A pan-African medium-range ensemble flood forecast system

摘要

The African Flood Forecasting System (AFFS) is a probabilistic floodforecast system for medium- to large-scale African river basins, with leadtimes of up to 15 days. The key components are the hydrological modelLISFLOOD, the African GIS database, the meteorological ensemble predictionsby the ECMWF (European Centre for Medium-Ranged Weather Forecasts) andcritical hydrological thresholds. In this paper, the predictive capability isinvestigated in a hindcast mode, by reproducing hydrological predictions forthe year 2003 when important floods were observed. Results were verified byground measurements of 36 sub-catchments as well as by reports of variousflood archives. Results showed that AFFS detected around 70 % of thereported flood events correctly. In particular, the system showed goodperformance in predicting riverine flood events of long duration(> 1 week) and large affected areas (> 10 000 km2) well inadvance, whereas AFFS showed limitations for small-scale and short durationflood events. The case study for the flood event in March 2003 in the SabiBasin (Zimbabwe) illustrated the good performance of AFFS in forecastingtiming and severity of the floods, gave an example of the clear and conciseoutput products, and showed that the system is capable of producing floodwarnings even in ungauged river basins. Hence, from a technical perspective,AFFS shows a large potential as an operational pan-African flood forecastingsystem, although issues related to the practical implication will still needto be investigated.