Science of the Total Environment 630 (2018) 211–221

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Assessing sea-level rise impact on saltwater intrusion into the root zone
of a geo-typical area in coastal east-central Florida
Han Xiao a,⁎, Dingbao Wang a, Stephen C. Medeiros a, Scott C. Hagen b, Carlton R. Hall c
a
Department of Civil, Environmental, and Construction Engineering, University of Central Florida, Orlando, FL, USA
b
Department of Civil and Environmental Engineering, Center for Computation & Technology, Center for Coastal Resiliency, Louisiana State University, Baton Rouge, LA, USA
c
Ecological Program, Integrated Mission Support Services, Kennedy Space Center, FL, USA

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• Saltwater intrusion due to sea-level rise

greatly increases root zone salinity level.
• FEMWATER models can simulate the
extent of saltwater intrusion.
• Root zone salinity level is highest in June
and lowest in October.

a r t i c l e i n f o a b s t r a c t

Article history: Saltwater intrusion (SWI) into root zone in low-lying coastal areas can affect the survival and spatial distribution
Received 2 December 2017 of various vegetation species by altering plant communities and the wildlife habitats they support. In this study, a
Received in revised form 14 February 2018 baseline model was developed based on FEMWATER to simulate the monthly variation of root zone salinity of a
Accepted 15 February 2018
geo-typical area located at the Cape Canaveral Barrier Island Complex (CCBIC) of coastal east-central Florida
Available online xxxx
(USA) in 2010. Based on the developed and calibrated baseline model, three diagnostic FEMWATER models
Editor: D. Barcelo were developed to predict the extent of SWI into root zone by modifying the boundary values representing
the rising sea level based on various sea-level rise (SLR) scenarios projected for 2080. The simulation results in-
Keywords: dicated that the extent of SWI would be insignificant if SLR is either low (23.4 cm) or intermediate (59.0 cm), but
Saltwater intrusion would be significant if SLR is high (119.5 cm) in that infiltration/diffusion of overtopping seawater in coastal low-
Sea-level rise lying areas can greatly increase root zone salinity level, since the sand dunes may fail to prevent the landward
Root zone migration of seawater because the waves of the rising sea level can reach and pass over the crest under high
Salinity (119.5 cm) SLR scenario.
FEMWATER
© 2018 Elsevier B.V. All rights reserved.
East-central Florida coast

1. Introduction

Coastal aquifer water quality is of great importance for human de-

velopment since more than 50% of the world's population lives within
⁎ Corresponding author. 60 km of the shoreline (Newmann et al., 2015). However, coastal
E-mail address: Han.Xiao@Knights.ucf.edu (H. Xiao). groundwater contamination from saltwater intrusion (SWI) has been

https://doi.org/10.1016/j.scitotenv.2018.02.184
0048-9697/© 2018 Elsevier B.V. All rights reserved.
212 H. Xiao et al. / Science of the Total Environment 630 (2018) 211–221
reported in many coastal areas and was deemed as a detrimental issue
affecting coastal water supplies, ecosystems, and economies (Barlow
and Reichard, 2010). The occurrence of SWI depends on coastal hydro-
geology, hydro-climatology such as sea levels and rainfall, and anthro-
pogenic activities such as groundwater pumping and land use change
(Bear et al., 1999). In recent years, sea-level rise (SLR) has been
highlighted as one of the main factors causing SWI (Chang et al., 2011;
Rasmussen et al., 2013; Werner and Simmons, 2009; Werner et al.,
2013), and the extent and intensity of SWI will further increase due to
the increase of the rate of SLR (IPCC, 2013). This is likely to cause envi-
ronmental problems including land salinization, groundwater quality
deterioration, closure or relocation of water supply wells, and ecosys-
tem degradation (Alizad et al., 2016; Bilskie et al., 2016; Hovenga
et al., 2016; Huang et al., 2015; Ketabchi et al., 2016; Kidwell et al.,
2017; Passeri et al. 2015a, 2015b & 2015c).
A saltwater/freshwater transition zone is formed between the fresh
and saline groundwater zones, and the density and salt concentration in
the transition zone varies spatially and temporally depending on local
hydrologic and geologic conditions (Bear, 1979). The migration of tran-
sition zone can be simulated by numerical models, such as SEAWAT
(Guo and Langevin, 2002) and SUTRA (Voss, 1984). These models
have been utilized to estimate the extent of SWI by modeling the migra-
tion of the transition zone under climate change and human develop-
ment (Datta et al., 2009; Hussain and Javadi, 2016; Kim et al., 2012;
Langevin, 2003; Lin et al., 2009; Mzila and Shuy, 2003; Qahman and
Larabi, 2006; Sanford and Pope, 2010; Xiao et al., 2016; Yu et al.,
2016). For example, Xiao et al. (2016) developed a SEAWAT model to
study the salinity distribution in the surficial aquifer of east-central Flor-
ida coast for assessing SLR and climate change impacts on SWI into sur-
ficial aquifer. However, note that both SEAWAT and SUTRA models can
simulate the extent of SWI into saturated zone (unconfined/confined
aquifer below water table), while are not capable of simulating SWI
into unsaturated zone (vadose zone above water table).
Root zone salinity affected by water table dynamics in coastal east-
central Florida is crucial to the survival of various vegetation species
(Purdum et al., 2002) and the distribution of plant communities and
habitats (Box et al., 1993; Foster et al., 2017; Saha et al., 2011; Saha
et al., 2015). Root zone salinity level provides a good measurement of
salt exposure to vegetation and an indication of vegetation impacts.
An increasing salinity level can cause serious consequences such as veg-
etation species dieback and limited recovery, shift in species composi-
tion from less to more salt-tolerant species, and reduction in biomass
production (Hall et al., 2014; Schmalzer, 1995; Steyer et al., 2007).
Therefore, it is of great importance to quantify the spatial and temporal
variation of root zone salinity and predict the extent of SWI under SLR
impacts, and numerical models capable of simulating salinity transport
under variable-density condition in unsaturated zone (or combined un-
saturated and saturated zones), such as FEMWATER (Lin et al., 1997),
are highly required.
In this study, two research questions were addressed: (1) what is the
current salinity level and where is the current saltwater/freshwater
transition zone in root zone of a geo-typical coastal area located at the
Cape Canaveral Barrier Island Complex (CCBIC) of coastal east-central
Florida?; and (2) how will the spatial and temporal variation of root
zone salinity and location of the saltwater/freshwater transition zone
change under various SLR scenarios? To answer these questions, a
three-dimensional finite-element variable-density FEMWATER model
was developed and calibrated using 2010 hydrologic conditions to sim-
ulate the monthly variation of root zone salinity in 2010 (baseline
model), and the extent of SWI into root zone under three SLR scenarios
projected to 2080 were estimated by developing three “diagnostic”
models based on the calibrated “baseline” model. The outcome of this
study contributes to ongoing research focused on forecasting vegetation
community responses to climate change (Foster et al., 2017), and serves
as an effective tool for climate change adaptation planning and decision-
making in the CCBIC and other low-lying coastal alluvial plains and bar-
rier island systems.
2. Overview of study area
2.1. Site description
The study area (i.e., model domain) covers approximately 0.45 km2
and is located in a biogeotypical coastal area located at the Cape
Canaveral Barrier Island Complex (CCBIC) of east-central Florida coast,
as shown in Fig. 1a. The lateral boundaries of the model domain are de-
termined from the water table contours simulated by a developed and
calibrated SEAWAT model (Xiao et al., 2016) as shown in Fig. 1b. From
Fig. 1b, the lateral boundaries correspond to (1) the groundwater divide
to the west; (2) the soil-water interface (coastline of the Banana River)
to the east; and (3) the no-flux boundary to the north and south. Unlike
the lateral boundaries, the vertical boundaries of the model domain are
determined physically. The upper boundary is the land surface (see
Fig. 1c) and the lower boundary is the bottom of the surficial aquifer,
therefore, both the unsaturated zone (from the land surface to water
table) and the saturated zone (from water table to the bottom of the
surficial aquifer) are included in the model domain. The land surface el-
evation varies from approximately −1.28 to 5.42 m NAVD88. Variation
in the local topography is relatively small since the study area is mainly
composed of broad and flat lowland. The land cover is mainly comprised
of shrub and brushland, grassland, and mixed scrub-shrub wetland. The
climate is classified as humid subtropical with hot/humid summers
(mean temperature 22 °C to 33 °C) and mild/dry winters (mean tem-
perature 10 °C to 22 °C), and the mean annual rainfall is 1366 mm
with a wet season lasting from June to October (Mailander, 1990). The
primary inflow is rainfall and the primary outflows are evapotranspira-
tion and submarine groundwater discharge to the Banana River
(Schmalzer et al., 2000). The Banana River, located on the southeast bor-
der of the study area, is a coastal lagoon connected with the Atlantic
Ocean through the Canaveral Lock to the east and Sebastian Inlet to
the southeast, and has a shallow, flat, seagrass covered bottom with
an average depth of 1.5 m, and the total dissolved solids (TDS) concen-
tration typically varies from 10 to 45 kg/m3 (Hall et al., 2014). The Ba-
nana River in this area is micro-tidal due to the distance to small
ocean passes, and its water level varies monthly from −0.33 to −
0.06 m NAVD88 approximately 4–5 cm higher than sea level of the At-
lantic Ocean (Foster et al., 2017; Smith, 1990 & Smith, 1993). The unsat-
urated zone in the study area is mainly composed of fine sand and the
water table depth varies from 0 to 2 m (Xiao et al., 2016), and the satu-
rated zone is mainly composed of fine to medium sand (Miller, 1986;
Williams and Kuniansky, 2016). The water table and local Banana
River level rise to the highest elevation late in the wet season (October)
and drop to the lowest elevation late in the dry season (May), mainly
controlled by the temporal variation of rainfall, evapotranspiration,
and the annual rise and fall of sea level (Foster et al., 2017).
2.2. Sea-level rise scenarios
The NASA Climate Adaptation Science Investigator (CASI)
Workgroup established in 2010 consists of scientists and engineers
from academia, private sectors, and non-governmental organizations,
and developed the scientific and technical basis for climate change ad-
aptation based on the Goddard Institute for Space Studies (GISS) global
climate model (GCM) Model E (Schmidt et al., 2006). Sea-level rise pro-
jections for the coastal areas located at the Cape Canaveral Barrier Island
Complex (CCBIC) of east-central Florida were regionalized and down-
scaled based on the GCM model as well as rapid ice-melt scenarios
(low, intermediate, and high) (Horton and Rosenzweig, 2010) using
the method described in Horton et al. (2011). The SLR scenarios applied
to the study area were developed by the Center for Climate Systems Re-
search, Earth Institute, Columbia University as part of the NASA CASI
 H. Xiao et al. / Science of the Total Environment 630 (2018) 211–221 213

Fig. 1. Study area and model domain: (a) Location; (b) Water table contours; (c) Land surface elevations.
214 H. Xiao et al. / Science of the Total Environment 630 (2018) 211–221

Program (Rosenzweig et al., 2014). Relative to the mean sea level in Table 2
2010, SLR scenarios in 2080 were projected to be 23.4, 59.0, and Fluid properties.

119.5 cm for the low, intermediate, and high ice-melt scenarios, Parameters Value (25 °C) Reference
respectively. Freshwater density 998 [kg/m3] Sharqawy et al.,
Seawater density 1025 [kg/m3] 2010
3. Model development Freshwater viscosity 0.0009 [kg/(m·s)]
Seawater viscosity 0.000925 [kg/(m·s)]
Concentration dependence 1 [m3/kg]
As mentioned above, a baseline model and three diagnostic models
coefficient a1
were developed in this study. The baseline model was developed and Concentration dependence 0.00073851 [m3/kg]
calibrated to simulate the monthly variation of root zone salinity in coefficient a2
2010, and the three diagnostic models were developed based on the cal- Concentration dependence 0.0000002054 [m3/kg]
ibrated baseline model by modifying the values of the boundary condi- coefficient a3
Concentration dependence 0
tions representing the rising Banana River water levels corresponding to coefficient a4
various SLR scenarios projected to 2080 to estimate the extents of SWI Concentration dependence 0.0096 [m3/kg]
into root zone in 2080, with the assumptions that: (1) the rainfall and coefficient a5
evapotranspiration patterns remain unchanged from 2010 to 2080; Concentration dependence 0.002 [m3/kg]
coefficient a6
(2) the Banana River water levels rise simultaneously in the same mag-
Concentration dependence −0.00000028409
nitude with SLR since the Banana River is connected to the Atlantic coefficient a7 [m3/kg]
Ocean at the Canaveral Lock and Sebastian Inlet; (3) the effect of astro- Concentration dependence 0.00000017045
nomic tides is ignored since the Banana River is micro-tidal environ- coefficient a8 [m3/kg]
ment; and (4) the effect of storm surges from cyclone swells is not
considered due to lack of data. Note that the baseline model and diag-
nostic models were simplified to focus on the effect of SLR on SWI into 13, the thickness of each layer was 1.0, 1.5, 2.0, and 2.5 m, respectively.
root zone based on these assumptions. Also, note that the term “SLR” The unsaturated zone was vertically discretized with a finer mesh than
in this study refers to the simultaneous rising of the Atlantic Ocean sea the saturated zone in order to enhance modeling accuracy and prevent
level and the Banana River water level. oscillations in the numerical output. In total, the three-dimensional
The boundary conditions were described in Section 3.2, and the mesh consists of 13,048 nodes and 21,762 elements. Although simula-
hydrogeologic parameters such as hydraulic conductivities, porosity, tion accuracy can further increase with finer mesh (smaller average
and dispersivities were summarized in Table 1. For the unsaturated spacing), the number of nodes and elements are required to be no
zone, soil properties were estimated based on soil types. The specified greater than 25,578 and 22,080 for executing FEMWATER (Lin et al.,
fluid properties such as freshwater/seawater density and viscosity, 1997).
and concentration dependence coefficients were summarized in The 12-month simulation period was divided into 12 stress periods
Table 2. so that the monthly variation of salinity distribution could be simulated.
The three-dimensional finite-element variable-density FEMWATER Within each stress period, further temporal discretization was intro-
program developed by Lin et al. (1997) was used as the simulation duced in the form of a time step specified to start at 0.1 day and increas-
code. Groundwater Modeling Systems (GMS), a graphical user interface ing by a multiplier of 1.1.
developed by AQUAVEO, was used to create model input files. The
model output was visualized using ArcGIS. 3.2. Boundary conditions/initial conditions

3.1. Spatial and temporal discretization Boundary conditions applied to the baseline model and the diagnos-
tic models included the Dirichlet boundary, the no-flux boundary, and
The criteria for the mesh design were that it: (1) shows the topo- the variable flux boundary. The Dirichlet boundary was assigned to
graphic variation of land surface; (2) represents the freshwater/saltwa- the lateral boundary to the southeast that represents the coastline of
ter transition zone; and (3) maximizes simulation accuracy while the Banana River (see the red line in Fig. 3). For the baseline model,
maintaining reasonable computer simulation time. the heads and TDS concentrations specified at the Dirichlet boundary
Horizontally, the two-dimensional unstructured finite element were obtained from field measurements in 2010 as shown in Fig. 4.
mesh consisted of 932 nodes and 1674 triangular elements as shown For the diagnostic models, the TDS concentrations remained unchanged
in Fig. 2. In order to increase the simulation accuracy especially in the but the heads were modified based on the SLR scenarios projected to
coastal region, the average spacing for Areas (1), (2), (3), and (4) was 2080. Note that coastal erosion and overtopping of seawater in coastal
40, 30, 20, and 10 m, respectively. Vertically, the model domain was di- low-lying areas due to SLR is inevitable in that sand dunes may not be
vided into 13 layers as shown in Fig. 2. The top elevation of Layer 1 was able to impede waves reaching and overtopping their crests, and were
set to the land surface elevation derived from the LiDAR data, and the approximated by comparing the land surface elevation of coastal low-
bottom elevation of Layer 13 was set to 10 m below land surface eleva- lands with the rising sea level in order to alter the locations of the
tion, since the thickness of the surficial aquifer is approximately 10 m. Dirichlet boundary as well as the boundary values with respect to vari-
The thickness of Layers 1 to 9 was 0.33 m uniformly. For Layers 10 to ous SLR scenarios. The no-flux boundary was assigned to all other lateral
boundaries (see the black lines in Fig. 3) as well as the bottom boundary.
Table 1 The variable flux boundary was assigned to the top of the model domain
Hydrogeologic parameters. to represent the precipitation and evapotranspiration flux. The mean
monthly precipitation and evapotranspiration flux in 2010 were ob-
Parameters Value Reference
tained from rain gauges operated by the St. Johns River Water Manage-
Horizontal hydraulic conductivity 15 [m/d] Xiao et al., 2016
ment District and the data collection sites of the U.S. Geological Survey
Vertical hydraulic conductivity 0.15 [m/d]
Porosity 0.43 [−] Florida Evapotranspiration Network, respectively (Fig. 5).
Soil bulk density 1600 [kg/m3] Initial conditions applied to the baseline model and diagnostic
Longitudinal dispersivity 6 [m] models included heads and concentrations. For the baseline model,
Transverse dispersivity 0.025 [m] the values of initial heads and concentrations were obtained from the
Molecular diffusion coefficient 0.00128 [m2/d]
output of the previously developed and calibrated SEAWAT model
 H. Xiao et al. / Science of the Total Environment 630 (2018) 211–221 215

Fig. 2. Two-dimensional (horizontal and vertical) mesh.

(Xiao et al., 2016). For the diagnostic models, the values of initial heads
and concentrations were obtained from the output of the developed and
calibrated baseline model.
3.3. Model calibration
Due to lack of groundwater monitoring wells within the study site,
the observed groundwater data (e.g., water tables, TDS concentrations)
were unavailable. Then, the simulated hydraulic heads from the
FEMWATER model were calibrated against the simulated hydraulic
heads from the developed and calibrated SEAWAT model (Xiao et al.,
2016), rather than the observed groundwater levels. The SEAWAT
model developed by Xiao et al. (2016) was a steady-state model that
simulated the annual-averaged heads and TDS concentrations in the
year of 2010, while the FEMWATER model developed in this study
was a transient model that simulated the monthly-averaged heads
and TDS concentrations in the year of 2010. Thus, before initiating the
calibration procedure, the simulated monthly-averaged heads from
the FEMWATER model were converted to annual-averaged heads. Be-
sides, the simulated heads from the bottom layer of the FEMWATER
model and the SEAWAT model were exported for comparison, while
the simulated heads from the top layer of the FEMWATER and
SEAWAT models were not exported for calibration because the top
layer of the FEMWATER model represented the top of unsaturated
zone above water table while the top layer of the SEAWAT model repre-
sented the top of saturated zone below water table. The locations of the
24 calibration targets were shown in Fig. 6.
Sensitivity analysis was conducted during the model calibration pro-
cess. Sensitivity analysis found that vertical hydraulic conductivity was
the dominant factor affecting the simulation results, since a slightly
change of the value can vary the simulation results. Simulation results
tend to be insensitive to other hydrogeologic parameters such as hori-
zontal hydraulic conductivity, porosity, and dispersivity. Therefore, the
values of vertical hydraulic conductivity were adjusted iteratively in

Fig. 3. Boundary conditions assigned to the lateral boundaries.

216 H. Xiao et al. / Science of the Total Environment 630 (2018) 211–221
Fig. 4. Monthly mean water level and TDS concentration of the Banana River in 2010.
Fig. 5. Monthly mean precipitation and evapotranspiration in 2010.
order to minimize the difference between the simulated heads from the
FEMWATER model and the simulated heads from the SEAWAT model
until a satisfactory agreement was achieved.
After calibration, a Nash-Sutcliffe model efficiency coefficient of 0.99
was achieved (Fig. 6), indicating a strong correlation between the hy-
draulic heads simulated by the FEMWATER and SEAWAT models. The
horizontal and vertical hydraulic conductivities were calibrated to 6
[m/d] and 0.022 [m/d], respectively.
Fig. 6. Locations of the calibration targets.
Note that the baseline model was calibrated in terms of the simu-
lated hydraulic heads only, while was not calibrated against the simu-
lated TDS concentrations. It was not necessary to do so in that the TDS
concentrations simulated by both the FEMWATER and SEAWAT models
at 21/24 calibration targets sites were approximately equal to zero.
4. Results
4.1. Root zone salinity simulated in 2010
In 2010, the simulated monthly variation of root zone salinity at four
points located 10, 20, 30, and 40 m far away from the coastline were
shown in Fig. 7. It can be observed from Fig. 7 that: (1) salinity level is
highest in June and lowest in October; and (2) salinity level reduces
from July to October and rises from November to June. Thus, it can be es-
timated from Fig. 7 that: (1) root zone salinity reduces in wet season be-
cause sufficient infiltrated rainwater can dilute and flush saltwater back
to the Banana River (rainfall from June to October accounting for more
than 70% of annual total rainfall); (2) root zone salinity rises in dry sea-
son (November to May) because infiltrated rainwater is insufficient and
unable to act as an effective freshwater barrier to impede SWI; (3) root
zone salinity in June and October is representative of the extent of SWI
since salinity level in June and October reaches its annual maximum and
minimum values. The spatial variation of root zone salinity in June and
October 2010, were shown in Fig. 8a and 9a, respectively. Note that
water quality of the saltwater/freshwater transition zone was further
classified as slightly saline (1–3 kg/m3), moderately saline
(3–10 kg/m3), and highly saline (10–35 kg/m3) according to NGWA
(2010). From Fig. 8a, a small portion (1.05%) of the root zone has its sa-
linity level higher than 1 kg/m3.
4.2. Root zone salinity projected to 2080
Projected to 2080, the spatial variation of root zone salinity in June
under low (23.4 cm) SLR, intermediate (59.0 cm) SLR, and high
(119.5 cm) SLR were shown in Fig. 8b, c, and d, respectively. From
Fig. 8b and c, the effects of low (23.4 cm) and intermediate (59.0 cm)
SLR were not distinctly different, since a small portion (1.21% and
1.43%) of the root zone has its salinity level higher than 1 kg/m3. How-
ever, from Fig. 8d, the effect of high (119.5 cm) SLR is very significant,
since 13.94% of the root zone has its salinity level higher than 1 kg/m3.
 H. Xiao et al. / Science of the Total Environment 630 (2018) 211–221
Fig. 7. Simulated monthly variation of root zone salinity at 10, 20, 30, and 40 m away from the coastline.
Projected to 2080, the spatial variation of root zone salinity in Octo-
ber under low (23.4 cm) SLR, intermediate (59.0 cm) SLR, and high
(119.5 cm) SLR were shown in Fig. 9b, c, and d, respectively. Compared
to Fig. 8b, c, and d, the impacts of SLR are very similar. A statistical anal-
ysis was conducted to find out the impacts of SLR on SWI into root zone,
the ratio (percentage) of influenced area (root zone salinity level higher
than 1 kg/m3) to total land area was calculated and tabulated in Table 3.
5. Implications
The astronomic tides of the Banana River were assessed by Bilskie
et al. (2017), whom developed and validated a two-dimensional,
depth-integrated ADCIRC model to numerically simulate the tidal hy-
drodynamics of the Indian River Lagoon (IRL) system (the Banana
River is an important component of the IRL system). Results from
Bilskie et al. (2017) indicated that: (1) tidal flows within the Banana
River are especially damped, since the Sebastian inlet chokes the tides
entering the Banana River and the expansive size and shallowness of
the estuary dissipates tidal energy; (2) the Banana River is micro-
tidal; and (3) tides in the Banana River is small at just millimeters in
range. Thus, the effect of astronomic tides on the extent of SWI into
root zone was ignored in this study.
Coastal areas in Florida get hit by storm surges from hurricanes and
cyclonic swells frequently in wet season from June to October, and
storm surge can cause significant seawater flooding and overtopping
on coastal lowlands in that waves can reach and pass over the crest of
the sand dunes if wave run-up and storm tide levels are high enough,
resulting in an increase of root zone salinity due to rapid infiltration/dif-
fusion of overtopping seawater (Lin et al., 2014). However, the effect of
storm surge on the extent of SWI into root zone was not accounted in
this study, since (1) there was no major hurricanes and cyclonic swells
recorded around the year of 2010; (2) storm surge monitoring data
were not available around the year of 2010; and (3) there was no
storm surge projections for the year of 2080.
As mentioned above, the objective of this study is to quantify the ef-
fect of SLR on the extent of SWI into root zone, whereas the effects of as-
tronomic tides and storm surges from cyclone swells are ignored. In
general, the FEMWATER simulation results corresponded with the
SEAWAT simulation results from Xiao et al. (2016), that is, the extent
of SWI due to eustatic SLR (without considering tides, storm surges or
event driven geomorphology) is significant if the SLR scenarios are
greater than 65–70 cm. If so, the sand dunes are not capable of
preventing rising seawater from migrating inland to inundate the
217
coastal low-lying areas. Then, overtopping of seawater causes salt to in-
filtrate/diffuse into root zone to rise the salinity level. Over time, root
zone salinity level might either decline gradually if fresh groundwater
is available for dilution/flushing, or continue to rise as a result of evapo-
transpiration if fresh groundwater is not introduced. Elevated root zone
salinity can influence the vegetation community by decreasing vegeta-
tion productivity, affecting soil physical/chemical characteristics, and
breaking the ecological balance of the area. Salinity is one of the domi-
nant factors controlling seed germination, plant growth, and water
and nutrient uptake (Shrivastava and Kumar, 2015). If root zone salinity
level is too high, water may not move to plant roots but may flow from
plant roots back to soil, resulting in plant dehydration causing yield de-
cline or even death of the plant (excessive chloride is toxic to plant).
Even if root zone salinity level is not too high, nitrogen uptake might
still be interfered, restricting plant growth and stopping plant reproduc-
tion. A continuous high level of root zone salinity can cause potential
consequences in the salt-affected areas of: (1) decline of natural vegeta-
tion productivity; (2) dieback and limited recovery of natural vegeta-
tion species; (3) shifts in natural vegetation species composition from
less salt-tolerant species to more salt-tolerant species; (4) changes of
landscape; and (5) degradation of biodiversity and destruction of natu-
ral habitat.
To minimize the effect of SLR on the extent of SWI into root zone, it is
important and necessary to prevent overtopping of seawater in coastal
low-lying areas to reduce the amount of saltwater infiltration/diffusion
into root zone. Construction and maintenance of artificial coastal de-
fense structures such as seawalls might be a good choice to stabilize
the beach and shoreline, and protect coastal low-lying areas from inun-
dation by SLR, but usually the cost is expensive. The micro-tidal study
area is not urbanized and there is no harbor, dock, or channel inlet/out-
let. Thus, it might not be necessary to spend a lot of money to build up
artificial coastal defense structures such as breakwaters/seawalls that
are primarily used to absorb shock of waves to protect coastline from
high energy waves. However, an economical but efficient way to pre-
vent SLR-induced shoreline erosion and overtopping of seawater in
coastal low-lying areas to protect root zone from SWI could be height-
ening and strengthening the existing sand dunes in coastal upland
areas and building up “new” sand dunes in coastal lowland areas to
act as effective coastal defense structures to withstand the rising sea
level.
Groundwater serves as a large subsurface water reservoir and is an
important source of water to support surface water bodies such as
creeks and wetlands, and the contribution of groundwater to surface
218 H. Xiao et al. / Science of the Total Environment 630 (2018) 211–221

Fig. 8. Simulated spatial variation of root zone salinity: (a) June 2010; (b) June 2080 (23.4 cm SLR); (c) June 2080 (59 cm SLR); (d) June 2080 (119.5 cm SLR).

water weighs more in the dry season (November to May). Within the
study area, surface water and groundwater are highly interconnected
because water table is very shallow and can be at or near the land sur-
face in low-lying areas, resulting in a high risk of surface water bodies
being contaminated by saline groundwater in that the porous and per-
meable surficial soil primarily composed of sand can transmit
groundwater and certain types of contaminants such as chloride to nat-
ural creeks, artificial ditches, and wetlands. An increase in chloride con-
centration in surface water bodies can create problems for plants and
animals by altering their metabolic pathways and rates of activities,
and the influence on coastal ecosystem can be at small or large scales
dependent on the magnitude of the changes in chloride concentration.
 H. Xiao et al. / Science of the Total Environment 630 (2018) 211–221 219

Fig. 9. Simulated spatial variation of root zone salinity: (a) October 2010; (b) October 2080 (23.4 cm SLR); (c) October 2080 (59 cm SLR); (d) October 2080 (119.5 cm SLR).

Therefore, future research should further evaluating the potential ef-
fects of SLR on altering human and ecosystem exposure pathways to
changes in chloride and other toxic or non-toxic contaminant concen-
trations by transmitting groundwater contaminants to surface water
bodies, with a focus on vulnerable sites located at coastal alluvial plains
in Florida and other coastal sites all over the world as well.
6. Conclusion
The purpose of this study was to diagnose the effects of SLR on salt-
water intrusion into the root zone of a geo-typical low-lying coastal al-
luvial plain area in the CCBIC in coastal east-central Florida in that the
root zone salinity level is crucial for the survival of vegetation species
220 H. Xiao et al. / Science of the Total Environment 630 (2018) 211–221
Table 3
Ratio of influenced area to total land area.
SLR scenarios Percentage (%) Percentage increase (%)
June 2010 (0 cm) 1.05 – Chang, S.W., Clement, T.P., Simpson, M.J., Lee, K., 2011. Does sea-level rise have an impact
June 2080 (23.4 cm) 1.21 0.16
June 2080 (59.0 cm) 1.43 0.38
June 2080 (119.5 cm) 13.94 12.89
October 2080 (0 cm) 1.01 – Foster, T.E., Stolen, E.D., Hall, C.R., Schaub, R., Duncan, B.W., Hunt, D.K., Drese, J.H., 2017.
October 2080 (23.4 cm) 1.15 0.14
October 2080 (59.0 cm) 1.37 0.36
October 2080 (119.5 cm) 13.83 12.82
and maintaining high biodiversity in the natural coastal environment. A
baseline FEMWATER model was developed and calibrated in terms of
hydraulic heads to simulate monthly variation of salinity distribution
in 2010. Based on the developed and calibrated baseline model, three
diagnostic FEMWATER models were developed to simulate the effect
of SLR on the extent of SWI into root zone by modifying the values of
the boundary conditions representing the rising Banana River water
level with respect to various SLR scenarios, while assuming that the
rainfall and evapotranspiration patterns from 2010 to 2080 remain un-
changed. It was indicated from the simulation results that the effect of
low (23.4 cm) and intermediate (59.0 cm) SLR would be insignificant
while the effect of high (119.5 cm) SLR would be very significant on in-
ducing SWI into root zone in that overtopping of seawater in coastal
low-lying areas caused by the rising sea level can infiltrate/diffuse into
root zone and rise salinity. Therefore, heightening/reinforcing the
existing sand dunes in coastal upland areas and building up “new”
sand dunes in coastal lowland areas could be an economical and effec-
tive approach to act as effective coastal defense structures to prevent
shoreline erosion and overtopping of seawater to alleviate the extent
of SWI into root zone.
The authors caution that astronomic tides as well as storm surges
from cyclone swells and changes in geomorphology through erosion
or accretion were not considered in this study and are known to dynam-
ically interact with SLR causing non-linear effects. The outcome of this
study contributes to ongoing research focused on forecasting vegetation
community responses to climate change (Foster et al., 2017), and serves
as an effective tool for climate change adaptation planning and decision-
making in the CCBIC and other low-lying coastal alluvial plains and bar-
rier island systems.
Acknowledgements
The authors thank the two anonymous reviewers for their insightful
comments that contributed to improving the manuscript. This research
was funded in part by the NASA's Kennedy Space Center, Ecological Pro-
gram, Climate Adaptation Science Investigators (CASI) Project (Award:
IHA-SA-13-006) and the Louisiana Sea Grant Laborde Chair Endow-
ment. The opinions, findings and conclusions expressed in this manu-
script were those of the author(s) and not necessarily those of the
NASA Kennedy Space Center and the Louisiana Sea Grant.
References
Alizad, K., Hagen, S.C., Morris, J.T., Medeiros, S.C., Bilskie, M.V., Weishampel, J.F., 2016.
Coastal wetland response to sea-level rise in a fluvial estuarine system. Earths Future
4 (11), 483–497.
Barlow, P.M., Reichard, E.G., 2010. Saltwater intrusion in coastal regions of North America.
Hydrogeol. J. 18 (1), 247–260.
Bear, J., 1979. Hydraulics of Groundwater. McGraw-Hill, New York.
Bear, J., Cheng, A., Sorek, S., Ouazar, D., Herrera, I., 1999. Seawater Intrusion in Coastal
Aquifers: Concepts, Methods and Practices (Theory and Applications of Transport in
Porous Media). Kluwer, Dordrecht, the Netherlands.
Bilskie, M.V., Hagen, S.C., Alizad, K., Medeiros, S.C., Passeri, D.L., Needham, H.F., Cox, A.,
2016. Dynamic simulation and numerical analysis of hurricane storm surge under
sea level rise with geomorphologic changes along the northern Gulf of Mexico. Earths
Future 4 (5), 177–193.
Bilskie, M.V., Bacopoulos, P., Hagen, S.C., 2017. Astronomic tides and nonlinear tidal dis-
persion for a tropical coastal estuary with engineered features (causeways): Indian
River lagoon system. Estuar. Coast Shelf S. 1–17.
Box, E.O., Crumpacker, D.W., Hardin, E.D., 1993. A climatic model for location of plant spe-
cies in Florida, USA. J. Biogeogr. 20 (6), 629–644.
on saltwater intrusion? Adv. Water Resour. 34, 1283–1291.
Datta, B., Vennalakanti, H., Dhar, A., 2009. Modeling and control of saltwater intrusion in a
coastal aquifer of Andhra Pradesh, India. J Hydro-environ Res 3, 148–159.
Modeling vegetation community responses to sea-level rise on Barrier Island sys-
tems: a case study on the Cape Canaveral Barrier Island complex, Florida, USA. PLoS
One 12 (8), e0182605.
Guo, W., Langevin, C.D., 2002. User's guide to SEAWAT: a computer program for simula-
tion of three-dimensional variable-density ground-water flow. Techniques of Water-
Resour Invest, Book 6. USGS, Reston, VA.
Hall, C.R., Schmalzer, P.A., Breininger, D.R., Duncan, B.W., Drese, J.H., Scheidt, D.A., Lowers,
R.H., Reyier, E.A., Holloway-Adkins, K.G., Oddy, D.M., Cancro, N.R., Provancha, J.A.,
Foster, T.E., Stolen, E.D., 2014. Ecological impacts of the space shuttle program at
John F. Kennedy Space Center, Florida. NASA/TM-2014-216639. NASA, Washington,
DC.
Horton, R., Rosenzweig, C., 2010. In: Rosenzweig, C., Solecki, W. (Eds.), Climate Risk Infor-
mation, Climate Change Adaptation in New York City: Building a Risk Management
Response. New York Academy of Sciences, pp. 148–228.
Horton, R., Gornitz, V., Bader, D.A., Ruane, A.C., Goldberg, R., Rosenzweig, C., 2011. Climate
hazard assessment for stakeholder adaptation planning in New York City. J. Appl.
Meteorol. Climatol. 50, 2247–2266.
Hovenga, P.A., Wang, D., Medeiros, S.C., Hagen, S.C., Alizad, K., 2016. The response of run-
off and sediment loading in the Apalachicola River, Florida to climate and land use
land cover change. Earths Future 4 (5), 124–142.
Huang, W., Hagen, S.C., Bacopoulos, P., Wang, D., 2015. Hydrodynamic modeling and anal-
ysis of sea-level rise impacts on salinity for oyster growth in Apalachicola Bay, Florida.
Estuar. Coast Shelf S. 156, 7–18.
Hussain, M.S., Javadi, A.A., 2016. Assessing impacts of sea level rise on seawater
intrusion in a coastal aquifer with sloped shoreline boundary. J Hydro-environ Res
11, 29–41.
IPCC, 2013. Climate Change 2013: The Physical Science Basis. Contribution of Working
Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate
Change. Cambridge University Press, Cambridge, United Kingdom, and New York,
NY, USA.
Ketabchi, H., Mahmoodzadeh, D., Ataie-Ashtiani, B., Simmons, C.T., 2016. Sea-level rise
impacts on seawater intrusion in coastal aquifers: review and integration. J. Hydrol.
535, 235–255.
Kidwell, D.M., Dietrich, J.C., Hagen, S.C., Medeiros, S.C., 2017. An Earth's future special col-
lection: impacts of the coastal dynamics of sea level rise on low-gradient coastal land-
scapes. Earths Future 5 (1), 2–9.
Kim, S.D., Lee, H.J., Park, J.S., 2012. Simulation of seawater intrusion range in coastal aqui-
fer using the FEMWATER model for disaster information. Mar. Georesour.
Geotechnol. 30 (3), 210–221.
Langevin, C.D., 2003. Simulation of submarine groundwater discharge to a marine estu-
ary: Biscayne Bay, Florida. Groundwater 41 (6), 758–771.
Lin, H.C.J., Richards, D.R., Talbot, C.A., Yeh, G.T., Cheng, J.R., Cheng, H.P., Jones, N.L., 1997.
FEMWATER: a three-dimensional finite element computer model for simulating
density-dependent flow and transport in variably saturated media. Technical Report
CHL-97-12.
Lin, J., Snodsmith, B., Zheng, C., Wu, J., 2009. A modeling study of seawater intrusion in Al-
abama Gulf Coast, USA. Environ. Geol. 57, 119–130.
Lin, N., Lane, P., Emanuel, K.A., Sullivan, R.M., Donnelly, J.P., 2014. Heightened hurricane
surge risk in northwest Florida revealed from climatological-hydrodynamic modeling
and paleorecord reconstruction. J. Geophys. Res.-Atmos. 119 (14), 8606–8623.
Mailander, J.L., 1990. Climate of the Kennedy Space Center and vicinity. NASATech. Memo.
103498. NASA, Washington, DC.
Miller, J.A., 1986. Hydrogeologic framework of the Floridan aquifer system in Florida and
in parts of Georgia, Alabama, and South Carolina. U.S. Geological Survey Professional
Paper 1403-B.
Mzila, N., Shuy, E.B., 2003. Studies on groundwater salinity distribution in a coastal
reclaimed land in Singapore. International Conference on Estuaries and Coasts, No-
vember 9–11, 2003, Hangzhou, China.
Newmann, B., Vafeidis, A.T., Zimmermann, J., Nicholls, R.J., 2015. Future coastal population
growth and exposure to sea-level rise and coastal flooding - a global assessment. PLoS
One 10 (3), e0118571.
NGWA, 2010. Brackish Groundwater. National Groundwater Association Information
Brief. http://www.ngwa.org/mediacenter/briefs/documents/brackish_water_info_
brief_2010.pdf (Accessed June 2016).
Passeri, D.L., Hagen, S.C., Bilskie, M.V., Medeiros, S.C., 2015a. On the significance of incor-
porating shoreline changes for evaluating coastal hydrodynamics under sea level rise
scenarios. Nat. Hazards 75 (2), 1599–1617.
Passeri, D.L., Hagen, S.C., Medeiros, S.C., Bilskie, M.V., 2015b. Impacts of historic morphol-
ogy and sea level rise on tidal hydrodynamics in a microtidal estuary (Grand Bay,
Mississippi). Cont. Shelf Res. 111, 150–158.
Passeri, D.L., Hagen, S.C., Medeiros, S.C., Bilskie, M.V., Alizad, K., Wang, D., 2015c. The dy-
namic effects of sea level rise on low-gradient coastal landscapes: a review. Earths Fu-
ture 3 (6), 159–181.
Purdum, E.D., Krafft, P.A., Anderson, J., Bartos, B., McPherson, S., Penson, G., Tramontana,
E., 2002. Florida waters: a water resources manual from Florida's water management
districts.
 H. Xiao et al. / Science of the Total Environment 630 (2018) 211–221
Qahman, K., Larabi, A., 2006. Evaluation and numerical modeling of seawater intrusion in
the Gaza aquifer (Palestine). Hydrogeol. J. 14, 713–728.
Rasmussen, P., Sonnenborg, T.O., Goncear, G., Hinsby, K., 2013. Assessing impacts of cli-
mate change, sea level rise, and drainage canals on saltwater intrusion on coastal
aquifer. Hydrol. Earth Syst. Sci. 17, 421–443.
Rosenzweig, C., Horton, R.M., Bader, D.A., Brown, M.E., DeYoung, R., Dominguez, O.,
Fellows, M., Friedl, L., Graham, W., Hall, C., Higuchi, S., Iraci, L., Jedlovec, G., Kaye, J.,
Loewenstein, M., Mace, T., Milesi, C., Patzert, W., Stackhouse, P.W., Toufectis, K.,
2014. Enhancing climate resilience at NASA centers: a collaboration between science
and stewardship. Bull. Am. Meteorol. Soc. 95 (9), 1351–1363.
Saha, A.K., Saha, S., Sadle, J., Jiang, J., Ross, M.S., Price, R.M., Sternberg, L., Wendelberger,
K.S., 2011. Sea level rise and South Florida coastal forests. Clim. Chang. 107, 81–108.
Saha, S., Sadle, J., Heiden, C., Sternberg, L., 2015. Salinity, groundwater, and water uptake
depth of plants in coastal uplands of Everglades National Park (Florida, USA).
Ecohydrology 8 (1), 128–136.
Sanford, W.E., Pope, J.P., 2010. Current challenges using models to forecast seawater in-
trusion: lessons from the eastern shore of Virginia, USA. Hydrogeol. J. 18, 73–93.
Schmalzer, P.A., 1995. Biodiversity of saline and brackish marshes of the Indian River la-
goon: historic and current patterns. B. Mar. Sci. 57 (1), 37–48.
Schmalzer, P.A., Hensley, M.A., Mota, M., Hall, C.R., Dunlevy, C.A., 2000. Soil, groundwater,
surface water, and sediments of Kennedy Space Center, Florida: background chemical
and physical characteristics. NASA/Technical Memorandum-2000-208583, NASA.
Schmidt, G.A., Ruedy, R., Hansen, J.E., Aleinov, I., Bell, N., Bauer, M., ... Yao, M.S., 2006.
Present-day atmospheric simulations using GISS model E: comparison to in-situ, sat-
ellite, and reanalysis data. J. Clim. 19, 153–192.
Sharqawy, M.H., Lienhard, V.J.H., Zubair, S.M., 2010. Thermophysical properties of seawa-
ter: a review of existing correlations and data. Desalin. Water Treat. 16, 354–380.
Shrivastava, P., Kumar, R., 2015. Soil salinity: a serious environmental issue and plant
growth promoting bacteria as one of the tools for its alleviation. Saudi J. Biol. Sci.
22, 123–131.
221
Smith, N.P., 1990. An introduction to the tides of Florida's Indian River Lagoon. II. Currents.
Florida. Scientist 56, 216–225.
Smith, N.P., 1993. Tidal and wind-driven transport between Indian River and Mosquito
Lagoon, Florida. Florida. Scientist 56, 235–246.
Steyer, G.D., Perez, B.C., Piazza, S.C., Suir, G., 2007. Potential consequences of saltwater in-
trusion associated with hurricanes Katrina and Rita: Chapter 6C in Science and the
storms-the USGS response to the hurricanes of 2005. Circular 1306-6C.
Voss, C.I., 1984. SUTRA – A finite-element simulation model for saturated-unsaturated,
fluid density-dependent ground-water flow with energy transport or chemically-
reactive single-species solute transport: U.S. Geological Survey Water-Resources In-
vestigations Report 84-4369.
Werner, A.D., Simmons, C.T., 2009. Impact of sea-level rise on sea water intrusion in
coastal aquifers. Groundwater 47 (2), 197–204.
Werner, A.D., Bakker, M., Post, V.E.A., Vandenbohede, A., Lu, C., Ataie-Ashtiani, B.,
Simmons, C.T., Barry, D.A., 2013. Seawater intrusion processes, investigation and
management: recent advances and future challenges. Adv. Water Resour. 51, 3–26.
Williams, L.J., Kuniansky, E.L., 2016. Revised hydrogeologic framework of the Floridan
aquifer system in Florida and parts of Georgia, Alabama, and South Carolina (ver.
1.1, March 2016): U.S. Geological Survey Professional Paper 1807. 140 p., 23 pls.
https://doi.org/10.3133/pp1807.
Xiao, H., Wang, D., Hagen, S.C., Medeiros, S.C., Hall, C.R., 2016. Assessing the impacts of
sea-level rise and precipitation change on the surficial aquifer in the low-lying coastal
alluvial plains and barrier islands, east-central Florida (USA). Hydrogeol. J. 24 (7),
1791–1806.
Yu, X., Yang, J., Graf, T., Koneshloo, M., O'Neal, M.A., Michael, H.A., 2016. Impact of topog-
raphy on groundwater salinization due to ocean surge inundation. Water Resour. Res.
52, 5794–5812.