Author(s): Reinaldo Garcia; Sergio Martinez-Aranda; Pilar Garcia-Navarro; Nick Calero
Linked Author(s): Pilar García-Navarro, Reinaldo García-Martínez
Keywords: Containment berm design; GPU-accelerated simulation; Oil storage tank failure
Abstract: Failures of oil storage tanks that trigger structural collapse and spills are uncommon, but several major incidents have caused severe environmental damage. In each case, inadequate containment berm capacity played a key role in the magnitude of the spill. Berms are generally designed to hold 120% of the tank volume, but this static criterion overlooks the dynamic oil wave produced during sudden tank failure, which can lead to overtopping even when nominal design requirements are met. This work proposes a high-performance computational approach for simulating tank-failure scenarios that accounts for dynamic wave propagation. The approach supports the evaluation and optimization of berms to ensure complete containment of released oil. The modeling framework uses OilFlow2D, a GPU-accelerated viscous-flow model that solves the governing equations for free-surface non-Newtonian fluids while tracking temperature-driven changes in oil properties, including density, viscosity, and yield stress. The finite-volume solution on flexible triangular meshes enables fast, high-resolution simulations. We validate the proposed approach with an application to the Ashland tank break that occurred in Pennsylvania, USA in 1988. Additional simulations will show how berm configuration can be optimized to improve containment and enhance spill-risk mitigation and emergency planning.
Year: 2026