Validation for solar wind prediction at Earth: Comparison of coronal and heliospheric models installed at the CCMC

Multiple coronal and heliospheric models have been recently upgraded at the Community Coordinated Modeling Center (CCMC), including the Wang-Sheeley-Arge (WSA)-Enlil model, MHD-Around-a-Sphere (MAS)-Enlil model, Space Weather Modeling Framework (SWMF), and heliospheric tomography using interplanetary scintillation data. To investigate the effects of photospheric magnetograms from different sources, different coronal models, and different model versions on the model performance, we run these models in 10 combinations. Choosing seven Carrington rotations in 2007 as the time window, we compare the modeling results with the Operating Mission as Nodes on the Internet data for near-Earth space environment during the late declining phase of solar cycle 23. Visual comparison is proved to be a necessary addition to the quantitative assessment of the models' capabilities in reproducing the time series and statistics of solar wind parameters. The MAS-Enlil model captures the time patterns of solar wind parameters better, while the WSA-Enlil model matches with the time series of normalized solar wind parameters better. Models generally overestimate slow wind temperature and underestimate fast wind temperature and magnetic field. Using improved algorithms, we have identified magnetic field sector boundaries (SBs) and slow-to-fast stream interaction regions (SIRs) as focused structures. The success rate of capturing them and the time offset vary largely with models. For this quiet period, the new version of MAS-Enlil model works best for SBs, while heliospheric tomography works best for SIRs. The new version of SWMF with more physics added needs more development. General strengths and weaknesses for each model are diagnosed to provide an unbiased reference to model developers and users.