Validation of IRI and IRTAM Models Based on Global Ionosphere Radio Observatory Data

The ionospheric model serves as a crucial pillar for both space science research and the engineering of radio systems. Precise understanding of the ionospheric model's performance is a foundational requirement for the attainment of high-precision corrections to ionospheric errors. Utilizing ionospheric data collected through Global Ionospheric Radio Occultation (GIRO), the International Reference Ionosphere (IRI) model and the International Reference Ionosphere Model (IRTAM) were subjected to analysis and evaluation. The outcomes of this evaluation indicate that: 1) The average absolute prediction errors of the ionospheric parameters foF2 and hmF2 of the IRI model in 32 regions including KING SALMON in 2011 were 2.8292 MHz and 48.7899 km respectively, while the prediction averages of the IRTAM model were 3.0790 MHz and 45.0783 km respectively. It can be seen from this that the reliability of the IRTAM model in predicting the foF2 parameter is slightly lower than that of the IRI model, but the difference is not significant. The error in predicting the hmF2 parameter is significantly lower than that of the IRI model, which means that the IRTAM model has a significant advantage in the performance of predicting the hmF2 parameter. 2) Within one day, with the fluctuation of solar activities, the fluctuation range of the prediction error maximum and minimum of the IRI model for hmF2 in different regions is between 2.15 and 4.46 times, while the error fluctuation range of the IRTAM model is within the range of 1.71 to 3.71 times. This comparison shows that solar radiation has a certain impact on the prediction accuracy of both models, but the degree of influence on the IRTAM model is relatively small, and its error fluctuation range is more convergent, reflecting the stability of this model in prediction accuracy is better than that of the IRI model. 3) For the prediction error of the hmF2 parameter, the average value of the IRTAM model is 0.1578, while that of the IRI model is 0.1674. For the foF2 parameter, the average prediction error of the IRTAM model is 0.3535, and that of the IRI model is 0.3360. In addition, the errors of the IRTAM model in predicting the B0 and B1 parameters are 0.3197 and 0.3824 respectively, which further confirms the advantage of the IRTAM model in prediction accuracy.