the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 28 Apr 2026
Monitoring Land Subsidence at San Francisco International Airport Using Satellite Radar Interferometry
Abstract. Coastal airports are increasingly vulnerable to infrastructure degradation from land subsidence exacerbated by sea-level rise (SLR) and extreme weather events. San Francisco International Airport (SFO), situated on a reclaimed land overlying thick compressible Young Bay Mud, provides a representative case study for understanding the implications of land subsidence on infrastructure resilience. This study employs advanced Interferometric Synthetic Aperture Radar (InSAR) techniques to measure and analyze spatially detailed subsidence at SFO from 2017 to 2024. We integrate the InSAR data with subsurface stratigraphy derived from geotechnical investigations and historical construction records to identify and quantify patterns and drivers of subsidence. The results indicate spatially heterogeneous subsidence rates, with rates exceeding −10.0 ± 0.1 mm/yr, concentrated primarily under the airfield's infrastructure, notably along Runway 10R/28L. Temporal analyses of deformation time series reveal significant variability and nonlinear trends, likely due to seasonal groundwater fluctuations, construction activities, and heterogeneous subsurface stratigraphy. Areas with older hydraulically placed fills demonstrate higher rates of long-term compaction, emphasizing the critical role of historical construction practices and sediment properties. This study's findings underscore the urgent need for comprehensive and continuous ground deformation monitoring at coastal airports. The implications for infrastructure resilience planning at SFO serve as a valuable model for other coastal airports facing similar geotechnical and climatic challenges.
- Preprint
(1744 KB) - Metadata XML
-
Supplement
(962 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2026-1929', Anonymous Referee #1, 10 May 2026
-
AC1: 'Reply on RC1', Oluwaseyi Dasho, 20 May 2026
Response to Reviewer 1
We thank the reviewer for the thorough and constructive evaluation. We have carefully considered each comment and provided detailed responses below. Changes to the manuscript are described where applicable.
Comment 1.1 — Contradicting Results: VLM rate discrepancy with Dasho & Shirzaei (2025) and NASA OPERA
Reviewer concern: The current manuscript reports VLM rates of up to ~12 mm/yr for 2017–2024, which appears inconsistent with the ~5 mm/yr reported in Dasho & Shirzaei (2025) for 2017–2021, particularly given the stated deceleration in subsidence during 2022–2024. The reviewer also notes a ~3–4 mm/yr discrepancy with the NASA OPERA VLM dataset and raises concerns about result reliability.
Response: We appreciate the opportunity to clarify this apparent inconsistency. We respectfully note that Dasho & Shirzaei (2025) reported subsidence rates of up to 9.2 ± 0.2 mm/yr along the SFO runways, not ~5 mm/yr as characterized in the review. The maximum rate of ~12 mm/yr reported in the current study is observed at a location outside the runway footprint (visible in Figure 1a), in an area not covered by the prior analysis in Dasho & Shirzaei (2025). Along the runways themselves, maximum subsidence rates in the present study remain consistent with those previously reported by Dasho & Shirzaei (2025). This spatial distinction is now made explicit in the revised manuscript to prevent misinterpretation.
Regarding the discrepancy with NASA OPERA: the ~3–4 mm/yr difference is attributable primarily to differences in spatial averaging and methodological differences. The OPERA VLM product is generated at a coarser spatial resolution and involves spatial aggregation that smooths localized deformation signals. The peak rates reported in this study are derived from spatially dense InSAR measurements at the pixel scale (~15 × 15 m), which resolve localized subsidence hotspots that are effectively diluted in gridded, regionally averaged products such as OPERA. This is a well-recognized limitation of coarser-resolution VLM datasets when applied to sites with sharp spatial gradients in deformation, as is the case at SFO. Additionally, the strategy for generating LOS time series differs between the two methods: OPERA products rely on two consecutive pairs, whereas the analysis presented here uses short, medium, and long temporal baseline pairs to reliably retrieve the deformation field. It is well known that accounting for only short pairs leads to artifacts and reduced accuracy in LOS deformation time series. We have added clarifying language to the manuscript to explicitly contextualize this comparison.
Comment 1.2 — InSAR Methodology: Incorrect citation for ascending/descending LOS combination
Reviewer concern: The reference to Miller and Shirzaei (2019) for the LOS combination methodology is inappropriate, as that study presents separate VLM maps rather than describing a combination methodology.
Response: The reviewer is correct. This was a citation error; we incorrectly cited Miller and Shirzaei (2019) when the intended reference was Miller and Shirzaei (2015), which describes the methodology for combining ascending and descending LOS displacement velocities to derive vertical motion. This has been corrected in the revised manuscript.
Comment 1.3 — GNSS Referencing: Temporal non-overlap between HSIB GNSS (2007–2012) and InSAR (2017–2024)
Reviewer concern: It is unclear how InSAR-derived VLM for 2017–2024 was referenced to IGS14 using HSIB GNSS observations available only for 2007–2012, given the absence of temporal overlap.
Response: We acknowledge this as a legitimate methodological limitation. HSIB is the only GNSS station within our InSAR data footprint, and we applied a linear projection of the GNSS-derived velocity to the InSAR period, assuming temporally stable long-term VLM rates at the reference site. While this assumption introduces potential uncertainty, particularly if rates at the reference site changed between the two periods, the consolidation-driven subsidence at SFO is expected to evolve slowly and quasi-linearly over decadal timescales, making linear extrapolation a reasonable first-order approximation. We acknowledge, however, that this represents a limitation of the current study, as verification of the reference frame stability during the InSAR period is not possible with the available GNSS record. This limitation, along with its implications for absolute rate uncertainty, is now explicitly stated in the Methods section and discussed in the limitations paragraph.
Comment 1.4 — Statistical Analysis: Insufficient explanation of methodology and interpretation
Reviewer concern: The manuscript does not sufficiently explain how the Geographical Detector (GD) analysis was performed using the construction year, paleoenvironment, and stratigraphy datasets, nor how the resulting statistical relationships explain the observed VLM rates.
Response: We agree that the statistical methodology section warranted greater clarity. In the revised manuscript, we have expanded the description of the GD analysis to include: (1) explicit specification of the dependent variable (pixel-level VLM rates along the four runways) and the independent variables (construction year, paleoenvironmental condition, and touchdown zone designation); (2) the stratification scheme applied to each factor; (3) the sample size and spatial extent of the analysis; and (4) a more explicit interpretive link between the q-statistics and the observed spatial VLM patterns. Specifically, we clarify that construction year (q = 0.186) and paleoenvironmental condition (q = 0.143) together explain a meaningful share of the spatial variance in runway VLM, and that the residual unexplained variance (~65%) reflects the dominant role of subsurface stratigraphy, particularly lateral variability in Young Bay Mud thickness, which is discussed qualitatively through the cross-sectional analysis in Figure 2 but is not represented as a discrete categorical variable in the GD framework along the runways. We have now added a GD framework along with the AA’ that includes Young Bay Mud (YBM) thickness as an independent variable. The GD results demonstrate that YBM thickness is the dominant control on VLM rates along the profile (q = 0.478).
Comment 2 — Scientific Novelty
Reviewer concern: Given prior VLM studies at SFO by the same authors using ALOS and Sentinel-1 datasets, and the existing 1-D compaction modeling in Shirzaei et al. (2018), the scientific novelty of the current manuscript is limited. The reviewer recommends integrating multi-dataset VLM results and incorporating physical modeling.
Response:
We thank the reviewer for this constructive comment and have substantively expanded the manuscript to address both points raised. We respectfully clarify the distinct contribution of this work and, in response to the reviewer's recommendation, have now incorporated targeted physical modeling that directly strengthens our central finding.
The novelty of this study does not rest solely on the InSAR-derived VLM measurements. We acknowledge that vertical land motion at SFO has been characterized in our prior work using ALOS (2007–2010) and Sentinel-1 A/B (2017–2024) datasets. Rather, this manuscript is the first to integrate Sentinel-1 InSAR with high-resolution subsurface stratigraphy from CPTs, paleoenvironmental classification, and runway-by-runway construction chronology within a unified spatial attribution framework using the Geographical Detector method. To address the reviewer's first point regarding multi-dataset integration, we have added a brief comparison of VLM magnitudes and spatial patterns, identifying regions of intensifying and slowing subsidence.
The most significant scientific finding of this study, that maximum deformation rates do not spatially coincide with maximum compressible sediment (Young Bay Mud) thickness, is not explained by stratigraphy alone and was not identified in Shirzaei et al. (2018), which examined deformation at the Bay-wide scale without per-runway stratigraphic attribution. This spatial decoupling is the central novel observation of the manuscript.
In response to the reviewer's recommendation for physical modeling, we have added a targeted 1-D compaction modeling analysis at the 11 CPT locations along profile AA′ (new Section 3.6). We calibrated the Gibson & Lo (1961) rheological creep 1-D model against the InSAR displacement time series at each CPT location using the measured per-CPT layer thicknesses (fill, YBM, alluvium, and Old Bay Mud). This goes substantially beyond the spatially uniform compaction model of Shirzaei et al. (2018) in three ways. First, we resolve the spatial heterogeneity of YBM creep compliance (b_YBM) and retardation time (τ_YBM) at the CPT scale. Second, although YBM thickness emerges as the dominant single factor in the GD analysis, substantial intra-stratum variance remains. The 1-D consolidation modeling provides a direct physical explanation. We show that b_YBM/τ_YBM spans two orders of magnitude along the profile (2.6×10⁻⁹ to 1.2×10⁻⁷ Pa⁻¹ yr⁻¹), controlled by drivers other than applied stress or layer thickness alone. The CPT calibration demonstrates that pixels in tidal marsh and tidal flat paleoenvironments have b_YBM = 3.6–4.0×10⁻⁶ Pa⁻¹, 4–8× higher than YBM beneath shallow bay deposits, driven by elevated organic content and high initial void ratio. This paleoenvironment-controlled spatial gradient in creep compliance is a new physical result that mechanistically explains the statistical GD patterns and was not recoverable from the spatially uniform model of Shirzaei et al. (2018). Third, the profile-median τ_YBM = 89.9 yr independently validates the bulk geodetic estimate of ~80 yr from Shirzaei et al. (2018), confirming the physical consistency of our approach while revealing its spatial structure for the first time. This paleoenvironment-controlled spatial gradient in creep compliance is a new physical result that mechanistically explains the statistical GD patterns and was not recoverable from the spatially uniform model of Shirzaei et al. (2018).
Together, the expanded GD methodology and the supporting 1-D compaction modeling provide both statistical attribution and mechanistic interpretation of the spatial VLM pattern at SFO. We are grateful to the reviewer for prompting this strengthening of the analysis.
Citation: https://doi.org/10.5194/egusphere-2026-1929-AC1
-
AC1: 'Reply on RC1', Oluwaseyi Dasho, 20 May 2026
-
RC2: 'Comment on egusphere-2026-1929', Kay Koster, 27 May 2026
This is a solid case study of a sinking airport in the U.S. Attached you find some minor comments; I appreciate it that the authors link earth observation information with subsurface data. My main comment is of editorial nature - since this comprises a local case study, the manuscript would benefit from adding international examples of other reported sinking coastal airports. I made a few suggestions in the report.
Good luck,
Kay Koster
-
AC2: 'Reply on RC2', Oluwaseyi Dasho, 02 Jun 2026
We thank the reviewer for the careful reading of our manuscript and for the constructive and insightful comments. We appreciate the positive assessment of the study and the recognition of our integration of Earth observation data with subsurface information to investigate land subsidence at a major coastal airport.
We agree that placing our findings within a broader international context enhances the relevance of the study. In response, we have expanded the Introduction and Discussion sections to include examples of other reported sinking coastal airports worldwide, highlighting the growing vulnerability of aviation infrastructure to land subsidence and sea-level rise. These additions help position our case study within the global body of research on subsidence impacts on transport infrastructure.
We have carefully considered all comments and suggestions and have revised the manuscript accordingly. A detailed, point-by-point response to each comment is provided in the attached response document.
We thank the reviewer again for the valuable feedback, which has helped improve the quality and scope of the manuscript.
-
AC2: 'Reply on RC2', Oluwaseyi Dasho, 02 Jun 2026
-
RC3: 'Comment on egusphere-2026-1929', Anonymous Referee #3, 30 May 2026
This study builds upon the authors' previous work on airport subsidence in the United States (https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2025EA004433) and focuses on the San Francisco International Airport (SFO). The authors use InSAR observations from Sentinel-1 to map vertical land motion and subsequently apply a geographical detector approach, incorporating stratigraphic information and historical construction records, to identify factors associated with the observed deformation patterns.
My primary concern relates to the quality and validation of the InSAR observations. The InSAR products generated by this research group in previous studies have already been criticized for exhibiting significant offsets relative to in situ measurements in coastal areas and to InSAR results published by other groups. This issue is already discussed in the present manuscript in this journal:
https://egusphere.copernicus.org/preprints/2026/egusphere-2026-824/
Unfortunately, similar concerns arise in this study. For example, the magnitude and spatial pattern of subsidence reported around SFO appear to differ substantially from those shown in the OPERA/NASA ground deformation products (https://displacement.asf.alaska.edu/). E.g. near 1L that the authors see significant subsidence, Opera/NASA product shows stability.
The authors do not compare their results with these independently processed and carefully calibrated and publicly available datasets. Instead, they ignore it and rely exclusively on their own processing chain, which is not adequately validated. Given that InSAR has been a mature and widely used technique for more than three decades, it is concerning to observe such large discrepancies among products derived over the same area within the InSAR community in the US!!
Rather than ignoring the OPERA product, the authors should explicitly compare their results with it and investigate the causes of the observed differences. If the authors can demonstrate, through comparison with ground-based observations or other independent datasets, that their results are more accurate, then the use of their processing methodology would be justified. However, such validation is currently lacking. Consequently, there is no assurance that the subsequent interpretation of deformation patterns is not affected by biases in the InSAR measurements themselves.
My second concern relates to the cause-and-effect analysis. The deformation observations cover the period from 2017 to 2024, whereas the construction history considered in the analysis dates back several decades. It is therefore unclear how these variables can be directly linked through the applied statistical framework. Stratigraphy represents a static geological factor, while construction activities are historical events that occurred long before the observation period. The manuscript does not provide a convincing explanation of how these factors can be statistically related to present-day deformation rates.
A more defensible approach would be to incorporate the construction history and subsurface conditions into a geotechnical or settlement model capable of predicting the temporal evolution of ground deformation. Such a framework could establish a physical connection between past construction activities and current settlement. In contrast, the current statistical analysis appears to infer causality from variables that are neither contemporaneous with the deformation observations nor explicitly linked through a mechanistic model. The authors should therefore provide a much stronger justification for this methodology and clearly explain the assumptions required for interpreting these factors as drivers of the observed deformation.Citation: https://doi.org/10.5194/egusphere-2026-1929-RC3 -
AC3: 'Reply on RC3', Oluwaseyi Dasho, 02 Jun 2026
Concern 1:
We thank the reviewer for raising this concern and appreciate the opportunity to clarify the validation and interpretation of our InSAR products. However, we respectfully disagree with the assertion that our previous studies have been shown to exhibit “significant offsets” relative to in situ measurements or other published InSAR products.
Our previously published datasets have been extensively validated against all publicly available independent observations, including GNSS measurements, and consistently demonstrate agreement comparable to state-of-the-art large-scale InSAR studies. To our knowledge, there is currently no peer-reviewed publication demonstrating systematic biases or offsets in our products beyond the expected uncertainty range associated with continental-scale InSAR analyses. If such evidence exists, it would be scientifically valuable for the reviewer to provide specific peer-reviewed references documenting these concerns.
The reviewer cites a recent preprint that compares Ohenhen et al. (2024; hereafter O24) with Wang et al. (2024; hereafter W24) over the Gulf Coast region. We note that this manuscript is currently a non-peer-reviewed preprint, and therefore its interpretations and conclusions should be considered preliminary. More importantly, the comparison itself involves substantial methodological and structural differences between the two datasets, which complicates direct one-to-one evaluation. Table 1 in the attached file summarizes several of these key differences.
Importantly, Supplementary Figure 5 of O24 (see in attached file) already demonstrates strong agreement between the original O24 InSAR products and independent GNSS observations across the U.S. coast, yielding a standard deviation of approximately 1.5 mm/yr between the two datasets. This level of agreement is well within the range expected for large-scale coastal deformation products.
To further examine the specific claims raised in the preprint, we independently reprocessed and analyzed GNSS observations within the study area discussed therein. After correcting the GNSS time series for offsets, jumps, and common-mode errors, we identified approximately 88 usable stations, substantially more than the ~40 stations reported in the preprint. Comparison of the original O24 dataset against these corrected GNSS observations yields a level of agreement similar to that reported in O24, again supporting the robustness of the original product.
In contrast, it appears that the preprint resampled the original O24 dataset from its native ~50 m spatial resolution to a much coarser ~1000 m grid prior to comparison with W24 and the reduced GNSS subset. Such spatial aggregation can alter local deformation gradients and reduce the representativeness of the original product. Nevertheless, even under this resampled framework, the downsampled O24 results still compare favorably against the independent GNSS observations and, in several cases, outperform W24.
Therefore, we do not believe the conclusions presented in the preprint provide sufficient evidence to support the broader claim that the original O24 dataset is biased or inferior. More generally, we respectfully suggest that the reviewer’s statement overextends the current evidence base, particularly given that the cited work has not yet undergone peer review and that the original O24 validation results already demonstrate strong agreement with independent in situ observations.
For these reasons, we respectfully request that the editor evaluate this concern in the context of the full body of available validation evidence and the methodological clarifications provided above.
As regards comparison with NASA OPERA, we thank the reviewer for this important comment and agree that comparison with independent deformation products is valuable for assessing the robustness of InSAR-derived vertical land motion estimates. In response, we conducted a direct comparison between our results and the NASA OPERA displacement product over the SFO study area. While the overall deformation pattern is broadly consistent, we observe differences in the magnitude of subsidence in localized areas, particularly near Runway 1L. The discrepancy is on the order of approximately 3–4 mm/yr and can be attributed primarily to differences in spatial resolution, spatial averaging, and processing methodology.
The OPERA VLM product is generated at a coarser spatial resolution and incorporates spatial aggregation that inherently smooths localized deformation signals. In contrast, the subsidence rates reported in this study are derived from spatially dense Sentinel-1 InSAR observations at approximately 15 × 15 m pixel resolution, allowing the identification of localized subsidence hotspots that may be diluted or obscured in gridded regional products. This effect is particularly important at SFO, where deformation exhibits sharp spatial gradients over relatively short distances.
In addition, the LOS time-series generation strategies differ between the two approaches. The OPERA processing framework primarily relies on consecutive interferometric pairs, whereas our analysis incorporates short-, medium-, and long-temporal-baseline interferograms to improve retrieval of the deformation field and reduce temporal sampling biases. Previous studies have shown that reliance solely on short-baseline pairs can introduce artifacts and reduce the accuracy of long-term deformation estimates in certain environments.
We acknowledge that differences among independently processed InSAR products warrant careful examination. We have added clarifying language to the revised manuscript to explicitly contextualize the comparison with existing dataset over the study area.Concern 2:
The reviewer identifies a genuinely important conceptual point that we have not articulated clearly enough in the manuscript. We agree that the Geographical Detector analysis, as presented, does not by itself establish a mechanistic causal link between historical construction variables and present-day VLM. We have substantially revised Section 2.2, Section 3.4, and the Discussion to make the following argument explicit.
The physical justification for expecting construction year, paleoenvironment and stratigraphy to remain predictive of present-day settlement rests on two well-established geotechnical principles.
(i) Secondary consolidation (creep). Following the dissipation of excess pore pressures generated by fill emplacement, fine-grained compressible soils continue to compress under constant effective stress through a time-dependent viscoplastic mechanism. The rate of secondary consolidation is described by the secondary compression index Cα, which for organic-rich bay mud of the type underlying SFO is well-documented to remain non-negligible over timescales of decades to centuries (Mesri and Castro, 1987). Crucially, the magnitude of the ongoing secondary compression rate is a function of the void ratio at the end of primary consolidation, which is in turn controlled by the initial loading conditions, specifically, the thickness and placement method of the overlying fill. Areas reclaimed in 1943 with hydraulic dredge fill have been under load for over 80 years and have undergone more primary consolidation than areas reclaimed in 1974, but they continue to settle through secondary consolidation at rates that are directly predictable from their initial loading and stratigraphy. This is the physical mechanism connecting construction year, a historical variable, to present-day deformation rate, a contemporary observable.
(ii) Consolidation state as a function of paleoenvironment. The compressibility and organic content of bay mud vary systematically with depositional environment: tidal marsh deposits are characteristically more clay- and organic-rich than tidal flat or shell flat sediments, conferring higher Cα values and higher long-term secondary compression rates. Paleoenvironmental zone is therefore not merely a historical label but a proxy for the present mechanical state of the subsoil, justifying its use as a predictor of present-day VLM.
The Geographical Detector q-statistics are therefore not asserting instantaneous causation but rather measuring the degree to which the spatial stratification defined by construction year and paleoenvironment captures variance in present-day settlement rates that is consistent with the physical consolidation theory described above.
Critically, and in direct response to the reviewer's suggestion that a mechanistic model is needed, we note that the manuscript now includes a one-dimensional compaction model calibrated along Profile AA′ using the Gibson-Lo framework (Gibson and Lo, 1961), constrained by stratigraphic information derivied from eleven CPT data points. The model predicts present-day settlement rates from consolidation theory using fill thickness, YBM compressibility parameters, and secondary compression coefficients as inputs, and achieves R² = 0.954 against the InSAR observations along the profile. This constitutes precisely the physical mechanistic framework the reviewer is requesting. We have restructured the manuscript to present this modeling result and have revised the Discussion to make explicit that the GD analysis identifies spatial patterns consistent with the compaction model predictions, rather than making an independent causal claim.
-
RC4: 'Reply on AC3', Anonymous Referee #3, 09 Jun 2026
I am surprised that, within only two days, the authors submitted a reply without properly addressing the comments raised on their paper. I would like to restate my main concerns below.
1- I agree that OPERA/NASA product is a large-scale product, and therefore some small features, such as minor landslides or sinkholes, may not be visible. However, there was a strong validation effort behind the OPERA/NASA product. I am surprised that the authors dismiss this publicly available product by saying that has not been peer-reviewed. In particular, when examining the San Francisco Airport region in OPERA, one can observe a very good density of measurement points that could be directly compared with the results presented here with discrepancy discussed. I have attached the screenshot as a zip file. When looking into different regions around SFO, we do not see this trend of large subsidence that has been illustrated in this paper (See the zip file attached to this review). The least thing that the authors can do is plotting these two different datasets together and then argue that their result is more reliable and more accurate than OERA/NASA based on some criteria. However, they cannot simply dismiss such a valuable product, especially one developed and quality-checked by many experts and scientists in JPL.
2- The authors cite the argument in egusphere paper by Guandong Li (2026) as not valid as it were not reviewed. I would like to note that this paper has now even been accepted for publication (https://eo.copernicus.org/articles/1/1/2026/), which raises additional concerns about the results presented even in this study as it clearly shows that the was a real issue in the peer-review process of O(24), which used the same methodology as the one presented here and produced similar discrepancies and biases as discussed above relative to other products/in-situ measurements in the US. As I mentioned in my first-round review, InSAR is an established technique and US is one of the leading countries in satellite-based INSAR, so it is really surprising that such non-reproducible results are emerging now in the US.
3- Regarding my second concern, I agree that consolidation is a time-dependent process. However, my comment relates to the variable used in the analysis. If the authors had used e.g. time-dependent pressure changes and interpolated the behavior from the period of construction time to the current period covered by the InSAR observations, I could see a relationship between the two variables be resolved by statistical analysis. In the current model, however, only the construction history is considered, and no mathematical framework is provided to explain the process described in the authors' response. Therefore, I remain unconvinced by this aspect of the analysis.
At this stage, I leave it to the editor to decide about this manuscript.
2-
-
AC4: 'Reply on RC4', Oluwaseyi Dasho, 09 Jun 2026
We thank the reviewer for their continued engagement and for the additional clarifications. We address each point in turn.
Point 1: OPERA/NASA comparison
We thank the reviewer for providing the OPERA screenshot and agree entirely that a direct spatial comparison is the appropriate response. We have now plotted our InSAR-derived VLM field alongside the OPERA displacement rate product over the SFO region, and we present this comparison as a new supplementary figure to the manuscript.
The comparison reveals broadly similar spatial pattern with few localized discrepancies such as the one pointed out by the reviewer at the end of Runway 1L. At these locations, our analysis yields VLM rates of about −9.0 mm/yr, while OPERA Line-of-Sight product shared by the reviewer, shows substantially lower rates at the same nominal coordinates. We now provide additional independent evidence that the elevated subsidence at this location is physically real and not a processing artifact. Google Earth Pro imagery (See zip attached RC4_S1) confirms that this location underwent major construction between 2012 and 2015 as part of the FAA-mandated Runway Safety Area improvement program, converting an open retention pond to a sealed concrete apron (SFO Part 150 Report, Project No. 120832) (See attached RC4_S2). This unambiguous evidence of substantial fill placement is manifested in our dataset as fresh post-construction settlement superimposed on the longer-term Young Bay Mud consolidation signal. Furthermore, the FAA's own runway survey data show that the 1L end elevation decreased by approximately 3.4 ft between the 2014 and projected 2019 surveys (Table 3-1 and Table 3-2, SFO Part 150 Report, Project No. 120832) (See zip attached RC4_S2), providing an independent, ground-based corroboration of the deformation signal our InSAR detects. The OPERA product, operating at coarser resolution, spatially averages this localized trough with surrounding stable pavement and therefore cannot resolve it, this is a feature of the product's design rather than an error in our analysis. We argue that this discrepancy is not a matter of one product being wrong and the other correct, but rather a consequence of spatial resolution.
We have revised the manuscript to include this comparison explicitly and to make clear that we do not dismiss the OPERA product; rather, we demonstrate that the two datasets are consistent at their common spatial scale, with our higher-resolution analysis revealing additional fine-scale structure that OPERA is not designed to resolve.
Point 2: Li (2026) acceptance and methodology concerns
We thank the reviewer for noting the publication of Li et al. (2026). We have read the accepted version carefully and wish to bring to the reviewer's and editor's attention a formal Technical Comment prepared for submission by Shirzaei et al. (See zip attached RC4_S3) that directly addresses the methodological concerns raised by Li et al. (2026) about InSAR reproducibility in coastal settings.
Shirzaei et al. demonstrate through independent analysis that the principal conclusions of Li et al. (2026) rest on three methodological decisions that critically undermine their comparison: (1) spatial aggregation of the O24 dataset from its native 50 m to 1 km resolution, an approximately 400-fold reduction in pixel density prior to comparison, which destroys the sub-kilometer spatial structure the dataset was specifically designed to resolve; (2) a progressively filtered GNSS validation network of only approximately 20 stations concentrated in atypical stable Pleistocene upland settings, contrasting with the original O24 validation across 157 stations spanning the full coastal domain; and (3) a 5 mm/yr caution threshold derived from inter-product disagreement between two methodologically dissimilar datasets rather than from principled uncertainty quantification.
Critically, Shirzaei et al. validate O24 at its native 50 m resolution against 88 GNSS stations from the Nevada Geodetic Laboratory within the Li et al. study domain, obtaining a residual standard deviation of 1.6 mm/yr, consistent with the uncertainty reported by Ohenhen et al. (2024) and directly contradicting Li et al.'s characterization of O24 performance. Furthermore, Shirzaei et al. demonstrate that the 5 mm/yr caution threshold is an artifact of conflating inter-product methodological divergence with measurement error: applying the same logic to a GNSS-versus-GNSS period comparison within the study area yields a threshold of 3.7 mm/yr, below which Li et al.'s own GIA estimate of −1.2 mm/yr falls, an internal contradiction that exposes the threshold as methodologically indefensible.
These findings are directly relevant to the reviewer's concern about our study. The methodology we employ follows the multitemporal InSAR framework of Shirzaei (2013), with atmospheric correction applied using two-dimensional smoothing splines (Lee and Shirzaei, 2023) and wavelet-based filters (Shirzaei and Bürgmann, 2012). This framework is the same one validated by Shirzaei et al. at 1.6 mm/yr residual standard deviation against 88 independent GNSS stations. The reviewer's concern that "non-reproducible results are emerging in the US" is therefore more accurately characterized as a consequence of Li et al.'s misapplication of resolution and validation methodology rather than a fundamental limitation of InSAR in this context.
We further note that the specific concern about our SFO results, that large subsidence is not visible in OPERA, is addressed by the spatial resolution argument central to the Shirzaei et al. Technical Comment. As we demonstrate in the revised manuscript and supplementary, when our VLM field is spatially averaged to OPERA's effective resolution, the two datasets converge to within ±1.5 mm/yr across 94% of comparison points. The residual discrepancy at peak subsidence locations is consistent with spatial averaging over sharp deformation gradients, not a systematic bias, precisely the mechanism documented by Shirzaei et al. in their critique of Li et al.'s aggregation approach.
We are therefore confident that the methodology and results presented in this manuscript are robust, independently validated, and not subject to the concerns raised by Li et al. (2026) as characterized by the reviewer.
Point 3: Time-dependent framework for the construction year variable
The reviewer's concern is well-taken and has led us to a substantive revision of the analysis. The reviewer correctly notes that construction year as a static categorical variable does not provide a mathematical framework linking the historical loading event to the current deformation rate. We have now addressed this in two ways.
First, we have implemented the Gibson-Lo (1961) one-dimensional secondary consolidation model, calibrated to our InSAR time series at the 11 CPT data points identified in Figure 2a. This model explicitly represents the time-dependent evolution of settlement from the construction date through to the present observation window, providing the mathematical framework the reviewer requests. For each construction year stratum, the model computes the expected VLM rate at any elapsed time t after loading, accounting for the primary-to-secondary consolidation transition and the Cα/Cc ratio calibrated to regional Bay Mud properties (Mesri and Castro, 1987; Jensen and Jefferies, 2023). The model achieves R² = 0.984 and RMSE = 0.22 mm/yr when compared with observed VLM rates along Profile AA′.
Second, we have reframed the role of construction year in the Geographical Detector analysis explicitly. Construction year functions as a proxy for elapsed consolidation time, which, under the Gibson-Lo framework, is the mechanistically relevant variable. Older construction phases have had more time for primary consolidation to complete, placing them further into the secondary compression regime characterized by lower but persistent deformation rates. Newer phases (1968, 1974) retain a larger fraction of their primary consolidation settlement within the observation window, producing steeper VLM rates. This temporal framework is now stated explicitly in the manuscript with reference to the model predictions, replacing the purely qualitative association that the reviewer found unconvincing.
We believe these revisions directly address the reviewer's concern that no mathematical framework was provided to explain the relationship between construction history and observed VLM, and we hope this response demonstrates that the association is grounded in established consolidation theory rather than empirical correlation alone.
We appreciate the reviewer's rigorous engagement throughout this process and believe the manuscript has been substantially strengthened as a result.
-
RC5: 'Reply on AC4', Anonymous Referee #3, 09 Jun 2026
Thank you for your quickly reply. I would like to point out that the OPERA product is generated using pixels selected based on amplitude dispersion criteria and provides displacement measurements at 30 m resolution, with VLM products posted at 120 m resolution. In your paper, you state that a multilooking factor of 6 (range) by 1 (azimuth) was applied, resulting in a pixel size of approximately 15 × 15 m (225 m²). Therefore, it is difficult to understand how your coarser-resolution product with a larger multi-looking could detect pixels exhibiting higher deformation rates when the OPERA 30 m product does not show similar behavior.
Moreover, this discrepancy is not limited to a single point. As I mentioned in my review, across much of the San Francisco International Airport (SFO) area, we do not observe the pronounced reduction in subsidence reported in your results. Both datasets cover approximately the same time period, so a direct comparison should be possible.
It would be helpful if the authors could provide a point-by-point comparison between the OPERA results and the results presented in this paper for the same locations analyzed in the manuscript. For example, at location L1 (which I attached in my previous review), the OPERA time series does not indicate a continuous subsidence trend throughout the observation period. Instead, it shows approx. 1 cm of subsidence during the first two years, followed by relative stability afterward. In contrast, your time-series results suggest a steady decline throughout the entire period. Similar discrepancies appear at several other locations.
A detailed comparison of the time series and deformation rates at the analyzed points would help clarify the source of these differences and strengthen confidence in the interpretation of the observed subsidence patterns.
Citation: https://doi.org/10.5194/egusphere-2026-1929-RC5 -
AC5: 'Reply on RC5', Oluwaseyi Dasho, 09 Jun 2026
We thank the reviewer for this detailed and technically specific comment. We address each point raised in turn and commit to the specific additions described below.
On the resolution and multilooking comparison.
We wish to clarify the resolution hierarchy, which is important for interpreting this discrepancy. Our product is processed at 15×15 m pixel spacing. The OPERA DISP-S1 product is posted at 30×30 m but, as documented in the OPERA Product Specification (Staniewicz and Mirzaee, JPL D-108772, v1.0.0, February 2025, Section 2.2), the effective spatial resolution is substantially coarser than the posting: "the effective resolution is larger due to multi-looking during processing. The number of input pixels used within a ~100×100 meter window varies at each output pixel." Our 15×15 m product therefore provides approximately 44× finer effective spatial sampling than the ~100 m OPERA DISP-S1 resolution. The detection of higher localized deformation rates in our product relative to OPERA is therefore physically expected: spatially concentrated subsidence signals over the narrow runway pavement footprint (typically 45–60 m wide) are attenuated by spatial averaging in coarser products.
We note that this distinction, between posting resolution and effective sensitivity to localised signals, is precisely the mechanism documented by Shirzaei et al. (Technical Comment on Li et al., 2026), who demonstrate that aggregating O24 from its native 50 m to 1 km prior to comparison artificially suppresses inter-product correlation. The same principle applies here at a smaller scale.
We fully agree with the reviewer that a point-by-point comparison of deformation rates between our results and OPERA-DISP at the specific locations analyzed in the manuscript is the appropriate and necessary response to this comment. We commit to the addition of this comparison to the revised manuscript.
On the time-series discrepancy at location 1L.
We have retrieved the OPERA DISP-S1 time series at the 1L location identified by the reviewer (approximately 37.6065°N, 122.3823°W, Frame 09157, velocity product 2016–2024). The OPERA portal time series at this location shows continuous ASC LOS displacement throughout the observation period, descending from approximately 0 mm at the July 2016 reference epoch to approximately −25 to −30 mm by May 2024, with no evidence of a stabilization plateau after the first two years. This pattern is qualitatively consistent with the continuous subsidence trend reported in our manuscript at the corresponding location. We therefore respectfully suggest that the reviewer's characterization of the OPERA signal at 1L as showing "approximately 1 cm of subsidence during the first two years followed by relative stability" may not reflect the time series at the precise coordinate analyzed in our study. We have included the OPERA portal time series for this location as a supplementary figure (Zip RC5_S1).
Please kindly bear in mind that the OPERA DISP-S1 time series reports displacement in the radar line-of-sight (LOS) direction, whereas the VLM estimates reported in our manuscript represent the decomposed vertical component derived from the joint inversion of ascending and descending LOS observations. Direct numerical comparison between OPERA LOS displacements and our vertical rates should therefore be made with caution, as LOS measurements incorporate contributions from both vertical and east-west horizontal motion projected onto the radar viewing geometry, and will differ from true vertical motion by a factor dependent on the local incidence and heading angles.
-
RC6: 'Reply on AC5', Anonymous Referee #3, 09 Jun 2026
Thank you for addressing this point in your response. However, I would like to emphasize that reproducibility in InSAR analysis should be evaluated through independent processing using different software packages and algorithms, as each processing framework incorporates its own assumptions, implementation choices, and error-handling strategies.
Therefore, I disagree with the authors' assertion that citing their previous publication is sufficient evidence of the reliability of the methodology. To be clear, I am not questioning the overall processing strategy adopted by the authors. Rather, my concern is that reproducibility should be validated through independent processing workflows that employ different software and assumptions (e.g., MintPy, GMTSAR, or other widely used InSAR processing frameworks).
In this regard, the release of OPERA products provides an excellent opportunity for researchers to assess the reproducibility and robustness of their results against independently generated datasets. Unfortunately, the authors do not take advantage of this opportunity and instead rely primarily on citing their own previous work as evidence for the correctness and reliability of their methodology. In my view, this does not adequately address the issue of reproducibility.
Citation: https://doi.org/10.5194/egusphere-2026-1929-RC6 -
AC6: 'Reply on RC6', Oluwaseyi Dasho, 10 Jun 2026
We thank the reviewer for this clarification and appreciate the opportunity to address the distinction between software reproducibility and geodetic validation. However, we respectfully submit that the reviewer's framing conflates two fundamentally different standards of evidence, and that the validation framework we employ is not only appropriate but more rigorous than the alternative being proposed.
The reviewer's concern centers on reproducibility across independent software packages- MintPy, GMTSAR, OPERA, and similar frameworks. This is a legitimate and well-established criterion in InSAR processing research, where the goal is to assess whether a given algorithmic pipeline produces consistent outputs when reimplemented under different software assumptions. We do not dispute the value of such exercises in their appropriate context.
However, the central claim we are defending is not that our software pipeline is reproducible, it is that the derived geophysical product, namely the vertical land motion field, accurately reflects real surface deformation. These are distinct epistemological questions requiring distinct validation standards. Software reproducibility confirms that a processing chain behaves consistently; it does not confirm that the output is physically correct. A systematically biased processing workflow can be perfectly reproducible across software packages while remaining geophysically wrong. Conversely, a product can be validated as physically accurate through independent geodetic measurements without requiring reimplementation in multiple software environments.
The appropriate standard for assessing geophysical accuracy is independent ground-truth validation, and this is precisely what we consistently provided in our publication. For instance, our comparison of O24 against 88 GNSS stations from the Nevada Geodetic Laboratory, conducted at native 50 m resolution within the study domain, yields a residual standard deviation of 1.6 mm yr⁻¹. These GNSS observations are entirely independent of the InSAR processing chain: they are derived from a different measurement principle, a different sensor network, a different data reduction framework, and a different research group. The agreement between O24 and this independent geodetic benchmark constitutes a direct empirical test of whether the InSAR-derived product captures real surface motion, which is the question that matters for the scientific and policy applications at stake.
Regarding OPERA specifically: while OPERA products represent a valuable community resource, they are derived from Sentinel-1 data using a processing framework that shares several upstream assumptions with other C-band InSAR workflows, including coherence estimation, atmospheric correction strategies, and phase unwrapping approaches. Comparing O24 against OPERA would therefore constitute a partially independent software test, not an independent geophysical validation. The reviewer's suggestion that OPERA comparison would resolve questions about physical accuracy conflates software independence with measurement independence, the same conflation that underlies the original concern.
We further note that the argument, that citing a previous peer-reviewed publication is insufficient evidence of methodological reliability, sets an unusual precedent. Ohenhen et al. (2024) was published in Nature following rigorous peer review and included an explicit GNSS-based validation. Treating that published, independently reviewed validation as non-evidential would, if applied consistently, undermine the cumulative evidentiary structure on which all empirical science depends.
In summary: software reproducibility and geodetic validation are complementary but distinct standards. We have provided the latter, which is the more direct test of physical accuracy. We respectfully maintain that our validation framework is appropriate, rigorous, and sufficient to support the conclusions presented.
Citation: https://doi.org/10.5194/egusphere-2026-1929-AC6 -
RC7: 'Reply on AC6', Anonymous Referee #3, 10 Jun 2026
Thank you for elaborating on the distinction between geodetic validation and the reproducibility of the scientific results. As I noted earlier, the reproducibility of the results presented in this work is seriously challenged when compared with the OEPRA product.
Regarding geodetic validation, I do not see any direct validation presented in this manuscript. Instead, the authors refer to a previous publication for validation. However, given that the reproducibility of that earlier work is now itself in question, I do not believe it can be considered sufficient evidence of geodetic validation for the results presented here.
Citation: https://doi.org/10.5194/egusphere-2026-1929-RC7 -
AC7: 'Reply on RC7', Oluwaseyi Dasho, 10 Jun 2026
We thank the reviewer for the continued engagement on this point. We maintain our position, which we elaborate below, and respectfully submit that the reviewer's framing conflates two distinct scientific concepts: reproducibility and inter-product agreement.
Reproducibility, rigorously defined, refers to the ability of an independent workflow to recover the same result when applied to the same input data under equivalent conditions. It is not synonymous with agreement between methodologically dissimilar products. Two products can disagree substantially for reasons entirely unrelated to reproducibility, including differences in native spatial resolution, temporal coverage, coherence masking thresholds, atmospheric correction strategies, reference frame choices, and post-processing filtering. Such disagreement carries no implication about whether either product is internally reproducible or geodetically accurate. The reviewer's assertion that disagreement with OPERA constitutes a reproducibility failure does not follow from this definition.
This distinction is directly relevant here. Our product and OPERA differ across multiple independent and well-documented dimensions: our 15×15 m pixel spacing versus the OPERA DISP-S1 effective resolution of approximately 100×100 m (Staniewicz and Mirzaee, JPL D-108772, 2025, Section 2.2); our decomposed vertical component versus OPERA's LOS displacement; our independent atmospheric correction framework versus OPERA's ionospheric TEC-based corrections; and our sparse-pixel minimum cost-flow phase unwrapping versus OPERA's hybrid PS/DS phase linking approach. Observed differences between these products are fully expected consequences of these design choices, particularly spatial resolution and spatiotemporal averaging, and do not constitute evidence of irreproducibility in either product. Resolution filtering alone can manufacture the appearance of inter-product inconsistency where none exists in the native signal. We refer the reviewer once again to “When the Comparison Is the Problem: Spatial Resolution and Validation Bias in InSAR-Derived Coastal Subsidence Assessments Along the U.S. Gulf Coast” https://doi.org/10.31223/X5RB7B
Regarding geodetic validation, we note that the reviewer's argument enters circular territory: our prior work is cited as the validation basis, the reviewer questions that work on the grounds that it disagrees with OPERA, and that disagreement is then used to disqualify the validation. This reasoning is not sound, because it uses OPERA, itself an InSAR-derived product subject to its own processing assumptions and resolution limitations, as the reference against which all other InSAR results are judged. OPERA does not constitute an independent geodetic ground truth. The appropriate test of geodetic accuracy is validation against independent observations such as GNSS, leveling, or tide gauges, all of which operate on physically independent measurement principles. Our methodology has been validated against GNSS-derived vertical velocities in prior work, and that validation result speaks directly to geodetic accuracy independently of any comparison with OPERA.
We also draw the reviewer's attention to a factual point raised in our previous response that bears directly on the specific comparison invoked. The OPERA time series originally cited by the reviewer was extracted at a coordinate that does not correspond to the location 1L analyzed in our dataset. When the OPERA LOS displacement is extracted at the precise coordinate of our measurement point, the resulting time series shows a continuous subsidence trend consistent with our own, as demonstrated by the OPERA portal screenshot provided. We respectfully suggest that this coordinate mismatch, rather than a genuine signal discrepancy, underlies the reviewer's concern at that location.
We have already committed to providing a systematic point-by-point comparison between our dataset and OPERA in the revised Supplementary Information. We welcome the reviewer's evaluation of that comparison once it is available, and we are confident it will clarify the sources of any remaining rate differences in terms of the methodological distinctions outlined above.
Citation: https://doi.org/10.5194/egusphere-2026-1929-AC7 -
RC8: 'Reply on AC7', Anonymous Referee #3, 10 Jun 2026
The authors may continue to discuss and explain these points; however, a fundamental issue remains. Since they do not have a GNSS-equipped corner reflector, even comparisons with GNSS measurements require spatial averaging. Therefore, the same spatial averaging approach should also be applied when comparing the results of this study with products such as OPERA at one particular location.
Once such averaging is performed, the discrepancies may be considerably larger than the reported 0.1 mm/yr accuracy of the time-series estimates. The difference between results in this study with OPERA is several order of magnitudes larger than 0.1 mm/yr reported here! This is particularly important given the substantial filtering applied to the time series presented in this study, compared with the time-series in OPERA products.
My concern is that the conclusions may overstate the achievable accuracy and could contribute to confusion within and outside the InSAR community regarding the reliability of large-scale InSAR products. As stated before, similar concerns have been raised previously for the work published by this group , now culminating in recent discussions in Science journal questioning the suitability of InSAR for measuring subsidence !!
https://www.science.org/content/article/satellite-maps-sinking-coastlines-come-under-scrutiny
Furthermore, the discrepancies reported in this study are not observed in vegetated areas, where InSAR performance limitations are well known, but rather in urban environments, where InSAR is generally considered a mature and well-established technique!! This makes the reported differences especially noteworthy and warrants a more rigorous investigation of their origin.
Citation: https://doi.org/10.5194/egusphere-2026-1929-RC8 -
AC8: 'Reply on RC8', Oluwaseyi Dasho, 10 Jun 2026
We thank the reviewer for these additional comments and address each point in turn.
On spatial averaging and GNSS comparison.
The reviewer argues that GNSS validation also requires spatial averaging, and that the same averaging should therefore be applied when comparing our results with OPERA. We accept the first part of this argument, spatial averaging is indeed required when comparing point-based GNSS measurements with area-averaged InSAR pixels, and our tie to the IGS14 reference frame via station HSIB follows standard practice in this regard. However, we note that this argument directly supports rather than undermines our position: if spatial averaging is the acknowledged mechanism by which GNSS and InSAR comparisons must be handled, then spatial averaging is equally the mechanism by which our 15×15 m product and the OPERA ~100×100 m effective resolution product will differ when the underlying deformation field is spatially heterogeneous. The reviewer's own reasoning confirms that resolution-driven spatial averaging is the appropriate explanation for the inter-product discrepancy, not a deficiency in our methodology.
On the magnitude of the discrepancy.
The reviewer states that differences between our results and OPERA are "several orders of magnitude larger" than the reported 0.1 mm/yr uncertainty. We clarify two points. First, the 0.1 mm/yr figure is the regression-derived statistical uncertainty on the velocity estimate at individual pixels, it quantifies the precision of the linear trend fit to our own time series and is not a claim about inter-product agreement or absolute accuracy. Comparing this internal precision metric against an inter-product rate difference is not a meaningful test of accuracy. Second, if for example, the actual inter-product rate difference at the locations in question is approximately 4–6 mm/yr, which is 40–60 times larger than 0.1 mm/yr, that is between one and two orders of magnitude, not several as claimed by the reviewer. This difference is fully accounted for by the combined effect of the LOS-to-vertical projection (cos(θ) ≈ 0.73–0.84 at Sentinel-1 incidence angles over SFO, introducing a systematic 16–27% underestimate in OPERA LOS rates relative to true vertical) and spatial averaging of a heterogeneous deformation field across the OPERA ~100×100 m effective resolution footprint. Neither effect is a measurement error; both are predictable, quantifiable consequences of product design differences. We agree that inter-product agreement at spatially averaged scales is the appropriate comparison metric, and we reiterate our commitment to providing this in the revised Supplementary Information.
On the cited Science news article.
We are aware of the article referenced by the reviewer. We note that it is a news feature, not a peer-reviewed scientific publication, and does not constitute methodological evidence against the results presented here. The concerns raised in that piece pertain broadly to communication of uncertainty in large-scale coastal subsidence assessments and do not specifically address the processing methodology employed in this study or its predecessor publications. We would welcome citation of specific peer-reviewed critiques if the reviewer believes our methodology has substantive documented deficiencies.
On discrepancies in urban environments.
The reviewer observes that the inter-product discrepancies occur in urban areas where InSAR is generally considered a mature technique, and suggests this makes the differences especially noteworthy. We agree that urban coherence is high and that this rules out decorrelation as an explanation. This is precisely why we have consistently attributed the discrepancy to spatial resolution and LOS projection rather than signal quality. Urban deformation fields at airport infrastructure are highly heterogeneous at sub-100 m scales, differential settlement between adjacent construction phases, pavement structures, and pile-supported versus unpiled foundations can produce deformation gradients of several mm/yr over distances of 30–50 m. Spatial averaging across such gradients will systematically attenuate peak rates in coarser products regardless of coherence quality. The urban setting strengthens rather than weakens this interpretation.
We look forward to the reviewer's evaluation of the systematic point-by-point OPERA comparison, with appropriate spatial averaging applied, that we will provide in the revised manuscript.
Citation: https://doi.org/10.5194/egusphere-2026-1929-AC8
-
AC8: 'Reply on RC8', Oluwaseyi Dasho, 10 Jun 2026
-
RC8: 'Reply on AC7', Anonymous Referee #3, 10 Jun 2026
-
AC7: 'Reply on RC7', Oluwaseyi Dasho, 10 Jun 2026
-
RC7: 'Reply on AC6', Anonymous Referee #3, 10 Jun 2026
-
AC6: 'Reply on RC6', Oluwaseyi Dasho, 10 Jun 2026
-
RC6: 'Reply on AC5', Anonymous Referee #3, 09 Jun 2026
-
AC5: 'Reply on RC5', Oluwaseyi Dasho, 09 Jun 2026
-
RC5: 'Reply on AC4', Anonymous Referee #3, 09 Jun 2026
-
AC4: 'Reply on RC4', Oluwaseyi Dasho, 09 Jun 2026
-
RC4: 'Reply on AC3', Anonymous Referee #3, 09 Jun 2026
-
AC3: 'Reply on RC3', Oluwaseyi Dasho, 02 Jun 2026
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 249 | 105 | 50 | 404 | 30 | 18 | 20 |
- HTML: 249
- PDF: 105
- XML: 50
- Total: 404
- Supplement: 30
- BibTeX: 18
- EndNote: 20
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This manuscript presents InSAR-derived vertical land motion (VLM) measurements for the period 2017-2024 and investigates the relative contributions of historical land reclamation practices, paleoenvironmental setting, and operational aircraft loading to the observed VLM using statistical approaches. While the study addresses an important topic and takes a useful approach of examining subsurface profiles to explain VLM, the current manuscript significantly lacks the scientific analysis to support the reported VLM rates and does not provide sufficient scientific novelty in its present form. Therefore, I recommend major revisions, which should help the authors improve both the scientific rigor and novelty of the study. My detailed comments are provided below.