The Multifaceted Optimization Algorithm (MOA) is a tool for generating corrected empirical model thermospheric densities during geomagnetic storms. It consists of a suite of Python functions that operate around the Spacecraft Orbit Characterization Kit (SpOCK), an orbital propagator developed by Charles D. Bussy-Virat, PhD, Joel Getchius, and Aaron J. Ridley, PhD at the University of Michigan, and it estimates new densities for the NRLMSISE-00 atmospheric model. MOA generates new model densities by estimating modifications to inputs to the NLRMSISE-00 model that minimize the orbit error between modeled spacecraft in SpOCK, and their actual altitudes as described in publicly-available Two-Line Element Sets (TLEs), made available online via Space-track.org.
MOA consists of three sub-process: (1) The Area Optimization Algorithm (AROPT), (2) the F10.7 Optimization Algorithm (FOPT), and (3) the Ap Optimization Algorithm (APOPT). AROPT computes the contribution to the drag of the modeled spacecraft due to their varying projected area. FOPT estimates modifications to the 10.7 cm solar radio flux in NRLMSISE-00, and APOPT estimates modifications to the Earth's magnetic activity in NRLMSISE-00. MOA finds these modifications across many spacecraft, and the medians of those modifications are then applied in NLRMSISE-00 along the orbit of another satellite to generate new densities for verification. In this instance, modifications are applied along the orbits of the Swarm spacecraft and compared to Swarm GPS-derived densities.
Brandt, D. A., Bussy-Virat, C. D., & Ridley, A. J. (2020). A Simple Method for Correcting Empirical Model Densities During Geomagnetic Storms Using Satellite Orbit Data. Space Weather, 18(12), e2020SW002565. https://doi.org/10.1029/2020SW002565
The research that produced this data focused on conducting a statistical comparison between horizontal winds modeled with GITM and those derived from the accelerometer aboard the GOCE satellite. The winds from GITM and GOCE were compared by constructing their respective probability densities under different levels of geomagnetic activity, and by distributing them as a function of geomagnetic activity, magnetic latitude, magnetic local time, day-of-the-year, and solar radio flux.
