Tide-gauge records of the 1883 Krakatau tsunami

Efim Pelinovsky, Alexander Stromkov, Irina Didenkulova

The eruption of Krakatau volcano, situated in the Sunda Strait occurred in August 27, 1883 has generated a very devastating tsunami. The blast, at 9:58 (local time) was largest from the series of the four main explosions during this day. Approximately 10 km 3 of solid rock was blown out of the volcano and about 20 km 3 of pyroclastic deposits spread out over 300 km 2 to an average depth of 40 m. A caldera 6 km in diameter and 270 m deep formed where the central island had once stood. This blast was the largest sound ever heard in recorded history, being heard up to 4,800 km away. The atmospheric shock wave traveled around the world seven times. Barometers in Europe and the United States measured significant oscillations in pressure over nine days following the blast. The total energy released by this eruption was equivalent to 200 megatons atomic bomb (8.4 ? 10 17 joules). It devastated the adjacent coastline of Java and Sumatra with a maximum runup height of 42 m, and penetrated 5 km inland over low-lying areas. At least 36,000 people were killed, mostly by the sea waves, and 300 villages were destroyed. Sea surface oscillations have been recorded worldwide in the Indian, Pacific and Atlantic Oceans. Various descriptions of tsunami wave manifestation can be found in [ Symons, 1888; Murty, 1977; Simkin and Fiske, 1983; Bryant, 2001; Choi et al., 2003 ].

Original data of tsunami observations in Indonesia and in the World were collected in [ Verbeek, 1885; Symons, 1888 ], and then reproduced in part [ Murty, 1977; Simkin and Fiske, 1983 ; Bryant, 2001 ]. These data are digitized with a time step of 2 min and presented in Original. It is important to mention that tsunami record at Batavia ( Jakarta ) has “cut-off” of large waves and we use two curves for Jakarta : the original and the smoothed curve as described in [ Symons, 1888 ].

We eliminate the tide components from the tide-gauge records by the following processing procedure. At the first stage the tidal component in each of the records is selected using the Godin-type filter, which consists of three consecutive running-mean filters with windows 100, 100 and 105 min, as it is recommended by [ Emery and Thomson, 1998 ]. After that, the selected tide components are subtracted from the initial records. The results of this work are presented in Filtered. Some "disturbances" on the filtered records are due to the high-frequency tide components that cannot be filtered out and should not be interpreted as tsunami related.

The first column in all data files is the time in hours. The second column is the sea level in meters.

References:

Bryant, T., Tsunamis , Cambridge University Press. 2001.

Choi, B.H., Pelinovsky, E., Kim, K.O., and Lee, J.S., Simulation of the trans-oceanic tsunami propagation due to the 1983 Krakatau volcanic eruption. Natural Hazards and Earth System Sciences , 2003, vol. 3, No. 5, 321 – 332.

Emery, W.J., Thomson, R.E. , Data analysis methods in physical oceanography, Pergamon GB, London , 1998.

Murty, T., Seismic Sea Waves – Tsunamis , Bull. Dep. Fisheries, Canada , 1977.

Pelinovsky, E. , Choi, B.H. , Stromkov, A. , Didenkulova, I. , Kim, H.S. , Analysis of tide-gauge records of the 1883 Krakatau tsunami, In book: Tsunamis: case studies and recent developments , Springer , 2005, Vol. 23 , 57-78.

Simkin, T., and Fiske, R.S., Krakatau 1883 - the volcanic eruption and its effects , Smithsonian Institution Press, Washington , D.C. , 1983.

Symons, G.J. (ed), The eruption of Krakatoa and subsequent phenomena. Report of the Krakatoa committee of the Royal Society. London : Trubner & Co, 1888.

Verbeek, R.D.M., Krakatau , Batavia 1885.