The primary goal of the SFS Project is to deliver a new, world-class operational forecast system that is more accurate, reliable, and accessible than the decades-old CFSv2. This new system, SFSv1, will provide enhanced predictions for temperature, precipitation, drought, fire, heat waves, hurricanes, and sea ice extent up to 12 months in advance, supporting decision-makers across the entire Weather Enterprise. To achieve this, the project is focused on three key outcomes:

1. Improved Forecast Skill & Reliability

(Delivering a more accurate forecast by reducing known model errors)

• Improve the model's core physics and dynamics to reduce systematic biases.
• Enhance the land-surface, ocean, and sea ice components to better represent their interactions with the atmosphere, minimize model drift, and ensure energy conservation.
• Develop an advanced ensemble framework that more realistically represents the full range of uncertainty in a seasonal forecast.

2. A Modern, Accessible & Community-Driven System

(Building the SFS on the Unified Forecast System (UFS) framework to make it accessible to everyone)

• Develop a flexible, efficient, and cloud-ready workflow to support rapid testing, reanalysis, and real-time operations.
• Release the full SFS application to the public and research community through the Earth Prediction Innovation Center (EPIC).
• Develop an integrated, METplus-based verification package to provide a standardized tool for evaluating SFS prototypes and operational forecasts.

3. Actionable Data for Stakeholders

(Ensuring the model's output is usable and valuable)

• Produce and release a comprehensive, multi-decade reanalysis and reforecast dataset (1981-present) that is essential for stakeholder calibration and product development.
• Develop a consistent initialization strategy for both reforecasts and real-time forecasts.
• Work directly with stakeholders to ensure SFS products and post-processing meet their specific needs.