Chou, Min‐Yang, Yue, Jia, Wang, Jack, Huba, J. D., El Alaoui, Mostafa, Kuznetsova, Maria M., Rastätter, Lutz, Shim, Ja Soon, Fang, Tzu‐Wei, Meng, Xing, Fuller‐Rowell, Dominic, and Retterer, John M.
This paper presents the validation of modeled total electron content (TEC) from 14 ionospheric models, including empirical, physics‐based, and data assimilation (DA) models, hosted by the NASA/NSF Community Coordinated Modeling Center (CCMC), NOAA Space Weather Prediction Center, and NASA Jet Propulsion Laboratory (JPL). This study aims to assess the current progress and capability of the CCMC‐hosted ionospheric models in capturing the storm time ionosphere during the low and moderate solar flux years. We focus on the low‐latitude ionosphere (i.e., ±40° in magnetic latitude) and compare the modeled TEC with the Madrigal TEC during the 2013 March and 2021 November storms. Multiple metrics are used to quantitatively assess the models' accuracy, precision, association, bias, and capability in capturing the TEC changes in response to the storms. The skill score based on the metric scores is further proposed to evaluate the overall performance of ionospheric models against the reference model (International Reference Ionosphere 2016; IRI‐2016). The results indicate that the DA model GLObal Total Electron Content and JPL Global Ionospheric Map models show good performance in modeling the TEC and reasonably reflect the storm time TEC changes spatially and temporally. The empirical models IRI‐2016 and 2020 show relatively good performance compared with the physics‐based models regarding the model‐data comparison; however, it is difficult to characterize the TEC changes caused by storms. The physics‐based models can simulate the storm effect in spatial and temporal TEC variations better than the empirical model. The performance of ionospheric models in capturing the storm time TEC anomaly is presented and discussed. Plain Language Summary: The Earth's ionosphere is highly variable due to atmospheric variations and solar activity. Forecasting the ionosphere is of particular importance because the ionosphere significantly impacts our daily lives. Scientists have developed multiple numerical models, such as empirical, physics‐based, data‐assimilation, and machine learning models, to improve our capability in capturing space weather and understanding the underlying physics responsible for ionospheric variability. NASA Goddard Space Flight Center Community Coordinated Modeling Center (CCMC) hosts an extensive suite of state‐of‐the‐art ionospheric models, providing the community access to these modern models to support space weather and space physics research. CCMC also acts as an unbiased evaluator to validate models for eventual use in space weather forecasting. This paper aims to validate the performance of recently onboarded CCMC, National Oceanic and Atmospheric Administration Space Weather Prediction Center, and NASA Jet Propulsion Laboratory ionospheric models during the geomagnetic storms, providing a baseline for users and modelers to understand the recent progress of ionosphere models. Key Points: Validation of ionospheric total electron content (TEC) by the state‐of‐the‐art ionospheric models hosted by NASA Community Coordinated Modeling Center, National Oceanic and Atmospheric Administration Space Weather Prediction Center, and NASA Jet Propulsion Laboratory (JPL)Multiple metrics and skill scores are used to assess the performance of ionospheric models in capturing storm time TEC anomalyGLObal Total Electron Content and JPL Global Ionospheric Map perform best, and physics‐based models perform better than the empirical model in capturing storm TEC variations [ABSTRACT FROM AUTHOR]