Tsunami: Approaching the Speed of Sound

Craig Brokensha picture
Craig Brokensha (Craig)
Swellnet Analysis

Most people identify tsunamis with destructive walls of water surging across low lying islands. This is often the case even when a tsunami is only a metre or two in height.

So how does a tsunami differ from the surfable waves we see across our beaches day to day?

The main factor is the period and speed of the tsunami. We know that long period groundswells are much stronger than short period windwells, and that they refract more easily into protected locations. Tsunamis can exhibit periods - the time between successive wave crests - of more than ten times that of a typical groundswell, commonly minutes to sometimes even hours.

Last night's series of major earthquakes off the West Sumatran coast (8.6 and 8.2 on the Richter scale) generated several small tsunamis; their small amplitude owing to a horizontal rather than vertical movement of the ocean floor. Nevertheless, wave periods of 18 minutes were recorded at monitoring stations on Cocos Island and Trincomalee in Sri Lanka.

With such large periods tsunamis are regarded as shallow water waves due to their wavelength - the distance between successive waves crests - being greater than the depth of the water it's moving through.

This means that there's a large amount of energy stored beneath the ocean to depths of thousands of metres, and once this energy starts to reach shallower depths such as an island it starts to pile up and surge.

Because a tsunami is a shallow water wave, a simple equation can be used to calculate its speed and hence arrival time. The travel time is calculated by taking the square root of gravity (9.8m/s^2) multiplied the average water depth. Using last night's tsunami as an example, the water depth off of Sumatra is approximately 5,000 metres, which means that the travel time is approximately 220 metres per second. That's an impressive 797 kilometres per hour, with the speed of sound coming in at 1,224 kilometres per hour.

This means that the island of Simeulue - closest to the epicentre - had only thirty three minutes notice to evacuate following the earthquake. The arrival of the tsunami was estimated at Cocos Island just two hours after the earthquake; while the Maldives had a little more time to spare: two hours and forty minutes.

So while a tsunami with a height of one metre doesn't sound threatening, its forward speed of 797 kilometres per hour can cause catastrophic damage as the surge moves inland for many minutes behind the initial impact.

Below is a small selection of locations and the time it would take for the tsunami to be felt relative to the earthquake off Sumatra.

Simeulue - 33 minutes
Sri Lanka - 1:42 minutes
Maldives - 2:38 minutes
Perth - 5:38 minutes
Durban - 9:23 minutes 



pensky's picture
pensky's picture
pensky Friday, 13 Apr 2012 at 12:24pm

thanks Craig - very interesting. How does the speed of a shallow (Tsunami) wave compare to a groundswell and then windswell wave (I've always been a bit confused by the distinction between groundswell v windswell because surely a groundswell is just a swell created by the wind from a large storm albeit further from the coast -is that essentially the distinction?)

Craig's picture
Craig's picture
Craig Friday, 13 Apr 2012 at 1:33pm

Hi Pensky,

Once the tsunami enters shallower water is slows down and when to about 10 metres deep it moves at around the same speed of a typical 14-15s groundswell 35-40km/h.

To work out the speed of a swell im km/h just multiply the Period by 1.5 to get to knots and then 1.85 to convert to km/h.

So a 10 second swell moves at around 27 km/h while a 16 second swell moves at around 44 km/h.

The difference is that behind the tsunami is a lot more energy spreading out kilometres behind the initial wave that will push in over land for minutes before slowly receding back.

Also the difference between windswell and groundswell is as you've stated, relative to the strength of the storm and hence wind speed, longevity, width and length within the storm (also known as fetch).

The stronger the winds, over a longer period of time and larger area of ocean, the bigger the wave periods will be.

If the storm is right off your coast though the sea state will have no time to organise itself and hence periods can still be low even though winds are extremely strong. Once the swell moves off away from the storm though that's when it organises itself and grows into a proper swell trains with sets etc.

It's tough to pick the point between windswells and groundswells as it's all relative to the coast. For example periods of 10 seconds in WA, SA and Vic would be classed as windswells to some but that's a strong swell on the East Coast. In saying that I'd say somewhere around 7-8 seconds would be the point where windswells start turning into more defined swells and then from 11 seconds up it could be classified as groundswell. It's a real tricky one.

Also here's another article that goes into more detail about how different swell trains and periods are created.. http://www.swellnet.com.au/news/2253-why-the-swell-train-is-often-late

pensky's picture
pensky's picture
pensky Friday, 13 Apr 2012 at 2:42pm

fantastic- appreciated Craig

bbbird's picture
bbbird's picture
bbbird Thursday, 11 Feb 2021 at 9:29pm

Thanks ...for another swell fact closer to home from Craig.
"Why The Swell Train Is Often Late"