Features

Tsunami danger closer home

K. S. Jayaraman

doi:10.1038/nindia.2009.8 Published online 21 January 2009

Indian Fans.

@ WHOI

The Asian tsunami in 2004 that killed 225,000 people was caused when a powerful earthquake shook the ocean floor off western Sumatra pushing up the overlying water. Now a team of researchers from Bangalore and New Delhi report that high magnitude earthquakes are not the only events that can generate tsunamis in the Indian Ocean1.

"The sources of potential danger are closer home," says Chittenipattu Rajendran, a geologist at the Indian Institute of Science (IISc) in Bangalore who led the study. "The danger that has been overlooked till now lurks on either side of the Indian subcontinent," he told Nature India.

"Both the Indus and the Bengal Fan areas in the north Arabian Sea and the Bay of Bengal respectively, are prone to submarine slides that pose additional tsunami threats in the region," the scientists report.

The Bengal Fan that is about 3000 km long, 1000 km wide and up to 20 km thick is the largest submarine fan in the world. It is filled with sediments eroded off Tibet and the Himalayas since 20 million years ago and transported by the Ganga-Brahmaputra River system. The Indus Fan (1500 km long, 960 km wide and 10 km thick) in the Arabian Sea receives sediments from the Himalayas as well as from Pakistan and Arabia.

Mountains of sediments sliding down the continental slope can generate tsunami, says Rajendran who is a Fellow at IISc's Centre for Earth Sciences "You do not require a big earthquake to initiate a slide. Once toppled, there is no stopping like a snow avalanche," he said. There are precedents in other parts of the world: the destructive tsunamis of Papua New Guinea (1998)2, Grand Banks, Canada (1929)3 and the Messina Tsunami (1908)4 in Southern Italy were all caused by underwater landslides triggered by earthquakes. Rajendran says that sediments and seismic faults — the two necessary ingredients for submarine landslides — co-exist in north Indian Ocean. The Makran subduction zone, which extends 800 km across the northern Arabian Sea is close to Indus Fan, and the Arakan subduction zone along Myanmar coast actually extends beneath the Bengal Fan.

"The new study makes the important point that we know lesser about sources of tsunami in the Indian Ocean than those in the Pacific," says Phil Cummins a geologist and tsunami expert at Geoscience Australia. "The fact that the two largest sedimentary bodies in the modern oceans, the Indus and Bengal Fans, lie in the Indian Ocean should alert us to the potentially much greater importance of landslide-generated tsunamis there," he told Nature India in an email interview. "The abundance of sediment makes the subduction systems along the Myanmar and Makran coasts unlike anything found in the Pacific, and this means there is a lot of science that needs to be done before we know much about their tsunamigenic potential."

The warning systems set up after the 2004 tragedy (one in Jakarta, Indonesia, and another in Hyderabad, India) rely on estimates of an earthquake's magnitude and vertical dislocation of the sea floor to determine whether a tsunami will be generated. "These models cannot predict tsunamis generated by submarine slides and should therefore incorporate such overlooked mechanisms," says Rajendran. His team included his wife Kusala, a geophysicist at IISc, and researchers M. V. Ramanamurthy and N. T. Reddy in the ministry of earth sciences in New Delhi.

The warning by the Indian scientists came from their research to pinpoint the reason for a four-hour delay between the 1945 Makran earthquake off Pakistan coast and preceding tsunami that lashed the coasts of Iran, Pakistan, Oman and India. "It took nearly four hours to reach Karachi instead of a few minutes it would have taken if it were caused by the earthquake," said Rajendran. They analysed archival data to conclude that the tsunami was not caused directly by the earthquake but actually by the submarine slide(s) it triggered. "The Makran subduction zone where the quake occurred has one of the thickest unconsolidated sediments," Rajendran said. "The time taken for slumping of the sediment along the continental slope delayed the arrival of the tsunami," he said.

"The Makran event challenges the rupture-centric views on tsunami generation and the hazard preparedness in the Indian Ocean region that ignores the alternate mechanism of landslides," the scientists report. The risk from landslides is compounded by the fact that the Bengal Fan is sitting atop huge reserves of gas hydrates — ice like substances consisting of water and natural gas — whose dissociation is known to have triggered the submarine slide elsewhere5. "Failures due to breaking of gas hydrates have not been reported so far in Bengal area, but that possibility should also be looked into," says Rajendran.

Brian Atwater of the US Geological Survey says there is great interest worldwide in landslides as causes of tsunamis. There is also a UNESCO proposal to study Makran submarine slides by a team of scientists from India, Iran and Pakistan but "it is too early to say much about this proposal," he told Nature India.

"It would be good to include impact of mud-slides in tsunami models," admits Satish Shetye, director of the National Institute of Oceanography in Goa that has developed tsunami models. "But, as I understand it, this is not easy to do."


References

  1. Rajendran, C. P. et al. Hazard implications of the late arrival of the 1945 Makran tsunami. Curr. Sci. 95, 1739-1743 (2008)
  2. Synolakis, C. E. et al. The Slump Origin of the 1998 Papua New Guinea Tsunami. P. R. Soc. A. 458, 763–789 (2002)
  3. Bornhold, B. D. et al. The Grand Banks Landslide-Generated Tsunami of November 18, 1929: analysis and numerical modeling. Geophys. Res. Abstr. 5, 01775 (2003)
  4. Billi, A. et al. Landslide triggered Italian tsunami previously linked to quake. Geophys. Res. Lett. 35, L06301(2008) | Article |
  5. Beget, J. E. Methane gas release from the Storegga submarine landslide linked to early Holocene climate change. Holocene 17, 291-295 (2007) | Article |