@InProceedings{OsanyinGuCaMiHaGr:2024:VaDaIn,
author = "Osanyin, Taiwo Olusayo and Guedes, F{\'a}bio Becker and Candido,
Cl{\'a}udia Nicoli and Migoya-Orue, Yenca and Habarulema, John
Bosco and Graciela, Molina",
affiliation = "{Instituto Nacional de Pesquisas Espaciais (INPE)} and {Instituto
Nacional de Pesquisas Espaciais (INPE)}",
title = "Validation of Data Ingestion Technique in the NeQuick 2 Model over
South America by Means of TEC, Ionosonde and Radio Occultation
Data",
booktitle = "Proceedings...",
year = "2024",
organization = "Conferencia Latinoamericana de Geof{\'{\i}}sica Espacial, 14.",
abstract = "In this study, an investigation of data ingestion into the NeQuick
2 model using networks of ground-based Global Navigation Satellite
(GNSS) has been carried out for different spatial resolutions over
South America. This serves as a benchmark for cost-effectiveness
in using ground-based GNSS observation for the update and
validation of NeQuick 2 for a near-real-time specification of the
state of the ionosphere. Statistical analysis based on spatial
index selection of GNSS data shows an improvement in both middle
and lowlatitude regions considering ingestion of data over the
regions separately. We observed a decrease in the RMSE between the
NeQuick 2 and GNSS observation as the number of ingested GNSS TEC
data increases over the test stations with an average improvement
of about 26 % and 28 % for both 34 and 120 observations. Our
results showed that the datadriven NeQuick 2 model is able to
describe the evolution of the Equatorial Ionization Anomaly during
both quiet (10 March 2015) and disturbed (17 March 2015)
geomagnetic conditions. The diurnal variation of vTEC, hmF2, and
Ne over the test stations showed considerable improvement after
data ingestion. Validation with Ionosonde measurements and COSMIC
Radio Occultation data showed significant improvement in
reconstructing the electron density of the bottomside and topside
ionosphere profiles after data ingestion. Between the altitude of
100-250 km, the NeQuick \$%\&2 is in good agreement with
ionosonde measurement with an average correlation coefficient of
0.95 and 0.83 at Sao Luis (2.58°S, 44.20°S), 0.94 and 0.91 at
Boavista (2.83°N, 60.70°S), and 0.75 and 0.71 at Campo Grande
(20.40°S, 54.50°S) during quiet and disturbed conditions.
Likewise, the biases obtained for the peak electron density
(\$%\&2 ) at different times are larger compared to those
estimated for the peak height (\ℎm\&2).",
conference-location = "Monterrey, Mexico",
conference-year = "08-12 Apr. 2024",
language = "en",
urlaccessdate = "2024, July 27"
}