Authors: Samuel J. Schonfeld (Air Force Research Laboratory), Lily A. Wood (University of New Mexico), Carl J. Henney (Air Force Research Laboratory)

The Solar Indices Forecasting Tool (SIFT) uses ADAPT (Air Force Data Assimilative Photospheric flux Transport) global magnetic maps to drive nowcasts and forecasts of the Sunspot Number (SSN), 10.7 cm radio flux density (F10.7), and Magnesium II core-to-wing ratio (Mg II). SIFT runs in near-real-time at the National Solar Observatory (NSO) providing automated daily 1-7 day forecasts of solar activity. Input to the ADAPT model includes photospheric magnetograms from the NISP (NSO Integrated Synoptic Program) ground-based instruments, GONG and VSM. We present updates to the SIFT model that improve its performance and expand its capabilities to forecast the Mg II index and seven EUV/FUV irradiance measurements available operationally from the EXIS/EUVS instrument on GOES-R series (16, 17, 18, and 19) satellites. These updated SIFT models implement a fully interpretable Explainable Boosting Machine (EBM) regression algorithm with non-linear model response which yields improved performance over the old multiple linear regression models, particularly for multi-day forecasts. These additions upgrade SIFT from an “Indices” to an “Irradiance” forecasting tool in preparation for an anticipated operational implementation at NOAA/SWPC.