Improving tsunami hazard assessment in central-north Chile using forward (TELEMAC) and inverse (TSUFLIND) models for the 2015 Illapel tsunami

• Published in: Applied ocean research 2024-09, Vol.150, p.104101, Article 104101
• Main Authors: León, Tomás, Wuppukondur, Ananth, Easton, Gabriel, Tang, Hui, Lau, Annie
• Format: Article
• Language: English
• Subjects: Hydrodynamics; Sediment transport; TELEMAC; TSUFLIND; Tsunami deposits
• ISSN: 0141-1187
• DOI: 10.1016/j.apor.2024.104101

•This study introduced a novel comparative numerical modelling methodology using both forward (TELEMAC) and inverse (TSUFLIND) models to simulate the 2015 Illapel tsunami's impact in Chile accurately.•Most probable earthquake source parameters were estimated by comparing modelled tsunami hydrodynamics from TELEMAC with field observations.•Tsunami sediment deposits were used to estimate flow characteristics in the TSUFLIND model.•Both models, TELEMAC and TSUFLIND, yielded consistent results, validating the predicted tsunami flow depths, velocities, and sediment depositions.•The study highlights the importance of combining geological evidence with numerical models to enhance tsunami hazard assessment, especially when direct hydrodynamic measurements are unavailable. The 2015 Illapel earthquake (Mw 8.3) generated a tsunami that significantly impacted the central-north Chilean coast. Previous studies have utilized numerical modelling to analyse this tsunami propagation, yet none have applied a comparative forward and inverse modelling approach to accurately estimate the earthquake source parameters, flow characteristics and sediment transport. A forward model TELEMAC was employed to simulate tsunami hydrodynamics and sediment transport, calibrating it against tide gauge records to identify the most accurate earthquake source scenario. Simultaneously, the inverse model TSUFLIND was utilized to deduce hydrodynamic properties during the inundation phase from sediment deposit characteristics. The performance of the models was evaluated by comparing flow characteristics. The approach was tested along the Tongoy estuary by comparing it against regional tide gauge data, alongside measurements of flow depth and run-up, and sediment deposit observations from five pits along the river (680 m to 1280 m inland). Along the five pits, TELEMAC results indicate the tsunami flow depths ranged between 3.22 m to 0.47 m, and onshore velocities ranged from 4.9 m/s to 2.21 m/s, with sediment deposition from 20 cm to 7 cm. On the other hand, TSUFLIND results were consistent with TELEMAC, with flow depths between 3.22 m to 0.68 m, velocities between 4.84 m/s to 2.05 m/s, and deposition thickness of 15 cm to 0.5 cm along the five pits. The agreement of the hydrodynamic results from both models highlights the importance of integrating geological evidence with numerical modelling where tide records or other hydrodynamic information cannot be obtained for tsunami events, highlighting the