Assessment of new thermospheric mass density model using NRLMSISE-00 model, GRACE, Swarm-C, and APOD observations

Abstract

Thermospheric mass density estimates from in-situ observations along satellite orbits are difficult to validate due to their inherent spatiotemporal sparse nature, and difficulties related to drag-force modeling and estimation of actual mass density state. Current upper atmospheric models are unable to accurately represent the actual thermospheric variability, and in-situ observations are far to fulfill the minimum requirements in practical applications. In this manuscript, the new Thermospheric Mass Density Model (TMDM) is based on the fit of solar flux, annual, Local Solar Time (LST), and magnetospheric proxies into the Principal Component Analysis (PCA) of 13 years of accelerometer-based mass density estimates derived from the GRACE (Gravity Recovery and Climate Experiment) mission. We employ the NRLMSISE-00 model and estimates from APOD (Atmospheric density detection and Precise Orbit Determination), Swarm-C, and GRACE satellites, and assess the new model, including statistical analyses, and a Precise Orbit Determination (POD) scheme. We compare 2 years of APOD and Swarm-C estimates, and study the dynamic orbit propagation of the 3 missions under different mass density input schemes and different magnetospheric activity conditions. The results with TMDM show similar differences in the dynamically propagated orbits from NRLMSISE-00 and in-situ observations. The statistical analyses show that NRLMSISE-00 overestimates about 20%, and TMDM underestimates about 20%, the in-situ observations.