January 22, 2001
AGU RELEASE NO. 01-5
For Immediate Release

WASHINGTON - The explosions that sank the Russian submarine Kursk on August 12, 2000, triggered shock waves that were recorded by a network of seismic stations in the Baltic region and beyond. Now, forensic seismologists have used these data to reconstruct the disaster. Writing in the January 23 issue of Eos, the weekly newspaper of the American Geophysical Union, Keith D. Koper and Terry C. Wallace of the University of Arizona and Steven R. Taylor and Hans E. Hartse of the Los Alamos National Laboratory report that, based on their analysis of seismograms, explosions, not impact, caused the Kursk to sink with the loss of all crew members.

The authors note that underwater explosions are highly efficient producers of seismic signals, and these have been long studied, including those generated by the sinking of a Soviet submarine in 1989. The Kursk seismic data possess features unique to underwater explosions, a strong indication that the Kursk did not sink because of a collision or other impact, they say. Seismic stations recorded two explosions that correspond to the Kursk disaster in time and place. The first explosion was 250 times smaller than the second one, which occurred 135 seconds later. The earlier explosion was clearly recorded only at a few nearby stations, while the second one released energy equivalent to around five tons of TNT and was recorded up to 5,000 kilometers [3,100 miles] away.

Koper and his colleagues note that this area of the Barents Sea rarely experiences any seismic activity, so it was highly unlikely that the seismic signals were caused by an earthquake. One point of careful analysis, they say, concerned whether the second event consisted of one massive explosion or several simultaneous smaller ones and perhaps also impact of the Kursk on the seafloor. The most compelling seismic evidence that the main Kursk event was dominated by an explosion was the observation of a "bubble pulse." This pulse results from oscillations of a bubble of hot gases unleashed by an explosion as it rises toward the surface. The spectral pattern produced by an underwater explosion and recorded by seismic stations provides strong evidence that the second explosion was one massive event, not several smaller ones.

The approximate size of the main Kursk explosion can be determined, thanks to a series of calibrated tests conducted by Israeli scientists in the Dead Sea in November 1999. The authors note that the largest Israeli explosion produced a signal similar to that of the Kursk, and both were recorded by a German seismic array, located at virtually equal distance between the Dead Sea and the Barents Sea.

Putting seismic data together with other reliable information, the authors are able to provide a more solid explanation of the Kursk disaster than if they were limited to the seismic data alone. It is believed that the first explosion occurred with the Kursk near the surface, as its periscope was filmed in the up position on the seafloor. Also, it had radioed for permission to fire ordnance just before the first explosion. That explosion produced a seismic record consistent with 250 kilograms [550 pounds] of high explosive, equivalent to the warhead of a modern torpedo. The scientists conclude that a torpedo misfired or exploded prematurely, and that the submarine absorbed a large fraction of the energy released.

As for the large second seismic signal, the authors conclude that it was not from impact with the seafloor, as the Kursk would have sunk the 80-100 meters [265-330 feet] much faster than the 135 seconds between the seismic signals. They say the boat may have remained above the seabed for a time after the first explosion or that the explosion occurred on the sea floor, but only after fire had finally reached other warheads on board. The main event is consistent with the explosion of four to eight SS-N-19 ship-to-ship missiles, which the Kursk carried, or one cruise missile tipped with conventional high explosive warheads.