A new catalogue of tropical cyclones of the northern Bay of Bengal and the distribution and effects of selected landfalling events in Bangladesh

摘要

Tropical cyclones devastate large areas, take numerous lives and damage extensive property in Bangladesh. Research on landfalling tropical cyclones affecting Bangladesh has primarily focused on events occurring since AD 1960 with limited work examining earlier historical records. We rectify this gap by developing a new tropical cyclone catalogue that maximizes the use of available sources. The catalogue consists of 304 tropical cyclones that occurred between AD 1000 and AD 2009 and made landfall along the coasts of Bangladesh, eastern India and Myanmar. One hundred and ninety-three events directly struck Bangladesh between AD 1484 and AD 2009, although the precise landfall location of six events is unknown. Of the remaining 187 events, Cox's Bazar, Chittagong, Noakhali, Barisal and Khulna were struck by 30, 46, 19, 41 and 51 tropical cyclones, respectively. There is a paucity of data about tropical cyclones before AD 1900 and this increases the further back in time we go. Inconsistencies in reported storm surge height, wind speed and exaggerations in the reporting of deaths are identified and discussed. Some 20 72 509 human deaths in Bangladesh are associated with 71 tropical cyclones that occurred between AD 1484 and AD 2009. Between AD 1923 and AD 2009, 11 tropical cyclones caused 94 35 000 people to become homeless and between AD 1961 and AD 2009, 10 tropical cyclones resulted in economic damage of over US$ 4.6 billion. Analysis of the deaths and damage associated with tropical cyclones in AD 1970, AD 1991 and AD 2007 indicates that while the number of deaths decreased between events, economic damage and the number of people made homeless increased. There are positive and significant correlations between increasing storm surge height and increasing human fatalities (r = 0.60, p < 0.01) and increasing human injuries and greater wind speed (r = 0.45, p < 0.01). Despite our best efforts, the catalogue is incomplete. As such, we suggest further deep' archival research coupled with regional geological studies of palaeostorm events to gain a more sophisticated understanding of the hazard. Our results have implications for both risk assessment and disaster risk reduction.