Sea-level rise (SLR), a consequence of climate change, poses significant challenges in the 21st century and beyond. Accurate assessment of populations currently and projected to be affected by SLR is crucial for effective adaptation planning and policy. While numerous studies have examined SLR impacts on coastal areas, estimates of affected populations vary widely, ranging from 88 million to 1.4 billion globally. This discrepancy arises from several factors, including differing definitions of "at-risk" spatial zones and differing temporal horizons associated with those zones. The choice of spatial zone—from mean sea level to the Low-Elevation Coastal Zone (LECZ)—significantly influences exposure estimates and the implied timeframe for impacts. Permanent inundation isn't the only immediate concern; regular tidal flooding (nuisance flooding) poses significant near-term disruptions. Existing studies often focus on a single spatial zone, neglecting the spatio-temporal continuity of SLR impacts. This study aims to address these limitations by developing a more comprehensive approach.

A systematic review of 46 studies assessing populations affected by SLR revealed a wide range of spatial zones used for exposure assessment. Three common zones emerged: (i) complete inundation (future high-tide line), (ii) extreme water levels (100-year floodplain), and (iii) the LECZ. The study notes that these zones imply varying temporal horizons for impact onset. Specified Levels of SLR (complete inundation) assumes impacts only upon permanent submergence. The LECZ provides the broadest estimate, encompassing potential hazards over millennia. Coastal Floodplains (100-year floodplain) represents a gradient of exposure to multiple hazards with a time horizon between these extremes. The review identified the limitations of using singular spatial zones for adaptation planning, as they overlook the spatio-temporal heterogeneity of SLR impacts.

This study utilizes a novel approach called Expected Annual Exposure (EAE) to address the limitations of using single spatial zones. The EAE integrates across the three most common spatial zones (MHHW, 100-year floodplain, and LECZ) into a single, continuous model, focusing on the annual probability of population exposure to flooding. The EAE is calculated by summing the range of annual exposure probabilities over space under a given Representative Concentration Pathway (RCP). This approach considers the annual probability of flooding, incorporating events ranging from frequent nuisance flooding to 100-year floodplains. To incorporate future population changes, the EAE model is combined with sub-county population projections for the United States from 2000 to 2100 under three RCPs (2.6, 4.5, and 8.5) and all five Shared Socioeconomic Pathways (SSPs). The methodology includes: 1. A small-area demographic projection model, using a proportional fitting algorithm and a mixed linear/exponential projection, to produce spatiotemporally consistent Census Block Group (CBG) population projections from 1940–2100, controlled to the SSPs. 2. A digital elevation model (DEM), using airborne lidar-derived data supplemented with USGS data, to classify exposure categories. Levee data is incorporated to refine the DEMs and to remove isolated regions within inundation surfaces. 3. SLR projections and flood event probability surfaces, using probabilistic SLR projections that incorporate local non-climatic factors such as isostatic adjustment and subsidence. Historical storm surge records are used to estimate return level curves at individual tide stations, and applied to all pixels between tide stations using a bathtub model. Generalized Pareto distributions (GPD) are fitted to historical data to estimate the annual probability of extreme flood events exceeding pixel elevations, accounting for SLR projections using Monte Carlo simulations. 4. Exposure computation, where for each CBG, the percentage of pixels on dry land covered by an inundation surface is calculated and multiplied by the projected population. EAE is calculated by multiplying the annual probability of exceedance by the block group's per-pixel population density and summing over all pixels.

The study's key findings highlight the significant and uneven increases in population exposure to coastal flooding due to SLR in the United States. In 2000, just over 600,000 people were exposed to annual flood events. By 2020, this number increased by 60% to 980,000, reflecting both population growth and SLR. Projections for 2100 under SSP2 and RCP 4.5 indicate a 325% increase in EAE to 4.1 million people (with a range of 2.3-6.4 million). The population below the high-tide line is projected to increase over 435% and the population in the 100-year floodplain by 160%. These increases in exposure significantly outpace overall coastal population growth, highlighting the amplified impact of SLR. Exposure unfolds unevenly across the US. In 2000, only two counties had over 100,000 people in the 100-year floodplain; by 2100, nine counties are projected to have 100,000 annually exposed to flooding, and 13 counties with 100,000 in the 100-year floodplain. Comparison of the three common spatial zones shows that counties with similar exposure profiles in one metric may have vastly different profiles in others. This highlights the importance of using multiple zones for a comprehensive assessment. The EAE approach is shown to be particularly useful for identifying the annual increase in populations directly affected by flooding events.

The findings underscore the limitations of relying on single spatial zones for assessing SLR impacts. The significant increases in population exposure projected for the United States, exceeding population growth rates, clearly demonstrate the urgency for adaptation strategies. The uneven distribution of exposure across counties also highlights the need for targeted adaptation planning tailored to specific vulnerabilities and contexts. The EAE metric, by integrating across multiple zones and focusing on annual flood exposure, provides a more nuanced and informative assessment for decision-makers than approaches that rely on a single zone. While the EAE doesn't replace other spatial zone assessments, it offers additional insights into the direct impacts of annual flooding, improving the understanding of current and projected risks.

This study demonstrates the significant and accelerating impact of SLR on coastal populations, particularly in the United States. Using multiple spatial zones for impact assessment, especially incorporating the EAE metric, is crucial for effective adaptation planning. The findings highlight the need for immediate and future adaptation strategies to address not only projected impacts in 2100, but also the currently increasing annual flood exposure affecting millions of coastal residents. Future research should investigate the social heterogeneity of populations across these spatial zones to better understand equity concerns in adaptation planning.

The study employs a bathtub model for flood modeling, which may overestimate exposure by not incorporating wave attenuation or the time it takes for water to reach its full extent. This limitation is acknowledged by the authors, but due to computational constraints, hydrodynamic modeling was not feasible at the scale of the study. The assumption of homogeneous populations within CBGs also presents a limitation; future research should consider intra-CBG social heterogeneity. The study does not account for adaptation measures that may be implemented in the future.