Development of magnitude correlation equations for the tsunamigenic zones of the Indian Ocean

• Published in: Journal of seismology 2023-06, Vol.27 (3), p.473-492
• Main Authors: Sabah, Nazeel, Shanker, Daya
• Format: Article
• Language: English
• Subjects: Catalogues; Correlation; Earth and Environmental Science; Earth Sciences; Earthquakes; Geophysics/Geodesy; Geotechnical Engineering & Applied Earth Sciences; Hazard assessment; Hydrogeology; Oceans; Regional development; Regression analysis; Regression models; Saturation; Seismic activity; Seismology; Structural Geology; Surface water waves; Surface waves; Tsunami hazard; Tsunamis; Weather hazards
• ISSN: 1383-4649; 1573-157X
• DOI: 10.1007/s10950-023-10151-x

There is a pressing need for a homogonous tsunami catalogue for the Indian Ocean as nearly 20% of tsunami events worldwide affect the region. Any study on tsunami hazard assessment necessitates a homogenous tsunamigenic earthquake catalogue. The existing records of strong tsunamigenic earthquakes have the magnitudes expressed in moment magnitude (M W ), body wave magnitude (m b ), local magnitude (M L ), and surface wave magnitude (M S ). This study deals with developing regional magnitude correlation equations for tsunamigenic earthquakes of the Indian Ocean. The present investigation estimates the threshold magnitude and focal depth for an earthquake to turn tsunamigenic. It is found that earthquakes above M W  ≥ 5.9 and focal depth ≤ 80 km have the potential to generate a tsunami in the region. The moment magnitude is the most proper scale to characterize the size of large tsunamigenic earthquakes as it is more directly related to the released energy and does not suffer saturation. Hence, equations have been developed to convert surface wave magnitude (M S ) to moment magnitude (M W ) using three types of regression models viz. standard regression (SR), inverse standard regression (ISR), and orthogonal standard regression (OSR). The efficacy of these models has been compared in terms of R-squared and residual analysis. This study indicates that OSR is the best-suited regression model for developing magnitude correlation equations for the three zones of the Indian Ocean region under study. Also, a single unified conversion equation for the whole of the Indian Ocean has been derived with rational accuracy.