRP (2000) [4], JRA (2002)[5], and Euro code (2012) [6], focused on bending strength of the pile to avoid bending failure due to lateral loads (combination of inertia and lateral spreading) In contrast to these conventional design codes, which consider bending mechanism as the main design consideration, recent research showed that buckling mechanism failure may be more critical design condition for an axially loaded pile due to loss of lateral support as a result of liquefaction of soil During design process, beam bending and column buckling are treated differently Bending is a stable mechanism as long as the pile remains elastic and secondary failure (e.g., local buckling) is not a possibility This failure mode depends on the bending strength (e.g., yield moment capacity and plastic moment capacity) of the member under consideration In contrast, buckling is an unstable mechanism which occurs suddenly and drastically when the elastic critical load is reached It is one of the most destructive mode of failure and depends on the geometrical properties of the member, i.e., slenderness ratio, rather than the strength of member 1.1 Liquefaction susceptibility of soil The soil can liquefy if it is saturated, able to contract under shear and not permeable enough to allow drainage for the Kashif Quamar Inqualabi et al / Procedia Engineering 173 (2017) 1739 – 1746 1741 build-up of excess pore water pressure along with other factors like intensity of earthquake, location of ground water table, soil type, particle size distribution, confining pressure, soil stratification etc Standard penetration test (SPT) has been used to assess liquefaction susceptibility This is also recommended by Euro code The steps recommended by Euro code are summarized below: Step The shear stress,߬ is assessed ߬ ൌ ͲǤͷߙܵߪ௩ (1) Here, α is the ground acceleration in terms of gravity, S is the soil parameter as specified in clause 4.2.2 of part 5(Euro code 8) based on the stratigraphic profile and intensity of earthquake ߪ௩ ൌ ߛܼis the total overburden pressure, where ߛ is the unit weight of soil and Z is the depth below the ground level This equation for seismic shear stress is applicable up to a depth of 20 m Step The SPT data is normalised to a reference effective overburden pressure of 100 kPa and to a 60% ratio of impact energy over theoretical free fall energy The normalization can be done by using equation as: ாோ ܰ ൌ ܰௌ் ටͳͲͲൗߪ ᇱ (2) ௩ Where ER is the measured energy ratio In Europe its typically 70% and in India it is 60% Step Determine the cyclic stress ratio using the equation 1.3 This method is calibrated for an earthquake of magnitude 7.5 and for earthquakes other than 7.5 magnitudes, cyclic stress ratio needs to be multiplied by a factor CM given in Euro Code part Step Based on the figure B.1 (Eurocode part 5), susceptibility to liquefaction can be assessed A relationship between stress ratios causing liquefaction and N1 (60) values for clean and silty sands for Ms = 7.5 earthquake is provided Point above the curve shows susceptibility to liquefaction and point below the curve shows non liquefiability of the soil 2.0 Assessment of liquefaction effects on an urban bridge 2.1 Analytical modelling A box girder bridge located on the bank of Yamuna River in New Delhi has been studied The region has the alluvial soil of Yamuna catchment The soil stratification shows deposits of fine sand to silty sand In the adjacent area presence of rock has been identified Below ground up to 0.5 m a fill is observed Below this up to a depth of 28.0 m, light grey fine sand is observed This is underlain by light brown silty sand to the final explored depth of 40 m The recommended depth of the designed pile is 32 m to 35 m The soil stratification along with corrected N-values is shown in Fig 10 Corrected SPT value (N) 20 30 Depth (m) 10 15 20 25 30 35 40 Fig 1: Soil strata profile The liquefaction susceptibility of the soil strata has been assessed using Euro code part methodology For seismic 1742 Kashif Quamar Inqualabi et al / Procedia Engineering 173 (2017) 1739 – 1746 zone IV, ground acceleration value is taken as 0.24 g The soil type conforms to class C of the Euro code which gives S coefficient as 1.15 For a depth of 14 m, ߪ௩ is 380.00 kN/m2 Other parameters are given below ఛ ߬ ൌ ͲǤͷ Ͳ כǤʹͶ ͳ כǤͳͷ ͵ כͺͲǤͲͲ = 68.172kN/m2 ߪ ᇱ ௩ ൌ ͳͻ ͳ כͶ െ ͻǤͺͳ ͻ כǤͷ ൌ ͳʹǤͺͳkN/m2 Now, ᇲ ൌ ͲǤ͵ͻ ఙ ೡ Thus, from Fig B.1 (Eurocode part 5), the soil is found to be liquefaction susceptible For the assessment of effects of liquefaction, two analytical models of soil conditions namely without liquefaction (with full soil confinement of piles) and with liquefaction (with lateral spread having no confinement) have been studied The structural details of the bridge are shown in Figs and Hemsley’s approach based on Winkler foundation has been used to define soil stiffness springs This approach has been used with good level of reliability for general purpose dynamic analysis The discrete soil springs with the stiffness k (based on the soil type), effective influence depth z, and pile diameter D the Winkler soil reaction modulus or spring constant k s along the length of pile foundation is determined using equations and Ks = Dks (3) Ki = ksZiBiD (4) Where, Bi = (Zi+1 – Zi-1)/2, and Zi+1 is the depth of the next spring node and Zi-1 is the depth of the previous spring node The length of pile below the bed level is divided into a no of elements based on the type of soil strata The spring with spring constant ks proportional to the depth of node below the bed level can be computed from the above equation Fig 2: Pier elevation and pile cap in plan Fig 3: Pile cap details Kashif Quamar Inqualabi et al / Procedia Engineering 173 (2017) 1739 – 1746 2.2 Fragility assessment Pushover analysis is carried out by finding the target displacement using response spectrum analysis The response spectrum is defined according to IRC-6-2014 with Z value as 0.24 (seismic zone IV) The directional combination (X+0.3Y and Y+0.3X) of the response spectrum has been taken into consideration The pushover load application is displacement controlled i.e the load is applied till either the target displacement is reached or the structure has failed The hinge definitions have been taken from FEMA 356 (2000) with P-M2-M3 interaction The hinge assignment is applied at the joints of small elements of pier and pile, pile cap joints, and at the depth where soil profile changes so as to capture the true potential failure location Based on the nonlinear static analysis i.e pushover analysis, different damage states have been derived in comparisons with the performance levels defined in FEMA 356 [7] and ATC 40 [8] Using the lognormal standard deviation value (0.6) as suggested by FHWA [9] retrofitting manual fragility curves have been developed Discrete probability of different damage states under a seismic activity of intensity VIII or PGA of 0.24 g has been calculated The damage states have been defined as per HAZUS99 which are mentioned below • • • • Slight Damage: Minor cracking and spalling to the abutment, cracks in shear keys at abutments, minor spalling and cracks at hinges, minor spalling at the column (damage requires no more than cosmetic repair) or minor cracking to the deck Moderate Damage: Any column experiencing moderate (shear cracks) cracking and spalling (column structurally still sound), moderate movement of the abutment (DSI] Fig 6: Fragility curve for non-liquefied condition 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 Slight Moderate Severe Complete 200 400 600 800 1000 Displacement (mm) 1200 Fig 7: Fragility curve at full liquefied condition 0.9 0.8 0.7 Probability P [D >DSI] 0.8 0.6 0.5 0.4 non-liquefied fully liquefied 0.3 0.2 0.1 No Slight Moderate Severe Damage State Fig 8: Change in Damage states due to liquefaction under PGA of 0.24 g Complete 1400 1746 Kashif Quamar Inqualabi et al / Procedia Engineering 173 (2017) 1739 – 1746 surface with some permanent drifts and out of plane failure due to excessive deflection at the top may be observed As the spectral displacement increases the fragility curve of liquefied condition shows a higher rate of increase in probability of severe and complete damages in comparison of non-liquefied condition To mitigate the effect of liquefaction soil improvement techniques such as vibro-compaction, dynamic compaction and stone columns can be used For retrofitting of existing structures, grouting can be used The pile should be provided with proper lateral reinforcement to take the additional dynamic horizontal thrust 4.0 Conclusions In the present study, the assessment of effect of liquefaction on a mono-pier with a group of piles has been successfully achieved Fragility curves have been developed to assess the seismic performance of the bridge and the following are concluded a) From the fragility curves it has been observed that during earthquake, consideration of liquefaction effect in modelling highlights greater vulnerability of the bridge b) The piers due to liquefaction may undergo excessive lateral deflection causing them to have slight to severe damage in the piles and the pier c) During soil liquefaction especially in the case of soil spread, there is 32% probability of occurrence of slight damage that it may experience spalling of cover to reinforcement, some cracking on the surface Liquefaction has also increased the probability of severe damage up to 4% i.e the bridge may be rendered failed as the bridge structure experiences substantial loss of strength Shear failure in the piles and columns, visible shear cracks on the surface with some permanent drifts and out of plane failure due to excessive deflection at the top may be observed d) The increased slenderness of the pile may cause buckling of the piles which warrants appropriate measures while designing such structures e) The assessment of damage states so discussed helps in assessing parameters for damage detection towards structural haelth assessment of bridges Acknowledgement The authors thankfully acknowledge the financial support provided by Department of Science and Technology (DST), New Delhi under grant DST/TSG/STS/2011/43 References [1] S Bhattacharya, Pile instability during earthquake liquefaction, Doctoral dissertation, University of Cambridge (2003) [2] D Lombardi, S Bhattacharya, Modal analysis of pile-supported structures during seismic liquefaction Earthquake Engineering Structural Dynamics 43 (2014) 119–138 [3] IS: 1893 (Part I), Criteria for earthquake resistant design of structures, general provisions and buildings, Bureau of Indian Standards, New Delhi, (2002) [4] NEHRP, Recommended provisions for seismic regulations for new buildings and other structures, Building Seismic Safety Council, Washington D.C (2000) [5] JRA: Design specifications of highway bridges, Part V: Seismic design, Japan Road Association, (2002) [6] Eurocode 8, Seismic design of buildings worked examples, Joint Research Centre, European Commission, Italy (2012) [7] FEMA: 356-2000, Pre standard and commentary for the seismic rehabilitation of buildings, American Society of Civil Engineers, Virginia, USA (2000) [8] ATC: 40, Seismic evaluation and retrofit of concrete buildings, Redwood City, Proposition 122 Seismic Retrofit Practices Improvement Program Report SSC 96-01, Applied Technology Council, California Seismic Safety Commission, U.S.A (1996) [9] FHWA-HRT-06-032, Seismic retrofitting manual for highway structures: Part – Bridges, MCEER University, Buffalo State University of New York, New York, USA (2006) [10] HAZUSMR MR4, FEMA-2003, Multi-hazard loss estimation methodology: Earthquake model, technical manual, Department of Homeland Security Emergency Preparedness and Response Directorate FEMA, Washington, D.C., USA (2003)