Elevated steel water tanks play a critical role in urban water supply systems by providing both storage capacity and adequate pressure within distribution networks. Due to their strategic importance, ensuring their structural integrity and continuous functionality during and after seismic events is essential. Unlike conventional buildings, elevated water tanks exhibit distinct dynamic characteristics arising from the interaction between the supporting structure, contained fluid, and supporting soil. The 2017 Ezgeleh earthquake in Kermanshah Province, Iran (November 12, 2017), caused noticeable damage to several elevated steel water tanks in Sarpol-e Zahab, highlighting the need for a more comprehensive understanding of their seismic behavior. This study investigates the seismic performance of five damaged elevated steel water tanks observed during the aforementioned earthquake, with particular emphasis on the role of soil–structure interaction (SSI). The research combines field observations with analytical modeling to evaluate the influence of foundation conditions on the dynamic response of these structures. The fluid inside the tanks is modeled using the widely accepted lumped mass-spring analogy, in which the liquid mass is divided into impulsive and convective (sloshing) components. This approach enables a realistic representation of hydrodynamic pressures exerted on the tank walls during seismic excitation. To assess the impact of soil–structure interaction, two different approaches are employed. The first approach is based on the cone model for shallow foundations, where the soil medium is idealized using a system of springs, dashpots, and lumped masses that simulate radiation damping and soil stiffness in various degrees of freedom. This method offers a balance between computational efficiency and engineering accuracy, making it suitable for practical applications. The second approach follows the simplified provisions recommended by the Iranian seismic design code (Standard No. 2800), which incorporates SSI effects through modification factors applied to structural properties.
The supporting structures of the tanks are modeled as braced steel frames, and nonlinear dynamic analyses are conducted to capture their inelastic response under seismic loading. Soil properties used in the SSI modeling are derived from site-specific geotechnical data, including shear wave velocity profiles obtained from previous microtremor measurements. The comparison between fixed-base and flexible-base models demonstrates that SSI significantly affects the seismic response of elevated tanks. In particular, it leads to an increase in lateral displacements and fundamental periods of vibration, which may amplify the overall seismic demand on the structure. Furthermore, the results obtained from the cone model and the code-based method are compared to evaluate their consistency and applicability. It is observed that while both methods capture the general trends of SSI effects, the cone model provides a more detailed and realistic representation of soil behavior, especially for sites with complex stratification. However, the code-based approach remains useful for preliminary design and routine engineering practice due to its simplicity. The findings of this study emphasize the importance of incorporating soil–structure interaction in the seismic analysis and design of elevated water tanks, particularly in regions with soft or layered soil conditions. Neglecting SSI may lead to underestimation of seismic demands and potential structural vulnerabilities. The results can contribute to improving current design practices and enhancing the resilience of water supply infrastructure in seismic regions.