Assessing the vulnerability of buildings to long-term sea level rise across the Global South

doi:10.21203/rs.3.rs-5227062/v1

Download PDF

Article

Assessing the vulnerability of buildings to long-term sea level rise across the Global South

https://doi.org/10.21203/rs.3.rs-5227062/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Future sea levels are expected to rise, resulting in the progressive inundation of coastal cities. Because the spatio-temporal progression of this inundation is complex, few estimates have been made of how sea level rise will impact specific features of the built environment beyond 2100. Here we provide a first-order assessment of the vulnerability of buildings to sea level rise from satellite observation in Africa, Southeast Asia, and South and Central America. We circumvent factors such as local subsidence and ice sheet dynamics by defining an inundation metric as a function of Local Sea Level Rise (LSLR), rather than time. Of the 840 million buildings in the study region, we find ~ 3.0 million at risk of inundation with 0.5m LSLR, ~ 45 million with 5m LSLR, and ~ 136 million with 20m LSLR. Our results highlight geographic variability in vulnerability and demonstrate the benefits that low-emissions pathways imply for preserving built environments.

Earth and environmental sciences/Climate sciences

Earth and environmental sciences/Environmental sciences

Earth and environmental sciences/Natural hazards

Earth and environmental sciences/Planetary science

Scientific community and society/Developing world

Scientific community and society/Geography

Social science/Environmental studies

Sea level rise (SLR) poses a prominent challenge to current and future generations, impacting the habitability of coastlines, trade routes, and major infrastructure systems near the ocean^1,2. Projected SLR will have economic consequences at local, regional and global scales, disrupting the availability of land and material resources, and forcing migration³. Nearly 40% of the global population lives within 100 km of the coast⁴, and the sizes of communities exposed to coastal flooding are rapidly increasing^5,6. While people can relocate relatively quickly, infrastructure cannot, and with a minimum of 0.5 m of global mean SLR projected by 2150, significant loss of coastal buildings, roads and other infrastructure is already inevitable¹.

The inundation of the built environment will not stop in 2150⁷. Because of the slow response timescale of the polar ice sheets to climate change, the incursion of coastal waters into built environments is projected to continue for many centuries. Although the built environment will continue to evolve, many coastal settlements are thousands of years old, including megacities such as Bangkok, and individual buildings can easily last centuries⁸. The fact that LSLR and the built environment both change on multi-century timescales underscores the importance of comparing the two directly over long time horizons^9,10. Prior studies have provided insight on how buildings could be impacted, including for example New Zealand’s cumulative building exposure to LSLR¹¹ and the flooding of affordable housing in the United States of America over the next few decades¹². These works have begun to paint a picture of local challenges to the built environment that will be caused by LSLR, but have been focused on relatively immediate time scales and restricted geographic scope. Furthermore, such detailed assessments are typically not available for low-income nations.

Projecting the long-term impact of future LSLR on coastal infrastructure for a given year in the future is challenging due to spatial variability and the uncertain timing and magnitude of future sea level changes. Current multi-century SLR projections along global coastlines differ substantially due to the wide range of possible future greenhouse gas emissions trajectories, uncertainty in the response of the Earth system to climate change, and complex local processes such as subsidence and erosion¹. For example, the IPCC’s 6th Assessment report² projected that global mean SLR in the year 2300 would be in the range 0.5–16.2 m higher than present (defined as a 1995–2014 baseline), a range so large that it is difficult for coastal practitioners and policy makers to interpret meaningfully ¹³. More recent literature suggests that the high end of this range could be even higher due to profound uncertainty in ice sheet processes and feedbacks^14,15. Furthermore, uncertainties in local sea level projections on multi-century timescales can be even larger than with projections of global mean SLR¹⁶, differing from site to site due to the spatially variable nature of contributing processes, as well as unequal spatial coverage of local sea level observations such as tide gauges that serve as input to local projections¹. Uncertainty can be further increased due to geophysical feedbacks that arise between LSLR and other relevant processes, such as tides¹⁷, that in turn require regional analysis to accurately assess¹⁸. Finally, understanding the potentially substantial impact of local processes to LSLR such as land subsidence has not yet been quantified across a regional scale and is not robustly included in global-scale spatially variable projections. For example, in the Jakarta Basin region in Indonesia, excessive groundwater pumping to meet residential and industrial needs is expected to result in seawater intrusion and land subsidence, causing inundation to occur before the current elevation values would estimate¹⁹. Together, these factors have left huge uncertainties on the amount of LSLR that can be expected at any particular location, by any particular year in the future, impeding progress in understanding the potential impacts on coastal settlements.

Even if the future time-trajectory of local changes in sea level were well-constrained, quantifying local inundation patterns demands highly resolved data on coastal settlements that has previously been lacking for much of the world. The required level of topographic and hypsometric detail can be confounded by trees and buildings in remote sensing data products, which can result in a systemic overestimation of the elevation within densely developed regions²⁰. Connecting patterns of inundation to direct impacts on infrastructure poses an additional challenge, as it requires determining the precise location and elevation of infrastructure at the scale of individual buildings, over vast spans of coastline. Fortunately, recent innovations in remote sensing and machine learning have developed greatly improved data products with which to assess the vulnerability of coastal infrastructure. Hawker et al (2022)²⁰ released the first global high resolution topography dataset which removed building and tree height bias, a necessary step to avoid systematically underestimating the water inundation risk per meter of LSLR in urban areas. Meanwhile, high precision building footprint data has recently become available from Google, providing the location of billions of buildings with current coverage across the Global South (see Supplemental SI for a comprehensive list of regions)²¹. Together, these new data sources allow for a major advance to be made beyond previous work such as Ning et al. (2021)²², Kulp & Strauss (2019)²³ and Bhuyan (2022)²⁴, which was more limited in geographical scope and/or specificity of results due to less precise data.

Here, we take advantage of these new datasets, together with a novel approach that circumvents much of the uncertainty in time-trajectories of LSLR, to assess the risk posed to coastal buildings per meter of LSLR throughout the Global South. Given the substantial uncertainties and evolving state of knowledge on the magnitude and timing of future SLR at specific locations, we focus on assessing the relative vulnerability of coastal infrastructure to LSLR, independent of temporal scale. This shift in focus allows us to produce a broad perspective that can be interpreted with ongoing local, regional and global SLR projection efforts to fit into the adaptive decision-making frameworks²⁵ that are typically applied in coastal adaptation efforts.

In order to estimate building inundation risk, the area, elevation, country, and location of each building centroid were retrieved from Google Earth Engine (see Methods). Using the centroid of each building we calculate its corresponding elevation, relative to local high water level (i.e. the water level at high tide). If this centroid falls below the local high water level with a given amount of LSLR, it is identified as ‘inundated’. Note that this definition does not account for the possibility that flooding of the base of a building could occur at lower levels of local mean sea level during extreme events such as storm surges. We also cannot account for unpredictable potential human measures such as the construction of flood protection. We therefore refer to this as an inundation metric, rather than a firm prediction. However, our metric is straightforward and easy to interpret in terms of existing infrastructure. The high-precision nature of the data allows for a scalable analysis of the physical consequences of emissions pathways to assess how future SLR will change global, regional and local spaces. The resulting analysis provides a new, broad, long-term vision of what SLR could mean for reshaping settlements across the Global South, without being inhibited by the unavoidable uncertainty of how much flooding will occur in a given year. Our result is less useful for short-term property management decisions than a decadal-timescale forecast, but it is highly relevant to long-lived infrastructure planning.

Distribution of inundated buildings across spatial scales

Our results show that the vulnerability of current infrastructure to LSLR (Fig. 1) varies spatially across coastal areas. Regions where significant, regular inundation is predicted with as little as 0.5m SLR include Northeastern Africa, and the majority of Southeast Asia and the Caribbean. The high concentration of buildings along the banks of major estuaries and deltas such as the Río de la Plata, Amazon, Gambia, and Nile rivers result in dramatic inland inundations compared to other regions, most prominent for SLR greater than 5 m. Note that because the gridded data are assembled building by building, results can be synthesized at a range of spatial scales (Fig. 1b and c).

Our analysis indicates pronounced nonlinearity in the rate at which buildings are inundated as a function of LSLR (Fig. 2). This nonlinearity cannot be accounted for alone by the extent of land area in each region due to the complexities that land use, population, economic activity, materials, and culture contribute to determining how urban environments develop (see Supplemental SII for more information). All regions show a relatively slow rate of loss at less than 2 m LSLR, rapidly increasing between 2 and 4 m, particularly in Southeast Asia. Of the 840 million buildings in the study region, ~ 3.0 million buildings are inundated by our metric with 0.5 m of LSLR, a value which is projected as the lower bound for global mean SLR by 2100 even under low emissions scenarios¹. In the early stages of LSLR, Africa is the continent with the highest number of buildings impacted in the dataset, but as LSLR intensifies, Southeast Asia comes to rapidly dominate the number of buildings lost (Fig. 2). The number of buildings inundated increases by nearly two orders of magnitude to 136 million buildings with 20 m of LSR. In terms of area, 305 million m² of current building footprints are at risk of being inundated with 0.5 m LSLR, increasing to 1.1 billion m² with 2 m LSR. Our estimates necessarily exclude the majority of North America, Europe, Oceana, Australia and the remainder of mainland Asia, as they are not included in the building dataset at present.

Impact on buildings across countries

Turning now to the country level, we find that for the majority of countries, 0.5 m of LSLR inundates only a small proportion of the current building stock centroids (Fig. 3), amounting to ~ 1% of all buildings in most coastal countries. The disparity between countries intensifies rapidly as LSLR increases, with 5 m of LSLR resulting in more than 80% loss of building stock in some countries (Fig. 3).

Our results provide what is, to our knowledge, the first large-scale assessment of the vulnerability of individually mapped buildings to inundation by large magnitudes of SLR. We find that the exposure of building stock varies widely between coastal countries, depending on the relationship between topography and building distributions. There is relatively little inundation for magnitudes of SLR that have been commonly considered (0.5–1 m), but the rate of inundation increases rapidly throughout much of the study area between 2 and 4 m of LSLR. Our results therefore highlight the large difference in building inundation likely to occur under low vs. high-emissions scenarios.

Previous research has estimated that 0.7–0.9 m global mean SLR is likely to occur by 2100, and 2.2–2.5 m by 2300, if the objectives of the Paris Agreement are met¹³. This magnitude of LSLR across coastlines in the study area alone would inundate on the order of 5 million existing buildings at high tide by the end of this century, approaching 20 million buildings by 2300. Importantly, this simple estimate does not account for geophysical and oceanographic effects (e.g. the effects of gravity, Earth rotation and deformation initiated by land-ice loss) that will cause LSLR to exceed the global mean across the majority of locations within the study area^7,26. Furthermore, non-climatic local factors such as erosion and subsidence will also increase LSLR in many places during this time frame. Exceeding temperature targets set by the Paris Agreement would only worsen this, having significant impacts on the resiliency of coastal infrastructure.

Combining our findings with sea level projections may provide an opportunity for policymakers and urban developers to anticipate future changes and plan accordingly. Depending on the extent of sea water inundation, there is potential for adaptation in some locations through the use of flood protection or land reclamation^27,28, which would decrease the amount of infrastructure at risk of inundation from LSLR. However, the long-term feasibility and costs of such measures are strongly dependent on the ultimate magnitude of SLR, which depends on the emissions trajectory²⁹. Our results may help to inform decisions between adaptation and retreat. We provide an interactive map of our results on Google Earth as a part of this publication to facilitate this (see Data and Code Availability). Note that while we do not consider extreme SLR events directly in this study, our gridded infrastructure-elevation mapping results could be modified to assess vulnerability of infrastructure to projected short-term flooding as well.

Multiple datasets of varying resolutions were used in our work, each with their own uncertainty, including the tidal range estimates, building location and floor area, and topographic elevation^11,12. In terms of tidal ranges, differences between tide gauge measurements and the FES14 tidal model predictions we adopt are less than 7 cm for all wave types, which is not significant on the multi-meter scale addressed here^30–32. To validate our use of high tide, an assessment of the extent of areas identified prematurely as flooded at 0 m LSLR, and was found to be > > 1%, meaning this assumption is likely not the cause of any errors in our results (details can be found in Supplemental SIV). Additionally, Tidal ranges are dynamic in nature and there is evidence to support that regional tidal variations will intensify with LSLR in some places and diminish in others¹⁷. Therefore, changes in the height of high tide should be considered when combining our results with LSLR projections. In terms of topography, the mean error for FABDEM is reported to be 0.1–0.5 m depending upon landscape, with highest errors occurring where there is high density canopy cover. Additionally there is limitations for FABDEM in areas without forest that have steep slopes, for which the reanalysis FABDEM conducted on the original COPDEM30 model will not apply. Finally, the existence, location, and size of the buildings contributes to uncertainty. Sirko et al. (2021)²¹ reports 95% precision for the Open Buildings Polygon Dataset, however due to it’s novelty, to our knowledge there is currently no peer-reviewed third-party comparison of this dataset. In order to validate our findings, we conducted a preliminary accuracy assessment (see Supplemental SIII for more details) and found the accuracy to be much higher than reported by Sirko et al.

At large scales it is likely that these primary sources of error are correlated and more likely to occur in similar locations, which should minimize their impact on aggregate measures. However, the uncertainties will have a larger effect on smaller scale, local results, particularly in areas with highly variable tides and regions where building roofs are made of visually similar material to the ground²¹, to a degree which is difficult to assess at present.

In this study, we adopt the simple approach of considering exposure as a function of LSLR rather than as a function of time to avoid tying our results to uncertain climate projections. However, it is important to emphasize that our metric does not estimate vulnerability to levels of global mean SLR, as LSLR and flood risks are spatiotemporally variable. For example, loss of ice from the Greenland and Antarctic ice sheets in the coming centuries will result in gravitational, Earth effects that will produce a non-uniform pattern of sea level change, with mid and low latitude regions far from the polar ice sheets expected to experience the greatest rise relative to the mean global value^7,33,34. As a result, the exposure for each country by meter of LSLR (Fig. 3) will not occur simultaneously and will accentuate the already present issue of inequality between mid and high-latitude impacts evident in Fig. 3²⁶.

Given that SLR can amplify the impacts of other processes such as high tide extent, subsidence, and storm surges³⁵, damage to coastal infrastructure may occur well before the area is under water at high tide. Subsidence, erosion, and tidal intensification will likely lead to the destruction of more buildings than our metric predicts. Future work could incorporate spatiotemporally variable sea level projections and extreme events to produce global maps that would provide a more accurate assessment of impact on buildings for a given scenario. We also note that, although we have focused on buildings, inundation of other forms of infrastructure may threaten the functionality of coastal communities, and their roles within the broader economy. These risks could be assessed by comparing our results with data on waste containment, transportation infrastructure, energy infrastructure, cultural monuments, or other demographic data at local and regional scales where available.

SLR will pose a major challenge to coastal societies throughout the coming centuries, impacting hundreds of millions of currently existing buildings and the lives and livelihoods of their residents. Our results provide a novel perspective on the impact of SLR scenarios across spatial scales, by mapping building inundation at 30 m resolution across hundreds of thousands of kilometers of coastline³⁶. Due to the high resolution of the datasets and the broad geographical scope, the results can be assessed from an area spanning multiple countries, to a single building on a city block, making our findings potentially useful for informing large and small-scale efforts for building resiliency against climate change impacts. As our results were not tied to any specific Representative Concentration Pathway (RCP) / Shared Socioeconomic Pathway (SSP) climate warming scenarios, our findings will remain relevant even as projections change with improved data and techniques^37,38.

We made use of Google’s ‘Open Buildings’ V2 data²¹ to quantify the number of buildings at risk of water inundation at high tide per meter of SLR. Google’s Open Buildings data is the result of a new learning model trained using 50-cm resolution satellite images to identify buildings in Africa, Latin America, Caribbean, South Asia and Southeast Asia. It currently spans 58 million square km and is being regularly updated with new areas. For the purposes of this research, we limited our scope in Southeast Asia to islands, as the data for mainland areas in Southeast Asia are currently incomplete.

To determine the elevation of each building identified from the Open Buildings dataset, we selected the Forest and Buildings Removed Digital Elevation Model (FABDEM). FABDEM is a new DEM that used machine learning to produce a bare-earth version of the Copernicus Digital Elevation Model DEMGLO COP30²⁰. Using EGM2008 WSG84 geoid model, FABDEM is the first 1 arc second global map of elevation with buildings and trees removed. The algorithm was trained on reference elevation data from 12 countries, covering a range of climate zones and urban extents. The accuracy of FABDEM has consistently matched or outperformed other global topography datasets of similar resolution using multiple comparison techniques, including comparison to differential GPS data³⁹, UAV-lidar derived elevation⁴⁰, field surveys⁴¹, random block design⁴², and ICESat-2 photon data⁴³. The removal of building height in FABDEM’s dataset is particularly valuable, as our goal is to assess whether the ground floor of buildings will be inundated for a given amount of sea level rise.

To account for tidal amplitudes, we adopted a similar technique to Hawker et al (2022)²⁰ by using the FES14 global tidal model to adjust the elevation of shoreline areas (< 20 m elevation) to reflect high tide rather than mean sea level. The Finite Element Solution (FES) 2014 is based upon the Toulouse Unstructured Grid Ocean model³¹ and provides tide amplitude a 1/16-degree global grid (~ 6.87 km at the equator). FES14 is a longstanding industry standard tidal model and has been selected for the CNES/NASA/ESA/EUMETSAT operational and reprocessing altimetry data de-aliasing correction and as the standard correction in ITRF2020 conventions by the International Earth Rotation and Reference Systems Service. Comparisons between FES14 and other global tidal models show that FES14 outperforms other options in tidal gauge tests for nearly all wave types³¹. FES14 uses the International Terrestrial Reference System, which is compatible with the WSG84 model that FABDEM uses for mean sea level⁴⁴ and has a root-mean-square accuracy of 6.44 cm on coastal shelves³¹. We added half of the maximum tidal amplitude observed in an 18-year period between 2003–2020 to the elevation on the coastlines. Erosion, storm surges, and tidal intensification are not considered, making these results a minimum. Future development patterns for buildings were also not considered.

Utilizing the above data, we calculated the centroid of each building and determined its corresponding elevation from the tidally adjusted FABDEM dataset. Using this centroid, we calculate each buildings corresponding elevation, relative to local high water level. If this centroid falls below the local high tide level with a given amount of LSLR, it is identified as ‘inundated’. Our analysis assumed no systematic bias in the elevation of building corners, so the mean over large numbers of buildings, the centroid elevations serve as a reliable metric. The area in square meters, elevation, country, and latitude-longitude of the centroid of each building were batch exported from Google Earth Engine Code Editor using a Python API. As the IPCC 6th assessment report projected SLR ranges from 0.5–16 m above preindustrial levels by 2300¹, our results quantified the number of existing buildings that will fall below high tide as a function of SLR over 0.5-20m. In order to convey spatial results, SLR ranges were binned in values expressed for different IPCC scenarios, both by Fox-Kemper et al. (2021)¹ for short-term projections, and DeConto et al. (2021)⁴⁵ for longer-term impacts including ice-sheet interactions.

Author Contribution

All authors devised the project, the main conceptual ideas and proof outline. M.W.S, N.G., and E.G. designed the sea level rise methodology and J.C. and M.W.S designed the spatial analysis methodology. M.W.S. performed the computations, and M.W.S. and E.G. prepared the figures. All authors contributed to writing the original draft and revising the manuscript.

Acknowledgement

We would like to acknowledge Lauryn Talbot for her contributions to the early stages of this project during her undergraduate research thesis.

Data & Code Availability

To provide a comprehensive picture of our results on local to regional and continental scales, we created an interactive global map including the location and orientation of the buildings, and a mask of seawater inundation for a given SLR scenario through the Google Earth Engine Interface. The raw data and Python scripts for this research will be made publicly available on Zenodo prior to publication.

Fox-Kemper, B. et al. Climate Change 2021: The Physical Science Basis. Working Group I. Ocean, Cryosphere, and Sea Level Change. 1211–1362 (2021) doi:10.1017/9781009157896.001.
Intergovernmental Panel On Climate Change (Ipcc). The Ocean and Cryosphere in a Changing Climate: Special Report of the Intergovernmental Panel on Climate Change. (Cambridge University Press, 2022). doi:10.1017/9781009157964.
Oppenheimer, Michael & Glavovic, Bruce C. IPCC Sixth Assessment Report Chapter 4: Sea Level Rise and Implications for Low-Lying Islands, Coasts and Communities. (2021) doi:10.1017/9781009157964.
Kummu, M. et al. Over the hills and further away from coast: global geospatial patterns of human and environment over the 20th–21st centuries. Environ. Res. Lett. 11, 034010 (2016).
Neumann, B., Vafeidis, A. T., Zimmermann, J. & Nicholls, R. J. Future Coastal Population Growth and Exposure to Sea-Level Rise and Coastal Flooding - A Global Assessment. PLOS ONE 10, e0118571 (2015).
Rentschler, J. et al. Global evidence of rapid urban growth in flood zones since 1985. Nature 622, 87–92 (2023).
Golledge, N. R. et al. Global environmental consequences of twenty-first-century ice-sheet melt. Nature 566, 65–72 (2019).
Andersen, R. & Negendahl, K. Lifespan prediction of existing building typologies. Journal of Building Engineering 65, 105696 (2023).
Lyon, C. et al. Climate change research and action must look beyond 2100. Global Change Biology 28, 349–361 (2022).
Magnan, A. K. et al. Sea level rise risks and societal adaptation benefits in low-lying coastal areas. Sci Rep 12, 10677 (2022).
Paulik, R., Stephens, S., Wild, A., Wadhwa, S. & Bell, R. G. Cumulative building exposure to extreme sea level flooding in coastal urban areas. International Journal of Disaster Risk Reduction 66, 102612 (2021).
Buchanan, M. K. et al. Sea level rise and coastal flooding threaten affordable housing. Environ. Res. Lett. 15, 124020 (2020).
van de Wal, R. S. W. et al. A High-End Estimate of Sea Level Rise for Practitioners. Earth’s Future 10, e2022EF002751 (2022).
Seroussi, H. et al. ISMIP6 Antarctica: A multi-model ensemble of the Antarctic ice sheet evolution over the 21st century. The Cryosphere 14, 3033–3070 (2020).
Aschwanden, A., Bartholomaus, T. C., Brinkerhoff, D. J. & Truffer, M. Brief communication: A roadmap towards credible projections of ice sheet contribution to sea level. The Cryosphere 15, 5705–5715 (2021).
Nauels, A., Meinshausen, M., Mengel, M., Lorbacher, K. & Wigley, T. M. L. Synthesizing long-term sea level rise projections – the MAGICC sea level model v2.0. Geoscientific Model Development 10, 2495–2524 (2017).
Wilmes, S.-B., Green, J. A. M., Gomez, N., Rippeth, T. P. & Lau, H. Global Tidal Impacts of Large-Scale Ice Sheet Collapses. Journal of Geophysical Research: Oceans 122, 8354–8370 (2017).
Hayden, A.-M. et al. Multi-Century Impacts of Ice Sheet Retreat on Sea Level and Ocean Tides in Hudson Bay. Journal of Geophysical Research: Oceans 125, e2019JC015104 (2020).
Abidin, H. Z. et al. Land subsidence characteristics of the Jakarta basin (Indonesia) and its relation with groundwater extraction and sea level rise. in Groundwater Response to Changing Climate (CRC Press, 2010).
Hawker, L. et al. A 30 m global map of elevation with forests and buildings removed. Environ. Res. Lett. 17, 024016 (2022).
Sirko, W. et al. Continental-Scale Building Detection from High Resolution Satellite Imagery. (2021) doi:10.48550/arXiv.2107.12283.
Ning, H. et al. Exploring the vertical dimension of street view image based on deep learning: a case study on lowest floor elevation estimation. International Journal of Geographical Information Science 36, 1317–1342 (2022).
Kulp, S. A. & Strauss, B. H. New elevation data triple estimates of global vulnerability to sea-level rise and coastal flooding. Nat Commun 10, 4844 (2019).
Bhuyan, K., Van Westen, C., Wang, J. & Meena, S. R. Mapping and characterising buildings for flood exposure analysis using open-source data and artificial intelligence. Nat Hazards (2022) doi:10.1007/s11069-022-05612-4.
Völz, V. & Hinkel, J. Sea Level Rise Learning Scenarios for Adaptive Decision-Making Based on IPCC AR6. Earth’s Future 11, e2023EF003662 (2023).
Sadai, S., Spector, R. A., DeConto, R. & Gomez, N. The Paris Agreement and Climate Justice: Inequitable Impacts of Sea Level Rise Associated With Temperature Targets. Earth’s Future 10, e2022EF002940 (2022).
Hinkel, J. et al. Coastal flood damage and adaptation costs under 21st century sea-level rise. Proceedings of the National Academy of Sciences 111, 3292–3297 (2014).
Siders, A. R., Hino, M. & Mach, K. J. The case for strategic and managed climate retreat. Science 365, 761–763 (2019).
Tamura, M., Kumano, N., Yotsukuri, M. & Yokoki, H. Global assessment of the effectiveness of adaptation in coastal areas based on RCP/SSP scenarios. Climatic Change 152, 363–377 (2019).
Edward D. Zaron & Shane Elipot. An Assessment of Global Ocean Barotropic Tide Models Using Geodetic Mission Altimetry and Surface Drifters. Journal of Physical Oceanography 51, (2020).
Lyard, F. H., Allain, D. J., Cancet, M., Carrère, L. & Picot, N. FES2014 global ocean tide atlas: design and performance. (2021) doi:10.5194/os-17-615-2021.
Ranji, Z., Hejazi, K., Soltanpour, M. & Allahyar, M. R. INTER-COMPARISON OF RECENT TIDE MODELS FOR THE PERSIAN GULF AND OMAN SEA. Coastal Engineering Proceedings 9–9 (2016) doi:10.9753/icce.v35.currents.9.
Mitrovica, J. X., Tamisiea, M. E., Davis, J. L. & Milne, G. A. Recent mass balance of polar ice sheets inferred from patterns of global sea-level change. Nature 409, 1026–1029 (2001).
Roffman, J., Gomez, N., Yousefi, M., Han, H. K. & Nowicki, S. Spatial and temporal variability of 21st century sea level changes. Geophysical Journal International 235, 342–352 (2023).
Vitousek, S. et al. Doubling of coastal flooding frequency within decades due to sea-level rise. Sci Rep 7, 1399 (2017).
C, L., R X, S. & Y H, Z. 2015: How Many Islands (Isles, Rocks), How Large Land Areas and How Long of Shorelines in the World? ——Vector Data based on Google Earth Images. Journal of Global Change Data & Discovery 3, 124–148 (2019).
Riahi, K. et al. The RCP greenhouse gas concentrations and their extensions from. Global Environmental Change 42, 153–168 (2017).
Meinshausen, M. et al. The RCP greenhouse gas concentrations and their extensions from 1765 to 2300. Climatic Change 109, 213 (2011).
Bhardwaj, A. Assessment of FABDEM on the Different Types of Topographic Regions in India Using Differential GPS Data. Engineering Proceedings 27, 79 (2022).
Marsh, C. B., Harder, P. & Pomeroy, J. W. Validation of FABDEM, a global bare-earth elevation model, against UAV-lidar derived elevation in a complex forested mountain catchment. Environ. Res. Commun. 5, 031009 (2023).
Vernimmen, R. & Hooijer, A. New LiDAR-Based Elevation Model Shows Greatest Increase in Global Coastal Exposure to Flooding to Be Caused by Early-Stage Sea-Level Rise. Earth’s Future 11, e2022EF002880 (2023).
Bielski, C., López-Vázquez, C., Grohmann, C. H., Guth, P. L. & Group, T. T. D. W. Novel approach for ranking DEMs: Copernicus DEMs improve open global topography. Preprint at https://doi.org/10.48550/arXiv.2302.08425 (2023).
Dandabathula, G., Hari, R., Ghosh, K., Bera, A. K. & Srivastav, S. K. Accuracy assessment of digital bare-earth model using ICESat-2 photons: analysis of the FABDEM. Model. Earth Syst. Environ. 9, 2677–2694 (2023).
GWG World Geodetic System and Geomatics Focus Group, Chair. Department of Defense World Geodetic System 1984. (2014).
DeConto, R. M. et al. The Paris Climate Agreement and future sea-level rise from Antarctica. Nature 593, 83–89 (2021).

No competing interests reported.

NPJUrbanSustainabilitytransferNCLIM23112492Supplementary.docx

Download PDF

Reviewers invited by journal
30 Oct, 2024
Editor assigned by journal
23 Oct, 2024
Submission checks completed at journal
15 Oct, 2024
First submitted to journal
08 Oct, 2024

You are reading this latest preprint version

Assessing the vulnerability of buildings to long-term sea level rise across the Global South

Status:

Version 1

Abstract

Figures

Introduction

Results

Distribution of inundated buildings across spatial scales

Impact on buildings across countries

Discussion

Methods

Declarations

Author Contribution

Acknowledgement

References

Additional Declarations

Supplementary Files

Status:

Version 1