Forecasting Solar F10.7: the state of the art and plans for far-side helioseismology integration

The NASA Living With a Star (LWS) Focused Science Topic "Beyond F10.7" is composed of a team of researchers organized around improving our current ability to predict the F10.7 index, improving our understanding of how the F10.7 index relates to solar X-ray, EUV, FUV, and UV irradiance, and in moving upper atmospheric models beyond the use of a single proxy index to represent key solar drivers of ionospheric composition and dynamics. As part of the Beyond F10.7 team, our group is focused on developing a Machine Learning (ML) model to regress solar far-side helioseismic images of sunspot active regions onto F10.7 values that can then be rotated forward in time to predict Earth-facing flux values up to 3--7 days in advance. In this presentation we first review the current F10.7 flux forecasting skill levels exhibited by the NOAA Space Weather Prediction Center to establish a baseline of predictive performance that our new model must surpass to be operationally useful. We then review the ML model architecture and preliminary performance in regressing from far-side helioseismic signals to hemispheric F10.7 flux values. We conclude with plans to extend the ML capability to include prediction of solar spectral irradiance over the EUV and FUV wavelengths of interest to upper atmospheric modelers and forecasters.