Evaluating Ionospheric TEC Predictions: A Comparative Analysis between IRI-2016 and IRI-2020 Models Over Indian Equatorial and EIA Regions During 2016-2018

Abstract

This study conducts a comprehensive comparative analysis of the ionospheric total electron content (TEC) prediction capabilities of the recent International Reference Ionosphere (IRI) models, specifically IRI-2016 and IRI-2020, by utilizing ground-based GPS measurements over the equatorial ionization anomaly (EIA) region, Varanasi (Geographic: 25.31° N, 82.97° E; geomagnetic: 16.54° N, 157.09° E) and the near equatorial region Bangalore (Geographic: 12.97° N, 77.59° E; geomagnetic: 4.86° N, 150.95° E) during the solar minimum period of 2016-18, corresponding to the 24th solar cycles. Through rigorous statistical evaluations, including diurnal, monthly, seasonal, and annual TEC variation comparison, the root mean squared error (RMSE) of dTEC, and correlation coefficient calculation, this research investigates the forecasting accuracy of Ne-Quick (the default topside option) of IRI-2016, Ne-Quick, and COR2 (the newly added option) of IRI-2020 models. Although the performance of all the topside options is equally good, the Ne-Quick topside of IRI-2016 shows better accuracy over Varanasi (RMSE = 6.09), whereas the prediction by COR2 topside is better over Bangalore (RMSE = 6.02). Furthermore, the overall correlation coefficient analysis highlights the superior performance of the Ne-Quick of the IRI-2016 model (r = 0.88) in capturing GPS TEC dynamics over the near equatorial and low-latitude regions during our study period. However, the performance of all the topside options in forecasting the TEC during the storm time is relatively poor, with a correlation coefficient around ~0.67. These findings offer significant implications for the ionospheric physics research community, providing a validated reference for the IRI models in predicting ionospheric TEC characteristics during solar minimum years over near equatorial and EIA regions.