High Tide Flooding (HTF) is a present and increasing hazard for coastal communities across the United States. NOAA provides HTF outlooks at U.S. tide gauges, however, many coastal communities lie relatively far from a tide gauge and therefore currently lack localized HTF guidance. In this study, we demonstrate an approach to generate spatially-continuous daily predictions of HTF at 400-500 m resolution out to a year into the future, by combining NOAA's monthly HTF outlook framework with the newly-released Coastal Ocean Reanalysis (CORA). Using CORA to derive daily HTF predictions at tide gauges, as compared to using gauge observations, results in average HTF model skill reduction of ≤5% using three different statistical metrics at one month lead time. Further, stations which obtain statistically skillful HTF predictions using gauge data also do so using CORA for 94% of cases. The results suggest that CORA could enable skillful HTF predictions away from tide gauges, supporting the possibility of providing high resolution HTF outlooks for much of the U.S. coastline. The potential value of these spatially continuous HTF predictions is illustrated by identifying communities near Charleston S.C. with different CORA-derived local HTF risk than that provided by the closest tide gauge. Finally, we describe outstanding questions and needs for the scaling of these results to an operational national-scale monthly HTF outlook.

Communities globally are experiencing an increase in high tide flood (HTF) frequency. The present-day impact of HTF for communities is expansive and recurrent, ranging from disrupted activities for infrastructure, inundated stormwater and wastewater systems, and increased public health hazards. Accurate estimates of the probability density functions (PDFs), especially for extreme water levels, are essential for quantifying risks of coastal flooding. In this work, we decompose still water levels measured at 148 tide gauge stations along the United States’ coasts and evaluate the characteristics of the nontidal residual (NTR) distributions. We compare the distribution of high-pass filtered water levels (hourly anomalies) to PDFs of a first-order autoregressive (AR1) process resulting in a Gaussian (normal) distribution and a non-Gaussian (skewed and heavy tailed) ‘Stochastically Generated Skewed’ (SGS) distribution that includes correlated additive and multiplicative noise (CAM noise). We find that the overall error computed between the PDFs and the observed anomalies is reduced at most stations when using the non-Gaussian PDF compared to the AR1 for both the bulk of the distribution and extreme values. We also show that the non-Gaussian SGS distribution is more robust at capturing extreme values in the case of sparse observations, compared to other distributions (kernel density) and extreme value analysis methods (i.e., Generalized Extreme Value and Generalized Pareto Distribution). Our non-Gaussian PDF allows us to diagnose how the shape of the distribution may evolve with climate change. Findings from this work will be implemented in the National Oceanic and Atmospheric Administration’s HTF monthly predictions and used to evaluate changes in forecast skill. This work has relevance for high tide flooding forecasts along the coast and inundation mitigation strategies, as well as estimating PDFs for other physical variables that exhibit heavy-tailed skewed distributions.

Forecasting seasonal sea levels along many coasts remains challenging, with generally lower skills than forecasts for the open oceans. We investigate the influence of ocean dynamics on forecasting monthly sea level anomalies for the United States Gulf Coast and East Coast using the Estimating Circulation and Climate of the Ocean (ECCO) system, which is initialized monthly from 1992 through 2017 and runs forward for 12 months under climatological atmospheric forcing. This approach, which we refer to as an ocean dynamic persistence forecast, demonstrates improved skill compared to both observed damped persistence and the ECMWF SEAS5 climate forecast system when evaluated against observations. At a lead of 4 months, dynamic persistence has the highest anomaly correlation coefficients at 22 out of 39 coastal locations (mostly south of Cape Hatteras). However, improvement in root mean square error is minimal, possibly due to reduced variability in ECCO associated with its climatology forcing and coarse resolution. This study suggests that dynamic persistence offers the potential to improve sea level forecasts beyond the capabilities of damped persistence and a state-of-the-art climate model.

Our ability to characterize and quantify the complex uncertainties surrounding future sea-level changes is crucial for coastal risk assessments and adaptation strategies. This study focuses on the role of steric and dynamic changes (i.e., sterodynamics) in sea level projections, particularly regarding their contribution to the uncertainty of global and regional sea level changes in relation to other components such as ice sheet dynamics. A probabilistic framework is used to estimate probability distributions of sea-level change for each component. Through variance decomposition, the total uncertainty in sea-level change is dissected into its constituent sources. Subsequently, the relative contribution of sterodynamics uncertainty is quantified across various regions, time frames, emission scenarios, and projection methodologies utilized to estimate future sea-level distributions. The contribution of sterodynamics to overall uncertainty reduces over time as the contribution from ice sheets becomes more pronounced. The spatiotemporal pattern of sterodynamic significance is not strongly dependent on future greenhouse gas emissions, yet its overall role is highly dependent on the representation (e.g., emulation) of ice sheets. When high-end, low-probability estimates of future Antarctic ice sheet contributions are excluded, sterodynamics remain a dominant source of regional sea-level uncertainty at the end of this century, particularly along the US East Coast and European coast. These regions are also identified as hotspots for future sea-level rise, indicating that sterodynamic processes will play a significant role in assessing coastal vulnerabilities there. This study suggests that ocean model development can most effectively reduce the overall uncertainty in future sea-level projections by focusing on these areas.

Coastal vertical land motion (VLM), including uplift and subsidence, can greatly alter relative sea level projections and flood mitigations plans. Yet, current projection frameworks, such as the IPCC Sixth Assessment Report, often underestimate VLM by relying on regional linear estimates. Using high-resolution (90-meter) satellite data from 2015 to 2023, we provide local VLM estimates for California and assess their contribution to sea level rise both now and in future. Our findings reveal that regional estimates substantially understate sea level rise in parts of San Francisco and Los Angeles, projecting more than double the expected rise by 2050. Additionally, temporally variable (nonlinear) VLM, driven by factors such as hydrocarbon and groundwater extraction, can increase uncertainties in 2050 projections by up to 0.4 meters in certain areas of Los Angeles and San Diego. This study highlights the critical need to include local VLM and its uncertainties in sea level rise assessments to improve coastal management and ensure effective adaptation efforts.

High tide flooding (HTF) occurs when astronomically driven water levels rise above flooding thresholds in coastal areas, which can happen on sunny days. In a warming climate, sea-level rise (SLR) is expected to change the frequency of HTF via a direct non-linear change in the mean water level. In this study, we investigate the impacts of SLR on HTF along the Gulf and Atlantic Coasts of the United States. We quantify the extent to which SLR is expected to exacerbate the HTF regime in the coming decades if no more flood protection is implemented. We estimate SLR at 10 km intervals using regression models, add it to numerically simulated tidal levels, and compare the results with estimated HTF thresholds. Our results provide continuous spatial coverage along the Atlantic and Gulf Coasts, showing a projected average rise of 0.35 +/- 0.10 m in tidal levels above the mean higher high water (MHHW) by 2050, whereas Chesapeake Bay is projected to experience a greater rise of 0.39 +/- 0.05 m, and Maine is expected to see a lower increase of 0.27 +/- 0.08 m. This yields an increase of 10 days/year and 110 days/year in HTF hours in the years 2050 and 2100, respectively. Moreover, Pamlico Sound and Chesapeake Bay are expected to experience the most significant changes in HTF frequency, with more than 90 days of HTF by 2050s. Our results show that HTF regime in mesotidal semidiurnal systems are, on average, more sensitive to the projected SLR than the rest.

Developing predictions of coastal flooding risk on subseasonal timescales (2-6 weeks in advance) is an emerging priority for the National Oceanic and Atmospheric Administration (NOAA). In this study, we assess the ability of two current operational forecast systems, the European Centre for Medium-Range Weather Forecasts Integrated Forecasting System (IFS) and the Centre National de Recherches M & eacute;t & eacute;orologiques climate model (CNRM), to make subseasonal ensemble predictions of the non-tidal residual component of coastal water levels at United States coastal gauge stations for the period 2000-2019. These models were chosen because they assimilate satellite altimetry at forecast initialization and attempt to predict the mean sea level, including a global mean component whose absence in other forecast systems complicates assessment of tide gauge reforecast skill. Both forecast systems have skill that exceeds damped persistence for forecast leads through 2-3 weeks, with IFS skill exceeding damped persistence for leads up to 6 weeks. Post-processing forecasts to include the inverse barometer effect, derived from mean sea level pressure forecasts, improves skill for relatively short forecast leads (1-3 weeks). Accounting for vertical land motion of each gauge primarily improves skill for longer leads (3-6 weeks), especially for the Alaskan and Gulf coasts; sea-level trends contribute to reforecast skill for both model and persistence forecasts, primarily for the East and Gulf coasts. Overall, we find that current forecast systems have sufficiently high levels of deterministic and probabilistic skill to be used in support of operational coastal flood guidance on subseasonal timescales.

In the United States, the National Oceanic and Atmospheric Administration’s National Ocean Service (NOAA/NOS) has developed a statistical model to predict the daily risks of high tide flooding (HTF), for forecast leads of up to one year, at 98 tide gauge locations along the US coastline. NOAA/NOS predicts the daily probability of exceedance of hourly water levels above a specified flood threshold by combining the tide prediction with the (extrapolated) linear trend of mean sea level and a probabilistic prediction of the anomalous hourly non-tidal residual (NTR). In turn, the NTR anomaly prediction is made up of two components: (1) a prediction of monthly mean NTR, currently based upon the observed autocorrelation function of linearly detrended NTR, with uncertainty based upon the observed standard deviation of monthly NTR; and (2) a prediction of the probability distribution function (PDF) of hourly NTR anomalies, which uses observed historical dependence upon the total water level and is assumed to be Gaussian. These forecasts are available at https://tidesandcurrents.noaa.gov/high-tide-flooding/monthly-outlook.html. In this presentation, we introduce an updated version of the NOAA/NOS HTF framework, with three key improvements: (1) the trend estimate is determined empirically and is allowed to be nonlinear; (2) monthly mean SLA is predicted by either an empirical or dynamical climate forecast model, and includes an ensemble spread; and (3) the PDF of hourly NTR anomalies is non-Gaussian and determined separately for each month from past observations using a “stochastically-generated skewed” (SGS) distribution. Skill of the updated version is compared to the original (currently operational) version at all tide gauge locations, and the impact of each of the improvements on skill is diagnosed. Further prospects for improvement of the HTF framework are also discussed.

The daily likelihood of High Tide Flooding (HTF) predicted by the National Oceanic and Atmospheric Administration (NOAA) for leads up to one year is expressed as the sum of a long-term trend, tides, and nontidal residuals (NTRs) whose probability density functions (PDFs) are assumed to be Gaussian (i.e., normally distributed). We analyzed observed detrended hourly NTR distributions at 148 NOAA tide gauges along the U.S. coastline and show that 98.7% of them are better characterized by ‘Stochastically Generated Skewed’ (SGS) distributions, a class of non-Gaussian (skewed, heavy-tailed) PDFs. In contrast to other methods that generate PDFs by fitting observed raw histograms, SGS distributions are determined through time series analysis. Observations are fit to a simple linear (autoregressive) time series model, driven by stochastic noise with a linear dependence upon the NTR anomaly. The PDF is then determined from the fitted model parameters. The SGS distributions improve upon the Gaussian PDF high-water probabilities at varying thresholds throughout the year along all U.S. coasts, with significantly better estimates along the U.S. East and Gulf coasts during summer (apart from large hurricane events) and along the U.S. West Coast during winter (even though variability there is often dominated by monthly time scales and many locations have nearly Gaussian PDFs). For evaluating extreme high-water event probabilities, the SGS distribution is no more sensitive to limited observations than kernel density estimation or Generalized Extreme Value methods. Tail probabilities for all three methods are generally similar. Our results may contribute to more robust and accurate HTF forecasts and, more broadly, provide additional insight in developing adaptation and mitigation strategies for future sea level conditions.

Sea level rise (SLR) affects coastal flood regimes and poses serious challenges to flood risk management, particularly on ungauged coasts. To address the challenge of monitoring SLR at local scales, we propose a high tide flood (HTF) thresholding system that leverages machine learning (ML) techniques to estimate SLR and HTF thresholds at a relatively fine spatial resolution (10 km) along the United States' coastlines. The proposed system, complementing conventional linear- and point-based estimations of HTF thresholds and SLR rates, can estimate these values at ungauged stretches of the coast. Trained and validated against National Oceanic and Atmospheric Administration (NOAA) gauge data, our system demonstrates promising skills with an average Kling-Gupta Efficiency (KGE) of 0.77. The results can raise community awareness about SLR impacts by documenting the chronic signal of HTF and providing useful information for adaptation planning. The findings encourage further application of ML in achieving spatially distributed thresholds. Using machine learning algorithms, this study estimates sea level rise and high tide flooding thresholds every 10 km along the United States' coasts, complementing conventional linear-/point-based estimates and offering insights for ungauged areas.

Realistic representation of monthly sea level anomalies in coastal regions has been a challenge for global ocean reanalyses. This is especially the case in coastal regions where sea levels are influenced by western boundary currents such as near the U.S. Atlantic Coast and the Gulf of Mexico. For these regions, most ocean reanalyses compare poorly to observations. Problems in reanalyses include errors in data assimilation and horizontal resolutions that are too coarse to simulate energetic currents like the Gulf Stream and Loop Current System. However, model capabilities are advancing with improved data assimilation and higher resolution. Here, we show that some current-generation ocean reanalyses produce monthly sea level anomalies with improved skill when compared to satellite altimetry observations of sea surface heights. Using tide gauge observations for coastal verification, we find the highest skill associated with the GLORYS12 and HYCOM ocean reanalyses. Both systems assimilate altimetry observations and have eddy-resolving horizontal resolutions (1/12°). We found less skill in three other ocean reanalyses (ACCESS-S2, ORAS5, and ORAP6) with coarser, though still eddy-permitting, resolutions (1/4°). The operational reanalysis from ECMWF (ORAS5) and their pilot reanalysis (ORAP6) provide an interesting comparison because the latter assimilates altimetry globally and with more weight, as well as assimilating ocean observations over continental shelves. We find these attributes associated with improved skill near many tide gauges. We also assessed an older reanalysis (CFSR), which has the lowest skill likely due to its lower resolution (1/2°) and lack of altimetry assimilation. ACCESS-S2 likewise does not assimilate altimetry, although its skill is much better than CFSR and only somewhat lower than ORAS5. Since coastal flooding is influenced by sea level anomalies, the recent development of skilful ocean reanalyses on monthly timescales may be useful for better understanding the physical processes associated with flood risks.

Accurate projections of future sea level depend on adequately representing contributions from the land and ocean. The Intergovernmental Panel On Climate Change-6th Assessment Report made significant progress with most contributors to relative sea level rise. However, coastal vertical land motion (VLM) remained unchanged from previous assessments, due in part to challenges with its spatial and temporal variability. Here, we outline a framework for estimating varying VLM in fine detail, and projecting it into the future. Along the 1700-km long California coast, we show that localized VLM, often missing in current frameworks, could double current sea-level projections by 2050 in parts of San Francisco and Los Angeles. We introduce a temporary variability metric to distinguish locations with stable from changing VLM trends, typically driven by human activities. This research underscores nonlinear VLM and its uncertainties in projections, and calls out for localized monitoring to address its impacts on relative sea-level changes.

Abstract Probabilities of coastal extreme water levels (EWLs) are increasing as sea levels rise. Using a time‐dependent statistical model on tide gauge data along U.S. and Pacific Basin coastlines, we show that EWL probability distributions also shift on an annual basis from climate forcing and long‐period tidal cycles. In some regions, combined variability (>15 cm) can be as large or larger than the amount of sea level rise (SLR) experienced over the past 30 years and projected over the next 30 years. Considering SLR and variability by 2050 at a location like La Jolla, California suggests a moderate‐level (damaging) flood today with a 50‐year return level (2% annual chance) would occur about 3–4 times a year during an El Nino nearing the peak of the nodal tide cycle. If interannual variability is overlooked, SLR related impacts could be more severe than anticipated based solely upon decadal‐scale projections.

Coastal water level information is crucial for understanding flood occurrences and changing risks. Here, we validate the preliminary version (0.9) of NOAA’s Coastal Ocean Reanalysis (CORA), which is a 43-year reanalysis (1979–2021) of hourly coastal water levels for the Gulf of Mexico and Atlantic Ocean (i.e., the Gulf and East Coast region, or GEC). CORA-GEC v0.9 was conducted by the Renaissance Computing Institute using the coupled ADCIRC+SWAN coastal circulation and wave model. The model uses an unstructured mesh of nodes with varying spatial resolution that averages 400 m near the coast and is much coarser in the open ocean. Water level variations associated with tides and meteorological forcing are explicitly modeled, while lower-frequency water level variations are included by dynamically assimilating observations from NOAA’s National Water Level Observation Network. We compare CORA to water level observations that were either assimilated or not, and find that the reanalysis generally performs better than a state-of-the-art global ocean reanalysis (GLORYS12) in capturing the variability on monthly, seasonal, and interannual timescales as well as the long-term trend. The variability of hourly non-tidal residuals is also shown to be well resolved in CORA when compared to water level observations. Lastly, we present a case study of extreme water levels and coastal inundations around Miami, Florida to demonstrate an application of CORA for studying flood risks. Our assessment suggests that NOAA’s CORA-GEC v0.9 provides valuable information on water levels and flooding occurrence from 1979–2021 in areas that are experiencing changes across multiple time scales. CORA potentially can enhance flood risk assessment along parts of the U.S. Coast that do not have historical water level observations.

Accelerated sea-level rise is an existential threat to coastal wetlands, but the timing and extent of wetland drowning are debated. Recent data syntheses have clarified future relative sea-level rise exposure and sensitivity thresholds for drowning. Here, we integrate these advances to estimate when and where rising sea levels could cross thresholds for initiating wetland drowning across the conterminous United States. Our results show that there is much spatial variation in relative sea-level rise rates, which impacts the potential timing and extent of wetlands crossing thresholds. High rates of relative sea-level rise along wetland-rich parts of the Gulf of Mexico and Atlantic coasts highlight areas where wetlands are already drowning or could begin to drown within decades, including large wetland landscapes within the Mississippi River delta, Greater Everglades, Chesapeake Bay, Texas, Georgia, and the Carolinas. Collectively, our results underscore the need to prepare for transformative coastal change. Coastal wetlands along the Gulf of Mexico and the Atlantic Coast of the United States could begin drowning within decades due to rising sea levels, according to a study of future sea-level rise scenarios.

Coastal regions face increasing threats from rising sea levels and extreme weather events, highlighting the urgent need for accurate assessments of coastal flood risk. This study presents a novel approach to estimating global extreme sea level (ESL) exceedance probabilities using a regional frequency analysis (RFA) approach. The research combines observed and modelled hindcast data to produce a high-resolution (∼1 km) dataset of ESL exceedance probabilities, including wave setup, along the entire global coastline (excluding Antarctica). The methodology presented in this paper is an extension of the regional framework of Sweet et al. (2022), with innovations introduced to incorporate wave setup and apply the method globally. Water level records from tide gauges and a global reanalysis of tide and surge levels are integrated with a global ocean wave reanalysis. Subsequently, these data are regionalised, normalised, and aggregated and then fit with a generalised Pareto distribution. The regional distributions are downscaled to the local scale using the tidal range at every location along the global coastline obtained from a global tide model. The results show 8 cm of positive bias at the 1-in-10-year return level when compared to individual tide gauges. The RFA approach offers several advantages over traditional methods, particularly in regions with limited observational data. It overcomes the challenge of short and incomplete observational records by substituting long historical records with a collection of shorter but spatially distributed records. These spatially distributed data not only retain the volume of information but also address the issue of sparse tide gauge coverage in less populated areas and developing nations. The RFA process is illustrated using Cyclone Yasi (2011) as a case study, demonstrating how the approach can improve the characterisation of ESLs in regions prone to tropical cyclone activity. In conclusion, this study provides a valuable resource for quantifying the global coastal flood risk, offering an innovative global methodology that can contribute to preparing for – and mitigating against – coastal flooding.

Sea level rise is increasing the frequency of high tide flooding in coastal communities across the United States. Although the occurrence and severity of high-tide flooding will continue to increase, skillful prediction of high tide flooding on monthly-to-annual time horizons is lacking in most regions. Here, we present an approach to predict the daily likelihood of high tide flooding at coastal locations throughout the U.S. using a novel probabilistic modeling approach that relies on relative sea-level rise, tide predictions, and climatological non-tidal residuals as measured by NOAA tide gauges. A retrospective skill assessment using the climatological sea level information indicates that this approach is skillful at 61 out of 92 NOAA tide gauges where at least 10 high tide flood days occurred from 1997–2019. In this case, a flood day occurs when the observed water level exceeds the gauge-specific high tide flood threshold. For these 61 gauges, on average 35% of all floods are accurately predicted using this model, with over half of the floods accurately predicted at 18 gauges. The corresponding False-Alarm-Rate is less than 10% for all 61 gauges. Including mean sea level anomaly persistence at leads of 1 and 3 months further improves model skill in many locations, especially the U.S. Pacific Islands and West Coast. Model skill is shown to increase substantially with increasing sea level at nearly all locations as high tides more frequently exceed the high tide flooding threshold. Assuming an intermediate amount of relative sea level rise, the model will likely be skillful at 93 out of the 94 gauges projected to have regular flooding by 2040. These results demonstrate that this approach is viable to be incorporated into NOAA decision-support products to provide guidance on likely high tide flooding days. Further, the structure of the model will enable future incorporation of mean sea level anomaly predictions from numerical, statistical, andmachine learning forecast systems.

Abstract Local sea-level changes deviate from the global mean and unforced variability often masks sea-level changes driven by greenhouse forcing. Both cause difficulties when local observations are compared to projections. We present two analyses of local sea level aiming at improving understanding local causes of sea-level rise and variability. First, we analyse local sea-level budgets at 557 tide-gauge locations from 1993-2019. On average, the sum of contributing processes explains 49% of the observed inter-annual variance. Sterodynamic processes explain most of the variability. The average observed trend is 2.6 mm yr-1 with contributors summing up to 2.7 mm yr-1. Secondly, we extrapolate the current trajectory of sea-level rise and estimate how unforced variability can mask or exaggerate future long-term sea-level changes. Unforced variability can cause sea-level changes up to multiple decimeters on 30-year time scales, and the differences between projections and the trajectory are thus generally not significant.