Introduction This chapter elaborates the overview of the research. Background of the research and the researchproblem are discussed and objective is stated based on the research problem.1.1Background Among number of natural disasters, flooding is considered asone of the most devastating disaster that affects many parts of the world. Floodingposes a serious damage on human population as well as on natural environment. Itexerts huge damage of infrastructure, properties and human life. Floods areconsidered as the most common, costly and deadly natural disaster (Patro et al,2009).
Due to climate change and rapid urbanization, the frequencyand magnitude of floods tend to be higher (Liu et.al, 2012). At the same timedevelopment of floodplains for settlements and other industrial activitiescontinues. With high urbanization, urban flooding due to insufficient drainagecapacity to drain out the excess runoff adds up to flash floods and riverfloods.Therefore flood mitigation plays an important role in safetyof the urban areas. Both structural and non-structural measures have been takenby relevant authorities to make sure the loss due to floods are controlled. As a non-structural measure of flood mitigation, floodforecasting and early warning systems are very important tools to manage and reducethe damages (Burlando et al., 1993).
Timely forecast are valuable forauthorities to make decisions efficiently and evacuate people to safer places(Werner et al., 2013). Whereas inaccurate and late forecasts will just wastethe efforts of management (M. G. Anderson & Burt, 1985). Commonly used method for flood forecasting is to record realtime rainfall records using rain gauges, converting them to discharges using a rainfallrunoff model and then forecasting water levels using a hydraulic model (Burlandoet al.
, 1993). For urban catchments with lower response time, flood forecastingwith real time rainfall records might not be sufficient. In such cases rainfallforecasts have to be used in place of real time rainfall so sufficient leadtime is given to authorities to take decisions (Harris et al. 2007). Flood forecasts are generally carried out using a two-step process.Rainfall runoff is modeled using a hydrological model and resulting floodhydrographs are converted to water level forecasts by using hydraulic models(Hicks and Peacock, 2005). In the current research, water level forecasting isdone by utilizing a previously calibrated runoff model and two-dimensional hydrodynamicmodel.Hydrodynamic modeling helps to understand flood hazard andthis is a useful tool in flood mitigation activities as well.
Hydrodynamicmodels route the water over the digital elevation model with rainfalls andupstream discharge hydrographs and downstream boundary conditions of waterlevels (Damayati, 2011).In SriLanka, Colombo is highly vulnerable to floods. Floodrisk of the Colombo and surrounding areas has increased by rapid urbanizationwhich has caused unrestricted development activities within flood prone areas.The expansion of human settlements in low lying flood plain areas and loss ofwetlands has made the region highly vulnerable to floods. Metro Colombo hasexperienced regular floods over the past 30 years, affecting a large number ofpeople annually (Patankar, 2017).Colombo city and the Kolonnawa region experience floodingdue to two major climatological/hydrological phenomena. The first phenomenon ishigh-intensity rainfall with short-term duration in the city area.
The secondis high discharge through Kelani River due to high rainfall in the uppercatchments of the Kelani River. When these two phenomena coincide the city ofColombo faces the worst flood conditions. During heavy rainfall, the rainwateris discharged under gravity through the urban water system into the KelaniRiver or directly to the ocean. However, when the Kelani River is high, theriver outfalls are blocked and the water levels in Metro Colombo will rise. In Sri Lanka, flood warning is still not based onsophisticated methods.
The trend of the river stage is judged by continuousmonitoring of river gauges. The current river stage, the trend noticed, thelatest rainfall observations within the catchment, the rainfall forecastingmade by the meteorological department, and the past experiences are used toissue flood warnings. In past several attempts have been made to develophydrodynamic models for Metro Colombo and Lower Kelani region (De Silva et al.2012). Most of the models have developed flood hazard maps and identified vulnerableareas. Though this research it is expected develop a detailed model which willbe used to forecast future flood events. 1.
2 Problem DefinitionThe Colombo metropolitan region, the commercial andfinancial center of the country, is home to 28% of the country’s 20 millionpeople and accounts for about 50% of gross domestic product. Thecity’s drive to emerge as a competitive player in the global economy has beenundermined by challenges such as urban flooding and there is a need to reducethe impact of floods if the country will be to tap into the competitiveadvantages of the Colombo metropolitan region in order to have an accelerategrowth in the future years. Floods in Colombo are caused by heavy rainfall on a drainagesystem with insufficient storage, conveyance and outflow capacity. Severeflooding has been recorded in the Greater Colombo basin in the last few decadeson several occasions.
1.3 Research ObjectivesConsidering the major issues discussed above, it is veryimportant to develop a proper flood forecasting system for early warning thatcan help save lives and livelihood of people. Hence, the main objective of thisresearch is to study the possibility of developing a reliable hydrodynamicmodel for Metro-Colombo basin. Thespecific objectives would be to develop a hydrodynamic model for real timeflood forecasting and as a tool for flood mitigation activities.