Experiences from co-developing hydrological forecasting systems across West Africa and eastern Siberia

<p>Changes in hydrology and society are growing concerns in West Africa and eastern Siberia. Floods cause loss of lives, damaged infrastructure, and reduced food security. A warming climate impacts river ice conditions, affecting ice roads and summertime river transportation. Operational hydrological forecasting systems are needed to boost climate adaptation and resilience in these regions.</p><p>Within the projects FANFAR (www.fanfar.eu) and HYPE-ERAS (www.hype-eras.org), we work with key stakeholders (e.g. hydrological services and disaster managers) to jointly design, develop, operate and evaluate hydrological forecasting systems in each domain.</p><p>In West Africa, we used Multi-Criteria Decision Analysis (MCDA) to engage users in the definition and development of the forecasting system. MCDA was employed to clarify and prioritize objectives and system configurations. We found that the most highly prioritized objectives were: high accuracy, clear flood risk information, reliable access, and timely production and distribution of the information. The stakeholders also contributed to forecast evaluation, e.g. by comparing forecasts with field experiences and local observations. Skills vary, with higher probability of detection (POD) in unregulated areas, and lower POD downstream of wetlands and reservoirs. Overall, 75% of the West African agencies considered the forecast accuracy to be &#8220;good&#8221; or &#8220;very good&#8221;. This also translated into changed operational warning practices, e.g. in Nigeria where FANFAR information was used to save thousands of lives and property on at least two occasions in 2020.</p><p>In eastern Siberia, a key target variable to forecast is river ice breakup date. To this end we combined meteorological data (HydroGFD and ECMWF 1-10day forecasts), hydrological modelling (Artic-HYPE), and local observations of river discharge, water level, ice thickness and river ice breakup dates. In both 2020 and 2021, the model successfully predicted the river ice breakup in the Lena River near Yakutsk up to 10 days in advance. We also found a systematic trend of too early breakup in upstream locations, and too late breakup in downstream locations, indicating problems in ice melt and/or growth calibration.</p><p>Here we summarize the research design, results and experiences gained in FANFAR and HYPE-ERAS; indicating that operational forecasting can be useful to boost climate adaptation and resilience.</p>