In 1972, H. Kanamori defined a special class of earthquakes, which he called “tsunami earthquakes,” whose tsunamis are significantly larger than expected from their seismic magnitudes, especially conventional ones. Such events generally feature an exceptionally slow progression of the seismic rupture along the earthquake fault and can be very treacherous because they lack the high frequencies felt by humans in the near-field, which serve as a natural warning for local populations, while hiding in their enhanced low-frequency spectrum the capability to generate disastrous tsunamis. Examples include the catastrophic events in Sanriku (Japan, 1896) and Unimak (Aleutian Islands, 1946). The real-time identification of tsunami earthquakes remains a challenge in modern tsunami warning, especially because these events are relatively rare; only a dozen have been documented in the past 113 years with only five since the advent of modern digital seismometers.
A case study. On September 2, 1992, an earthquake occurred off-shore Nicaragua with magnitudes mb = 5.3 and Ms= 7.2. Note the disparity between the body- and surface-wave magnitudes. The former meant that the earthquake was deprived of the high frequencies typical of ground shaking and felt by humans in the near-field. Indeed, in some coastal communities, the earthquake was not even felt by the population, who thus had no natural warning of the impending disaster. Its higher surface-wave magnitude indicates a “red” source, enriched in lowfrequency energy, as was later confirmed by a Global Centroid-Moment-Tensor (CMT) solution equivalent to Mw = 7.6, measured at periods of 135 s. The earthquake generated a tsunami that ran up to more than 10 m and killed 170 people on the shores of Nicaragua.1 Similar scenarios took place in Sanriku, Japan (1896; 27,000 dead), Java (1994, 2006), and Peru (1996); other tsunami earthquakes have been described in the Kuril Arc (1963, 1975), the Aleutians (1946), and Tonga (1982).2
A major challenge regarding tsunami earthquakes is to identify them in real time from their seismic records. Once an estimate of the seismic moment is obtained, the earthquake is analyzed for possible extended source duration by computing an estimate of the highfrequency energy carried in its P-waves. The result allows a comparison between the behavior of the source in the bass and treble parts of its spectrum, and if an anomaly is detected, identifies the earthquake as a violator of scaling laws, that is, as a tsunami earthquake, whose tsunami potential is greater than would be expected by its initial seismic waves. This algorithm, which uses the concept of the slowness parameter Θ3 has been implemented at the Pacific Tsunami Warning Center (PTWC).4 It was used to successfully identify in real time the slowness of the Java earthquake of July 17, 2006.
Another, more general challenge is to understand the origin of the anomalous rupture in tsunami earthquakes and in particular in what geological environments they can occur. At least two different (and somewhat contradictory) scenarios have been proposed, involving the
activation of splay faults rupturing in sedimentary prisms (Sanriku, Japan; Kurils),5 or rupture propagating in a jagged mode along poorly coupled interfaces in sediment-starved environments (Nicaragua, Peru).6 In turn, this raises the questions whether any subduction zone can be the site of a tsunami earthquake, and conversely whether the documented occurrence of such events (most often comparatively small in the magnitude 7.5 range) precludes the occurrence of mega-earthquakes as would be suggested by the available historical record in Nicaragua or Java. Despite active research efforts in this domain, we presently have no firm answers in this respect.
1. Abe, K., K. Abe, Y. Tsuji, F. Imamura, H. Katao, I. Yohihisa, K. Satake, J. Bourgeois, E. Noguera, and F. Estrada. 1993. Field survey of the Nicaragua earthquake and tsunami of September 2, 1992. Bulletin of the Earthquake Research Institute University of Tokyo 68(1):23-70.
2. Polet, J. and H. Kanamori. 2000. Shallow subduction zone earthquakes and their tsunamigenic potential. Geophysical Journal International 142(3):684-702.
3. Newman, A.V. and E.A. Okal. 1998. Teleseismic estimates of radiated seismic energy: The E/M0 discriminant for tsunami earthquakes. Journal of Geophysical Research 103(B11):26885-26898.
4. Weinstein, S.A. and E.A. Okal. 2005. The mantle wave magnitude Mm and the slowness parameter THETA: Five years of real-time use in the context of tsunami warning. Bulletin of the Seismological Society of America 95(3):779-799.
5. Tanioka, Y., L.J. Ruff, and K. Satake. 1997. What controls the lateral variation of large earthquake occurrence along the Japan trench?Island Arc 6(3):261-266.
6. Fukao, Y. 1979. Tsunami earthquakes and subduction processes near deep-sea trenches. Journal of Geophysical Research 84(B5):2303-2314.