Trapped waves of the 27 November 1945 Makran tsunami: observations and numerical modeling

Abstract

The 27 November 1945 earthquake in the Makran Subduction Zone triggered a destructive tsunami that has left important problems unresolved. According to the available reports, high waves persisted along the Makran coast and at Karachi for several hours after the arrival of the first wave. Long-duration sea-level oscillations were also reported from Port Victoria, Seychelles. On the other hand, only one high wave was reported from Mumbai. Tide-gauge records of the tsunami from Karachi and Mumbai confirm these reports. While the data from Mumbai shows a single high wave, Karachi data shows that Electronic supplementary material The online version of this article (high waves persisted for more than 7 h, with maximum wave height occurring 2.8 h after the arrival of the first wave. In this paper, we analyze the cause of these persistent high waves using a numerical model. The simulation reproduces the observed features reasonably well, particularly the persistent high waves at Karachi and the single high wave at Mumbai. It further reveals that the persistent high waves along the Makran coast and at Karachi were the result of trapping of the tsunami-wave energy on the continental shelf off the Makran coast and that these coastally-trapped edge waves were trapped in the alongshore direction within a *300-km stretch of the continental shelf. Sensitivity experiments establish that this along-shore trapping of the tsunami energy is due to variations in the shelf width. In addition, the model simulation indicates that the reported long duration of sea-level oscillations at Port Victoria were mainly due to trapping of the tsunami energy over the large shallow region surrounding the Seychelles archipelago.

Effective coastal boundary conditions for tsunami wave run-up over sloping bathymetry

“…In our result, the NSWE model is employed in the last beach slope region. The period of this wave is approximately 2 h, and it coincides with the observed tsunami at the Makran coast, according to Neetu et al (2011). In nature, one would not expect a tsunami of a monochromatic wave train.…”

“…12), akin to a real coastal bottom topography, to find the analytical run-up amplification due to resonance effects. We follow this bathymetry profile, with tan α = 0.0036, tan β = 0.0414, h 0 = 2500 m, and h 1 = 200 m. These choices roughly characterize the Indian coast bathymetry (Neetu et al, 2011). The EBC point is located at the edge of the last beach slope.…”

“…These choices are roughly characterizing the Indian coast bathymetry (Neetu et al, 2011). The EBC point is located at the edge of the last beach slope.…”

Localised continental shelf waves: geometric effects and resonant forcing

“…A slightly less clear-cut example may be given by the numerical study of Neetu et al (2011) who consider the response of the coastal region off the Makran coast of Pakistan to an offshore earthquake on the continental shelf. The instantaneous shift in bottom topography, near a narrow shelf region, forces a localised disturbance, with maximum amplitude near the region of maximum shelf slope gradient, which persists for the duration of their numerical simulations (10hrs) with at least 25% of the total energy in the computational domain concentrated in the localised disturbance, suggesting that the instantaneous shift in bottom topography transfers energy into a lowest mode CSW.…”

Effective coastal boundary conditions for tsunami simulations

“…We follow this bathymetry profile, with tan α = 0.0036, tan β = 0.0414, h 0 = 2500m, and h 1 = 200m. These choices roughly characterize the Indian coast bathymetry [Neetu et al, 2011]. The EBC point is located at the edge of the last beach slope.…”