| dc.description.abstract | Space weather disturbances lead to significant changes in the neutral density and have a substantial impact on the thermosphere-ionosphere system due to the interaction between neutral particles and plasma. These subsequently affect the satellite drag, which plays an important role in satellite orbit maintenance, collision avoidance, and satellite traffic management. However, we still lack the ability to reliably predict the dynamic density of the upper atmosphere. The Iterative Driver Estimation and Assimilation (IDEA) data assimilation technique was employed with the Whole Atmosphere Model (WAM) to enhance neutral density specification in the upper thermosphere. Given the intensity of the November 2003 storm, two changes were necessary in WAM prior to applying IDEA. The first was to allow the Kp geomagnetic index to exceed 9 and the second was to modify the relationship between Kp and the solar wind parameters used to drive the model. With these changes in place, results show that WAM ingesting the accelerometer estimates of neutral density from the Challenging Mini-Satellite Payload (CHAMP) satellite effectively captured the thermospheric neutral density at the CHAMP’s altitude. Furthermore, data assimilation outputs were also validated against an independent neutral density data set from the Global Ultraviolet Imager (GUVI) limb-scan daytime airglow observations aboard the Thermosphere Ionosphere Mesosphere Energetics and Dynamics (TIMED) satellite, and the strong agreement within 270-320 km altitude shows the utility of the IDEA technique and a potential use case of GUVI data set in the IDEA system. An experiment was conducted in which WAM ingested GUVI-derived neutral density at 300 km, and IDEA-GUVI data assimilation outputs closely matched GUVI observations throughout the storm period, while another data source should be incorporated to supplement the information. Given the limited spatial and temporal coverage of satellite measurements, High Accuracy Satellite Drag Model (HASDM) density database was employed for robust, global-scale, and long-term data integration into the IDEA scheme. Results show that the IDEA-HASDM effectively eliminated the model bias and brought the model density into the agreement with CHAMP during the quiet time. The prominent discrepancy, however, between WAM and CHAMP densities arises from the density dips at high latitudes during the peak of the storm. This discrepancy might result from different locations of observed and modeled high-latitude density hole structures, while further investigations are required to address this issue in the future. Additionally, to capture the rapid changes in the thermosphere, a reduced Kp estimation window from 3 to 1.5 hours was experimented. Meanwhile, the estimated F10.7 value remained unchanged at 24 hr in order to limit the correlation between Kp and F10.7 corrections. While this adjustment significantly reduced the model bias, it resulted in worse root-mean-square errors and standard deviation, which suggests that estimating six 1.5-hour Kp values instead of three 3-hour Kp lowers the observability of these interrelated and correlated parameters. For the future transition from a nowcasting to a forecasting system, solar and Joule heating scale factors were used as the estimators in IDEA, allowing the model to ingest the observed solar wind drivers. The good agreement between IDEA and CHAMP shows that the two scale factors can equally be integrated in the IDEA system.
This dissertation demonstrates the utility of the IDEA scheme based on WAM and that using various data sources, such as neutral density estimates from accelerometers and airglow limb scan measurement and HASDM database. The improvements made to WAM can further enhance our understanding of the thermosphere-ionosphere coupling, bolster the whole atmosphere nowcasting and forecasting capability, and improve the accuracy of LEO satellite orbit determination. | en_US |