Tidal measurements in the Gulf of Mexico: intercomparison of coastal tide gauge, insular GNSS reflectometry and SAR altimetry

• Published in: GPS solutions 2022, Vol.26 (1), Article 22
• Main Authors: Martinez-Felix, Carlos A., Vazquez-Becerra, G. Esteban, Geremia-Nievinski, Felipe, Millan-Almaraz, Jesus R., Franco-Ochoa, Cuauhtemoc, Melgarejo-Morales, Angela, Gaxiola-Camacho, J. Ramon
• Format: Article
• Language: English
• Subjects: Altimeters; Atmospheric Sciences; Automotive Engineering; Correlation; Earth and Environmental Science; Earth Sciences; Electrical Engineering; Fruits; Gauges; Geophysics/Geodesy; Global navigation satellite system; Original Article; Reflectometry; Reliability analysis; Root-mean-square errors; Satellite altimetry; Satellite observation; Sea level; Shallow water; Signal to noise ratio; Space Exploration and Astronautics; Space Sciences (including Extraterrestrial Physics; Synthetic aperture radar; Tidal range
• ISSN: 1080-5370; 1521-1886
• DOI: 10.1007/s10291-021-01207-x

Sea level can be monitored using conventional tide gauges, ground-based Global Navigation Satellite System Reflectometry (GNSS-R), and synthetic aperture radar (SAR) satellite altimetry. Each technique has its advantages and disadvantages, so an intercomparison is a good exercise for cross-validation. We analyze the reliability of a standard geodetic-grade GNSS receiver and antenna and two Sentinel-3A SAR products to measure sea level variations in the Gulf of Mexico. We considered a one-year period over a 120 km wide region between an island and mainland, composed of reefs and shallow waters. First, signal-to-noise ratio (SNR) observations from a GNSS station (CN26) on the island were analyzed. Second, high-resolution SAR altimetry data (20 Hz) of the Sentinel-3A satellite were acquired based on two processors, SARvatore and Peachi. The above results were compared to a conventional tide gauge located on the mainland at the shoreline of Progreso, Yucatan. The GNSS-R relative sea level had a correlation of 0.84 and root-mean-square error (RMSE) of 7.8 cm with the tide gauge, while SAR altimetry products had a correlation and RMSE of 0.86 12.4 cm for SARvatore and 0.85, 12.8 cm for Peachi. Furthermore, correlation and RMS between GNSS-R and SAR were found to be 0.94, 8.6 cm for SARvatore and 0.95, 7.2 cm for Peachi. Meanwhile, SARvatore and Peachi achieved an internal correlation of 0.92 and RMSE of 8.9 cm. Moreover, scale variations caused by the differences in the tidal range were quantified by linear regression slope with respect to the tide gauge, which amounted to 0.554, 0.843 and 0.814, respectively, for GNSS-R, SARvatore and Peachi. Finally, derived daily GNSS-R observations in conjunction with SAR altimetry were coherent with respect to the reference tide gauge indicating that GNSS-R can operate as an auxiliary validation instrument for SAR altimetry coastal measurements at ungauged locations.