Predicting geo-effectiveness two days prior to CME impact with EUHFORIA

The EUropean Heliospheric FORecasting Information Asset (EUHFORIA, Pomoell and Poedts, 2018), a physics-based and data-driven heliospheric and CME propagation model can predict the solar wind plasma and magnetic field conditions at Earth. It contains several flux-rope CME models, such as the simple spheromak models and the more advanced FRi3D and toroidal CME models. This enables the prediction of the sign and strength of the magnetic field components upon the arrival of the CME at Earth and, thus, the geo-effectiveness of the CME impact. EUHFORIA has been coupled to several global magnetosphere models like OpenGGCM, GUMICS-4, and Gorgon-Space. In addition, the synthetic data at L1 (from the EUHFORIA simulation) can be used as input for empirical models and neural networks to predict the geomagnetic indices like Disturbance-storm-time (Dst) or Kp that quantify the impact of the magnetized plasma encounters on Earth’s magnetosphere. Hence, we also coupled EUHFORIA to empirical models (Obrien and McPherron, 2000b, and Newell et al, 2006) and machine learning (NARMAX, and the models from Wintoft et al. (2017 and 2021)) based models to predict the geomagnetic indices. We then compare the results of these models to observational data to evaluate their performance in predicting the geo-effect indices. To quantify these comparisons, we use the advanced dynamic time warping method. Since we use synthetic data from the EUHFORIA simulations, we can obtain the input parameters for running the geomagnetic indices models two to three days in advance, unlike the 60-90 minutes lead time of the real-time measurements. 

We perform ensemble modelling considering the L1 monitor precision in its orbit as well as the uncertainty in the initial CME parameters (longitude and latitude) at launch, for error quantification. This is done by evaluating the geomagnetic index models using synthetic data from the virtual satellites around L1 in EUHFORIA’s simulation domain. In addition, we also investigate the impact of the spatio-temporal resolution of EUHFORIA output in forecasting the geomagnetic indices, exploiting the adaptive mesh refinement feature in ICARUS (Baratashvili et al., 2022). Overall, this study validates various space weather forecasting model chains and checks the best compatibility and predictive capabilities using EUHFORIA data for operational space weather forecasting.