Memorandum on the Sumatran Tsunami of 26 December 2004

3.

RECORDING THE TSUNAMI ALONG SOUTH AFRICAN COAST

Although the main impact was felt in South-east Asian countries, the tsunami also affected the coasts of some African countries. A newspaper, The Cape Times of 28 December 2004, reported that a high tide surge was experienced at Struisbaai on the Southern coast of South Africa (IOL website, 2004). The effect of the tsunami was, in fact, registered around the entire South African coastline. The Hydrographic Office of the South African Navy operates water level recorders at most of the major ports around the South African coast (see Figure 7). These recorders are used mainly for measuring the daily tidal variations. These data sets are then analysed to predict future tide levels which are used by the maritime community. Unfortunately, the recorder in Durban was out of commission at the time of the event. The

Hydrographic

Office

analysed

the

water

level

measurements

for

December 2004. This analysis entailed the separation of the normal tidal variation from the water level measurements. Once the tidal component has been removed the sea level variation remains which, most probably, is the result of the tsunami effect. The time-series of these variations are presented in Figure 8. It is interesting to note that variations reached a peak in the Port of Port Elizabeth (PE). The reason for the peak at PE is not clear and further analyses are still required. However, a possible explanation may be found by considering the continental shelf along the South African coast. The approaching tsunami can be seen as a long-period wave, with a wave length of anything from 20 to 300 km. As it reached the South African East coast (Richards Bay to East London), little shoaling may have taken place as the continental shelf is fairly narrow. However, towards PE, the shelf widens considerably, thereby allowing shoaling to take place. The extent of this shoaling might also have resulted in an increase in the wave height in the vicinity of PE before friction, as a result of the

1

Memorandum on the Sumatran Tsunami of 26 December 2004

effect of the sea-bottom topography (bathymetry) on the propagating waves, would have reduced the energy of the westerly propagating waves. Further to the west the effect of the tsunami was still significant, as indicated by the Mossel Bay time-series (Figure 8), but the increasing friction effect of the sea-bottom, may have resulted in a reduction in the wave energy. Water level measurements at Port Nolloth on the West coast indicate that the effect of the tsunami was even felt this far north (Figure 8). These recordings, therefore, indicate how effectively a tsunami can propagate over long distances and around continents. The graphs in Figure 8 also show that the disturbance lasted for more than three days, which can be attributed to the repeat long-period ripple effects of the initial disturbance generated by the earthquake. Although no direct water levels were recorded at Durban, two sets of sea-surface measurements are available.

The CSIR operates two Acoustic Doppler Current

Profiler ADCP) instruments on behalf of the National Ports Authority of Durban. Both these instruments are located on the sea-bottom. The data from these instruments, which include wave and current measurements, are normally used for daily port operations. The positions of the two ADCPs are shown in Figure 9. The purpose of these instruments is to record wave and current information. However, the ADCP instrument also permits the recording of surface water levels, thereby providing information of the tidal variation. Note that the water levels are averaged over a 20 minute period while the SA Navy recordings are averages of measurements over 3 minutes. Consequently, the data from the ADCPs cannot be compared directly with that from the water level recorders. However, the ADCP data still show that the normal tidal variation was disturbed during the afternoon of 26 December 2004. The tidal variation as recorded by the two ADCP systems is presented in Figure 9. Using the measurements available along the South African coast, and the approximate time of the earthquake (Section 2.2), it was possible to calculate the

2

Memorandum on the Sumatran Tsunami of 26 December 2004

speed or celerity of the tsunami as it propagated across the Indian Ocean. The estimates of the celerity are summarised in Table 1.

Table 1: Summary of distance, time and celerities Station Richards Bay Durban East London Port Elizabeth Mossel Bay

Location

ETA

(Lat/Long)

(in SAST) th

Approx. travelled

Approx.

Calculated

distance (km)

Travel time (hrs)

celerity (km/h)

28.812°S; 32.055°E

14:03 (26 )

7 700

11

700

29.877°S; 30.991°E

th

32.991°S; 27.850°E 33.940°S; 25.588°E 34.248°S; 21.888°E

14:00* (26 )

7 800

11

709

th

8 200

11.5

713

th

8 400

12.2

689

th

8 800

13.2

667

th

9 110

14.6

624

th

9 130

14.8

617

th

9 240

15.2

608

th

9 640

21.2

455

14:30 (26 ) 15:12 (26 ) 16:12 (26 )

Simons Town

34.193°S; 18.450°E

17:34 (26 )

Cape Town

33.934°S; 18.431°E

17:48 (26 )

Saldanha

32.998°S; 17.968°E

18:13 (26 )

Port Nolloth

29.289°S; 16.879°E

00:10 (27 )

ETA: Estimated Time of Arrival SAST: South African Standard Time Based on the data presented in Table 1 it is estimated that the tsunami travelled to the South African coast, between Richards Bay and PE, at a speed of approximately 700 km/h. This is supported by the observed estimated time of arrival (ETA) at Richards Bay and East London. Note that the Durban measurements are too coarse for a proper estimate of the ETA. The estimated celerities based on the stations further South and West along the coast indicate slower speeds.

This can be

attributed to the friction effect induced by the broadening of the continental shelf from the vicinity of Port Elizabeth westwards. The celerity (c) of approximately 700 km/h is also supported by the equation for estimating a long-period gravity wave in shallow water, which is given by:

c = gd where g = the gravitational constant and d = the water depth

3

Memorandum on the Sumatran Tsunami of 26 December 2004

Taking an average water depth of about 4 000 m (for the Indian Ocean), a celerity of 713 km/h is obtained, which is close to the celerity obtained from the observed time of arrival at the stations on the East coast. From all the measurements available, it is clear that a disturbance in the waters along the South African coast occurred on 26 December 2004. This disturbance was most probably the result of the tsunami that originated about 8000 km to the East of South Africa (Figure 10). The effect of the tsunami is further illustrated in Figure 11, which shows two images from video footage taken in the Port of East London with the surveillance camera system on the afternoon of 26 December 2004 (G Kingsley Wilkens, Security Manager Port of East London, pers. comm.).

The images shows a yacht being

pushed against another moored craft and a moored surface buoy being dragged under the water as a result of the outgoing surge.

The surveillance camera system

provided detailed information on the surge motions in the harbour as a result of the tsunami.

4

Memorandum on the Sumatran Tsunami of 26 December 2004

4.

SUMMARY AND CONCLUSIONS

This report gives a brief summary of the 26 December 2004 tsunami that occurred off the West coast of Northern Sumatra in the Indian Ocean and its consequences along the coast of South Africa. More detailed information on this tsunami is available from numerous websites (see Section 2.2). The earthquake responsible for the destructive tsunami registered at least 9 on the Richter Scale and is considered to be one of the worst natural disasters in modern history. Although the main impact was felt in South-east Asian countries, the tsunami also affected the coastlines of some African countries, including Somalia, Tanzania and South Africa. The tsunami was recorded around the South African coastline by water level recorders operated by the Hydrographic Office of the South African Navy and by the CSIR ADCPs at the Port of Durban. The data collected around the coast indicated that the tsunami travelled around the coast, from Richards Bay on the East coast, to Port Nolloth on the West coast. The data further indicated that the tsunami covered the distance from the Sumatra to the South African East coast in approximately 11 hours at a speed of about 700 km/h. A question frequently posed, not just in South Africa but elsewhere as well, is whether South Africa is safe from tsunamis, especially from subduction zone earthquakes. To answer this question, a thorough study of the potentially earthquake prone geological structures is required.

For example, apart from the Sunda Arc

region, a subduction zone is also active in the South Sandwich Island region, located approximately 4 000 km South-west of South Africa.

Fortunately, only moderate

earthquakes occur in this region since the two plates generate less stress than that experienced in the Sunda Arc region. Considering that this phenomenon has now been experienced and even in the past, it is clear that tsunamis will occur in future again. Therefore, a tsunami warning system may well be relevant to this coastal region of Southern Africa. Sufficient warning time could be provided by such a system as the tsunami generation areas 5

Memorandum on the Sumatran Tsunami of 26 December 2004

are relatively far away from the South African coast. The implementation should therefore be investigated. The probability of a major tsunami devastating the South African coast is very low.

However, the consequence of the impact could be

catastrophic. Acknowledgements: The author wish to thank the Hydrographic Office of the South African Navy for providing the water level data recorded at Richards Bay, East London, Port Elizabeth, Mossel Bay, Simons Town, Cape Town, Saldanha and Port Nolloth.

6

Memorandum on the Sumatran Tsunami of 26 December 2004

REFERENCES Cummins, P (2004).

Small threat, but warning sounded for tsunami research.

AusGeo News, Issue No. 75, September 2004. Demark, T. (2005).

Website: http://homepage.mac.com/demark/tsunami/11.html. Images ©

DigitalGlobe.

HAMZAH LATIEF (2005). Presentation: TSUNAMI ACEH 2004, Tsunami Research Group, Kelompok Penelitian dan, Pengembangan Kelautan, Institut Teknologi Bandung, HAMZAH LATIEF dkk. IOL website (2004). Website: http://www.iol.co.za. Freak waves hit SA as a result of tsunami NOAA (2005). Website: http://www.pmel.noaa.gov/tsunami/indo_1204.html. The Tsunami Research Program of the Pacific Marine Environmental Laboratory. NOAA.

USGS (2005).

Website:

http://earthquake.usgs.gov/eqinthenews/2004/usslav/.

Earthquake

Hazards Program.

Wikipedia (2005). Website: http://en.wikipedia.org/wiki/2004_Indian_Ocean_earthquake

7

Map showing the locations of operational water level recorders along the coast

Figure 7

Sumatra Earthquake (M = 9.0)

Sea level

100 cm

South Africa

December 2004

Source: received from the Hydrographic Office

Sea-level variations measured with the water level recorders of the Hydrographic Office of the SA Navy

Figure 8

Position of Offshore ADCP

Position of Harbour ADCP

Durban ADCP: Water Level 1.0 0.8 0.6

Water Level (m)

0.4 0.2 0.0 -0.2 -0.4 -0.6 -0.8 -1.0 25 (00:00)

25 (12:00)

26 (00:00)

26 (12:00)

27 (00:00)

27 (12:00)

28 (00:00)

28 (12:00)

29 (00:00)

Date (SAST) Offshore ADCP

Harbour ADCP

Map based on SAN chart 1030, produced by Hydrographer, SA Navy

Location of the offshore and harbour ADCP’s at the Port of Durban with the tidal variation measurements from 25 to 28 December 2004

Figure 9

± 8 000

Figure Map illustrating the distance between South Africa and the origin of the December 2004 tsunami event

10

Moored surface buoy

Moored surface buoy

Source:

Video footage received from G Kingsley Wilkens, Security Manager of NPA East London

Images captured from video footage taken by the NPA surveillance camera system in the Port of East London

Figure 11