Seismic, Wind and Extreme-Event Engineering

Introduction and Context

Bridges are long-life infrastructure systems exposed to hazards whose intensity, duration and consequences vary greatly. Earthquakes may impose rapid cyclic demands and foundation movements; wind may cause static, dynamic and aeroelastic effects; floods can produce scour, debris impact and hydraulic forces; fire, blast, landslides and vehicle or vessel impact can cause highly localized damage. Unlike ordinary gravity-load design, extreme-event engineering must address uncertainty, low-probability hazards, nonlinear response, interacting failure modes and the consequences of loss of service.

Modern bridge engineering therefore extends beyond resistance to individual load effects. It seeks to provide ductility, controlled damage, alternative load paths, inspectability, rapid recovery and acceptable residual risk. This chapter develops a performance-oriented framework for evaluating and designing bridges under seismic, wind and other extreme events. It is intended for postgraduate students who must integrate structural analysis, geotechnical engineering, hydraulics, materials, construction, risk management and infrastructure governance.

Learning Objectives

On completing this chapter, the reader should be able to:

1. Explain the principal hazard mechanisms affecting bridge systems.

2. Distinguish force-based, displacement-based, performance-based and risk-informed approaches.

3. Evaluate seismic ductility, capacity design, isolation and damping strategies.

4. Assess wind-induced static, dynamic and aeroelastic effects.

5. Identify and mitigate scour, flood, debris, fire, blast, landslide and impact hazards.

6. Apply principles of redundancy, robustness, resilience and progressive-collapse prevention.

7. Develop a rational extreme-event assessment and emergency-management strategy.

1. Hazard, Vulnerability and Risk

A useful risk model separates three elements:

• Hazard is the probability and intensity of an event at a location.
• Exposure describes the bridge, users, networks and economic activities subject to the hazard.
• Vulnerability describes the probability and extent of damage conditional on hazard intensity.

Risk may be expressed conceptually as

\[

R=\sum_i P(E_i)\,C_i,

\]

where \(P(E_i)\) is the annual or specified-period probability of event state \(E_i\), and \(C_i\) is its consequence. This is an expected-value representation, not a complete description of societal risk. It assumes that event states and consequences have been defined consistently; correlation, uncertainty and distributional effects may require more advanced probabilistic models.

Bridge performance should be stated by limit states rather than by strength alone. Typical states are:

• Operational: traffic continues with little or no intervention.
• Immediate occupancy: limited inspection and repair are needed.
• Life safety: significant damage is permitted, but collapse and unacceptable loss of life are prevented.
• Collapse prevention: near-collapse conditions are tolerated only under rare extreme actions.

A performance objective combines a hazard level with an intended limit state. Critical routes, emergency bridges and bridges whose failure would isolate communities generally require more demanding objectives than low-consequence structures.

2. Seismic Engineering of Bridges

2.1 Seismic demand and bridge response

Earthquake action arises from ground acceleration, velocity and displacement, together with spatially varying support movements. A bridge is a multi-degree-of-freedom system involving deck mass, piers, bearings, abutments, foundations, soil and adjacent approach fills. Its response depends on stiffness, mass distribution, damping, hysteresis, torsion, pounding and soil–structure interaction.

For a linear single-degree-of-freedom approximation, the equation of motion is

\[

m\ddot u+c\dot u+ku=-m\ddot u_g,

\]

where \(u\) is relative displacement, \(m\) is mass, \(c\) is viscous damping, \(k\) is stiffness and \(\ddot u_g\) is ground acceleration. The natural circular frequency is \(\omega_n=\sqrt{k/m}\), the period is \(T=2\pi/\omega_n\), and the damping ratio is \(\zeta=c/(2\sqrt{km})\). This idealization is appropriate for conceptual analysis or a dominant mode, but bridge design normally requires multimode analysis and nonlinear component models.

A response spectrum gives the maximum response of idealized oscillators as a function of period and damping. It does not represent a time history and cannot by itself capture spatially varying ground motion, pounding, permanent ground displacement or detailed hysteretic behavior. Modal response-spectrum analysis is efficient for elastic global response; nonlinear static or dynamic analysis is preferable where redistribution, cyclic deterioration or large inelastic displacement governs.

2.2 Ductility and capacity design

Ductility is the ability to undergo substantial inelastic deformation while retaining resistance. It is not simply a material property: it depends on confinement, shear strength, anchorage, lap-splice arrangement, local buckling, low-cycle fatigue and detailing.

The displacement ductility ratio is commonly written as

\[

\mu_\Delta=\frac{\Delta_{\max}}{\Delta_y},

\]

where \(\Delta_{\max}\) is maximum displacement and \(\Delta_y\) is the displacement at first significant yield. The ratio is meaningful only when yield and maximum displacement are defined consistently. A high calculated ductility ratio is unacceptable if brittle shear failure, unconfined concrete crushing or bar anchorage failure occurs first.

Capacity design deliberately makes selected components ductile and protects other components from brittle failure. In a reinforced-concrete pier, plastic hinges may be permitted near the base and possibly the top, while the design shear, joint, foundation and connection forces are derived from probable plastic resistance rather than only from elastic analysis. The hierarchy is typically:

1. Identify intended dissipative regions.

2. Provide confinement and stable flexural reinforcement in those regions.

3. Design shear and connections for amplified flexural capacity.

4. Check foundations, bearings, seats and adjacent components for compatible displacement and force.

5. Verify that global mechanisms are preferable to local brittle mechanisms.

The overstrength factor \(\Omega_o\) is used in many code frameworks to represent actual strength exceeding nominal design strength. Its value and use are code-specific and must not be transferred between standards without verification.

2.3 Bearings, joints and unseating

Seismic vulnerability frequently arises at interfaces rather than in main members. Important failure modes include bearing loss, insufficient seat width, restrainer rupture, expansion-joint failure, pounding between adjacent spans, shear-key damage and abutment displacement. Design should provide a continuous load path for longitudinal and transverse actions, positive restraint where required, adequate displacement capacity and access for inspection.

Pounding is governed by relative movement, phase differences and gap dimensions. It can be reduced through compatible stiffness and mass, adequate separation, energy-absorbing devices or restrainers designed for realistic demand. Restrainers must not introduce an unintended brittle force path or transfer excessive demand to weak diaphragms.

2.4 Isolation and supplemental damping

Seismic isolation lengthens the effective period and limits force transmission by inserting flexible and energy-dissipating devices between the superstructure and substructure. Common systems include elastomeric bearings, lead-rubber bearings, friction pendulum devices and viscous dampers.

For an equivalent linear isolated system, the effective period is approximately

\[

T_e=2\pi\sqrt{\frac{M}{K_e}},

\]

where \(M") is effective seismic mass and \(K_e\) is effective lateral stiffness. The approximation assumes a dominant isolated mode and an equivalent linear representation. Real devices may be strongly nonlinear, temperature-dependent, velocity-dependent or friction-dependent; qualification testing, aging assessment and displacement checks are essential.

Isolation does not eliminate seismic risk. Large displacement must be accommodated at joints, utilities, restrainers and abutments. Stability under vertical load, uplift, torsion, accidental restraint and inspection access must be checked. Supplemental damping, usually through viscous or hysteretic devices, can reduce displacement and force demand, but device failure, leakage, replacement and interaction with thermal movements require explicit consideration.

2.5 Foundations and soil–structure interaction

Deep foundations can experience inertial forces from the bridge and kinematic forces from soil deformation. Liquefaction may reduce bearing capacity, induce lateral spreading and impose large permanent demands on piles. Abutment backfill can mobilize passive resistance, but this resistance is displacement-dependent and should not be assumed to act elastically at small movement.

Seismic evaluation must therefore integrate geotechnical investigations, liquefaction screening, cyclic resistance, lateral spreading analysis, pile–soil interaction and post-earthquake settlement. A structurally strong pier supported by unstable soil does not constitute a robust bridge system.

3. Wind Engineering

3.1 Wind actions

Wind produces mean drag, fluctuating buffeting, vortex shedding, flutter, galloping and, during construction, unusual temporary configurations. A basic drag estimate is

\[

F_D=\tfrac12\rho V^2 C_D A,

\]

where \(F_D\) is drag force, \(\rho\) is air density, \(V\) is reference wind speed, \(C_D\) is drag coefficient and \(A\) is projected area. The expression assumes a sufficiently defined reference velocity and a quasi-steady force model. It is not sufficient for long flexible bridges, where turbulence, aerodynamic derivatives, spatial correlation and motion-induced forces are important.

Wind speed varies with height and terrain. Gust factors account, in a simplified manner, for fluctuations around mean wind. For long-span bridges, wind-tunnel testing, computational fluid dynamics and aeroelastic analysis may be required. Scale effects, Reynolds number, turbulence simulation and boundary-layer representation must be addressed before accepting test results.

3.2 Aeroelastic instability

Flutter is a coupled instability involving aerodynamic forces and structural motion, commonly in vertical bending and torsion. Vortex shedding can produce lock-in when shedding frequency approaches a structural frequency. The approximate shedding frequency is

\[

f_s=St\frac{V}{D},

\]

where \(f_s\) is shedding frequency, \(St\) is Strouhal number, \(V\) is wind speed and \(D\) is a characteristic dimension. The relation is an idealized estimate for a defined body and flow regime; it is not a substitute for bridge-specific aerodynamic testing.

Mitigation includes streamlined deck sections, fairings, guide vanes, spoilers, tuned mass dampers, increased torsional stiffness, aerodynamic openings and control of construction stages. Aerodynamic stability must be checked for the completed bridge and every critical temporary configuration, including partially erected decks, lifting operations and incomplete bracing.

4. Flood, Scour and Hydraulic Extreme Events

Flood damage commonly results from a combination of hydraulic force, debris impact, scour, buoyancy, sediment transport, bank erosion and access failure. Scour is often the governing hazard for bridge collapse during floods.

A rational evaluation distinguishes:

• long-term degradation of the river bed;
• contraction scour caused by reduced flow area at the crossing;
• local scour around piers and abutments; and
• pier or abutment instability caused by erosion and lateral migration.

Scour depth is not a universal constant and depends on sediment size, flow intensity, pier geometry, bedforms, duration and foundation arrangement. Empirical equations require calibration to their stated ranges and should not be extrapolated blindly. Foundations should extend below credible scour levels or be protected with appropriately designed countermeasures such as riprap, grout, guide banks or sacrificial elements.

Hydraulic assessment should use design hydrographs, not only peak discharge where duration matters. Debris may block openings and amplify local scour. Inspection plans should include post-flood sounding, underwater assessment, monitoring of bed elevation and trigger levels for closure. Climate change may alter rainfall intensity, flood frequency, sediment regimes and sea levels; adaptation should therefore consider nonstationarity and the possibility of future intervention.

5. Fire, Blast, Impact and Landslides

5.1 Fire

Fire can reduce concrete strength, cause explosive spalling, weaken prestressing steel and produce large thermal gradients. A bridge fire assessment should consider fuel type, duration, ventilation, shielding, member restraint and heat transfer. Protection may include increased cover, fire-resistant materials, sacrificial barriers, drainage of flammable liquids and separation of fuel routes from critical members. Post-fire evaluation requires visual inspection, temperature history where available, rebound or ultrasonic testing used cautiously, and direct sampling when structural decisions depend on residual properties.

5.2 Blast

Blast effects include reflected pressure, impulse, fragmentation, progressive collapse and damage to bearings or diaphragms. Structural design for blast is normally risk-based and security-sensitive. Measures include standoff distance, vehicle barriers, sacrificial components, continuity, compartmentalization and detailing for alternate load paths. A single pressure value is inadequate unless the explosive characteristics, stand-off, geometry and venting assumptions are specified.

5.3 Vehicle, vessel and train impact

Impact demand depends on mass, speed, contact geometry, direction, deformability and probability of occurrence. Protection can be provided by fenders, dolphins, barriers, redundant pier arrangements, collision-resistant geometry and navigation controls. Impact design should address both direct member damage and redistribution after local failure. Inspection and monitoring are particularly important because apparently minor impact can damage reinforcement, bearings or foundations.

5.4 Landslides and rockfall

Landslides may impose lateral soil pressure, foundation displacement, debris impact and loss of approach embankments. Risk reduction begins with geomorphological mapping, drainage control, slope stabilization and avoidance of unstable alignments. Stabilization may involve retaining systems, anchors, piles, buttresses, surface-water management and debris barriers. Because ground movement may be progressive, instrumentation and trigger-action response plans are valuable complements to structural resistance.

6. Robustness, Redundancy and Resilience

Redundancy provides multiple load paths so that local damage need not cause system collapse. Robustness is the ability to resist disproportionate consequences from localized damage. Resilience additionally includes preparation, absorption, recovery and adaptation.

A robust bridge system generally has:

• continuity between deck, bearings, piers and foundations;
• ductile and inspectable connections;
• alternate vertical and lateral load paths;
• protection of critical nodes and utilities;
• compartmentalization of hazards;
• tolerable damage states and replaceable components; and
• emergency access and maintainable monitoring systems.

Redundancy is not automatically beneficial. Additional members can increase mass, seismic demand, aerodynamic complexity and maintenance burden. A rational design identifies credible initiating damage, assesses alternate-path capacity and checks dynamic amplification during sudden load redistribution. Nonlinear alternate-path analysis may be necessary for critical bridges.

Resilience metrics may include downtime, repair cost, functionality retained, safety risk and recovery time. A network perspective is essential: a bridge that remains structurally sound may still be unusable if approaches, power, communications or adjacent crossings fail.

7. Integrated Assessment and Design Workflow

A professional extreme-event study can follow this sequence:

1. Define consequences and performance objectives. Identify users, emergency routes, lifeline functions, acceptable downtime and regulatory requirements.

2. Characterize hazards. Establish seismic spectra and records, wind climate, flood frequency and scour, fire scenarios, impact threats, blast scenarios and slope hazards.

3. Develop the system model. Include superstructure, substructure, foundations, bearings, joints, soil, water and relevant neighboring assets.

4. Screen vulnerabilities. Prioritize irregularity, poor detailing, unseated spans, weak foundations, scour exposure, aerodynamic sensitivity and nonductile connections.

5. Select analysis methods. Use preliminary hand calculations and linear models for screening; use nonlinear static, dynamic, fluid–structure, soil–structure or probabilistic models where warranted.

6. Design or retrofit. Apply ductility, capacity design, isolation, damping, confinement, scour protection, barriers, redundancy and monitoring.

7. Check constructability and temporary states. Many failures occur during erection, diversion, demolition or repair.

8. Plan inspection and recovery. Define sensors, trigger levels, closure criteria, rapid inspection procedures, spare components and repair sequences.

9. Verify and document. Record assumptions, uncertainties, code provisions, model validation, residual risk and decisions requiring authority approval.

8. Illustrative Applied Case: Seismic Retrofit of a Multi-Span Viaduct

Background

Consider an illustrative, fictional six-span reinforced-concrete viaduct built before modern seismic detailing requirements. It carries an emergency route across an alluvial valley. The deck is continuous, supported on slender single-column piers with older reinforcement details. Elastomeric bearings have limited transverse restraint, and geotechnical investigation indicates potential liquefaction and lateral spreading near one abutment.

Problem

A preliminary assessment identifies inadequate confinement in potential plastic-hinge regions, insufficient transverse displacement capacity at bearings, vulnerable shear keys and uncertain pile performance. A conventional elastic strengthening scheme would increase pier stiffness and force demand without resolving foundation vulnerability.

Analysis

A three-dimensional model is developed with nonlinear pier hinges, bearing properties, abutment interaction and simplified soil springs. Multiple ground-motion records are selected to represent the site hazard and scaled consistently with the governing design spectrum. Sensitivity analyses vary soil stiffness, friction at bearings and abutment participation. The governing response is controlled by transverse displacement and lateral spreading rather than by elastic member force alone.

Approach and solution

The retrofit adopts a system strategy:

• confinement jackets are installed at selected pier ends;
• shear-critical regions and anchorage are strengthened so that flexural yielding precedes brittle failure;
• seismic isolation bearings and supplemental restrainers are introduced after checking thermal and construction movements;
• seat extensions and positive connection details reduce unseating risk;
• piles near the spreading zone are strengthened and the approach ground is improved;
• a scour and liquefaction monitoring plan is added;
• post-earthquake inspection access and replaceable bearing details are incorporated.

Outcome

Nonlinear analysis indicates that the intended plastic mechanism forms in confined pier regions while shear, bearings and foundations retain compatible capacity. Residual displacement is reduced by isolation, although joint movement and utility flexibility remain design constraints. The project is accepted subject to staged construction, device testing, post-event inspection and a documented closure protocol.

Learning points

The case demonstrates that seismic retrofit is not merely a member-strengthening exercise. The governing vulnerability may lie in bearings, soil, approaches or interfaces. Performance objectives, uncertainty studies and recovery planning are as important as calculated peak force.

9. Limitations, Governance and Contemporary Practice

Hazard models contain epistemic uncertainty from incomplete data and aleatory variability from event-to-event randomness. Nonlinear simulations may appear precise while depending strongly on constitutive models, damping assumptions, record selection, boundary conditions and component degradation. Independent review, sensitivity analysis and transparent documentation are therefore essential.

Design must comply with the applicable national bridge, seismic, wind, hydraulic, fire and occupational-safety regulations. Authorities should define approval responsibilities, inspection intervals, closure powers, emergency procurement procedures and public communication protocols. Ethical practice requires that uncertainty and residual risk be communicated honestly, especially where retrofit budgets force prioritization among bridges.

Digital twins, structural-health monitoring, remote sensing, satellite deformation measurement, drones and machine-learning damage detection can improve surveillance. AI can assist anomaly detection, image classification, traffic disruption prediction and prioritization, but it should not replace engineering judgment. Training-data bias, false negatives, cybersecurity, sensor failure and explainability must be addressed. Human verification remains necessary for safety-critical decisions.

Best practice is consequently risk-informed and lifecycle-oriented: design for controlled damage, inspectability and repair; protect critical interfaces; use validated analysis; consider cascading hazards and climate change; rehearse emergency actions; and update assessments as monitoring and event data become available.

Chapter Summary

Extreme-event bridge engineering addresses hazards whose probability may be low but whose consequences are severe. Sound assessment combines hazard characterization, vulnerability modelling, consequence evaluation and explicit performance objectives. Seismic design emphasizes ductility, capacity hierarchy, confinement, displacement compatibility, isolation, damping and foundation performance. Wind design must distinguish static drag from buffeting, vortex shedding and aeroelastic instability, especially for flexible or long-span bridges. Flood resilience requires serious attention to scour, debris, sediment movement and post-flood inspection. Fire, blast, impact and landslide design require scenario-based assessment, protection of critical components and alternate load paths.

Redundancy, robustness and resilience extend the objective from preventing member failure to preserving system function and enabling recovery. Reliable practice integrates structural, geotechnical, hydraulic, aerodynamic, construction and emergency-management decisions. Advanced tools such as monitoring, digital twins and AI are useful when validated and governed appropriately, but they supplement rather than replace professional responsibility.

Key Terms

Hazard: A potentially damaging natural or human-induced event.

Vulnerability: Susceptibility of a bridge to damage under a specified hazard.

Performance objective: A required level of functionality or damage for a defined hazard.

Ductility: Capacity for inelastic deformation while maintaining resistance.

Capacity design: Design method that protects selected components by forcing yielding into intended ductile regions.

Plastic hinge: A localized region capable of sustaining inelastic rotation.

Displacement ductility: Maximum displacement divided by yield displacement.

Seismic isolation: Separation of structural response from ground motion using flexible and dissipative devices.

Supplemental damping: Added energy dissipation provided by mechanical or hysteretic devices.

Pounding: Collision between adjacent structural components during dynamic movement.

Aeroelastic instability: Self-excited structural response produced by interaction with airflow.

Vortex shedding: Alternating vortices generated as flow passes a body.

Scour: Erosion of bed or bank material around hydraulic structures.

Liquefaction: Loss of soil stiffness and strength caused by cyclic loading and excess pore pressure.

Robustness: Ability to limit disproportionate consequences from local damage.

Redundancy: Provision of multiple load paths or alternative resisting components.

Resilience: Capacity to prepare for, withstand, recover from and adapt to disruption.

Progressive collapse: Disproportionate spread of failure from local damage.

Risk-informed design: Decision-making based on probabilities, consequences and uncertainty.

Structural-health monitoring: Systematic measurement and interpretation of structural condition.

Review Questions

Short Answer Questions

1. Distinguish hazard, vulnerability and risk in bridge engineering.

2. Why is ductility insufficient unless brittle shear and connection failures are prevented?

3. State two important limitations of a response spectrum.

4. What is local scour, and why can it cause bridge collapse?

5. Distinguish redundancy from resilience.

Long Answer Questions

1. Explain capacity design for a reinforced-concrete bridge pier, including the intended hierarchy of failure.

2. Discuss the principles, benefits and limitations of seismic isolation and supplemental damping.

3. Explain the principal wind-induced mechanisms affecting long-span bridges and appropriate mitigation measures.

4. Develop an integrated assessment procedure for a bridge exposed to flood, scour, debris impact and climate change.

5. Discuss how robustness, inspection, monitoring and emergency planning should be incorporated into bridge lifecycle management.

Analytical / Application Questions

1. A bridge has mass \(M=2.0\times10^6\,kg\) and effective lateral stiffness \(K_e=8.0\times10^6\,N/m\). Estimate its isolated period using \(T_e=2\pi\sqrt{M/K_e}\), and state the assumptions behind the calculation.

2. A pier is exposed to wind with \(\rho=1.225\,kg/m^3\), \(V=30\,m/s\), \(C_D=1.2\) and projected area \(A=40\,m^2\). Calculate the quasi-steady drag force and identify why this result may be inadequate for a flexible bridge.

3. Prepare a prioritized retrofit strategy for a bridge with poor seismic detailing, liquefaction-prone foundations, inadequate seat width and severe flood-scour exposure. Justify the sequence using performance, constructability and recovery criteria.

Multiple Choice Questions with Answers

Question 1

Which statement best describes capacity design in seismic bridge engineering?

A. It forces all components to remain elastic under every earthquake.

B. It concentrates inelastic action in selected ductile regions and protects other components from brittle failure.

C. It eliminates the need to design foundations for seismic actions.

D. It replaces displacement checks with wind-load checks.

Correct Answer: B. It concentrates inelastic action in selected ductile regions and protects other components from brittle failure.

Question 2

For a single-degree-of-freedom model, which expression gives the natural circular frequency?

A. ωn = k/m

B. ωn = m/k

C. ωn = √(k/m)

D. ωn = √(m/k)

Correct Answer: C. ωn = √(k/m)

Question 3

A primary design concern introduced by seismic isolation is:

A. The complete elimination of all bridge displacement.

B. The need to accommodate larger movements at joints and connections.

C. The removal of all foundation demands.

D. The prevention of thermal movement by definition.

Correct Answer: B. The need to accommodate larger movements at joints and connections.

Question 4

The approximate vortex-shedding relation is:

A. fs = StV/D

B. fs = StD/V

C. fs = VD/St

D. fs = V/(StD)

Correct Answer: A. fs = StV/D

Question 5

Which mechanism is most directly associated with loss of river-bed material around a pier?

A. Flutter

B. Pounding

C. Local scour

D. Low-cycle fatigue

Correct Answer: C. Local scour

Question 6

A bridge drag estimate uses ρ = 1.225 kg/m³, V = 20 m/s, CD = 1.0 and A = 10 m². What is the estimated drag force?

A. 0.245 kN

B. 2.45 kN

C. 24.5 kN

D. 245 kN

Correct Answer: C. 24.5 kN

Question 7

Which condition most directly threatens pile foundations during earthquake-induced lateral spreading?

A. Increased paint deterioration

B. Large permanent ground displacement and lateral soil demand

C. Reduced aerodynamic drag

D. Improved bearing confinement

Correct Answer: B. Large permanent ground displacement and lateral soil demand

Question 8

Robustness primarily refers to a bridge’s ability to:

A. Limit disproportionate consequences from localized damage.

B. Increase traffic speed during normal operation.

C. Avoid all maintenance activities.

D. Convert wind loads into seismic loads.

Correct Answer: A. Limit disproportionate consequences from localized damage.

Question 9

For M = 2.0×10^6 kg and Ke = 8.0×10^6 N/m, the approximate isolated period is:

A. 0.10 s

B. 0.50 s

C. 3.14 s

D. 6.28 s

Correct Answer: C. 3.14 s

Question 10

Which is an appropriate limitation on using AI for bridge safety decisions?

A. AI outputs never require validation.

B. Training-data bias and false negatives can affect reliability.

C. AI cannot process sensor data.

D. AI automatically establishes legal design requirements.

Correct Answer: B. Training-data bias and false negatives can affect reliability.