Abstract:
Megacity Dhaka, the capital of Bangladesh has been ranked top in urban population density. Urban waterlogging from intense rainfall is a recurring phenomenon in Dhaka. In recent years, it has become a common picture in Azimpur, BUET and Lalbag that if there is a rainfall of moderate to severe intensity, a major portion of the area goes under water. This study is an initiative to investigate the potential of Rainwater Harvesting in minimizing the waterlogging problem in BUET Campus.
Bangladesh University of Engineering and Technology (BUET) campus located in southern Dhaka, experiences acute waterlogging problem even for a very short duration of intense rainfall. BUET is situated in a low-lying area compared to the surroundings. The natural slope of BUET campus is from north to south with a natural depression in the southern part where runoff is driven by gravity. Rainwater Harvesting (RWH) can be a viable approach to reduce the waterlogging problem of Dhaka. To assess the impact of incorporating RWH system in urban areas to reduce waterlogging and save water, the entire BUET campus has been selected as a study area.
Hydrodynamic models help in simulating runoff depth, estimating inundation extents, and indicating waterlogging hazard areas. Researchers have observed a positive relationship between RWH and urban flood mitigation. This research focuses on the identification of regions in the study area that are vulnerable to frequent waterlogging by creating a 1D-2D coupled inundation model and simulating rainwater harvesting as low impact development solutions for overcoming such vulnerability using Mike+ software developed by DHI. By simulating rainfall events using IDF curves for 2, 5 and 10 years return period 1 hour duration rainfall events, the model identifies areas where drainage network reached maximum capacity and subsequent overflow creating waterlogging.
Analyzing the land elevation and land use patterns of the frequently waterlogged areas gives an idea about the possible reasons for waterlogging in these areas. The imperviousness of the study area is very high due to low soil permeability and paved surfaces. The hydrological model showed that BUET campus is very vulnerable to overflowing during low to medium rainfall events in low lying areas.
Low impact development solutions like RWH are generally suggested for primary treatment or cleaning first flush from low-intensity small duration rainfalls and delaying peak runoff at source. From this study it has been found that RWH does not have significant effect in reducing waterlogging in the study area but has some effect on delay of peak and reduction of water depth in 2 year and 5-year return period rain. For a 2-year return period rain when runoff from 50% of hard runoff area of the roofs is stored in rain barrel instead of flowing out, the catchment discharge reduces to 50% of base scenario flow. Inundation maps for the 2, 5- and 10-year return period without RWHS shows that 10-year return period 1 hour duration rain event has the maximum water depth (0.45m maximum) in multiple regions compared to the 2 and 5-year events of same duration. Among the simulated events with RWHS the 5-year return period rain event shows a 0.003 m to 0.01 m decrease in depth 0.092 km2 area and 0.01 to 0.067 m decrease in 0.051 km2 area which is the maximum decrease among the 2-, 5- and 10-year return period rain events simulated. For the 2-year return period rain event, usage of RWHS decreases inundation water depth in significant extent of areas but for 10-year return period rainfall event the impact of RWHS in reducing inundation depth is less.
The waterlogging problem in the study area is mostly due to topographical condition and elevation differences. A combination of solutions with RWHS, underground stormwater reservoirs and sufficient stormwater drainage network is necessary to mitigate the problem. However, RWH appears to be an excellent way to supplement water demand while delaying the peak runoff.