Global positioning system interferometric reflectometry for accurate tide gauge measurement: Insights from South Beach, Oregon, United States. (August 2020)

• Record Type: Journal Article
• Authors: Ansari, Kutubuddin; Bae, Tae-Suk; Inyurt, Samed
• Is Part Of: Acta astronautica. Volume 173(2020)
• Page Start: 356
• Page End: 362
• Publication Date: 2020-08
• Subjects: Tide gauge -- Global positioning system -- Singular spectrum analysis -- Auto-regressive moving average; Astronautics -- Periodicals; Outer space -- Exploration -- Periodicals; Astronautics; Periodicals; 629.405
• DOI: 10.1016/j.actaastro.2020.04.060
• Languages: English
• ISSNs: 0094-5765

Abstract

Abstract: Global positioning system (GPS) stations located along coastal areas have the ability to measure tide gauge records by reflected signal reception from the sea water surface. In this study we used the GPS signal-to-noise ratio data from the GPS station (44.6257 ⁰N, 124.0455 ⁰W) located at South Beach, Oregon, United States, to estimate the tide gauge records using seasonal measurements. First, we derived the tide gauge record from the station and used traditional tide gauge data from the National Water Level Observation Network (NWLON) sentinel station (Station ID: 9435380) located in Oregon for validation purposes because it was closest to the GPS station. Our results show that the GPS tide gauge and NWLON tide gauge correlate well with the correlation coefficient of 0.942 and the root mean square of their residuals was about 12.90 cm. The corresponding tide gauge prediction by auto-regressive moving average and singular spectrum analysis models are evaluated for their effectiveness over the station. The comparative analysis demonstrates that the GPS tide gauge has improved correlation with auto-regressive moving average predicted tide gauge (correlation coefficient of ~0.981 and root mean square of ~4.80 cm), and singular spectrum analysis predicted tide gauge (correlation coefficient of ~0.998 and root mean square of ~0.88 cm) compared to the NWLON tide gauge (correlation coefficient of ~0.942 and root mean square of ~12.90 cm) values. We believe the outcomes from this study contribute to a better understanding of the numerical modeling of tide gauge records as well as other measurements based on reflectometry techniques. Highlights: Proposes novel mathematical options for tidal gauge forecasting. Uses ARMA-TG application for more accurate tidal gauge forecasts. Near one-to-one tidal gauge forecast correlation: SSA-TG technique to GPS. Uses GPS signal-to-noise ratio (SNR) data to estimate the TG records. Findings indicate that ARMA-TG and SSA-TG are suitable for TG forecasting.