Life-Cycle Risk, Resilience, and Sustainability of Individual and Spatially Distributed Structures

Abstract

Field investigations after recent large earthquakes have confirmed that several structures were severely damaged and collapsed not only by the earthquake, but also by the subsequent tsunami, landslide, or fault displacement. Effect of material degradation due to chloride attack on structural performance should be considered when structures are located in a harsh environment. In addition, climate change has produced typhoons and hurricanes with extreme intensity in recent years. Sea-level rise could cause severe storm surges and tsunamis, and global warming is accelerating the deterioration of structures. When structures are exposed to these different types of hazards, it can be difficult to ensure their safety and additional performance indicators such as risk and resilience are needed. Several lessons were learned about the importance of investigating individual structures from the perspective of ensuring network functionality. A probabilistic life-cycle framework for quantifying the loss of functionality of road networks including bridges is needed. A risk-based decision-making approach at the network level is required to identify the dominant hazard and the vulnerable structures that require strengthening and retrofitting. After a catastrophic event, the functionality of transportation networks can be significantly degraded, resulting in catastrophic economic impacts. To quantify the promptness of recovery, it has become common to use the concept of resilience. In addition, the economic, environmental, and social impacts of disaster waste management systems need to be examined in terms of sustainability. Consequences related to resilience and sustainability need to be investigated and implemented in the risk assessment of road networks under multiple hazards. Life-cycle design and assessment methodologies can incorporate risk, resilience, sustainability and multiple hazards, learning from the lessons of past disasters. This keynote paper provides an overview of measures to ensure the functionality of individual and spatially distributed structures under multiple hazards from the perspectives of reliability, risk, resilience and sustainability.

1 Introduction

As observed in recent disasters in Japan (e.g., 2011 Great East Japan earthquake, 2016 Kumamoto earthquake, and 2024 Nota-Hanto earthquake), since transportation networks, including bridges, play a crucial role in evacuating affected people and transporting relief goods and materials, the functionality of the network after a disaster needs to be investigated. A significant amount of research has shifted the focus from investigating the performance of individual infrastructure components to that of entire distributed civil infrastructure systems and networks [1]. Rapid recovery of critical infrastructure system functionality after an extreme event is always a goal of paramount importance [2, 3].

To quantify the speed of recovery, it has become common to use the concept of resilience. Resilience emphasizes the impact of infrastructure damage, failure, and societal recovery under low-probability and high-consequence hazards. Sustainability, on the other hand, focuses on current and future resource management and addresses the impacts of planning and development on the economy, society, and the environment. A sustainable infrastructure system must incorporate resilience and adaptability to ensure its long-term sustainability. In fact, some of the most promising solutions for resilience are also sustainable in nature [4].

The design and assessment of structures has focused on addressing the most dominant hazard at the site of interest. However, the possibility that structures may experience multiple hazards of different types during their lifetime needs to be considered. Bridge design methodology needs to shift to a more comprehensive approach of addressing multiple hazards to ensure adequate performance under different mechanical and environmental scenarios [5]. Quantifying the reliability and risk of each bridge under multiple hazards can help prioritize retrofit activities for bridges in a network.

Significant progress has been made in the field of earthquake engineering. However, further research is needed to develop concepts and methods for designing and evaluating resilient and sustainable bridges and bridge networks in a life-cycle context. This paper provides an overview of life-cycle design and assessment methodologies for bridges under multiple hazards, with emphasis on independent and interacting hazards, based on lessons learned from recent major earthquakes in Japan. Several performance indicators that need to be implemented in practical design and assessment are presented. Finally, case studies are used to illustrate the concepts and methods presented.

2 Lessons from Recent Large Earthquakes in Japan

Figure 1 shows several destructive earthquakes that have occurred in Japan since the 1995 Kobe earthquake. The lessons learned from these earthquakes indicate that in addition to seismic safety, other performance indicators of individual bridges and bridge networks need to be considered. The 2024 Noto-Hanto earthquake, as well as the 2011 Great East Japan earthquake, caused structural damage due to strong excitation, tsunami, landslides, and liquefaction. The 2024 Noto-Hanto earthquake demonstrated that an earthquake is a source of multiple hazards which can cause a variety of adverse effects on structures and infrastructure systems. In this section, several types of damage are reported, mainly from the results of the field investigation conducted after the 2024 Noto-Hanto earthquake.

Fig. 1.

Large earthquakes in Japan after the 1995 Kobe earthquake.

Compared to the ground motion- and tsunami-induced damages to bridges during the 2011 Great East Japan earthquake, the number of bridges affected by the 2024 Noto-Hanto earthquake was limited. Figure 2 shows the damage to Ukai Bridge with three simple skewed girders constructed in 1960. Total length of Ukai Bridge is 54 m. Three simple girders for pedestrians were parallel to the decks for vehicles. As shown in Fig. 2, two of the pedestrian girders collapsed, which may be caused by the deck rotation seismic response of skewed decks for vehicles. However, one of them was washed away by the tsunami as shown in Fig. 3. It was displaced more than approximately 50 m from its original position. Tsunami waves created water pressures due to the impulse of breaking waves, and dynamic pressures that varied with wave velocity and height [6]. It is sufficient force to blow away even concrete members of considerable weight. The damage to the Ukai Bridge shown in Figs. 2 and 3 indicates the need for countermeasures that take into account the effects of both strong ground motions and tsunamis. However, as shown in Fig. 3, it is extremely difficult with current bridge technology to prevent damage to bridges from a tsunami strong enough to blow away the superstructure. Therefore, it is important to avoid placing bridges in high tsunami hazard areas or to create routes that pass through areas without high tsunami hazard to ensure the functionality of the network in terms of redundancy.

Around the Ukai Bridge, there were many old wooden houses that were built before seismic design methods were established in Japan. They did not appear to have been seismically retrofitted. As shown in Fig. 4, many of them were severely damaged by earthquake excitation, tsunami, and/or liquefaction. The Noto Peninsula experienced an earthquake with the moment magnitude of 6.7 in 2007, which damaged many wooden houses. Since some of them were not repaired and retrofitted, the 2024 Noto-Hanto earthquake may have caused further damage. In addition, water and sewage pipes were seriously damaged in many areas of the Noto Peninsula. Figure 4 shows a manhole that has popped out due to liquefaction. Although it is difficult to seismically strengthen underground water and sewer pipes, the 2024 Noto-Hanto earthquake demonstrates the importance of retrofitting them to ensure the continued use of tap water and toilets, which are essential to the daily lives of people in the affected areas.

Fig. 2.

Sidewalk collapse of Ukai Bridge subjected to ground motion and/or tsunami caused by 2024 Noto-Hanto earthquake. Note that the picture was taken by the second author.

Fig. 3.

Washout of Ukai Bridge sidewalk due to tsunami caused by the 2024 Noto-Hanto earthquake. Note that the picture was taken by the second author.

Fig. 4.

Collapse of classical wooden Japanese houses due to ground motion, tsunami and/or liquefaction caused by the 2024 Noto-Hanto earthquake. Note that the picture was taken by the second author.

Fig. 5.

Bridge redeterioration due to chloride attack after repair near Noto Penisula (Note that the picture was taken in 2017 by the second author)

The effect of corrosion on the deterioration of bridge capacity under seismic hazard must be considered. The Noto Peninsula faces the Sea of Japan and is known as one of the harshest environments in Japan due to airborne chloride. During the winter season, the wind mostly blows from the west (i.e., Japan Sea), bringing airborne chloride. Many concrete bridges deteriorate due to steel corrosion. Recently, some of the bridges that were repaired to remove chloride in the concrete and replace the cover concrete with corrosion cracks are deteriorating again, as shown in Fig. 5. As discussed in Sect. 3, it is important to understand that the seismic demand depends on the seismic hazard, while the seismic capacity depends on the other hazard, such as the hazard associated with airborne chlorides [7]. The seismic performance of existing bridges in a harsh environment cannot be expected to be the same as that at the time of construction [8].

Fig. 6.

Tsunami damage in Suzu City and disaster waste generated by the 2024 Noto-Hanto earthquake. Note that the picture was taken by the second author.

In Japan, following the tsunami caused by the 2011 Great East Japan earthquake, large areas of farmland were flooded with salty water and contaminated with marine sediments, resulting in long-term soil contamination of highly fertile agricultural land with metals and metalloid compounds [9]. As a result of the earthquake and subsequent tsunami, approximately 23 million tons of disaster debris were generated, with more than 12 million m³ of tsunami deposits remaining in the flooded area. The amount of disaster waste generated by the 2024 Noto-Hanto earthquake is limited compared to that generated by the 2011 Great East Japan earthquake, but it must prevent the affected region from recovering from the disaster. Figure 6 shows an example of the disaster waste generated by the 2024 Noto-Hanto earthquake.

The structural and geotechnical utilization of the concrete and soil fraction in the disaster debris and tsunami deposits has posed a great challenge to engineers because (a) the removal of the debris and tsunami deposits is an urgent task that must be completed within a few years [10], and (b) although a large amount of waste concrete and soil can be recycled and used in reconstruction, their properties have temporal and spatial variations [11]. If poorly managed, these wastes can have significant environmental and public health impacts that can affect the overall recovery process and undermine sustainability [12].

3 Toward Life-Cycle Based Design and Assessment of Bridges and Bridge Networks Under Multiple Hazards

3.1 Progress of Structural Performance Methodology and Associated Performance Indicators

Figure 7 shows the evolution of structural design methodology from deterministic allowable stress design to life-cycle based design and assessment of transportation networks that include bridges. Traditionally, structural safety in design is quantified by comparing the structural capacity to the load. In the allowable stress design method, the designer simply ensures that the structural components have a fraction of the elastic limit state for the service loads. Static and linear elastic analyses are performed to estimate the demand at the component level. With the development of computer technology and computer simulation capability, and with the lessons learned from disasters, structural design methodology has progressed to include consequences of structural failure, various performance indicators, and life-cycle concepts for bridges and bridge networks under multiple hazards. More details of these advances are reviewed in [13].

Fig. 7.

Progress of structural design methodology: from the classical allowable stress design method toward the life-cycle based design and assessment of network involving bridges under multiple hazards.

It is not feasible to design a bridge that will remain intact under all hazards that may affect its performance. Under excessive interacting hazards on bridges, such as seismic and tsunami hazards, and seismic and landslide hazards, it is quite difficult to identify the solution in terms of structural control to prevent the failure of bridges with damage due to strong ground motion under the cascading giant tsunami or huge landslide. The technology may not exist to increase the structural ductility and integrity of bridges against damage and collapse, even if additional requirements beyond those provided in current structural codes are required. Therefore, the concept of risk and resilience should be introduced. Before presenting studies that investigate risk, resilience and sustainability through quantitative approaches and application to case studies of bridge structures as described later, an idea of the contribution to risk reduction and resilience that comes from structural control is provided herein.

To maximize the post-event operability of bridges against extremely large earthquake excitations, the novel RC bridge pier with the sliding pendulum system has been proposed [14,15,16,17]. As a sliding pendulum, the upper component of the bridge pier moves on a sliding surface of the lower component of the bridge pier under strong excitation. In order to achieve cost-benefit, no flexible isolator layers were included in the bridge pier to extend the natural period; only conventional concrete and steel were used in addition to 3D printing technology. Computational and experimental investigations demonstrated that a damage-free bridge pier could be achieved under strong earthquakes.

Fig. 8.

Example of robust structures against catastrophic hazards (a) and (b) taken after the 2016 Kumamoto earthquake; and (c) taken after the reconstruction of the abutment. Note that all pictures were taken by the second author.

Fig. 9.

Washout of superstructure due to the 2020 Kyushu flood and reconstruction of temporary bridge using the survived bridge piers in Kuma River. Note that the picture was taken by the second author.

Compared with the conventional girder bridges, a rigid frame bridge has the potential of being a robust and resilient structure against catastrophic actions such as fault displacement, huge landslide, and hydrodynamic forces associated with tsunami and flood. Since the rigid frame structures do not have bearings between the superstructure and substructure, they could prevent the washout of the superstructure due to the tsunami attack compared with the conventional girder bridge. Figures 8(a) and (b) show that the abutment of the Aso-Choyo Bridge could not support the PC box girder after it was transversely displaced by the landslide during the 2016 Kumamoto earthquake. However, since both the superstructure and substructure were rigidly connected and behaved as a continuous unit, the severe damage to the Aso-Choyo Bridge due to the landslide was not observed. Only after the reconstruction of the abutment, as shown in Fig. 8(c), the Aso-Choyo Bridge became functional and could contribute to the disaster recovery of the affected region.

Considering the possibility that many bridge superstructures would be washed away during a future tsunami or flood event, solid technologies must be developed to enhance the disaster resilience of communities under seismic or flood hazards [13]. In Japan, since the structural details of bridge substructures (i.e., bridge piers and foundations) have been determined to ensure sufficient seismic performance, they could survive hydrodynamic attacks, as shown in Fig. 9. Surviving substructures were reused to construct a temporary bridge superstructure after, for example, the 2011 Great East Japan earthquake and the 2020 Kyushu flood, as shown in Fig. 9. A technology for the rapid construction of temporary bridge superstructures is needed to ensure rapid restoration and improve disaster resilience. Accelerated bridge construction (ABC) is an important research topic for this purpose. ABC using prefabricated elements saves on-site construction time and improves work zone safety [18].

3.2 Multiple Hazard Issues

A strong earthquake could cause multiple disasters, including damage to structures due to strong ground motions and/or liquefaction, and washout of structures due to subsequent tsunamis and landslides. In addition, seismic performance would be degraded due to material corrosion, fatigue, and scour caused by the flood, among others. Comparing the life-cycle reliability and risk of structures in a network under multiple hazards is useful for identifying significant hazard scenarios. A discussion of various design and analysis aspects for bridges under multiple hazards in a life-cycle context is provided herein.

Independent Hazards

Different types of hazards such as independent hazards, correlated hazards, concurrent hazards, and cascading hazards have been studied in the literature. The seismic reliability of corroded bridges or bridges damaged by flood scour is an example of the independent hazard cases. Figure 10 shows a general seismic fragility analysis procedure that considers the effect of steel corrosion on the deterioration of RC components. However, it’s quite difficult to apply the procedure to evaluate the seismic capacities of corroded RC columns for developing the numerical fragility assessment. Since experimental results on spatial cross-sectional area loss of corroded steel and bond strength between concrete and corrosion product and between corrosion product and uncorroded rebar are still scarce [19,20,21,22], model error associated with seismic capacity prediction is quite large. Further experimental and computational research is needed to develop the numerical model for the seismic fragility of corroded piers or bridges with scour-induced damage.

Fig. 10.

Procedure of seismic fragility analysis considering effect of chloride-induced steel corrosion on deterioration of RC components.

Interacting Hazards

To estimate the reliability of structures under both seismic and tsunami hazards, or mainshock and aftershock as an example of interacting hazards, the structural vulnerability to the subsequent action needs to be evaluated considering the effects of ground motion-induced damage on the reduction of structural capacity. Figure 11 shows a damage sequence of piers subjected to earthquake excitation and subsequent tsunami. The residual displacement and the stiffness and strength degradation due to the ground motion are considered as the initial conditions when performing the pushover analyses to develop the fragility curves. Bridges become more vulnerable to tsunami due to ground motion-induced damage [13, 23].

Fig. 11.

Damage process of piers subjected to earthquake excitation and subsequent tsunami.

Climate Change Effects

In the wake of recent catastrophes exacerbated by climate change effects, the impact of climate change on civil infrastructure has received increasing attention around the world. For hazards intensified by the effects of climate change, predicting future conditions is significantly complicated because future climate data derived from general circulation models (i.e., GCMs), such as future temperature and precipitation, cannot be easily related to structural demand and capacity under future hazards. The lack of reliable climate models means that previous studies cannot adequately quantify the increasing risk to civil infrastructure under the impact of climate change.

Fig. 12.

Structures and infrastructure systems exposed to multiple hazards and climante change

By comparing risk and resilience with and without climate change, the threat can be quantified. Figure 12 illustrates a schematic approach for estimating the reliability, risk, and resilience of bridges. This framework requires individual techniques such as hydrological modeling using geographic information, reliability assessment of bridges under flood and scour hazards considering climate change, impact of higher temperature, humidity and CO₂ concentration effects on the deterioration of concrete structures, and functionality loss assessment of road network. In addition, the issue of climate non-stationarity and its impact on climate loads, load combinations, and structural reliability should be addressed.

4 Illustrative Examples

Several case studies presented in the literature by the authors and their co-workers are introduced herein to illustrate how to ensure life-cycle reliability, risk, and resilience of bridges and bridge networks under multiple hazards and climate change impacts.

Fig. 13.

Life-cycle reliability of RC bridge piers under seismic and airborne chloride hazards in case of Uwajima and Sakata Cities [13].

Figure 13 shows the relationship between the cumulative-time failure probability and the time after construction of bridge piers in Sakata and Uwajima Cities [13]. To investigate the effect of corrosion on the failure probability, the cumulative-time failure probabilities without deterioration are also shown in Fig. 13. The geometry and structural details of the RC bridge pier in the two cities are the same. Initially, the failure probability of the bridge piers depends only on the seismic hazard, and the RC bridge pier located in Uwajima City has the higher failure probability. However, the probability of exceeding a prescribed amount of airborne chloride in Sakata City is much higher. The cumulative-time failure probability of the RC bridge pier in Sakata City increases with time due to chloride attack. The difference between the cumulative-time failure probability associated with both seismic and airborne chloride hazards and that associated with seismic hazard alone is greater for Sakata City. Finally, at 50 years after construction, the cumulative-time failure probability of the RC bridge pier in Sakata City is higher, even though the seismicity in Sakata City is lower than that in Uwajima City. To ensure the seismic reliability for the entire lifetime of the RC bridge pier in a marine environment, it is extremely important to consider the effect of corrosion on the seismic capacity and demand.

Fig. 14.

Spatially correlated seismic and tsunami hazards [24].

When assessing the probabilistic connectivity of a large road network exposed to both ground motion and tsunami, it is important to consider the spatial correlations between hazard intensities, as shown in Fig. 14 [24]. To evaluate the joint probabilities of bridge states (i.e., passable and impassable states) and the probabilistic connectivity of road networks, the total probability theorem is used to integrate spatially correlated seismic and tsunami hazard assessments into the fragility estimates that consider the cascading effects of ground motion- and tsunami-induced damage to bridges.

As discussed in Sect. 2, the amount of disaster waste is one of the most important performance indicators for quantifying a community’s resilience. Disaster waste can have a significant negative impact on the environment in the affected regions and hinder the post-disaster recovery process. Adequate disaster waste management should be developed in Japan prior to the occurrence of the Nankai Trough earthquake. The seismic and tsunami intensities caused by the anticipated Nankai Trough earthquake are expected to be significantly greater than those caused by the 2011 Great East Japan earthquake. Figure 15 illustrates the risk curves associated with disaster waste from buildings collapsed by ground motion and tsunami due to the anticipated Nankai Trough earthquake [10]. The risk of disaster waste depends on the number and location of structures and the intensity of the earthquake and tsunami. In the areas, such as Kuwana City, where the tsunami inundation depth is negligible, the risk curve is determined by the seismic effect only. Meanwhile, in the areas such as Owase City where the tsunami hazard is severe, the amount of disaster waste is underestimated if the effects of both seismic and tsunami impacts are not considered.

Fig. 15.

Risk curves associated with disaster waste of buildings collapsed by ground motion and tsunami due to the anticipated Nankai Trough earthquake [10]

Fig. 16.

Example of retrofitting prioritization of bridges in a road network (a) under seismic and tsunami hazards; and (b) seismic and flood hazards

The results of this case study show that a large amount of disaster waste would be generated in Mie Prefecture by the Nankai Trough earthquake with a high exceedance probability. Countermeasures to strengthen disposal capacity and facilitate cooperation with other prefectures before the event are important to reduce the adverse effects of disaster waste.

Under budget constraints, it is important to efficiently identify bridges that should be retrofitted. When they are exposed to multiple hazards, this process becomes more complicated. There are several performance indicators that can be used to prioritize retrofitting, including the reliability of individual bridges, the probabilistic connectivity of a road network, and risk and resilience considering the corresponding consequences. Figures 16(a) and (b) shows an example of performance indicator calculations to prioritize bridges in a road network under seismic and tsunami hazards as the interacting hazards [23], and under seismic and flood hazards as the independent hazards [25], respectively.

Fig. 17.

Tsunami inundation maps without/with sea-level rise [29]

Climate change and the resulting more extreme weather events are likely to impact structural performance. This phenomenon can pose direct threats to infrastructure systems, as well as significant indirect impacts on those who rely on the services these assets provide. Such threats are path- and location-dependent, as they are highly dependent on current and future climate variability, location, asset design life, function, and condition [26]. Framework for probabilistic tsunami hazard assessment considering the effects of sea-level rise due to climate change was presented in [27]. It was applied to a region where the ground motion and tsunami intensities would be extremely large during the anticipated Nankai Trough earthquake [28,29,30]. Figure 17 shows an example of tsunami inundation maps without and with sea-level rise [29].

5 Conclusions

In the aftermath of a catastrophic event, the functionality of individual and spatially distributed structures such as transportation networks can be significantly degraded, resulting in catastrophic economic impacts. The concept of resilience is needed to quantify the time required for recovery. In addition, the economic, environmental, and social impacts of disaster waste management systems need to be studied in terms of sustainability. Consequences related to resilience and sustainability need to be investigated and implemented in the risk assessment of the bridge network under multiple hazards. The life-cycle design and assessment methodology can incorporate all the key concepts such as risk, resilience, sustainability, and multiple hazards, learning from the lessons of past disasters.

Finally, the concepts and methods presented were illustrated on both single bridges and bridge networks with emphasis on earthquake, tsunami, flooding, and continuous deterioration in addition to climate change. A risk-based decision-making approach at the network level is required to identify the dominant hazard and the vulnerable structures that require strengthening and retrofitting.