Comprehensive Probabilistic Tsunami Hazard Assessment in the Makran Subduction Zone

TL;DR

This study presents a comprehensive probabilistic tsunami hazard assessment for the Makran subduction zone, accounting for uncertainties and variability, revealing significant hazard levels and regional differences along the Iranian and Pakistani coastlines.

Contribution

It introduces a novel probabilistic framework for tsunami hazard assessment in Makran, integrating epistemic and aleatory uncertainties with scenario modeling and wave propagation analysis.

Findings

Hazard curves show high variability across locations.

Probability of waves exceeding 3 m reaches up to 91% for 1000-year return period.

Ignoring aleatory variability underestimates long-term hazard risk.

Abstract

After the 2004 and 2011 tsunamis came unprecedented to the scientific community the role of probabilistic tsunami hazard assessment (PTHA) in tsunami-prone areas came to the fore. The Makran subduction zone (MSZ) is a hazardous tsunami-prone region; however, due to its low population density, it is not as prominent in literature. In this study, we assess the threat of tsunami hazard posed to the coast of Iran and Pakistan by the MSZ and present a comprehensive PTHA for the entire coast regardless of population density. We accounted for sources of epistemic uncertainties by employing event tree and ensemble modeling. Aleatory variability was also considered through probability density function. Further, we considered the contribution of small to large magnitudes and used our event trees to create a multitude of scenarios as initial conditions. Funwave-TVD was employed to propagate these scenarios. Our results demonstrate that the spread of hazard curves for different locations on the coast is remarkably large, and the probability that a maximum wave will exceed 3 m somewhere along the coast reaches $\{13.5, 25, 52, 74, 91\}\%$ for return periods $\{50,100, 250, 500, 1000\}$, respectively. Moreover, we found that the exceedance probability could be higher at the west part of Makran for a long return period if we consider it as active as the east part of the MSZ. Finally, we demonstrated that the contribution of aleatory variability is significant, and overlooking it leads to a significant hazard underestimation, particularly for a long return period.