Misconception of coastal resilience caused by inconsistent resolution in bathymetry mapping

Misconception of coastal resilience caused by inconsistent resolution in bathymetry mapping | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Misconception of coastal resilience caused by inconsistent resolution in bathymetry mapping Bo Miao, Peter Arlinghaus, Ha Thi Minh Ho-Hagemann, Corinna Schrum, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6254537/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Long-term time series of bathymetric data of coastal zones are indispensable for analysing coastal morphological resilience to climate change. Despite the increasing popularity of utilizing high-resolution gridded bathymetric digital elevation models for coastal management, potential errors in analysing the long-term trend of mean elevation change from historical bathymetric datasets spanning a period of multiple years to decades have attracted little attention. Here, we demonstrate that inconsistency in the spatial resolution of small-scale topographic features characterized by sharp bathymetric gradients, such as tidal creeks and streams, could produce an artificial false trend of mean elevation change that is on the same or even higher order of the sea level change rate. Neglecting this inconsistency may lead to a misconception of coastal resilience to sea level rise and misguide planning and implementation of coastal protection strategies. We provide an analytical method to identify such inconsistency in time series of gridded digital elevation models and a homogenization method to minimise the associated errors. Our methods are broadly applicable to reduce errors in bathymetric analysis and improve quantitative assessment of coastal resilience to climate change. Geomorphology Coastal resilience Sea level rise Bathymetry mapping Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Low-lying coastal zones constitute the most dynamic part of the surface Earth and are constantly changing their morphology driven by both natural and anthropogenic forcing. Relative sea-level rise (RSL), caused jointly by climate change and vertical land movement, has been identified as a major threat to human life and socio-economic stability in coastal zones worldwide 1 – 3 . The global-mean relative sea-level has been rising at a mean rate of 2.6 mm yr − 1 over the recent two decades and exhibits exceptionally high rates in subsiding low-lying coastal zones at 7.8 to 9.9 mm yr − 1 4 . This will likely lead to more flooding, submersion and erosion of the coastal zones as well as degradation of coastal ecosystems 5 . The resilience of coastal morphology is largely dependent on whether sedimentation in the system can keep pace with the rising sea level 6 – 8 . For quantitative assessment of sedimentation rate in the coastal zones, sediment budget analysis based on long-term time series of high-resolution bathymetric digital elevation models (DEM) is indispensable 9 – 11 . Bathymetric data of coastal zones is usually an integration of measurements by multiple instruments (Fig. 1 a) adopting different technologies including optical sensing (e.g., satellite, LiDAR-light detection and ranging) and acoustic sensing (e.g., single beam, multibeam and side-scan sonars) that are of varying accuracy, spatial resolution and limitations 11 – 13 . Integrating measurement datasets which are highly volatile in spatial and temporal resolution to uniformly gridded DEM requires interpolation in both space and time 14 . Despite various approaches for combining spatial and temporal interpolation of multiple datasets to generate consistent time series of DEM 12 , 14 , 15 , potential errors in analysing the long-term trend of mean elevation change from gridded DEM spanning a period of multiple years to decades have not been assessed in existing literature. Identifying such errors and their correction are key for quantifying sedimentation rate in the coastal zones and assessing their resilience to the rising sea level that are fundamental for developing coastal management strategies. In this study, we analysed 25 annual high-resolution bathymetric DEM spanning from 1998 to 2022 on a 10 m × 10 m regular grid for a complex coastal zone consisting of estuaries, barrier islands, and part of the largest tidal flat systems of the world, the Wadden Sea. We identified an inconsistency in the spatial resolution of sampling between earlier and later years, despite all measurement data are integrated to a regular grid of 10 m × 10 m. This inconsistency results in errors in the estimated long-term trend of mean elevation change of both intertidal and subtidal areas. The errors are comparable to and in some tidal basins even higher than the long-term RSL rate in the region. We then propose solutions to minimise the errors and discuss a global implication of the outcomes. Results Inconsistency in small-scale topographic features By comparing the time series of gridded DEM, we identified an inconsistency in resolution of small-scale topographic features (e.g., creeks and streams) between earlier and later years. Taking the Jade Bay as one of the 24 tidal basins (Supplementary Fig. 1) in the German Wadden Sea for example, the number of identified tidal channels (including creeks and streams) increases remarkably from 310 for the year 1998 to 488 for the year 2019 (Supplementary Fig. 2). A similar increase is seen for all other 23 tidal basins. Such increase is mostly attributed to a better resolution of small-scale topographic features such as creeks and streams that are of a width of a few meters to a few tens of meters in the DEM for later years (Fig. 1 b). To confirm an improved resolution of small-scale topographic features in later years, we performed Fast Fourier Transform (FFT) on each of the annual bathymetric DEM for each tidal basin to convert the topographic signal from the original space domain to a frequency domain (see Methods). In the resultant frequency domain, topography characterized by sharp gradients such as edges of channels are represented by high frequencies, while low frequencies correspond to a spatially smoothed topography (Supplementary Fig. 3). A generally increasing trend of frequency is seen in the time series of DEM for each tidal basin (Supplementary Fig. 4), indicating that more small-scale topographic features that are characterized by sharp bathymetric gradients are resolved in later years compared to earlier years (Fig. 1 b). Impact of sharp topographic gradients on the mean elevation of tidal basins We identified a strong positive correlation ( r > 0.5, P < 0.01) in the annual time series of DEM between the normalized high frequency (HF) in the frequency domain and the mean elevation of the intertidal area in 19 of the 24 tidal basins, whilst a strong negative correlation ( r < -0.5, P < 0.01) between the HF and the mean elevation of the subtidal area is found in 13 of the 24 tidal basins (Extended Data Table 1 and Supplementary Figs. 5&6). Statistically significant positive correlations between the HF and the mean elevation of intertidal areas are also seen in the rest 5 tidal basins despite with r < 0.5. By contrast, among the rest 11 tidal basins which do not exhibit strong negative correlations between the HF and the mean elevation of subtidal areas, positive correlations are found in 4 of them. Considering that the HF is an indicator of resolution of small-scale topographic features with sharp bathymetric gradients, its significant correlation with the mean elevation of intertidal/subtidal areas implies their potential causal linkage. To disentangle their relationship, we have designed a set of numerical experiments to understand the dependence of the mean elevation of intertidal and subtidal zones on sampling resolution (see Methods). Results from the experiments (Fig. 2 ) show that the calculated mean elevation of intertidal areas increases along with an increase in sampling resolution represented by HF ( r > 0.95), whilst the calculated mean elevation of subtidal areas decreases along with increasing HF ( r < -0.95). The difference in the calculated mean elevation of intertidal areas exceeds 10 mm between low and high HF on a 10 m × 10 m grid and is slightly smaller (~ 9 mm) on a 100 m × 100 m grid. For subtidal areas, the difference in the mean elevation exceeds 20 mm between low and high HF on both grids. Besides the strong dependence of the mean elevation on the sampling resolution, it is worth to note that the calculated mean elevation of intertidal areas is persistently lower than the “ground truth” value with decreasing sampling resolution leading to increasing deviation. By contrast, the calculated mean elevation of subtidal areas is persistently higher than the “ground truth” value with decreasing sampling resolution also leading to larger deviation. Not only sampling resolution but also gridding resolution introduces errors in the elevation (Fig. 2 ). For both intertidal and subtidal areas, a coarser gridding resolution (e.g. from 10 m × 10 m to 100 m × 100 m) leads to larger deviation of the calculated mean elevation from the “ground truth”. The mean elevation of intertidal area is 5–10 mm lower in the 100 m × 100 m grid than the 10 m × 10 m with the same coverage of sampling points. By contrast, the mean elevation of subtidal area is 15–20 mm higher in the 100 m × 100 m grid than the 10 m × 10 m in the same sampling condition. Homogenization of long-term time series of bathymetric DEM More homogeneous time series of DEM in terms of sampling resolution can be derived by filtering of the frequency domain (see Methods and Supplementary Figs. 7 & 8). We adopted two solutions to homogenize the long-term time series of DEM so that errors associated with inconsistent resolution of small-scale bathymetric gradients are minimized (Supplementary Figs. 9 & 10). Linear regression trends of the mean elevation changes of intertidal and subtidal areas of the 24 investigated tidal basins based on the original data show that almost all intertidal areas except one (Basin 24) are in accretion (0.7–19.3 mm yr − 1 ) for the period 1998–2022 (Extended Data Table 2). By contrast, 18 out of 24 subtidal areas are in erosion (-2 – -30 mm yr − 1 ) and the rest are in accretion (0.1–40 mm yr − 1 ) for the same period. A subset of the time series (mainly between 2010 and 2022) with comparable level of sampling resolution derived from Solution #1 indicates remarkably reduced accretion rates (0.1–10 mm yr − 1 ) of the intertidal areas in most basins in recent years, with 7 of the previously accreting intertidal areas even turned to erosion state (-0.1 – -10 mm yr − 1 ). Moreover, erosion not only occurs in more subtidal areas (20 out of 24) but also enhances in 8 subtidal areas in recent years (2010–2022) compared to earlier years (1998–2009). In general, most tidal basins exhibit a systematic shift toward erosion, namely less accretion in the intertidal areas and more erosion in the subtidal areas, in recent years (2010–2022) compared to earlier years (1998–2009). A comparison of the linear regression trends of the mean elevation changes between the original data and the homogenized data (Solution #2) for the period 1998–2022 reveals that despite accretion still occurs in most intertidal areas (21 out of 24) in the homogenized data, the rates (0.3–19 mm yr − 1 ) are considerably lower compared to the original data (Fig. 3 , Extended Table 2). The reduction in the accretion rate ranges between 0.2 and 6.8 mm yr − 1 among the different intertidal areas (Extended Data Fig. 1 ). Erosion remains in most subtidal areas (17 out of 24) in the homogenized data. However, the change in the rate of subtidal areas is less clear than that of intertidal areas, with enhanced and reduced erosion rate in 8 and 9 subtidal areas, respectively. The difference in the mean elevation change rates of intertidal areas between the original data and the homogenized data, when compared with the RSL rate, can lead to different and even opposite assessments of coastal resilience (Fig. 4 ). In the original data, the mean elevation change rates of intertidal areas in 16 tidal basins are higher than the upper range of RSL rate (mean + standard deviation) for the period 1998–2022 and are categorized as “Shallowing” indicating that these areas are elevated at a rate surpassing the RSL and thus becoming shallower. Five tidal basins are categorized as “Quasi-stable” because the mean elevation change rates of their intertidal areas fall in the range between the upper and lower range of RSL rate (mean ± standard deviation). Only 2 tidal basins are categorized as “Deepening” due to a lower mean elevation change rate than the lower range of RSL rate (mean - standard deviation). By contrast, the number of “Deepening” tidal basins increases to 5 in the homogenized data, whilst only 9 tidal basins (compared to 16 in the original data) are in “Shallowing” state. Implications for global assessment of coastal resilience to sea level rise Low-lying coastal zones with tidal flats are distributed worldwide 17 (Supplementary Fig. 1). About 16% of global tidal flats’ surface area was lost between 1984 and 2016 owing to multiple anthropogenic and climate stressors according to an analysis by Murray et al 17 . Moreover, coastal resilience is determined by not only its horizontal dimension (surface area) but also its vertical dimension (depth range between intertidal and subtidal areas) 18 . Higher tidal ranges promoted by sea level rise can lead to an increased vertical dimension of tidal basins by erosion in the subtidal channels and sedimentation on the intertidal flats 19 , 20 . Effect of this natural process, superposed by an artificial trend associated with inconsistent resolution of small-scale topographic features as revealed in our study, may lead to overestimation of sedimentation rates on the intertidal flats and excessive optimistic assessment of their resilience to RSL 9 . Our findings imply that contemporary and future adaptation needs for coastal protection are likely much higher than previously assessed. The false artificial trend in the mean elevation change associated with inconsistent resolution of small-scale sharp-gradient topographic features is inherent in long-term time series of bathymetric DEM generated by integration of multiple data sources. There exists no single approach or instrument for high-resolution bathymetric mapping covering both intertidal and subtidal areas. A recent review shows that the current research on bathymetric mapping in shallow waters is characterized by multi-platform, multi-sensor, and multi-model trends 11 . Thus, the integration of multiple data sources is a crucial step for deriving high-resolution DEM of coastal zones. Despite various approaches for combining spatial and temporal interpolation of multiple datasets have been developed in recent decades 11 , 12 , 14 , 15 , potential inconsistency in resolving small-scale sharp-gradient topographic features is often overlooked and our methods for identification of such inconsistencies in the time series of DEM and their homogenization present a valuable contribution to such effort. Methods Long-term time series of bathymetric data Annual bathymetric DEM at a 10 × 10 m grid for the German Wadden Sea from 1998 to 2022 as products from the project EasyGSH-DB 14 ( https://doi.org/10.48437/02.2020.K2.7000.0001 ) and its follow-up project TrilatWatt ( https://trilawatt.eu/en/data/data-products/ ) were analyzed in this study. Data from more than 20000 measuring campaigns were merged and interpolated by spatio-temporal interpolation using the Functional Seabed Model (FSM), a data-based hindcast simulation model for the bathymetric development of the subaquatic surface 14 , 21 . Sources of measurement include shipborne echo sounder, LiDAR and profile measurement surveys, with a highly varying measuring resolution ranging from less than 1 m in the nearshore intertidal area to ~ 100 m in deep subtidal channels 14 . The uncertainty of the measurements varies among different instruments and was estimated to be ~ 20 cm on average for the Wadden Sea basins 22 . However, the overall uncertainty is reduced due to the large number of grid cells as well as the large area considered 9 . In processing the annual datasets for individual tidal basins, we have removed those which are featured by artifacts (Supplementary Fig. 11) to further ensure the data consistency. RSL change rates and map of coastal morphological resilience Wahl et al. 23 estimated the mean RSL trends at 30 tide gauge stations along the North Sea for the period 1993–2011 23 . Maximum RSL rise trends are seen in the German Bight including the Wadden Sea, with 2.2 ± 2.5 mm yr − 1 at Norderney (southwest of the German Wadden Sea) and a north-eastward increase to 6.6 ± 3.2 mm yr − 1 at Hörnum. The trends include the component of the vertical land motion induced by the glacial isostatic adjustment (GIA) which was estimated to range between − 0.36 and − 0.57 mm yr − 1 in our study area based on a global GIA model 24 . High-resolution time series of vertical land motion from the European Ground Motion Service (EGMS) product based on Synthetic Aperture Radar Interferometry (InSAR) data derived from Sentinel-1 at the same locations for the period 2015–2021 ( https://egms.land.copernicus.eu/ ) show a general agreement with the estimated values despite of an overall underestimation by less than 1 mm yr − 1 in the latter. Therefore, the same values of the mean RSL trends from Wahl et al. 23 were adopted in this study. A coastal morphological resilience classification map (Fig. 4 ) is produced based on a comparison between the mean RSL trends and the mean elevation change trends of the intertidal areas of each tidal basin for the period 1998–2022. This map helps identify areas that are not able to keep pace with the RSL. Fast Fourier Transform (FFT) and Inverse Fast Fourier Transform (I-FFT) Fast Fourier transform (FFT) is an efficient implementation of the discrete Fourier transform (DFT). The DFT maps a sequence either in the time domain or in the spatial domain into the frequency domain 25 , while the inverse discrete Fourier transform (IDFT) performs the opposite 26 . After shifting the zero-frequency component to the centre of the spectrum, the frequency domain exhibit two major features 27 , namely, (a) higher frequencies represent sharper small-scale gradients and are located further away from the center of the domain, and (b) the brightness indicates the proportion of each corresponding frequency in the entire domain, with increasing brightness for a larger proportion. An example illustrating the application of FFT and inverse FFT (I-FFT) to topography is given in Supplementary Fig. 3. A low-pass filtering of the frequency domain and its I-FFT would generate a smoothed topography, whilst topographic features of sharp bathymetric gradients such as edges of channels and creeks are captured by an I-FFT of the high-pass filtered frequency domain. Because the high frequency band may vary over a large range among different DEM, normalization was applied to measure its variation in the annual time series of DEM for each tidal basin. The normalization consists of three steps. First, we removed the common low frequency band of each DEM by high pass filtering. Afterward, we calculated the mean amplitude of the remaining high frequencies in the domain over the 25 annual time series (1998–2022). In the last step, we normalized the mean amplitude of high frequencies for each annual time series by confining the value range between 0.1 (the DEM with the lowest mean amplitude of high frequencies) and 0.9 (the DEM with the highest mean amplitude of high frequencies). The derived normalized high frequency (NHF) is used as a quantitative indicator for comparing the resolution of small-scale sharp bathymetric gradients in the annual time series of DEM for each tidal basin. Detection of intertidal and subtidal areas A tidal basin consists of a subtidal part and an intertidal part 28 . We followed the study by Benninghoff &Winter 9 to distinguish intertidal and subtidal areas of each tidal basin. The intertidal area is defined as the part that is submerged at mean high water level (MHW) but emerged at mean low water level (MLW). Consequently, the mean elevation of the intertidal area ( \(\:{\text{H}}_{\text{i}\text{n}\text{t}}\) ) is calculated by: $$\:{\text{H}}_{\text{i}\text{n}\text{t}}=\text{M}\text{L}\text{W}+\frac{\text{V}\text{s}\left(\text{z}\:>\:\text{M}\text{L}\text{W}\right)-\text{V}\text{s}\left(\text{z}\:>\:\text{M}\text{H}\text{W}\right)}{\text{A}\left(\text{z}\:>\:\text{M}\text{L}\text{W}\right)-\text{A}\left(\text{z}\:>\:\text{M}\text{H}\text{W}\right)}$$ , where z is the bed elevation, Vs (z > MLW) and Vs (z > MHW) refer to the sediment volume above the MLW and the MHW, respectively, and A (z > MLW) and A (z > MHW) correspond to the area above the MLW and the MHW, respectively. The calculation loops though all grid cells of a tidal basin. Similarly, the mean elevation of the subtidal area ( \(\:{\text{H}}_{\text{s}\text{u}\text{b}}\) ) is given by: $$\:{\text{H}}_{\text{s}\text{u}\text{b}}=\text{M}\text{L}\text{W}-\frac{\text{V}\text{w}\left(\text{z}\:<\:\text{M}\text{L}\text{W}\right)}{\text{A}\left(\text{z}\:<\:\text{M}\text{L}\text{W}\right)}$$ , where Vw (z < MLW) is the water volume below the MLW. The values of MLW and MHW for each tidal basin were derived from Benninghoff &Winter 9 and listed in Supplementary Table 1. Linear regression analysis was applied to the annual time series of \(\:{\text{H}}_{\text{i}\text{n}\text{t}}\) and \(\:{\text{H}}_{\text{s}\text{u}\text{b}}\) to derive their trends. Detection of tidal channels and creeks We adopted the same procedure from our previous study 29 for channel detection based on image analysis. The procedure consists of 5 consecutive steps, including presentation of the bathymetry in a greyscale, background noise removal by anisotropic diffusion, binary image generation based on an adaptive thresholding method and correction for channel continuity, identification of intersections and end points of all channels, and quantification of channel number and length. Experiments on relationship between mean elevation and sampling resolution In signal processing, it is known that the sampling rate must be at least twice the bandwidth of the signal to avoid aliasing, so-called the Nyquist–Shannon sampling theorem 30 . For bathymetric sampling, this refers to the sampling resolution which is needed to resolve the small-scale topographic features characterized by sharp bathymetric gradient. Clearly, lower sampling resolution would lead to higher distortion of topography and increased loss of sharp bathymetric gradients. We designed a set of numerical experiments to derive a quantitative understanding of the impact of sampling resolution on the mean elevation of intertidal and subtidal areas. We assumed a gridded DEM (10 m× 10 m) as the “ground truth” of the topography of a tidal basin (Jade Bay as the test case) and performed sampling with different spatial resolutions on this “ground truth” topography. We firstly divided the whole domain into a 100 m × 100 m resolution grid. This means each grid cell contains maximum 100 sampling points. We then started with a sampling resolution of 10 random samples per grid cell and gradually increased the number of sampling points till approaching the “ground truth”, i.e. all 100 points in each grid cell are sampled. For each resolution 100 times of random sampling were performed. Each of the derived dataset was then interpolated to a 10 m × 10 m grid and a 100 m × 100 m grid by inverse distance weighting (IDW) using the 5 nearest sampled data points for each grid point, respectively. Intertidal and subtidal areas of the resultant gridded DEM were then identified, and their mean elevation were calculated and compared to the “ground truth” values. Solutions for minimizing the errors associated with sharp bathymetric gradients Two solutions are proposed in this study for minimizing the errors in the estimation of mean elevation change trends of coastal zones associated with inconsistent resolution of small-scale bathymetric gradients in the long-term time series of DEM. Solution #1 is to select only those DEM with NHF on a similar level for the estimation. In this case, only a subset of DEM from the long-term time series is selected and therefore the estimated trend may not be representative for a longer term. The other solution (Solution #2) is to lower down the NHF of the DEM for later years by removing the high frequencies (low-pass filtering) in the frequency domain so that the resultant NHF is comparable to the DEM for earlier years. This sacrifices the data quality for the layer years but allows to derive a consistent trend covering the entire long-term time series. Declarations Data Availability The original annual bathymetric DEM at a 10 ´ 10 m grid for the German Wadden Sea from 1998 to 2022 were derived from the project EasyGSH-DB (https://doi.org/10.48437/02.2020.K2.7000.0001) and its follow-up project TrilatWatt (https://trilawatt.eu/en/data/data-products/) and are openly accessible at the respective websites. Low-pass filtering of the DEM was done by executing the codes shared on Zenodo https://zenodo.org/records/14943963. Code availability The codes for processing and analyzing bathymetric data as well as for producing the figures and tables are available at Zenodo https://zenodo.org/records/14943963. Acknowledgements This study is a contribution to the Helmholtz research programme POF IV “The Changing Earth – Sustaining our Future” on “Topic 4: Coastal zones at a time of global change”. B.M. is supported by grants from the China Scholarship Council. Author contributions W.Z. conceived the study and designed the numerical experiments. P. A. developed the methodology. B.M. and P.A. collected data and performed the experiments. B.M. analysed the data and experiment results and wrote the first draft under the supervision by W. Z. P.A., H. H. and C.S. contributed to the evaluation and discussion of the data. All authors contributed to revision of the manuscript and approved the final version. Competing interests The authors declare no competing interests. References Oppenheimer, M. et al. Sea Level Rise and Implications for Low-Lying Islands, Coasts and Communities. Ch. 4 (Cambridge Univ. Pres, 2019). Vousdoukas, M.I. et al. Sandy coastlines under threat of erosion. Nat. Clim. Chang. 10 , 260-263 (2020). https://doi.org/10.1038/s41558-020-0697-0 Thiéblemont, R. et al. Assessing current coastal subsidence at continental scale: Insights from Europe using the European Ground Motion Service. Earth's Future 12 , e2024EF004523 (2024). https://doi.org/10.1029/2024EF004523 Nicholls, R.J. et al. A global analysis of subsidence, relative sea-level change and coastal flood exposure. Nat. Clim. Chang. 11 , 338-342 (2021). https://doi.org/10.1038/s41558-021-00993-z van de Wal, R. et al. Sea Level Rise in Europe: Impacts and consequences, in: Sea Level Rise in Europe: 1st Assessment Report of the Knowledge Hub on Sea Level Rise (SLRE1). State Planet 3-slre1 , 5 (2024). https://doi.org/10.5194/sp-3-slre1-5-2024 Kirwan, M., & Megonigal, J. Tidal wetland stability in the face of human impacts and sea-level rise. Nature 504 , 53-60 (2013). https://doi.org/10.1038/nature12856 Schuerch, M. et al. Future response of global coastal wetlands to sea-level rise. Nature 561 , 231-234 (2018). https://doi.org/10.1038/s41586-018-0476-5 Brand, M. W., Buffington, K., Rogers, J.B., Thorne, K., Stein, E. D., & Sanders, B. F. Multi-decadal simulation of marsh topography under sea level rise and episodic sediment loads. J. Geophys. Res. Earth Surf. 127 , e2021JF006526 (2022). https://doi.org/10.1029/2021JF006526 Benninghoff, M. & Winter, C. Recent morphologic evolution of the German Wadden Sea. Sci. Rep. 9 , 9293 (2019). https://doi.org/10.1038/s41598-019-45683-1 Cesbron, G., Melet, A., Almar, R., Lifermann, A., Tullot, D. & Crosnier, L. Pan-European Satellite-Derived Coastal Bathymetry—Review, User Needs and Future Services. Front. Mar. Sci. 8 ,740830 (2021). https://doi.org/10.3389/fmars.2021.740830 He, J., Zhang, S., Cui, X. & Feng, W. Remote sensing for shallow bathymetry: a systematic review. Earth Sci. Rev. 258 , 104957 (2024). https://doi.org/10.1016/j.earscirev.2024.104957 Li Z. X. et al. Exploring modern bathymetry: A comprehensive review of data acquisition devices, model accuracy, and interpolation techniques for enhanced underwater mapping. Front. Mar. Sci. 10, 1178845 (2023). http://dx.doi.org/10.3389/fmars.2023.1178845 Wölfl A. C. et al. Seafloor mapping - the challenge of a truly global ocean bathymetry. Front. Mar. Sci . 6, 283(2019). https://doi.org/10.3389/fmars.2019.00283 Sievers, J., Milbradt, P., Ihde, R., Valerius, J., Hagen, R., & Plüß, A. An integrated marine data collection for the German Bight – Part 1: Subaqueous geomorphology and surface sedimentology (1996–2016). Earth Syst. Sci. Data 13 , 4053-4065 (2021). https://doi.org/10.5194/essd-13-4053-2021 Włodarczyk-Sielicka, M., Bodus-Olkowska, I. & Łącka, M. The process of modelling the elevation surface of a coastal area using the fusion of spatial data from different sensors. Oceanologia 64 , 22-34 (2022). https://doi.org/10.1016/j.oceano.2021.08.002 Bradski, G. The OpenCV library. Dr. Dobbs J. 25 , 120-126 (2000). Murray, N.J. et al. The global distribution and trajectory of tidal flats. Nature 565 , 222-225 (2019). https://doi.org/10.1038/s41586-018-0805-8 Hanegan, K.C., FitzGerald, D.M., Georgiou, I.Y. & Hughes, Z.J. Long-term sea level rise modeling of a basin-tidal inlet system reveals sediment sinks. Nat Commun 14 , 7117 (2023). https://doi.org/10.1038/s41467-023-42895-y Friedrichs, C. T. Tidal Flat Morphodynamics: A Synthesis. Treatise on Estuarine and Coastal Science 3 , 137-170 (2011). Zhang, W. & Arlinghaus, P. Climate, Coast, and Morphology. Oxford Research Encyclopedia of Climate Science . (Oxford University Press, 2022). https://doi.org/10.1093/acrefore/9780190228620.013.814 Milbradt, P., Valerius, J. & Zeiler, M. Das funktionale Bodenmodell: Aufbereitung einer konsistenten Datenbasis für die Morphologie und Sedimentologie, Die Küste 83 AufMod , 19-38 (2015). https://doi.org/20.500.11970/101736 Valerius, J., Feldmann, J., van Zoest, M., Milbradt, P. & Zeiler, M. Documentation of morphological products from the AufMod project Functional Seabed Model, data format: Text files (CSV, XYZ). Bundesamt für Seeschiffart und Hydrographie. 11pp (2013). Wahl, T. et al. Observed mean sea level changes around the North Sea coastline from 1800 to present. Earth Sci. Rev. 124, 51-67 (2013). https://doi.org/10.1016/j.earscirev.2013.05.003 Peltier, W.R. Global glacial isostasy and the surface of the ice-age Earth: the ICE-5G(VM2) model and GRACE. Annu. Rev. Earth Planet. Sci. 32 , 111–149 (2004). http://dx.doi.org/10.1146/annurev.earth.32.082503.144359 Rao, K.R., Kim, D.N. & Hwang, J.J. Fast Fourier Transform-Algorithms and Applications. (Springer, Berlin, 2011). Kumar, G.G., Sahoo, S.K. & Meher, P.K. 50 Years of FFT Algorithms and Applications. Circuits Syst Signal Process 38 , 5665–5698 (2019). https://doi.org/10.1007/s00034-019-01136-8 Das, A. Interpretation and Processing of Image in Frequency Domain. In: Guide to Signals and Patterns in Image Processing. Springer, Cham (2015). Mahavadi, T. F., Seiffert, R., Kelln, J. & Fröhle, P. Effects of sea level rise and tidal flat growth on tidal dynamics and geometry of the Elbe estuary, Ocean Sci. 20 , 369–388 (2024). https://doi.org/10.5194/os-20-369-2024 Arlinghaus, P., Zhang, W. & Schrum, C. Small-scale benthic faunal activities may lead to large-scale morphological change- A model based assessment. Front. Mar. Sci. 9 , 1011760 (2022). https://doi.org/10.3389/fmars.2022.1011760 H. Nyquist. Certain Topics in Telegraph Transmission Theory. Transactions of the American Institute of Electrical Engineers 47, 617-644 (1928). https://doi.org/10.1109/T-AIEE.1928.5055024 Additional Declarations The authors declare no competing interests. Supplementary Files SupplementMiaoetal.docx Supplementary Figures and Tables ExtendedDataTableandFigure.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6254537","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":430588559,"identity":"9f5828ed-a6bb-4b4c-8fc5-69282668eab8","order_by":0,"name":"Bo Miao","email":"","orcid":"","institution":"Helmholtz-Zentrum Hereon","correspondingAuthor":false,"prefix":"","firstName":"Bo","middleName":"","lastName":"Miao","suffix":""},{"id":430588560,"identity":"dc52c249-4a7e-48fc-ab48-321eccf4a5e4","order_by":1,"name":"Peter Arlinghaus","email":"","orcid":"","institution":"Helmholtz-Zentrum Hereon","correspondingAuthor":false,"prefix":"","firstName":"Peter","middleName":"","lastName":"Arlinghaus","suffix":""},{"id":430588561,"identity":"4f1b8fb7-b9d0-4b2a-b8fa-d68da05550a8","order_by":2,"name":"Ha Thi Minh Ho-Hagemann","email":"","orcid":"","institution":"Helmholtz-Zentrum Hereon","correspondingAuthor":false,"prefix":"","firstName":"Ha","middleName":"Thi Minh","lastName":"Ho-Hagemann","suffix":""},{"id":430588562,"identity":"7fdabd3d-56cb-4354-a294-98899f05e3f3","order_by":3,"name":"Corinna Schrum","email":"","orcid":"","institution":"Helmholtz-Zentrum Hereon","correspondingAuthor":false,"prefix":"","firstName":"Corinna","middleName":"","lastName":"Schrum","suffix":""},{"id":430588563,"identity":"53885bf4-43c2-46b5-8e1d-7725d0f5b82b","order_by":4,"name":"Wenyan Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+klEQVRIiWNgGAWjYDADAxDxsQHK4yFWC+NMkrUw8xKjxeB47zGJnzsY7M3Zzx5+bbvDJp+fvfcAw5sKPFrOnEuT7D3DwGzZk5dmnXsmzXJmz7kExjlncGsxu5FjJsHbxsBmcCDHzDi37bCBwY0cA2agCG4t99+YSf5tY+AxOP/GzNiy7b+BPVjLP3y28JhJA82UABpu/Jix7YCBgQRISwNuLfZncoytZdskgO55Y8bY25ZsIHHmjMHBOcdwa5FsP2N4822bjb3B+RzjDz/b7Az423sMH7ypwa0FCFgkGBgkQAw2CZjQAbwagBH4AZ0xCkbBKBgFowAFAAAZl08YHlMDBgAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-6239-8312","institution":"Helmholtz-Zentrum Hereon","correspondingAuthor":true,"prefix":"","firstName":"Wenyan","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2025-03-18 15:20:54","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-6254537/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6254537/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":78884080,"identity":"76ee459a-31d6-4a2d-a015-1bc160233057","added_by":"auto","created_at":"2025-03-20 09:17:55","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":451577,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTechnologies for bathymetry mapping in coastal zones and example of inconsistency in sampling resolution. a, \u003c/strong\u003eSketch showing shipborne and airborne measurements. Integration of data from different sources may lead to inconsistency in resolution of small-scale topographic features as illustrated by a comparison of the topography of a tidal basin (Jade Bay) between 1999 (\u003cstrong\u003eb\u003c/strong\u003e) and 2019 (\u003cstrong\u003ec\u003c/strong\u003e). The sharp-gradient topographic features of the Jade Bay were obtained using the Sobel operator for edge detection through Python's CV2 library\u003csup\u003e16\u003c/sup\u003e.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6254537/v1/f9b07172fe5adf7325699db2.png"},{"id":78885381,"identity":"f80d177b-1e57-4f85-840f-42ff8b19cf7b","added_by":"auto","created_at":"2025-03-20 09:33:56","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":443294,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eNumerical experiments showing the relationship between sampling resolution and the mean elevation of intertidal and subtidal areas of a tidal basin. \u003c/strong\u003eSampling resolution is represented by the normalized high frequency in the frequency domain. \u003cstrong\u003ea\u003c/strong\u003e, Variation of the calculated mean elevation of intertidal (y-axis on the left) and subtidal (y-axis on the right) areas on a 10 m × 10 m grid interpolated from varying sampling densities. \u003cstrong\u003eb\u003c/strong\u003e, Similar to (\u003cstrong\u003ea\u003c/strong\u003e) but on a 100 m × 100 m grid. Sampling density starts from 10 random samples per 100 m × 100 m grid cell (i.e. 10% of the total size of data points as indicated by the x-axis) and gradually increases toward a full coverage of data points representing the “ground truth” (see Methods). The “ground truth” values of the mean elevation of the intertidal and subtidal areas are indicated by the blue and red horizontal lines, respectively.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6254537/v1/8827517fe2aac34b110edb9c.png"},{"id":78884082,"identity":"ad72c845-6609-42d0-8c98-385de858915c","added_by":"auto","created_at":"2025-03-20 09:17:56","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":460029,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLinear regression trends of mean elevation changes of intertidal and subtidal areas of the German Wadden Sea tidal basins\u003c/strong\u003e. \u003cstrong\u003ea,\u003c/strong\u003e Schematic showing two solutions for minimizing the errors associated with inconsistent sampling resolution. \u003cstrong\u003eb\u003c/strong\u003e, Mean elevation change rate of the tidal basins for the period 1998-2022 derived from the original data. \u003cstrong\u003ec,\u003c/strong\u003e Mean elevation change rate of the tidal basins derived from Solution #1 by selecting the years with comparable sampling resolution. \u003cstrong\u003ed\u003c/strong\u003e, Mean elevation change rate of the tidal basins for the period 1998-2022 derived from Solution #2 by low-pass filtering.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6254537/v1/4a0efdfbf8b48d65f3e7f476.png"},{"id":78884086,"identity":"e7b11e9d-58de-4526-825e-1f9ab1df7c61","added_by":"auto","created_at":"2025-03-20 09:17:56","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":283207,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAssessment of the resilience of the German Wadden Sea tidal basins against RSL for the period 1998-2022.\u003c/strong\u003e \u003cstrong\u003ea,\u003c/strong\u003e Result based on the original data, \u003cstrong\u003eb, \u003c/strong\u003eResult based on the homogenized data\u003cstrong\u003e \u003c/strong\u003e(Solution #2). Areas with mean accretion rate larger than the upper range of RSL rate (mean + standard deviation) are categorized as “Shallowing” (in green). Areas with mean elevation change rate between the upper and lower range of RSL rate (mean ± standard deviation) are categorized as “Quasi-stable” (in yellow). Areas with mean elevation change rate lower than the lower range of RSL rate (mean - standard deviation) are categorized by “Deepening” (in red).\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6254537/v1/beef04c1ac35d294f6269dfe.png"},{"id":78886378,"identity":"61e7c973-e52d-4cb8-9cb1-3c01924f015d","added_by":"auto","created_at":"2025-03-20 09:41:57","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2604939,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6254537/v1/eec363b9-99c4-4e0e-9575-b61fd8b5233d.pdf"},{"id":78884100,"identity":"5d565c0a-b684-447c-9329-380cda057e7d","added_by":"auto","created_at":"2025-03-20 09:17:56","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":10532768,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Figures and Tables\u003c/p\u003e","description":"","filename":"SupplementMiaoetal.docx","url":"https://assets-eu.researchsquare.com/files/rs-6254537/v1/3cf3e45bbac12214c1f1df76.docx"},{"id":78884083,"identity":"1c118b62-2b65-4c64-bcc7-2ce1adc911d6","added_by":"auto","created_at":"2025-03-20 09:17:56","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":267317,"visible":true,"origin":"","legend":"","description":"","filename":"ExtendedDataTableandFigure.docx","url":"https://assets-eu.researchsquare.com/files/rs-6254537/v1/c9b1721d08a7626845655390.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eMisconception of coastal resilience caused by inconsistent resolution in bathymetry mapping\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eLow-lying coastal zones constitute the most dynamic part of the surface Earth and are constantly changing their morphology driven by both natural and anthropogenic forcing. Relative sea-level rise (RSL), caused jointly by climate change and vertical land movement, has been identified as a major threat to human life and socio-economic stability in coastal zones worldwide\u003csup\u003e\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. The global-mean relative sea-level has been rising at a mean rate of 2.6 mm yr\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e over the recent two decades and exhibits exceptionally high rates in subsiding low-lying coastal zones at 7.8 to 9.9 mm yr\u003csup\u003e\u0026minus;\u0026thinsp;1 4\u003c/sup\u003e. This will likely lead to more flooding, submersion and erosion of the coastal zones as well as degradation of coastal ecosystems\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe resilience of coastal morphology is largely dependent on whether sedimentation in the system can keep pace with the rising sea level\u003csup\u003e\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. For quantitative assessment of sedimentation rate in the coastal zones, sediment budget analysis based on long-term time series of high-resolution bathymetric digital elevation models (DEM) is indispensable\u003csup\u003e\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eBathymetric data of coastal zones is usually an integration of measurements by multiple instruments (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea) adopting different technologies including optical sensing (e.g., satellite, LiDAR-light detection and ranging) and acoustic sensing (e.g., single beam, multibeam and side-scan sonars) that are of varying accuracy, spatial resolution and limitations\u003csup\u003e\u003cspan additionalcitationids=\"CR12\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. Integrating measurement datasets which are highly volatile in spatial and temporal resolution to uniformly gridded DEM requires interpolation in both space and time\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Despite various approaches for combining spatial and temporal interpolation of multiple datasets to generate consistent time series of DEM\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e, potential errors in analysing the long-term trend of mean elevation change from gridded DEM spanning a period of multiple years to decades have not been assessed in existing literature. Identifying such errors and their correction are key for quantifying sedimentation rate in the coastal zones and assessing their resilience to the rising sea level that are fundamental for developing coastal management strategies.\u003c/p\u003e \u003cp\u003eIn this study, we analysed 25 annual high-resolution bathymetric DEM spanning from 1998 to 2022 on a 10 m \u0026times; 10 m regular grid for a complex coastal zone consisting of estuaries, barrier islands, and part of the largest tidal flat systems of the world, the Wadden Sea. We identified an inconsistency in the spatial resolution of sampling between earlier and later years, despite all measurement data are integrated to a regular grid of 10 m \u0026times; 10 m. This inconsistency results in errors in the estimated long-term trend of mean elevation change of both intertidal and subtidal areas. The errors are comparable to and in some tidal basins even higher than the long-term RSL rate in the region. We then propose solutions to minimise the errors and discuss a global implication of the outcomes.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eInconsistency in small-scale topographic features\u003c/h2\u003e \u003cp\u003eBy comparing the time series of gridded DEM, we identified an inconsistency in resolution of small-scale topographic features (e.g., creeks and streams) between earlier and later years. Taking the Jade Bay as one of the 24 tidal basins (Supplementary Fig.\u0026nbsp;1) in the German Wadden Sea for example, the number of identified tidal channels (including creeks and streams) increases remarkably from 310 for the year 1998 to 488 for the year 2019 (Supplementary Fig.\u0026nbsp;2). A similar increase is seen for all other 23 tidal basins. Such increase is mostly attributed to a better resolution of small-scale topographic features such as creeks and streams that are of a width of a few meters to a few tens of meters in the DEM for later years (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb). To confirm an improved resolution of small-scale topographic features in later years, we performed Fast Fourier Transform (FFT) on each of the annual bathymetric DEM for each tidal basin to convert the topographic signal from the original space domain to a frequency domain (see Methods). In the resultant frequency domain, topography characterized by sharp gradients such as edges of channels are represented by high frequencies, while low frequencies correspond to a spatially smoothed topography (Supplementary Fig.\u0026nbsp;3). A generally increasing trend of frequency is seen in the time series of DEM for each tidal basin (Supplementary Fig.\u0026nbsp;4), indicating that more small-scale topographic features that are characterized by sharp bathymetric gradients are resolved in later years compared to earlier years (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eImpact of sharp topographic gradients on the mean elevation of tidal basins\u003c/h3\u003e\n\u003cp\u003eWe identified a strong positive correlation (\u003cem\u003er\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.5, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01) in the annual time series of DEM between the normalized high frequency (HF) in the frequency domain and the mean elevation of the intertidal area in 19 of the 24 tidal basins, whilst a strong negative correlation (\u003cem\u003er\u003c/em\u003e \u0026lt; -0.5, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01) between the HF and the mean elevation of the subtidal area is found in 13 of the 24 tidal basins (Extended Data Table\u0026nbsp;1 and Supplementary Figs.\u0026nbsp;5\u0026amp;6). Statistically significant positive correlations between the HF and the mean elevation of intertidal areas are also seen in the rest 5 tidal basins despite with \u003cem\u003er\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.5. By contrast, among the rest 11 tidal basins which do not exhibit strong negative correlations between the HF and the mean elevation of subtidal areas, positive correlations are found in 4 of them.\u003c/p\u003e \u003cp\u003eConsidering that the HF is an indicator of resolution of small-scale topographic features with sharp bathymetric gradients, its significant correlation with the mean elevation of intertidal/subtidal areas implies their potential causal linkage. To disentangle their relationship, we have designed a set of numerical experiments to understand the dependence of the mean elevation of intertidal and subtidal zones on sampling resolution (see Methods).\u003c/p\u003e \u003cp\u003eResults from the experiments (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) show that the calculated mean elevation of intertidal areas increases along with an increase in sampling resolution represented by HF (\u003cem\u003er\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.95), whilst the calculated mean elevation of subtidal areas decreases along with increasing HF (\u003cem\u003er\u003c/em\u003e \u0026lt; -0.95). The difference in the calculated mean elevation of intertidal areas exceeds 10 mm between low and high HF on a 10 m \u0026times; 10 m grid and is slightly smaller (~\u0026thinsp;9 mm) on a 100 m \u0026times; 100 m grid. For subtidal areas, the difference in the mean elevation exceeds 20 mm between low and high HF on both grids. Besides the strong dependence of the mean elevation on the sampling resolution, it is worth to note that the calculated mean elevation of intertidal areas is persistently lower than the \u0026ldquo;ground truth\u0026rdquo; value with decreasing sampling resolution leading to increasing deviation. By contrast, the calculated mean elevation of subtidal areas is persistently higher than the \u0026ldquo;ground truth\u0026rdquo; value with decreasing sampling resolution also leading to larger deviation.\u003c/p\u003e \u003cp\u003eNot only sampling resolution but also gridding resolution introduces errors in the elevation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). For both intertidal and subtidal areas, a coarser gridding resolution (e.g. from 10 m \u0026times; 10 m to 100 m \u0026times; 100 m) leads to larger deviation of the calculated mean elevation from the \u0026ldquo;ground truth\u0026rdquo;. The mean elevation of intertidal area is 5\u0026ndash;10 mm lower in the 100 m \u0026times; 100 m grid than the 10 m \u0026times; 10 m with the same coverage of sampling points. By contrast, the mean elevation of subtidal area is 15\u0026ndash;20 mm higher in the 100 m \u0026times; 100 m grid than the 10 m \u0026times; 10 m in the same sampling condition.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eHomogenization of long-term time series of bathymetric DEM\u003c/h3\u003e\n\u003cp\u003eMore homogeneous time series of DEM in terms of sampling resolution can be derived by filtering of the frequency domain (see Methods and Supplementary Figs.\u0026nbsp;7 \u0026amp; 8). We adopted two solutions to homogenize the long-term time series of DEM so that errors associated with inconsistent resolution of small-scale bathymetric gradients are minimized (Supplementary Figs.\u0026nbsp;9 \u0026amp; 10).\u003c/p\u003e \u003cp\u003eLinear regression trends of the mean elevation changes of intertidal and subtidal areas of the 24 investigated tidal basins based on the original data show that almost all intertidal areas except one (Basin 24) are in accretion (0.7\u0026ndash;19.3 mm yr\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) for the period 1998\u0026ndash;2022 (Extended Data Table\u0026nbsp;2). By contrast, 18 out of 24 subtidal areas are in erosion (-2 \u0026ndash; -30 mm yr\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) and the rest are in accretion (0.1\u0026ndash;40 mm yr\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) for the same period. A subset of the time series (mainly between 2010 and 2022) with comparable level of sampling resolution derived from Solution #1 indicates remarkably reduced accretion rates (0.1\u0026ndash;10 mm yr\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) of the intertidal areas in most basins in recent years, with 7 of the previously accreting intertidal areas even turned to erosion state (-0.1 \u0026ndash; -10 mm yr\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e). Moreover, erosion not only occurs in more subtidal areas (20 out of 24) but also enhances in 8 subtidal areas in recent years (2010\u0026ndash;2022) compared to earlier years (1998\u0026ndash;2009). In general, most tidal basins exhibit a systematic shift toward erosion, namely less accretion in the intertidal areas and more erosion in the subtidal areas, in recent years (2010\u0026ndash;2022) compared to earlier years (1998\u0026ndash;2009).\u003c/p\u003e \u003cp\u003eA comparison of the linear regression trends of the mean elevation changes between the original data and the homogenized data (Solution #2) for the period 1998\u0026ndash;2022 reveals that despite accretion still occurs in most intertidal areas (21 out of 24) in the homogenized data, the rates (0.3\u0026ndash;19 mm yr\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) are considerably lower compared to the original data (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Extended Table\u0026nbsp;2). The reduction in the accretion rate ranges between 0.2 and 6.8 mm yr\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e among the different intertidal areas (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Erosion remains in most subtidal areas (17 out of 24) in the homogenized data. However, the change in the rate of subtidal areas is less clear than that of intertidal areas, with enhanced and reduced erosion rate in 8 and 9 subtidal areas, respectively.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe difference in the mean elevation change rates of intertidal areas between the original data and the homogenized data, when compared with the RSL rate, can lead to different and even opposite assessments of coastal resilience (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). In the original data, the mean elevation change rates of intertidal areas in 16 tidal basins are higher than the upper range of RSL rate (mean\u0026thinsp;+\u0026thinsp;standard deviation) for the period 1998\u0026ndash;2022 and are categorized as \u0026ldquo;Shallowing\u0026rdquo; indicating that these areas are elevated at a rate surpassing the RSL and thus becoming shallower. Five tidal basins are categorized as \u0026ldquo;Quasi-stable\u0026rdquo; because the mean elevation change rates of their intertidal areas fall in the range between the upper and lower range of RSL rate (mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation). Only 2 tidal basins are categorized as \u0026ldquo;Deepening\u0026rdquo; due to a lower mean elevation change rate than the lower range of RSL rate (mean - standard deviation). By contrast, the number of \u0026ldquo;Deepening\u0026rdquo; tidal basins increases to 5 in the homogenized data, whilst only 9 tidal basins (compared to 16 in the original data) are in \u0026ldquo;Shallowing\u0026rdquo; state.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eImplications for global assessment of coastal resilience to sea level rise\u003c/h3\u003e\n\u003cp\u003eLow-lying coastal zones with tidal flats are distributed worldwide\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e(Supplementary Fig.\u0026nbsp;1). About 16% of global tidal flats\u0026rsquo; surface area was lost between 1984 and 2016 owing to multiple anthropogenic and climate stressors according to an analysis by Murray et al\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Moreover, coastal resilience is determined by not only its horizontal dimension (surface area) but also its vertical dimension (depth range between intertidal and subtidal areas)\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. Higher tidal ranges promoted by sea level rise can lead to an increased vertical dimension of tidal basins by erosion in the subtidal channels and sedimentation on the intertidal flats\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. Effect of this natural process, superposed by an artificial trend associated with inconsistent resolution of small-scale topographic features as revealed in our study, may lead to overestimation of sedimentation rates on the intertidal flats and excessive optimistic assessment of their resilience to RSL\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Our findings imply that contemporary and future adaptation needs for coastal protection are likely much higher than previously assessed.\u003c/p\u003e \u003cp\u003eThe false artificial trend in the mean elevation change associated with inconsistent resolution of small-scale sharp-gradient topographic features is inherent in long-term time series of bathymetric DEM generated by integration of multiple data sources. There exists no single approach or instrument for high-resolution bathymetric mapping covering both intertidal and subtidal areas. A recent review shows that the current research on bathymetric mapping in shallow waters is characterized by multi-platform, multi-sensor, and multi-model trends\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. Thus, the integration of multiple data sources is a crucial step for deriving high-resolution DEM of coastal zones. Despite various approaches for combining spatial and temporal interpolation of multiple datasets have been developed in recent decades\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e, potential inconsistency in resolving small-scale sharp-gradient topographic features is often overlooked and our methods for identification of such inconsistencies in the time series of DEM and their homogenization present a valuable contribution to such effort.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eLong-term time series of bathymetric data\u003c/h2\u003e \u003cp\u003eAnnual bathymetric DEM at a 10 \u0026times; 10 m grid for the German Wadden Sea from 1998 to 2022 as products from the project EasyGSH-DB\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.48437/02.2020.K2.7000.0001\u003c/span\u003e\u003cspan address=\"10.48437/02.2020.K2.7000.0001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and its follow-up project TrilatWatt (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://trilawatt.eu/en/data/data-products/\u003c/span\u003e\u003cspan address=\"https://trilawatt.eu/en/data/data-products/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) were analyzed in this study. Data from more than 20000 measuring campaigns were merged and interpolated by spatio-temporal interpolation using the Functional Seabed Model (FSM), a data-based hindcast simulation model for the bathymetric development of the subaquatic surface\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. Sources of measurement include shipborne echo sounder, LiDAR and profile measurement surveys, with a highly varying measuring resolution ranging from less than 1 m in the nearshore intertidal area to ~\u0026thinsp;100 m in deep subtidal channels\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. The uncertainty of the measurements varies among different instruments and was estimated to be ~\u0026thinsp;20 cm on average for the Wadden Sea basins\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. However, the overall uncertainty is reduced due to the large number of grid cells as well as the large area considered\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. In processing the annual datasets for individual tidal basins, we have removed those which are featured by artifacts (Supplementary Fig.\u0026nbsp;11) to further ensure the data consistency.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eRSL change rates and map of coastal morphological resilience\u003c/h3\u003e\n\u003cp\u003eWahl et al.\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e estimated the mean RSL trends at 30 tide gauge stations along the North Sea for the period 1993\u0026ndash;2011\u003csup\u003e23\u003c/sup\u003e. Maximum RSL rise trends are seen in the German Bight including the Wadden Sea, with 2.2\u0026thinsp;\u0026plusmn;\u0026thinsp;2.5 mm yr\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e at Norderney (southwest of the German Wadden Sea) and a north-eastward increase to 6.6\u0026thinsp;\u0026plusmn;\u0026thinsp;3.2 mm yr\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e at H\u0026ouml;rnum. The trends include the component of the vertical land motion induced by the glacial isostatic adjustment (GIA) which was estimated to range between \u0026minus;\u0026thinsp;0.36 and \u0026minus;\u0026thinsp;0.57 mm yr\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e in our study area based on a global GIA model\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. High-resolution time series of vertical land motion from the European Ground Motion Service (EGMS) product based on Synthetic Aperture Radar Interferometry (InSAR) data derived from Sentinel-1 at the same locations for the period 2015\u0026ndash;2021 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://egms.land.copernicus.eu/\u003c/span\u003e\u003cspan address=\"https://egms.land.copernicus.eu/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) show a general agreement with the estimated values despite of an overall underestimation by less than 1 mm yr\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e in the latter. Therefore, the same values of the mean RSL trends from Wahl et al. \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e were adopted in this study.\u003c/p\u003e \u003cp\u003eA coastal morphological resilience classification map (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) is produced based on a comparison between the mean RSL trends and the mean elevation change trends of the intertidal areas of each tidal basin for the period 1998\u0026ndash;2022. This map helps identify areas that are not able to keep pace with the RSL.\u003c/p\u003e\n\u003ch3\u003eFast Fourier Transform (FFT) and Inverse Fast Fourier Transform (I-FFT)\u003c/h3\u003e\n\u003cp\u003eFast Fourier transform (FFT) is an efficient implementation of the discrete Fourier transform (DFT). The DFT maps a sequence either in the time domain or in the spatial domain into the frequency domain\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e, while the inverse discrete Fourier transform (IDFT) performs the opposite\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. After shifting the zero-frequency component to the centre of the spectrum, the frequency domain exhibit two major features\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e, namely, (a) higher frequencies represent sharper small-scale gradients and are located further away from the center of the domain, and (b) the brightness indicates the proportion of each corresponding frequency in the entire domain, with increasing brightness for a larger proportion.\u003c/p\u003e \u003cp\u003eAn example illustrating the application of FFT and inverse FFT (I-FFT) to topography is given in Supplementary Fig.\u0026nbsp;3. A low-pass filtering of the frequency domain and its I-FFT would generate a smoothed topography, whilst topographic features of sharp bathymetric gradients such as edges of channels and creeks are captured by an I-FFT of the high-pass filtered frequency domain.\u003c/p\u003e \u003cp\u003eBecause the high frequency band may vary over a large range among different DEM, normalization was applied to measure its variation in the annual time series of DEM for each tidal basin. The normalization consists of three steps. First, we removed the common low frequency band of each DEM by high pass filtering. Afterward, we calculated the mean amplitude of the remaining high frequencies in the domain over the 25 annual time series (1998\u0026ndash;2022). In the last step, we normalized the mean amplitude of high frequencies for each annual time series by confining the value range between 0.1 (the DEM with the lowest mean amplitude of high frequencies) and 0.9 (the DEM with the highest mean amplitude of high frequencies). The derived normalized high frequency (NHF) is used as a quantitative indicator for comparing the resolution of small-scale sharp bathymetric gradients in the annual time series of DEM for each tidal basin.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eDetection of intertidal and subtidal areas\u003c/h2\u003e \u003cp\u003eA tidal basin consists of a subtidal part and an intertidal part\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. We followed the study by Benninghoff \u0026amp;Winter\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e to distinguish intertidal and subtidal areas of each tidal basin. The intertidal area is defined as the part that is submerged at mean high water level (MHW) but emerged at mean low water level (MLW). Consequently, the mean elevation of the intertidal area (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{H}}_{\\text{i}\\text{n}\\text{t}}\\)\u003c/span\u003e\u003c/span\u003e) is calculated by:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:{\\text{H}}_{\\text{i}\\text{n}\\text{t}}=\\text{M}\\text{L}\\text{W}+\\frac{\\text{V}\\text{s}\\left(\\text{z}\\:\u0026gt;\\:\\text{M}\\text{L}\\text{W}\\right)-\\text{V}\\text{s}\\left(\\text{z}\\:\u0026gt;\\:\\text{M}\\text{H}\\text{W}\\right)}{\\text{A}\\left(\\text{z}\\:\u0026gt;\\:\\text{M}\\text{L}\\text{W}\\right)-\\text{A}\\left(\\text{z}\\:\u0026gt;\\:\\text{M}\\text{H}\\text{W}\\right)}$$\u003c/div\u003e\u003c/div\u003e,\u003c/p\u003e \u003cp\u003ewhere \u003cem\u003ez\u003c/em\u003e is the bed elevation, Vs (z\u0026thinsp;\u0026gt;\u0026thinsp;MLW) and Vs (z\u0026thinsp;\u0026gt;\u0026thinsp;MHW) refer to the sediment volume above the MLW and the MHW, respectively, and A (z\u0026thinsp;\u0026gt;\u0026thinsp;MLW) and A (z\u0026thinsp;\u0026gt;\u0026thinsp;MHW) correspond to the area above the MLW and the MHW, respectively. The calculation loops though all grid cells of a tidal basin. Similarly, the mean elevation of the subtidal area (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{H}}_{\\text{s}\\text{u}\\text{b}}\\)\u003c/span\u003e\u003c/span\u003e) is given by:\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:{\\text{H}}_{\\text{s}\\text{u}\\text{b}}=\\text{M}\\text{L}\\text{W}-\\frac{\\text{V}\\text{w}\\left(\\text{z}\\:\u0026lt;\\:\\text{M}\\text{L}\\text{W}\\right)}{\\text{A}\\left(\\text{z}\\:\u0026lt;\\:\\text{M}\\text{L}\\text{W}\\right)}$$\u003c/div\u003e\u003c/div\u003e,\u003c/p\u003e \u003cp\u003ewhere Vw (z\u0026thinsp;\u0026lt;\u0026thinsp;MLW) is the water volume below the MLW.\u003c/p\u003e \u003cp\u003eThe values of MLW and MHW for each tidal basin were derived from Benninghoff \u0026amp;Winter \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e and listed in Supplementary Table\u0026nbsp;1. Linear regression analysis was applied to the annual time series of \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{H}}_{\\text{i}\\text{n}\\text{t}}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{H}}_{\\text{s}\\text{u}\\text{b}}\\)\u003c/span\u003e\u003c/span\u003e to derive their trends.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eDetection of tidal channels and creeks\u003c/h2\u003e \u003cp\u003eWe adopted the same procedure from our previous study\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e for channel detection based on image analysis. The procedure consists of 5 consecutive steps, including presentation of the bathymetry in a greyscale, background noise removal by anisotropic diffusion, binary image generation based on an adaptive thresholding method and correction for channel continuity, identification of intersections and end points of all channels, and quantification of channel number and length.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eExperiments on relationship between mean elevation and sampling resolution\u003c/h2\u003e \u003cp\u003eIn signal processing, it is known that the sampling rate must be at least twice the bandwidth of the signal to avoid aliasing, so-called the Nyquist\u0026ndash;Shannon sampling theorem\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. For bathymetric sampling, this refers to the sampling resolution which is needed to resolve the small-scale topographic features characterized by sharp bathymetric gradient. Clearly, lower sampling resolution would lead to higher distortion of topography and increased loss of sharp bathymetric gradients.\u003c/p\u003e \u003cp\u003eWe designed a set of numerical experiments to derive a quantitative understanding of the impact of sampling resolution on the mean elevation of intertidal and subtidal areas. We assumed a gridded DEM (10 m\u0026times; 10 m) as the \u0026ldquo;ground truth\u0026rdquo; of the topography of a tidal basin (Jade Bay as the test case) and performed sampling with different spatial resolutions on this \u0026ldquo;ground truth\u0026rdquo; topography. We firstly divided the whole domain into a 100 m \u0026times; 100 m resolution grid. This means each grid cell contains maximum 100 sampling points. We then started with a sampling resolution of 10 random samples per grid cell and gradually increased the number of sampling points till approaching the \u0026ldquo;ground truth\u0026rdquo;, i.e. all 100 points in each grid cell are sampled. For each resolution 100 times of random sampling were performed. Each of the derived dataset was then interpolated to a 10 m \u0026times; 10 m grid and a 100 m \u0026times; 100 m grid by inverse distance weighting (IDW) using the 5 nearest sampled data points for each grid point, respectively. Intertidal and subtidal areas of the resultant gridded DEM were then identified, and their mean elevation were calculated and compared to the \u0026ldquo;ground truth\u0026rdquo; values.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eSolutions for minimizing the errors associated with sharp bathymetric gradients\u003c/h2\u003e \u003cp\u003eTwo solutions are proposed in this study for minimizing the errors in the estimation of mean elevation change trends of coastal zones associated with inconsistent resolution of small-scale bathymetric gradients in the long-term time series of DEM. Solution #1 is to select only those DEM with NHF on a similar level for the estimation. In this case, only a subset of DEM from the long-term time series is selected and therefore the estimated trend may not be representative for a longer term. The other solution (Solution #2) is to lower down the NHF of the DEM for later years by removing the high frequencies (low-pass filtering) in the frequency domain so that the resultant NHF is comparable to the DEM for earlier years. This sacrifices the data quality for the layer years but allows to derive a consistent trend covering the entire long-term time series.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData Availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe original annual bathymetric DEM at a 10 \u0026acute; 10 m grid for the German Wadden Sea from 1998 to 2022 were derived from the project EasyGSH-DB (https://doi.org/10.48437/02.2020.K2.7000.0001) and its follow-up project TrilatWatt (https://trilawatt.eu/en/data/data-products/) and are openly accessible at the respective websites. Low-pass filtering of the DEM was done by executing the codes shared on Zenodo https://zenodo.org/records/14943963. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe codes for processing and analyzing bathymetric data as well as for producing the figures and tables are available at Zenodo https://zenodo.org/records/14943963.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study is a contribution to the Helmholtz research programme POF IV \u0026ldquo;The Changing Earth \u0026ndash; Sustaining our Future\u0026rdquo; on \u0026ldquo;Topic 4: Coastal zones at a time of global change\u0026rdquo;.\u003c/p\u003e\n\u003cp\u003eB.M. is supported by grants from the China Scholarship Council.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eW.Z. conceived the study and designed the numerical experiments. P. A. developed the methodology. B.M. and P.A. collected data and performed the experiments. B.M. analysed the data and experiment results and wrote the first draft under the supervision by W. Z. P.A., H. H. and C.S.\u0026nbsp;contributed to the evaluation and discussion of the data. All authors contributed to revision of the manuscript and approved the final version.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eOppenheimer, M. et al. Sea Level Rise and Implications for Low-Lying Islands, Coasts and Communities. Ch. 4 (Cambridge Univ. Pres, 2019).\u003c/li\u003e\n\u003cli\u003eVousdoukas, M.I. et al. Sandy coastlines under threat of erosion. \u003cem\u003eNat. Clim. Chang.\u003c/em\u003e\u003cstrong\u003e10\u003c/strong\u003e, 260-263 (2020). https://doi.org/10.1038/s41558-020-0697-0\u003c/li\u003e\n\u003cli\u003eThi\u0026eacute;blemont, R. et al. Assessing current coastal subsidence at continental scale: Insights from Europe using the European Ground Motion Service. \u003cem\u003eEarth\u0026apos;s Future\u003c/em\u003e\u003cstrong\u003e12\u003c/strong\u003e, e2024EF004523 (2024). https://doi.org/10.1029/2024EF004523\u003c/li\u003e\n\u003cli\u003eNicholls, R.J. et al. A global analysis of subsidence, relative sea-level change and coastal flood exposure. \u003cem\u003eNat. Clim. Chang.\u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, 338-342 (2021). https://doi.org/10.1038/s41558-021-00993-z\u003c/li\u003e\n\u003cli\u003evan de Wal, R. et al. Sea Level Rise in Europe: Impacts and consequences, in: Sea Level Rise in Europe: 1st Assessment Report of the Knowledge Hub on Sea Level Rise (SLRE1). \u003cem\u003eState Planet\u003c/em\u003e\u003cstrong\u003e3-slre1\u003c/strong\u003e, 5 (2024). https://doi.org/10.5194/sp-3-slre1-5-2024\u003c/li\u003e\n\u003cli\u003eKirwan, M., \u0026amp; Megonigal, J. Tidal wetland stability in the face of human impacts and sea-level rise. \u003cem\u003eNature\u003c/em\u003e\u003cstrong\u003e504\u003c/strong\u003e, 53-60 (2013). https://doi.org/10.1038/nature12856\u003c/li\u003e\n\u003cli\u003eSchuerch, M. \u003cem\u003eet al.\u003c/em\u003e Future response of global coastal wetlands to sea-level rise. \u003cem\u003eNature\u003c/em\u003e\u003cstrong\u003e561\u003c/strong\u003e, 231-234 (2018). https://doi.org/10.1038/s41586-018-0476-5\u003c/li\u003e\n\u003cli\u003eBrand, M. W., Buffington, K., Rogers, J.B., Thorne, K., Stein, E. D., \u0026amp; Sanders, B. F. Multi-decadal simulation of marsh topography under sea level rise and episodic sediment loads. \u003cem\u003eJ. Geophys. Res. Earth Surf.\u003c/em\u003e\u003cstrong\u003e127\u003c/strong\u003e, e2021JF006526 (2022). https://doi.org/10.1029/2021JF006526\u003c/li\u003e\n\u003cli\u003eBenninghoff, M. \u0026amp; Winter, C. Recent morphologic evolution of the German Wadden Sea. \u003cem\u003eSci. Rep.\u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 9293 (2019). https://doi.org/10.1038/s41598-019-45683-1\u003c/li\u003e\n\u003cli\u003eCesbron, G., Melet, A., Almar, R., Lifermann, A., Tullot, D. \u0026amp; Crosnier, L. Pan-European Satellite-Derived Coastal Bathymetry\u0026mdash;Review, User Needs and Future Services. \u003cem\u003eFront. Mar. Sci.\u003c/em\u003e\u003cstrong\u003e8\u003c/strong\u003e,740830 (2021). https://doi.org/10.3389/fmars.2021.740830\u003c/li\u003e\n\u003cli\u003eHe, J., Zhang, S., Cui, X. \u0026amp; Feng, W. Remote sensing for shallow bathymetry: a systematic review. \u003cem\u003eEarth Sci. Rev.\u003c/em\u003e\u003cstrong\u003e258\u003c/strong\u003e, 104957 (2024). https://doi.org/10.1016/j.earscirev.2024.104957\u003c/li\u003e\n\u003cli\u003eLi Z. X. et al. Exploring modern bathymetry: A comprehensive review of data acquisition devices, model accuracy, and interpolation techniques for enhanced underwater mapping. \u003cem\u003eFront. Mar. Sci.\u003c/em\u003e\u003cstrong\u003e10, \u003c/strong\u003e1178845 (2023). http://dx.doi.org/10.3389/fmars.2023.1178845\u003c/li\u003e\n\u003cli\u003eW\u0026ouml;lfl A. C. et al. Seafloor mapping - the challenge of a truly global ocean bathymetry. \u003cem\u003eFront. Mar. Sci\u003c/em\u003e. \u003cstrong\u003e6, \u003c/strong\u003e283(2019). https://doi.org/10.3389/fmars.2019.00283\u003c/li\u003e\n\u003cli\u003eSievers, J., Milbradt, P., Ihde, R., Valerius, J., Hagen, R., \u0026amp; Pl\u0026uuml;\u0026szlig;, A. An integrated marine data collection for the German Bight \u0026ndash; Part 1: Subaqueous geomorphology and surface sedimentology (1996\u0026ndash;2016). \u003cem\u003eEarth Syst. Sci. Data\u003c/em\u003e\u003cstrong\u003e13\u003c/strong\u003e, 4053-4065 (2021). https://doi.org/10.5194/essd-13-4053-2021\u003c/li\u003e\n\u003cli\u003eWłodarczyk-Sielicka, M., Bodus-Olkowska, I. \u0026amp; Łącka, M. The process of modelling the elevation surface of a coastal area using the fusion of spatial data from different sensors. \u003cem\u003eOceanologia\u003c/em\u003e\u003cstrong\u003e64\u003c/strong\u003e, 22-34 (2022). https://doi.org/10.1016/j.oceano.2021.08.002\u003c/li\u003e\n\u003cli\u003eBradski, G. The OpenCV library. \u003cem\u003eDr. Dobbs J.\u003c/em\u003e\u003cstrong\u003e25\u003c/strong\u003e, 120-126 (2000).\u003c/li\u003e\n\u003cli\u003eMurray, N.J. et al. The global distribution and trajectory of tidal flats. \u003cem\u003eNature\u003c/em\u003e\u003cstrong\u003e565\u003c/strong\u003e, 222-225 (2019). https://doi.org/10.1038/s41586-018-0805-8\u003c/li\u003e\n\u003cli\u003eHanegan, K.C., FitzGerald, D.M., Georgiou, I.Y. \u0026amp; Hughes, Z.J. Long-term sea level rise modeling of a basin-tidal inlet system reveals sediment sinks. \u003cem\u003eNat Commun\u003c/em\u003e\u003cstrong\u003e14\u003c/strong\u003e, 7117 (2023). https://doi.org/10.1038/s41467-023-42895-y\u003c/li\u003e\n\u003cli\u003eFriedrichs, C. T. Tidal Flat Morphodynamics: A Synthesis. \u003cem\u003eTreatise on Estuarine and Coastal Science\u003c/em\u003e\u003cstrong\u003e3\u003c/strong\u003e, 137-170 (2011).\u003c/li\u003e\n\u003cli\u003eZhang, W. \u0026amp; Arlinghaus, P. Climate, Coast, and Morphology. \u003cem\u003eOxford Research Encyclopedia of Climate Science\u003c/em\u003e. (Oxford University Press, 2022). https://doi.org/10.1093/acrefore/9780190228620.013.814\u003c/li\u003e\n\u003cli\u003eMilbradt, P., Valerius, J. \u0026amp; Zeiler, M. Das funktionale Bodenmodell: Aufbereitung einer konsistenten Datenbasis f\u0026uuml;r die Morphologie und Sedimentologie, \u003cem\u003eDie K\u0026uuml;ste\u003c/em\u003e\u003cstrong\u003e83 AufMod\u003c/strong\u003e, 19-38 (2015). https://doi.org/20.500.11970/101736\u003c/li\u003e\n\u003cli\u003eValerius, J., Feldmann, J., van Zoest, M., Milbradt, P. \u0026amp; Zeiler, M. Documentation of morphological products from the AufMod project Functional Seabed Model, data format: Text files (CSV, XYZ). \u003cem\u003eBundesamt f\u0026uuml;r Seeschiffart und Hydrographie.\u003c/em\u003e\u003cstrong\u003e11pp\u003c/strong\u003e (2013).\u003c/li\u003e\n\u003cli\u003eWahl, T. et al. Observed mean sea level changes around the North Sea coastline from 1800 to present. \u003cem\u003eEarth Sci. Rev.\u003c/em\u003e\u003cstrong\u003e124,\u003c/strong\u003e 51-67 (2013). https://doi.org/10.1016/j.earscirev.2013.05.003\u003c/li\u003e\n\u003cli\u003ePeltier, W.R. Global glacial isostasy and the surface of the ice-age Earth: the ICE-5G(VM2) model and GRACE. \u003cem\u003eAnnu. Rev. Earth Planet. Sci.\u003c/em\u003e\u003cstrong\u003e32\u003c/strong\u003e, 111\u0026ndash;149 (2004). http://dx.doi.org/10.1146/annurev.earth.32.082503.144359\u003c/li\u003e\n\u003cli\u003eRao, K.R., Kim, D.N. \u0026amp; Hwang, J.J. Fast Fourier Transform-Algorithms and Applications. (Springer, Berlin, 2011).\u003c/li\u003e\n\u003cli\u003eKumar, G.G., Sahoo, S.K. \u0026amp; Meher, P.K. 50 Years of FFT Algorithms and Applications. \u003cem\u003eCircuits Syst Signal Process\u003c/em\u003e\u003cstrong\u003e38\u003c/strong\u003e, 5665\u0026ndash;5698 (2019). https://doi.org/10.1007/s00034-019-01136-8\u003c/li\u003e\n\u003cli\u003eDas, A. Interpretation and Processing of Image in Frequency Domain. In: Guide to Signals and Patterns in Image Processing. Springer, Cham (2015).\u003c/li\u003e\n\u003cli\u003eMahavadi, T. F., Seiffert, R., Kelln, J. \u0026amp; Fr\u0026ouml;hle, P. Effects of sea level rise and tidal flat growth on tidal dynamics and geometry of the Elbe estuary, \u003cem\u003eOcean Sci.\u003c/em\u003e\u003cstrong\u003e20\u003c/strong\u003e, 369\u0026ndash;388 (2024). https://doi.org/10.5194/os-20-369-2024\u003c/li\u003e\n\u003cli\u003eArlinghaus, P., Zhang, W. \u0026amp; Schrum, C. Small-scale benthic faunal activities may lead to large-scale morphological change- A model based assessment. \u003cem\u003eFront. Mar. Sci.\u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 1011760 (2022). https://doi.org/10.3389/fmars.2022.1011760\u003c/li\u003e\n\u003cli\u003eH. Nyquist. Certain Topics in Telegraph Transmission Theory. \u003cem\u003eTransactions of the American Institute of Electrical Engineers \u003c/em\u003e\u003cstrong\u003e47, \u003c/strong\u003e617-644 (1928). https://doi.org/10.1109/T-AIEE.1928.5055024\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Helmholtz-Zentrum Geesthacht Centre for Materials and Coastal Research","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Coastal resilience, Sea level rise, Bathymetry mapping","lastPublishedDoi":"10.21203/rs.3.rs-6254537/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6254537/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eLong-term time series of bathymetric data of coastal zones are indispensable for analysing coastal morphological resilience to climate change. Despite the increasing popularity of utilizing high-resolution gridded bathymetric digital elevation models for coastal management, potential errors in analysing the long-term trend of mean elevation change from historical bathymetric datasets spanning a period of multiple years to decades have attracted little attention. Here, we demonstrate that inconsistency in the spatial resolution of small-scale topographic features characterized by sharp bathymetric gradients, such as tidal creeks and streams, could produce an artificial false trend of mean elevation change that is on the same or even higher order of the sea level change rate. Neglecting this inconsistency may lead to a misconception of coastal resilience to sea level rise and misguide planning and implementation of coastal protection strategies. We provide an analytical method to identify such inconsistency in time series of gridded digital elevation models and a homogenization method to minimise the associated errors. Our methods are broadly applicable to reduce errors in bathymetric analysis and improve quantitative assessment of coastal resilience to climate change.\u003c/p\u003e","manuscriptTitle":"Misconception of coastal resilience caused by inconsistent resolution in bathymetry mapping","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-20 09:17:51","doi":"10.21203/rs.3.rs-6254537/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"86cf9dd4-6eae-4349-95e2-51bda335f2a2","owner":[],"postedDate":"March 20th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":45860839,"name":"Geomorphology"}],"tags":[],"updatedAt":"2025-04-03T15:20:09+00:00","versionOfRecord":[],"versionCreatedAt":"2025-03-20 09:17:51","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6254537","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6254537","identity":"rs-6254537","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}