Coastal Flood Vulnerability Research Articles

Assessment of flood mitigation strategy based on integrated approach of remote sensing and coastal vulnerability geospatial modeling at the coastal plain of Suriname

Suriname is the smallest South American nation with a low-lying coastal plain that is vulnerable to inundation from the Atlantic Ocean and inland rivers, as well as pluvial flooding primarily due to rainfall. Paramaribo, the capital of Suriname, has the highest population density, and its demographics extend into the surrounding districts of Wanica and Commewijne. Suriname has experienced flood disasters almost annually, which has exacerbated in recent years, posing a significant socioeconomic challenge. The country must balance the need for flood disaster adaptation and climate resilience with the potential impact on its resources and well-being of settlement areas. Policymakers and other stakeholders are working to address environmental impacts on the coast, but there is still a need for a comprehensive approach to monitor and manage flood impacts. This research has three objectives. The first is to analyze flood frequency events from 2021 to 2023 using multi-temporal satellite image processing from Sentinel-1 SAR (Synthetic Aperture Radar). The second is to generate a Coastal Flood Vulnerability Index (CFVI) for floods using a geospatial multi-criteria analysis approach based on exposure, sensitivity, and adaptive capacity components. The third objective is to assess the mitigation strategy for floods in settlement areas based on an integrated analysis of CFVI and stakeholder perception. The research methodology uses a multi-criteria analysis regarding settlement areas and ranking each component by expert opinion in an equation derived from Intergovernmental Panel on Climate Change (IPCC) assessment report at district level. The CFVI indices rely on secondary data acquisition from national and global datasets or referenced works. Interviews were conducted to better understand the stakeholder’s perspectives that are at a strategic or governing level, and to evaluate the existence of flood early-warning and other adaptation capabilities. A flood mitigation strategy is then suggested for the most vulnerable district by CFVI score.

Analisis Faktor-Faktor Kerentanan Banjir Pesisir (Studi pada Pesisir Pantai Utara Jawa Tengah)

Flooding is a natural phenomenon that occurs in Indonesia, including the coastal areas of Pantura) which are complicated problems for the government that have not been fully handled properly. Pantura coastal flooding occurs due to rainwater flooding and tidal flooding, causing vulnerability. This study aims to determine the factors that cause vulnerability and determine the adaptation strategy for Pantura coastal flooding. This type of research is quantitative research because research is in the form of case studies based on descriptive assessments and the existence of quantitative data based on numbers and research is carried out by observation and data collection using data collection techniques and data analysis with SPSS Version 25.00 with 183 respondents. The research location is the coastal area of Pantura. Primary data collection method with questionnaires, interviews, observations, and documentation,. Based on the results of the study, the factors of coastal flood vulnerability are classified into physical vulnerability of buildings and infrastructure, biophysical and hydrological vulnerability, vulnerability of community capability and safety and flood vulnerability. The validity test of the vulnerability of coastal flooding shows all valid and reliable indicators because it has a validity value of > 0.300 and the reliability test value of the cronbrach alpha research instrument is above 0.700.

Evaluating geomorphological changes and coastal flood vulnerability of the Nijhum Dwip Island using remote sensing techniques

This research critically assesses the spatiotemporal changes in morphology of the Nijhum Dwip Island caused by the dynamic wave action of the Bay of Bengal. Using image analysis techniques on Landsat Satellite Data (ETM+, OLI), the nature of shoreline alterations, erosion-accretion, and lob shifting from 2003 to 2021 has been identified. The findings reveal that since 2013, the island's morphology and geographical extent have undergone considerable changes, and it has gradually reached its current stable state in the Bay of Bengal. The fluvial sediments brought by the Ganges-Brahmaputra-Meghna (GBM) river system influenced by tide is mainly responsible for the sedimentary accretion in and around the island. Due to the complex wave action, the island has experienced alternating erosion and accretion domination scenarios. Compared with its extent in 2003, up until 2021, it has lost approximately 2.57 km2 area, erosion and accretion mostly occurring in the southern part, geometrically transforming the island from a rectangular to a subrounded shape. The depocenter shifting of the island was predominantly in the south-eastern direction during the whole period but has shifted to reverse direction in 2021. The minimum and maximum lob shifting occurred in 2006 and 2015, respectively, at approximately 600 m and 1800 m. ALOS PALSAR DEM (12.5m) in combination with secondary data from the Bangladesh Meteorological Department has been used to develop and simulate flood inundation scenarios under different surge heights. The projections illustrate that in the case of 2 m, 3 m, and 4 m surge heights, approximately 46.45%, 56.51%, and 61.42% of the total land will go underwater, respectively. In all scenarios, cropland and bare ground are the most affected classes.

Climate Scenarios for Coastal Flood Vulnerability Assessments: A Case Study for the Ligurian Coastal Region

Extreme sea levels and coastal flooding are projected to be among the most uncertain and severe consequences of climate change. In response, a wide development of coastal vulnerability assessment methodologies has been observed in research to support societal resilience to future coastal flood risks. This work aims to explore the scope of application of index-based methodologies for coastal vulnerability assessment, in terms of their suitability to convey information on variations in climate variables potentially leading to sea-level changes and inundation. For this purpose, the InVEST Coastal Vulnerability model was coupled for the first time with the ERA5 reanalysis and used to develop a case study assessment of the biophysical exposure component of vulnerability to coastal flooding for Liguria, an Italian coastal region facing the Mediterranean Sea. Different scenarios of wind speed and wave power were created in order to test the sensitivity of this approach to climate data inputs. The results support the applicability of this approach to provide a preliminary grasp of local vulnerability to coastal inundation. Yet, this work also highlights how the method’s data aggregation and indicator computation processes result in its insensitivity to wind and wave variations, and therefore in its unsuitability to reproduce climate scenarios. The implications of these findings for research methodology and regarding the operationalisation of vulnerability assessment results are discussed.

Coastal flood vulnerability assessment, a satellite remote sensing and modeling approach

Although there are numerous case studies assessing coastal vulnerability, many of these studies have been performed in places where notable efforts have been carried out to provide information on the different variables that affect the coast. However, this is not the case for most places worldwide given the lack of long-term datasets. This study makes use of information from satellite remote sensing and analytical models to derive two vulnerability indices along a 9.5 km stretch of the coast of Langue de Barbarie, Saint Louis, Senegal (Western Africa). The first is a coastal vulnerability index (CVI) to sea level rise due to climate change and results in a five-category classification: Very Low, Low, Moderate, High, and Very High. The second is a flood vulnerability index (FVI) to coastal flooding due to extreme events and results in a three-category classification: Low, Moderate, and High. Results for the CVI index show that 70% of the coast presents High and Very High vulnerability values, largely located in the most densely populated areas. The FVI is assessed for one of the most energetic storms for the 1979–2021 period which occurred in February 2018 using a beach configuration of March 2021. Results show that 29% of the coastline presents High FVI values (i.e., are likely to be overtopped) concentrated in the central sector of the most-populated districts. This provides relevant tools to improve coastal management when in situ data are not available.

Integrated coastal subsidence analysis using InSAR, LiDAR, and land cover data

Land subsidence is an important cause of relative sea-level rise along the Gulf Coast. There is a lack of effective monitoring of coastal subsidence with high accuracy and high spatial resolution for improving coastal risk assessment and mitigation. This study is the first attempt to integrate satellite interferometric synthetic aperture radar (InSAR) and airborne light detection and ranging (LiDAR) methods to investigate the spatiotemporal pattern of coastal subsidence. The study area is around Eagle Point, Texas, a region known for its fast rate of relative sea-level rise in recent decades. From 2006 to 2011, the line-of-sight velocities were up to −33 mm/year based on ascending ALOS-1 PALSAR-1 images. From 2016 to 2021, the vertical velocities were up to −34 mm/year based on ascending and descending Sentinel-1 images. Additional details of the subsidence pattern were revealed by incorporating the surface difference derived from 1-m airborne LiDAR results. Comparisons of the InSAR-derived velocities from image time series and the LiDAR-derived surface changes from time-lapsed observations were conducted at different spatial levels with linkages to land cover patterns and topography. The results showed that local subsidence rates could vary significantly below the spatial resolution of InSAR results, indicating a valuable role of airborne LiDAR results in extending InSAR results to parcel and building levels and explaining subpixel uncertainties. Also, subsidence appeared to be stronger in vegetated areas than in developed areas and negatively correlated with surface imperviousness. The magnitude of subsidence was not correlated with elevation along selected transect lines. Overall, this study demonstrated the benefits of combining InSAR results with other geospatial datasets to characterize coastal subsidence. In particular, the high vertical accuracy of InSAR results and the high spatial resolution of airborne LiDAR results could be complementary, highlighting the necessity of multi-resolution data fusion to support studies on coastal flood vulnerability, infrastructure reliability, and erosion control.

Dynamic modeling of sea-level rise impact on coastal flood hazard and vulnerability in New York City's built environment

There is a consensus that future sea-level rise (SLR) will increase the exposure of population and assets to coastal flooding. However, the extent to which SLR affects flood hazards and human vulnerability to flooding in the built environment is not well understood. This study investigates the effects of future SLR on coastal flood hazards and human vulnerability to flooding in New York City's built environment. With a focus on a hurricane-induced flood event, we utilize a building-scale hydrodynamic model to simulate flood hazards under different 21st-century SLR scenarios. We further implement a human vulnerability model to reveal how the physical vulnerability of individuals to flooding would respond to the effects of SLR on flood hazards. We find that SLR would result in a substantial increase in not only the floodwater depth but also the floodwater velocity in the study area. For example, under a 1.04 m SLR scenario, the increase in the max floodwater speed exceeds 2.7 m/s (1271%) in 5% of the area that was flooded under the no-SLR scenario (control run). Model results show that, due to nonlinear interactions, the floodwater depth simulated by the hydrodynamic model for a SLR scenario could substantially differ from the depth estimated based on a linear addition of the SLR to the control-run floodwater depth. We find that the effects of SLR on flood hazards would, in turn, substantially affect the extent, intensity, and duration of human physical vulnerability to flooding, which could potentially increase the number of injuries and mortalities.

Quantifying coastal flood vulnerability for climate adaptation policy using principal component analysis

With increasing population growth and urban sprawl, many coastal lowlands are unprecedentedly vulnerable to climate change and its impacts, such as rising sea levels, increasing extreme storm events, and coastal flooding. Quantifying coastal flood vulnerability serves as a tool to identify a system’s weakness, monitor its change, and support making targeted climate adaptation policies. The assessment framework proposed in this research uses principal component analysis (PCA) and a weighting method to build a composite indicator of flood vulnerability index and evaluate the vulnerability for 256 coastal census tracts and 24 municipalities along the coast of Connecticut, USA. The research uses Keiser-Meyer-Olkin (KMO) test and Bartlett’s test of sphericity to test sample adequacy and performs data standardization for all indicators. Through PCA, 30 coastal vulnerability-related indicators were grouped into four major dimensions: hazard exposure, socio-economic, physical/land use and land cover, and natural. The findings highlight the variations of flood vulnerability across highly urbanized areas, suburban areas, and rural areas; and the gradient from coastal low-elevation region to high-elevation inland area. This variance is unevenly caused by different dimensions although they may trade-off with each other when aggregated, the dominant dimensions play a significant or decisive role in the vulnerability assessment. This research built an automatic and objective assessment framework that is flexible enough to be applied at a smaller scale so as to obtain detailed analysis and it can be used as a decision-making support system.

Assessing Multi-Hazard Vulnerability and Dynamic Coastal Flood Risk in the Mississippi Delta: The Global Delta Risk Index as a Social-Ecological Systems Approach

The tight coupling of the social-ecological system (SES) of the Mississippi Delta calls for balanced natural hazard vulnerability and risk assessments. Most existing assessments have approached these components in isolation. To address this, we apply the Global Delta Risk Index (GDRI) in the Mississippi Delta at high-resolution census tract level. We assess SES spatial patterns of drought, hurricane-force wind, and coastal flood vulnerability and integrate hazard and exposure data for the assessment of coastal flood risk. Moreover, we compare current coastal flood risk to future risk in 2025 based on the modelled effects of flood depth, exposure, and changes in ecosystem area in the context of ongoing efforts under the 2017 Louisiana Coastal Master Plan. Results show that the Master Plan will lead to decreases in risk scores by 2025, but the tracts that are currently the most vulnerable benefit less from risk reduction efforts. Along with our index output, we discuss the need for further advancements in SES methodology and the potential for catastrophic hazard events beyond the model parameters, such as extreme rainfall events and very strong hurricanes. Assessing SES risk components can lead to more targeted policy recommendations, demonstrated by the need for Master Plan projects to consider their unequal spatial effects on vulnerability and risk reduction.

Vulnerability of Florida residential structures to hurricane induced coastal flood

This paper presents a methodology to develop hurricane induced coastal flood vulnerability functions for residential construction based on empirical tsunami fragility functions. A force equivalency mapping procedure first transforms the tsunami fragility functions into coastal flood fragility functions. Following the quantification of the damage states and the incorporation of repair costs, the coastal flood fragility functions translate into coastal flood vulnerability functions. Insurance claims data from the National Flood Insurance Program and vulnerability functions independently derived by the US Army Corps of Engineers are employed to validate single-story on-grade timber and reinforced masonry structure model outputs. The limitations of the model and future developments are discussed.

A participatory community case study of periurban coastal flood vulnerability in southern Ecuador.

BackgroundPopulations in coastal cities are exposed to increasing risk of flooding, resulting in rising damages to health and assets. Adaptation measures, such as early warning systems for floods (EWSFs), have the potential to reduce the risk and impact of flood events when tailored to reflect the local social-ecological context and needs. Community perceptions and experiences play a critical role in risk management, since perceptions influence people’s behaviors in response to EWSFs and other interventions.MethodsWe investigated community perceptions and responses in flood-prone periurban areas in the coastal city of Machala, Ecuador. Focus groups (n = 11) were held with community members (n = 65 people) to assess perceptions of flood exposure, sensitivity, adaptive capacity, and current alert systems. Discussions were audio recorded, transcribed, and coded by topic. Participatory maps were field validated, georeferenced, and digitized using GIS software. Qualitative data were triangulated with historical government information on rainfall, flood events, population demographics, and disease outbreaks.ResultsFlooding was associated with seasonal rainfall, El Niño events, high ocean tides, blocked drainage areas, overflowing canals, collapsed sewer systems, and low local elevation. Participatory maps revealed spatial heterogeneity in perceived flood risk across the community. Ten areas of special concern were mapped, including places with strong currents during floods, low elevation areas with schools and homes, and other places that accumulate stagnant water. Sensitive populations included children, the elderly, physically handicapped people, low-income families, and recent migrants. Flood impacts included damages to property and infrastructure, power outages, and the economic cost of rebuilding/repairs. Health impacts included outbreaks of infectious diseases, skin infections, snakebite, and injury/drowning. Adaptive capacity was weakest during the preparation and recovery stages of flooding. Participants perceived that their capacity to take action was limited by a lack of social organization, political engagement, and financial capital. People perceived that flood forecasts were too general, and instead relied on alerts via social media.ConclusionsThis study highlights the challenges and opportunities for climate change adaptation in coastal cities. Areas of special concern provide clear local policy targets. The participatory approach presented here (1) provides important context to shape local policy and interventions in Ecuador, complimenting data gathered through standard flood reports, (2) provides a voice for marginalized communities and a mechanism to raise local awareness, and (3) provides a research framework that can be adapted to other resource-limited coastal communities at risk of flooding.

A methodology for urban micro-scale coastal flood vulnerability and risk assessment and mapping

One of the most dangerous challenges to settlements in the UK comes from flooding. Currently, there is extensive map coverage of flood hazards zones in the UK; however, it is increasingly recognised that risk associated with natural hazards cannot be reduced solely by focussing on the hazard. There is also an urgent need for methods of evaluating and mapping flood vulnerability and risk in detail. Despite its significance, conventional flood risk assessment methodologies often underestimate likely levels of vulnerability in areas prone to hazards, yet it is the degree of vulnerability within a community that determines the consequences of any given hazard. The research presented proposes a general methodology to assess and map Coastal Flood Vulnerability and Risk at a detailed, micro-scale level. This captures aspects that are considered crucial and representative of reality (socio-economic, physical and resilient features). The methodology is then applied to a UK case study (city of Portsmouth). Environment Agency flood hazard data, National Census socio-economic data and Ordnance Survey topographic map data have been used to evaluate and map coastal flood vulnerability, examining neighbourhoods within census wards. This led to a subsequent analysis of Coastal Flood Risk, via the combination of a Coastal Flood Vulnerability Index and a Coastal Flood Hazard Index, for the Portsmouth ward Hilsea. This, consequently, identifies potential weaknesses that could lead to more effective targeting of interventions to improve resilience and reduce vulnerability in the long term and provides a basis for hazard and environmental managers/planners to generate comprehensive national/international vulnerability and risk assessments.

Modeling residential coastal flood vulnerability using finished-floor elevations and socio-economic characteristics

Densely populated coastal regions are vulnerable to threats associated with climate change and variability, especially storms. In the United States, millions of people are repeatedly at risk of flooding and because this number will only continue to grow, the identification of the intersection of social vulnerability and physical risk to flood inundation is essential for both coastal planning and adaptation purposes. Although a key tool to identify vulnerable populations, most vulnerability models are built at the county or coarser scales, thereby hindering the effectiveness of mitigation and adaptation planning at community scales, which are more socially and physically diverse than what county-scale analyses can reveal. We present an integrated social and physical model of vulnerability at the block-group level of geography using census data to measure social variability based population and housing data and physical exposure based on the intersection of finished floor elevation of all buildings in coastal North Carolina, USA with flood hazards maps. We identify, in a spatially-explicit manner and at multiple levels of governance, areas of high social vulnerability and their intersection with areas of high physical exposure to inundation. We found that in the 28 coastal counties of North Carolina, 45.3% of the structures within the 100-year floodplain were structurally exposed to potential damage from inundation. Supporting our hypothesized patterns of vulnerability to inundation, a significant clustering of highly vulnerable block-groups were located in Albemarle and Eastern Carolina coastal regions, yet high vulnerability outliers were also located at significant distance away from the highly physically-exposed coastline. Our findings suggest that the high-resolution block-group level analysis identified multiple levels of vulnerability to inundation at the sub-county scale and provide essential information for effective hazard mitigation within scales ranging from the community to transboundary governing bodies.

Development and Implementation of an Assessment Framework for Coastal Flood Vulnerability

Development and Implementation of an Assessment Framework for Coastal Flood Vulnerability

Effects of DEM Resolution on Modeling Coastal Flood Vulnerability

This article examines whether Digital Elevation Model (DEM) resolution affects the accuracy of predicted coastal inundation extent using LISFLOOD-FP, with application to a sandy coastline in New Jersey. DEMs with resolution ranging from 10 to 100 m were created using coastal elevation data from NOAA, using the North American Vertical Datum of 1988. A two-dimensional hydrodynamic flood model was developed in LISFLOOD-FP using each DEM, all of which were calibrated and validated against an observed 24-h tidal cycle and used to simulate a 1.5 m storm surge. While differences in predicted inundated area from all models were within 1.0%, model performance and computational time worsened and decreased with coarser DEM resolution, respectively. This implied that using a structured grid model for modeling coastal flood vulnerability is based on two trade-offs: high DEM resolution coupled with computational intensity, but higher precision in model predictions, and vice versa. Furthermore, water depth predictions from all DEMs were consistent. Using an integrated numerical modeling and GIS approach, a two-scale modeling strategy, where a coarse DEM is used to predict water levels for projection onto a fine DEM was found to be an effective, and computationally efficient approach for obtaining reliable estimates of coastal inundation extent.

Urban neighbourhood flood vulnerability and risk assessments at different diurnal levels

Diurnal changes within communities can significantly alter the level of impacts during a flood, yet these essential daily variations are not currently catered for within flood risk assessments. This paper develops a flood vulnerability and risk model that captures crucial features of flood vulnerability; integrating physical and socio‐economic vulnerability data, combined with a flood hazard analysis, to give overall flood risk at neighbourhood scale, at two different times of day, for floods of different magnitudes. The flood vulnerability and risk model, the resulting diurnal coastal flood vulnerability and risk indexes, and corresponding maps for the ward of Hilsea (Portsmouth, United Kingdom), presented within this paper, highlight three previously unidentified neighbourhoods in particular in the northwest of the Hilsea ward, which have the highest levels of risk during both time zones and for flood events of different magnitude. Critically, these neighbourhoods lie further inland and not directly on the Hilsea coastline, yet by analysing at this resolution (including diurnal impacts), substantial levels of underlying vulnerability were identified within these areas.

DEM Resolution Effects on Coastal Flood Vulnerability Assessment: Deterministic and Probabilistic Approach

AbstractFlood modeling is highly influenced by topography data set. Lower resolution digital elevation models (DEMs) are usually used because of their availability and less computational burden when they are used in modeling applications. However, low accuracy of these DEMs yields to even lower accuracy in flood risk analysis through spatial modeling. This study aims to explore the DEM resolution effects on coastal flood risk assessments. For this purpose, deterministic and probabilistic approaches are employed for flood inundation modeling utilizing hydrologically connected bathtub method. High‐resolution light detection and ranging (LiDAR) DEM is considered as the most accurate data from which different resolution maps are obtained using resampling techniques. This is incorporated into an error analysis framework along with U.S. Geological Survey (USGS) National Elevation Dataset (NED) DEMs. The probabilistic framework is developed by simulating the spatial variability of elevation errors compared to LiDAR DEM through a Monte Carlo‐based method called sequential Gaussian simulation. The proposed methodology is applied to the lower Manhattan in New York City. By integrating the flood model into the developed framework, flood inundation probability at each grid cells is obtained. Furthermore, using the concept of accuracy‐efficiency tradeoffs, a framework for selecting a suitable spatial resolution for probabilistic flood risk assessment has been suggested. The results show that by exercising a range of options presented in this paper, a broader insight into mapping resolution can be provided, improving flood assessment, evacuation zones, and mitigation plans depending upon the data availability in a region.

An uncertainty-based framework to quantifying climate change impacts on coastal flood vulnerability: case study of New York City.

The continued development efforts around the world, growing population, and the increased probability of occurrence of extreme hydrologic events have adversely affected natural and built environments. Flood damages and loss of lives from the devastating storms, such as Irene and Sandy on the East Coast of the USA, are examples of the vulnerability to flooding that even developed countries have to face. The odds of coastal flooding disasters have been increased due to accelerated sea level rise, climate change impacts, and communities' interest to live near the coastlines. Climate change, for instance, is becoming a major threat to sustainable development because of its adverse impacts on the hydrologic cycle. Effective management strategies are thus required for flood vulnerability reduction and disaster preparedness. This paper is an extension to the flood resilience studies in the New York City coastal watershed. Here, a framework is proposed to quantify coastal flood vulnerability while accounting for climate change impacts. To do so, a multi-criteria decision making (MCDM) approach that combines watershed characteristics (factors) and their weights is proposed to quantify flood vulnerability. Among the watershed characteristics, potential variation in the hydrologic factors under climate change impacts is modeled utilizing the general circulation models' (GCMs) outputs. The considered factors include rainfall, extreme water level, and sea level rise that exacerbate flood vulnerability through increasing exposure and susceptibility to flooding. Uncertainty in the weights as well as values of factors is incorporated in the analysis using the Monte Carlo (MC) sampling method by selecting the best-fitted distributions to the parameters with random nature. A number of low impact development (LID) measures are then proposed to improve watershed adaptive capacity to deal with coastal flooding. Potential range of current and future vulnerability to flooding is estimated with and without consideration of climate change impacts and after implementation of LIDs. Results show that climate change has the potential to increase rainfall intensity, flood volume, floodplain extent, and flood depth in the watershed. The results also reveal that improving system resilience by reinforcing the adaptation capacity through implementing LIDs could mitigate flood vulnerability. Moreover, the results indicate the significant effect of uncertainties, arising from the factors' weights as well as climate change, impacts modeling approach, on quantifying flood vulnerability. This study underlines the importance of developing applicable schemes to quantify coastal flood vulnerability for evolving future responses to adverse impacts of climate change.

Hydrodynamic versus GIS modelling for coastal flood vulnerability assessment: Which is better for guiding coastal management?

Under climate change, sea-level rise and increasing storm surge intensity will increase the likelihood of floods for low-lying coastal areas. Therefore, Coastal Flood Vulnerability Assessments (CFVA) are necessary to inform decision making for coastal management, where applicable. Two types of tools, hydrodynamic models and Geographic Information Systems (GIS), are commonly employed for these evaluations. Using Pigeon Point, Southwest Tobago, as a case study, this paper compares the application of both to determine the best approach for CFVA that can be used to guide coastal management. For hydrodynamic modelling, available and surveyed data on bathymetry, topography and tides along with current and future estimates of mean sea-level were used to create a structured and an unstructured grid model via the LISFLOOD-FP and TELEMAC-2D codes, respectively. For the GIS assessment, a coastal digital elevation model was developed using the same data on bathymetry and elevation. These models were used to project the present and future impact of storm surges on coastal flood extent at Pigeon Point. Outputs generated were compared and analysed within a GIS. Using predictions from TELEMAC-2D (full shallow water equations) as the baseline for comparison, average results acquired revealed that projections generated by all three models were consistent (less than 5.00% difference in flood predictions). To add certainty to the modelling results, all models were used to simulate an observed spring high tide event for model validation purposes. The Root Mean Squared Error (RMSE) was calculated as an indication of model performance. RMSE values indicated that all models were consistent and matched well to the field observations. However, further analysis revealed that inherent in the use of GIS for CFVA is hydraulic connectivity issues due to exclusion of flow dynamics, which leads to over-estimation in flood extent. Acknowledging the theory that over-estimation leads to over-management, it is suggested that hydrodynamic models are better suited for detailed CFVA, while GIS can be used rapidly as a potential indicator of flood exposure for large sites.

Modelling coastal flood vulnerability: Does spatially-distributed friction improve the prediction of flood extent?

This paper examines whether the application of spatially-distributed versus static friction in hydrodynamic modelling increases the accuracy of predicted coastal flood extent using Pigeon Point, southwest Tobago, as a case in point. A two-dimensional hydrodynamic flood model is created from acquired and surveyed bathymetric, topographic and tidal data via the LISFLOOD-FP model code. Using a Landsat 8 image of the study area, a Maximum Likelihood (ML) supervised classification was performed to distinguish different land cover classes within the study site. The classified Landsat 8 image was further processed by assigning friction values to each land cover class to create a spatially-distributed friction file in ASCII format for use in LISFLOOD-FP. Using the flood model developed, simultaneous simulations were performed to assess the impact of storm surges (varying levels) on coastal flood extent at Pigeon Point utilising a static friction value, which broadly defined the area (i.e., 0.02), and the spatially-distributed friction file generated. Model outputs were compared to determine the extent of difference in flood prediction obtained from the application of static versus spatially-distributed friction through a Geographic Information System (GIS) based analysis. The flood model developed was subsequently applied to simulate an observed spring tide event using both static and spatially-distributed friction value(s) defined and model performance in each case was evaluated using the Root Mean Squared Error (RMSE) approach. Collated results indicated that using spatially-distributed over static friction do not increase accuracy of predicted coastal inundation extent, nor improve model performance. However, it appears to provide more insight on flood timings, which can be useful for coastal management.