Page 1 of 19
European Journal of Applied Sciences – Vol. 12, No. 4
Publication Date: August 25, 2024
DOI:10.14738/aivp.124.17312.
Lotero, S. S., Bloetscher, F., Hoque, M., Liu, W., Meeroff, D. E., Mitsova, D., Nagarajan, S., Salazar, S., Su, H., Teegavarapu, R., Xie,
Z., Yong, Y., & Zhang, C. (2024). Incorporating Flood Inundation to Flood Risk Modeling. European Journal of Applied Sciences,
Vol - 12(4). 241-259.
Services for Science and Education – United Kingdom
Incorporating Flood Inundation to Flood Risk Modeling
Stephanya Salazar Lotero
Florida Atlantic University
Frederick Bloetscher
Florida Atlantic University
Mushfiqul Hoque
Florida Atlantic University
Wiebo Liu
Florida Atlantic University
Daniel E. Meeroff
Florida Atlantic University
D. Mitsova
Florida Atlantic University
S. Nagarajan
Florida Atlantic University
S. Salazar
Florida Atlantic University
Hongbo Su
Florida Atlantic University
Ramesh Teegavarapu
Florida Atlantic University
Zhixiao Xie
Florida Atlantic University
Yan Yong
Florida Atlantic University
Caiyun Zhang
Florida Atlantic University
ABSTRACT
Coastlines are particularly vulnerable to flood under multiple conditions,
including heavy precipitation, high sea levels and tropical storm surge. These
Page 2 of 19
Services for Science and Education – United Kingdom 242
European Journal of Applied Sciences (EJAS) Vol. 12, Issue 4, August-2024
conditions should be considered to assess and manage inundation with more
effectiveness. As a part of a larger research effort to develop a watershed level
screening tool to identify areas with potential for flooding, incorporating readily
available data on topography, ground, and surface water elevations, tidal data for
coastal communities, soils, and rainfall data. However, the most destructive
impacts are more likely to be from storm surge – moving waves of water that
inundate the coast. While SLOSH and other models can be used for this purpose,
this effort reviewed prior storms to determine the impact surge actually has on
coastal communities in South Florida. What was found was that in urban areas, a
wedge of surge waters could be created. Likewise, a wedge could be created for
natural areas, although natural area extension was double that of urban areas. The
wedge can be added to storm events, acting as inputs for Cascade 2001 software to
calculate the headwater height for probabilities of inundation.
Keywords: Storm surge, flooding, modeling, watershed
INTRODUCTION
Flooding is the most common and costly disaster in the United States (FEMA 2018). Among
the flood options is a nuisance flooding (small amounts in the street), tidal flooding (which
typically occurs during exceptionally high tides, causing seawater to spill onto land and
inundate low-lying areas until the tide recedes), rainfall events caused by heavy precipitation,
and storm surge and high waves during coastal storms (CDC, 2017). These events can occur
concurrently, creating major challenges for coastal resiliency. As a result, communities in
coastal areas are at risk of socio-economic and health impacts from increased coastal flooding.
Figure 1 shows the number of nuisance flooding events and the less frequent moderate and
significant floods that occurred in the 1950s as compared to the 2010s. Since the 1950s,
almost all the sites shown on the map have increased in the number of days per year with
floods.
According to NOAA, the U.S. experienced a record of 28 weather and climate disasters, as with
a total cost of at least $92.9 billion. The most recurrent disaster type was severe storm, with 19
events, followed by flooding, with four events. From a broader perspective, over the last 5-
years (2019-2023), an annual average of 20 billion-dollar disasters have occurred with an
annual average cost of $120.6 billion (NOAA NCEI, 2024). Worse, flood risks are growing in
most coastal regions in the U.S. and is expected to continue to grow.
On 28th September 2022, Hurricane Ian landed in southwestern Florida as a category 4
hurricane. At least 150 people died, and the area incurred over $112 billion in damages,
making it the costliest and third costliest in Florida and the United States, respectively, as
reported by the US National Hurricane Center. Figures 1-4 show examples of the damage
wrought by Hurricane Ian on Fort Myers Beach and adjacent Sanibel Island. Storm surge on
Sanibel Island reach a maximum of 16 feet per NAA records. Sanibel Island is a curved island
that is 1 mile wide at its widest off the coast of Fort Myers. The island’s 6400 people were left
without water, sewer and power, along with the loss of many structures.
Page 3 of 19
243
Lotero, S. S., Bloetscher, F., Hoque, M., Liu, W., Meeroff, D. E., Mitsova, D., Nagarajan, S., Salazar, S., Su, H., Teegavarapu, R., Xie, Z., Yong, Y., &
Zhang, C. (2024). Incorporating Flood Inundation to Flood Risk Modeling. European Journal of Applied Sciences, Vol - 12(4). 241-259.
URL: http://dx.doi.org/10.14738/aivp.124.17312
Figure 1: Frequency of Flooding Along U.S. Coasts, 2010-2015 Versus 1950-1959 (U.S. EPA,
2016)
Figure 2: Damage from Hurricane Ian storm surge
Page 4 of 19
Services for Science and Education – United Kingdom 244
European Journal of Applied Sciences (EJAS) Vol. 12, Issue 4, August-2024
Figure 3: Damage from Hurricane Ian
Figure 4: Damage from Hurricane Ian
Page 5 of 19
245
Lotero, S. S., Bloetscher, F., Hoque, M., Liu, W., Meeroff, D. E., Mitsova, D., Nagarajan, S., Salazar, S., Su, H., Teegavarapu, R., Xie, Z., Yong, Y., &
Zhang, C. (2024). Incorporating Flood Inundation to Flood Risk Modeling. European Journal of Applied Sciences, Vol - 12(4). 241-259.
URL: http://dx.doi.org/10.14738/aivp.124.17312
Figure 5: Damage from Hurricane Ian
Many factors are responsible for coastal flooding, such as development along the coast.
Another factor is climate change. Rising temperatures from climate change contribute to
increases in sea level, severe storms, and storm surge and change in precipitation patterns
(CDC, 2017). All these changes increase the risks associated with flooding.
Figure 6: Past and project sea level rise in southeast Florida (Bloetscher et al 2021)
-20
0
20
40
60
80
100
1900 1950 2000 2050 2100
Inches above NAVD88 Datum=0
Year
Key West Historic Virginia Key Historic
IPCC AR5 Median USACE High
NOAA High Linear (Key West Historic)
Linear (Virginia Key Historic)
Page 7 of 19
247
Lotero, S. S., Bloetscher, F., Hoque, M., Liu, W., Meeroff, D. E., Mitsova, D., Nagarajan, S., Salazar, S., Su, H., Teegavarapu, R., Xie, Z., Yong, Y., &
Zhang, C. (2024). Incorporating Flood Inundation to Flood Risk Modeling. European Journal of Applied Sciences, Vol - 12(4). 241-259.
URL: http://dx.doi.org/10.14738/aivp.124.17312
Figure 8: Vulnerability of Sea Level Rise Along the Coast (Carter et al., 2014)
These heavy precipitation events are already occurring more often in the U.S. due to our
warming climate. Figure 9 shows how the annual number of heavy downpours, defined as
extreme two-day precipitation events, for the contiguous United States has increased –
particularly between the 1950s and the 2000s (CDC,2017).
Figure 9: Change in Number of Extreme Precipitation Events (USGCRP, 2016)
Compounding sea level rise and increased precipitation, storm surge is an abnormal rise in
the water that is over and above the regular tide level. Storm surges are caused by wind,
waves, and low atmospheric pressure from severe storms, such as hurricanes. Storm surges
can be particularly damaging when they occur at the same time as the daily high tide. Storm
Page 8 of 19
Services for Science and Education – United Kingdom 248
European Journal of Applied Sciences (EJAS) Vol. 12, Issue 4, August-2024
surge is dependent on tides, movement of the storm, wind speeds, direction and a number of
other factors.
This research aims to add to the screening tool developed by Bloetscher, et al (2021) and
Zhang et al (2020), and builds upon more specific projects by Romah, (2012), Wood (2015)
and Rojas (2020).
METHODS
From Rojas (2020), a screening tool was introduced to assess the risk due to flooding in the
coastal communities using maps of open space, impervious area, and rainfall; a runoff model
after the 3-day 25 years storm was created (Figure 10). These flooding results were then used
to develop risk contours for flooding. However, these results are limited to the application of
sea level rise (from Romah, 2011 and Wood, 2015), plus the soil capacity derived by Rojas
(2020) and the application of rainfall to these scenarios via the South Florida Water
Management District’s Cascade 2001 modeling software.
Figure 10: Screening Tool for the Flood Risk
Storm Surge Modelling
Storm surge is one of the major natural hazards, which can cause damage to coastal
infrastructure and local environments. Because field measurement stations are often limited,
numerical modeling has often been used to study storm surge dynamics, coastal morphology,
and coastal hazard impacts (Irish et al., 2011; Vijayan et al.,2021). Verification of the models is
difficult due to the limited nature, and varying parameters associated with storm events and
the surges generated.
ADCIRC produces reliable results while simulating coastal storm surge, as shown by several
studies worldwide (Vijayan et al.,2021; Yin et al., 2017; Westerink et al., 2008; Lin et al., 2010;
Fritz et al. 2010). It simulates water levels and velocities by solving the coupled equations of
depth-integrated generalized wave continuity equation (GWCE) and two-dimensional depth- integrated (2DDI) momentum equations. These equations are solved by the finite element
Page 10 of 19
Services for Science and Education – United Kingdom 250
European Journal of Applied Sciences (EJAS) Vol. 12, Issue 4, August-2024
of storm surge considered. Several things were noted when the surge was compared to the
land use, noting virtually all these sites had relatively low, level, coastal topography,
consisting mainly of dune that may be developed:
• Storm surge can go 4 mile inland in areas with no coastal development,
• Storm surge can go 2 mi inland in low lying areas with limited development, and
• Storm surge goes about 1 mi inland in areas that are developed, noting most have
water internally so there is a bayside surge to contend with.
Figure 11: Storm Surge in Naples due the hurricane Irma
Figure 12: Storm Surge in Fort Myers due the hurricane Charley
Figure 13: Storm Surge in Naples due the hurricane Irma
Page 11 of 19
251
Lotero, S. S., Bloetscher, F., Hoque, M., Liu, W., Meeroff, D. E., Mitsova, D., Nagarajan, S., Salazar, S., Su, H., Teegavarapu, R., Xie, Z., Yong, Y., &
Zhang, C. (2024). Incorporating Flood Inundation to Flood Risk Modeling. European Journal of Applied Sciences, Vol - 12(4). 241-259.
URL: http://dx.doi.org/10.14738/aivp.124.17312
Figure 14: Storm Surge in Fort Myers due the hurricane Charley
Figure 15: Storm Surge in St. Augustine due the hurricane Mathew
Figure 16: Apalachicola due to Hurricane Mathew
Page 14 of 19
Services for Science and Education – United Kingdom 254
European Journal of Applied Sciences (EJAS) Vol. 12, Issue 4, August-2024
roadway on the island. City staff confirmed at 2-3 of water existed on all City streets was
Hurricane Ian went by.
Time of Concentration
After creating a surge wedge, ArcHydro tools were used to generate the drainage line and
drainage points for the basin. This was done to determine the flow direction and the longest
drainage path exiting the basin. Using the length of the longest drainage path, the time of
concentration was determined.
Generation of the Risk Maps
At this stage, Cascade 2001 was used to determine the flood risk. Cascade 2001 is a
hydrologic/hydraulic routing model developed by the South Florida Water Management
District that allows the analysis of multiple cascading basins connected by discharge
structures. The model is organized in five main windows – Project, Off-site Receiving Body,
Basin, Structure, and User-defined Rainfall Distribution – that allow access and incorporation
of the project's unique characteristics (South Florida Water Management District, 2001).
The "Project Window" includes information concerning the entire project, such as name,
reviewer, time step, iterations, print time step, time window, etc. The "Off-site Receiving Body
Window" specifies downstream boundary conditions for the basins, including receiving
bodies' hydrographs. The "Basin Window" allows the incorporation of the previously
discussed GIS information for each basin such as area, time of concentration, ground storage,
initial stage, stage-storage relationship, and rainfall characteristics. The "Structure Window"
supports three structures - gravity, pump station, and gated spillway - to connect the multiple
basins. Finally, the "User-defined Rainfall Distribution Window" allows the creation of custom
rainfall distributions, specifying the time and corresponding ratio of the cumulative rainfall
for the event (South Florida Water Management District, 2001). For example, for the larger
project, FAU ran analyses for a 3-day 25-year, 1 day 10 year and 1 day 100-year rainfall event.
The time of concentration, area in acres, topography, soil storage, and precipitation of each
basin is used as an input for Cascade 2001.
Using the same methodology, the headwater height was determined. After determining the
headwater height, a spatially-temporally quantified understanding of nuisance-destructive
flood potential in the area given observed values. “Risk”, then, is a function of compounding
geo-hydrological features, namely, SW, GW, tides, topography, and time of year. A GIS-based
algorithm and kriging spatial interpolation will produce layers of the greatest observable
hydrographic surfaces. These outputs can then be compared with high resolution topographic
model informed by LiDAR to develop digital elevation models that reflect the observed risk
landscape. These models can then be combined into CASCADE to produce vector and volume
information, in combination with soils, vegetation and percent impervious surfaces, allowing
the observed model outputs to be extrapolated into a more predictive context.
The uncertainties associated with the DEM vertical accuracy, estimated depths to
groundwater table, and the modeling approach itself are incorporated in the RMSE
computation. One can create a z-score surface from which to derive probabilities of
inundation:
Page 16 of 19
Services for Science and Education – United Kingdom 256
European Journal of Applied Sciences (EJAS) Vol. 12, Issue 4, August-2024
Figure 22: Groundwater and Surface water stations of the study area
Figure 23: Water Table elevation of the study area.
Figure 24: Soil storage capacity of the study area.