Technical note on the multi-GNSS, multi-frequency and near real-time ionospheric TEC monitoring system for South America

Abstract

<p>Development of regional services able to provide ionospheric total electron content (TEC) maps with a high spatial resolution, and in near real-time, are of high importance for applications and the research community. We provide here the methodologies, and a preliminary assessment, of such a system.The system relies on the public Global Navigational Satellite Systems (GNSS) infrastructure in South America, it incorporates data from multiple constellations (currently GPS, GLONASS, Galileo and BeiDou), it employs multiple frequencies, and it produces continental wide TEC maps with a latency of just few minutes. A year-round comparison of the produced maps with several products issued by the International GNSS Service (IGS) resulted in mean biases lower than 1 TEC units (TECU), whereas their evaluation against direct and independent GNSS-based slant TEC measurements resulted in accuracies of the same magnitude.</p>

A Multi‐GNSS, Multifrequency, and Near‐Real‐Time Ionospheric TEC Monitoring System for South America

“…Also, to lesser extent, with those from UPC and Center for Orbit Determination in Europe. In general, all comparisons result in mean differences and standard deviations larger than the obtained in the year‐round analysis (Mendoza et al, ). Indeed, the rapid spatial and temporal variability of the physical conditions, naturally associated to a geomagnetic storm, are responsible for the larger discrepancies between the different products and, of course, also between them and the actual TEC they are trying to represent.…”

“…Indeed, the ionosphere over Brazil and southern South America is adequately covered but a large area in the northwest of the subcontinent is, systematically, not optimally observed (Figure 1, right). Although our analysis indicates that the produced maps are also reliable over this area (see Mendoza et al, 2019), the system could clearly benefit in the future from the availability, and subsequent addition, of openly accessible real-time GNSS data streams from stations located in that region. In fact, the system has been designed to be highly scalable, in order to incorporate both additional navigational systems (constellations) and ground stations (networks) without the penalty of a significantly increased latency.…”

“…In general, the near‐real‐time TEC maps (Figure and Table ) are available online less than 3 min after the last GNSS observation is acquired (i.e., timeliness of ∼3 min), that is, about 10 min after the mean observational epoch of the issued product (i.e., ∼10‐min latency). These short intervals, achieved thanks to the aggressive parallelization of the analysis system, includes the observational and orbital RINEX (Gurtner & Estey, ) files concatenation, the data cleaning (the preprocessing), the Interfrequency Biases calibration, and the vertical TEC representation (the postprocessing, see Mendoza et al, ). In practice, plots of the current TEC map and the employed satellite‐receivers Ionosphere Pierce Point traces, together with a summary of the latest analysis, are updated every 15 min and can be accessed anonymously from wilkilen.fcaglp.unlp.edu.ar/ion/latest.png (or alternatively in Spanish from wilkilen.fcaglp.unlp.edu.ar/ion/ultimo.png).…”

“…A plot of the most recent TEC map can be accessed anonymously from wilkilen.fcaglp.unlp.edu.ar/ion/latest.png(or alternatively in Spanish from wilkilen.fcaglp.unlp.edu.ar/ion/ultimo.png), whereas registered users can retrieve the TEC maps produced by the system, in IONEX and NetCDF formats, from wilkilen.fcaglp.unlp.edu.ar/ion/magn/ website. A detailed list of all the employed and openly accessible data streams is given in Mendoza et al (). All IGS products were retrieved from ftp://cddis.gsfc.nasa.gov/gnss/products/ionex/ website.…”

“…In general, the near-real-time TEC maps ( Figure 1 and Table 1) are available online less than 3 min after the last GNSS observation is acquired (i.e., timeliness of ∼3 min), that is, about 10 min after the mean observational epoch of the issued product (i.e., ∼10-min latency). These short intervals, achieved thanks to the aggressive parallelization of the analysis system, includes the observational and orbital RINEX (Gurtner & Estey, 2017) files concatenation, the data cleaning (the preprocessing), the Interfrequency Biases calibration, and the vertical TEC representation (the postprocessing, see Mendoza et al, 2019). In practice, (Wessel et al, 2013) Note.…”