PSI - Issue 78

Maria Zucconi et al. / Procedia Structural Integrity 78 (2026) 839–844

840

Keywords: Consequence functions, Seismic-tsunami correlation, Probabilistic risk assessment, Repair cost ratios, Loss estimation; RC infill wall losses.Type your keywords here, separated by semicolons ;

1. Introduction The analysis of losses resulting from natural hazards is gaining strategic relevance in structural engineering, particularly when dealing with multiple interacting phenomena (Goda and Abilova 2016; Goda and De Risi 2017; Romano et al. 2018, 2021; Song and Goda 2019; Zucconi et al. 2022). Among these, the combined effects of earthquakes and subsequent tsunamis represent a critical scenario in terms of vulnerability and economic consequences. Estimating the impact of such events is essential for supporting emergency planning, designing mitigation strategies, and informing territorial risk governance. Over the years, several countries have developed structured methodologies to quantify tsunami-related losses. In the United States, loss estimation tools, such as those developed by FEMA classify damage based on flooding depth and flow characteristics, assigning repair cost ratios to specific building types (FEMA - Hazus 5.1 2022). In Japan, the Ministry of Land, Infrastructure, Transport and Tourism (MLIT) has collected extensive post-event data, following the 2011 tsunami, allowing for the definition of empirical repair cost ranges linked to different damage levels and construction typologies (Goda and De Risi 2018). Despite the robustness of these models, their transferability to other contexts is limited, due to differences in materials, design standards, and building morphology. In the Italian context, although a rich dataset exists concerning building damage from seismic events (Dolce et al. 2021, [CSL STYLE ERROR: reference with no printed form.]) — collected through national post-earthquake assessment programs — no equivalent database is available for tsunami damage. This gap hinders the direct development of empirical tsunami loss models calibrated on local structures. To address this challenge, the present study introduces a methodology that integrates available seismic loss data with tsunami-specific engineering classifications to estimate realistic repair cost ratios in tsunami scenarios. The proposed approach defines a mapping between seismic and tsunami damage states, enabling the construction of consequence functions even in the absence of direct tsunami observed data. The methodology is applied to a representative reinforced concrete (RC) residential building, selected to reflect the structural characteristics commonly found in Italian coastal areas exposed to tsunami risk. By analyzing the influence of building height and damage extent on expected losses, the framework aims to support scalable applications for territorial loss scenarios and contribute to multi-hazard risk assessment in data-poor environments. 2. Methodological Framework In the absence of empirical tsunami damage data for Italian buildings, this study adopts a consequence-based approach by leveraging post-earthquake reconstruction data already available in the national context. In particular, the loss estimation framework is built upon the correlation between observed seismic damage — classified using the EMS 98 scale (Grünthal 1998)and derived from AeDES surveys (Dolce et al. 2014) — and corresponding repair cost ratios (Cr%). These ratios have been previously calibrated using statistical analyses conducted on buildings affected by recent seismic events, including the 2009 L’Aquila earthquake (Zucconi et al. 2025). To transfer this information into a tsunami context, a set of engineering-based tsunami damage states (DSt1 to DSt5) has been adopted, following the classification proposed by Del Zoppo et al. (2025) (Del Zoppo et al. 2025). These damage states are defined according to specific structural and non-structural response criteria observed through numerical simulations on RC buildings. On one hand, DSt1 and DSt2 refer to damage affecting only non-structural components — primarily infill walls — with increasing severity: from minor cracking in DSt1 to out-of-plane failure of infills in DSt2. On the other hand, DSt3 and DSt4 are associated with structural damage mechanisms. DSt3 corresponds to the attainment of shear capacity in vertical members, while DSt4 is linked to the development of flexural yielding in the longitudinal reinforcement. Finally, DSt5 represents a condition of global collapse, including the potential loss of bearing capacity at the ground story. By associating each DSt with a compatible seismic damage level — based on similarities in the affected components and failure mechanisms — it becomes possible to infer tsunami-related Cr% using the well-documented Italian seismic