Operational training on High Impact Weather Assessment Toolkit for Bangladesh Meteorological Department forecasts

Date 29 Mar 2022 to 31 Mar 2022
Venue Bangladesh Meteorological Department, Dhaka, Bangladesh
Contact Persons servirhkh@icimod.org
Type Training

About the training

Our SERVIR Hindu Kush Himalaya (SERVIR-HKH) Initiative is collaborating with Bangladesh Meteorological Department (BMD), the NASA SERVIR Science Coordination Office (SCO), and the NASA Short-term Prediction Research and Transition Center (SPoRT) to provide operational training on the High Impact Weather Assessment Toolkit (HIWAT) for BMD professionals.

Due to COVID-19, the training will be hybrid in nature. The training will comprise three live sessions, which will begin with an online lecture followed by an interactive question and answer period focused on the topic of the day and administered by the subject matter experts. Demonstrations using data products for each live session will start with an interactive question and answer period followed by a discussion on the lecture material. The rest of the live session will be devoted to hands-on exercises analyzing data products conducted by SERVIR-HKH representatives either in-person or virtually.


This training will focus on building BMD’s capacity to independently operationalize and use the HIWAT at BMD to simulate extreme weather hazards, issue forecasts, and effectively manage the system.


Upon completion of the training, participants will have a better understanding of operating and interpreting HIWAT during the pre-monsoon and monsoon seasons, and using HIWAT to help issue extreme weather alerts and develop forecasts based on HIWAT outputs.


Professionals from BMD and ICIMOD will be participating in this training. Participants will require advanced knowledge of atmospheric modelling, high performance computing cluster system architecture, hardware, and information technology. Strong expertise in command line and beginner to intermediate-level skills in Python programming language are required for the hands-on exercises.




Extreme weather events claim many human lives and cause economic losses every year in the Hindu Kush Himalaya (HKH) region. The HKH region is a hotspot for some of the most intense thunderstorm activities taking place globally. However, until recently countries such as Bangladesh have lacked locally and regionally relevant observations and numerical modelling capabilities to monitor and forecast extreme weather events such as severe thunderstorms.

HIWAT uses a mesoscale numerical weather prediction model and the Global Precipitation Measurement (GPM) constellation of satellites to assess high-impact convective weather events over the HKH region. The toolkit includes a suite of ensemble model forecasts to constrain the uncertainties and provides a 54-hour probabilistic forecast for improved decision-making. HIWAT provides outlooks for lightning strikes, high impact winds, high rainfall rates, hail, and other weather events.

With dedicated involvement and efforts from ICIMOD’s SERVIR-HKH Initiative, NASA SERVIR SCO, principal investigators and co-investigators of the NASA Applied Sciences Program, and USAID, HIWAT has been developed and established as an operational system to provide high-resolution ensemble forecast guidance of severe thunderstorm hazards for BMD.


Tentative agenda

All timestamps are in Bangladesh Standard Time (BST )

  • Day 1 – Tuesday, 29 March 2022
    Time Topic
    08:00–11:00 Background on convection-allowing ensemble modelling, and description of HIWAT model configuration and products
      Training overview and introduction of the participants – Abdul Mannan, BMD

    Lectures – Jonathan Case and Jayanthi Srikishen

    A high-level overview of convection-allowing ensemble modelling

    Goals of addressing severe weather hazard uncertainties with HIWAT

    HIWAT configuration and product descriptions

    Interactive – All

    Exercises: Examination of case studies and interpretation of various HIWAT products and fields

    BMD solidify use cases of interest for Day 2-3 testing

    Q & A

    HIWAT Model Viewer for 2018–2021: https://hmv.servirglobal.net/

  • Day 2 – Wednesday, 30 March 2022
    Time Topic
    8:00–10:00 Core modelling system for HIWAT: Unified Environmental Modeling System (UEMS) running the Weather Research and Forecasting (WRF) Model
      Lectures – Jonathan Case and Jayanthi Srikishen

    Q & A session from Day 1

    Orientation to the UEMS modelling framework running the WRF model

    Configuration and management of model runs with the UEMS/WRF framework

    Start to finish: Setting up modelling domains for HIWAT using the UEMS

    Interactive – All

    Exercise: Set up a new modelling domain in the UEMS.

    Establish run-time parameters

    Configure physics options

    Set up post-processing details

    Q & A

  • Day 3 – Thursday, 31 March 2022
    Time Topic
    8:00-10:30 HIWAT product generation: Understanding settings, wrapper scripts, and Python packages for generating products and probability maps
      Lectures – Jonathan Case and Jayanthi Srikishen

    Q & A session from Day 2

    Input files that can be edited by the user

    settings_hiwat.txt (model settings and deterministic products)

    settings.py (Python ensemble package settings)

    Start with cron entries: wrapper scripts

    Deterministic graphics with Grid Analysis and Display System (GrADS) templates and plotting script

    Ensemble product generation with Python package

    Interactive – All

    Exercise: Edit the settings files to change ensemble product generation

    Q & A

    10:30–11:00 Closing session