Earthquakes injure millions and simultaneously disrupt the infrastructure to protect them. This perspective argues that the current post-disaster investigation paradigm is insufficient to protect communities’ health effectively. We propose the Earthquake Survival Chain as a framework to change the current engineering focus on infrastructure to health. This framework highlights four converging research opportunities to advance understanding of earthquake injuries, search and rescue, patient mobilizations, and medical treatment. We offer an interdisciplinary research agenda in engineering and health sciences, including artificial intelligence and virtual reality, to protect health and life from earthquakes.
Efforts have been made in the past to enhance building exposure models on a regional scale with increasing spatial resolutions by integrating different data sources. This work follows a similar path and focuses on the downscaling of the existing SARA exposure model that was proposed for the residential building stock of the communes of Valparaíso and Viña del Mar (Chile). Although this model allowed great progress in harmonising building classes and characterising their differential physical vulnerabilities, it is now outdated, and in any case, it is spatially aggregated over large administrative units. Hence, to more accurately consider the impact of future earthquakes on these cities, it is necessary to employ more reliable exposure models. For such a purpose, we propose updating this existing model through a Bayesian approach by integrating ancillary data that has been made increasingly available from Volunteering Geo-Information (VGI) activities. Its spatial representation is also optimised in higher resolution aggregation units that avoid the inconvenience of having incomplete building-by-building footprints. A worst-case earthquake scenario is presented to calculate direct economic losses and highlight the degree of uncertainty imposed by exposure models in comparison with other parameters used to generate the seismic ground motions within a sensitivity analysis. This example study shows the great potential of using increasingly available VGI to update worldwide building exposure models as well as its importance in scenario-based seismic risk assessment.
In an earthquake emergency, injured people require hospital emergency services and timely medical treatment. Earthquake-related patients often have trauma injuries and stress-linked (ischemic) illnesses that require multiple healthcare procedures, such as minor orthopedic treatment, surgical treatment of fractures, and thrombolysis or thrombectomy. Hospital operation models have been proposed to examine these healthcare procedures; however, they exhibit two fundamental gaps that hinder their ability to assess critical service areas after earthquakes. First, these models rest heavily on emergency procedures based on injury severity rather than type. Second, healthcare demands and injury profiles are often modeled after moderate earthquakes in the United States without including mass casualty data after large earthquakes. This approach has led to an oversimplified representation of the utilization of hospital emergency services and resources, especially for the various profiles of trauma patients after earthquakes globally. This research presents a new hospital operations model based on patient injury type and worldwide earthquake epidemiology data to fill these two gaps. We built the model using discrete event simulations to capture patients' healthcare metrics that vary rapidly after the earthquake. We then used these metrics to study how healthcare outcomes vary with different levels of functional capacity in specific hospital service areas. Our case study showed that ED-level patients wait 40% longer to receive treatment when half of the examination rooms in the emergency department (ED) are lost and 35% less time if examination rooms increase by 50%. Also, the waiting time for hospital-level procedures reduces up to 65% if the bed occupancy rate drops by 15%, e.g., if reverse triage procedures are activated through a discharge of 15% of patients hospitalized before the earthquake. Our findings provide a valuable tool for decision-making in hospital preparedness as it links expected patient volumes to key healthcare metrics according to hospital functional capacity.
The modelling of residential building portfolio exposure model for risk and loss estimations due to natural hazards often do not receive as much attention as other components in the risk chain (e.g. hazard intensity distribution, physical vulnerability). Large-scale (nation or region-wide) exposure models, for instance, are often based on information derived from census and aggregated over geographical administrative units. Moreover, it is customary to employ specific exposure/vulnerability schemas that entail a set of mutually exclusive, collectively exhaustive (MECE) building classes, each associated with a fragility/vulnerability model focusing on the specific reference hazard (e.g. HAZUS). In order to improve the reliability of these models, particularly when the composition of the portfolio is expected to be heterogeneous, individual building observations may be required. This process is relevant in order to constrain and validate the underlying model assumptions. The assignment of single-hazard building classes within a given schema is usually obtained through expert elicitation (e.g., a skilled surveyor). However, if the very same building has to be classified under another vulnerability schema, either for the same hazard (e.g. EMS98 and HAZUS for seismic risk) or, in a multi-risk context, for a different hazard (e.g. tsunami, lahars), this might require a different expertise and the uncertainty of the resulting models could even increase. We propose an innovative method to decouple the collection of exposure information from the development of exposure models in terms of specific vulnerability classes (schemas). Taking advantage of the methodology suggested by Pittore et al., 2018, individual building attributes are observed in the field for a set of surveyed buildings and described in terms of the GEM v2.0 taxonomy, a widely used and well-established faceted building taxonomy (Brzev et al., 2013). The assignment of a class is carried out in a post-processing stage and within a fully probabilistic framework by evaluating the level of compatibility between the observed building attributes and the classes available within the considered schema. The proposed methodology has been exemplified in Chile and Peru within the framework of the RIESGOS project. Expert structural engineers from CIGIDEN (Chile) and the Universidad de la Sabana (Colombia) carried out a Rapid Remote Visual Screening Survey using the RRVS web tool (e.g. Haas et al., 2016). In the case of seismic risk we focused on three schemas, namely SARA (a custom schema developed within the GEM-SARA Project in South America), and the well-known EMS-98 and HAZUS. The tsunami-focused schema proposed by Suppasri et al. (2013) has been also implemented. Preliminary results for Gran Valparaiso (Chile) and Metropolitan Lima (Peru) study areas show the potential of the proposed methodology for streamlining the development of multi-hazard exposure models and significantly improving the transparency of the risk assessment procedures and the propagation of related uncertainties. The importance of extending the building taxonomy to encompass multi-hazard attributes is also discussed.
Conventional earthquake fatality models are commonly implemented using a binary representation of building collapse, an assumption that overlooks severe failure modes and systematically underestimates fatalities. To address this limitation, this study advances the collapse volume ratio—a continuous metric quantifying the reduction of survivable space during collapse—into a probabilistic framework that directly links structural failure to fatality risk. Although previously introduced and quantified in limited studies, we extend its role from a primarily descriptive measure to an analytical component of performance‐based earthquake engineering modeling. The framework is designed for versatility, allowing implementation with different levels of data availability, including site‐specific collapse observations, collapse fragility functions, or complete‐damage fragility functions. Its applicability is demonstrated by leveraging post‐earthquake data from reinforced concrete moment frame buildings affected by the 2023 Kahramanmaraş earthquakes and through case studies of mid‐1970s nonductile and post‐1997 ductile concrete space‐frame archetypes in California. Results show that conventional approaches, constrained by binary collapse assumptions, may underpredict fatalities by up to fortyfold at high shaking intensities. In contrast, the proposed framework reproduces observed fatality rates, estimating that under maximum considered earthquake shaking, nonductile buildings may experience fatality rates of 65%, compared to 11% for ductile buildings. These findings provide quantitative evidence of the life‐safety benefits of modern seismic design and highlight the urgent need for retrofit policies addressing older, nonductile construction.