|Title||Seasonal variability and predictability of agro-meteorological indices: Tailoring onset of rainy season estimation to meet farmers’ needs in Ghana|
|Author(s)||Gbangou, Talardia; Ludwig, Fulco; Slobbe, Erik van; Hoang, Long; Gordana Kranjac-Berisavljevic, Kranjac-Berisavljevic|
|Source||Climate Services 14 (2019). - ISSN 2405-8807 - p. 19 - 30.|
Water Systems and Global Change
|Publication type||Refereed Article in a scientific journal|
|Keyword(s)||Forecast categories - Inter-annual variability - Onset dates - Seasonal forecasts|
Reliable information on onset of the rainy season is important for local agriculture planning in Ghana. We examine the (i) trend and variability of onset in local observations to better understand the need for onset forecast information and (ii) performance of ECMWF System 4 seasonal climate forecast in reproducing this variability and discriminating tercile categories of onset dates across Ghana. The analyses focused on two pilots locations of interest among the fourteen synoptic stations studied, namely Ada and Tamale located in the coastal savanna and in northern Ghana. Two different onset date definitions were tested to suite with uncorrected and bias-corrected forecasts in order to test the predictability. The definitions were tailored to suit with forecast start dates, local climate data availability and cropping calendar. Results show a significant decreasing trend in historical onset dates towards more recent times (i.e 1986–2010) at Tamale station. Also, historical onset dates exhibit a significant increasing variability towards more recent time at Ada station. System 4 shows some ability for reproducing local onset variability with significant correlational relationship between forecasted and observed onset dates at some locations including Ada station. The forecasting system also has significant skill in predicting early and late onset dates categories (i.e H-K score > 0) at the pilot stations. In conclusion, the use of onset agro-meteorological index, based on System 4 as climate service in Ghana, has a potential value for decision making when considering categorical based forecasts.