Solar observation-based model for a multiday predictions of interplanetary shock and CME arrivals at earth

We present the design, implementation, present status, and preliminary use of an operational forecaster-friendly solar wind prediction system. We have tested this architecture in our real-time fearless forecast 'prediction study of interplanetary shock arrival time (SAT) at Earth. This study has progressed on a continuous basis since 1997, near the rise of Solar Cycle 23. Comparisons of predicted SAT with observed SAT provide a measure of model forecast skill. We present our kinematic model's forecast skill statistics as compared with several other SAT models and propose important reference metrics for evaluating this and any other interplanetary modeling system. Our prediction system is presently being extended, via a hybrid approach, to include a three-dimensional magnetohydrodynamic (MHD) model. This procedure will accommodate proxy inputs for simulating interplanetary disturbances that are associated with various forms of solar activity (flares, disappearing filaments, coronal mass ejection imagery, shocks). The objective is to provide a capability in the near term to predict, shortly after the launch of large-scale structures at the Sun, the time-dependent interplanetary magnetic field vector (including the north-south component, Bz) at Earth. This is a key requirement for predicting geomagnetic storms.