Towards weather and climate services that integrate indigenous and scientific forecasts to improve forecast reliability and acceptability in Ghana

The livelihood of many farmers across the globe is affected by climate variability and change. Providing weather and seasonal climate information is expected to support farmers to make adaptive farming decisions. Yet, for many farmers, scientific forecast information provided remains unreliable for decision-making. Scholars have called for the need to integrate indigenous and scientific forecasts to improve forecast information at the local level. In Northern Ghana, scientific forecast information from meteorological agency is unacceptable to farmers, making them rely on indigenous forecasts for adaptive decisions. This study proposed an integrated probability forecasting (IPF) method that integrates indigenous and scientific forecasts into a single forecast. As a proof of concept, we tested the reliability of IPF using binary forecast verification method and evaluated its acceptability to farmers through internally consistent multiple response questions. Results of the reliability test show that IPF performed on average better than indigenous and scientific forecasts at a daily timescale. At the seasonal timescale, IPF and indigenous forecast performed better than Scientific forecast, although in terms of probability IF showed better results overall. Majority of the farmers (93%) prefer the IPF method as this provides a reliable forecast, requires less time, and at the same time resolves the contradictions arising from forecast information from different sources. The results also show that farmers already integrate (complementary) scientific and indigenous forecasts to make farming decisions. However, their complementary approach does not resolve the issue of contradictory forecast information. From our proof of concept, we conclude that integrating indigenous and scientific forecasts can potentially increase forecast reliability and uptake.

Automated summary

Farmers' livelihoods are affected by climate variability and change, but scientific forecast information is often unreliable for decision-making. Integrated indigenous and scientific forecasts can increase reliability and acceptance. A study proposes an integrated probability forecasting method and tests its reliability and acceptability through verification and farmer surveys.