A coastal geodetic GNSS station for tectonic and sea-level variation study in the South China Sea

Qiu, ﻿Qiang; Li, Linlin; Ye, Mai; Yang, Hongqiang; Song, Xingyu; Zheng, Tingting; Cao, Jinghe; Wang, Peitao; Chen, Zhanying; Zhang, Zhiwen; Xiao, Kuilin; Huang, Xing; Huang, Yadong; Zheng, Chuanyang; OuYang, Zhiyuan; Fu, Xiaoming; Wang, Muzhong; Wang, Zhen; Zhang, Zitao; Cui, Haiping

doi:https://doi.org/10.5194/egusphere-2025-3563

Preprints

https://doi.org/10.5194/egusphere-2025-3563

Preprints

16 Sep 2025

| 16 Sep 2025

Status: this preprint is open for discussion and under review for Natural Hazards and Earth System Sciences (NHESS).

A coastal geodetic GNSS station for tectonic and sea-level variation study in the South China Sea

Qiang Qiu, Linlin Li, Mai Ye, Hongqiang Yang, Xingyu Song, Tingting Zheng, Jinghe Cao, Peitao Wang, Zhanying Chen, Zhiwen Zhang, Kuilin Xiao, Xing Huang, Yadong Huang, Chuanyang Zheng, Zhiyuan OuYang, Xiaoming Fu, Muzhong Wang, Zhen Wang, Zitao Zhang, and Haiping Cui

Abstract. South China Sea (SCS) separates the land from north to south by nearly two-thousand km in distance. In this framework, land-based observation is limited, thus, little is known about the tectonics in this widely gapped area. Here, we present a coastal geodetic GNSS station that is recently deployed in Nansha southern SCS. The station measures mm precision of land displacements that show ~4–5 mm range of variation in horizontal components and ~1 cm range of changes in vertical nearly a year. These displacements are characterized by a flat trend, with occasionally disturbed by subcentimeter changes. The GNSS-IR retrieved sea-level variations over both short- and long-term time span are comparable with tide gauge recordings, offering a complement equipment to detect amplitude variations and trend adjustments in both relative and absolute sea levels. Although varying with time, the average rate is commensurate with that of the global mean sea-level changes. With such capacity, we filter the sea-level retrievals to be equally sampled as a conventual tide gauge using a Kalman filter, and conduct experimental tests to investigate whether it can capture sea-level disturbances from extreme events i.e., tsunamis. Our results show that as long as the disturbances are twice larger than the standard deviation of the filtered time series, then such disturbances are successfully detected. This criterion could be largely reduced if the GNSS site was built particularly for GNSS-IR. In any case, the GNSS-IR detectability is particularly helpful to capture sea-level disturbances triggered by quiet submarine landslides as they often don’t send sensible signals as earthquakes do. Our station adds an extra connecting pod to fill in the existing few stations and pushes one-step forward to link the separated land, and affords another example to study regional tectonics and sea-level variations simultaneously.

Received: 24 Jul 2025 – Discussion started: 16 Sep 2025

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 10939 KB)

Supplement (5557 KB)

Download & links

Preprint (10939 KB)
Metadata XML
Supplement (5557 KB)
BibTeX
EndNote

Status: open (until 24 Nov 2025)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2025-3563', Anonymous Referee #1, 09 Oct 2025 reply
Review to “A coastal geodetic GNSS station for tectonic and sea-level variation study in the South China Sea”

General comments:

Overall this is a well structured investigation into the capabilities of a single GNSS station in the South China Sea. The overall presentation is thorough, with reasonable figures that do a good job in showcasing their methods and results. The authors present a processed time series, estimates of long-term motion, and sea level from GNSS-IR. As this station is located in a relatively remote and politically contentious region, the data is of particular importance. Additionally, the authors have done significant work in trying to assess the ability of this station to supplement tide gauge data from region and potentially serve as part of a larger tsunami-warning network. Although this station’s location is not ideal for this purpose, its seems to be able to provide valuable early-warning data in one of the three scenarios tested, namely a submarine landslide offshore of Borneo. The other two scenarios, earthquakes on the Manila subduction zone, seem unlikely to be detectable over the observed noise, unless the tsunami generated is much larger than the current understanding of the Manila subduction zone would imply. The authors also make a claim that the position time series recorded by this GNSS station indicates the region is relatively “stable” though, there is only ~1 year of data, and without assessing the observed motion in the context of other nearby (or nearby-ish) sites, this seems a bit dubious. Additionally, there are a number of outliers in their GNSS-IR estimated sea level that were removed with no explanation, and if included could change their results. Lastly, the authors seem to overstate their observational capacity and over-generalize many of their results, before eventually walking them back to a reasonable stance in the discussion and conclusion. It is for these reasons that I recommend major revisions.

Specific Comments:

Larger-scale and Paper-wide comments
How are outliers defined in the GNSS-IR sea level estimates? While its great to see that the authors are able to use GNSS-IR to retrieve sea level estimates despite a non-optimal site, they do not define how they select outliers, be it in some definition of error from the GNSS-IR data, distance from the average of other “nearby” observations in time, or something else.

Description as GNSS linear trend implying “stability” is inaccurate, as the station’s velocity is strongly dependent on the reference frame it is tied to, and the stability that it implies in the text does not come from interpretation of this velocity estimate in the context of others nearby in the SCS, Philippines, Borneo, or Vietnam. While this is mentioned briefly in the discussion section, the implication of stability is vastly overstated and (in my opinion) incorrect, especially in light of the potential transient signal observed and the limited duration of the geodetic time series.

The actual detectability of tsunamis presented in this paper seems to be largely overstated until the discussion. While case 3 (tsunami from an underwater landslide offshore Borneo) seems to be the only incident that could reasonably be detected by this station, the authors imply that detection is possible for the other events, when as I understand it, they seem to be indistinguishable from normal variations in sea level at the observation frequency possible with this specific GNSS site.

Data: GNSS and tide gauge data should be available in some form, even if only the time series of position and time series of sea level.

Place names of disputed territories should use (or at least acknowledge) most commonly used names, especially as this is an English language publication for an international audience, and ideally the manuscript should not implicitly endorse one territorial claim over another.
Nansha Islands – Spratly Islands

Zeng Mu An Sha submarine landslide – Brunei Slide

All acronyms/Initialisms should be defined when they are first used. Use of definitions is inconsistent, with some things defined, others not.

Abstract:
Wording could be cleaned up a bit

Introduction
Figure 1: choice of cities is interesting. On Borneo, the authors have marked (I think) Bandar Seri Begawan and Bintulu, yet Miri, Kuching, and Kota Kinabalu are all larger cities (and broadly coastal with the exception of Kuching).

Having an outline of the Dangerous grounds block would help here.

Methods – 2.1 GNSS and displacement
Line 131: Is it ITRF14?

Methods – 2.2 GNSS-IR for sea-level retrievals
Equations: Not all variables are defined and there is mismatch between the variables in the equations and those in figure 2.
A space between sin and e would be helpful while reading through

Equation (2-1): While D is effectively the reflected height (2H) projected onto the line of sight vector, that’s not easily apparent from figure 2 as implied.

Methods –2.3 Tsunami detection using synthetic sources
Line 191-193: It would be good to mention that your azimuthal range is somewhat constrained, indicating that although this site is not ideally situated for sea-level/tsunami studies it can still be useful, especially if combined with other sites.

Lines 227-232: Sentences are quite confusing. Rephrasing for clarity would be helpful.

Results – 3.1 Land displacements
Line 246: Can you mention what this is relative to here?

Line 249-251: Maybe describe this as a “potential transient signal” for clarity. Also, this signal appears to be quite small and not exactly correlated between the horizontal components and vertical components. While maybe a real signal, this could just be due to a changing skyview (from construction or plant growth), land reclamation, subsidence, a loose antenna, etc. To confirm an actual physical process, additional datapoints (SAR, further GNSS stations, etc) are needed.

Results – 3.2 GNSS-IR sea level retrievals and its validation
Line 270-271: Are sea-level oscillations tidal or the sea-state? A bit nit-picky, but I wouldn’t use “intriguingly” here, it seems logical with lower sampling rates and greater variation you’d have less resolvability.

Figure 4,5,6: How are the outliers defined?

Line 292-294: This sentence is a bit confusing, rephrase for clarity.

Figure 6: It would be helpful to have all the sea-level time series used in this paper and their statistics shown somewhere (either in chart or plots in the supplement). Does the Kalman filtering include the GNSS-IR outliers?

Results –3.3 GNSS-IR detection on tsunami waves
Line 324: “Amplitudes in the rupture zone” does this indicate slip in the rupture zone, or wave height?

Line 362-363: I disagree. While for case 3 I think the detection capabilities seem evident in figure 8, this is not true for case 1 and 2, where the frequency analysis is indistinguishable from normal tidal variations. Additionally, the wave form also looks like it could just be noise in the range of normal variation of the sea surface elevation. Furthermore, in figure S2, the indicated abnormalities seem quite similar to other variations in the frequency plots. If the authors’ goal was to use this station for tsunami detection, it seems (from visual inspection) like there would be many false positives.

Discussions
Line 373-385 Do you have any hypothesis for the mechanism of this observed signal?

Line 385: I’m unfamiliar with the term “trainset”. Potentially rephrase? Maybe it is a typo of “transient"?

Line 385-391: I disagree. The authors report a velocity on the order of ~8mm/year to the east-northeast, which could be significant, depending on the deformation rates of nearby stations (on Palawan, in Vietnam, elsewhere in the SCS). In fact, most GNSS stations will dominantly show linear rates representative of long-term tectonic motion. Furthermore, the linear trend will vary strongly dependent on the reference frame/which euler pole the station’s velocity is relative to.

Line 397: Again, I disagree that the so-called “flat” displacements are representative of regional stability. To determine that, the authors need to compare their results to nearby GNSS velocity vectors, and also likely need a bit longer time series. 1 year, with a potential transient, is too short. With the authors’ statement in lines 398 to 400, I think they understand this on some level, but should probably remove or tone down the hypothetical implications of a linear trend in a time series, unless they compare their observations to other GNSS velocity estimates around the SCS. Additionally, earthquakes can happen anywhere, claims of “stability” especially using data from one sight is suspect.

Figure 10: Could you plot all three on one plot? The Kalman-filter soothed GNSS-IR, GNSS IR daily averages, and the tide gauge?

Line 460-461: What are these rates representative of? Sea level rise/Relative sea level (to the ground)? Or something else? If this is less than a year’s rate, are there any oceanographic phenomena that might explain an additional order of magnitude in comparison to the long term rates (line 469).

Technical Corrections
Overall the paper is well written in good English, but there are several instances where grammatical mistakes make some sentences a bit confusing and hard to follow, going through the manuscript for grammar would be helpful to future readers. Beyond the few instances mentioned above, here is non-exhaustive list of places where a rewriting for grammar and clarity would be helpful.
Line 58-68: Should be split into multiple sentences

Line 106: “is located”

Line 227-232: Reword for clarity

Line 288: “the same”

Line 289-290: reword for clarity

Line 292: reword for clarity, maybe “the measurements exhibit less scatter in…”

Line 401-404: reword for clarity/shorten

Reply
Citation: https://doi.org/10.5194/egusphere-2025-3563-RC1

Supplement

https://doi.org/10.5194/egusphere-2025-3563-supplement

Viewed

Total article views: 1,194 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
1,119	64	11	1,194	25	13	8

HTML: 1,119
PDF: 64
XML: 11
Total: 1,194
Supplement: 25
BibTeX: 13
EndNote: 8

Views and downloads (calculated since 16 Sep 2025)

Month	HTML	PDF	XML	Total
Sep 2025	1,022	34	4	1,060
Oct 2025	97	30	7	134

Cumulative views and downloads (calculated since 16 Sep 2025)

Month	HTML	PDF	XML	Total
Sep 2025	1,022	34	4	1,060
Oct 2025	97	30	7	134

Viewed (geographical distribution)

Total article views: 1,037 (including HTML, PDF, and XML) Thereof 1,037 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 29 Oct 2025

Short summary

To address critical data gaps on land stability and sea levels in the vast South China Sea, a new GPS station was deployed. It measured small, stable land shifts (<1 cm yearly) and accurately tracked sea-level rise, matching the global average. It detects abnormal sea-level surges from underwater landslides or tsunamis if large enough. Serving as a vital link in a sparse network, this station gives vital data to study regional earth movements, coastal hazards, and sea-level trends.


Total:	0
HTML:	0
PDF:	0
XML:	0