Development and error analysis of nonlinear ionospheric removal algorithm for ionospheric electron density determination using broadband RF data

[1] The first documented, empirical comparisons are provided of four methods to retrieve total electron content (TEC) that use broadband, impulsive events detected by satellite in the lower very high frequency range (20–150 MHz). The four TEC retrieval methods are the quasi‐longitudinal approximation (i.e., Taylor expansion) of the Appleton‐Hartree (A‐H) dispersion relation to the first and second orders, as well as the nonlinear ionospheric removal algorithm (NIRA) that utilizes the A‐H dispersion equation directly to model the propagation of an electromagnetic wave through the ionosphere. NIRA solves not only for TEC between the ground source and satellite, but also for higher‐order ionospheric terms, such as electron density, ionospheric thickness, and angle between wave vector and magnetic field. Regimes of validity for each TEC retrieval method are analyzed by comparison of the parameters retrieved from synthetic data with a known ionosphere and from RF FORTE satellite data measurements of a ground‐based broadband transmitter. Results include a comparison between TEC and infinite frequency time of arrival (to) determined by NIRA and determined by using the first‐ and second‐order terms from the Taylor expansion of the A‐H equation. Plasma density, ionospheric thickness, and angle between magnetic field and wave vector as determined by the two NIRA methods are also compared.