Integrating user- and data-driven weather forecasts to develop legitimate, credible, and salient information services...

Climate-related risks and variability pose significant challenges to the livelihoods and food security of smallholder farmers practicing rainfed agriculture. Many smallholders have limited access to weather information from climate services, and this information is often not tailored to their specific context and needs. Therefore, they rely on local ecological knowledge. This study utilizes the second generation of climate services, which provide demand-driven forecast information systems through mobile apps. We present three cases from agricultural communities in Guatemala, Bangladesh, and Ghana where we collaborated with farmers to develop local weather forecasts (LF) and combined them with scientific weather forecasts (SF) to create hybrid weather forecasts (HF). The integration of user-driven forecasts (LF) and data-driven forecasts (SF) enhances the legitimacy of the service, thereby increasing farmers' trust and credibility by providing skilful forecasts. Furthermore, our results demonstrate that the hybrid weather forecast approach facilitates climate-smart, adaptive agricultural decision-making, enhancing the resilience and capacity of smallholder farmers in the Global South to adapt to a changing climate.