Assessing GNSS ionospheric models at low latitudes: BDGIM, NeQuick-G, and Klobuchar

Abstract

Single-frequency Global Navigation Satellite System (GNSS) users primarily rely on broadcast models to adjust their observations for ionospheric delay, one of the major sources of error in GNSS positioning. In this study, we assess the performance of Global Positioning System (GPS)’s Klobuchar, Galileo’s NeQuick-G, and BeiDou Navigation Satellite System (BDS)’s BeiDou Global Ionospheric delay correction Model (BDGIM) models at low latitudes. Our analysis is based on data from the years 2020 and 2021, during the ascending phase of the solar cycle. We compared total electron content (TEC) estimates against regional ionospheric maps (RIM) and calibrated and differential TEC derived from dual-frequency observations of 10 GNSS reference stations located in Brazil. The reference stations were also used to perform single point positioning (SPP). Our findings demonstrate BDGIM’s superior performance over the other two models across all satellite elevation angles and throughout the seasons of the year, achieving a correction rate of 80%. While Klobuchar outperforms NeQuick-G under disturbed ionospheric conditions, it faces challenges in accurately estimating ionospheric delay during nighttime, when NeQuick-G shows strength. Our results also showed that NeQuick-G model tends to underestimate electron density, whereas Klobuchar and BDGIM models tend to overestimate it. In terms of positional accuracy, the maximum error reduction achieved by correcting for the ionosphere is 49% with Klobuchar, 45% with NeQuick-G, and 51% with BDGIM.