The Hindu Kush Himalayan region is extremely vulnerable to various types of water-induced disasters, particularly floods, drought and landslides. Due to its fragile geology, rugged terrain, and South Asian monsoon patterns, Nepal is prone to floods, landslides, and glacial lake outburst floods (GLOFs). Around 80% of the annual precipitation falls during the monsoon season, between June and September. While extreme rainfall events due to climate change are on the rise, flood risk communication, such as early warning systems, need strengthening.
Natural hazards often result in disasters due to inefficient and ineffective risk communication. In Nepal, the investment required to gauge numerous flood-prone medium and small rivers, and the short lead time provided by instrumental flood early warning systems (EWS) are two major challenges.
Existing technical flood EWS provides hazard based warning and are susceptible to failure due to inadequate readiness level. The warnings are also limited to major rivers and many flood prone rivers/ rivulets are without flood warning provision.
The service intends to use downscaled global flood discharge forecasts with routing models to forecast flood water levels in medium sized watersheds of Nepal, with 10–15 days lead time. The goal is to enhance response interventions through effective planning and preparedness, and reduce loss of property and life.
To improve risk communication in Nepal, a flood early warning system is being piloted in five watersheds across Nepal— Bagmati, Kandra, Kamala, Kankai, and Macheli — in partnership with Mercy Corps Nepal and the Department of Hydrology and Meteorology (DHM), Government of Nepal.
ICIMOD, Brigham Young University, and NASA’s Jet Propulsion Laboratory are working together to develop a system that provides flood warning information for rivers in the pilot watersheds by using information generated by the Global Flood Awareness System (GloFAS), a joint initiative of the European Commission and the European Centre for Medium-Range Weather Forecasts (ECMWF).
The GloFAS product is downscaled in order to make it relevant for the pilot watersheds in Nepal before routing, using a Routing Application for Parallel computation of Discharge (RAPID) model. The system incorporates digital elevation models (DEM) and exposure (gridded population and infrastructure) data. Historical river hydrology data, flood inundation scenarios, and impact information are used to validate and calibrate the model. This system is able to forecast flood water levels at any stretch of a given river for the next 10–15 days.