ISRO Research Could Aid Tsunami Warning System


Scientists with the Indian Space Research Organisation (ISRO) have combined satellite data with various ocean circulation and wave propagation models, along with seismic information, to create a simulation of the Dec. 26 Indian Ocean tsunami that closely matches up with the real thing.

The research could lead to better predictions of tsunamis like the earthquake-triggered tidal wave that brought widespread destruction to the Indian Ocean littoral region, especially Indonesia. The primary source of satellite data used in creating the simulation was the U.S.-French Topex-Poseidon ocean-altimetry satellite, but imagery from other remote-sensing spacecraft was used as well.

Preliminary findings of the work were published in the Feb. 10 issue of Current Science, a journal of the Indian Academy of Sciences in Bangalore. The paper, authored by Vijay Agarwal and colleagues at ISRO’s Space Applications Centre in Ahmedabad, said the tsunami provided a test case to see if the model can simulate the entire life cycle of such events.

“The results are very encouraging,” the scientists wrote. “Though our study was done after the event one should expect that this model could work in real time.”

Agarwal and his colleagues are conducting their research in consultation with India’s Department of Ocean Development, which has committed 1.25 billion rupees ($28 million) to a project to set up a tsunami warning system for the Indian Ocean region by 2007.

The model, called Tidal Ocean Atmospheric Surge and Tsunami, or TOAST, was specially designed for the Indian Ocean region, Agarwal said in a March 21 telephone interview. TOAST actually is an amalgam of different models related to tidal movements, ocean circulation, wave propagation and ocean-atmosphere interaction. Agarwal and his colleagues stitched the pieces together in such a way as to provide predictions of ocean surges in coastal areas.

The TOAST model is based in part on the U.S. National Oceanic and Atmospheric Administration’s Modular Ocean Model, Agarwal said. The Modular Ocean Model is a basic, 3D ocean-circulation model and was modified with permission for the Indian Ocean region, he said.

“The model requires geological information pertaining to the earthquake and uses explicit bathymetry, to approximately 4.5 kilometers horizontal resolution, based on Topex altimeter data combined with Earth observation data from other satellites,” Agarwal said in the interview. “The ocean components can be run live with satellite data or even after the event, as it happened in the tsunami case, when the analysis was performed several days later.”

The TOAST model closely approximated some the events and phenomena and associated with the actual tsunami, Agarwal said. The height of the simulated waves closely matched the real thing based on tide gauge readings, and the timing of the coastal strike matched up reasonably well with the time that was actually observed, he said.

However, the model underestimated the extent of the coastal surge due to what Agarwal said was “uncertain bathymetric information in the territorial regions.” The comparison between the simulation and the actual tsunami lend confidence to the TOAST model , but more work is needed to be able to better predict coastal inundation , he said. This will require using ships to more accurately chart ocean depths in the coastal areas, he added.

“It is a project we intend to complete in two months,” Agarwal said, referring to the ship survey. The model is flexible enough to accommodate ocean altimetry data that will become available once ISRO’s planned radar imaging satellite becomes operational. That satellite is slated to launch in 2006.

“The model could be run a priori for various seismic possibilities to generate potential inundation scenarios, which could form the basis for forewarning/advisories,” its developers claim in the Current Science paper.

Nevertheless, it remains unclear when or even if the TOAST model might be incorporated into India’s planned Tsunami warning system.

“We are aware of this model,” says K. Radhakrishna, director of the Hyderabad-based National Centre for Ocean Information Services at the Department of Ocean Development. “We have yet to see how we can integrate it in the proposed warning system.”