Acknowledgement Table of Contents TOC o “1-3” h z u Acknowledgement PAGEREF _Toc522926062 h iTable of Contents PAGEREF _Toc522926063 h iiLists of Tables PAGEREF _Toc522926064 h ivLists of Figures PAGEREF _Toc522926065 h ivAbstract PAGEREF _Toc522926066 h vList of Acronyms PAGEREF _Toc522926067 h viiChapter-1

Acknowledgement Table of Contents TOC o “1-3” h z u Acknowledgement PAGEREF _Toc522926062 h iTable of Contents PAGEREF _Toc522926063 h iiLists of Tables PAGEREF _Toc522926064 h ivLists of Figures PAGEREF _Toc522926065 h ivAbstract PAGEREF _Toc522926066 h vList of Acronyms PAGEREF _Toc522926067 h viiChapter-1

Acknowledgement
Table of Contents TOC o “1-3” h z u Acknowledgement PAGEREF _Toc522926062 h iTable of Contents PAGEREF _Toc522926063 h iiLists of Tables PAGEREF _Toc522926064 h ivLists of Figures PAGEREF _Toc522926065 h ivAbstract PAGEREF _Toc522926066 h vList of Acronyms PAGEREF _Toc522926067 h viiChapter-1: Background/Introduction PAGEREF _Toc522926068 h 11.1 Introduction PAGEREF _Toc522926069 h 11.2 Causes of Flood PAGEREF _Toc522926070 h 21.3 Review of the Flood in Bangladesh PAGEREF _Toc522926071 h 31.4 Review of the Flood Forecasting System PAGEREF _Toc522926072 h 31.5Objectives of the study PAGEREF _Toc522926073 h 51.6Scope of the study PAGEREF _Toc522926074 h 51.7Rationale of the study PAGEREF _Toc522926075 h 51.8Limitation of the study PAGEREF _Toc522926076 h 51.9Outline of the study PAGEREF _Toc522926077 h 6Chapter-2: Literature Review PAGEREF _Toc522926078 h 72.1 Introduction PAGEREF _Toc522926079 h 72.2Structural and non-structural flood protection measures PAGEREF _Toc522926080 h 92.3River floods and early warnings PAGEREF _Toc522926081 h 102.4Flood warning errors and credibility PAGEREF _Toc522926082 h 132.5Uncertainty in flood risk assessment PAGEREF _Toc522926083 h 142.6Agriculture and Flooding: Bangladesh PAGEREF _Toc522926084 h 142.7Conclusion PAGEREF _Toc522926085 h 16Chapter-3: Method and Materials PAGEREF _Toc522926086 h 183.1 Methodology PAGEREF _Toc522926087 h 183.2Research Methods PAGEREF _Toc522926088 h 193.3Data PAGEREF _Toc522926089 h 193.3.1Secondary Data PAGEREF _Toc522926090 h 193.3.2 Primary Data PAGEREF _Toc522926091 h 203.4 Processing of Data PAGEREF _Toc522926092 h 203.5 Ethical Consideration PAGEREF _Toc522926093 h 21Chapter-4: Data Analysis and Findings PAGEREF _Toc522926094 h 224.1 Introduction PAGEREF _Toc522926095 h 224.2Evaluation of Flood Bulletin PAGEREF _Toc522926096 h 234.3Flood Forecasting Performance based on Secondary Data PAGEREF _Toc522926097 h 244.4Flood Forecasting Performance based on Primary Data PAGEREF _Toc522926098 h 26Flood message and dissemination: PAGEREF _Toc522926099 h 26Sufficiency of information and expectations: PAGEREF _Toc522926100 h 27Data, technology, interagency and manpower Sufficiency: PAGEREF _Toc522926101 h 30Chapter-5: Conclusion and Recommendations PAGEREF _Toc522926102 h 345.1Conclusions PAGEREF _Toc522926103 h 345.2Recommendations PAGEREF _Toc522926104 h 35References PAGEREF _Toc522926105 h 36Annexure-1: Flood warning message of BWDB PAGEREF _Toc522926106 h 39Annexure-2: Questionnaire to Conduct Survey PAGEREF _Toc522926107 h 47Lists of Tables TOC h z c “Table 3.” Table 3.1 List and Composition of the Respondents PAGEREF _Toc522923389 h 20 TOC h z c “Table 4.” Table 4. 1 Flood forecasting performance for 2011, 2012, 2013 and 2014 PAGEREF _Toc522923289 h 25Table 4. 2 Outcome of the flood forecasting center and information dissemination PAGEREF _Toc522923290 h 26Table 4. 3 Sufficiency, expectation and strong points of the forecasting system PAGEREF _Toc522923291 h 28Table 4. 4 Data, technology, interagency coordination and manpower sufficiency PAGEREF _Toc522923292 h 31
Lists of Figures TOC h z c “Figure 1.” Figure 1. 1 Area affected by flood in different year PAGEREF _Toc522923616 h 3 TOC h z c “Figure 3.”
Figure 3. 1 Methodology for evaluating flood forecasting performance PAGEREF _Toc522923668 h 19
TOC h z c “Figure 4.” Figure 4. 1 Mapping for the questionnaire survey PAGEREF _Toc522923686 h 22

AbstractFloods are normal events in the deltaic plains of Bangladesh. Although the lifestyle of the people in Bangladesh is well adapted to normal monsoon flood phenomena, the damages due to inundation, riverbank erosion or breach of embankment, etc. still occur in various regions in almost every monsoon season. They often have disastrous consequences: major damage to infrastructure, great loss of property, crops, cattle, poultry etc., human suffering and impoverishment of the poor. With every major flood in Bangladesh, food security and poverty situation has been worsening. Flood forecasting and warning is recognized as a vital non-structural measures to aid the mitigating the loss of lives, crops and properties caused by the annual flood occurrence. However, due to different shortcomings including limited upstream hydro-meteorological information, detail & accurate digital elevation model and limited technological development, the services were fully not satisfactory to all corners. Information on flash flood was limited due to technological limitation and non-availability of the real time data at a much shorter interval than the usual.

The present study aims to assess the performance of flood forecasting system of Bangladesh by means of primary and secondary data. The primary data were collected through questionnaire survey by adopting Purposive and Snowball sampling procedure interviewing twenty five (25) respondents from four (4) government organizations and local people. A well prepared questionnaire was supplied to them in advance and then sought their comments/suggestions of some issues related to flood forecasting system of Bangladesh. The secondary data were collected by literature review, flood bulletin of Bangladesh Water Development Board. To evaluate the performance of the flood forecasting system of Bangladesh, this study uses generalized block-and-flow module. The module is made up of four basic components: Objectives, Evaluative criteria, Analysis synthesis and Evaluation.
The results suggests that FFWC issued daily flood bulletin from May to October with a forecast lead-time of 24hrs, 48hrs and 72hrs, 96 hrs and 120 hrs (upto 5 days) along with warning messages and flood inundation maps. The forecast performance is very good up to 72 hours. With the increasing lead-time forecast performance is gradually decreasing. The flood warning message is disseminated through different news media, news agencies, fax, e-mail, web-site (www.ffwc.gov.bd) and IVR (1090) through mobile phone.

The available data, technology, inter-agency coordination and human resources capacity is not sufficient to meet the present demand. In order to improve the situation it is necessary to improve the data sharing mechanism with upstream country along with utilization of satellite data and automation of data collection system. To improve the insufficiency of the technology and improving the accuracy of the flood forecasting performance it is necessary to improve the rainfall forecasting system. The present forecasting model covers Bangladesh only. However, Bangladesh is located at the outlet of the Ganges, Brahmaputra and Meghna Basin. Basin-wide model development with the aid of satellite based information is very crucial for improving the flood and water management of Bangladesh.

The severity of flooding is greatest when the peak floods of the major rivers coincide with these effects. Climate changes could influence the frequency and magnitude of flooding. A higher sea level will inhibit the drainage from the rivers to the sea and increase the impact of tidal surges. Deforestation in hilly catchments causes more rapid and higher runoff, and hence more intense flooding. The springtides of the Bay of Bengal retard the drainage of floodwater into the sea and locally increase monsoon flooding. A rise of MSL at times during the monsoon period due to effect of monsoon winds also adversely affect the drainage and raise the flood level along the coastal belt. Many of these factors required active cooperation with the upstream countries for improving the prediction and management of flood. In addition basin-wide model development with the help of satellite data can also give better information for flood management.

List of AcronymsBMD :Bangladesh Meteorological Department
BWDB :Bangladesh Water Development Board
CB : Cell Broadcasting
CWC:Central Water Commission
DAE :Department of Agricultural Extension
DDM :Department of Disaster Management
DEM : Digital Elevation Model
FAP: Flood Action Plan
FEC :French Engineering Consortium
FF :Flood Forecast
FFWC :Flood Forecasting and Warning Center
FFWS:Flood Forecasting and Warning Services
GIS: Geographic Information Services
GO:Governmental Organization
IVR :Interactive Voice Response
SOD :Standing Orders for Disaster
LGED: Local Government Engineering Department
MoDMR :Ministry of Disaster Management and Relief
NGO: Non-governmental Organization
NOAA:National Oceanic and Atmospheric Administration
VGD: Vulnerable Group Development
VGF: Vulnerable Group Feeding
Chapter-1: Background/Introduction1.1 IntroductionBangladesh is the part of world’s most dynamic hydrological and the biggest active delta system. The topography, location and outfall of the three great rivers system (the Ganges, the Brahmaputra and the Meghna) shape the annual hydrological cycle of the country and fate of millions of people. Too much and too little water in monsoon and dry season respectively is the annual phenomenon. Flood is a regular monsoon event whose depth and duration are the deciding factors whether it affecting beneficially or adversely. Monsoon inflow along with rainfall historically shapes the civilization, development, environment, ecology and the economy of the country. Extreme events of flood adversely affect the development, economy, food security, poverty and almost every sector. In flood management, Bangladesh has been taken structural and non-structural measures. One of the main non-structural measures is the flood forecasting and warning service.

Bangladesh Water Development Board (BWDB) is responsible for Flood Forecasting in Bangladesh. Flood Forecasting and Warning Center (FFWC) of BWDB is carrying out this duty. The FFWC was established in 1972 and is fully operative in the flood season, from April to October every year, following the Standing Orders for Disaster (SOD) of the Government of Bangladesh. The FFWC is acting as the focal point on flood forecasting and warning services in co-ordination with concerned ministries and agencies like Ministry of Disaster Management and Relief (MoDMR), Bangladesh Meteorological Department (BMD), Department of Disaster Management (DDM), Department of Agricultural Extension (DAE) etc. during the monsoon for flood disaster management.

Flood forecasting and warning is recognized as a vital non-structural measures to aid the mitigating the loss of lives, crops and properties caused by the annual flood occurrence. Flood forecasting and warning services enable and persuade people, community, agencies and organizations to be prepared for the flood and take necessary actions in advance to increase safety and reduce or protect damages of lives and properties. It also helps agencies, departments, communities and people to enhance preparedness and to motivate vulnerable communities to undertake preparedness and protective measures.
However, due to different shortcomings including limited upstream hydro-meteorological information, detail ; accurate digital elevation model and limited technological development, the services were fully not satisfactory to all corners. Information on flash flood was limited due to technological limitation and non-availability of the real time data at a much shorter interval than the usual.

1.2 Causes of FloodThe river system of Bangladesh is one of the most extensive in the world, and the Ganges and the Brahmaputra are amongst the largest rivers on earth in terms of catchment size, river length and discharge. The total catchment area is approximately 1.6 million km2. The river systems drain about 1 Trillion m3 of water annually. Most of the rivers are characterized by having sandy bottoms, flat slopes, substantial meandering, banks susceptible to erosion and channel shifting.
There are four climatic distinct seasons (a) winter: December to February (b) pre-monsoon: March to May, (c) monsoon: June to September and (d) post monsoon: October to November. Over 80% of the rainfall occurs during the monsoon or rainy season also known as flood season. The normal annual rainfall of the country varies approximately from 1200 mm in the west to over 5000 mm in the east. Long periods of steady rainfall persisting over several days are common during the monsoon, but sometimes local high intensity rainfall of short duration also occurs.

Floods in Bangladesh occur for number of reasons. The main causes are excessive precipitation, low topography and flat slope of the country. There are other factors as well which include (a) geographic location and climatic pattern, (b) confluence of three major rivers, the Ganges, the Brahmaputra and the Meghna, (c) coincidence of flood peak in major rivers, (d) influence of tides and cyclones, (e) long term environmental changes, (f) deforestation in the upstream area, (g) climate change, etc.

The springtides of the Bay of Bengal retard the drainage of flood water into the sea and locally increase monsoon flooding. A rise of tide level during the monsoon period due to effect of monsoon winds also adversely affect the drainage and raise the flood level along the coastal belt.

1.3 Review of the Flood in BangladeshMany parts of the Asia during monsoon frequently suffer from severe floods. Some parts of India and Bangladesh experience floods almost every year with considerable damage. The floods of 1954, 1955, 1974, 1987, 1988, 1998, 2004, 2007 and 2017 in Bangladesh caused enormous damages to properties and considerable loss of life (BWDB, 2011, 2012, 2013, 2014, 2015, 2016). The floods of 1987, 1988 1998, 2004 and 2007 flood caused heavy damage. Figure 1.1 shows the percent of total area of Bangladesh affected by the flood since 1954.
209550090297010% area line
10% area line
152527062674520% area line
20% area line

Figure 1. SEQ Figure_1. * ARABIC 1 Area affected by flood in different year1.4 Review of the Flood Forecasting SystemThe Flood Forecasting and Warning Centre under BWDB collect hydrological monitoring data of 90 representative water level stations and 59 rainfall stations throughout the country. The principal outputs are the daily statistical bulletin of floods, river situation, a descriptive flood bulletin, forecast for 24, 48, 72, 96 and 120 hours at 54 monitoring points on the major rivers, special flood report along with different graphical and statistical presentation during the monsoon season.
Step by step development has been made in the flood forecasting and warning services in Bangladesh, started from 1972. Before 1990, forecast for six locations Bahadurabad, Serajgonj, Aricha, Goalondo, Bhagyakul and Hardinge Bridge. After the devastating flood of 1987 and catastrophic flood of 1988, it was deeply realized that the forecast formulation should be introduced in the process of river modelling. In view of the above, the forecasting system was updated. The brief history of the forecasting in Bangladesh is given below.

1972: FFWC Established under BWDB
Real Time Flood Monitoring at 10 Stations/Points along the Brahmaputra, Ganges and Padma rivers
Flood Forecast (FF) with few hours lead time at 6 points by Gauge Correlation along Brahmaputra and Padma rivers
1992: MIKE11-FF Model Introduced
FF with one day lead time at 16 points/locations 1995-96
MIKE11 Super Model with GIS FF at 30 locations with lead time up to 2-days
2000-04: Strengthening FFWS Expansion of FF areas coverage
Flood monitoring covers entire country
Improved accuracy and extend Lead Time up to 3-days
Improved dissemination
2005-07: Probabilistic medium range FF with lead time upto 10-days initiated at 18 points/locations
2007-09: Further extension of FFWS
Mike 11 Super Model with GIS introduced with flood map generation facility FF at 38 locations on 21 Rivers up to 3-days Lead Time
Flood Inundation Mapping
Improvement of probabilistic medium range FF up to 10-days at 18 points
From 2012 onward: Strengthening and Improvement of FFWS
Flood Forecasting at 54 locations on 29 rivers with Extended Lead Time up to 5-days
Structure based FF for 4-selected projects up to 5-days lead time
Improved and more user-friendly web-site with Bangla language IVR system for dissemination based on mobile phone introduced.

1.5Objectives of the studyThe flood forecast is intended to alert the people of the locality about the predicted water levels of floodwater 3-days ahead of its occurrence. An accurate forecast would be one where the forecast level and corresponding observed level at the stipulated time are within a small range of variation. The main objective of the present study is to assess the performance of flood forecasting system of Bangladesh. The study will also suggest some policy initiatives for improving the flood forecasting system of Bangladesh.
1.6Scope of the studyPresently early warning on floods provides a lead time of 24, 48 and 72 hours. There are needs and expectations for increasing lead time forecast for cropping decisions, such as early harvesting, or to implement a contingency crop plan or protect infrastructure and preserve livelihoods. Assessment of the performance will lead to develop a framework to further improve the flood forecasting system of Bangladesh.

1.7Rationale of the studyFloods are normal events in the deltaic plains of Bangladesh. Although the lifestyle of the people in Bangladesh is well adapted to normal monsoon flood phenomena, the damages due to inundation, riverbank erosion or breach of embankment, etc. still occur in various regions in almost every monsoon season. They often have disastrous consequences: major damage to infrastructure, great loss of property, crops, cattle, poultry, etc, human suffering and impoverishment of the poor. With every major flood in Bangladesh, food security and poverty situation has been worsening.
1.8Limitation of the studyThe main limitation of the present study is the availability of data and sufficient time. Thus the present study will assess the performance of the flood forecasting in Bangladesh qualitatively. Access to data and information is another limitation of the present study. Due to its complexity in nature, researchers are not in a position to capture all those changes instantly.

1.9Outline of the studyThe study contains some background information on flood forecasting along with objectives, rationale, scope and limitations of the study. All of these have been included in Chapter one. In the second chapter, an attempt has been made to figure out the gap in the context of flood forecasting by reviewing the existing literature. Chapter three contains the methods and materials that have been followed in preparing the study. Analytical part of this study has been included in the fourth chapter with some analytical tools and techniques in details for the study. This chapter also contains the major findings of the study. Chapter five contains some recommendations along with some concluding remarks. At last, some references and appendices have concluded the study.

Chapter-2: Literature Review2.1 IntroductionFloods are a major barrier to development in Bangladesh, and since 1954 governments have given great emphasis to addressing this problem. The country is vulnerable to several forms of severe natural events, but the destructive effects of floods have long been highlighted as the main obstacle to the economic improvement of the nation. The importance of having effective and end user friendly early warning systems is widely accepted as one valuable preparedness measure to manage disaster risk. The Hyogo Framework for Action (2010-2015) made early warning a Priority for Action and the post 2015 framework for Disaster Risk Reduction is expected to continue this attention.
Bangladesh is a very flood prone country: one-fifth up to one-third is being flooded during the monsoon period. Essential to enhancing preparedness is the establishment of a well-functioning people centered early warning system. In order to realize the maximum value of the system, each component (risk knowledge, monitoring and warning service, communication and dissemination and response capability) must function effectively. Communication and dissemination of the warning message is often downgraded as a less important component in comparison to developing the technical forecasting system itself.
A people-centered early warning system comprises four key elements. These are knowledge of the risks; monitoring, analysis and forecasting of the hazards; communication or dissemination of alerts and warnings; and local capabilities to respond to the warnings received (UNISDR, 2006). Ultimately an early warning system will only be effective if all components are effective. The communication and dissemination component has been recognized as the component which lacks sufficient attention and results in a huge gap between the information produced by national level forecasting agencies and the information that is actually received and acted upon by the flood affected communities.

Currently, the Government of Bangladesh is most concerned with the threat of more frequent and severe floods due to climate change and sea-level rise, and as a consequence millions of marginal farmers will be seriously affected. At the World Climate Conference 3 in Geneva in 2009 the Government urged the international community to provide more technological and financial support for upholding the community-based adaptation mechanisms in order to reduce flood vulnerability in Bangladesh and in other countries facing similar threats.

Studies into flood hazards, and human adjustments to floods, emerged in the 1940s with researchers such as White (1945) conducting some pioneering work. Islam (1974, 1980) focused on agricultural adjustment to floods in Bangladesh. Individuals and communities adjust to circumstances, and farmers in Bangladesh have long had to adapt to severe natural events. This community-based autonomous adaptation is a feature of life in Bangladesh but, to date, vulnerability and adaptation to climate change have been absent from flood-hazard research. Since 1980 a handful of projects by Islam (1980, 1990, 1995), Paul and Rasid (1993), Paul (1995), and Chowdhury (1988) have investigated agricultural adjustments and flood damage to rice crops in Bangladesh, but with the exception of the Flood Action Plan (FAP 16), government research initiatives in this regard have been negligible. The studies completed by independent individual researchers have not been sufficiently comprehensive to understand the extent of damage (including household damage) due to extreme flood events (EFEs), and vulnerability and adaptation issues have never been adequately identified at the rural-community level. Some independent water-resource researchers such as Miah (1988), Ahmad and Ahmed (2003, 2004), Ahmad and Rasheed (1998), Ahmad et al. (2001), Ahmad et al. (1994), Baqee (1997), Rasheed (2004, 2008), Ninno and Roy (1999), Ninno et al. (2001), and Warrick et al. (1996) have focused on flood-hazard research and regional co-operation, but the vulnerability of riverine communities and the capacity to adjust to change still require much closer examination.

The Bangladesh government focused on both structural and nonstructural flood management methods in order to reduce vulnerability to flooding (Paul, 1995). The prudent approach is to have a combination of the two methods as suggested by Rasheed (2008). In addition to structural and non-structural approaches, it is important that the flood-warning system be upgraded and that real-time data be exchanged with other upper riparian countries to better address the flood management issue (Ahmad et al., 1994, 2001; Ahmad and Ahmed, 2003, 2004). The FAP and a French engineering consortium (FEC), in association with the Bangladesh Water Development Board, have studied various flood issues, particularly a project for flood-control drainage and flood-control drainage and irrigation. The FAP 19 dealt with GIS. Other FAP studies like 2-6, 11-16, 20 and 23 all dealt, in one way or another, with community and agricultural adaptation to floods. The FAP and FEC also conducted a feasibility flood-control survey, a char-land study (the most comprehensive char-land study was made under FAP 16), and a hydrological study, but a few exceptions that have focused on agricultural adjustment processes, i.e. how farmers adapt (especially crop planting) with flood characteristics: timing, frequency and duration (Younus, 2012b) and the impacts of floods when flood adaptation fails on rural households at the community level (Younus and Harvey, 2014). In the light of the growing urgency of this issue a number of questions arise: to what extent will Bangladeshi farming systems be able to adjust to extreme floods? Is it important to understand adaptation issues for future community-based adaptation planning?
2.2Structural and non-structural flood protection measuresA range of flood protection and management measures exist, falling into structural (‘hard’) or non-structural (‘soft’) approaches. The former refer to large-scale defenses, such as dikes, dams and flood control reservoirs, diversions and floodways, and improving drainage channel capacity. Structural defenses have a very long tradition, with dams and dikes having been built for millennia. Constructing reservoirs where excess water can be stored allows a regulated temporal distribution of streamflow, reducing the natural peak flow. The physical dimensions of structural flood protection measures, such as levees, are based on probability theory to withstand a predicted ‘design flood’ of a certain magnitude, e.g. a 100-year flood (although this is difficult to determine in practice), in a given location. The longer the assumed return period of the design flood, the better the level of protection and the greater the costs. This raises various value judgments, such as whether to design dikes to withstand a 100-year flood or perhaps a 1000-year flood. The latter solution would give better protection but be far more costly. Moreover, it is misleading to expect complete flood protection or total certainty of outcomes. Dikes protect well against small- and medium?size floods but when a deluge is of disastrous size and dikes break, losses in a levee-protected landscape can be higher than in the absence of a levee due to the false feeling of security that levees can generate among the riparian population and the high damage potential in apparently (but not completely) safe areas. No matter how high a design flood is, there is always a possibility of a greater flood occurring, inducing losses. A dike designed for a 100-year flood is likely to fail if a 1 000-year flood occurs. Several developed countries have costly structural protection facilities to withstand a high, rare flood. Reinforced dikes or super-dikes of 300–500 meters width play an important part in flood protection of major cities in Japan, where a very high level of safety must be assured (Kundzewicz and Takeuchi, 1999). Even higher protection levels are achieved in the low-lying Netherlands. The Flood Defense Act of 1996 set high safety standards with the return period of design flood set at 1 250 years for middle to upper rivers in the Netherlands and 2000 years for lower river reaches (Pilarczyk, 2007).

‘Soft’ non-structural flood protection measures include source control (watershed management), laws and regulations, zoning, economic instruments, efficient flood forecast warning systems, flood risk assessment systems, awareness-raising information and flood-related databases. Source control modifies the formation of floodwater by catching water where it falls, enhancing infiltration, reducing impermeable areas and increasing storage in the watershed. These measures counteract the adverse effects of urbanization, such as reduced storage potential, growth in the runoff coefficient and flood peak, and acceleration of a flood wave. Restoring, retaining or enhancing water storage capacity in the river system (floodplains, polders and wetlands) is also important.

Appropriate schemes of insurance, which distribute risks and losses over many people and over a longer time, coupled with aid that can compensate for uninsurable losses, are additional important components of flood preparedness. Flood-risk maps developed for the insurance industry are used to help estimate insurance premiums for properties. Post-flood disaster aid, based on voluntary solidarity contributions, national assistance and international help, is essential to restore the livelihoods of survivors.

The permanent evacuation of floodplains is virtually unthinkable in most countries. This is definitely true for Bangladesh, a densely populated and low-lying country, which ranks as the most flood-prone country on earth. The people of Bangladesh, growing rapidly in number, have to live with regular floods. Most of the country is made up of floodplains and soil fertility depends on regular flood inundation. In 1998, more than two thirds of the country was inundated. New flood embankments, even if affordable, would occupy scarce and highly demanded land. Thus, the options include reinforcing the existing structural defenses and enhancing and optimizing non-structural measures, including the forecast-warning system. As this example makes clear, optimum strategies for flood protection must be site-specific.

2.3River floods and early warningsThe term ‘flooding’ denotes a potentially destructive abundance of water in a normally dry location. Various categories of flood exist. They include, for example, situations where intense precipitation overwhelms urban sewer and drainage systems. In contrast, groundwater flooding occurs when the water table reaches the ground surface in a location where it does not normally do so. And coastal and estuary flooding occurs when high sea levels (due to high tides, storm surges or tsunamis) cause the coastal line to recede. River floods occur when water inundates areas outside the river channel, where there is potential to cause damage. They can be caused by several mechanisms, such as rain (intense or long?lasting), snowmelt (possibly with rain), glacier melt, glacial lake outbursts, dam breaks (breach) and tidal surges. Unexpected flow obstructions such as landslides, ice jams, beaver dams or debris can also cause flooding upstream. Floods differ considerably from some of the other hazards addressed.
River floods are natural phenomena, manifesting the natural spatial and temporal variability of the river water level and discharge, which can take on extremely high values from time to time. River floods, jeopardizing settlements located in floodplains, have been a continued hazard for humanity and can be identified in old myths and narratives. For millennia, people have settled in river valleys to till fertile soils, benefit from flat terrain, access water supplies easily and use water for transport. Riparian people have historically lived in harmony with nature, benefiting from benign floods and the valuable services they provide to fisheries, wetlands, wildlife and agriculture.

The notion of ‘early warning of floods’ can be interpreted in at least two ways. The first refers to a short-term flood preparedness system, where a ‘flood warning’ is a technical term, denoting a means of reducing flood damage to people and property. In this sense, a flood warning contains specific timely information, based on a reliable forecast, that a high water level is expected at a particular location and time. It aims to ensure that emergency actions, such as strengthening dikes or evacuation, can be undertaken. A ‘flood alert’, usually issued before a ‘flood warning’, is less specific and has the broader aim of raising vigilance. A warning should be issued sufficiently early prior to the potential inundation to allow adequate preparation. The appropriate timeframe is affected by the catchment size relative to the vulnerable zones. The warning should also be expressed in a way that persuades people to take appropriate action to reduce damage and costs of the flood.

The other interpretation of ‘early warning’ in the context of floods is a statement that a high water level or discharge is likely to occur more frequently in the future. Technically, this constitutes a ‘prediction’ of a change in flood frequency compared to a reference period. An early warning of this type could, for example, predict that at a site of concern the current 100-year flood (river flow exceeded once in 100 years on average) may become a 50-year flood within some defined future time horizon. Such an early warning, over a longer time scale, is (or should be) an important signal for decision-makers that the required level of protection is unlikely to be maintained in the future unless flood preparedness is improved.

Flood warnings should contain more information than flood forecasts. This includes recommendations or orders for affected populations to take actions, such as evacuation or emergency flood-proofing, specifically designed to safeguard life and property (Smith and Ward, 1998). According to Nigg (1995), warning systems must fulfill two basic functions:
assessment (from the moment that a specific hazard is detected to the point when a risk message is developed for the threatened locality);
dissemination (issuing and transmitting the warning message to a target audience).

A warning, converting scientific forecasts into lay language, is a communication that a hazard will produce specific risks for a specific population (Nigg, 1995). Affected communities should not just be told about the hazard but also informed in such a way that they are persuaded to take specific remedial action in time. Such warnings should be ‘populist’ in tone and communication but their design is actually a skilled task requiring careful forethought and design. There is an important difference between flood alerts and flood warnings. The latter have much shorter lead times that must be accurate to maintain public confidence. It is often noted that forecasts have advanced markedly, while progress in warnings has lagged behind despite recent advances in some countries, such as automatic telephone and text-messaging services and the provision of detailed forecasting services on the internet. Nigg (1995) argues that to formulate warning messages:
the basis of the warning must be credible;
the warning message must explain the degree to which a specific area is at risk;
people must be told what they can do to reduce their exposure to danger.

Speed of reaction to warnings is essential because there may be a short time to implement emergency pre?flood actions, such as strengthening or deploying defenses and evacuation, before the risk becomes high. Other useful criteria or indicators of warning quality are the penetration of the warning (the proportion of those who need information that receive it) and degree of satisfaction.

2.4Flood warning errors and credibilityTwo types of warning errors occur: first, when a warning is issued but the risk does not materialize; second, when no warning is issued but a risk and the ensuing disaster occur. The first does not include situations where the risk has materialized but the disaster has not. For instance, flood warning in the Netherlands in January 1995 resulted in massive evacuation. A disaster did not arrive, as the levees withstood the high water load but the warning and the evacuation were justified and perceived by the population as the right decision. Nigg (1995) noted that officials often hesitate to issue warnings due to fear of error, especially when warning systems are just developing or when there is still a great deal of uncertainty about the occurrence of the future event.

Person-specific warnings may trigger more urgent responses than communication directed at the general public. There can be challenges, however, in establishing more focused communication mechanisms. Factors influencing the efficiency of message dissemination include the credibility of the source (which may vary according to the recipient of the message); the accessibility of the chosen communication channel; the usefulness or redundancy of the information; and the communication of a system’s resistance to floods. Essentially, the information should be conveyed in wording that the public can understand and empathize with and via channels that the public finds credible.

The projections (predictions) of flood hazards in the more distant future indicate considerable changes in the anticipated risk (Hall et al., 2003, 2005; Hirabayashi et al., 2008; Dankers and Feyen, 2008; Kundzewicz et al., 2010a, 2010b). Early warnings of increasing flood risk in the decades ahead are necessary to upgrade defense strategies, for example by undertaking the time-consuming and resource-intensive task of strengthening levees. Climatic and non-climatic factors affect future flood risk and adaptation needs. For example, land-cover changes and urbanization can increase flood risk regardless of any change in climate. Observations and climate projections show widespread increases in the contribution of very wet days to total annual precipitation in the warming atmosphere. Increasing flood magnitudes are projected in areas where floods result from heavy rainfall; decreasing flood magnitudes are anticipated where they result from spring snowmelt. Floods corresponding to a 100-year return in the control period may become more or less frequent in the future climate. Where they increase, upgraded flood prevention measures will be needed to ensure the required protection level.
Complete flood safety is impossible in low?lying areas adjacent to rivers. Flood risk can be considerably restricted; however, if an adequate preparedness system is built, consisting of a site?specific mix of measures. Flood-related research, financed by regional, national and international funding institutions, administrative authorities, water agencies, and the insurance and reinsurance industry, is indispensable to optimize preparedness systems.

2.5Uncertainty in flood risk assessmentVarious studies (e.g. Beven, 2006a; Sivakumar, 2008) have judged the uncertainty analysis in hydrological studies to be highly unsatisfactory. Although disagreeing in significant respects, all called for the promotion of uncertainty analysis of measurements and modelled results in hydrological studies; uncertainty analysis should not be an add-on element-an afterthought of little importance. There is considerable uncertainty regarding uncertainty estimation. A good start would be greater rigor and consistency in analyzing and reporting uncertainties. One weakness arises from the deficiencies of hydrological models and available observation records for model validation. There is an overwhelming scarcity of homogeneous long?term observation records. The inherent uncertainty in analyzing any set of flood flows also stems from the fact that directly measuring the range of extreme flows can be challenging because, for example, rating curves are not available for the high flow range, gauges are destroyed by flood waves or observers are evacuated. Recourse to indirect determination is therefore necessary (Kundzewicz et al., 2010b). Uncertainty in future projections of river flooding is very high (Kundzewicz et al., 2010b), and grows the further we look into the future. In the near?term, climate model uncertainties play the dominant role, while over longer time horizons uncertainties due to greenhouse gas emission scenarios become increasingly significant. Uncertainty in practical flood-related projections is also due to a spatial and temporal scale mismatch between coarse-resolution climate models and the finer scale of a drainage basin. Scale mismatch renders downscaling (disaggregation) necessary. In fact, much more refined data are necessary for the ‘point’ scale of a locality (e.g. a small riparian town), which is the level at which costly adaptation is undertaken.

2.6Agriculture and Flooding: BangladeshBangladesh is particularly flood-prone, and during typical annual floods about 20-22% of the country is inundated between June and October, even in years of normal rainfall (Rasheed, 2008). However, as a consequence of climate change Bangladesh faces more frequent floods. The floods of 1988 and 1998 were especially severe, inundating 61 percent and 68 percent of the land respectively. Not only were those floods more widespread but they were of much longer duration. For example, the 1998 flood continued for over 90 days. Ahmad et al. (1994a) have predicted that by the year 2030 Bangladesh will be 0.5–20C warmer if emissions continue at the present rate. Moreover, climate models indicate that average monsoon rainfalls are likely to rise by 10–15% by 2030, although the authors cited above emphasized that there is uncertainty in these predictions regarding magnitude, rates, and regional patterns.

The Bangladesh economy is mostly rural, though its contribution to total GDP is steadily decreasing. In 1994, 35% of GDP was derived from agriculture but by 2007 agriculture, including its sub-sectors (livestock, fisheries and forestry) made up only 21.8% of GDP. Agriculture uses two-thirds of available land and employs about 75% of the workforce, and in 1996 it comprised about 80% of the nation’s export earnings (BBS, 1996). Within the rural sector, crops account for about three-quarters of the total production. Moreover, non-modernized agro-based industries are dependent on the agricultural sector for raw materials. Ninno et al. (2001) examined how food security in Bangladesh was maintained following the “flood of the century”, and they suggested measures that might be applicable to other developing countries facing similar natural disasters. Crop yields in Bangladesh are amongst the lowest in the world, although the soil in the river basin is quite fertile. This is because of the dependence on traditional methods of which the main technique is autonomous crop cultivation. There is no large-scale commercial farming in Bangladesh, all farming systems being based on subsistence methods in small land-holdings. Methods of cultivation and yearly flood hazards are factors which have a negative influence on food availability. Crop production and the resultant availability of food depend entirely on environmental factors-particularly flooding and its nature, frequency, and severity. Farmers make their decisions based on traditional predictive factors related to flooding characteristics. If their decisions are in accord with natural phenomena then there is usually sufficient production for their needs. This is a cyclical and continuous process in the agricultural system of Bangladesh.

Though farmers farming decisions are mostly based on traditional flooding characteristics and the crop productions of flood basins were entirely dependent on environmental factors up to the 1950s, but the development of dry-season irrigation and flood protection measures (e.g. the coastal embankment project) had tremendous influence on crop production since 1960s. Rice production in the dry season now exceeds production in the monsoon season. Also Bangladeshi farmers are highly sensitive to market prices, demand and supply.

2.7ConclusionFlooding cannot be totally prevented. The occurrence of a flood need not be considered a ‘failure’ and, conversely, minimization of losses may constitute a ‘success’. The first lesson is that there are always lessons to be learned, from every flood. Building awareness and understanding of the reasons for a system’s failure or inadequate performance, and identifying weak points using a holistic perspective is very important. Flood forecasting and warning systems fail because links in the chain perform poorly or not at all. The observation system may fail, the forecast may be grossly in error, the warning message may be wrong, the communication of a warning may be deficient and the response may be inadequate. A single weak point in a system, which otherwise contains many excellent components, may render the overall system performance unsatisfactory.

The components in flood forecasting and warning systems must be adequately integrated but responsibility may reside with different agencies. This requires adequate collaboration and coordination between multiple institutions, which is challenging. In emergency situations, it may become evident that distribution of roles of agencies is unclear and possibly redundant. The institutional framework is a key socio-economic determinant of a nation’s vulnerability against natural disasters.

People’s experience of flooding may reduce damage in the next flood. Where large floods occur in the same location twice in a short time period, losses during the second flood are typically far lower than those during the first (Kundzewicz and Takeuchi, 1999). The first flood will provide lessons for diverse groups: riparian homeowners; farmers with fields on the floodplain; professionals in the affected water district; legislators; spatial planning (zoning) officers; and public administrators at the country, province, town and community levels. Although flood events and human failures provide valuable lessons, memories can fade quickly after a flood. Typically, a destructive flood generates enthusiasm for strengthening flood preparedness systems and heavy expenditure follows. Following a deluge, the relevant authorities elaborate ambitious plans and launch works but lessons are soon forgotten.
Only informed stakeholders can make rational decisions and agree on an acceptable flood protection strategy, being aware of both costs and benefits. There may be conflicting interests between those living in floodplains and demanding efficient and very costly protection, and the rest of the nation. It is therefore important to prepare using a variety of different means, notably: rigorous implementation of zoning-using regulations to develop flood hazard areas and leaving floodplains with low-value infrastructure; strengthening existing defenses; building or enhancing flood mitigation monitoring systems; forecasting; issuing and disseminating warnings; evacuation; relief and post-flood recovery; flood insurance; capacity-building (improving flood awareness, understanding and preparedness); enhancing a participatory approach, including consultation on the preparedness strategy and the level of flood protection, and household-level flood-proofing and mitigation measurements for both newly built and existing properties.

Chapter-3: Method and Materials3.1 MethodologyEvaluation is the function whereby the eventual and/or actual results of a specific course of action are assessed. The assessment is done against the evaluative criteria. Evaluative criteria are set independently of the specific alternatives and are stated in parameters which furnish direct measurements on the results of an alternative course of action vis-à-vis the objectives of the system. ‘Objective’ is used in its very general form to mean the desired state of a system.
Performance evaluation has to provide answers to the following questions:
Does the system operate to meet the demand of the decision and beneficiaries makers?
How does it actually work?
Does it operate ‘satisfactorily’?
Can it be improved upon?
By what means?
What are the effects of treatment and change?
There is a need for the analyst/evaluator to be in a position to delineate carefully the system under evaluation, i.e. identify its goals, inputs, outputs, components and processes and to describe with the maximum possible precision the meaning of a “satisfactory” performance. The notion of a system operating satisfactorily relates to both its “output” and its specific operations. A direct relationship is postulated between output and the operation of a system. In other words, if a system operates as it ought, its output is expected to correspond and vice versa. However, this relationship may not always hold.
To evaluate the performance of the flood forecasting system of Bangladesh, this study uses generalized block-and-flow module. The module is made up of four basic components: Objectives, Evaluative criteria, Analysis synthesis and Evaluation. Objectives guide the behavior of the system which strives to attain them through its output (decision/acts). Depending on the degree of the system’s autonomy, the objectives are either set by the system itself or prescribed. Analysis/synthesis is the function whereby alternative courses of action, through which the system is supposed to attain its objectives, are generated. The module use in this study are shown in Figure 3.1.

Figure 3. SEQ Figure_3. * ARABIC 1 Methodology for evaluating flood forecasting performance3.2Research MethodsThe study aims to assess the performance of flood forecasting system of Bangladesh and to suggest some policy initiatives for improving the flood forecasting system of Bangladesh based on interviewing concerned officials of various government organizations as well as reviewing existing practices. Therefore, mainly qualitative and partly quantitative research method is more suitable in this regard and hence an exploratory comparative analysis has been used for the study.

3.3DataBoth the primary and secondary data has been collected and used in preparing this seminar paper.
3.3.1Secondary DataIn this study, flood forecasting performance has been reviewed and then it has been compared with the related policies, rules, acts of Bangladesh which has been treated as secondary sources of data. Other sources of secondary data have been collected from newspaper articles, books, journals and internet based publications.
3.3.2 Primary Data
In addition to secondary data, some primary data has also been collected for this study from various sources – mainly from key government offices by adopting Purposive and Snowball sampling procedure. It was anticipated earlier that shortages of time for data collection and possibility of non-availability of some of the concerned officials may hamper the process of collecting primary data. However, it was possible to collect primary data through interviewing twenty five (25) respondents from four (4) government organizations and local people. A well prepared questionnaire was supplied to them in advance and then sought their comments/suggestions of some issues related to flood forecasting system of Bangladesh. Table 3.1 gives composition of the respondents:
Table 3. SEQ Table_3. * ARABIC 1 List and Composition of the RespondentsKey Respondents Number of Respondents
1 Implementing Agency Officials (BWDB) 13
2 Policy formulating Officials (District Administration) 01
3 Other government organization (DAE, LGED) 02
4 Local People 09
Total 25
3.4 Processing of DataBased on Miles and Huberman (1994), I have followed three step procedures for interview data analysis which are storing, managing and processing. At first stage, I have made grouping for all data based on the variables, indicators and measures. Then at the second stage, I have looked closely the contents to get the sequence of a particular variable which match theoretically guided sequence found from literature survey. At this stage, emphasis has also given in searching more explanation from the key respondents based on actual field level experience which can explain the policy delivery. Later on at the third stage while writing I made triangulation of selected information, opinions and views with other sources. In case of hard facts, triangulation was done with documentary evidence and in case of soft facts, triangulation has been done with other respondent’s information. Secondly, the quantitative data has been processed with the help of MS Excel. The data collected along with the information obtained through observation and interviews were used to provide frequency and percentage distribution. Cross tabulation was conducted to analyze the data.
3.5 Ethical ConsiderationThroughout this research work, some policies and relevant of some countries like India, Sri Lanka and Netherlands have been reviewed and included. From the ethical point of view, prior permission should be sought from appropriate authority but this will not be followed here for time constraint and other practical reasons.

Chapter-4: Data Analysis and Findings4.1 IntroductionIn conducting this research work, both primary and secondary data has been used. Primary data has been collected through structured questionnaires. In conducting the questionnaire, two groups of five groups of respondents are identified. Group 1 includes personnel from BWDB who are responsible for producing flood forecasting and warning information and its dissemination. They are also responsible for improving the performance of the flood forecasting. Group 2 includes officials from BWDB who are working in the field level. In this group respondents are divided into two sub-groups: respondents from pre-monsoon flash flood affected area and respondent from normal monsoon flood affected area. Group 3 includes respondent from other implementing agency such as Department of Agriculture Extension (DAE) and Local Government Engineering Department (LGED). Group 4 includes respondents responsible for disaster management in field level such as officials from district administration. The fifth group of respondents is from the local people who are directly affected by the ravages of the flood. The mapping of the respondents is given below in Fig. 4.1.

Figure 4. SEQ Figure_4. * ARABIC 1 Mapping for the questionnaire surveyA total of eighteen (18) questions were prepared for the respondents from government officials covering areas of current initiatives, warning dissemination, expected information, strong sides of the flood information, usefulness of the information, way of using information, data sufficiency, improvement of data availability, sufficiency of technology, ways of improving technology, manpower sufficiency, ways of improving human resources capacity, inter-agency coordination and major drawbacks of flood forecasting system. On the other hand, a total of eight (8) questions were prepared for the respondents from local people.
In addition to the answer of the question the basic information of the respondent such as Name, Age, Address, Sex, Education, Marital Status, Occupation, Monthly Income, Mobile No were also collected.

The respondents were purposively selected from different government organizations having familiarities with the concept flood forecasting. The following sections deals with discussion on these aspects. The next section deals with analyzing the views of the respondents along with some published materials related with relevant policies/approaches in other countries. Based on this section, section three draws policy recommendation followed by conclusion.

4.2Evaluation of Flood BulletinFlood Forecasting and Warning Centre (FFWC) of BWDB is mandated for preparation of flood forecasting, early warning and its dissemination in Bangladesh (BWDB Act-2000). In order to meet the needs and expectations of flood forecast with increased lead times for cropping decisions, such as early harvesting, or to implement a contingency crop plan or protect infrastructure and preserve livelihoods. Although the livelihood of the people in Bangladesh is well adapted to normal monsoon flood, the damages due to inundation, riverbank erosion or breach of embankment, etc. still occur in various regions in almost every monsoon. They often have disastrous consequences: major damage to infrastructure, loss of property, crops, cattle, poultry etc., human suffering and impoverishment of the poor. With every major flood in Bangladesh, food security and poverty situation adversely affected.

Presently, FFWC has issued daily flood bulletin from May to October with a forecast lead-time of 24hrs, 48hrs and 72hrs, 96 hrs and 120 hrs (upto 5 days) along with warning messages and flood inundation maps. FFWC provides flood forecast 54 locations along the major river systems of Bangladesh. Location specific and impact based forecast is essential to reduce flood damage significantly. FFWC has also started to provide location specific forecast for flood infrastructure like Brahmaputra Right Embankment, Meghna-Dhonagoda Irrigation Project. Further improvement is needed to increase reliability of location specific forecast. Longer lead-time is always desirable to minimize loss and damage. But, there is lots of uncertainty with forecasting. FFWC is trying to increase forecast lead-time so that people can be benefited from flood forecast message.
The forecast of FFWC is very good up to 72 hours. With the increasing lead-time forecast performance affected due to some uncertainties in boundary estimation of flood model. The forecasting evaluation shows that was quite satisfactory for the major river system and needs further attention for small and flashy rivers.

The flood warning message was disseminated through different news media, news agencies, fax, e-mail, web-site (www.ffwc.gov.bd) and IVR (1090) through mobile phone. The flood forecast information has been used by various communities and organizations: national and international disaster management and relief operators, many Government agencies, NGOs and BWDB itself.

A sample flood bulletin developed by FFWC, BWDB is shown in Annexure-1. The flood bulletin provides following information as listed below
River situation and forecasting for major river basin at a glance
Number of locations where river is flowing above danger level. This also includes name of the station, name of the river, today’s water level, rise/fall of water levels in last 24 hours, danger level, above danger level in cm.

Rainfall situation in last 24 hours which includes location and amount of rainfall
Rainfall information in last 24 hours in Indian territory such as in Sikkim, Assam, Meghalaya and Tripura
River situation as of present date up to 9 A.M. which includes information on total stations, rising stations, falling stations, unchanged stations, stopped gauge, not measures gauge, information not available.

Detail forecasting information of 55 water level stations and 70 rainfall stations.

Details of 5 days forecasting information.
4.3Flood Forecasting Performance based on Secondary DataThe performance evaluation is conducted on a scale of ‘good’, ‘average’, ‘not satisfactory’, ‘poor’ and ‘very poor’ using data of 2011, 2012, 2013 and 2014 (BWDB 2011, 2012, 2013, 2014). The forecast quality gradually deteriorated where forecast intervals were moved further away from the time of forecast. Usually as lead time increases the accuracy decreases. This means that forecasts were the best at 24-hour interval (i.e. 24 hrs/1-day lead time) followed by 48-hrs interval and then 72-hrs (3-days). The performance of flood forecasting is summarized below in Table 4.1.

Table 4. SEQ Table_4. * ARABIC 1 Flood forecasting performance for 2011, 2012, 2013 and 2014Year Lead time (hrs) Total Station Good Average Not Satisfactory Poor Very poor
2011 24 31 23 4 2 2 –
48 18 9 – 3 1
72 13 8 8 – 2
2012 24 31 23 4 2 2 –
48 18 9 – 3 1
72 13 8 8 – 2
2013 24 31 23 4 2 2 –
48 18 9 – 3 1
72 13 8 8 – 2
2014 24 54 50 3 1 – –
48 38 11 3 2 –
72 27 14 5 6 2
96 20 13 9 6 6
120 11 21 6 4 12
The performance of the flood forecasting is evaluated for 2011, 2012 and 2013 and the summary is shown in Table 4.1. The total stations 31 and among the stations 23 stations were good with 24 hours lead time. Flood Forecast generated at 54 stations/points located within the model area (including some boundary stations) are evaluated for 2014. The forecast statistics along with their performance are provided in Tables 4.1. From the tables it may be seen that for 1-day forecast 98.15% for the stations are within the range of Good and Average. For 5-days forecast 59.26% stations are in the range of Good and Average for the monsoon of 2014. Especially few stations near boundary show poor to very poor performance.

4.4Flood Forecasting Performance based on Primary DataFlood message and dissemination:
Bangladesh is the part of world’s most dynamic hydrological and the biggest active delta system. The topography, location and outfall of the three great rivers shapes the annual hydrological cycle of the land. Too much and too little water in a hydrological cycle is the annual phenomenon. Regular monsoon event is the flood, the depth and duration of inundation are the deciding factors whether it affecting beneficially or adversely. Monsoon inflow along with rainfall historically shapes the civilization, development, environment, ecology and the economy of the country. Extreme events of flood adversely affect the development, economy, poverty and almost every sector. In flood management, Bangladesh has been taken structural and non-structural measures. One of the main nonstructural measures is the flood forecasting and warning. The FFWC is acting as the focal point on flood forecasting and warning services in co-ordination with concerned ministries and agencies like Ministry of Disaster Managment and Relief, BMD, DDM, DAE etc. during the monsoon for flood disaster management. Table 4.2 summaries the responses of questionnaire survey on the outcomes of the Flood Forecasting and Warning Center (FFWC) initiatives together with the information dissemination.

Table 4. SEQ Table_4. * ARABIC 2 Outcome of the flood forecasting center and information dissemination
FFWC initiatives Outcomes
(Sample size, N=25) Information of Dissemination
(Sample size, N=25)
Outcomes No. of Response Mode No. of Response
5days flood forecast 21 BWDB web site 20
Collect water level data 15 Mobile based SMS 20
Collect rainfall data 15 Electronic media (e.g. Television, Radio) 20
10days flood forecast 13 E-mail 20
Inundation Map for Bangladesh 10 Interactive Voice Response (IVR) System 13
Structure based forecast at 4 major locations. 5 Flood App 5
flood bulletin 5
The response from the questionnaire survey suggests that the FFWC of BWDB provides information on 5days flood forecast, 10days flood forecast, Inundation Map for Bangladesh, Structure based forecast at 4 major locations. They also collect real time water level and rainfall data. 21 respondents (out of 25) know that FFWC provides flood warning message before 5 days ahead. Respondent from BWDB only knows about 10days flood forecast. 15 respondents from BWDB, LGED and DAE know that the water level and rainfall data are collected by FFWC of BWDB. Inundation map is very important for flood management. However, less than 50 percent of the respondents, even the respondent of BWDB are ignorant of it. Inundation map could have been a very good tool to manage the flood more efficiently.
Response regarding the information dissemination suggests that most of the respondent (20 out of 25) know that flood information is disseminated to the local people, policy makers, other research organization through BWDB website, Mobile based SMS, electronic media such as TV and Radio and e-mail. With support from the Bangladesh Disaster Management Bureau (BMD), Cell Broadcasting (CB) has been started from July-2011 for flood warning message dissemination. Instant Voice Response (IVR) method is used; anyone can call 10941 from Teletalk mobile and hear a recorded Bangla Voice Message regarding days flood situation. As normal call charge applicable, the voice message is given within one minute duration. This method of innovative type disaster message dissemination is awarded in the Digital Innovation Fair 2011. However respondent from BWDB only mentioned about the IVR system. Only 5 respondents who are involved and working in the FFWC mentioned about the recently developed ‘flood app’ and ‘flood bulletin’ as the means of information dissemination.

Sufficiency of information and expectations:
The objectives of flood forecasting and warning services are to enable and persuade people, community, agencies and organizations to be prepared for the flood and take necessary actions to increase safety and reduce or protect damages of lives and properties. Its goal is to alert the agencies, departments, communities and people to enhance their preparedness and to motivate vulnerable communities to undertake preparedness and protective measures. In order to achieve the objectives it is very important to assess the sufficiency level of present forecasting system. Table 4.3 summaries the sufficiency level along with the expectation and strong points of the present forecasting system in Bangladesh.
Table 4. SEQ Table_4. * ARABIC 3 Sufficiency, expectation and strong points of the forecasting systemIndicators Response No. of Response
Sufficiency of current initiatives Yes 4
No 21
Expected information Increase lead time 20
Improve accuracy 20
Improve dissemination system 20
Depth and duration 15
Real time information 15
Affected area 15
Strong point of flood forecasting Deterministic forecast system 15
Dissemination system 15
Good Manager 10
Skill manpower 10
3-day flood forecast 80% accuracy 8
4-5 day flood forecast 60% accuracy 6
Usefulness of flood information Water level data 20
Rainfall data 20
Forecasting information 16
Inundation map 10
Means to use flood information Saving the human life & properties 20
Saving the flood control embankment & other structures 20
Expected water level, affected area 18
Gate operation of barrage 16
Flood preparedness 10
Flood relief 10
Emergency river bank protection 10
Interestingly, about 80 percent of the respondents mentioned that the present flood forecasting system is not sufficient to meet the demand of the present day. According to the respondent the lead time of the flood forecasting should be increased so that the flood information can be disseminated earlier to take appropriate measures for flood management. In addition 80 percent of the respondent think that the accuracy are needed to improve in order to avoid issuing misleading information. About 60 percent of the respondent suggests for providing information on depth and duration of flood together with real time information and affected area. The strong point of the flood forecasting system includes deterministic forecast system, dissemination system, good manager, skill manpower, 3-day flood forecast 80% accuracy, 4-5 day flood forecast 60% accuracy. In fact the daily forecast bulletin is prepared up to 5 days for important locations and region-wise flood warning messages. The bulletins are disseminated to more than 600 recipients including different ministries, offices (central & district level), individuals, print & electronic news media, development partners, research organizations, NGO’s etc. including President’s & Prime Minister’s Secretariat. Whenever, the forecast river stage cross the DL, the concern field offices and limited key officials are informed through mobile SMS. Interactive Voice Response (IVR) through mobile has been initiated since July 2011 through Teletalk. Now, all the mobile operators have started the IVR since 2015. About 80 percent of the respondent thought that the water level and rainfall data are very useful followed by forecasting information and inundation map. The flood information can be used to saving the human life ; properties, saving the flood control embankment ; other structures, guessing the expected water level, affected area, gate operation of barrage and other structures, flood preparedness, flood relief and emergency river bank protection.
The flood forecast is intended to alert the people of the locality about the predicted water levels of floodwater 3-days ahead of its occurrence. An accurate forecast would be one where the forecast level and corresponding observed level at the stipulated time are within a small range of variation. Flood inundation occurs due to overtopping or overflowing of flood water to the river banks. In our country, this situation at a particular place occurs when the river water level exceeds the danger level of that particular place. During normal flooding, it is expected and observed that flood plain along the major rivers becomes inundated and after that flood water progressively enters the adjacent residential and commercial areas depending upon the location and severity of flood situation. Flood inundation for whole country is a macro level product showing a general overview of flood situation of the whole country due to coarse resolution Digital Elevation Model (DEM). A detail, authentic and finer resolution DEM shall significantly improve generation of inundation maps and providing local level forecast. FFWC presented flood model domain does not cover coastal part, so model result is not appropriate for inundation analysis or verification of that part.

FFWC operates MIKE 11 FF Flood super model for flood forecasting purposes which excluded coastal region. Moreover periodic updating of river geometry is necessary for the best performance. Catchment characteristics, river morphology and climatology had been changed significantly which were not incorporated in the model in recent time. That’s why current inundation map provides underestimation as well as overestimation in some places.

Data, technology, interagency and manpower Sufficiency:
The real time hydrological data (85 water stations and 56 rainfall stations) is collected by SSB wireless, fixed & mobile telephone from the BWDB hydrological network. Water level for non-tidal stations is collected five times daily at 3 hourly intervals during day time from 6:00 AM to 6:00 PM, and for tidal stations collected hourly. Rainfall is collected daily period beginning at 9 AM. Limited WL, rainfall and forecasts of upper catchments from Indian stations are also collected through internet, e-mail, and from BMD. Estimation of water level at the model boundaries and rainfall for the catchments are required input to the model up to the time of Forecast (24, 48, 72, 96 & 20hrs). For the rainfall estimation, satellite images from NOAA and IMD is used. In addition a dedicated land line radar link with BMD provided frequent (five minutes interval) rainfall information. However, most of the respondent (about 60 percent) supposed that the present data is not sufficient to improve the lead time and accuracy of the flood forecasting. As such they think that data sharing with upstream country along with the automation of data collection system can improve the situation. The other alternative means to fill the data gap is to use the satellite information. Table 4.4 summaries the responses of the respondent on data, technology, interagency coordination and manpower sufficiency.

Collected/observed water level and rainfall data are given input to the computer database and checked. The water level and rainfall estimation has to be prepared. During monsoon (June to October) water level of few stations of upper catchments of Ganges, Brahmaputra, Teesta, Dharala and Barak rives has been received since 2010 from CWC India through e-mail. The basis for water level estimation is considering trend Hydrograph extrapolated upto the period of forecast from previous few days data, response characteristics of rivers, effect of rainfall on water level and Indian available water level & forecasts data. Rainfall estimation based on previous 2-day’s rainfall and analysis of information collected. After input required data and boundary-estimated data to the model, model run started. It takes about 30 to 40 minutes time to complete the calculations.
Table 4. SEQ Table_4. * ARABIC 4 Data, technology, interagency coordination and manpower sufficiency
Indicators Response No. of Response
Data sufficiency Yes 10
No 15
Improving Data availability Data sharing with upstream country 20
Automation of data collection 20
Use satellite data 20
Technology Sufficiency Sufficient 15
Not sufficient 10
Means to improve Technology advanced hydrodynamic modeling software 15
upgraded computer, printers and other machineries 13
Numerical Weather Products (NWP) 13
real time data from field and use of advanced use GIS-RS 12
Inter-agency coordination Adequate 4
Not adequate 21
Means to improve inter-agency coordination Seminar 20
Workshop 20
Meeting 15
Joint project 10
Manpower Sufficiency Sufficient 10
Not sufficient 15
Improving Human Resources Capacity Short Training 20
Long term training 20
Higher study 20
Workshop 15
On-job training 15
Conference and symposium 10
Responses from the primary data suggest that the technology is not sufficient. The respondent supposed that the advanced software is required and computation time is required to decrease. In addition the rainfall prediction system is to be improved. The respondent also thinks that the present interagency coordination is not adequate.
Flood happens in a relatively slower pace with a good chance of early warning. However, there is a good scope to improve the flood early warning mechanism to make it more people centric. Although stakeholders and community in Bangladesh are quite capable of responding to events such as floods, cyclones etc., there is a need for conducting flood response preparedness planning in advance. One reason for conducting response preparedness planning is because it will facilitate a rapid emergency response by allowing planners more time for advance preparedness measures for response. Time becomes more valuable once an emergency occurs, so planning before the emergency is very important, when workloads may be less and institutions involved are more flexible in accommodating the needs. Using Flood Response Preparedness Plan, in advance of Flood emergency, stakeholders will be able to:
Consider the likely consequences of a flood emergency before it occurs
Consider different risk scenarios to identify spatial response needs to suit prepositioning
Conduct capacity assessment to identify the key resources, both human and physical, needed for any flood emergency response.

Identify the critical areas for immediate action
Another benefit to response preparedness planning is that, before an emergency, there is comparatively more time to consider all the aspects of problems that are likely to arise. Once the emergency has occurred, it may be very difficult to bring all of the players together to discuss the needs. Agreement on policies and procedures in the response preparedness planning stage may help clarify applicability and resolve contradictions that may occur. It will help in filling the policy gaps in providing institutional mandates where needed. Rapid decision making on operational issues after an emergency is important because delays may cost lives. A good Flood Response Preparedness Plan ensures better preparedness for any emergency that may occur, even one that is very different from the scenarios in the plan.

Apart from the exposure to flood and other hazards, persistence of poverty is one of key underlying factor of vulnerability of rural households in the disaster-prone areas of Bangladesh. Flood is one of the factors and processes that have prevented certain groups of people in ecologically vulnerable areas escaping from extreme poverty. Flood refers to the state of seasonal unemployment and deprivation, especially in the northern districts and middle part of Bangladesh. Other socio-economic conditions such as lack of non-agro based employment, landlessness, malnutrition and lower level of education prevent these population to reduce vulnerabilities. Lack of access to health service, educational institutes, market and other institutions create a progressive vulnerability for these populations. Flood-prone zones are the worst off among different disaster prone areas in terms of food shortages, the incidence of extreme poor, insufficient income, illiteracy, and a high concentration of wage labors. Therefore, as expected, access to government programs like the VGD/VGF is the highest in the flood-prone zones. On the contrary, infrastructural services particularly that of roads, are more prevalent in the ecologically favorable areas.

Chapter-5: Conclusion and Recommendations5.1ConclusionsThe flood is a recurrent problem in Bangladesh. The characteristic of river varies from river to river and differs from region to region. The country is an active delta; it has numerous networks of rivers, canals and coast creeks with extensive flood plains through which surface water of about 1.72 million sq-km drains annually. Floods are normal monsoon phenomena in the deltaic plains of Bangladesh. Although the livelihood of the people in Bangladesh is well adapted to normal monsoon flood, the damages due to inundation, riverbank erosion or breach of embankment, etc. still occur in various regions in almost every monsoon. They often have disastrous consequences: major damage to infrastructure, loss of property, crops, cattle, poultry etc., human suffering and impoverishment of the poor. With every major flood in Bangladesh, food security and poverty situation adversely affected.

FFWC issued daily flood bulletin from May to October with a forecast lead-time of 24hrs, 48hrs and 72hrs, 96 hrs and 120 hrs (upto 5 days) along with warning messages and flood inundation maps. FFWC provides flood forecast 54 locations along the major river systems of Bangladesh. Location specific and impact based forecast is essential to reduce flood damage significantly.

The flood warning message is disseminated through different news media, news agencies, fax, e-mail, web-site (www.ffwc.gov.bd) and IVR (1090) through mobile phone. The flood forecast information has been used by various communities and organizations: national and international disaster management and relief operators, many Government agencies, NGOs and BWDB itself.

Improvement is needed to increase reliability of location specific forecast. Longer lead-time is always desirable to minimize loss and damage. The forecast of FFWC is very good up to 72 hours. With the increasing lead-time forecast performance affected due to some uncertainties in boundary estimation of flood model. The forecasting evaluation shows that were quite satisfactory for the major river system and needs further attention for small and flashy rivers.

Flood inundation for whole country is a macro level product showing a general overview of flood situation of the whole country due to coarse resolution Digital Elevation Model (DEM). A detail, authentic and finer resolution DEM shall significantly improve generation of inundation maps and providing local level forecast.

The available data, technology, inter-agency coordination and human resources capacity is not sufficient to meet the present demand. In order to improve the situation it is necessary to improve the data sharing mechanism with upstream country along with utilization of satellite data and automation of data collection system. To improve the insufficiency of the technology and improving the accuracy of the flood forecasting performance it is necessary to improve the rainfall forecasting system.

The present forecasting model covers Bangladesh only. However, Bangladesh is located at the outlet of the Ganges, Brahmaputra and Meghna Basin. Basin-wide model development with the aid of satellite based information is very crucial for improving the flood and water management of Bangladesh.
5.2RecommendationsFloods in Bangladesh occur for number of reasons. The main causes are excessive precipitation, low topography and flat slope of the country. The other factors include geographic location and climatic pattern, confluence of three major rivers, the Ganges, the Brahmaputra and the Meghna, the construction of embankments in the upstream catchments reduces the capacity of the flood plains to store water. The unplanned and unregulated construction of roads and highways in the flood plain without adequate opening creates obstructions to flow. The frequent development of low pressure areas and storm surges in the Bay of Bengal can impede drainage. The severity of flooding is greatest when the peak floods of the major rivers coincide with these effects. Climate changes could influence the frequency and magnitude of flooding. A higher sea level will inhibit the drainage from ther ivers to the sea and increase the impact of tidal surges. Deforestation in hilly catchments causes more rapid and higher runoff, and hence more intense flooding. The springtides of the Bay of Bengal retard the drainage of floodwater into the sea and locally increase monsoon flooding. A rise of MSL at times during the monsoon period due to effect of monsoon winds also adversely affect the drainage and raise the flood level along the coastal belt.

Many of these factors required active cooperation with the upstream countries for improving the prediction and management of flood. In addition basin-wide model development with the help of satellite data can also give better information for flood management.
References
Ahmad, Q. and Ahmed, A. (2003) Regional cooperation in flood management in the ganges-Brahmaputra-meghna region: Bangladesh perspective. Natural Hazards, 28, 181–198.

Ahmad, Q. K and Ahmed, A. (2004) Regional cooperation on water and environment in the ganges basin: Bangladesh perspectives. In M Mirza (ed.), The Ganges Water Diversion: Environmental Effects and Implications. The Netherlands/USA: Kluwer Academic Publishers.

Ahmad, Q., Biswas, A., Rangachari, R. and Sainju, M. (2001) A framework for sustainable development of the GBM region. In QK Ahmad, AK Biswas, R Rangachari and M M Sainju (eds.), Ganges-Brahmaputra-Meghna Region: A Framework for Sustainable Development. Dhaka, Bangladesh: The University Press Limited.

Ahmad, Q. and Rasheed, K. (1998) Regional perspectives in flood management, presented at national seminar on Flood ’98 and Management of Floods in Future, organized by National Committee for Relief and Rehabilitation, University of Dhaka, BUP, LGED and BFDR.

Ahmad, Q, Warrick, R. Ericksen, N. and Mirza, M. (1994a) A national assessment of the implications of climate change for Bangladesh: A synthesis. Asia Pacific Journal on Environment and Development, 1(1).Baqee, A. (1997) Coping with floods and erosion in Bangladesh Charlands. Asia Pacific Journal on Environment and Development, 4(2), December.Beven, K.J., 2006a, ‘On undermining the science?’, Hydrological Processes, (20) 3 141–3 146.

BWDB (2011) Annual Flood Report 2011. Flood Forecasting and Warning Centre, Bangladesh Water Development Board online. Available at: http://ffwc.gov.bd/images/annual11.pdfBWDB (2012) Annual Flood Report 2012. Flood Forecasting and Warning Centre, Bangladesh Water Development Board online. Available at: http://ffwc.gov.bd/images/annual12.pdfBWDB (2013) Annual Flood Report 2013. Flood Forecasting and Warning Centre, Bangladesh Water Development Board online. Available at: http://ffwc.gov.bd/images/annual13.pdfBWDB (2014) Annual Flood Report 2014. Flood Forecasting and Warning Centre, Bangladesh Water Development Board online. Available at: http://ffwc.gov.bd/images/annual14.pdfBWDB (2015) Annual Flood Report 2015. Flood Forecasting and Warning Centre, Bangladesh Water Development Board online. Available at: http://ffwc.gov.bd/images/annual15.pdfBWDB (2016) Annual Flood Report 2016 Flood Forecasting and Warning Centre, Bangladesh Water Development Board online. Available at: http://ffwc.gov.bd/images/annual16.pdfChowdhury, M. (1988) The 1987 flood in Bangladesh: An estimate of damage in twelve villages. Disasters, 12, 294–300.

Dankers, R. and Feyen, L., (2008) Climate change impact on flood hazard in Europe: An assessment based on high resolution climate simulations. Journal of Geophys. Res., doi:10.1029/2007JD009719.

FAP 16 (1995) Bangladesh flood action plan: Environmental study FAP 16: Completion report: A summary of activities 1991–1995. Ministry of Water Resources, Flood Plan Coordination Organization (FPCO), Irrigated Support Project for Asia and the Near East (ISPAN), Sponsored by the USAID.

Hall, J.W., Meadowcroft, I.C., Sayers, P.B. and Bramley, M.E. (2003) Integrated flood risk management in England and Wales. Natural Hazards Review, (4/3) 126–135.

Hall, J. W., Sayers, P. B. and Dawson, R. J. (2005) National-scale assessment of current and future flood risk in England and Wales, Natural Hazards, (36) 147–164.

Hirabayashi, Y., Kanae, S., Emori, S., Oki, T. and Kimoto, M. (2008) ‘Global projections of changing risks of floods and droughts in a changing climate. Hydrological Science Journal, (53/4) 754–773.
Islam, M (1974) Tropical cyclones: Coastal Bangladesh. In GF White (eds.), Natural Hazards: Local, National and Global. New York: Oxford University Press, pp. 19–25.

Islam, M (1980) Agricultural adjustment to flooding in Bangladesh. The National Geographical Journal of India, Vol. 26/1–2.Islam, M (1990) Environmental perception and agriculture. In S Huq, A Rahman and G Conway (eds.), Environmental Aspects of Agricultural Development in Bangladesh. Dhaka: University Press Limited.

Islam, M (1995) Environment, Land Use and Natural Hazards in Bangladesh. Dhaka University, Bangladesh.

Islam, N (2001) The open approach to flood control: The way to the future in Bangladesh. Futures, 33(8–9), 783–802.

Kundzewicz, Z.W., Hirabayashi, Y. and Kanae, S. (2010a) River floods in the changing climate -observations and projections. Water Resources Management (24/11) 2 633–2 646, doi:10.1007/s11269-009-9571–6.

Kundzewicz, Z.W., Lugeri, N., Dankers, R., Hirabayashi, Y., Döll, P., Pi?skwar, I., Dysarz, T., Hochrainer, S. and Matczak, P. (2010b) Assessing river flood risk and adaptation in Europe — review of projections for the future. Mitigation and Adaptation Strategies for Global Change, doi:10.1007/s11027-010- 9213-615:641–656.

Kundzewicz, Z.W., Takeuchi, K. (1999) Flood protection and management: Quo vadimus?, Hydrological Science Journal, (44) 417–432.

Nigg, J. (1995) Risk communication and warning systems. In: Hotlick-Jones, T., Amendola, A. and Casale, R. (eds), Natural Risk Smith, K. and Ward, R., 1998, Floods. Physical Processes and Human Impacts, Wiley, Chichester.Ninno, C. and Roy, D. (1999) The 1998 flood and household food security: Evidence from rural Bangladesh. Food Management and Research Support Project.Ninno, C., Dorosh, P., Smith, L. and Roy, D. (2001) The 1998 floods in Bangladesh —Disaster impacts, household coping strategies and response. Research Report I22, Washington, D.C: International Food Policy Research Institute.

Paul, B. (1995) Farmers responses to the flood action plan (FAP) of Bangladesh — an Empirical Study. World Development, 23(2), 299–309.

Paul, B. and Rashid, H. (1993) Flood damage to rice crop in Bangladesh. Geographical Review, 83, 150–159.

Pilarczyk, K.W. (2007) Flood protection and management in the Netherlands. In: (by Vasiliev O.F., van Gelder, P.H.A.J.M., Plate, E.J. and Bolgov, M.V. (eds), Extreme Hydrological Events: New Concepts for Security (NATO Science Series, IV: Earth and Environmental Sciences), (78), Springer, The Netherlands, pp. 385–407.

Rasheed, K. (2004) Flood mitigation measures: An exploratory case study of community role and perception. Journal of the Bangladesh National Geographical Association, 32, 1–7.

Rasheed, K. (2008) Bangladesh: Resource and Environmental Profile. Bangladesh: AH Development Publishing House.

Sivakumar, B. (2008) Undermining the science or undermining Nature?, Hydrological Processes, (22) 893–897.

UNISDR: Developing early warning systems: a checklist, United Nations International Strategy for Disaster Reduction, Geneva, Switzerland, available at: http://www.unisdr.org/2006/ppew/info-resources/ewc3/checklist/English.pdf, 2006.

UNISDR Sendai Framework for Disaster Risk Reduction 2015 – 2030. Third UN World Conference in Sendai, Japan, 18th March 2015, available at: http://www.unisdr.org/files/43291_sendaiframeworkfordrren.pdf, 2015.

Warrick, R., Bhuiya, A. and Mirza, M. (1996b) The greenhouse effect and climate change. In RA Warrick and QK Ahmad (eds.), The Implications of Climate and Sea-Level Change for Bangladesh. Kluwer Academic Publishers.White, G. F. (1945) Human Adjustment to Flood. Chicago: University of Chicago.

Younus, M. (2012a) Farmers household information in Bangladesh: Significant indicators for assessing vulnerability and adaptation. The Journal of Bangladesh National Geographical Association, 37(1–2).

Younus, M. (2012b) Autonomous crop adaptation processes to extreme floods: A case study in Bangladesh. In conference proceedings on water and climate: Policy implementation challenges, Canberra, Australia, National Convention Centre, 1–3rd May, 2012.

Younus, M. and Harvey, N. (2013) Community-based flood vulnerability and adaptation assessment: A case study from Bangladesh. Journal of Environmental Assessment Policy and Management, 15(3), 32 pages.Younus, M. and Harvey, N. (2014). Economic consequences of failed autonomous adaptation to extreme floods: A case study from Bangladesh. Local Economy, 29(1–2), 22–37.

Annexure-1: Flood warning message of BWDB

Annexure-2: Questionnaire to Conduct SurveyQuestionnaire to Conduct Survey
forSeminar paper
onPeople’s Perception about Bangladesh Railway as a Mode of Transportation
A Survey plan to conduct as a participant of 120th Advance Course on Administration and Development (ACAD) as part of ACAD course for fulfilling the requirement of preparing seminar paper. Data collection from the survey will be used for academic purpose. The identity of the respondent will not be disclosed. To complete the survey a questionnaire is developed with set of questions. There are two types of questions includes in the questionnaire, some questions are open ended and the remaining can answer through providing tick mark. It will be highly appreciated if you participate in survey and provide your valuable feedback.

Questionnaire for Officials
Part -A : Personal Information-
1. Name:
2. Age:
3. Address:
4. Sex: Male /Female
5. Education:
6. Marital Status:
7. Occupation:
8. Monthly Income:
9. Mobile No.-
Questionnaire for Officials
What are the current initiatives/actions of your organizations in flood forecasting?
How flood warning is disseminated to the local people?
Do you think that the current initiatives are sufficient?
149225044450005778504445000Yes No
What are the information expected from flood forecasting system?
a)
b)
c)
What are the strong points in flood forecasting system?
a)
b)
c)
Which information is useful in flood forecasting system?
a)
b)
c)
How information can be used for flood management?
a)
b)
c)
Do you think that the data is sufficient for flood forecasting?
a)
b)
c)
How to improve data availability for flood forecasting?
a)
b)
c)
Do you think that the technology is sufficient for flood forecasting?
a)
b)
c)
How to improve the technology for flood forecasting?
a)
b)
c)
Do you think that the manpower is sufficient for flood forecasting?
a)
b)
c)
How to improve the capacity human resources for flood forecasting?
a)
b)
c)
Do you think that the present inter-agency coordination is sufficient for flood forecasting?
a)
b)
c)
How to improve the inter-agency cooperation in flood forecasting?
a)
b)
c)
What are the major drawbacks in flood forecasting system?
a)
b)
c)
How the local people can be involved in flood management?
a)
b)
c)
What can be done for the further improvement of flood forecasting system?

Questionnaire to Conduct Survey
forSeminar paper
onPeople’s Perception about Bangladesh Railway as a Mode of Transportation
A Survey plan to conduct as a participant of 120th Advance Course on Administration and Development (ACAD) as part of ACAD course for fulfilling the requirement of preparing seminar paper. Data collection from the survey will be used for academic purpose. The identity of the respondent will not be disclosed. To complete the survey a questionnaire is developed with set of questions. There are two types of questions includes in the questionnaire, some questions are open ended and the remaining can answer through providing tick mark. It will be highly appreciated if you participate in survey and provide your valuable feedback.

Part -A : Personal Information-
1. Name:
2. Age:
3. Address:
4. Sex: Male /Female
5. Education:
6. Marital Status:
7. Occupation:
8. Monthly Income:
9. Mobile No.-
Questionnaire for Beneficiaries
Do you know about the flood forecasting and warning?
a)
b)
c)
How do you get the flood forecasting and warning?
a)
b)
c)
What is your rating on forecasting and warning information?
a)
b)
c)
Do you think the flood forecasting and warning easily understandable?
a)
b)
c)
Do you evacuate based on forecasting and warning information?
a)
b)
c)
What are the major drawbacks in flood forecasting system?
a)
b)
c)
How the local people can be involved in flood management?
a)
b)
c)
What can be done for the further improvement of flood forecasting system?