Quasi zenith satellite system-reflectometry for sea-level measurement and implication of machine learning methodology

Abstract

The tide gauge measurements from global navigation satellite system reflectometry (GNSS-R) observables are considered to be a promising alternative to the traditional tide gauges in the present days. In the present paper, we deliver a comparative analysis of tide-gauge (TG) measurements retrieved by quasi-zenith satellite system-reflectometry (QZSS-R) and the legacy TG recordings with additional observables from other constellations viz. GPS-R and GLONASS-R. The signal-to-noise ratio data of QZSS (L1, L2, and L5 signals) retrieved at the P109 site of GNSS Earth Observation Network in Japan (37.815° N; 138.281° E; 44.70 m elevation in ellipsoidal height) during 01 October 2019 to 31 December 2019. The results from QZSS observations at L1, L2, and L5 signals show respective correlation coefficients of 0.8712, 0.6998, and 0.8763 with observed TG measurements whereas the corresponding root means square errors were 4.84 cm, 4.26 cm, and 4.24 cm. The QZSS-R signals revealed almost equivalent precise results to that of GPS-R (L1, L2, and L5 signals) and GLONASS-R (L1 and L2 signals). To reconstruct the tidal variability for QZSS-R measurements, a machine learning technique, i.e., kernel extreme learning machine (KELM) is implemented that is based on variational mode decomposition of the parameters. These KELM reconstructed outcomes from QZSS-R L1, L2, and L5 observables provide the respective correlation coefficients of 0.9252, 0.7895, and 0.9146 with TG measurements. The mean errors between the KELM reconstructed outcomes and observed TG measurements for QZSS-R, GPS-R, and GLONASS-R very often lies close to the zero line, confirming that the KELM-based estimates from GNSS-R observations can provide alternative unbiased estimations to the traditional TG measurement. The proposed method seems to be effective, foreseeing a dense tide gauge estimations with the available QZSS-R along with other GNSS-R observables.

Advancing Sea Surface Height Retrieval through Global Navigation Satellite System Reflectometry: A Model Interaction Approach with Cyclone Global Navigation Satellite System and FengYun-3E Measurements

“…Subsequently, an improved ResNet network was established for the purpose of retrieving SSH. Another method employed in machine learning involves utilizing parametric mode decomposition with spaceborne data from the QZSS-R platform [41], using L1, L2, and L5 signals to retrieve the respective correlation coefficients. Furthermore, different spaceborne platforms operate in different orbits, resulting in different geometry configurations of reflected signals on the Earth's surface over time.…”

Multi-GNSS Combination Multipath Reflectometry Based on IVMD Method for Sea Level Retrieval

“…The experimental results showed that the Beidou reflection signal can also be used for sea level monitoring [11]. Ansari et al, confirmed that the KELM-based estimates from multi GNSS-R observations, comprising QZSS-R and other complementary measurements from GPS-R and GLONASS-R, provided alternative unbiased estimations to the traditional tide-gauge measurement [12]. Furthermore, more error sources have been corrected to improve the accuracy of sea level monitoring [13][14][15].…”