HYDRO-CLIMATE INFORMATION SERVICES IN GHANA FARMER SUPPORT APPLICATION IMPLEMENTATION AND EVALUATION

Abstract

Agriculture in Africa depends more on rainfall than irrigation. In ancient days, older people in farming communities relied on their perceptions, instincts and observations of the surrounding biota (flora and fauna) to plan their agricultural activities. These ideas, experiences and knowledge of local people had gained attention in the discussions of climate change and adaptation strategies in Africa essentially relating to agricultural sustenance. Evidence from literature shows that in recent times forecast which is based on modern technology (scientific forecast) had received a lot of attention and has been adopted by many rural farmers in the developing world due to its precision. Several researchers had inferred that the way forward to adaptive and reliable weather forecasting is by integrating indigenous forecast with scientific forecast. This study was carried out in the Yapalsi and Nakpanzoo rice valleys to evaluate the benefits and drawbacks of the Farmer Support Application (FSApp) which is a tool developed through a participatory approach to address the climate information needs of farmers with regards to rainfed agriculture. This FSApp works by receiving scientific forecast from meteoblue and local forecast from the farmers and displaying both scientific and local forecast information to farmers for making informed decisions pertaining to agriculture. To achieve the aims of the research, standard rain gauges were installed in both communities to record rainfall data. The rice farmers in both valleys were then trained to use the FSApp for making daily decisions which involved the daily entry of forecast by farmers using indigenous ecological indicators. All the data entered into the FSApp were recorded as local forecast data and stored on servers. At the end of the season, the local forecast data was retrieved from the server, scientific forecast data was obtained from meteoblue and the predictive accuracy of the FSApp was assessed using the rain gauge data as the reference. The final evaluation was also carried out using focus group discussions and questionnaires to assess the outcome of the project on the farmer's agricultural livelihood. Results indicated that an integration of Scientific Forecast Knowledge (SFK) and Local Forecast Knowledge (LFK) amounted to the highest skill score of 0.62 followed by SFK with a score of 0.61 and LFK with a score of 0.50. These skill score values show a significant predictive skill of the various forecast scales (SFK, LFK and Integration of SFK and LFK). The best accuracy of predictions was observed when LFK was integrated with SFK relative to the sole skills of SFK and LFK even though a significant difference was not observed between the scores of SFK and the integrated score. It was also revealed that farmers relied more on celestial bodies (Sun and Moon) for local weather prediction such that, 85.9% of indicators used by farmers for predictions were observations of celestial bodies. Results revealed that co- production played a vital role in the adaption of the farmer support app. Farmers demonstrated a significant (high) level of knowledge about weather phenomena and knowledge sharing among farmers was observed to have increased. The farmer support app performed as expected and therefore recommended to other farming communities that rely on rainfed agriculture.