Machine Learning Based Modeling of Thermospheric Mass Density

In this study, we propose a machine learning based approach to construct an empirical model of thermospheric mass densities, based on the MultiLayer Perceptron and bi-directional Long Short-Term Memory for ensemble learning model (MBiLE). The MBiLE model was trained by using only the thermospheric mass density from Swarm C satellite at similar to 450 km altitude. To assess the performance of the MBiLE model, the model predictions were compared with observations from several satellites, namely, the Swarm C, the Challenging Minisatellite Payload (CHAMP) and the Gravity Field and Steady-State Ocean Circulation Explorer (GOCE) satellites. The determination coefficients (R2) for the three satellites are 0.98, 0.99, and 0.98, respectively. The MBiLE model predicts the thermospheric mass density well not only at Swarm C altitude but also at lower altitudes. Earlier empirical models based on multivariate least-square-fitting approach failed to achieve this good altitude generalization (e.g., Liu et al., 2013, ; Xiong et al., 2018a, ). Further tests have been made by checking the MBiLE model prediction deviations in relation to magnetic local time, day of year, solar flux level, and magnetic activities. No obvious dependences are found for these parameters. Comparing with the NRLMSIS-2.0 model, the MBiLE model improves prediction accuracy by 91%, 66%, and 56% at the three satellites altitudes. The results indicate that the MBiLE model has the ability to predict well the thermospheric mass density over a wide altitude range, for example, from 224 to 528 km, offering potential for atmospheric research applications. Accurate prediction of thermospheric mass density is of significant importance for various space activities, for example, the safe operation of low Earth orbital satellites. To address the challenge of predicting thermospheric mass density over a broader range of altitudes, we propose an integrated model based on machine learning algorithms. By training the model solely with data from only the Swarm C satellite, we achieve high-quality predictions of thermospheric mass density covering a wider altitude range. Remarkably, the model demonstrates robustness against other parameters and exhibits excellent stability. In comparison with the latest version of NRLMISIS model, the prediction from our machine learning model has improved the accuracy by about 91%, 66% and 56% when taking the satellite measurements from Swarm C, CHAMP and GOCE as reference. These results highlight the efficacy and potential of our integrated learning model in accurately predicting thermospheric mass density, thereby benefiting future atmospheric research and enhancing space mission planning. Based on observations from Swarm C and machine learning approach we constructed the MBiLE model for predicting thermosphere mass density Despite using only Swarm C data at about 450-500 km altitude, the MBiLE model predictions well with satellite observations within 230-500 km Our MBiLE model exhibits superior altitude generalization, exceeding earlier models based on multivariate least-square-fitting approach