FOOD SECURITY IN WEST AFRICA
Climate impact indicators
Global essential climate variables/indicators used
The seasonal forecast ensemble (starting with ECMWF system 5) will be used. The climate variables are daily precipitation, mean and extreme temperatures (minimum and maximum) at the seasonal scale. The West African national rainfall gauge data merged with satellite data (TAMSAT and CHIRPS), the re-analysis climate data generated (HydroGFD2.0) by SMHI will also be used.
Regional/local indicators used and produced
The current seasonal forecasts are delivered by the PRESAO process (Seasonal forecasts in West Africa). Each year, a consensual seasonal forecast on rainfall and river discharge for the West-African region is delivered.
Global data to regional/local scale
Step 1: The data collection will include seasonal forecasts (ECMWF first then others as available), West African reference data (from AGRHYMET), HydroGFD2.0 reference dataset and hydrological discharge data. These data will be quality checked; reference datasets (check key characteristics), seasonal forecast models (only if we get several models), characteristic: accumulated P for entire season, sums for individual months/weeks/days, maxima for entire season. Bias adjustment & preparation of all datasets to be used in the forecasting: one bias-adjustment method (quantile mapping) and one reference dataset (West Africa reference data or HydroGFD2.0) and/or Intercomparison: 2 bias-adjustment methods (+DBS from SMHI) and/or two reference datasets (both HydroGFD2.0 and West Africa reference data).
Step 2: The HYPE model will be recalibrated using (a) climatological reference data (b) the reference data for bias-adjusting the seasonal forecasts (if different).
Step 3: The hydrological seasonal forecasts will be produced using several approaches: Meteorological approach: modelled seasonal forecasts (e.g. ECMWF system 5). (a) raw seasonal forecasts, (b) bias-adjusted seasonal forecasts; Climatological approach: (a) standard all years or (b) selecting analogous years/months.
Step 4: Several channels will be used to disseminate the seasonal hydrological forecasts produced. It is initially a forum bringing together all stakeholders, community radio stations, newsletters and special notes.
Step 5: The assessment of the predictive capacity will include dry spells, accumulated volumes, extreme wet and dry events (e.g. maximum daily, 3-day and weekly rainfall and discharge during the season), assess predictive capacity at local scale and compare the different seasonal forecasting approaches.
Step 6: The ensembles will be revisited by putting less weight on low-performing ensemble members.