Performance Evaluation of a Data Assimilation Method Estimating the External Forcing of the I-T System

Uncertainties in solar and geomagnetic energy inputs are a significant source of error when specifying the state of the thermosphere-ionosphere system with global physics-based models. To address this, the Iterative Driver Estimation and Assimilation (IDEA) data assimilation scheme was created to estimate corrections to these inputs through comparison of model outputs with accelerometer measurements of thermosphere mass density. While this assimilative technique has been shown to significantly increase the model-data agreement of neutral mass density for the NCAR TIE-GCM and NOAA WAM-IPE global models, the effects of the data assimilation on the full model state require more examination. Here, we compare the performance of assimilative runs of TIE-GCM and WAM-IPE with the performance of control runs using observed F10.7 and Kp values. The model-data agreement of both the thermosphere and ionosphere for each run are assessed, while the inclusion of two models serves to separate model-specific effects with those of the IDEA technique. Alongside knowledge of the influence of F10.7 and Kp variability in the models, these results serve to guide IDEA development so that the performance improvements of the data assimilation extend beyond neutral mass density to the full model state.