POLICY & PRACTICE

Reduced death rates from cyclones in Bangladesh: what more needs to be done?

Les taux de mortalité liés au passage de cyclones sont en diminution au Bangladesh: quelles mesures supplémentaires envisager?

Tras la reducción de las tasas de mortalidad por ciclones en Bangladesh: ¿qué otras acciones son necesarias?

Ubydul Haque^I,*; Masahiro Hashizume^I; Korine N Kolivras^II; Hans J Overgaard^III; Bivash Das^IV; Taro Yamamoto^I

^IDepartment of International Health, Institute of Tropical Medicine, Nagasaki University, 1-12-4 Sakamoto, Nagasaki City, 852-8523, Japan
^IIDepartment of Geography, Virginia Tech, Blacksburg, United States of America
^IIIDepartment of Mathematical Sciences and Technology, Norwegian University of Life Sciences, Ås, Norway
^IVLocal Government Engineering Department, Sher-e-Bangla Nagar, Dhaka, Bangladesh

ABSTRACT

Tropical storms, such as cyclones, hurricanes and typhoons, present major threats to coastal communities. Around two million people worldwide have died and millions have been injured over the past two centuries as a result of tropical storms. Bangladesh is especially vulnerable to tropical cyclones, with around 718 000 deaths from them in the past 50 years. However, cyclone-related mortality in Bangladesh has declined by more than 100-fold over the past 40 years, from 500 000 deaths in 1970 to 4234 in 2007. The main factors responsible for these reduced fatalities and injuries are improved defensive measures, including early warning systems, cyclone shelters, evacuation plans, coastal embankments, reforestation schemes and increased awareness and communication. Although warning systems have been improved, evacuation before a cyclone remains a challenge, with major problems caused by illiteracy, lack of awareness and poor communication. Despite the potential risks of climate change and tropical storms, little empirical knowledge exists on how to develop effective strategies to reduce or mitigate the effects of cyclones. This paper summarizes the most recent data and outlines the strategy adopted in Bangladesh. It offers guidance on how similar strategies can be adopted by other countries vulnerable to tropical storms. Further research is needed to enable countries to limit the risks that cyclones present to public health.

RÉSUMÉ

Les tempêtes tropicales, comme les cyclones, ouragans et typhons, constituent une menace majeure pour les populations côtières. Près de deux millions de personnes dans le monde ont trouvé la mort et des millions d'autres ont été blessées à cause du passage de tempêtes tropicales au cours des deux siècles derniers. Le Bangladesh est tout particulièrement vulnérable aux cyclones, ces derniers ayant causé 718 000 décès au cours des 50 dernières années. La mortalité liée aux cyclones au Bangladesh a toutefois été divisée par plus de cent au cours des 40 dernières années, passant de 500 000 décès en 1970 à 4234 en 2007. Les principaux facteurs de cette diminution du nombre des personnes décédées ou blessées sont de meilleures mesures de défense, notamment les systèmes d'alerte rapide, des abris capables de résister aux cyclones, les plans d'évacuation, les digues littorales, les programmes de reboisement et une sensibilisation et une communication accrues. Bien que les systèmes d'alerte aient été améliorés, évacuer avant un cyclone demeure un défi. D'importants problèmes résultant en effet de l'analphabétisme, du manque de sensibilisation et d'une mauvaise communication. Malgré les risques potentiels liés aux changements climatiques et aux tempêtes tropicales, il existe peu de connaissances empiriques sur le développement de stratégies efficaces réduisant ou atténuant les effets des cyclones. Le présent document résume les données les plus récentes, et présente la stratégie adoptée au Bangladesh. Il donne des indications sur la façon dont des stratégies similaires peuvent être adoptées dans d'autres pays vulnérables aux tempêtes tropicales. Des recherches supplémentaires sont nécessaires afin de permettre aux pays de limiter les risques que les cyclones font courir à la santé publique.

RESUMEN

Las tormentas tropicales, como los ciclones, huracanes y tifones, suponen una amenaza importante para las comunidades costeras. Durante los dos últimos siglos, unos dos millones de personas de todo el mundo han perdido la vida y varios millones han resultado heridas como resultado de las tormentas tropicales. Bangladesh es especialmente vulnerable a los ciclones tropicales, causantes de aproximadamente 718 000 muertes en los últimos 50 años. No obstante, la mortalidad relacionada con los ciclones en Bangladesh se ha reducido más de cien veces en los últimos 40 años, pasando de 500 000 muertes en 1970 a 4234 en 2007. Entre los principales factores que han contribuido a esta reducción en el número de muertes y heridos se encuentran unas mejores medidas defensivas, incluyendo los sistemas de alarma anticipada, los refugios anticiclónicos, los planes de evacuación, los muros de contención costeros, los esquemas de reforestación y el aumento de la concienciación y la comunicación. Aunque los sistemas de alarma han mejorado, la evacuación antes de un ciclón sigue constituyendo todo un reto cuyos mayores problemas son el desconocimiento, la falta de concienciación y las dificultades en la comunicación. A pesar de los riesgos potenciales del cambio climático y de las tormentas tropicales, existen pocos conocimientos empíricos sobre cómo desarrollar estrategias eficaces para reducir o mitigar los efectos de los ciclones. Este documento resume los datos más recientes y describe la estrategia adoptada en Bangladesh. Además, proporciona orientación sobre cómo otros países vulnerables a las tormentas tropicales pueden adoptar estrategias similares. Es necesario seguir investigando para permitir que los países limiten los riesgos que los ciclones suponen para la salud general.

Background

Cyclones and storm surges threaten coastal communities worldwide. The World Meteorological Organization defines a tropical cyclone as "a non-frontal synoptic scale cyclone originating over tropical or subtropical waters with organized convection and definite cyclonic surface wind circulation".¹ More specifically, a storm in the south-east Indian Ocean is cyclonic when the sustained wind speed is more than 33 nautical miles per hour (> 62 km/h). The storm surge represents a major cause of death and injury during a cyclone. A storm surge is the difference between the water level under the influence of a disturbance (storm tide) and the normal level that would have been reached in the absence of the meteorological disturbance.² Over the past two centuries, around two million people worldwide have died and millions have been injured as a result of tropical storms, including cyclones, hurricanes and typhoons.³ Globally, the number of cyclones has increased more than threefold (Fig. 1) from 1970 to 2006.⁴ The strength and number of major cyclones may be increasing because of higher sea surface temperatures associated with global warming.⁵ Tropical cyclones and storm surges are particularly severe in the Bay of Bengal region.

We review the impact of cyclones on health and livelihoods in Bangladesh, in light of the progress made and the challenges that remain. We explore cyclone-related severity and death in Bangladesh over the past 50 years, and discuss the experiences of other cyclone-afflicted countries. We also consider how we can learn from international experience to reduce the adverse health impacts of natural disasters. Finally, we recommend mitigation and adaptation strategies, and future research needs.

Cyclones in Bangladesh

Bangladesh is especially vulnerable to cyclones because of its location at the triangular shaped head of the Bay of Bengal,⁶ the sea-level geography of its coastal area, its high population density and the lack of coastal protection systems. During the pre-monsoon (April-May) or post-monsoon (October-November) seasons, cyclones frequently hit the coastal regions of Bangladesh. About 40% of the total global storm surges are recorded in Bangladesh,⁷ and the deadliest cyclones in the past 50 years, in terms of deaths and casualties, are those that have struck Bangladesh.⁸

The number and severity of cyclones in Bangladesh and the associated mortalities have varied greatly during the past 50 years (Table 1). The two deadliest cyclones occurred in 1970 and 1991, with > 500 000 and almost 140 000 deaths, respectively. However, during the past 20 years, Bangladesh has managed to reduce deaths and injuries from cyclones. For example, the most recent severe cyclone of 2007 caused 4234 deaths, a 100-fold reduction compared with the devastating 1970 cyclone.

In addition to the immediate death and suffering caused by such disasters, cyclones also have direct and indirect impacts on general public health, livelihoods, infrastructure, the economy and sociocultural foundations. They can affect access to food and drinking water, and increase the transmission risks of infectious diseases, such as diarrhoea, hepatitis, malaria, dengue, pneumonia, eye infections and skin diseases,¹² thus contributing to the interruption of livelihoods. Surface water, the main source of drinking water in coastal regions of Bangladesh, becomes contaminated by saline intrusion and poor sanitation systems.^13,14 Open latrines and poor sanitation are common in rural Bangladesh and coastal areas, and cyclones make this situation worse.¹³ The lack of safe drinking water may be the most important cause of the spread of waterborne diseases after a cyclone. Other causes are through indirect impacts such as damaged infrastructure, population displacement, reduced food production and the release of contaminants into the water (e.g. from storage and waste disposal sites). Childhood malnutrition is already a serious issue in Bangladesh, and the loss of crops and reduced access to fish compounds the problem. Indirect health-related impacts, such as increased suicide and crime rates, and adverse pregnancy outcomes, are clearly associated with cyclones.¹⁵ These tend to increase in the post-disaster period, as a result of post-traumatic stress and depression. Literacy rates are low and poor knowledge of environmental health issues creates additional problems following a cyclone.

Progress and challenges

In the past 50 years, Bangladesh has learnt how to adapt to recurrent cyclones and has succeeded in significantly reducing cyclone-related deaths. This has been achieved by modernizing early warning systems, developing shelters and evacuation plans, constructing coastal embankments, maintaining and improving coastal forest cover and raising awareness at the community level.

Cyclone preparedness has improved following the launch of the Cyclone Preparedness Programme by the Bangladesh Red Crescent Society in 1970. The programme's goal is to minimize the loss of lives and property in cyclonic disasters by strengthening and developing disaster preparedness and response capacity in coastal communities, and by increasing the effectiveness of volunteers. The programme's activities include: disseminating cyclone warning signals issued by the Bangladesh Meteorological Department through an extensive telecommunication network; providing and assisting in first aid, rescue, relief and rehabilitation operations; and coordinating and building community capacity, disaster management and development activities. The Bangladesh Disaster Management Bureau also issues cyclone alerts through the national media as soon as a cyclone is detected by environmental satellites. The Bangladesh Meteorological Department has three radar stations in Dhaka, Khepupara and Cox's Bazar that transmit minute-by-minute weather updates.¹⁶ The Department also receives information from the National Oceanic and Atmospheric Administration in the United States of America (USA) and from a Japanese satellite via the Bangladesh Space Research and Remote Sensing Organization.¹⁶ The effective early warning system provided by the government in advance of Cyclone Sidr in 2007 enabled the successful evacuation of coastal communities resulting in fewer than expected deaths. Initiatives at central and local governmental, nongovernmental and community levels seem to be key for success in minimizing cyclone-related mortality.

Apart from early warning systems, other measures such as cyclone shelters and coastal embankments have contributed to reducing death rates in Bangladesh. Prior to 2007, the country had 1500 shelters, each capable of offering refuge to up to 5000 people in coastal districts. After Cyclone Sidr, the Bangladesh government initiated the construction of 2000 new cyclone shelters in 15 low-lying coastal districts, but the number and location of shelters remain inadequate for the population.¹⁵ Bangladesh has more than 700 km of coastline. Since 1960 a series of embankments have been constructed to protect coastal regions, including around 4000 km of coastal embankments surrounding the Bay of Bengal and offshore islands.¹⁷ Coastal vegetation was found to be protective during Cyclone Sidr when mangrove forests saved the south-western part of Bangladesh and, during a different storm, reduced the death toll from a cyclone in India in 1999.¹⁸ Reforestation of approximately 1200 km² of mangrove forests in Bangladesh has been carried out to mitigate cyclone risk.¹⁹ Under the Cyclone Preparedness Programme, Bangladesh has implemented awareness campaigns to disseminate information about cyclone warning signals and preparedness measures, using meetings, discussions, posters, leaflets, film shows and demonstration performances.²⁰ Although there is currently no scientific evidence regarding the precise impact of shelters, coastal embankments or awareness programmes on cyclone-related mortality, they appear to have saved millions of lives. Continued technological advances will increase preparedness and help mitigate the effect of cyclones in Bangladesh.

Despite improvements in warning systems, pre-cyclone evacuation remains a challenge. Illiteracy, lack of awareness and communication problems mean that some people do not understand or follow the warnings. Instead of moving to cyclone shelters, people in coastal areas often still believe in a wait-and-see approach.^21,22 Fear of property loss and previous false warnings also limit the numbers who evacuate to shelters.²¹ Others refuse to evacuate because of the poor condition of the public cyclone shelters, attributes of the warning message, individual perceptions and beliefs, including thinking that their house can withstand a cyclone.²² Building structures of concrete or brick prevent human loss, as people who shelter in such structures generally survive, while the death rate can be double in populations without access to sturdy shelters.²² Maintenance of and access to cyclone shelters are important factors in enabling people to quickly find adequate protection. For example, only two out of every five shelters were usable during the 1991 cyclone, because of flooding.²³ People also had lack of access to shelters during Cyclone Sidr in 2007.

Dissemination of warning messages presents another challenge as most residents in coastal areas of Bangladesh have no access to radio or television. Some Bangladeshis rely on natural warning signs, such as unusual animal behaviour and weather and ocean patterns, to prepare for the impacts of a cyclone,²³ however these signs may be unreliable and inconsistent. In remote areas, the use of megaphones by volunteers (more than 20 000 during cyclone alerts)²² is not always reliable due to wind direction affecting sound transmission, and batteries for megaphones and microphones may not be locally available. Significantly, households with radios had lower death rates during cyclones than those without radios.²²

Experiences from other countries

By examining the impacts of and responses to cyclones in other countries, we can improve our understanding of effective strategies for preventing the loss of life. Cuba has significantly improved its pre- and post-cyclone early warning and evacuation systems and health services and has introduced a cyclone preparedness programme for primary school children. Universal education and the eradication of illiteracy are important to improve awareness of the risks associated with hurricanes and the understanding of government warnings. Cuba also has a population with a very high level of civil participation and a comprehensive primary health care system.²⁴

In early May 2008, Cyclone Nargis struck Myanmar with sea surges and wind speeds > 200 km/h; more than 140 000 people died or were presumed dead and almost 2.4 million people were seriously affected.^25,26 There was an international warning of the approaching Cyclone Nargis several days before its landing, but poor dissemination of information and lack of governmental responsibility were thought to have contributed to the outcome. Local authorities and populations were not proactive in their planning and response. No information on cyclone shelters in Myanmar had been published before the cyclone, and there was a lack of awareness and political will and a poor health infrastructure. Private organizations had to quickly decide how to become involved in relief distribution, with some organizations taking on relief work as a completely new task. There were delays in evacuating people and the international community was not allowed to access the most affected areas. A lack of boats also contributed to the problem. Interestingly, however, emergency projects after Cyclone Nargis opened up the way for peace building efforts in areas that had previously been difficult to access by the international community.²⁷

In February 2011, Cyclone Yasi hit Queensland, Australia. The cyclone was 500 km wide with an eye of 100 km in diameter and 285 km/h wind speeds. Local and district disaster management committees initiated their disaster management plans in advance. The media played a vital role in informing the public about weather events, assistance and evacuation locations. Evacuation, including hospitals, was completed more than four hours before the cyclone struck. Aircraft were prepared for evacuations after the cyclone. Considering the magnitude of its destruction capability, not a single person died as a direct result of the cyclone. This was achieved through thorough preparedness and early warning systems.²⁸

Even with public warnings before Hurricane Katrina in the USA, twothirds of the more than 1800 fatalities were reportedly caused by drowning as a result of cyclone-related storm surges and floods.²⁹ The early evacuation of 1589 people from New Orleans to Oklahoma was done based on the results of a rapid needs assessment.³⁰ Due to the successful coordinated evacuation of hospitals in the city of New Orleans, no patient deaths or injuries were reported.³¹

Outbreaks of cholera, diarrhoea, malaria and dengue have been common after cyclones in India and in several African and Central American countries.^32-36 Careful preparation for epidemics before the arrival of a cyclone is important to ensure a rapid response and control of outbreaks. Along with high death rates, the Philippines has experienced outbreaks of leptospirosis infection caused by coastal flooding after typhoons.³⁷ Basic hygiene kits were distributed to affected communities following the typhoon to reduce waterborne disease. Early warning systems and evacuation programmes have recently improved, and better coordination of relief efforts was also reported to have reduced typhoonrelated health injuries and increased relief distribution. In October 2010, the early evacuation of 3066 people before Typhoon Megi saved lives.³⁸

Cyclones are also responsible for many indirect traumas and mental disorders in different parts of the world. A high incidence (30.6%) of post-traumatic stress disorder was reported after a cyclone that struck India in 1999,³⁹ and a high prevalence of post-traumatic stress disorder and major depressive symptoms have also been reported following cyclones in India, Nicaragua, Sri Lanka and the USA.^40-44 These mentalhealth problems might have long-term impacts on health. However, these issues have so far been neglected in Bangladesh because of limited resources and poor health infrastructure. Post-disaster psychological care services should be developed including: screening of affected populations; prioritizing interventions on the basis of risk assessment; providing trauma/grief-focused interventions; and monitoring recovery.

Cyclone-prone countries should consider investing in the construction of coastal embankments and reforestation programmes, as Bangladesh and other countries have done. Awarenessbuilding programmes and learning from previous cyclones have saved lives in Bangladesh.

Although observational evidence from previous years does not show a clear trend in the number of tropical cyclones occurring, climate change is likely to cause an increase in the intensity of tropical storms.⁴⁵ It is crucial that countries that experience regular cyclones consider the Bangladeshi experience to minimize the loss of human lives.

Conclusions

Instead of developing large cyclone shelters, a dense network of small, sturdy and safe multipurpose buildings should be developed. Considering the population density, cyclone shelters should be established within a 2 km walking distance of households and villages. Geographic Information Systems and remote sensing technology should be used to determine the best locations in terms of factors such as access, road networks and population density. Schools, mosques, local government buildings or other locations where people congregate represent potential locations for these shelters. This should be given the highest priority in cyclone-preparedness programmes.

Bangladesh is now fully covered by mobile telecommunication networks; distributing cyclone warning messages via mobile phones is thus a good option. Colourful hot air balloons can be used to convey cyclone-warning messages in remote and coastal areas of Bangladesh.

The potential for the breakdown of water and sanitation systems during a cyclone should be considered carefully in the planning, design and implementation of future housing developments. This will help prevent vector- and waterborne disease outbreaks.

Initiatives to collect and store drinking water should also be considered. Harvesting rain water during a cyclone can be an option.

Coastal embankment projects should be extended to all coastal areas. Existing embankments should be repaired and maintained. Careful planning with sufficient sluice gates, especially in the south-eastern area of Bangladesh, will protect against both flash floods and storm surges during a cyclone,⁴⁶ and will also help protect cropland, fisheries and livestock.

Operational research should be conducted on the precise impacts of cyclone shelters, coastal embankments and awareness programmes on cyclonerelated mortality. Additionally, research should be conducted on how to reduce drowning-related deaths during floods caused by cyclones.

The development of a 500 metre coastal mangrove forest zone will further reduce the vulnerability to cyclones, which is especially important given the likelihood of a rise in sea level and an increase in tropical storm frequency and strength due to climate change.⁴⁷

Based on the elevation of houses/residential areas in relation to nearby streams, maps of areas at high risk of flooding can be prepared to use during evacuations ahead of cyclone-related coastal surges.

Planners, policy-makers and development practitioners should endeavour to incorporate local knowledge into environmental and adaptation strategies. The building code in coastal zones should be changed to ensure that concrete houses are raised 3 metres off the ground. More broadly, a more compact development style may be recommended.

To increase people's awareness of the severity of cyclone hazards, the Bangladesh government and nongovernmental organizations should further strengthen the existing awareness programme and initiate educational campaigns in coastal districts to ensure prompt use of public shelters during cyclones. Awareness should focus on public health and hygiene issues. The awareness programme could target primary school children, following the Cuban model, which represents an excellent example for Bangladesh. Some operational research should be conducted in this regard.

People's misconceptions about the strength of their houses, a lack of interest in moving to a cyclone shelter and other potential risk factors should be identified through qualitative research. The design and delivery of community cyclone-preparedness education should be based on these research findings. Community-based volunteer intervention programmes should be introduced without further delay.

Cyclone-related loss in terms of economic and human capital is exacerbated by poverty and poor infrastructure in coastal areas of Bangladesh. Donor agencies, politicians and planners in Bangladesh should take this into account in future planning of coastal zones.

Industrialized countries and newly emerging industrialized countries should provide financial support to vulnerable countries to help them adapt to and mitigate cyclone-related risks. At the same time, all countries should reduce their emissions of carbon dioxide and other greenhouse gases.

Acknowledgements

Ubydul Haque is also affiliated with the Department of Mathematical Sciences and Technology, Norwegian University of Life Sciences, Ås, Norway.

Competing interests: None declared.

References

1. Severe weather information centre. Geneva: World Meteorological Organization; 2010. Available from: http://severe.worldweather.org/tc/swi/acronyms.html [accessed 25 October 2011]         

2. Tropical cyclone operational plan for the Bay of Bengal and the Arabian sea. Geneva: World Meteorological Organization; 2007 (WMO/TD-No.84.I-4).         

3. Shultz JM, Russell J, Espinel Z. Epidemiology of tropical cyclones: the dynamics of disaster, disease, and development. Epidemiol Rev 2005;27:21-35. doi:10.1093/epirev/mxi011 PMID:15958424         

4. The international disaster database [Internet site]. Brussels: Centre for Research on the Epidemiology of Disasters; 2011. Available from: http://www.emdat.be/result-country-profile [accessed 25 October 2011]         

5. Romm J. Climate change is increasing the frequency of category 5 storms. Grist, 4 September 2007. Available from: http://www.grist.org/article/hurricanes-are-getting-stronger-thanks-to-global-warming [accessed 25 October 2011]         

6. Murty TS, Neralla VR. On the recurvature of tropical cyclones and the storm surge problem in Bangladesh. Nat Hazards 1992;6:275-9. doi:10.1007/BF00129512        

7. Murty TS. Storm surges meteorological ocean tides. Can J Fish Aquat Sci 1984;212:897.         

8. Quadir DA, Iqbal AM. Tropical cyclones: impacts on coastal livelihoods. Gland: International Union for Conservation of Nature; 2008.         

9. Cyclone Shelter Preparatory Study: feasibility study. Brussels: European Commission; 1998.         

10. Dasgupta S, Huq M, Khan ZH, Ahmed MM, Mukherjee N, Khan MF, et al. Vulnerability of Bangladesh to cyclones in a changing climate: potential damages and adaptation cost (World Bank Policy Research Working Paper no. 5280). Washington: The World Bank; 2010.         

11. Karmakar S. The impact of tropical cyclones on the coastal regions of SAARC countries and their influence in the region. Dhaka: SAARC Meteorological Research Centre; 1998.         

12. Sommer A, Mosley WH. The Lancet-Saturday 13 May 1972. 1972. Epidemiol Rev 2005;27:13-20. PMID:15958423        

13. Haque CE, Blair D. Vulnerability to tropical cyclones: evidence from the April 1991 cyclone in coastal Bangladesh. Disasters 1992;16:217-29. doi:10.1111/j.1467-7717.1992.tb00400.x PMID:20958747        

14. Unicef team. Health effects of the 1991 Bangladesh cyclone: report of a UNICEF evaluation team. Disasters 1993;17:153-65. doi:10.1111/j.1467-7717.1993.tb01142.x PMID:20958764        

15. Bangladesh to build 2 000 cyclone shelters. Reuters, 8 January 2008. Available from: http://www.reuters.com/article/2008/01/08/usbangladesh-cyclone-shelters-idUSDHA10588420080108 [accessed 25 October 2011]         

16. Haque CE. Climatic hazards warning process in Bangladesh: experience of and lessons from the 1991 April cyclone. Environ Manage 1995;19:719-34. doi:10.1007/BF02471954        

17. Statistics of Bangladesh water development board 1998. Dhaka: Bangladesh Water Development Board; 2000.         

18. Das S, Vincent JR. Mangroves protected villages and reduced death toll during Indian super cyclone. Proc Natl Acad Sci USA 2009;106:7357-60. doi:10.1073/pnas.0810440106 PMID:19380735        

19. Saenger P, Siddiqi NA. Land from the sea: the mangrove afforestation program of Bangladesh. Ocean Coast Manage 1993;20:23-39. doi:10.1016/0964-5691(93)90011-M        

20. Harun-Al-Rashid A. Cyclone preparedness programme. Dhaka: Bangladesh Red Crescent Society; 1997.         

21. Bern C, Sniezek J, Mathbor GM, Siddiqi MS, Ronsmans C, Chowdhury AM et al. Risk factors for mortality in the Bangladesh cyclone of 1991. Bull World Health Organ 1993;71:73-8. PMID:8440041        

22. Chowdhury AM, Bhuyia AU, Choudhury AY, Sen R. The Bangladesh cyclone of 1991: why so many people died. Disasters 1993;17:291-304. doi:10.1111/j.1467-7717.1993.tb00503.x PMID:20958772        

23. Howell P. Indigenous early warning indicators of cyclones: potential application in coastal Bangladesh (Disaster studies working paper 6). London: Benfield Hazard Research Centre; 2003.         

24. Miranda DS, Choonara I. Hurricanes and child health: lessons from Cuba. Arch Dis Child 2011;96:328-9. doi:10.1136/adc.2009.178145 PMID:20861403        

25. Lateef F. Cyclone Nargis and Myanmar: a wake up call. J Emerg Trauma Shock 2009;2:106-13. doi:10.4103/0974-2700.50745 PMID:19561970        

26. Stone R. Myanmar. One year after a devastating cyclone, a bitter harvest. Science 2009;324:715. doi:10.1126/science.324_715 PMID:19423794        

27. Listening to voices from inside: Myanmar civil society's response to cyclone Nargis. Phnom Penh: Center for peace and conflict studies; 2009.         

28. Noble N. A whirlwind response tropical cyclone Yasi: a success story for Australian EMS. JEMS 2011;36:54-9. PMID:21550497        

29. Jonkman SN, Maaskant B, Boyd E, Levitan ML. Loss of life caused by the flooding of New Orleans after Hurricane Katrina: analysis of the relationship between flood characteristics and mortality. Risk Anal 2009;29:676-98. doi:10.1111/j.1539-6924.2008.01190.x PMID:19187485        

30. Rodriguez SR, Tocco JS, Mallonee S, Smithee L, Cathey T, Bradley K. Rapid needs assessment of Hurricane Katrina evacuees - Oklahoma, September 2005. Prehosp Disaster Med 2006;21:390-5. PMID:17334185        

31. Gallagher JJ, Jaco M, Marvin J, Herndon DN. Can burn centers evacuate in response to disasters? J Burn Care Res 2006;27:596-9. doi:10.1097/01. BCR.0000235462.17349.03 PMID:16998390        

32. Kondo H, Seo N, Yasuda T, Hasizume M, Koido Y, Ninomiya N et al. Post-flood epidemics of infectious diseases in Mozambique. Prehosp Disaster Med 2003;17:126-33.         

33. Morrow MG, Johnson RN, Polanco J, Claborn DM. Mosquito vector abundance immediately before and after tropical storms Alma and Arthur, northern Belize, 2008. Rev Panam Salud Publica 2010;28:19-24. doi:10.1590/S1020-49892010000700003 PMID:20857016        

34. Pan American Health Organization. Impact of Hurricane Mitch on Central America. Epidemiol Bull 1998;19:1-13. PMID:10330785        

35. Bhunia R, Ghosh S. Waterborne cholera outbreak following Cyclone Aila in Sundarban area of West Bengal, India, 2009. Trans R Soc Trop Med Hyg 2011;105:214-9. doi:10.1016/j.trstmh.2010.12.008 PMID:21353273        

36. Mason J, Cavalie P. Malaria epidemic in Haiti following a hurricane. Am J Trop Med Hyg 1965;14:533-9.         

37. McCurry J. Philippines struggles to recover from typhoons. Lancet 2009;374:1489. doi:10.1016/S0140-6736(09)61888-2 PMID:19891040        

38. Philippines: Typhoon Megi situation report no. 1. New York: Office for the Coordination of Humanitarian Affairs; 2010.         

39. Kar N, Mohapatra PK, Nayak KC, Pattanaik P, Swain SP, Kar HC. Post-traumatic stress disorder in children and adolescents one year after a super-cyclone in Orissa, India: exploring cross-cultural validity and vulnerability factors. BMC Psychiatry 2007;7:8. doi:10.1186/1471-244X-7-8 PMID:17300713        

40. Goenjian AK, Molina L, Steinberg AM, Fairbanks LA, Alvarez ML, Goenjian HA et al. Post-traumatic stress and depressive reactions among Nicaraguan adolescents after hurricane Mitch. Am J Psychiatry 2001;158:788-94. doi:10.1176/appi.ajp.158.5.788 PMID:11329403        

41. Kar N, Bastia BK. Post-traumatic stress disorder, depression and generalised anxiety disorder in adolescents after a natural disaster: a study of comorbidity. Clin Pract Epidemiol Ment Health 2006;2:17. doi:10.1186/17450179-2-17 PMID:16869979        

42. Patrick V, Patrick WK. Cyclone '78 in Sri Lanka - the mental health trail. Br J Psychiatry 1981;138:210-6. doi:10.1192/bjp.138.3.210 PMID:7272612        

43. Rhodes J, Chan C, Paxson C, Rouse CE, Waters M, Fussell E. The impact of hurricane Katrina on the mental and physical health of low-income parents in New Orleans. Am J Orthopsychiatry 2010;80:237-47. doi:10.1111/j.19390025.2010.01027.x PMID:20553517        

44. Ruggiero KJ, Amstadter AB, Acierno R, Kilpatrick DG, Resnick HS, Tracy M et al. Social and psychological resources associated with health status in a representative sample of adults affected by the 2004 Florida hurricanes. Psychiatry 2009;72:195-210. doi:10.1521/psyc.2009.72.2.195 PMID:19614556        

45. Climate change 2007: synthesis report. Valencia: Intergovernmental panel on climate change; 2007.         

46. Choudhury NY, Paul A, Paul BK. Impact of coastal embankment on the flash flood in Bangladesh: a case study. Appl Geogr 2004;24:241-58. doi:10.1016/j. apgeog.2004.04.001        

47. Liao C, Luo Y, Fang C, Li B. Ecosystem carbon stock influenced by plantation practice: implications for planting forests as a measure of climate change mitigation. PLoS ONE 2010;5:e10867. doi:10.1371/journal.pone.0010867 PMID:20523733        

Submitted: 16 March 2011
Revised version received: 4 September 2011
Accepted: 6 September 2011
Published online: 24 October 2011

* Correspondence to Ubydul Haque (e-mail: ubydul.haque@umb.no).