"When we saw the [storm] track, I said, 'Uh oh, this is not going to be good,'" said Mark Lander, a meteorology professor at the University of Guam.
"It would create a big storm surge. It was like Katrina going into New Orleans."
Forecasters began tracking the cyclone April 28 as it first headed toward India. As projected, the storm took a sharp turn eastward. But it didn't follow the typical cyclone track, which leads to Bangladesh or Myanmar's mountainous northwest.
Instead, the cyclone swept into the low-lying Irrawaddy River Delta in central Myanmar. The result was the worst disaster ever in the impoverished country.
It was the first time such an intense storm is known to have hit the delta, said Jeff Masters, co-founder and director of meteorology at the San Francisco-based Web site Weather Underground.
He called it "one of those once-in-every-500-years kind of things."
"The easterly component of the path is unusual," Masters said. "It tracked right over the most vulnerable part of the country, where most of the people live."
When the storm made landfall early Saturday at the mouth of the Irrawaddy River, the cyclone's battering winds pushed a wall of water as tall as 12 feet (3.7 meters) some 25 miles (40 kilometers) inland, laying waste to villages and killing tens of thousands. Most of the dead were in the delta, where farm families sleeping in shacks barely above sea level were swept to their deaths.
Almost 95 percent of the houses and other buildings in seven townships were destroyed, Myanmar's government says. UN officials estimate 1.5 million people were left in severe straits.
"When you look at the satellite picture of before and after the storm, the effects look eerily similar to Hurricanes Katrina and Rita in how it inundated low-lying areas," said Ken Reeves, director of forecasting for AccuWeather.com.
The Irrawaddy Delta "is huge, and the interaction of water and land lying right at sea level allowed the tidal surge to deliver maximum penetration of sea water over land," Reeves said.
"Storms like this do most of their killing through floods, with salt water being even more dangerous than fresh water."
The delta had lost most of its mangrove forests along the coast to shrimp farms and rice paddies over the past decade. That removed what scientists say is one of nature's best defenses against violent storms.
"If you look at the path of the [cyclone] that hit Myanmar, it hit exactly where it was going to do the most damage, and it's doing the most damage because much of the protective vegetation was cleared," said Jeff NcNeely, chief scientist for the International Union for Conservation of Nature.
"It's an expensive lesson, but it has been one taught repeatedly," he said. "You just wonder why governments don't get on this."
Some environmentalists suggested global warming may have played a role.
Last year the UN's Intergovernmental Panel on Climate Change concluded that warming oceans could contribute to increasingly severe cyclones with stronger winds and heavier rains.
"While we can never pinpoint one disaster as the result of climate change, there is enough scientific evidence that climate change will lead to intensification of tropical cyclones," said Sunita Narain, director of the India-based environmental group Centre for Science and Environment.
"Nargis is a sign of things to come," she said.
"The victims of these cyclones are climate change victims, and their plight should remind the rich world that it is doing too little to contain its greenhouse gas emissions."
Weather experts, however, are divided over whether global warming is a factor in catastrophic storms.
At a January conference of the American Meteorological Society, some experts postulated that warmer ocean temperatures may actually reduce the strength of cyclones and hurricanes.
Masters, at Weather Underground, said Wednesday that, in the case of Nargis, the meteorological data in the Indian Ocean region "is too short and too poor in quality to make judgments about whether tropical cyclones have been affected by global warming."
Despite assertions by Myanmar's military government that it had warned people about the storm, critics contend the junta didn't do enough to alert the delta and failed to organize any evacuations, perhaps resulting in unnecessary deaths.
"Villagers were totally unaware," said 38-year-old Khin Khin Myawe, interviewed in the hard-hit delta town of Labutta.
"We knew the cyclone was coming but only because the wind was very strong. No local authorities ever came to us with information about how serious the storm was."
The India Meteorological Department, one of six regional warning centers set up by the World Meteorological Organization (WMO), began sending regular storm advisories April 27. The information appeared in Myanmar's state-run newspapers, on radio, and on television 48 hours ahead of the storm.
But the international advisories said nothing about a storm surge. And Myanmar, unlike its neighbors Bangladesh and India, has no radar network to help predict the location and height of surges, the WMO said.
There also wasn't any coordinated effort on the part of the junta to move people out of low-lying areas, even though information was available about the expected time and location of landfall.
"How is it possible that there was such a great death toll in the 21st century, when we have imagery from satellites in real time and there are specialized meteorology centers in all the regions?" said Olavo Rasquinho of the UN Typhoon Committee Secretariat.
Bangladesh has a storm-protection system that includes warning sirens, evacuation routes, and sturdy towers to shelter people—measures that were credited with limiting the death toll from last year's Cyclone Sidr to 3,100.
Atiq A. Rahman, executive director of the Bangladesh Center for Advanced Studies and a disaster specialist, said Myanmar's death toll would have been lower if it had had such a system.
"Taking some action to move people from affected areas would have dramatically helped reduce the numbers of causalities. Absolutely," Rahman said.
But junta officials and some weather experts said evacuating a large area with millions of residents would have been nearly impossible, given the poor roads, the distance to some villages, and the likely refusal of some families to leave.
"Even if they warned them, they can't go anywhere. Or they are afraid to go anywhere, because they are afraid of losing their property," said Lander, the University of Guam professor.
"It is debatable how much of a mass exodus you could have had."
Lily Hindy contributed to this report.