The thermosphere is heated by incident extreme ultraviolet (EUV) radiation from the Sun. Accurately forecasting this driver of thermospheric density is of paramount importance to spacecraft operations, such as collision avoidance, in an increasingly crowded low Earth orbit environment. This study uses deep learning techniques to forecast solar irradiance from three solar EUV/UV image channels (9.4nm, 33.5nm, and 160.0nm) taken by the Solar Dynamics Observatory up to four days in advance. The proposed model is able to forecast 23 wavelength bands from 0.05nm to 121nm (produced from the FISM2 empirical irradiance model (Chamberlin et al., 2020)) with a mean absolute percentage error of 6.0% and an average improvement in the mean absolute percentage error of 36.45% over the baseline persistence model at a 4 day time horizon. The study further validates the model and derives insights about its internal working by testing and investigating the importance of its core components.

Extreme ultraviolet images taken by the Atmospheric Imaging Assembly on board the Solar Dynamics Observatory make it possible to use deep vision techniques to forecast solar wind speed - a difficult, high-impact, and unsolved problem. At a four day time horizon, this study uses attention-based models and a set of methodological improvements to deliver an 11.1% lower RMSE error and a 17.4% higher prediction correlation compared to the previous work testing on the period from 2010 to 2018. Our analysis shows that attention-based models combined with our pipeline consistently outperform convolutional alternatives. Our model has learned relationships between coronal holes’ characteristics and the speed of their associated high speed streams, agreeing with empirical results. Our study finds a strong dependence of our best model on the position in the solar cycle, with the best performance occurring in the declining phase.