On-the-job training: Development of localised flash flood prediction tool for upper Meghna River Basin, Bangladesh

Date 17 Jun 2024 to 21 Jun 2024
Venue ICIMOD, Kathmandu
Contact Persons Manish Shrestha
Type Training
Programmes SG 2, AAE, SG3


About the event

The Flood Forecasting and Warning Center (FFWC) and the International Centre for Integrated Mountain Development’s (ICIMOD) SERVIR-HKH Initiative have developed a hydrological model to utilise the High-Impact Weather Assessment Toolkit (HIWAT) for flash flood forecasting. The model will be able to capture localised weather phenomena and generate flood information 54 hours in advance for the upper Meghna River Basin in Bangladesh.

One key aspect of SERVIR-HKH is the capacity building of its partners in the use of the developed model. As part of this activity, we are organising an on-the-job training for FFWC officers.


The training aims to provide participants better understanding of:

  • The functionality of the J2000 hydrological model
  • Use of HIWAT rainfall to generate flood forecast
  • Get feedback to be incorporated into the model

Expected outcomes

Upon completion of the training, the participants will be able to run the model in their institute and validate the model forecast. We also expect the participants to be able to transfer the acquired knowledge to other staff in the institute. They will also gain knowledge about various tools developed under the SERVIR-HKH initiative.

Expected participants

Nominated officials from FFWC are expected to join this training. For the training, each participant should have a laptop with a Windows 10 or higher operating system. They should have a basic understanding of GIS, QGIS installed, hydrology, forecasting and data analyses. A good understanding of English is also required as all resource materials will be in the English medium.

Resource persons

Birendra Bajracharya, Chief of Party, SERVIR-HKH

Kiran Shakya, Geospatial Application Development Specialist

Manish Shrestha, Hydrologist

Prabhat Shrestha, Application Developer


Floods cause loss of life and massive damage to infrastructure every year, especially in the northeast part of Bangladesh. They also have a significant impact on the country’s economy due to agricultural damage. In 2019, flash floods in the districts of Sylhet and Sunamganj affected over 50,000 people and caused massive damage to crops. Similarly, heavy monsoon rains in 2020 affected 21 districts in the north and northeast parts of the country, impacting over 700,000 families and causing damage to infrastructure and agricultural products). In 2021, Sylhet, Sunamganj, and Habiganj districts were inundated, affecting over 200,000 people.

Given the significant impact of floods, a Flood Early Warning System (FEWS) is crucial as it provides notice of impending floods, giving people precious time to evacuate and take action to save their valuables. The Flood Forecasting and Warning Center (FFWC) under the Bangladesh Water Development Board (BWDB) is the main agency responsible for flood management in Bangladesh. Partnering with the FFWC, we have utilised HIWAT precipitation data to develop a hydrological model for flash flood forecasting in the upper Meghna River Basin. We have employed an open-source J2000 hydrological model, which is expected to provide better flash flood forecasting due to increased resolution compared to the current system at the FFWC.