SAGE builds upon the legacy of the Sub-seasonal to Seasonal (S2S) Prediction Project (2013–2023), which significantly advanced our understanding of forecast predictability, skill, and uncertainty on sub-seasonal timescales (2 weeks to 2 months ahead). While the S2S project made great strides in the physical science of predictability, challenges remain in effectively applying these forecasts to real-world decision-making.

Key challenges include:

•     Limited integration of S2S forecasts into operational decision-making across sectors such as agriculture, health, disaster risk reduction, and energy.

•     Difficulties in communicating forecast uncertainty,  particularly in situations where forecast skill varies based on climate regimes and atmospheric conditions.

•     A need for co-designed forecast products, where end-users are actively involved in developing tools tailored to their specific needs.

•     Advancements in Machine Learning (ML) and Artificial Intelligence (AI) require careful evaluation to ensure they enhance—not just automate—forecast skill and usability.

Recognizing these challenges, SAGE was created to bridge the gap between research and application. The project is designed to enhance both the scientific foundations and practical use of S2S forecasts, ensuring they are reliable, accessible, and actionable for decision-makers worldwide. Through interdisciplinary collaboration, SAGE aims to transform scientific progress into tangible benefits for sectors that rely on climate-sensitive planning.