Combining ground and satellite data to forecast flood in Nepal

03 Aug 2023

Every year, the western region of Nepal faces the wrath of the Karnali River. As the rains fall and the river swells, floods wreak havoc upon villages, destroying lives, livelihoods, and property.

To help mitigate this problem, the government has made substantial investment in flood early warning systems. Early warning can reduce the risk faced by communities living downstream. The use of remote sensing and geospatial technology can help strengthen these vital systems.

The fury of the Karnali River

The mighty Karnali River originates in the Himalayas and flows 1080 km south. It is the longest river system of Nepal. The river flows through three provinces of Nepal – Karnali, Lumbini, and Sudurpaschim. At Chisapani, it splits into two tributaries – Kauriala River to the west and Girwa River to the east – between which lies fertile land. The two streams merge again across Nepal’s border to form the Ghaghara River.

Downstream of the Karnali River
Figure 1: Downstream of the Karnali River

Every monsoon season, the Karnali basin experiences heavy rainfall, followed by severe floods that cause immense suffering. Lives are lost, infrastructures, homes, and crops are destroyed, and entire villages are displaced. In 2017, floods affected more than 150,000 people (at least five people lost their lives) in Kailali and Bardiya districts.1 More recently, in October 2022, nearly 700 families were displaced by floods in Tikapur and Janaki municipalities of Kailali.2 During the same period, more than 800 families in Rajapur and Geruwa municipalities of Bardiya were affected.3

Early warning system in the Karnali

Guage-reader
Figure 2: Manual and automatic gauge readers in the Karnali River at Chisapani (Photo: Manish Shrestha)

The Department of Hydrology and Meteorology (DHM) has set up both automatic and manual rain and river gauges on the bank of the Karnali River at Chisapani. Parbati Gurung, a DHM staff, takes the readings every hour during the monsoon season and twice a day during the dry season. Parbati Didi, as the locals call her, is a familiar name in the downstream community. She is the caretaker and gauge reader of the hydro-met station. After collecting the information, she shares it with the DHM and the downstream community via a well-established communication channel. During the monsoon when the water level at Chisapani reaches a warning level of 10 m and danger level of 10.8 m4, an SMS alert is sent to the municipal offices, Armed Police Force, Nepal Red Cross, and the downstream communities. The downstream communities of Tikapur, Janaki, Rajapur, Geruwa, and Madhuwan municipalities have 3–4 hours lead time to prepare for a potential flood.

SMS-channel
Figure 3: SMS channel for early warning dissemination (Photo: Utsav Maden)

SERVIR-HKH initiatives

Flood forecasting with a few days lead time allows people to plan and prepare for flood. SERVIR-HKH, a joint initiative of NASA and USAID hosted at ICIMOD, has developed weather and flood forecasting systems. The High Impact Weather Assessment Toolkit (HIWAT) provides information on extreme precipitation and the probability of thunder and lightning with 54 hours lead time. The Streamflow Prediction Tool (SPT-Nepal), a customised GEOGloWS tool, provides flood discharge 10 days in advance. The SPT-Nepal is suitable for larger, perennial rivers such as the Karnali and its larger tributaries. However, several streams originate in the Mahabharat range (mid-hills) and the Churiya range (lower hills) and are seasonal. During monsoon, the volume of water in these streams rises significantly, causing flash floods across the country. To capture these flash flood events, SERVIR-HKH has developed a Flash Flood Forecasting Tool (FFPT). The tool uses the HIWAT’s rainfall data to generate discharge 54 hours in advance. We have also developed a Flood Inundation Mapping Tool to identify flooded areas during an event. The Flood Inundation Mapping Tool uses Earth observation data to map flooded areas and provides information on the severity of the event, thus aiding flood response.

Connecting space to village

Training on the use of SERVIR-HKH
Figure 4: Training on the use of SERVIR-HKH science applications for disaster preparedness in Nepalgunj (Photo: Utsav Maden)

We seek to build local people’s capacity to use SERVIR-HKH science applications for disaster preparedness. To that end, we have provided several trainings to our partners from Nepal Red Cross and Practical Action at the community level. After the training, some of the participants were able to use the tools to forecast flood events. Lautan Chaudhary, Program Officer for STRONG Project, says, “Since I received the training, I have been using HIWAT and flash flood prediction tools. I think the tools are very useful for predicting floods in specific river basins.”

Our Flood Inundation Mapping tool provides near real-time data for flood mapping. An important feature of the tool is that it is user friendly. According to Ramesh Gautam, Project Officer of Nepal Flood Resilience Project, “Even a layman can get a sense of the general scenario of flood impact in an area using the Flood Inundation Mapping Tool.” He further adds, “I regularly look at this site [Flood Inundation Mapping Tool] to assess the situation in our working areas, located in the lower Karnali River watershed, below the Chisapani Bridge. The tool shows the number of times these areas have been flooded during the monsoon season.”

We are working with our partners to develop a mobile app that can disseminate forecasts from our tools to the last mile. We believe that empowering communities and local governments to use satellite data can improve decision making and help save lives and properties.

References

1 Building Flood Resilience through Livelihood Diversification in Lower Karnali Basin

2 700 families displaced due to floods in Tikapur

3 Karnali flood displaces over 800 households in Bardiya

4 Red Cross, Acting in anticipation against unseasonal floods in Nepal

Authors

Manish Shrestha

Hydrologist
Utsav Maden

Utsav Maden

Knowledge Management and Communication Officer