An Adaptive Prediction System for Specifying Solar Wind Conditions Near the Sun

The ambient solar wind flows and fields influence the complex propagation dynamics of coronal mass ejections in the interplanetary medium and play an essential role in shaping Earth's space weather environment. A critical scientific goal in the space weather research and prediction community is to develop, implement and optimize numerical models for specifying the large-scale properties of solar wind conditions at the inner boundary of the heliospheric model domain. Here we present an adaptive prediction system that fuses information from in situ measurements of the solar wind into numerical models to better match the global solar wind model solutions near the Sun with prevailing physical conditions in the vicinity of Earth. In this way, we attempt to advance the predictive capabilities of well-established solar wind models such as the Wang-Sheeley-Arge model. We perform a statistical analysis of the resulting solar wind predictions for the years 2006 to 2015. The proposed prediction scheme improves all the coronal/heliospheric model combinations investigated by approximately 15-20 percent in terms of various comprehensive prediction validation measures. We discuss why this is the case, and conclude that our findings have important implications for future practice in applied space weather research and prediction.