Extreme Water Levels in Bangladesh: Past Trends, Future Projections and their Impact on Mortality

Abstract:

Climate change is expected to have an impact on meteorological and therefore hydrological extremes, thereby possibly altering the vulnerability of exposed populations. Our study focuses on Bangladesh, which is particularly vulnerable to changes in extremes due to both the large population at risk, as well as geographical characteristics such as the low-rising slope of the country through which the outflow of the combined catchments of the Ganges, Brahmaputra and Meghna rivers (GBM, ~1.75 million km²) is channeled.
Time series of daily discharge and water level data for the past 100 years were analyzed with respect to trends in frequency, magnitude and duration, focusing on rare but particularly high-risk events using extreme-value theory. Mortality data is available for a five-year period (2003-2007), with a distributed lag non-linear model used to examine possible connections between extreme water levels and mortality.
Then, using output from the Community Climate System Model CCSM4, projections were made regarding future flooding due to changes in precipitation intensity and frequency, while also accounting for the backwater effect of sea-level rise. For this, the upper catchment precipitation as well as monthly mean thermosteric sea-level rise at the river mouth outflow were taken from the four CCSM4 1° 20th Century ensemble members as well as from six CCSM4 1° ensemble members for the RCP scenarios RCP 2.6, 4.5, 6.0 and 8.5.
Results show that while e.g. the mean water level did not significantly rise during the past 100 years, a change in extreme water levels can be detected. In addition, annual minimum water levels have decreased, which is of particular importance as there is a significant connection to an increase in mortality for low water levels. While mortality does not seem to increase significantly due to extreme floods, our results indicate that return levels projected for the future shift progressively, with the effect being strongest for RCP 8.5. Further measures to strengthen the resilience of the exposed population are therefore required to ensure that climate change effects do not overwhelm the population’s coping capacities.