| GISdevelopment.net ---> Application ---> Natural Hazard Management |
Development of flood warning system
Farah Aziz, Nitin Tripathi, Mark Ole and Michiru Kusanagi
Asian Institute of Technology, Bangkok, Thailand
nitinkt@ait.ac.th
Farah Aziz, Nitin Tripathi, Mark Ole and Michiru Kusanagi
Asian Institute of Technology, Bangkok, Thailand
nitinkt@ait.ac.th
Abstract
Bangladesh is considered to be a country miserably affected by recurring floods with devastating dimensions exposing the national economy in the hands of nature. Complete flood control in the geographical context, particularly in the deltaic form of Bangladesh is not at all a feasible option. Structural methods of flood protection are neither economically viable nor these are environment friendly. Therefore, non-structural methods are becoming popular in mitigating flood disaster. A good way to prevent/ reduce damages occurred from flood is to develop a Flood Forecasting and Warning System in the affected area. This study is an attempt to fulfill such requirement by the extensive use of GIS, providing spatial information for assessment of flood vulnerability.
A methodology is developed by integrating Danish Hydrodynamic Model MIKE 11 and GIS. The MIKE 11 flood forecasting system comprising of three modules of NAM, HD and FF. FF module is used for flood foresting. The Geographic Information System (GIS) in this has proven to be a very effective tool for describing, analysing, modeling and integrating forecasted flood levels with other related information such as topographic, thematic and attribute information. It offers new opportunities to develop and implement a user-friendly, interactive decision support system for flood forecasting and identifying the affected areas using dynamic spatial modeling.
The study concludes that an effective Warning System that can release warning in advance, i.e. 72hrs, 48hrs and 24hrs. It can change the existing scenario substantially and render informed decision making in adopting proper measures towards disaster preparedness, mitigation, control, planning and management. This kind of advance warning can help the authorities for better flood preparedness and also effective flood mitigation.
Introduction
Floods are the major disaster affecting many countries in the world year after year. It is an inevitable natural phenomenon occurring from time to time in all rivers and natural drainage systems, which not only damages the lives, natural resources and environment, but also causes the loss of economy and health. The impact of floods has been increased due to a number of factors, with rising sea levels and increased development on flood plain (Sanders and Tabuchi, 2000). Recurring flood losses have handicapped the economic development of both developed and developing countries.
Damage from flooding has been increasing each year resulting in loss of lives, property and production as well as affecting activities in the flooded areas. Large and long duration of flooding can be considered as the economic loss of the country. The non-structural methods of mitigation of flood hazards are very cost effective as compared with structural ones (dams and dikes). Among non-structural methods, modern flood forecasting and the association with real-time data collection systems have increasingly found favor with countries prone to flood hazards. Flood risk mapping is required to provide information concerning flood risk areas to residents in flood prone areas and to establish flood protection and evacuation system. In determination of Decision Support System for flood risk assessment, it is of utmost importance to apply the most efficient methods in flood forecasting and warning system associated with real-time data collection system. (Sirivilal Kanchanaham, 1997)
From Bangladesh perspective, floods are among the most regular and hazardous natural disasters in terms of human suffering and economic losses. It has been identified that, timely flood forecasts and warnings are key elements to aid disaster preparedness, which in turn will reduce flood damages and human sufferings in a great extent.
The existing systems of collecting information on floods extent and their effects are not very reliable. The system depends heavily on field information, which sometimes is erroneous and at times cannot even be collected until the recession of the floodwaters (EGIS, Dhaka, June 1998). Information from GIS can be used to extract some types of information, which are otherwise difficult to access by traditional methods, particularly for flood-forecasting and floodwater movement. This present study implies the application of Geographic information Systems technologies to develop a new model for forecasting floods rather than flood mapping for flood risk assessment in flood prone countries for regular monitoring of damages. The importance of the flood forecasting and warning is widely recognized as a vital non structural measures to aid the mitigating the loss of life, crops and property caused by the annual flood occurrence.
Flood Mitigation
Flooding cannot be completely avoided, but damages from severe flooding can be reduced if effective flood prevention scheme is implemented. This can be achieved if the sufficient information for flood forecasting is acquired both in time and in quality. Hydrologic applications of GIS range from synthesis and characterization of hydrologic tendencies to the prediction of response to hydrologic events. The payoff comes from the multiple ways in which the data can be used once it is made to be digitally accessible in a GIS (Tawatchai, 1999).
Use of GIS will provide supplementary data in Hydrology for such analysis and will lead to easier interpretation and understanding of flood phenomena and characteristics. The use of Digital Elevation Model (DEM) can be effectively used for simulation to get a complete model of the study area.
The objective is to develop an effective flood warning system integrating GIS, hydrologic models, emergency response strategies, and expert knowledge into the system.
Fig.1: Development of flood warning system
Study Area
Sundarganj Thana of Gaibandha district in Bangladesh has been selected as my study area. It is located in the northern part of the country. Two major rivers bound Sunderganj, Brahmaputra (Jamuna) in the eastern and Teesta in the northern part of Bangladesh. Total area is 418.74 Sq. km and population is 364,432. Sunderganj thana consists of 15 unions and Tarapur is one of them. The size of urban area of Sunderganj thana is 5.0 Sq. km, whose population is 9, 940. Area of household in Tarapur union is 6.42 Sq. km; population is 25,796(Source: BPC, District: Gaibandha, June 1994)
Floods in Sunderganj
The area is subjected to flood almost every year from both rivers the Teesta and the Brahmaputra, with both beneficial and adverse effects. The river Teesta causes flash flood, monsoon flood is caused by the river Brahmaputra. The terrain is basically alluvium flood plain and is not much stable as the river courses changes continuously. The region is characterized by shallow depressions and valleys of moribund river channels created by a long morphological history of changes in the river courses.
The rivers in the extreme northern area (Teesta, Dudkumar and Dharla) have steeper gradients (1 in 2000) than elsewhere and most of their catchments lie in India and Bhutan. They frequently cause flash floods. The part of the region along Brahmaputra suffers severely from river flooding caused by breaches, mainly in the main Brahmaputra River and to limited scale in the Teesta River. Inside the region, flooding and drainage patterns of the internal rivers which have catchment areas basically within Bangladesh and have very flat gradients of 1 in 5000. Internal rainfall flooding is therefore common. The severity of flooding may be exacerbated by rainfall within the region. The region experienced a devastating rainfall floods in September-October 1995. Such severe rainfall flood also occurred in 1922. (Source: International Commission on irrigation and Drainage, Bangladesh national Committee of ICID)
The health conditions of the villagers of the area were found to be serious due to severe flood years of 1987, 1988, 1991, 1993 and 1998. Seventy percent of people suffered from dysentery, diarrhea, typhoid and the remainder had fever, the rest of them showed skin infections. People in this area termed the stress in floods as
- Damage of crops
- Considerable loss of households
- Threatening to human health
Fig.2: Customised Dynamic GIS based flood warning system
Activities of Flood Forecasting and Warning System in Bangladesh
After the devastating flood of 1988, the government of Bangladesh took initiative to modernize the operation of Flood Forecasting & Warning Center (FFWC) by adopting the state of art technology and integrating it into the forecast and warning dissemination process. The present flood forecasting and warning system in operation is composed of 4 main elements, which are:
- Real-time rainfall and water level data collection
- Meteorological forecasting
- Flood forecasting
- Flood warning dissemination
Hydrodynamic Model MIKE 11 for Flood Forecasting
Rainfall Runoff Model NAM
NAM is a precipitation-runoff model. This model was developed by Nielsen and Hansen (1973) at the Institute of Hydrology and Hydraulic Engineering at the Technical University of Denmark. NAM simulates the rainfall-runoff process in catchments. It operates by continuously accounting the moisture content in four different and mutually inter-related storages, i.e.: Snow Storage, Surface Storage, low Zone Storage, Ground Water Storage representing physical elements of catchment. NAM comprises of a set of linked mathematical statements describing, in simplified quantitative form, the behavior of the land phase of hydrological cycle. The model is defined as a deterministic, conceptual, lumped type of model with moderate data requirements. The model area can be divided into a number of sub-catchments. Each catchment is treated as one unit so parameters are representative of average values for the entire catchment. NAM is based on a set of linked mathematical equations, both empirical and semi empirical. NAM simulates rainfall-runoff process in rural catchments.
HD Model
The MIKE 11 hydrodynamic module, an implicit, finite difference model for the computation of unsteady flows in rivers and estuaries. The model can describe sub critical as well as supercritical flow conditions through a numerical scheme, which adapts according to the local flow conditions (in time and space).
MIKE 11 HD applied with the dynamic wave description solves the vertically integrated equations of conservation of continuity and momentum (the Saint Venant’s equations) , based on the following assumptions:
- Water is incompressible and homogeneous; i.e., negligible variation in density;
- The bottom slope is small, thus the cosine of the angle makes with the horizontal may be taken as 1
- The wavelengths are large compared to the water depth. This ensures that the flow everywhere can be regarded as having a direction parallel to the bottom, i.e., vertical accelerations can be neglected and a hydrostatic pressure variation along the vertical can be assumed
- The flow is sub-critical (Super critical flow is modeled in MIKE 11, but using more restrictive conditions.
Flood forecasting systems, producing real-time forecasts of river flows and levels, provide a cost-effective solution to many flood management problems. It has been designed to perform the calculations required to predict the variation in discharges and water levels in a river system as a result of catchment rainfall and inflow/outflow through boundaries in the river system. From the MIKE 11 –FF module, the hydrodynamic model HD and the rainfall-runoff model NAM are controlled. All necessary input data and runtime parameters can be specified from MIKE11-FF. The MIKE 11- FF module has a real time data management facility with direct access databases and user-designed data entry menus. It reduces to a minimum the time required for entering of real-time and forecast data and other needed information for the simulations. In the add-on module, MIKE 11 FF, flood-forecasting techniques are used to predict local flood levels and river discharges. The forecasts can be used to set up control strategies for reservoir operation, which will prevent or reduce flooding in the downstream reaches and avoid unnecessary waste of water resources. Moreover, the forecasts form the basis for the dissemination of warnings to local authorities and the public. The forecasts provide information on the time scale and the extent of the flood inundation expected locally. Consequently, flood-forecasting techniques constitute a viable and important tool within flood management.
Development of Flood Warning System
Methodology of development of warning system is summarized in the flow chart (Figure 1). Forecasted water levels have been imported to MIKE GIS to map flood inundation maps. These maps are intersected with land use/ settlement maps to delineate flood inundation areas using techniques involved in GIS.
MIKE 11 and GIS are successfully integrated in Arc View GIS environment for retrieving near-real time flood level information. The study illustrates that an effective warning system can release warning in advance say 72, 48 and 24 Hrs in advance. It can change the existing perspective of flood preparedness and mitigation substantially and render information for better decision making for saving lives of people.
Customised and Automated Evacuation System
Adequate flood warning is a crucial element in emergency management, and operations of all of the systems around flood forecasting that the flood forecasting actually deals with.
The Network Analyst was employed to calculate the best way to get from one location to another or the best way to visit several locations. We can specify the locations by pointing to places on the screen, by entering address, or by using point information. We can decide the order they are visited, or we can let Network Analyst find a visiting sequence for us.
Finding best route
Best Route means different alternate route in different situations. For example, while people are affected under disaster, they need to be rescued by any means. So the selection of the quickest route to the hospital or to the cyclone shelter is very important at that moment.
The network Analyst supports different objectives, such as traveling quickly, traveling by the shortest route, traveling by the most interesting route, or any other criteria we specify. WE need only to specify an appropriate cost field in our line theme’s feature table when solving the problem. We can also make a route according to our personal preferences by specifying locations to pass along the way. The cost field we use can be in any units of distance or travel time, such as minutes or hours if we are finding a route that minimizes travel time, or kilometers or miles if we are finding a route that minimizes travel distance.
Finding the Closest Facility
“Closest facility” refers to anything providing a certain type of service that is closest to a given location, known as event. When finding the closest facility, we can specify whether the direction of travel is from the event to the facility, or whether it’s from the facility to the event. Direction of travel can be an important distinction because traffic pattern, one-way streets, and prohibited turns may make a facility more remote in one direction than in another. Identifying the closest facilities is only part of the solution. We also need to know the best way to get to or from them. Whenever we solve a closest facility problem, the Network Analyst also finds the best routes to travel to or from the facilities.
Customisation
The model was customised to make it user friendly, interactive, using Arc view script Avenue and Dialog Designer. To customize Arc View for our own use, we have created default settings for each default type (View, Table, Chart, Layout, Script, Project) and also create new user interfaces.
The project window contains a scrolling list of types, which by default contains View, Table, Chart, Layout and Script. With Customise Types, we can remove some of these types, we can change their names, we can change the icon that appears in the scrolling list, and we can reorder the types (e.g., move Scripts before Views). We can also create new types. Through the project window, we can create “virtual” documents and organize our project in ways that make sense for our application.
The methodology and database has been customised in Arc View GIS for user friendly interface and easy implementation. Figure 2 shows the customised window of the Dynamic GIS based Flood Warning System.
References
- BPC, 1994, Bangladesh Population Census, 1994, Zila: Gaibandha: 1994, Bangladesh Bureau of Statistics
- EGIS, 1998, Microwave Remote Sensing Applications for Flood Monitoring. Environmental and GIS Support Project for Water Sector Planning, Dhaka.
- Sanders R. and Tabuchi S., October, 2000, Decision Support System for Flood Risk Analysis for River Thames, United Kingdom, Photogrammetric Engineering & Remote Sensing Journal, Vol. 66, No. 10
- Sirivilai K., 997, Application of Flood Plain Mapping and Risk Assessment by the Integration of Fused Multisensor Remote Sensing Derived information within a GIS, AIT thesis No. SR – 97- 12, Bangkok, Thailand.
- Tawatchai, T., 1999, Use of Geographic Information Systems and Remote Sensing in Hydrological modeling. A Publication of the School of Civil Engineering, Asian Institute of technology, Vol. 1, No. 1.pp.7