events; and (3) an in-place drowning which turns the island
into a submarine deposit body. Although there exist some
cases to support the latter two modes, the most common
mode of barrier island transgression is the continuous landward migration through the combined effects of shoreface
erosion, overwash, and inlet floods (including storm
breaching). Through a continuous “rolling over” itself,
the barrier island eventually merges with the mainland,
with its upper layers of sediment eroded and recycled.
Barrier regression refers to an offshore expansion of the
landform and shoaling growth. It is a result of an excess of
sediment supply to the island. Sediment supply mainly
comes from three sources: river input, longshore drift,
and onshore migration of submarine sandbars. In the process of barrier island regression, the outer (ocean-ward)
shoreline progrades seaward, while the inner (lagoonward) shoreline remains relatively stable, forming a wide
low-lying plain characterized by multiple dune ridges,
normally with the most seaward foredune ridges
possessing the highest elevation. Such high foredune
ridges may prevent overwash and thus help to protect the
island from storm erosion; however, meanwhile they also
block the transport of sediment to the backshore and may
accelerate the erosion on the inner shoreline during
eustatic sea-level rise over the long term. Accompanied
by a decreased sediment supply, this may lead to
a switch of the barrier evolution to a transgression phase.
Numerical modeling
Due to high sensitivity to boundary conditions, natural
barrier islands serve as an ideal laboratory for numerical
studies of multi-scale physical processes on the coastal
morphological evolution. They can also be studied as
proxies of long-term climate change (Zhang et al., 2014).
Morphogenesis and evolution of barrier islands are complicated due to the influence of many processes occurring at
different temporal and spatial scales as discussed previously. Due to difficulties resolving all relevant processes
and their interplay in an integrated numerical model, simplifications are usually used in mathematical descriptions
of these processes and their corresponding scales.
The most common numerical models available for
study of barrier island evolution are 2-dimensional vertical
(2DV) cross-shore profile models. In these models, morphological response of a cross-shore coastal profile to
actions of several key processes is used to represent the
evolution of the whole barrier island. The coastal profile
is selected in such a way that it should be able to represent
typical characteristics of the barrier island and its adjacent
environments. The profile starts from a high terrestrial
point at the mainland and extends seaward to an offshore
closure point. Outer areas beyond these two points are presumed unchanged and do not impose any effect on the barrier island system during the time span of interest. After
a setup of the initial profile shape and other parameters
(e.g., sediment composition, grain size, substrate lithology), response of the profile shape and underlying
stratigraphy to influences of different processes is calculated through a set of equations. Depending on the equations adopted, 2DV cross-shore profile models can be
further classified into two different types: process based
and behavior oriented. Process-based models apply a set
of differential equations to describe the wave transformation, sediment transport, and subsequent bed elevation
change on the profile. Impacts of storm surge, eustatic
sea-level change, and tectonic movement are implemented
in the equations through a parameterization of boundary
conditions (i.e., incoming wave properties and water
level). Examples of process-based 2DV models can be
found in Masetti et al. (2008), Rosati et al. (2010), and
Zhang et al. (2013). On the contrary, behavior-oriented
models (e.g., Roy et al., 1994; Cowell et al., 1995; Storms
et al., 2002; Stolper et al., 2005; Moore et al., 2010)
describe the profile change by a set of empirical functions
of changes of sediment supply, sea level, and shoreface
geometry, without simulating the detailed processes
involved in sediment transport.
The validity of 2DV cross-shore profile models is based
on three pre-assumptions: (1) a zero net sediment exchange
at the boundary (thus sediment is conserved in the profile);
(2) evolution of the shoreface part of the profile evolves
toward a predefined shape (the so-called equilibrium),
which is determined by the grain size of the shoreface sediment and the mean wave climate (e.g., Bruun, 1962; Dean,
1991; Dean, 1997); and (3) alongshore uniformity of offshore wave parameters and nearshore isobaths along the
coastline (thus the gradient of longshore sediment transport
rate is zero and does not affect the profile change).
2DV cross-shore profile models have proven to be
more useful in providing detailed insights into the fundamental driving mechanisms of barrier island development
than conceptual models. However, one should always
keep in mind the limitations of validity which may hinder
application of a 2DV model to a real case. Another factor
affecting the reliability of a 2DV model is an exclusion of
inlet effects, which are most critical in controlling barrier
island morphodynamics according to Leatherman
(1979). An extension of an individual profile to an area
might overcome these limitations; however, this requires
much greater effort in bridging the different scales that
are involved in barrier island morphodynamics. Development of such models is still at an early stage. An example
of such models is presented by Zhang et al. (2012, 2014).
A hybrid and parallel coupling of process-based and
behavior-oriented modules provides a way to resolve the
relevant processes at their corresponding scales with an
affordable computational expense. In the model, wave
processes (propagation, transformation, refraction, and
breaking), currents, and subaqueous suspended sediment
transport are solved in process-based modules, while subaerial aeolian transport, bed-load transport, and land-sea
transition processes (e.g., cliff erosion) are simulated
either in behavior-oriented manners or by cellular automata approach. The model was applied to investigate the
morphogenesis and evolution of a Holocene barrier island
50
BARRIER ISLAND
into a submarine deposit body. Although there exist some
cases to support the latter two modes, the most common
mode of barrier island transgression is the continuous landward migration through the combined effects of shoreface
erosion, overwash, and inlet floods (including storm
breaching). Through a continuous “rolling over” itself,
the barrier island eventually merges with the mainland,
with its upper layers of sediment eroded and recycled.
Barrier regression refers to an offshore expansion of the
landform and shoaling growth. It is a result of an excess of
sediment supply to the island. Sediment supply mainly
comes from three sources: river input, longshore drift,
and onshore migration of submarine sandbars. In the process of barrier island regression, the outer (ocean-ward)
shoreline progrades seaward, while the inner (lagoonward) shoreline remains relatively stable, forming a wide
low-lying plain characterized by multiple dune ridges,
normally with the most seaward foredune ridges
possessing the highest elevation. Such high foredune
ridges may prevent overwash and thus help to protect the
island from storm erosion; however, meanwhile they also
block the transport of sediment to the backshore and may
accelerate the erosion on the inner shoreline during
eustatic sea-level rise over the long term. Accompanied
by a decreased sediment supply, this may lead to
a switch of the barrier evolution to a transgression phase.
Numerical modeling
Due to high sensitivity to boundary conditions, natural
barrier islands serve as an ideal laboratory for numerical
studies of multi-scale physical processes on the coastal
morphological evolution. They can also be studied as
proxies of long-term climate change (Zhang et al., 2014).
Morphogenesis and evolution of barrier islands are complicated due to the influence of many processes occurring at
different temporal and spatial scales as discussed previously. Due to difficulties resolving all relevant processes
and their interplay in an integrated numerical model, simplifications are usually used in mathematical descriptions
of these processes and their corresponding scales.
The most common numerical models available for
study of barrier island evolution are 2-dimensional vertical
(2DV) cross-shore profile models. In these models, morphological response of a cross-shore coastal profile to
actions of several key processes is used to represent the
evolution of the whole barrier island. The coastal profile
is selected in such a way that it should be able to represent
typical characteristics of the barrier island and its adjacent
environments. The profile starts from a high terrestrial
point at the mainland and extends seaward to an offshore
closure point. Outer areas beyond these two points are presumed unchanged and do not impose any effect on the barrier island system during the time span of interest. After
a setup of the initial profile shape and other parameters
(e.g., sediment composition, grain size, substrate lithology), response of the profile shape and underlying
stratigraphy to influences of different processes is calculated through a set of equations. Depending on the equations adopted, 2DV cross-shore profile models can be
further classified into two different types: process based
and behavior oriented. Process-based models apply a set
of differential equations to describe the wave transformation, sediment transport, and subsequent bed elevation
change on the profile. Impacts of storm surge, eustatic
sea-level change, and tectonic movement are implemented
in the equations through a parameterization of boundary
conditions (i.e., incoming wave properties and water
level). Examples of process-based 2DV models can be
found in Masetti et al. (2008), Rosati et al. (2010), and
Zhang et al. (2013). On the contrary, behavior-oriented
models (e.g., Roy et al., 1994; Cowell et al., 1995; Storms
et al., 2002; Stolper et al., 2005; Moore et al., 2010)
describe the profile change by a set of empirical functions
of changes of sediment supply, sea level, and shoreface
geometry, without simulating the detailed processes
involved in sediment transport.
The validity of 2DV cross-shore profile models is based
on three pre-assumptions: (1) a zero net sediment exchange
at the boundary (thus sediment is conserved in the profile);
(2) evolution of the shoreface part of the profile evolves
toward a predefined shape (the so-called equilibrium),
which is determined by the grain size of the shoreface sediment and the mean wave climate (e.g., Bruun, 1962; Dean,
1991; Dean, 1997); and (3) alongshore uniformity of offshore wave parameters and nearshore isobaths along the
coastline (thus the gradient of longshore sediment transport
rate is zero and does not affect the profile change).
2DV cross-shore profile models have proven to be
more useful in providing detailed insights into the fundamental driving mechanisms of barrier island development
than conceptual models. However, one should always
keep in mind the limitations of validity which may hinder
application of a 2DV model to a real case. Another factor
affecting the reliability of a 2DV model is an exclusion of
inlet effects, which are most critical in controlling barrier
island morphodynamics according to Leatherman
(1979). An extension of an individual profile to an area
might overcome these limitations; however, this requires
much greater effort in bridging the different scales that
are involved in barrier island morphodynamics. Development of such models is still at an early stage. An example
of such models is presented by Zhang et al. (2012, 2014).
A hybrid and parallel coupling of process-based and
behavior-oriented modules provides a way to resolve the
relevant processes at their corresponding scales with an
affordable computational expense. In the model, wave
processes (propagation, transformation, refraction, and
breaking), currents, and subaqueous suspended sediment
transport are solved in process-based modules, while subaerial aeolian transport, bed-load transport, and land-sea
transition processes (e.g., cliff erosion) are simulated
either in behavior-oriented manners or by cellular automata approach. The model was applied to investigate the
morphogenesis and evolution of a Holocene barrier island
50
BARRIER ISLAND