Climate-informed timely planting in rainfed systems for supporting sustainable food security in sub-Saharan Africa

Leveraging rainfed agriculture is key for achieving global food security while promoting sustainable agricultural production. Less energy, carbon and water-intensive than its irrigated counterpart, rainfed systems are nevertheless highly vulnerable to climate variability and change. Prior information on rainy season onset is thus vital for timely planting in sub-Saharan Africa (SSA), a practical and low-cost pathway to strengthen food security under a changing climate. Here, we aim to strengthen the role of climate services on farmers' resource allocation optimization and climate-related risk mitigation by identifying a suitable start for the rainfed maize growing season in SSA with a lead-time of several months. Our results show that the Maximum Wavelet Power Date (MWPD) constitutes an objective, predictable and consistent indicator across scales and precipitation datasets, maximizing the probability of meeting local agro-climatic onset criteria. Prediction performance four months in advance was within a ± 14-day window in 64–69% of the testing cases, although it decreased for early-onset seasons. Hence, expert knowledge and proper uncertainty communication must support the operational use of the MWPD. From local to regional scales, from traditional to scientific knowledge, our approach aims to contribute to United Nations Sustainable Development Goal 2 through the provision of easily interpretable and robust information for climate-resilient rainfed systems. Practical implications The Maximum Wavelet Power Date (MWPD) shows strong potential for integration into climate services, supporting agricultural decision-making in rainfed systems in sub-Saharan Africa (SSA). Its ability to provide a consistent and objective indicator of agro-climatic rainy season onset across datasets and spatial scales addresses a key challenge in translating climate information into actionable advice. The demonstrated predictive skill several months in advance is particularly relevant for operational use. Differences between observed and predicted MWPD four months ahead were below 14 days in most cases, suggesting that MWPD-based products could inform early recommendations for land preparation, input planning, and planting decisions. The regionalization of relevant agro-climatic conditions further supports service provision, by enabling meteorological services and regional agencies to deliver tailored advice rather than general recommendations. In addition, the approach explicitly addresses data uncertainty by demonstrating cross-dataset robustness, which is critical for building user trust in data-sparse regions. As such, MWPD can act as an intermediate product within climate-service delivery chains, linking seasonal data to location-specific agricultural guidance. However, in the SSA context, the operational adoption of this approach will depend on overcoming significant challenges, including sparse monitoring networks, limited institutional capacity, difficulties in communicating uncertainties, and the need to adapt recommendations to local agronomic practices. Minimum implementation requirements include access to quality-controlled rainfall datasets or seasonal forecast products, basic capacity to run or interpret MWPD predictions, institutional linkage with a National Meteorological Service or Regional Climate Centre, and an extension or advisory channel capable of translating probabilistic information into locally relevant recommendations. Fortunately, the continent has a strong foundation of local and regional expertise playing a key role in supporting decision-making on agricultural and climate services, with organizations such as the Centre Climatique Régional pour l'Agriculture, Hydrologie et la Météorologie (AGRHYMET), the Climate Prediction and Applications Centre (ICPAC) of the Intergovernmental Authority on development (IGAD) and the Southern African Development Community - Climate Service (SADC-CS). The predictive performance of the model varies across onset timing scenarios, underscoring the importance of accompanying MWPD-based information with contextual interpretation from these local and regional experts. Their involvement is thus crucial in providing the necessary understanding to inform decision-making. The flexibility of the approach allows for adaptation to different crops and local thresholds, facilitating co-production with stakeholders and the integration of local knowledge, which is critical for effective decision-making in African agricultural services.