Evaluation of total electron content prediction using three ionosphere-thermosphere models

Prediction of ionospheric state is a critical space weather problem. We expand on our18 previous research of medium-range ionospheric forecasts and present new results on eval19uating prediction capabilities of three physics-based ionosphere-thermosphere models (TIE20GCM, CTIPe and GITM). The focus of our study is understanding how current mod21eling approaches may predict the global ionosphere for geomagnetic storms (as studied22 through 35 storms during 2000-2016). Prediction approach uses physics-based model23ing without any manual model adjustment, quality control or selection of the results. Our24 goal is to understand to what extent current physics-based modeling can be used in to25tal electron content (TEC) prediction and explore uncertainties of these prediction ef26forts with multi-day lead times. The ionosphere-thermosphere model runs are driven by27 actual interplanetary conditions, whether those data come from real-time measurements28 or predicted values themselves. These model runs were performed by the Community29 Coordinated Modeling Center (CCMC). JPL-produced global ionospheric maps (GIMs)30 were used to validate model TEC estimates. We utilize the True Skill Statistic (TSS)31 metric for the TEC prediction evaluation, noting that this is but one metric to assess32 predictive skill and that complete evaluations require combinations of such metrics. The33 meanings of contingency table elements for the prediction performance are analyzed in34 the context of ionosphere modeling. Prediction success is between about 0.2 and 0.5 for35 weak ionospheric disturbances, and decreases for strong disturbances. We evaluate the36 prediction of TEC decreases and increases. Our results indicate that physics-based mod37eling during storms shows promise in TEC prediction with multi-day lead time.