This interactive tool was created by adopting a next-generation modeling approach. The Details tab provides descriptions of the important aspects used to create this forecast.

This Wave-Driven Flooding Viewer for East Babeldaob, Palau, was developed at the request of the Palau National Weather Service (NWS). The Viewer was created by the Wave-driven Coastal Processes Research (WaveCPR) group, a part of the Pacific Islands Ocean Observing System (PacIOOS) at the University of Hawaiʻi at Mānoa (UH) School of Ocean and Earth Science and Technology (SOEST). Primary funding was provided through a grant from the United Nations Environment Programme (UNEP) – Green Climate Fund (GCF) partnership to PacIOOS (Melissa M. Iwamoto, Director, and Principal Investigator; and, Douglas Luther, Co-Investigator for Wave-driven Inundation Forecast Tools). Additional funding was provided by PacIOOS and the Cooperative Institute for Marine and Atmospheric Research (CIMAR) at UH.

Contact

Please contact us at info@pacioos.org if you have any questions.

1. Overview
2. Scenarios
3. Data
    3.1. Bathymetry and Topography Data
    3.2. Sea Level Data
    3.3. Wave Data
    3.4. Mangrove Data
4. Model Setup
    4.1. Wave Forcing
    4.2. Water Level
    4.3. Mangrove Representation
5. Flood Depth Outputs
6. Assumptions and Limitations
References

1. Overview

The East Babeldaob, Palau Wave-Driven Flooding with Sea Level Rise tool is an interactive coastal hazard assessment platform designed to evaluate projected flooding along the eastern coastline of Babeldaob under rising sea levels and energetic wave conditions.

The flooding maps cover the eastern coastlines of the states of Ngarchelong, Ngaraard, Ngiwal, Melekeok, and part of Ngchesar. These areas are particularly vulnerable due to their low-lying coastal terrain, direct exposure to Pacific swell, and variable reef protection along the eastern shoreline.

Unlike static “bathtub” sea level rise maps that only account for gradual increases in still water levels, this tool integrates both static sea level rise and dynamic wave-driven flooding. The projections account for:

• Mean Higher High Water (MHHW) baseline conditions
• Sea level rise increments (0–10 ft, applied in 1-foot increments)
• Wave setup, run-up, and overtopping during energetic swell events

Flood depths (water height above the land surface) are calculated relative to Mean Higher High Water (MHHW = 0.65 m above mean sea level), derived from tide gauge observations at Malakal. Sea level rise increments are applied relative to this reference level to evaluate future flood exposure.

The modeling framework simulates how waves interact with reefs, shoreline topography, and mangrove forests to influence total water levels and inland inundation. By incorporating both sea level rise and dynamic wave processes, the tool provides a more comprehensive representation of future coastal flood exposure.

2. Scenarios

The tool evaluates coastal flooding under different combinations of sea level rise (SLR), wave conditions, and mangrove presence.

Three wave scenarios are considered:

1. No waves (static flooding): Flooding driven by still water levels only, defined as MHHW plus SLR. This represents the baseline “bathtub” condition without wave effects.
2. Annual wave conditions: Flooding driven by the combined effects of still water levels (MHHW + SLR) and representative energetic wave conditions that occur approximately once per year. This scenario reflects typical high-wave events that contribute to recurring coastal flooding.
3. Extreme wave conditions: Flooding driven by the combined effects of still water levels (MHHW + SLR) and extreme wave forcing based on Typhoon Bopha (2012). This scenario represents a high-impact yet plausible event, with the potential to produce severe coastal flooding.

All scenarios are evaluated across SLR increments from 0 to 10 ft, in 1-ft steps. These increments represent progressively higher future sea levels, allowing users to see how flooding may expand and deepen as sea level rises. Simulations are also performed with and without mangrove forests to assess their role in reducing dynamic wave-driven flooding.

Figure 1. (Click on image for larger view.) Conceptual illustration of wave-driven flooding scenarios under a 3-ft sea level rise (SLR). Flooding is shown for (top) still water levels only (MHHW + 3-ft SLR), (middle) annual wave conditions, and (bottom) extreme wave conditions. The horizontal extent of flooding increases with wave forcing, and flood depth represents the depth of water above the ground surface, with elevations referenced to MHHW.

3. Data

3.1. Bathymetry and Topography Data

The bathytopographic dataset is derived from a 1-m resolution digital elevation model (DEM) generated from PALARIS lidar data and is used to represent both land topography and nearshore bathymetry. The dataset extends offshore to approximately −50 m depth, capturing the reef and nearshore environment that strongly influences wave transformation. For implementation in the numerical model, the lidar-derived DEM was processed and resampled onto a regular 5-m × 5-m computational grid to ensure numerical efficiency while maintaining adequate resolution.

The original dataset is referenced to an orthometric vertical datum (EGM2008 geoid). For consistency with local sea level conditions, elevations were adjusted to the local mean sea level (MSL) at Malakal using a correction derived from the University of Hawaiʻi Sea Level Center (UHSLC), based on UHSLC benchmarks.

3.2. Sea Level Data

Sea level data from the Malakal tide gauge, provided by the UHSLC, were used to estimate Mean Higher High Water (MHHW) at the site. As long-term tidal observations are not available along the eastern coast of Babeldaob, the Malakal station in Koror is used as the regional reference for defining tidal datums.

To compute MHHW, the time series was first detrended to remove long-term sea level rise. For each tidal day (~25 hours), the higher of the two daily high waters was extracted. MHHW was then calculated as the average over a full 18.6-year lunar nodal cycle, with the averaging window centered around mid-2001.

The resulting MHHW value is 0.65 m above mean sea level and is used as the reference water level in the modeling framework, with SLR increments applied relative to this baseline.

Figure 2. (Click on image for larger view.) Estimation of Mean Higher High Water (MHHW) from the Malakal tide gauge. The detrended water level time series (blue) and extracted daily higher high waters (green) over an 18.6-year nodal cycle are shown. MHHW (0.65 m) is computed as the average of daily higher high waters and is indicated by the orange line.

3.3. Wave Data

Offshore wave conditions were derived from ERA5 reanalysis data (ECMWF Reanalysis v5; Hersbach et al., 2020), which provides continuous hourly wave records at a spatial resolution of approximately 0.25° (~28 km). A representative grid point located offshore of eastern Babeldaob (7.5°N, 134.75°E; water depth ~3,700 m) was selected to characterize incoming wave conditions.

High-resolution regional wave hindcasts and long-term in situ observations are limited for Palau. Available buoy records begin in 2015 and do not provide a sufficiently long record for statistical analysis. As a result, ERA5 data were used to define representative wave conditions for the simulations. From this dataset, significant wave height, peak wave period, and wave direction were extracted to characterize both annual and extreme wave scenarios.

To define representative annual wave conditions, a Generalized Extreme Value (GEV) analysis was applied to the ERA5 hindcast data. The analysis was restricted to wave events arriving from the northeast quadrant (45°–75°), which corresponds to the dominant wave direction and the primary exposure window of the eastern Babeldaob coastline. Annual maxima were used to estimate characteristic significant wave height (H_s) and peak period (T_p). The resulting annual wave conditions are defined by H_s = 2.52 m and T_p = 17 s, representing energetic but regularly occurring wave forcing.

To define extreme wave conditions, the most energetic historical event in the ERA5 hindcast record was identified, based on the combined occurrence of high significant wave height, long peak period, and an incident wave direction aligned with the exposed eastern coastline. The selected event occurred on December 2, 2012, during Typhoon Bopha and was characterized by a strong eastward wave direction (~90°), significant wave heights of approximately 9 m, and a combined sea state consisting of shorter-period wind waves (~10 s) and a dominant long-period swell component (T_p ≈ 23 s). The severity of this event is supported by observations from the Palau Typhoon History report (CRRF, 2014), which document significant impacts along Palau’s eastern coast, including high wave conditions, storm surge, and reef scouring. This event represents a rare but physically plausible upper-bound wave forcing and is used to define the extreme wave scenario in the modeling framework.

3.4. Mangrove Data

Mangrove distribution data for Palau were obtained from the Global Mangrove Watch dataset. Multiple temporal snapshots are available (1996–2020); the 2020 dataset was used in this study as the most recent representation of mangrove extent. This dataset includes 215 polygons delineating mangrove areas, covering most coastal regions of Babeldaob as well as smaller islands to the south.

4. Model Setup

Numerical simulations were conducted using a Boussinesq-type phase-resolving model (Mihami, 2023; Mihami and Roeber, 2026; Roeber et al., 2010) to resolve nonlinear wave propagation, transformation across the reef, and wave-driven inundation along the eastern coast of Babeldaob. The computational domain spans approximately 8.5 km in the cross-shore direction and 36 km alongshore, discretized on a 5 m × 5 m structured grid.

Figure 3. (Click on image for larger view.) Model domain and simulated free surface. The red outline indicates the extent of the phase-resolving computational domain (5 m × 5 m grid; ~8.5 km cross-shore × 36 km alongshore). The color shading shows the free-surface elevation at t = 2 h 20 min, illustrating a saturated wave field and associated inundation extent for the +2-ft SLR scenario under annual recurring wave conditions. Insets provide zoomed views of inundation patterns in Ngaraard (top) and Ngiwal (bottom).

4.1. Wave Forcing

Waves were generated within the computational domain using an internal spectral wavemaker located along the offshore (eastern) boundary. Additional lateral wavemakers were applied along the northern and southern boundaries to allow oblique wave propagation at the domain edges. Wave forcing was defined using a TMA (Texel–Marsen–Arsloe) spectrum, with wave parameters (significant wave height, peak period, and wave direction) set for each wave scenario as described above. Directional spreading was applied using a Gaussian distribution centered on the peak direction, with a 30° angular spread.

4.2. Water Level

The baseline water level was set to Mean Higher High Water (MHHW = 0.65 m above mean sea level), with sea level rise increments applied as uniform offsets. This represents the background water level over which waves propagate under different sea level rise scenarios.

4.3. Mangrove Representation

Mangrove forests play an important role in reducing wave-driven flooding by dissipating wave energy through drag generated by their complex root and canopy structures. In phase-resolving wave models, this effect is commonly represented as an additional dissipation term, which can be approximated using enhanced bottom friction due to its similar quadratic dependence on flow velocity (e.g., Mendez and Losada, 2004; Wamsley et al., 2010).

Mangrove areas were defined using the 2020 Global Mangrove Watch dataset. A higher Manning’s roughness coefficient (n = 0.14 m¹ᐟ³ s⁻¹) was assigned within these regions to represent increased drag due to vegetation, while non-mangrove areas were assigned a baseline value of n = 0.02 m¹ᐟ³ s⁻¹. The selected mangrove roughness is consistent with values reported in previous studies of wave and surge attenuation in dense mangrove systems (e.g., Zhang et al., 2012; Xu et al., 2018; Chen et al., 2020).

This approach provides a bulk representation of mangrove-induced dissipation suitable for large-scale simulations. Future field measurements along the eastern coast of Palau are planned to further refine these parameter values.

5. Flood Depth Outputs

The flooding maps show the near-maximum water depth above the land surface reached during each simulation. For each scenario (wave condition, mangrove configuration, and sea level rise increment), the model was run for a total of 2 hours and 20 minutes, including an initial 20-minute spin-up period to establish a fully developed wave field. Model outputs were recorded every 3 seconds to resolve wave-driven variability in water levels. A minimum wetting depth of 0.1 m was applied to exclude very shallow water layers.

Flood depths were estimated using a 2% exceedance runup approach. At each grid cell, the 98^th percentile of the free surface elevation time series was used to represent extreme water levels associated with wave runup. Flood depth was then computed relative to the local ground elevation, with all values referenced to Mean Higher High Water (MHHW). This approach captures the upper range of wave-driven flooding while filtering out short-duration fluctuations.

6. Assumptions and Limitations

The results presented here provide scenario-based projections of coastal flooding under defined input data and modeling assumptions. The main limitations of this study are outlined below:

• Bathymetry and topography are assumed to represent present-day conditions and remain unchanged. Future changes such as reef degradation, sediment transport, or shoreline evolution are not included.
• Wave conditions are based on present-day climate and do not account for potential future changes in wave climate, including shifts in storm patterns or swell characteristics.
• Flooding is driven only by waves and still water levels. Other processes such as rainfall, surface runoff, and groundwater inundation are not included.
• Mean Higher High Water (MHHW) is derived from the Malakal tide gauge on the western side of Palau due to the lack of long-term observations along the eastern coast, which may introduce uncertainty in local water level representation.
• Mangrove effects are represented using a uniform roughness coefficient and do not account for variations in vegetation structure, density, or species.
• The DEM represents a bare-earth surface and does not include vegetation or buildings. As a result, small-scale features that may influence local flooding are not resolved.

References

• Chen, Q., L. Zhu, F. Shi, and S. Brandt (2020). Boussinesq modeling of combined storm surge and waves over wetlands forced by wind. Coastal Engineering Proceedings, 36v, waves.6. https://doi.org/10.9753/icce.v36v.waves.6.
• Coral Reef Research Foundation (CRRF) (2014). A summary of Palau’s typhoon history (1945–2013). Technical Report, 17pp. https://coralreefpalau.org/wp-content/uploads/2017/05/CRRF-Palau-Typhoon-History-2014-1.pdf.
• Hersbach, H., B. Bell, P. Berrisford, S. Hirahara, A. Horányi, J. Muñoz-Sabater, J. Nicolas, C. Peubey, R. Radu, D. Schepers, A. Simmons, C. Soci, S. Abdalla, X. Abellan, G. Balsamo, P. Bechtold, G. Biavati, J. Bidlot, M. Bonavita, G. De Chiara, P. Dahlgren, D. Dee, M. Diamantakis, R. Dragani, J. Flemming, R. Forbes, M. Fuentes, A. Geer, L. Haimberger, S. Healy, R.J. Hogan, E. Hólm, M. Janisková, S. Keeley, P. Laloyaux, P. Lopez, C. Lupu, G. Radnoti, P. de Rosnay, I. Rozum, F. Vamborg, S. Villaume, and J.-N. Thépaut (2020). The ERA5 global reanalysis. Quarterly Journal of the Royal Meteorological Society, 146, 1999–2049. https://doi.org/10.1002/qj.3803.
• Mendez, F.J. and I.J. Losada (2004). An empirical model to estimate the propagation of random breaking and nonbreaking waves over vegetation fields. Coastal Engineering, 51, 103–118. https://doi.org/10.1016/j.coastaleng.2003.11.003.
• Mihami, F.-Z. (2023). Développement d’un modèle de vagues littorales dispersives pour l’évaluation opérationnelle des risques côtiers [Development of a phase-resolving computer model for operational nearshore wave assessment]. Doctoral Thesis. Université de Pau et des Pays de l’Adour, Pau, France. 219 pp. https://doi.org/10.70675/bc75acf0z8bdez496ez9091z2ca6d1bb864e.
• Mihami, F.-Z. and V. Roeber (2026). Exploring a conservative staggered scheme for Boussinesq-type equations: Insights into numerical diffusion, dispersion, and wave-breaking. Coastal Engineering, 204, 104880. https://doi.org/10.1016/j.coastaleng.2025.104880.
• Roeber, V., K.F. Cheung, and M.H. Kobayashi (2010). Shock-capturing Boussinesq-type model for nearshore wave processes. Coastal Engineering, 57, 407–423. https://doi.org/10.1016/j.coastaleng.2009.11.007.
• Wamsley, T.V., M.A. Cialone, J.M. Smith, J.H. Atkinson, and J.D. Rosati (2010). The potential of wetlands in reducing storm surge. Ocean Engineering, 37, 59–68. https://doi.org/10.1016/j.oceaneng.2009.07.018.
• Xu, H., X. Liu, F. Li, S. Huang, and C. Liu (2018). A novel multislope MUSCL scheme for solving 2D shallow water equations on unstructured grids. Water, 10, 524. https://doi.org/10.3390/w10040524.
• Zhang, K., H. Liu, Y. Li, H. Xu, J. Shen, J. Rhome, and T.J. Smith (2012). The role of mangroves in attenuating storm surges. Estuarine, Coastal and Shelf Science, 102–103, 11–23. https://doi.org/10.1016/j.ecss.2012.02.021.

Back to Top

The development of this flooding viewer would not have been possible without the generous support and collaboration of our Republic of Palau partners. Special thanks to the Palau Automated Land and Resources Information System (PALARIS) for providing high-resolution topobathymetric data, and to David Idip (PALARIS) for coordination and assistance. We also thank the Palau Weather Service for their support, and the Coral Reef Research Foundation (CRRF) for advice and assistance in identifying at-risk coastal areas and deploying sensors for validation data collection. We are especially grateful to Xavier Erbai Matsutaro, National Climate Change Coordinator at the Palau Office of Climate Change, for his strong support and valuable guidance. This work would not have been possible without the many wonderful local contacts who contributed to this effort.