Regional modeling of ionospheric peak parameters using GNSS data—An update for IRI

Abstract

The F2 layer of the ionosphere plays an essential role in radio communication and positioning applications as well as in space weather research. In particular, the characteristics of the F2 layer are defined by the F2 peak density NmF2NmF2, the F2 peak height hmF2hmF2 as well as the scale height HF2HF2. The International Reference Ionosphere (IRI) is the internationally recognized and recommended standard for predicting these critical parameters. However, IRI provides median monthly values of these parameters based on the International Radio Consultative Committee (CCIR) or the International Union of Radio Science (URSI) models, which were developed from data of the worldwide network of ionosondes during the years 1954 to 1958. These models provide monthly averages, and therefore, they are required to be updated with up-to-date measurements to get more accurate predictions. In this contribution, we provide a procedure to improve the parameters NmF2,hmF2 and HF2HF2 from modern space geodetic measurements, which can serve to update IRI. Specifically, we model these key parameters spatio-temporally by tensor product of B-spline expansions, and estimate the model coefficients using two types of GNSS data, namely, ground-based two-frequency GPS observations of total electron content and electron density profiles retrieved from ionospheric GPS radio occultation measurements acquired by the FORMOSAT-3/COSMIC mission. In this manner, the solution of the model parameters benefits from different sensitivities as well as from different spatio-temporal resolutions of the two observation techniques. The model is applied exemplarily over a South American region on three selected days under high solar activity (1 July 2012), moderate solar activity (16 July 2011) and low solar activity (16 July 2008), respectively. A comparison of our results with ionosonde data, a comparison of vertical total electron content (VTEC) values and a cross validation show the strengths of our approach and the potential to update the IRI model.