This is an inter-disciplinary project to expand existing ionospheric data assimilation techniques to develop a new state-of-the-art imaging and data assimilation technique for the plasmasphere. It will use a variety of observations of plasmaspheric densitites to produce time-evolving maps of the electron density i nthe plasmasphere. These maps will then be fed into first-principles models of the plasmaspehre to produce corrected and improved characterisations and predictions of plasmasphere dynamics. The method will be tested and validated via simulations. Finally, the technique will be used on real GPS and in-situ measurements to address questions of ionosphere-plasmasphere coupling through refilling and erosion, and for characterizations of plasmasphere plumes. The work will be carried out at Augsburg College under the mentorship of Dr. Murr and in collaboration with Dr. Bust of Atmospheric & Space Technology Research Associates and Dr. Huba of the Space Physics Section at the Naval Research laboratory.
The project will involve undergraduate students at Augsburg College. The project will produce a practical methodology to predict the state of the plasmasphere more accurately and more efficiently. Plasmasphere dynamics is an important ingredient in understanding space weather effects in the ionosphere. Therefore, the new model will help improve our space weather forecasting capabilities.
The major goal of the project is to develop a computationally efficient imaging and data assimilation technique for estimation of electron density in the plasmasphere and the top ionosphere. This information is critical for addressing the theoretical problems of ionosphere-plasmasphere coupling through refilling and erosion. Knowledge of plasmasphere density structure is also needed for characterization of plasmasphere plumes, which play an important role in the dynamics of the magnetosphere-ionosphere coupling. This study provides more accurate nowcast and forecast of the plasmasphere state, and hence is important in mitigating the adverse effects of the space weather on human activities and operational systems such as high-frequency communication and navigation systems using GPS satellites. In the data assimilation framework of this study, we use upward looking GPS total electron content (TEC) measurements from the Constellation Observing System for Meteorology, Ionosphere, and Climate (COSMIC) satellites as well as a background model for the plasmasphere. The COSMIC mission is a constellation of six satellites in the polar orbit (inclination of 72 degrees). The upward looking TEC data is an integration of electron density along the line-of-sight (LOS) between COSMIC and GPS altitudes. By discretization of the space in 3D (latitude, longitude, and altitude) and calculating the length of each LOS passing through each grid voxel, we compute a forward model, which relates the unknown electron density and the TEC data. We then form an optimization problem where deviations from the data and the background model are both penalized. Finally, we apply least-squares techniques to the optimization problem at hand to seek optimal estimates of electron density. With this setting, the solution will follow the data more closely in data-rich regions whereas in data-poor regions, where data is not available, the solution will be close to the background model. Due to limited availability of data, we are required to apply regularization techniques where we apply a priori knowledge on the estimation solution. In this study, to ensure smoothness of the solution along the altitude, we applied a smoothness constraint. To incorporate regularization into our analysis, we modified the optimization problem at hand to penalize sharp altitudinal gradients of the solution, as well. We used several background models for this study such as the International Reference Ionosphere model with the Izmiran expansion, Global Core Plasma model, and the SAMI2 model. The first two models are empirically driven climatological models, whereas the third one is a physics-based first-principle model. We found that due to limited availability of data, estimation results depend heavily on the choice of the background model and the background errors. Using the GCPM, we compared the regularized electron density estimates with the in situ density measurements from the Defense Meteorological Satellite Program (DMSP) and with the extracted densities from the Van Allen Probe measurements in geomagnetically quiet and active days, respectively. Reasonable agreements were shown in both cases.
