Tropical Cyclone Nargis, which exacted a staggering human toll on the politically isolated and poor country of Myanmar, has demonstrated once again that there is an urgent need for a more robust infrastructure in developing countries for issuing and disseminating warnings of natural hazards. Unlike when the deadly Indian Ocean tsunami hit in 2004, government officials had at least one and a half days of warning before Nargis' arrival as a Category Three or Four cyclone on May 3rd. However, it's doubtful that the Burmese military-led government was able to get word to the residents of the hardest hit Irrawaddy delta region in time for them to protect themselves. Even if the warnings had reached this area, it's unclear what, if any, storm shelter options were available to people.
To get a firsthand perspective on how to improve the timeliness and effectiveness of weather warnings in developing countries, I turned to Peter J. Webster of the Georgia Institute of Technology, who is one of the lead scientists involved in a flood warning project in Bangladesh, Myanmar's low lying neighbor. Webster has directed the Climate Forecast Applications in Bangladesh (CFAB) project since 1999, and he is also a prominent scholar of tropical cyclones. For example, one study he served as the lead author of in 2005 broke new ground and proved controversial when it demonstrated a potential link between global climate change and tropical cyclone intensity.
The following are portions of our Q&A, which was conducted via email over the weekend. In it, Webster shares his perspective on what went wrong in Myanmar, what is going right in neighboring Bangladesh, and how climate change may complicate matters.
Keep reading for Andrew Freedman's Q&A with Dr. Peter Webster. Also, see our full forecast through the weekend.
CWG: The World Meteorological Organization (WMO) has reported that it gaveâ€¨ Burmese authorities 36-hours warning of the cyclone's strength and likely location of landfall. Based on your experiences in Bangladesh and elsewhere, what are some of the issues that the government of a poor, developing nation faces in order to get the word out to the people who are most likely toâ€¨ be affected by such a storm? â€¨
But I wonder if the IMD forecasts and warning were like shouting a warning to a deaf and blind person crossing a freeway that he/she is about to [be] run down.
PW: The WMO has a number of regional centers, one of which is located in India and is responsible for all of the North Indian Ocean forecasts and warnings. At the Indian Meteorological Department (IMD), forecasts are made with horizons of two to three days. Hence the warning of the "WMO" 36 hours in advance... But I wonder if the IMD forecasts and warning were like shouting a warning to a deaf and blind person crossing a freeway that he/she is about to [be] run down.
In order for the forecast to be useful, it has to be passed on to the governmental authorities who, in turn, need to pass it on to the people who are most in danger. And note too, that the people who are most in peril are often the poorest of the poor. So a forecast is one thing. Getting it to the people is something else. This "end-to-end" forecasting system requires planning. To be fair, tropical cyclones (and major ones at that) are rare in Burma-Myanmar. But, perhaps the point is why didn't Myanmar have a general disaster plan which they could have implemented for Nargis?
Something can be learned by the Bangladesh experience. In 1971 the Bohar Tropical Cyclone took 300,000+ lives principally drowned by storm surge. In 1991, about 100,000 were lost in similar circumstances. An almost identical storm to the 1971 [one] landed last November and about 10,000 were lost. Why this decrease in death rate with very similar storms? Bangladesh tried, with international help, to do something about it. Along the coast, cyclone shelters were built and sea walls (rather mounds) were erected and earth mounds were constructed. People had somewhere to go if they could be warned...
How is the danger ([of a] flood or tropical cyclone) communicated? There was essentially no communication in 1971 and by radio in 1991. This latter improvement has the flaw that someone has to be at a radio receiver and somehow this message has to be communicated down to the household. No easy job. BUT, in the event of the 2007 storm (Sidr) and the floods of last year, communication was swift all the way to the villager. This was accomplished by the development of cell phone networks. For example, our 8-day forecasts of flooding along the Brahmaputra [River] were communicated within hours to all those concerned.
Although the 10,000 lost with Tropical Cyclone Sidr was large, it was reduced by better forecasts and effective communication. I think that even with the improvements that have been made (forecasts and communication) there always will be a large loss of life. After all, delta regions are flat and there are limited places to where people can be evacuated. There are no freeways!
CWG: How does this cyclone compare with past cyclones in the Indian Ocean? â€¨â€¨
PW: Nargis fits well with the storms of the recent past. It was a major tropical cyclone (category four) when it hit but unusual in its east-northeast trajectory. What is dangerous about a category four tropical cyclone is that it has the potential for very large storm surges. And the trajectory couldn't be worse. It moved eastward along an east-west delta coastline so that the northward winds ahead of the storm piled up vast amount of water onto the essentially flat [land]. The situation could not be worse.
CWG: Is there any evidence of a climate change-related trend in cyclones inâ€¨ the Indian Ocean?
So my guess, after looking back over the last few years is that we are seeing a signal of global warming in the North Indian Ocean.
PW: Records over the Indian Ocean are not the best. There are a number of gaps in the records even in the satellite era in the early 1970s. But we do know two things: the storms that have occurred in the past few years (Gonu, Sidr and Nargis) have all been category four and five, and the sea surface temperature (SST) in the Indian Ocean has warmed more than any other tropical ocean basin.
We published a paper in Science in 2005 (you can get a copy here, follow the link to "downloadable papers.") We noted that [the] intensity of tropical cyclones had increased over the 1970-2004 period and that the number of category fours and fives has doubled. We found this increase in intensity to be consistent with the increase of SST in that a warmer ocean provides more fuel to create stronger storms. So my guess, after looking back over the last few years is that we are seeing a signal of global warming in the North Indian Ocean.
CWG: How effective have local resident point persons been in Bangladesh inâ€¨-disseminating flood forecast information in areas that lack theâ€¨ infrastructure (electricity etc.) through which to spread the word aboutâ€¨ impending flooding? â€¨â€¨
PW: As I mentioned earlier: enormously important. Cell phone technology solves the electricity problem of course. Earlier in the year I was in India in some very remote parts of the country. But, no matter how large the town or how small the village, I could communicate locally and overseas by cell phone. Towers a few miles apart are sound investments compared to land lines.
CWG: What are some of the key steps that a government of a poorer, low lying country can take to improve their capacity to respond to weather and climate-related disasters, and their resiliency when such disasters occur? â€¨â€¨
PW: The first thing is national and governmental commitment as has happened in Bangladesh. The second is the willingness to accept that there is a need for outside help. The third is to realize that each country has special vulnerabilities and that there are solutions for each. The fourth is to develop a communications network to transmit in an efficient manner state-of-the-art forecasts to where they can be used.
One of the problems for Bangladesh is how do you balance flooding problems with the impact of tropical cyclones? You build sea mounds to help protect from storm surge but that same sea mound will slow the release of water following flooding. But the common issue is that the storm shelters and mounds where people can go work for both floods and storm surges.