Applications of Medium Range Probabilistic Flood Forecast for Societal Benefi ts: Lessons Learnt from Bangladesh - Reducing Disaster: Early Warning Systems for Climate Change - page 161

9.1 Introduction

Damages due to natural disasters have increased drastically in the last few decades
(Thieken et al. 2007 ). As a result of devastating fl oods, fl ood forecasting has received
increased impetus in recent years (Blöschl 2010 ). Advances in meteorological,
hydrological and engineering sciences are fast generating a range of new methodolo-
gies for forecasting weather and fl ood events, including ensemble prediction systems
(EPS) and new hydrological or hydrodynamic models (Drobot and Parker 2007 ).
However, many of these advanced prediction systems have not yet been incorporated
into operational forecast systems. Consequently, operational forecasts have not yet
been integrated into decision-making processes in order to reduce disaster risks. In
the real world, it has been observed that not all people notice warnings or are able to
understand the meaning of probabilistic forecasts well enough to consider them-
selves at risk (Parker et al. 2009 ; Molinari and Handmer 2011 ).
Despite advances in forecasting, hydro-meteorological 'surprises' have resulted
in loss of lives as well as property in every severe hazard case for any country (e.g.
Indian Ocean tsunami in 2004, cyclone NARGIS in 2008, Pakistan Flood in 2010
etc.). In most of these instances, the failure of early warning system was observed
(Parker et al. 2009 ). Scientifi c information has always had a certain degree of uncer-
tainty (Krzysztofowicz 2002 ). Of all natural disasters, fl oods impact the greatest
number of people across the world. Accordingly, fl ood predictions, as well as esti-
mates of fl ood risk, are uncertain too. Generation of medium range fl ood forecasts
is highly challenging and uncertain as it depends on precipitation forecasts. The
warning problem is made particularly complicated by the uncertainty in the fl ood
forecast used within the decision-making chain for issuing fl ood warnings (Moore
et al. 2006 ). Thus, their use is very limited. Even though the information may be
available, the community impacts remain very high due to constraints in informa-
tion fl ow, low capacity at local levels to understand the information and lack of
awareness on response options. The effectiveness of an early warning information
relies on how these uncertainties are being translated, managed and communicated
(Babel et al. 2013 ; Weick 1988 ). Hence, a decision-making process is essential in a
social context where roles and responsibilities are clearly shared among all decision
makers to contextualize the warning information to evoke appropriate responses
(Morss and Eugene 2007 ).
Bangladesh is a fl at deltaic country located at the lower part of the basins of
three large alluvial rivers, the Ganges, the Brahmaputra and the Meghna. It includes
57 trans-national rivers from several countries of South Asia. Total river basin area
of these three rivers is 1.7 million km 2 (Fig. 9.1 ). Numerous tributaries and dis-
tributaries of these rivers and extensive fl oodplains are the main physiographic
feature of the country. In fact, about four-fi fths of the country is a fl oodplain. As a
result of fl at topography of the fl oodplain, one-fi fth to one-third of the country is
fl ooded by overfl owing rivers during the monsoon when rainfall is also very high.
This annual phenomenon of rainfall and river fl ooding has played an active role in
shaping the landscape, economy, society and culture of the country. During the
monsoons (from late June to early October), Bangladesh experiences two forms of