Tsunamis and Early Warning Systems

A tsunami, also known as a seismic sea wave, is a series of waves in a water body caused by the displacement of a large volume of water, generally in an ocean or a large lake. Earthquakes, volcanic eruptions and other underwater explosions (including detonations of underwater nuclear devices), landslides, glacier calvings, meteorite impacts and other disturbances above or below water all have the potential to generate a tsunami. Once tsunami waves arrive to coastal areas, their heights magnify and become deadly and devastating. Their destructive power can be enormous and they can affect entire ocean basins; the 2004 Indian Ocean tsunami was among the deadliest natural disasters in human history with at least 230,000 people killed or missing in 14 countries bordering the Indian Ocean. Another devastating tsunami happened in 2011 in Japan with 18,550 people confirmed to be killed/missing with highest wave height of 40 m. Due to increase in the number of tsunamis in the last decades, it becomes a topic of great societal concern worldwide, as evidenced by any number of recent events including Indonesia (2004), Chile (2010) and Japan (2011).

Statistically speaking, the human exposure to tsunami in the US is 550000 people per year. Amount of GDP (Gross Domestic Product) present in hazard zones that are thereby subject to potential losses by tsunami for USA is 13 Billion USD. In the United States, history and geologic evidence show that tsunamis are a significant threat. Since the beginning of the 20th century, 34 tsunami events have caused more than 500 deaths and over $1.7 billion (2014 dollars) in damage to U.S. coastal states and territories. Recent studies indicate that in the future, a large U.S. tsunami could affect millions of people and cause tens of billions of dollars in damage. Tsunamis cannot be prevented, but their impacts on life, property, and the economy can be greatly reduced by an efficient Tsunami Early Warning System (TESWS).

The State of the art of tsunami warning procedures currently relies on seismic and sea level measurements (Detection of tsunami wave). Nevertheless seismic networks may lead to issue warnings that can be later canceled by the sea level records. The near-field prediction of tsunami propagation is challenging, given the limited time to spread the alarm: it is therefore unfeasible to wait the measurement of the tsunami itself before spreading the alert. The records of the faster hydro-acoustic waves can cope with the shortening decision time of spreading the alarm and improve the accuracy of operational TEWS. In our work, we have developed the first numerical model able to reproduce hydro-acoustic waves with a high level of accuracy compared to traditional models. Due to low computational costs of the proposed model, it can be used for enhancement of TWES and improve its performance significantly. Our  idea would be a practical solution for detection of tsunami events, especially where the generation zone and destination are close and conventional TWES cannot act in proper time to warn in danger people.  

Tsunami Precursors:

Tsunamigenic fast movements of the sea-bed generate pressure waves in weakly compressible sea water, namely hydro-acoustic waves, which travel at the sound celerity in water (about 1500 m/s). These waves travel much faster than the counterpart long free-surface gravity waves (tsunami) and contain significant information on the source. Measurement of hydro-acoustic waves can therefore anticipate the tsunami arrival and significantly improve the capability of Tsunami Early Warning Systems (TEWS). However, applications to real cases require detailed numerical modelling in order to clearly define the time series at point A due to a source at point B.

Three-dimensional models are straightforward to use, but require unrealistic computational times when applied to large-scale geographical areas, i.e. they cannot be used for a systematic investigation on an oceanic scale of prediction. The problem is further complicated by the effects of compressible viscous sediment layer at the sea bottom, which have a deep influence on the hydro-acoustic waves propagation over large distances. In our work, we derived two numerical models suitable for reconstruction of hydro-acoustic waves/tsunami over real large-scale geographies. The Mild-Slope Equation in Weakly Compressible fluid [Sammarco et al. 2013; Abdolali et al. 2014] was developed first. It reduced the computational problem from three to two dimensions, hence reducing dramatically the computational costs. Then the capabilities of the model were extended by including the effects of a sediment layer at the bottom [Abdolali et al. 2015a]. Finally, three applications to real bathymetries were implemented.  In the first, the model is applied to simulate the hydro-acoustic wave propagation in the central and eastern Mediterranean Sea, generated by two main destructive historical earthquakes: the 365 AD Crete event and the 1693 Sicily event [Abdolali et al. 2014; Cecioni et al. 2015]. In the second application the model is used to reproduce the 28 October 2012 7.8 Mw earthquake occurred off the West coast of Haida Gwaii archipelago, Canada. For this event deep water field measurements are available for comparison [Abdolali et al 2015 b]. In the third application, the model has been used to reproduce tsunami and hydro-acoustic wave fields of Tohoku Oki 2011, Japan, using MSEDWC [Abdolali et al. 2015 c]. On the basis of the numerical results for these real cases, several conclusions on the possible use of hydro-acoustic waves as support to Tsunami Early Warning Systems can be drawn.

This work is particularly valuable for detection of near field generated tsunamis where there is a short time from occurrence of event and arrival of tsunami to vulnerable areas. By providing the developed model, which detects precursor component of tsunami waves, locally generated tsunamis triggered by submarine earthquake/mass failure can be distinguished in proper time and coastal authorities have time to warn people to evacuate the in danger areas. The proposed model increases the accuracy of operational TEWS and therefore leads to a significant decrease in false alarms/increase the reliability of warning systems.

Contrary to traditional models, which use simplified assumptions for decrease of computational costs, our model considers a higher level of complexity of problem leading to a higher level of accuracy and at the same time retaining the same level of computational costs. Therefore, it can be used for real-time simulation of tsunamigenic events. The main advantages of the proposed model to traditional models are:

- Consideration of compressibility of water.

- Taking into the account the role of underlying sedimentary layers, which have a big influence of model results.

- Consideration of spatiotemporal distribution of bottom displacement, which plays a significant role on tsunami generation mechanism.

- Low computational costs of two dimensional numerical model compared to fully three dimensional models.

In addition, the model results can be used for interpretation of in-situ observatories and help to identify the characteristics of generation source i.e. earthquake, Submarine mass failure and the characteristics of complex stratified sedimentary layers structure at sea bottom.