Dhaka city is frequently affected by flash floods almost every heavy rainfall. The heavy rainfall is also caused by convective storms and/or tropical cyclones along with seasonal rainfall.This heavy rainfall waterlogs the city. This waterlogging buildings become at risk of building collapse. One example of building collapse occurred in 2013 at Savar. The reason of falling down was poorly engineered and designed as commercial buildings nearby lakes. This accidents crossed a total death toll of 1100 (the news of Rana plaza can be followed in the given link). Moreover from the study site description, it can be seen that the land use has frequently grown to a large metropolitan where numerous number of population occupying Dhaka city in multipurpose activities such as business, education, treatment, hospitals etc. Along with economy growth, the area such as canal market and other open space is taken for constructing buildings. However, the drainage system for rainwater is not constructed well.  Eventually, waterlogging issues and flash floods are increasing vulnerabilities to the buildings that can result in collapse which will cause damages to properties and deaths. Furthermore, the waterlogged city also an adverse situation and suffering for city people at work and their busy lives. In conclusion for flash flood assessment, Dhaka city should be addressed with much more importance in improving rainfall forecasts and drainage systems. 

​

the link: http://www.globallabourrights.org/campaigns/factory-collapse-in-bangladesh 

The observation rainfall data is only for Dhaka city to assess flash floods. However to be noted, the model forecasts are for generally for Bangladesh. The rainfall distribution for Dhaka city is show in figure 4 (a) and (b). Figure 4 (a) shows monthly variations for 15 years and evolution of rainfall distribution. The rainfall data were daily (mm/d) but were calculated for monthly mean to show seasonal distribution. Figure 4 (b) shows a trend of climatic variation by looking at the monthly mean. The dark line represents the monthly mean for 15 years data and the rectangular box shows the most rainfall in that range. The blank circles are outliers. The two lines from the rectangular box is the quartile range where the lowest one show the minimum 1st quartile and last one is 3rd (75%) quartile. This figure shows nearly a normal distribution to understand climatic conditions whereas the figure 4 (a) shows the peak rainfall as well. This peak values showed that month occurred flash floods. There two different studies of flood and rainfall trend in Bangladesh using climate and regional model. The model and research description is given in table 1. In Islam (2009) paper, there has been studied in a regional climate model named as Providing Regional Climates for Impacts Studies (PRECIS) for generating rainfall scenario for Bangladesh climate. This model is used for SAARC regions. 

This model is to provide a general trend for climatic change impact assessments in terms of rainfall for floods and temperature for droughts. The PRECIS model forecast is shown in figure 5. The model has forecasted by studying a climatic trend from 1960 to 2000 year. The initial observations were taken from Bangladesh meteorological department (BMD) of 27 stations to obtain a general trend for Bangladesh. Since Bangladesh does not have evenly distribution of rainfall the trend from the model seems very high, Dhaka city has lower and less rainfall comparing to entire countries. Thus for assessing Dhaka city flash floods, there is a necessity to study specifically rainfall distributions. Moreover, the climatic trend cannot provide the fluctuations that are shown in observed rainfall [fig 5 (a)]. The PRECIS model also has biases that were calibrated and statistically calculated to plot the climatic rainfall trend. To avoid such biases, the best model physics should be chosen for Bangladesh weather. To specify. Kain-Fristch (KF) seems better cumulus parameterization schemes for Bangladesh that is studied for Tropical cyclones convective rainfall (Fahad, 2015 and Mallik, 2015).  A similar study has been taken in atmospheric global circulation model (AGCM) and forecasted of rainfall trend in Bangladesh (Rahman and Ferdousi, 2011). This model results are shown in figure 5 (b). AGCM results are for the late 20th century which seems having the same trend obtained in figure 5 (a). However, the Dhaka city rainfall is usually lower than the average Bangladesh trend [shown in figure 5 (c)]. As mentioned in introduction section of Bangladesh geography, the rainfall is unevenly distributed over the country. Thus, it is important to assess flash floods in Dhaka with appropriate forecast and warning systems rather than generalizing for Dhaka city from typical studies of Bangladesh. 

One of the reason of flash floods is heavy rainfall that happens when convective clouds by convective storms, thunderstorm at different seasons happened in Bangladesh and also affects Dhaka city. Thus, eventual advancement in rainfall forecasting is also a way to improve forecasting for thunderstorm as well. Eventually, the study emphasises for short term forecasting by improving Numerical Weather Prediciton (NWP) in Bangladesh in flood and cyclone impacts assesments. The following suggestions specifically for Dhaka City Flash Flood Assemsent and future areas of studies are given. 

​

•Improve drainage Systems and keep free natural lakes, canals, rivers [examples shown in fig 6 (a), (b), and (c)] 

•Maximum rainfall should be considered in building infrastructures

•Improvement of short term rainfall forecast is important 

•QPF study is suggested, rainfall forecast with NWP Model (WRF) with specific Cumulus Parameterization schemes: KF, GD, GF, BMJ 

•Improvement in warning systems and educating public to act with awareness 

​

The term project would continue researching on finding the appropriate model of forecasting in Bangladesh and obtain best choices of CPs