Regional modeling of the impacts of tidal flooding in the context of mean sea level rise on low-lying in the Global South

Silva, Thiago Cavalcante Lins; Diniz, Marco Túlio Mendonça; Araújo, Paulo Victor do Nascimento; Dos Santos, José Yure Gomes; Ferreira, Bruno; Da Silva, Jucielho Pedro

doi:10.5194/egusphere-2025-4929

Preprints

https://doi.org/10.5194/egusphere-2025-4929

Preprints

08 Jan 2026

| 08 Jan 2026

Regional modeling of the impacts of tidal flooding in the context of mean sea level rise on low-lying in the Global South

Thiago Cavalcante Lins Silva, Marco Túlio Mendonça Diniz, Paulo Victor do Nascimento Araújo, José Yure Gomes Dos Santos, Bruno Ferreira, and Jucielho Pedro Da Silva

Abstract. This study assessed the risks and impacts of rising average sea levels on Brazil's semi-arid coastline in a low-lying coastal area with limited response potential, using freely available data and based on the central hypothesis that, even in conservative scenarios, there will be risks with significant impacts. The methodology integrated DEM calibration, geodetic validation of tide gauge data, flood modeling, and overlay with real estate grids to quantify damage. The results showed relative stability of astronomical tides, with projected extremes of up to 2.975 m and 3.454 m, respectively, for a 20-year return period. Meteorological tides showed low values (≈ 0.11 m), although with episodic variability. The modeling indicated that up to 14 % of the total area (about 730 km²) could be affected in extreme scenarios, with progressive flooding of solar salt pans and low-lying urban areas. Cities such as Areia Branca, Macau, and Porto do Mangue are at the highest risk, with a 60–80 % probability of flooded days in severe scenarios. Economic losses were estimated at approximately R$ 36 million in residences (≈ US$ 6.7 million) and R$ 158 million in land (≈US$ 29 million), with Areia Branca being the most impacted municipality. Towns such as Barra, Cristóvão, and Baixa Grande also experienced significant risks and damage. The findings reinforce the usefulness of open data for regional risk analysis, even recognizing limitations in spatial resolution and vertical uncertainties. The methodology proved promising, replicable, and useful for supporting adaptive policies in regions with low institutional technical capacity.

Received: 05 Oct 2025 – Discussion started: 08 Jan 2026

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

Thiago Cavalcante Lins Silva, Marco Túlio Mendonça Diniz, Paulo Victor do Nascimento Araújo, José Yure Gomes Dos Santos, Bruno Ferreira, and Jucielho Pedro Da Silva

Status: final response (author comments only)

RC1:
'Comment on egusphere-2025-4929', Anonymous Referee #1, 31 Jan 2026

The study evaluated the risks and impacts of rising mean sea levels on Brazil’s semi-arid, low-lying coastline using freely available datasets. The authors proposed a framework that integrates calibrated digital elevation models (DEM), validated tide-gauge data, flood modeling, and spatial overlays with real estate grids to estimate the impacts of tidal flooding. Overall, this study is meaningful for more informed coastal flood risk management, especially for areas with limited institutional and technical capacity. However, I still have several concerns and suggestions as follows.

1) Lines 90-94: As the authors mentioned, quite a few studies have been conducted to investigate the flooding risk analysis for low-lying coastal areas. Then what makes this study different from the literature?

2) It is suggested to add a list of acronyms mentioned in the manuscript. The full term of the acronym is only presented the first time it appears, e.g., SGB, however, does not make sense since it is short for "Brazilian Geodetic System". Should it be "BGS"?

3) Equation (1): Exposure is also an important factor in determining the degree of risk. Is it possible to incorporate the effect of exposure in your risk analysis? How would that affect the findings in this study?

4) In the caption of some figures, it is suggested to remove the unnecessary statement like "Map prepared by the authors (2025)".

5) Section 3.1: It is suggested to add a table to present the information of different datasets used in this study and make the corresponding text more concise.

6) Figure 3, Equations (3) and (4), Table 3, Table 4, etc.: To avoid confusion, please change the comma "," to decimal point "." in the elevation numbers.

7) Lines 245-248: The bathtub technique does not account for hydrodynamics and it is very likely that it will lead to overestimates in flood inundation extents. The applicability of the bathtub technique should be justified in more detail. Moreover, it should be noted that various uncertainty sources in the physics-based flood modeling process should not be ignored (Please refer to the paper below). Also, too many references are cited here. It is suggested to remove some old ones.

Reference:

"Uncertainty analysis and quantification in flood insurance rate maps using Bayesian model averaging and hierarchical BMA" (https://doi.org/10.1061/JHYEFF.HEENG-58)

8) Lines 264-266: 0 to 5 is actually six classes. According to Table 1, the hazard class value is from 1 to 5.

9) Table 2 and Table 3: Please add units to the numbers. Also in Table 2, the Kendall’s Tau is a normalized S. Why is S positive, while the Kendall’s Tau is negative?

10) Figure 4: Please add the area name to each sub-figure. For the bottom panel, how would you explain the observed data points of the tidal level are much lower than the fitted values?

11) Figure 6: To make the flooding probability clearer, it would be better to remove the ocean area from the figures.

12) Figure 9: It is suggested to add a scale bar to each sub-figure.

13) Should Figure 13 be Table 6?

14) Line 540: What does “addressing climate change” mean?

Citation: https://doi.org/10.5194/egusphere-2025-4929-RC1
- AC1:
  'Reply on RC1', Thiago Silva, 06 Mar 2026
  General Answers
  We thank the Editor-in-Chief for the opportunity and the reviewers for their insightful and constructive contributions. Their comments were fundamental in improving the methodological clarity, conceptual consistency, and presentation of the results. Below we respond point by point. The revisions also contributed to strengthening the scientific and applied relevance of the manuscript. It is hoped that the study can serve as a methodological reference for future research on coastal flood risk, especially in coastal cities with limited financial resources, where approaches based on accessible data and replicable methods are particularly necessary.
  Responses to Reviewer 01
  1) Thank you for your comment. The unique feature of this study is not only the analysis of coastal flood risk, but also the structured methodological integration for contexts with low technical availability, combining:
  - Multi-criteria calibration of free DEMs;
  - Validation with local tide gauges;
  - Systematic application of the bathtub technique under MSLR scenarios;
  - Overlay with real estate grids as a proxy for property exposure.
  It is a valid strategy and an important differential in terms of assessing low-lying coasts in the context of the global south with poor data infrastructure.
  While previous studies analyze coastal risk, few:
  - Integrate DEM calibration with local validation;
  - Apply a replicable structure explicitly focused on the Global South;
  - Incorporate spatial proxies of economic exposure in semi-arid areas with low institutional infrastructure. Finally, the methodology applied will become a benchmark to be replicated in several coastal cities with low financial resources.
  We will include an additional paragraph at the end of the introduction explaining these contributions.
  2) We fully agree. Acronyms will be better managed in the manuscript, including a table with all acronyms used in the manuscript. Regarding the observation about SGB: in fact, the Brazilian Geodetic System corresponds to BGS, not SGB. Its use will be reviewed to maintain institutional and linguistic consistency.
  3) We appreciate the pertinent observation.
  In fact, the formulation adopted in Equation (1) structures risk as a function of hazard (H) and vulnerability (V), not explicitly including exposure as a multiplicative term, an approach that has also been adopted in several studies in the literature (e.g., Araújo et al., 2021 - cited in the manuscript), and which is in line with the conceptual framework presented in the IPCC Sixth Assessment Report (IPCC, 2022).
  However, exposure was considered in the spatial stage of the analysis, through the intersection between potentially flood-prone areas and the urban real estate grid. This procedure allowed us to quantify the assets located in hazard zones, characterizing a spatial operationalization of exposure, even though it was not mathematically formalized in the risk equation.
  However, we can clarify further in the text that exposure is operationalized spatially. We also emphasize that the explicit inclusion of exposure as a multiplicative term could alter the absolute magnitude of the risk values, but would not change the spatial pattern of the priority areas identified, which is the main result of the study.
  A brief section discussing this conceptual framework will be included.
  4) We agree. The statement will be removed from the subtitles.
  5) We fully agree. A table will be added containing:
  Source
  
  Spatial resolution
  
  Temporal scale
  
  Reference system
  
  Data type
  
  Application in the study
  
  Collection point
  
  The text will be condensed as suggested.
  6) We agree. All commas will be replaced by decimal points.
  7) We appreciate the extremely pertinent observation.
  We recognize that the bathtub technique may overestimate flooded areas by not considering hydrodynamics. However, its application was based on:
  Regional scale;
  
  Absence of high-resolution bathymetric data;
  
  Context of applicability in regions with technical restrictions;
  
  Objective of initial risk screening.
  
  We will include:
  Expanding the methodological justification for choosing the static approach;
  
  Including explicit discussion of its limitations and potential overestimations;
  
  Inserting the suggested reference on uncertainties;
  
  Reducing older citations, prioritizing recent literature.
  
  We emphasize that, in the context of the study, the technique was used as a preliminary and comparative assessment tool between different DEMs, and not as a substitute for physics-based hydrodynamic modeling. We also emphasize that previous studies, published in NHESS by co-authors of this work, applied the bathtub technique in regions of the study area and obtained satisfactory results (Araújo et al., 2021 - cited in the manuscript).
  8) Correction accepted. We will adjust the description to maintain consistency with Table 1.
  9) Thank you for your comment.
  The units will be included in Tables 2 and 3, as suggested.
  Regarding Table 2, to facilitate verification and ensure consistency between the statistics presented, the column “Denominator (S)”, which represents the root of the variance of S, we propose replacing the value of S to facilitate verification. This modification makes it easier to verify the relationship between S, Z, and Kendall's Tau, while maintaining the mathematical consistency of the results.
  At the discretion of the checker, the values of S were as follows: for ATM (-214), ATAB (-95), MTM (-21), and MTAB (-498).
  This change does not alter the values or interpretations discussed in the manuscript.
  10) The following will be added:
  Area name and tide variable.
  
  In the lower panel, the points represent the observed annual maximums, while the curve corresponds to the theoretical quantiles estimated by the Gumbel distribution adjusted by maximum likelihood. The parametric fit does not consist of an interpolation of empirical data, but rather a probabilistic modeling of the extreme behavior of the series.
  Thus, differences between the observed values and the fitted curve are expected, since the model seeks to represent the overall statistical pattern of the sample, and local deviations between empirical and theoretical quantiles may occur. These differences reflect the variability inherent in finite series of annual maximums and do not indicate inconsistency in the adjustment.
  An additional explanation will be included in the text to clarify this behavior.
  11) We agree. The ocean area will be masked to improve visualization.
  12) A scale bar will be included.
  13) Formatting error confirmed. Will be corrected.
  14) The expression will be replaced by:
  “mitigate and adapt to the impacts of climate change”.
  
  Citation: https://doi.org/10.5194/egusphere-2025-4929-AC1
RC2:
'Comment on egusphere-2025-4929', Anonymous Referee #2, 12 Feb 2026
This study evaluates the risk of tidal flooding in the low-lying coastal areas in Brazil in the context of sea-level rise using open-source data. The evaluation framework can be helpful in areas with limited technical methods. However, the method used for flooding risk analysis in this study is not explained clearly, and there are several errors in the manuscript. My suggestions and questions are as follows.
The title indicates that this study evaluates the low-lying areas in the Global South, yet its study area is limited to Brazil. Why was Brazil chosen? Are the low-lying areas in Brazil the only representative cases for this study in the whole Global South? Should the title be “in Brazil” rather than “in the Global South”?

Equation 1 and section 3.6: Details should be explained on the “hazard” and “vulnerability”. What is the total flood modeling that derives “hazard”? How is “vulnerability” calculated? What factors are considered in deriving “vulnerability”?

The section about materials and methods is very confusing to read. The methods are not explained in detail. Necessary equations and modeling methods are not explained. The connections between the collected data and methods are not clear. It is recommended to add a flow chart for the risk evaluation processes for clarity.

There are a lot of acronyms in the manuscript. Some of them seem unnecessary since they are only used once (e.g., BSC, FT, GEV), and some are not given (e.g., SMC-Brazil). It is recommended to check all acronyms and compile a list of them for clear reference.

Section 3.1: The description of the collected data is very confusing. A table including the information on all the datasets is recommended.

Section 3.1.1: Elevation data should be demonstrated in a figure.

Section 3.1.5: Please explain the reason for using MSLR projections on three scales, and how they are “systematized”.

Section 3.1.6 and 3.1.7: Please add figures for land use data and urban data. What exactly is the urban data, images taken by UAVs, or something more?

In all figure captions, it is unnecessary to declare that the figures are prepared by the authors unless figures from other sources are used. These statements should be removed.

The numbering of sections 3.2 to 3.6 is incorrect. Section 3.3 is missing.

Table 2: Why is the header of the first column “Dates”? What is the meaning of “Initially, differences can be noted between the different processes in the region” in the caption?

Figure 4: Please add the location and whether it is the astronomical or meteorological tide in each sub-figure.

Figure 7: It is inappropriate to cite an online news report using a figure. It should be properly cited in words, and a formal reference item should be included in the References section with the title, the news reporter, the publication time, the URL to the news article, the access date, etc.

Figure 4, 5, 6, 11, and 12: Please label the sub-figures with alphabetical identifiers.

Please check all the tables and add units to numbers.
Citation: https://doi.org/10.5194/egusphere-2025-4929-RC2
- AC2:
  'Reply on RC2', Thiago Silva, 06 Mar 2026
  Responses to Reviewer 02
  1) Thank you for your comment.
  Brazil was selected as a study area because it presents characteristics representative of various coastal contexts in the Global South, i.e., a cross-section of the proposed geographical area, including:
  - Extensive coastline with semi-arid stretches and low topographical altitude;
  - High socioeconomic vulnerability in coastal areas;
  - Availability of consistent public data;
  - Institutional and coastal planning challenges similar to those faced by other countries in the Global South.
  However, we recognize that the original title may suggest an empirical generalization for the entire Global South, while the study is specifically applied to the Brazilian context. Thus, to avoid overly broad interpretations, we propose adjusting the title to:
  We propose adjusting to:
  “Regional modeling of the impacts of tidal flooding in the context of average sea level rise in low-lying areas of Brazil's semi-arid coast”.
  2) We appreciate the comment. Section 3.6 will be restructured to explicitly clarify the modeling of risk and vulnerability.
  Risk was operationalized as the spatial combination of flood hazard and the vulnerability of exposed elements. This approach has also been adopted in several studies in the literature (e.g., Araújo et al., 2021 - cited in the manuscript) and is consistent with the conceptual framework presented in the IPCC Sixth Assessment Report (IPCC, 2022).
  The hazard was defined based on the extreme hydrodynamic scenario, consisting of: (i) projected mean sea level rise (MSLR), (ii) extreme astronomical tide, (iii) extreme meteorological tide, and (iv) the vertical uncertainty of the digital elevation model (RMSE), incorporated as a conservative margin in the delimitation of the potentially floodable area.
  Vulnerability was estimated based on land use and land cover classification, assigning relative levels of susceptibility to different classes, following an approach similar to that adopted in recent studies in the international literature. This strategy represents the physical vulnerability associated with different types of land use in the face of flooding.
  The revised section will detail the classification criteria and the form of spatial integration between hazard and vulnerability in the derivation of risk.
  3) We agree completely.
  The following will be included:
  Complete methodological flowchart;
  
  Reorganization of subsections;
  
  Textual revision;
  
  Explanation of the connections between data and methods.
  
  4) We agree. Acronyms used only once will be removed.
  SMC-Brazil will be explained appropriately.
  5) We fully agree. A table will be added containing:
  Source
  
  Spatial resolution
  
  Temporal scale
  
  Reference system
  
  Data type
  
  Application in the study
  
  Collection point
  
  The text will be condensed as suggested.
  6) A figure illustrating the Digital Elevation Models (DEMs) used will be included.
  7) Thank you for your comment.
  Section 3.1.5 will be revised to clarify that the mean sea level rise (MSLR) projections on three scales (global, intermediate, and regional) were used exclusively as additional vertical increments in the composition of flood elevations, in order to clarify the different scenarios inherent in the projections, especially in relation to the quality of the projection data.
  The final flood elevation for each scenario was obtained by summing: (i) the return period of the extreme astronomical tide, (ii) the return period of the extreme meteorological tide, (iii) the increment corresponding to the MSLR projection at the scale considered, and (iv) the RMSE of the DEM, incorporated as a margin of vertical uncertainty.
  Each MSLR rate therefore generated a distinct elevation scenario, with the delimitation of flood-prone areas being carried out by classifying the rasters in the GIS based on these final elevations.
  The revised section will explicitly detail this systematization.
  8) Figures illustrating land use and land cover data, as well as urban data used in the analysis, will be added.
  We clarify that urban data corresponds to vector layers derived from real estate cadastral databases, produced from the interpretation of high-resolution orthomosaics obtained by UAV and GNSS surveys (for geodetic calibration purposes). The images were processed photogrammetrically to generate georeferenced orthomosaics, from which built features and urban infrastructure were delimited.
  The revised section will detail the vector nature of this data and the procedure for acquiring and processing the images.
  9) We agree.
  10) Agreed. It will be corrected.
  11) The header of the first column will be corrected from “Datas” to “Data”, in order to correctly reflect the content presented in the table.
  As for the phrase “Initially, differences between the different processes in the region can be observed,” we recognize that the wording is not appropriate for the table caption, as it introduces preliminary interpretation in a space that should be strictly descriptive. The text will be revised to make it more objective and compatible with the informative function of the caption.
  12) It will be included.
  13) We agree.
  The figure will be removed or replaced by:
  - Formal citation in the text;
  - Complete reference in the bibliography section.
  We will follow the journal's writing guidelines.
  14) It will be implemented.
  15) It will be reviewed in its entirety.
  
  Citation: https://doi.org/10.5194/egusphere-2025-4929-AC2

Thiago Cavalcante Lins Silva, Marco Túlio Mendonça Diniz, Paulo Victor do Nascimento Araújo, José Yure Gomes Dos Santos, Bruno Ferreira, and Jucielho Pedro Da Silva

Viewed

Total article views: 451 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
278	142	31	451	12	13

HTML: 278
PDF: 142
XML: 31
Total: 451
BibTeX: 12
EndNote: 13

Views and downloads (calculated since 08 Jan 2026)

Month	HTML	PDF	XML	Total
Jan 2026	164	38	11	213
Feb 2026	50	69	13	132
Mar 2026	64	35	7	106

Cumulative views and downloads (calculated since 08 Jan 2026)

Month	HTML	PDF	XML	Total
Jan 2026	164	38	11	213
Feb 2026	50	69	13	132
Mar 2026	64	35	7	106

Viewed (geographical distribution)

Total article views: 443 (including HTML, PDF, and XML) Thereof 443 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 19 Mar 2026

Short summary

This study examines the risks of sea level rise on Brazil's semi-arid coast, a vulnerable, low-lying region. Using open data, we modelled tidal flooding and assessed the social and economic impacts on towns and salt pan areas. The results revealed significant exposure even in conservative scenarios, emphasising the urgent need for adaptation in coastal areas of the Global South.


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