First Results of a New Inversion Tool for Thermospheric Neutral Mass Density Computations During Severe Geomagnetic Storms

Abstract In Low Earth Orbit (LEO), atmospheric drag is the dominant source of trajectory prediction error below approximately 700 km altitude, primarily due to inaccuracies in thermospheric density models. This work presents a method for temporally deconvolving Energy Dissipation Rates (EDRs) to produce single‐satellite density estimates, which can be used in future studies to investigate storm‐time density response and to inform improvements in thermospheric density modeling. An inversion tool algorithm estimates corrections to the ap index using the Naval Research Laboratory Mass Spectrometer and Incoherent Scatter radar Extended model (NRLMSISE‐00, commonly referred to as MSIS) as the background density model within an Ensemble Kalman Filter (EnKF) framework. EDRs from calibration satellites, satellites with known and relatively constant ballistic coefficients, are processed individually to invert a time series of single‐satellite perigee densities. The performance of the inversion tool is evaluated through comparisons with MSIS, the High Accuracy Satellite Drag Model (HASDM), and accelerometer‐derived densities during two periods: (a) two consecutive G4 storms in May 2005 during the declining phase of solar cycle 23 near solar minimum, and (b) the May 2024 Gannon G5 storm near the solar maximum of cycle 25. The inversion tool estimated effective densities show strong agreement with observed effective densities at perigee altitudes ranging from ∼200 to ∼750 km. However, the inversion tool does not reliably temporally deconvolve densities for satellites with very low perigee altitudes (