Our family

Entries in Burma (1)


Anatomy of a complex disaster: Cyclone Nargis

Cyclone Nargis was a category 4 cyclone which hit the Irrawaddy delta of Burma on the 2nd and 3rd of May, with winds of up to 215km/hr. Due to negligible preparation and no communication of the threat to the public, around 140 000 people were killed, and another 2.4 million strongly affected. The national response following Cyclone Nargis was slow and unfairly distributed, and the Burmese government sabotaged the international aid response. These anthropomorphic factors made Cyclone Nargis a complex disaster, with damage and destruction orders of magnitude greater than they could have been. Particular emphasis in this essay is placed on dissecting the factors which created a complex disaster out of a natural event, and how they could have been avoided or mitigated.


Historical contributions to the complexity of the disaster

In analysing Cyclone Nargis as a complex disaster it is important to consider the context, as the same cyclone in many other parts of the world would not have wrought the same destruction. At a national level, the context is the situation of Burma prior to Cyclone Nargis striking. Burma is a large South-East Asian country, with a population of around 55 million. Burma is extremely poor, with a GDP per capita of only $1,040, giving it an international rank of 162 in economic development. Economic growth in Burma has been stagnant compared to the surrounding “Asian Tiger” economies. This economic under-development is due to poor governance and a long-running civil war over the past fifty years.

Prior to 1962, and subsequent to independence from the British Empire, Burma had a democratic civilian government. However, in 1962, General Ne Win led a military coup, allowing a military junta to seize power. Over the subsequent decades, the military junta has becoming increasingly centralised and authoritarian, with restrictive laws placed on the population. Reports of human rights abuses resulted in Burma becoming a pariah state. In 1990, the first free multiparty elections were held, however the military government refused to step down after losing the election, and instead imprisoned the election opposition leader, and Nobel Peace Prize winner, Aung San Suu Kyi. The imprisonment of Aung San Suu Kyi, sporadically renewed (including in the immediate aftermath of the cyclone) created very poor international relationships between Burma and the developed world, with sanctions being placed on Burma by both the EU and the US. In late 2007 large anti-government protests were held around the country and were widely reported in the international press, along with the violent backlash by the junta, against the Burmese government’s wishes, creating further tension between the government and the international community.

Another relevant national factor is the long-running civil war fought between the Burmese government and the Karen National Union, representing the Karen ethnic group. This stretches back to 1948, since Burma gained independence from Great Britain, making it the longest running active war in the world. The Karen claim the right to self-determination while the Burmese government (both the previous democratic government and the current junta) claim national cohesiveness. Religion also plays a role in the war, with the Karen being predominately Christian while the official state religion is Buddhism. The centre theatre for the civil war is Karen State, a division of Burma to the east of the country, in the hilly forested region that borders Thailand. The war has resulted in over 500,000 people being displaced, either internally or as refugees to Thailand, and was escalated by the Burmese government in 2006. While Karen State has been the centre of the civil war, due largely to advantageous terrain for asymmetric war, the other major Karen population centre, the Irrawaddy delta, has also been involved, being a source of recruits and the site of attempted rebellion.

The Irrawaddy river delta was the region most heavily affected by Cyclone Nargis. This was an unfortunate region of Burma to have been struck for several reasons. One is the population composition of the Irrawaddy Division. The area is heavily populated, with 6.5 million people and a population density more than double that of the rest of the country (182/km2 vs 75/km2). Ethnically, the region is largely Bamar (the dominant ethnic group of Burma) and Karen, creating tensions with the Burmese government as outlined above. A second detrimental factor is the geography of the region, with low lying land, mostly below five metres above sea level. A third factor is the poverty of the region, being one of the economically least developed regions of Burma, with 26% of households below the poverty line. Despite the poverty, the region is highly fertile. It was once called the “Rice Bowl” of the British Empire and before the cyclone produced 65% of the national rice crop, making an agricultural disaster in the Irrawaddy Division highly detrimental to the entire country.

The combination of poverty, poor geography and a history placing the ethnic composition at odds with the government therefore placed the Irrawaddy delta at a high risk for complications arising in the context of a natural disaster. Furthermore, the low point in international relations between the Burmese government and the rest of the world contributed to the reduced effectivity of the ‘responsibility to protect’ provisions of international bodies, as detailed later on.


Preparedness of the health system for disaster

The health situation in Burma before the cyclone outperformed the economic development of the country, perhaps due to the abundant agricultural production. The DALY (disability-adjusted life years) average was 51.6 years, only slightly below average for countries in the region and, with a ranking of 139, better than many countries with greater economic development. Regardless of this good condition relative to the economic situation, the mortality rates were high. The adult mortality rate was 334 males per 1000 and 219 females per 1000, and the mortality rate under 5 was 105 per 1000 births. In particular, Burma had a high burden of infectious disease, with endemic malaria (in 80% of the country), Japanese encephalitis, diphtheria, pertussis and rubella. Polio is still present in the country, with 15 confirmed cases in 2007. Vaccination for both measles and polio was sub-optimal, with around 80% coverage (below the 90-95% coverage required to block an epidemic). Overall, infectious diseases accounted for five of the top ten causes of death in Burma prior to the cyclone. These diseases also have a disproportionate effect on total years of life lost in Burma, due to heightened death rates among youths. In particular, >62% of the death rate among the under fives is due to infectious disease. A further problem is the wide prevalence of sub-clinical Vitamin A deficiency.

The poor health situation in Burma also hides a far more dire situation in the Burmese health system. While the DALY average ranking is 139 out of 191 and the economic development ranking is 162 out of 191, the health system ranking is 190 out of 191, making it the second to worst health system in the world, behind only Sierra Leone. This is largely due to low government spending, far below the average of even low income countries in both dollar expenditure and as a proportion of GDP. Making matters worse is the unfair distribution of health services, with a fairness of financial contributions ranking of 190 out of 191.

As a result, with a high infectious disease burden, sub-optimal vaccination, sub-clinical vitamin A deficiency and almost non-existent public health infrastructure, the population of Burma before Cyclone Nargis was highly susceptible to negative health outcomes of disasters, with little capacity for absorbing epidemiological insult, except in the area of malnutrition, where the population was relatively well-off prior to Cyclone Nargis.


Cyclone Nargis and its immediate impact

Cyclone Nargis built up in late April over the warm waters of the Andaman Sea, south of Burma. The heating of air over warm water produces a high condensation low pressure system due to the rising of warm air. The low pressure system became a deep depression on the 27th of April, and moved west into the Bay of Bengal. On the 28th of April the winds picked up over warm waters and the deep depression formed a cyclonic storm. The cyclone peaked with winds at 160km/hr on the 29th of April, and was predicted to hit India or Bangladesh as a weakened tropical storm. On the 1st of May, however, Cyclone Nargis turned east and starting approaching Burma, picking up strength and speed. At this time, the Indian Meteorological Department identified that the storm was approaching the Irrawaddy delta and communicated the warning to the Department of Meteorology and Hydrology in Burma. The Burmese government published an altered official forecast in “The New Light of Myanmar”, predicting a far weaker storm than the advice they were given: “the severe tropical cyclone Nargis … is forecast to cross the coast during the next 36 hours … Under the influence of this storm, rain or thundershowers will be widespread …  frequent squalls with rough seas will be experienced off and along the Myanmar coast. Surface wind speed in squalls may reach 50 mph…”.

When Cyclone Nargis struck the Irrawaddy delta on the 2nd of May, it had surpassed its previous peak, with winds of 215km/hr making it a category four cyclone. The cyclone was able to stay at peak strength as it moved eastwards across the Irrawaddy river delta due to the proximity of the warm water along the extensive shallow continental shelf of the Andaman Sea (Figure 5). During the 2nd and 3rd of May, as Cyclone Nargis tracked along the Irrawaddy delta coastline, the cyclone was able to build up a storm surge of seawater 4 metres high. As the river delta is at sea-level for 200km along the coast, and the majority of land in the Irrawaddy Division is below 5 metres in elevation, the storm surge allowed an influx of seawater as far as 40km from the coast. It was only when Cyclone Nargis turned north and left the oceanic influences of the Andaman Sea that it lost its destructive power. Therefore, for population susceptibility, geographic vulnerability and meteorological propensity, “The path of the storm could not have been worse”.

Without any attempt at evacuation or storm threat mitigation, the effect of Cyclone Nargis was extreme. Around 140 000 people were killed, 80% of them due to the storm surge, which washed away 100 villages. Another 2.4 million people were severely affected, by injury (more than 30,000) or damage to property. The winter rice crop and seed stocks were lost and the fertile soils were contaminated with salt from the ocean surge. 50,000 acres were so salinated that they will not be recoverable and 500,000 acres were rendered unavailable for monsoon planting. 82% of homes in the region were destroyed (57% totally and 25% partially).

The neglect of the military junta in providing storm protection infrastructure and the disinformation broadcast in regard to the strength of the cyclone were the most significant factors in the death toll caused by the cyclone. The vast majority of deaths occurred during the storm itself, 80% by the storm surge alone. This means that even a perfect government and international aid response after the cyclone struck (which by no means occurred, see below) would only have been able to slightly mitigate the damage caused. Effective preparation to mitigate the immediate damage of the storm was within the available resources of the Burmese government. Tropical cyclone formation in the North Indian Ocean has been increased in recent decades, probably due to climate change effects warming the surface waters, indicating the increased risk to the region. In November 2007 an even stronger storm (category five) formed in the same area, Tropical Cyclone Sidr. It struck Bangladesh and killed 3500 people. Previous cyclones in 1970 and 1991 had killed 300,000 and 190,000 respectively. Cyclone Sidr was different because of the low-technology low-cost preparations that had been put in place. An extended cyclone forecast was utilised and a pre-prepared national emergency network was initiated. The warning was spread to remote areas by an established mobile phone network and then communicated to the public via megaphones. Notably, in the 1991 Bangladesh cyclone, 40% of family members in areas without shelters died, while only 3.4% in areas with shelters died. This led to the extensive building of a series of storm shelters and water control systems that in 2007 allowed over two million people to evacuate and shelter through the storm, giving a death toll that was 2% that of Cyclone Nargis. This demonstrates the over-whelming necessity of prior preparation and the complete failure of the Burmese government in this regard.


The response to Cyclone Nargis

The standard post-disaster government response to a cyclone should have consisted of immediate rescue/retrieval, evacuation, triage and clinical care, definitive surgery and rehabilitation. This would then be followed by public health measures such as the provision of clean water and sanitation, measles immunisation, simple treatment of dehydration for diarrhoea, supplementary feeding for malnourished, micronutrient supplements and public health surveillance. However the Burmese government failed to implement any of these standard measures in any meaningful way. The government set up 45 temporary camps as showcases for the international community, but failed to provide support to the wider affected population, especially areas where the Karen live. Support was provided to villages where the generals were born, such as Kyone Kuu, where Prime Minister Thein Sein came from, having its school and hospital rebuilt. One week after the cyclone only 1 in 10 needing assistance had been given any aid and resources had been diverted by the insistence of the Burmese government on holding a national referendum. One month after the cyclone and 55% of families had less than 1 day’s food, 63% lacked access to clean water, 82% had destroyed or damaged homes and 22% of household were under psychological stress. Only 1.3 million out of the 2.4 million affected had received even very basic assistance, most of which was inadequate and below minimum requirements.

In addition to neglecting the basic responses to a disaster, the Burmese government initiated delays in receiving aid that were "unprecedented in modern humanitarian relief efforts". While calls had been made to allow aid in from the 4th of May, the government did not accept any aid until the 10th of May, when it allowed supplies to land in Rangoon for government distribution. The first direct aid was only agreed upon on the 20th of May, from ASEAN (the Association of South-East Asian Nations). On the 22nd of May, United Nations (UN) Secretary-General Ban Ki-moon was able to secure an agreement with the Burmese government to open the borders, with all UN aid workers’ visas approved on the 29th of May, nearly a month after the cyclone struck. These actions by the Burmese government significantly slowed the delivery of aid, as sufficient cargo for the immediate response had been available on stock in Bangkok within days, and prior experience has shown the ability of the international aid community to organise emergency relief in less than a week. The opening up to international aid allowed a rapid expansion in aid delivery, with the Red Cross progressing from having reached only 26,000 people by May the 30th to reaching 191,000 by June the 2nd. While the actions of the military junta in preventing international aid reaching the people were undoubtedly against the interests of the nation as a whole, they are probably understandable in the context of an unelected regime interested first in preserving power and worried that the entry of international organisations, especially US and Christian-based organisations, would ultimately be used to bring about regime change.

Once the international aid community had been allowed into the country they implemented a cluster approach under the UN Office for the Coordination of Humanitarian Affairs (OCHR). The cluster approach broke the response down into groups such as food, health, sanitation, emergency shelters, logistics and early recovery, with each cluster being lead by a specialised agency to coordinate the different players. The effort was relatively successful in providing aid to certain regions, and outbreaks of disease and excess mortality were rare and possibly within baseline levels. Despite improvements in logistical coordination compared to previous disasters, significant problems were encountered. Funding for the emergency phase was only 40.8% available after one month and the Sichuan earthquake on May the 12th caused an increase in the global price of relief items, causing shortages in the supply chain. UN staff were allowed entry, but there were delays for NGO visa applications of up to three additional weeks and staff were limited to areas permitted by the government with close monitoring of aid distribution. Baseline data on health and geographical distribution of need was extremely poor, hampering logistics. On top of this, there was frequent moving of the population (both spontaneous and government-organised) further complicating logistics.


Redevelopment after Cyclone Nargis

The OCHR has developed a coordinated redevelopment plan to continue the recovery of the Irrawaddy delta after Cyclone Nargis, running until April 2009. The program comprehensively covers the immediate needs of security, water, sanitation, nutrition, health and food, as well as the long-term redevelopment needs of livelihood recreation, shelter and education. These last aspects are highly important, as the Burmese government starting pushing internally displaced people (IDP) out of camps as early as four weeks after the cyclone. They announced that any farmers who could not restart farming would lose their land, forcing people to return to their villages even if they did not have enough food or seed rice or farming equipment, on the grounds that “people must restart their own traditional way of living and … cannot continually be dependent on relief from outside as it would result in people not being able to stand on their own two feet any more”. There are even reports that villagers who evacuated to Rangoon and refused to go back were threatened with military force. Despite the urgent need this has created for redevelopment of functional villages, the agriculture and shelter programs are the least well developed. The agriculture / livelihood program is dramatically under-funded (with only 24% of the program funded), and the shelter program is also winding up due to lack of funding, despite only 1/3 of families considering their shelters safe and the majority of families without the funds for even minor repairs before the next monsoon.

One important point for aid agencies to consider during the reconstruction effort is that refugee and IDP camps have improved to have lower mortality than surrounding regions, with improvements in logistics and water, sanitation and nutrition. While the Irrawaddy delta has not yet reached its previous standard of living, it is important that an aid-induced distortion is not produced, and long-term aid produces a national, rather than regional, increase in living standards.


Concluding remarks and lessons for the future

Undoubtedly, the lion’s share of the blame for the disaster of Cyclone Nargis lies in the hands of the Burmese government. The government failed to prepare for the cyclone with storm and health infrastructure and did not pass on the meteorological warnings it had been given to the residents of the Irrawaddy delta, leading to a cataclysmic death toll on the days of the disaster. Furthermore, the Burmese government failed to implement any of the standard post-disaster mechanisms to any meaningful extent, and actively prevented international aid for rapidly reaching the people in need, creating further unnecessary suffering. It has been argued that the Burmese government’s failure to adequately respond to the disaster and then to block international aid constituted a criminal act against humanitarian law.

This is not to say, however, that the international aid response could not have been improved. While the Burmese government was not acting in the best interests of its citizens, there were alternative routes for access. One possibility was to threaten invoking the 2005 resolution of the UN General Assembly, allowing intervention without permission of the government in circumstances when the government manifestly fail to protect their populations from … crimes against humanity. This course of action was suggested by the French government on the 7th of May, but was dismissed. Of course, it cannot be known whether this action would have met military resistance by the Burmese government, potentially creating a net detrimental impact on the people of Burma. Alternatively, with the relative success achieved by ASEAN, the United Nations and other international aid bodies could have moved faster to work through the regional power structures, rather than rely on the international legal framework distrusted by the military junta. Moreover, it can be argued that the climate of distrust between the Burmese government and certain international institutions, such as the United Nations, the US and Christian organisations, was predictable based on the historical context, and a more politic course of funnelling aid through trusted intermediaries should have been taken, rather than arrogantly assuming the Burmese government would accept aid regardless of the current relationship with source. Another route for providing discrete aid would have been to focus more on the national staff already on the ground, through operational adaptation, strengthening of capacity and remote management, however this resource was under-valued. Once international aid organisations were allowed into the region, despite the relative success of the cluster approach in logistical management, there were still inefficiencies created by a lack of strategic coordination and consultations with local communities.

One final lesson that can be drawn from Cyclone Nargis is the limitation of surge capacity within the international aid system. The combination of Cyclone Nargis, with 140,000 dead, and the Sichuan earthquake, with 70 000 dead, within ten days severely constrained the ability of international aid organisations to scale up programmes in response to needs. With the likelihood of increasing natural and conflict disasters with global warming, resource limitation and overpopulation, international aid organisations will need to start increasing funding and stand-by capacity and should heavily invest in disaster aversion.