Geomorphology 191 (2013) 6474

Contents lists available at SciVerse ScienceDirect

Geomorphology
journal homepage: www.elsevier.com/locate/geomorph

A hydrologic and geomorphic model of estuary breaching and closure

Andrew Rich , Edward A. Keller
Department of Earth Science, University of California, 1006 Webb Hall, Santa Barbara, CA 93106, USA

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

Article history: To better understand how the hydrology of bar-built estuaries affects breaching and closing patterns, a model
Received 27 September 2012 is developed that incorporates an estuary hydrologic budget with a geomorphic model of the inlet system.
Received in revised form 25 January 2013 Erosion of the inlet is caused by inlet ow, whereas the only morphologic effect of waves is the deposition
Accepted 10 March 2013
of sand into the inlet. When calibrated, the model is able to reproduce the initial seasonal breaching, seasonal
Available online 16 March 2013
closure, intermittent closures and breaches, and the low-streamow (closed state) estuary hydrology of the
Keywords:
Carmel Lagoon, located in Central California. Model performance was tested against three separate years of
Estuary hydrology water-level observations. When open during these years, the inlet was visually observed to drain directly
Inlet breaching and closure across the beach berm, in accordance with model assumptions. The calibrated model predicts the observed
Model 48-h estuary stage amplitude with root mean square errors of 0.45 m, 0.39 m and 0.42 m for the three
Coastal processes separate years. For the calibrated model, the probability that the estuary inlet is closed decreases
exponentially with increasing inow (streamow plus wave overtopping), decreasing 10-fold in probability
as mean daily inow increases from 0.2 to 1.0 m 3/s. Seasonal patterns of inlet state reect the seasonal pat-
tern of streamow, though wave overtopping may become the main hydrologic ux during low streamow
conditions, infrequently causing short-lived breaches. In a series of sensitivity analyses it is seen that the
status of the inlet and storage of water are sensitive to factors that control the storage, transmission, and
inow of water. By varying individual components of the berm system and estuary storage, the amount of
the time the estuary is open may increase by 57%, or decrease by 44%, compared to the amount of time the
estuary is open during calibrated model conditions for the 18.2-year model period. The individual compo-
nents tested are: berm height, width, length, and hydraulic conductivity; estuary hypsometry (storage to
stage relationship); two factors that control wave-swash sedimentation of the inlet; and sea level rise. The
elevation of the berm determines the volume of water that must enter the estuary in order to breach, and
it modulates the wave-overtopping ux and frequency. By increasing estuary storage capacity, the estuary
will breach less frequently (27% change in time open for modeled excavation scenario) and store water
up to 3 months later into the summer. Altering beach aquifer hydraulic conductivity affects inlet state, and
patterns of breaching and water storage. As a result of sea-level rise of 1.67 m by 2100, and a beach berm
that remains in its current location and accretes vertically, the amount of time the estuary remains open
may decrease by 44%. Such a change is an end-member of likely scenarios given that the berm will translate
landwards. Model results indicate that the amount of time the estuary is open is more sensitive to changes in
wave run-up than the amount of sand deposited in the inlet per each overtopping wave.
2013 Elsevier B.V. All rights reserved.

1. Introduction and therefore engineering practices such as the impoundment of

sediment by dams (Willis and Griggs, 2003), beach groins and jetties.
Bar-built, coastal lagoons in the 21st century will experience a As shown here, the existence and characteristics of beach barriers fun-
number of stressors that will affect their function. These include damentally controls estuary dynamics. Predicting how these stresses
0.421.67 m of sea-level rise (Dalrymple et al., 2012), changing pre- may impact estuaries is important because estuaries are considered
cipitation and temperature patterns (Seager and Vecchi, 2010), and to be the most valuable biome on earth per-area (Costanza et al.,
a growing human population that is largely concentrated near the 1997). This paper seeks to determine: 1) if the breaching and closing
coast. As population growth continues, so too will the anthropogenic processes can be modeled with a hydrologic and geomorphic model,
impact on streamow (Beighley et al., 2003), which is a strong driver 2) how estuarine variability, such as differences in beach-berm heights
of the function of bar-built estuaries. Similarly, the growth and decay or estuary storage, affect estuary function, and 3) how sea-level rise will
of beach barriers is linked to sediment availability (Carter et al., 1989) affect estuary function.
Bar-built coastal lagoons exist in microtidal (Cooper, 2001) and
Corresponding author. Tel.: +1 805 893 4207. mesotidal environments, are wave-exposed, and often occur where
E-mail address: [email protected] (A. Rich). streamow is highly seasonal, such as in California (Elwany et al.,

0169-555X/$ see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.geomorph.2013.03.003
 A. Rich, E.A. Keller / Geomorphology 191 (2013) 6474 65

1998), Australia (Ranasinghe and Pattiaratchi, 2003), and South the summer, waves approach from similar directions, but with smaller
Africa (Cooper, 2001). Estuaries are fronted by high supratidal Hs and shorter wavelengths. Waves coming from the Northwest arrive
beach-berms when there is ample coarse littoral sediment (Kjerfve to the Carmel Beach unrefracted and unimpeded (Storlazzi and Field,
and Magill, 1989; Cooper, 2001). For estuaries fronted by a supratidal 2000). For three scenarios, ranging 27 in deepwater wave direction
beach berm, wave energy is the major control on berm-height and each of different wavelength, refraction and shoaling modeling
(Takeda and Sunamura, 1982), though aeolian processes may con- of waves traveling from deepwater to 15 m depth reveals that waves
tribute via dune formation in the backbeach. Lower beach-berms undergo less than a 10% change in wave height and wavelength off-
occur when nearshore conditions are dissipative (Cooper, 2001). shore of the Carmel Lagoon (Laudier et al., 2011).
There is a considerable amount of research concerning the The Carmel River drains 660 km 2 of mountainous topography, and
processes that maintain an open inlet for larger, more tidally domi- the ow is highly variable (Fig. 1). As measured 5 km upstream of the
nated systems (Escofer, 1940; Jarrett, 1976; van de Kreeke, 1985; Carmel Lagoon, more than 90% of yearly streamow occurs between
Ranasinghe and Pattiaratchi, 2003). Relatively little work has investi- January and May, and yearly mean streamow varies from zero ow
gated the controls on estuary breaching and closing, especially of to 14.5 m 3/s. Such variability is often related to anomalies in tropical
smaller, bar-built estuarine systems, and such research has been pre- sea-surface temperatures (Cayan et al., 1999), though the prevalence
dominantly empirical. For a Southern California lagoon, Elwany et al. of low ows in the Carmel River are also the result of groundwater
(1998) showed that streamow is the major control on inlet state, pumping of the alluvial aquifer (Kondolf et al., 1987), and until 2003
and that tidal prism and wave processes play a smaller role during by abstractions of water at two upstream, nearly sediment-lled res-
closure. By analyzing a record of estuary closures of unprecedented ervoirs (MPWMD, 2008). To protect nearby properties from ooding,
detail and length, Behrens (2012) showed that the length of time the the Carmel Lagoon is sometimes articially breached preceding the
estuary remains closed before breaching is a function of streamow, rst major streamow of the year, at which time it would imminently
estuary storage at breaching levels and barrier seepage, whereas breach otherwise (James, 2005). Once breached, the Carmel Lagoon
inlet closure probability is best predicted using a ratio of inlet dis- inlet may migrate up to 500 m north, 200 m to the south, or, about
charge to sediment transport by waves. The understanding of inlet 50% of the time, it will remain in its initial breach location, directly
morphodynamics was improved by Baldock et al. (2008) who was across the beach berm (James, 2005). The estuary was excavated and
able to accurately predict channel inlet elevation change by sand widened in 1997 and 2004 in order to increase habitat area (James,
deposition using predictions of shore-normal, wave-swash run-up 2005), resulting in a 24% increase in estuary storage at typical breaching
heights. Other authors emphasize the role of wave overtopping to in- levels (James, 2005; Hope, 2007).
duce breaching (Hart, 2007) or as a counterbalance to evaporative The Monterey Peninsula Water Management District (MPWMD)
losses (Rustomji, 2007) while Kirk (1991) emphasizes the impor- has operated a water-level recorder since 1993 in the Carmel Lagoon;
tance of barrier material in transmitting oods. Together these and the USGS operates a streamow gage 5 km upstream of the estuary;
other studies demonstrate that estuary function is a result of many two wave buoys measure waves within 50 km; and tides are mea-
interacting processes, and that isolating the effect of a single property sured 9 km north of the estuary.
on function is difcult. Therefore, the approach taken in this work is to
develop a model of lagoon breaching and closing, and then to investi- 3. Methods
gate how individual properties of the estuary inuence its function.
The model addresses one of the processes that breach bar-built The breaching and closing model is based upon a hydrologic
coastal lagoons, namely uvial erosion of the beach barrier by channel- mass-balance approach for estuary storage, with hydrologic uxes
ized ow through the estuary inlet. The two processes not addressed from streamow, wave-overtopping, inlet discharge, evaporation,
are wave erosion and seepage-induced transport by exltrating and groundwater ow through the barrier. The model consists of
groundwater on the beachface. Seepage forces increase directly with the following components: 1) a water balance for the estuary basin
groundwater hydraulic gradient (Howard and McLane, 1988), and and 2) a channel inlet component, for which channel geometry, uvial
therefore inversely with beach width (Kraus et al., 2008). Wave erosion erosion and wave deposition are calculated.
can cause breaching in two ways: incision of the beach berm by
overtopping waves or by landward erosion of the beachface as a result 3.1. Estuary mass balance
of backswash. The erosive ability of overtopping waves is a function
of the height of overwash (Donnelly et al., 2006), and the geometry The volume of water in the estuary through time is a function of
and slope of the back-berm (Pierce, 1970); minor overwash builds the the difference in the uxes of water owing into or out of the estuary.
berm, whereas steep, narrow berms promote erosion. The mass balance is written as:
This paper is organized as follows: a description of the study site;


development of the hydrologic and geomorphic model; and a results
S qriver qovertop qinlet qgroundwater qevaporation t 1
section highlighting the calibrated model results, followed by a sensi-
tivity analysis of the model, and a discussion of drivers of estuary
change. where S is the change in storage, t is the timestep, qriver is
streamow, qovertop is the ux from wave overtopping of the beach
2. Study site berm, qinlet is ow through the berm inlet, qgroundwater is groundwater
ow between the estuary and ocean through the beach barrier, and
The Carmel Lagoon is fronted by a steep, reective beach composed qevaporation is evaporation of water from the estuary surface. River
of coarse sand derived from the Carmel Watershed (Storlazzi and ow and evaporation (DWR, 1974) are model boundary conditions,
Field, 2000). The beach sits in a coastal embayment, with a southern whereas the other uxes are dependent upon estuary, tidal or wave
longshore sediment transport direction (Howell, 1972) and 100 m conditions.
water depths within half of a kilometer. Nearby beaches have been
observed to accrete and erode up to 60 m in a season (Bascom, 3.2. Channel inlet discharge
1951), and there is potentially a long-term net erosion of the beach
(Storlazzi and Field, 2000). From September to May, deepwater Hs Channel inlet ow only occurs when the estuary or tidal level is
generally range from 1.3 to 3.7 m, sometimes exceeding 9 m, with higher than the inlet, and its velocity is determined by Manning's
directions of 270310 and wavelengths ranging 1017 s. During equation. The depth, direction, and slope of inlet ow depend upon
66 A. Rich, E.A. Keller / Geomorphology 191 (2013) 6474

Fig. 1. Site Map. a. LiDAR DEM (csc.noaa.gov) of Carmel Lagoon showing Carmel River, beach-berm elevation, and location of waterlevel gage within the estuary. Wave rose shows
incident wave directional distribution. b. location map of Carmel Lagoon.

the tide level, estuary stage, and the height of the inlet (Fig. 2). The A h
width of the channel is calculated using hydraulic geometry relation- idt
ships, relating ow magnitude to channel width (O'Brien, 1931; chw
B ea La
Hughes, 2002; Behrens et al., 2009), and is calculated as: go
on
l
W
0:002 0:153
Q d50 S 2 ne
an
Ch
et
where and are coefcients, Q is ow discharge, d50 is median grain Inl
size, and S is the slope of the channel (Lee and Julien, 2006). The Be
model assumes that the inlet channel ows directly across the r m Beach
le
beach berm (i.e. no migration of channel), and that the beach width ng Berm
th
remains constant. The channel inlet elevation cannot exceed the (B
L)
elevation of the beach berm.
beachface slope,

3.3. Inlet erosion and sedimentation lagoon

B
inlet
Changes to the channel inlet elevation are calculated by treating Fixed (MSL)
depth slope (s)
the inlet as an alluvial channel that is subject to deposition of sand
Rc Rinlet
by wave-swash deposition and erosion by uvial transport: R (wave run-up)

z qswash qfluvial t 3
stillwater level
where qswash is wave-swash deposition, and quvial is uvial erosion.
Fig. 2. Beach berm and denitions of variables. A. Beach width, berm height and length
The change in channel inlet height occurs at the estuary side of the
are constant for each simulation. B. Cross section through the inlet channel showing
inlet, whereas the beachside of the inlet remains xed at mean tide wave run-up (R), freeboard height of inlet channel (Rinlet), freeboard height of the
(Fig. 2). The vertical erosion of sand by inlet discharge is calculated beach-berm (RC; for wave overtopping calculation), stillwater level, elevation of the
by adapting the bedload transport equation of Meyer-Peter and inlet (inlet), and elevation of the lagoon waterlevel (lagoon). Changes in inlet elevation
Mller (1948) as: are a result of Z (see text). Inlet water depth and slope are dependent upon lagoon,
inlet, beach width, and the elevation of the tide when it is above mean sea level
3=2
(MSL). The left edge of the berm remains xed through all simulations at MSL. inlet
qfluvial crit 4 cannot erode lower than MSL.
 A. Rich, E.A. Keller / Geomorphology 191 (2013) 6474 67

where is a coefcient, crit is the shear stress below which transport In this work the effect of refraction and shoaling in the near shore
does not occur for the bedload material, and is shear stress caused are not incorporated. Rather, wave data observed at NOAA buoy
by the water owing through the inlet. crit is predicted using the for- 46042 located 57 km NW of the lagoon was transformed linearly to
mulation of Miller et al. (1977). Scripps buoy 46239 located 27 km to the SE of the lagoon (Fig. 1).
Deposition of sand onto the channel inlet by overtopping wave- This was achieved through a linear regression of 2.5 years of contem-
swash is calculated using the methods of Baldock et al. (2008). poraneous Hs and Tp data between both buoys.
While longshore transport processes are clearly important in the
sedimentation of inlets, cross-shore, swash-driven sedimentation is 3.5. Estuary storage
likely more important for the closure of supratidal inlet systems,
especially on embayed coasts where longshore currents are weak Converting estuary storage to water-level, and vice-versa, is
(Ranasinghe and Pattiaratchi, 1999). The probability of a wave performed using a hypsometric relationship of water storage to
overtopping the channel inlet is a function of the wave run-up (R) of water-level. Two hypsometric curves (James, 2005; Hope, 2007)
breaking wave bores, and the height difference between the stillwater were used for the model simulations to account for changes to the es-
level and the channel inlet (Hunt, 1959) (Rinlet; Fig. 2), where: tuary bathymetry due to a restoration project that excavated a portion
p of the estuary. Both hypsometric curves are extended beyond their
R C tan Hs  L0 : 4 original coverage using LiDAR observations from 2010 (csc.noaa.gov).

C is a calibration coefcient, Hs is signicant offshore wave height, 3.6. Goundwater ow

tan is beachface slope, and L0 is offshore wavelength. Assuming
a Rayleigh distribution of wave run-up, the probability of a wave Groundwater ow between the estuary and the ocean is calculated
overtopping the beach-berm is: analytically by assuming Dupuit conditions for groundwater ow,
h i
2
PR > Rinlet exp Rinlet =R : 5 Initial Conditions
Input Variables
Estuary storage (S0) Hypsometric curve
Inlet elevation (inlet, 0) Max berm height
The probability of waves overtopping the freeboard height of Beach width
the inlet channel (Rinlet) is calculated at each time step because the Berm length
inlet elevation is dynamic. To calculate the ux of deposited sand Convert estuary storage (S) to elevation Hydraulic conductivity
from wave overwash for a model timestep, the following method is (lagoon); then subtract Hevap Beachface slope
applied: Inlet roughness
OUTPUT

qswash htsand  N  P 6 lagoon > inlet Calculate channel depth,

or yes channel width and
where htsand is the height of sand deposited per wave (m) that tide >inlet channel velocity
overtops the berm inlet and N is the number of waves that arrive at
the beachface per timestep as determined from the wave period
Calculate channel inlet
(Baldock et al., 2008).
no discharge (Qinlet)

3.4. Wave overtopping and wave transformations

Calculate groundwater Calculate groundwater
The hydrologic ux from wave overtopping is calculated using the discharge (Qgw) discharge (Qgw)
wave overtopping model of Van der Meer and Janssen (1995), which
was calibrated specically for the Carmel Lagoon (Laudier et al.,
2011). The form of the wave overtopping equation is dependent Calculate change in estuary Calculate change in estuary
upon the Irribaren number (): storage (S) storage (S)
S = (Qstr + Qgw + Qover)t S =(Qstr + Qinlet + Qgw + Qover)

p1 q
Q A expBRc = r Hs  tan gH3S BL 2 7a
qfluvial=0 Calculate fluvial erosion
q from inlet discharge
Q C expDRC =r HS  gH3S BL >2 7b (qfluvial)
p
where tan= HS =L0 , RC is the freeboard height of the beach-
berm (Fig. 2B), Lo is the offshore wavelength, A, B, C and D are Si+1 = Si + S
calibration coefcients, r is a beach condition reduction factor, HS
is the signicant wave height, and BL is the length of the berm Calculate swash
over which overtopping occurs (Fig. 2A; Van der Meer and Janssen, deposition into channel
inlet (qswash)
1995; Laudier et al., 2011). In the original formulation, signicant
waveheight and peak wavelength at the toe of the beachface were
used (Van der Meer and Janssen, 1995), whereas Laudier et al. Calculate elevation change
of channel inlet
(2011) used refracted wave conditions at 15 m depth. Laudier et al.
inlet=(qswash-qfluvial)t
(2011) implemented the overtopping equation in a quasi-2D manner
by calculating the ux for 5 m intervals along dGPS beach surveys. For
this work, the ux is calculated in a 1D manner where the beachface inlet, i+1= inlet, i + inlet
slope and berm height are constant along the length of the berm OUTPUT
and for the duration of each model run. An ad-hoc adjustment was
applied to the wave overtopping whereby it is limited to a total
peak ux of 5 m 3/s. Fig. 3. Flow diagram of a model timestep.
68 A. Rich, E.A. Keller / Geomorphology 191 (2013) 6474

Table 1 and closing period, the following parameters were calibrated: htsand,
Data sources for model boundary conditions. , C, and . The model was calibrated by minimizing the root
Model Input Source Station Frequency and mean square error between the observed and modeled 48-h stage
units amplitude on the basis that it most closely reects the dynamics of
Tides NOAA/NOS/CO-OPS Monterey, Ca. 9413450 60-min, m NAVD88 the lagoon, which is the fundamental purpose of this study (Janssen
Streamow USGS Carmel River near 15-min, m3/s and Heuberger, 1995). The hydraulic conductivity of the beach
Carmel, 11143250 aquifer and the magnitude of the groundwater inow are adjusted
Waves National Data Buoy Monterey 46042 and 60-min, Hs and Tp
to maximize model t during the low-ow, closed period.
Center and California Point Sur 46239
Data Information
Project 4. Results
Evaporation DWR, 1974 Central Coast Coastal Inches per month
Valleys and Plains 4.1. Model conrmation

Breaching events are identied in the observed records when

the water-level falls precipitously, and seasonal closure events are
a vertical beachface, and vertical backside of the beach berm. A identied when the water-levels cease to oscillate. As seen in Figs. 4
vertical beachface eliminates the ltering effect of a sloping beachface (WY1996) and 5 (WY2002), the breaching model is capable of
(Turner, 1993), and a vertical back-beach produces constant ground- predicting the initial breaching, seasonal closure, intermittent clo-
water ow lengths in contrast to varied ow lengths that arise with sures and breaches, and the low-streamow estuary hydrology.
a sloped back-beach. Tidal and lagoon elevations are the boundary Fig. 6 shows only the portion of water year 2004 when the estuary
conditions for groundwater ow. When streamow is below a thresh- was continuously opening and closing in order to more clearly high-
old, a groundwater inow component from the catchment is modeled light model performance during such conditions. For water years
to contribute to estuary storage. 2002 and 2004, the inlet channel was observed to remain in roughly
the same location (Fig. 7), draining directly across the ~ 60 m berm
3.7. Model setup and down the beachface (James, 2005), in agreement with model as-
sumptions. In these years, during the open period, the 48-h root mean
The routine for each model timestep is portrayed in Fig. 3, along square error is 0.50 and 0.64 m, respectively (Table 2). In other years,
with the input variables necessary to run the model. Table 1 lists the inlet meanders to the north or south, increasing the inlet length
the model boundary conditions, or the data necessary to derive up to 10 the direct conguration, causing signicant differences
the boundary conditions and their sources. All model scenarios between the observed and modeled water level. This is illustrated
were performed at 15-min timesteps and all model inputs were in Fig. 8 where the inlet conguration was observed in its direct
interpolated to this interval. conguration from February 1st to April 6th, and, following a
10 m 3/s streamow storm, the inlet terminus was observed to
3.8. Model calibration meander 500 m from its initial position (James, 2005), as has been
observed in the Russian River Lagoon (Behrens et al., 2009). Prior
The model is calibrated during water year 2003 when multiple to the increase in inlet length (Fig. 8), the model predicts rapid
observations of the inlet indicate that it was immobile and drained uctuations in estuary stage, similar to observed data, with an RMSE
directly across the beach berm (James, 2005). During the opening of 0.45 m. When the inlet length increases following the storm

20
Input (m3/s)

streamflow
10
wave overtopping

0
10/95 11/95 12/95 01/96 02/96 03/96 04/96 05/96 06/96 07/96 08/96 09/96 10/96
10
Hs (m)

5
0
10/95 11/95 12/95 01/96 02/96 03/96 04/96 05/96 06/96 07/96 08/96 09/96 10/96
4.5

4 berm inlet
stage (m, NAVD88)

3.5 model

observed
3

2.5

1.5

10/95 11/95 12/95 01/96 02/96 03/96 04/96 05/96 06/96 07/96 08/96 09/96 10/96

Fig. 4. Model validation for water year 1996. Top box is observed streamow (USGS gage 11143250) and predicted wave overtopping ux, and middle box is transformed signif-
icant wave height (see text for explanation). Bottom box shows model results and observed water level at Carmel Lagoon. Water level data is from Monterey Peninsula Water
Management District (James, 2005). Berm inlet height is a model output.
 A. Rich, E.A. Keller / Geomorphology 191 (2013) 6474 69

Hs (m) Input (m3/s) 20

missing data streamflow
10 wave overtopping

0
10/01 11/01 12/01 01/02 02/02 03/02 04/02 05/02 06/02 07/02 08/02 09/02 10/02
10
5
0
10/01 11/01 12/01 01/02 02/02 03/02 04/02 05/02 06/02 07/02 08/02 09/02 10/02
4.5

4
berm inlet
stage (m, NAVD88)

3.5 model

3 observed

2.5

1.5

10/01 11/01 12/01 01/02 02/02 03/02 04/02 05/02 06/02 07/02 08/02 09/02 10/02

Fig. 5. Model validation for water year 2002. See Fig. 4 for description.

event, the model continues to predict >1 m uctuations in estuary of time that the estuary is open is controlled principally by the magni-
stage, whereas the observed data indicates a muted amplitude oscil- tude of inow, which consists of streamow and overtopping. The
lation due to the reduced hydraulic efciency of the channel. With inlet is considered open when the estuary waterlevel is higher than
the increased channel length, the model error nearly doubles to the inlet. As mean daily inow increases from 0 to 2 m 3/s, the proba-
0.87 m. A similar event occurs in April 1996 (Fig. 4), which explains bility that the lagoon is closed decreases exponentially (Fig. 9); 98.5%
the large deviation between the model and observed data for that of open days occur when estuary inow is greater than 0.5 m 3/s and
period. Despite the poor model performance for the elongated inlet only 12.5% of closed days occur above 0.5 m 3/s. The segregation
conditions, the model is capable of predicting the seasonal closure of inlet state by inow highlights the importance of inow on
date within 10.5 days. inlet state, though the small overlap between estuary state and inow
indicates that factors other than inow inuence estuary state, albeit
4.2. The effect of inow on inlet state to a small degree. Demonstrating the relative importance of wave
overtopping versus streamow is achieved by running the model
Using the calibrated conditions, the model was run for an 18.2-year with zero wave-overtopping and normal streamow, and subse-
period, from 1993 to 2011. Under calibrated conditions, the amount quently with normal wave overtopping and zero streamow. With
Input (m3/s)

15 streamflow
10 wave overtopping
5
0
01/04 02/04 03/04 04/04 05/04
10
Hs (m)

5
0
01/04 02/04 03/04 04/04
4.5
berm height
4
model
stage (m, NAVD88)

3.5
observed

2.5

1.5
01/04 02/04 03/04 04/04 05/04

Fig. 6. Model validation during open period of water year 2004. See Fig. 4 for description.
70 A. Rich, E.A. Keller / Geomorphology 191 (2013) 6474

Hs (m) Input (m3/s)

20
streamflow
10 wave overtopping
0
04/00 05/00 06/00 07/00
10

0
04/00 05/00 06/00 07/00

~60m channel ~500m channel

stage (m, NAVD88)

RMSE =0.45m RMSE =0.87m
3.5

2.5

2
model
1.5 observed

04/00 05/00 06/00 07/00

Fig. 8. Effect of inlet length on model performance. Model inlet length is 60 m for
entire simulation. Channel inlet observations denoted as stars (James, 2005).

open 40100% of the time though for March the estuary is open
100% for 14 of the 18 years. By fall to early winter, streamow input
volumes often decrease to zero (15 of 18 years), causing the estuary
to convert to a coastal lake with water levels often > 1 m below
the berm. During this period, wave overtopping becomes important
in regards to the hydrology of the estuary, providing, along with
groundwater seepage through the berm and estuary periphery, the
main inux of water into the estuary. In some years (5 of 18),
streamow input remains zero in January while wave overtopping
reaches its peak, allowing breaches to occur from overtopping alone.
Such breaches are short-lived because there is not sustained inlet
ow to erode the inlet and the concurrent high wave energy quickly
rebuilds the inlet.

500
Fig. 7. Observations of inlet locations for water years 1996, 2002 and 2004.
Figures modied by permission from James (2005). 400
# of days open

300

zero wave-overtopping the inlet remains open for 38.9% of the time, 200
and with zero streamow it remains open for 0.1%. Under current
streamow and predicted wave overtopping, it is modeled to remain 100
open for 39.5% of the time. A similar relationship is observed on a
monthly basis; in January through April, volumetrically the greatest 0
0 1 2 3
wave overtopping and streamow months, the estuary is generally
500

Table 2 400
# of days closed

Model performance statistics. Root mean square error (RMSE) of the observed and pre-
dicted 48-h estuary stage amplitude. The rst statistic corresponds to when the estuary 300
is actively opening and closing, and the second statistic is for the entire water year n = 2119
(October to September). 200 3
n = 258 (when input = 0m /s)
Water year Period of statistic RMSE Period of statistic RMSE
(m) (m) 100

Calibration
0
2003 13/Dec/200217/Jul/2003 0.44 Entire water year 0.37 0 1 2 3
mean daily inflow
Validation
(streamflow + wave overtopping,m3/s)
1996 15/Dec/199505/Jun/1996 0.65 Entire water year 0.45
2002 01/Dec/200128/May/2002 0.50 Entire water year 0.39
Fig. 9. Hydrologic inow and days open. Histograms of model results of the number of
2004 29/Dec/200330/Apr/2004 0.64 Entire water year 0.42
days open and closed for streamow plus wave overtopping.
 A. Rich, E.A. Keller / Geomorphology 191 (2013) 6474 71

4.3. Model sensitivity

To determine the effect of individual model components on estu- 6

ary function, the calibrated model is re-run while only changing
one variable at a time. The model period is 18.2 years long and encap-
sulates long-term uctuations in the El Nio Southern Oscillations 5

elevation (m, NAVD88)

thereby by eliminating the bias such uctuations may introduce.
The wave buoy ceased to function during portions of the intense
4
1998 El Nio winter, and this non-consecutive 360 day period was
removed from the analysis. The variation in the model components
was chosen to reect physically realistic variability in such compo-
3
nents. The calibrated values, model sensitivity values, and model re-
sults are shown in Table 3, and the full hypsometry and excavated
hypsometry curves are shown in Fig. 10. 2
The hypsometry sensitivity results demonstrate that the accom- MHHW
modation space within the estuary plays a fundamental role in its
Carmel Hypsometry
functioning. For the full hypsometry conditions, the frequency of 1
MSL excavated hypsometry
estuary waterlevel uctuations and breaching events increases
full hypsometry
(Fig. 11). In the excavated hypsometry conditions, the estuary retains
stored water three months later than the full hypsometry conditions, 0 MLLW
and for a 3-week period, it breaches nine-fold less than the full 0 500,000 1,000,000 1,500,000
hypsometry. Excavating the current estuary would reduce the time volume (m3)
open by 27% and lling it with sediment would increase time open
by 18%. Fig. 10. Observed and Modeled Carmel Lagoon Hypsometry. Carmel hypsometry
is from James (2005), and excavated hypsometry and full hypsometry were used
From three observations Carmel lagoon berm heights appear to
in the sensitivity analyses. The tidal data MHHW, MSL, and MLLW represent the
vary from 4 to 5 m NAVD88, though a dune formed in 2010 in the mean higher-high water, mean sea level, and mean lower-low water, respectively
back-beach with elevations of 5.5 m NAVD88. Berm-height for the (tidesandcurrents.noaa.gov).
calibrated model simulations was 4.2 m NAVD88. The effects of a de-
creased (increased) berm-height on estuary breaching are two-fold:
1) an increase (decrease) in wave overtopping frequency and magni- to breaching levels; at such timescales, the role of water transmission
tudes, and 2) a decrease (increase) in volume of water necessary to through the beach berm in estuary functioning is unimportant. There-
breach the berm. For 4 of the 18 years, the calibrated model permit- fore, fairly common daily streamow volumes are capable of breaching
ted large wave-overtopping events to initialize short-lived breaches even the high-berm scenarios. The impact of monotonically increasing
up to a month earlier than the higher berm. Once the estuary is beach-berm heights from near tidal levels to ~6 m above mean
breached, the high-berm scenario behaves similar to the calibrated higher-high water is an exponential-like decrease in percent time
conditions (4.2 m NAVD88). When the berm height is decreased open of the estuary inlet (Fig. 12). As shown in Table 3, during the
from 4.2 m NAVD 88 to 3.2 m NAVD 88, wave overtopping becomes 18.2-year model scenario the inlet of the high-berm scenario is open
a strong determinant of estuary functioning, causing short, frequent only 3% less than the calibrated conditions, whereas a low-berm height
breaches, even when streamow is still zero. At a 95% daily stream causes a 25% increase in the time the estuary is open.
ow exceedance (14.7 m 3/s), it takes 0.2, 0.5 and 1.3 days to ll the Other results from the long-term modeling are non-linear, as well.
estuary (with no outow and zero initial storage) when the berm A ten-fold increase in hydraulic conductivity causes a 44% change
height is 3.2, 4.2 and 5.5 m NAVD88, respectively. This daily in time open, but a ten-fold decreased hydraulic conductivity causes
streamow magnitude is exceeded 15 out of the 18 years, whereas only a 10% increase in time open (Table 3). An increased hydraulic
the 99% daily exceedance ow is reached in 12 of the 18 years. At conductivity causes breaches to occur only during high streamow
the 99% ow exceedance, it takes 12 h to ll the high-berm estuary events, whereas during the low-ow summer conditions, estuary

Table 3
Calibrated values of the breaching model, their range of values tested in the sensitivity analysis and their effect on the amount of days the estuary is open during the 18.5-year
modeling scenario. Model values used in the sensitivity analysis are either shown directly or are a factor of the Calibrated Value (e.g. 10). Four of the model parameters were
not tested in the sensitivity analysis.

Calibrated value Sensitivity analysis

Model value Number of days open % change from best t

High value Low value High Low High Low

htsand, m/wave 0.0017 2 0.5 2400 2654 3% 7%

, erosion 2.15E07
C, run-up adjustment 0.47 2 0.5 1754 3059 29% 23%
, hydraulic geometry 0.43
, hydraulic geometry 0.67
b, berm ht, m NAVD88 4.2 5.5 3.2 2408 3100 3% 25%
Beach width, m 60 240 10 3907 2102 57% 15%
Hydraulic conductivity, k (m/s) 0.0062 10 0.1 1388 2742 44% 10%
Berm length, m 160 400 10 2268 2700 9% 9%
Beach face slope, m/m 0.09
Hypsometry See Fig. 10 Excavated Full 1809 2928 27% 18%
hypsometry hypsometry
Groundwater inow, m3/s 0.08
Sea level rise (m) 0 1.67 0.42 1388 2136 44% 14%
72 A. Rich, E.A. Keller / Geomorphology 191 (2013) 6474

Input (m3/s) 20
wave overtopping
streamflow
10

0
11/03 12/03 01/04 02/04 03/04 04/04 05/04 06/04 07/04 08/04 09/04 10/04 11/04 12/04 01/05
stage (m, NAVD88)

4 current hypsometry
full hypsometry
3

11/03 12/03 01/04 02/04 03/04 04/04 05/04 06/04 07/04 08/04 09/04 10/04 11/04 12/04 01/05
stage (m, NAVD88)

4 current hypsometry
excavated hypsometry
3

11/03 12/03 01/04 02/04 03/04 04/04 05/04 06/04 07/04 08/04 09/04 10/04 11/04 12/04 01/05

Fig. 11. Estuary Hypsometry Sensitivity. Model results showing the effect of estuary hypsometry on breaching and closing dynamics. Bottom two boxes show model output from
the calibrated model (current hypsometry) and the altered estuary hypsometry scenarios. Note the increase in breaching frequency in the full hypsometry conditions, and the
increased water retention in the excavated hypsometry conditions.

storage is negligible and strongly inuenced by tidally modulated the storage, inow, and transmission of water. For the estuary system
groundwater ow (Fig. 13). Breaching occurs throughout the year, itself, realistic variations in each of the factors investigated here pro-
and water storage, and consequently residence times, are great duces more than a 7% change in the amount of time the estuary is
during the reduced hydraulic conductivity conditions. During high open. Understanding how these factors affect estuary function allows
streamow, the calibrated model and the reduced hydraulic conduc- for improved restoration and management of these systems for their
tivity scenario behave similarly. ecological value. For example, with the use of the model developed
Sea level is predicted to rise by between 0.42 and 1.67 m by 2100 here, resource managers were provided quantitative predictions of
(Dalrymple et al., 2012), and will likely cause the beach-berm to the effects of a proposed estuary excavation project on the breaching
translate landward. The effects of sea level rise on estuary function and closing regime of a coastal estuary in Santa Barbara, Ca. (Rich,
were evaluated by increasing tidal elevations and berm elevations 2012).
by the sea level rise predictions, but keeping the beach berm in its Existing at the nexus of the coastal environment and watershed
current location. By not translating the berm landwards, the predic- outlet, the function of these systems is affected by a number of
tions should be interpreted as the maximal effect that sea-level rise drivers. These drivers can be grouped as: 1) water and sediment
may have on the time open. The 0.42 m sea-level rise scenario causes discharge, 2) land use, 3) climate and sea level, 4) estuary geomor-
a 14% decrease in time open and the 1.67 m sea-level rise causes a phology, and 5) coastal and barrier beach change. For the calibrated
44% decrease. Carmel Lagoon model, when beach berm heights adequately reduce
the wave overtopping ux, streamow is the primary determinant
5. Discussion on inlet state and on estuary storage. Elwany et al. (1998) demon-
strated that the long-term pattern of inlet state correlates well
The model results indicate that the functioning of bar-built, coastal with streamow and Behrens (2012) showed that both closure
lagoons are strongly dependent upon a variety of factors that control and breaching timing are persistently inuenced by streamow. It
is therefore expected that processes affecting runoff ow-durations
such as groundwater pumping (Winter et al., 1998) and urbanization
100 (Ferguson and Suckling, 1990) will strongly impact estuary
function.
80 It is not clear if global warming will lead to a more El Nio-like cli-
mate (Collins, 2005), with its attendant increased yearly streamow
% time open

(Cayan et al., 1999) and increased wave heights along western North
60 America (Allan and Komar, 2006). If El Nios do become more fre-
quent, they will cause lagoons to remain open longer because estuary
40 closure is more sensitive to increases in streamow than increases in
wave height (Behrens, 2012). If the beach berm remains in its current
location, sea-level rise will cause the estuary to function more like a
20
coastal lake.
Other watershed processes will also impact estuary function,
0 primarily by controlling estuary geomorphology and hypsometry.
2 3 4 5 6 7 8 The size (volume, area, etc.) of coastal estuaries in active tectonic
berm height (m, NAVD88)
environments scale with watershed precipitation and stream gra-
Fig. 12. Sensitivity analysis of beach berm height on the percent of time that the inlet is
dient (Rich and Keller, 2012). The nature of this scaling is a func-
open. Each point represents the results of an 18.2-year model run for given beach berm tion of a variety of factors, with lithology, tectonics, and sediment
height. ux likely as key drivers. Sediment ux is inuenced by land use
 A. Rich, E.A. Keller / Geomorphology 191 (2013) 6474 73

20
Input (m3/s) streamflow
wave overtopping
10

0
11/03 12/03 01/04 02/04 03/04 04/04 05/04 06/04 07/04 08/04 09/04 10/04 11/04 12/04 01/05
stage (m, NAVD88)

4 calibrated hydr. cond.

10x calibrated hydr. cond.
3

11/03 12/03 01/04 02/04 03/04 04/04 05/04 06/04 07/04 08/04 09/04 10/04 11/04 12/04 01/05
stage (m, NAVD88)

0.1x calibrated hydr. cond.

3
calibrated hydr. cond.
2

11/03 12/03 01/04 02/04 03/04 04/04 05/04 06/04 07/04 08/04 09/04 10/04 11/04 12/04 01/05

Fig. 13. Sensitivity analysis of beach hydraulic conductivity. Bottom two boxes show model output from the calibrated model (calibrated hydr. cond.) and the altered beach hydraulic
conductivity scenarios.

changes and reservoir construction (Syvitski et al., 2005), and 6. Conclusions

signicant sediment aggradation has been observed in a Central
California lagoon, likely as a result of land use change (Revell et The model results show that the breaching and closing of the
al., 2010). bar-built, Mediterranean climate, Carmel Lagoon can be predicted
Operative estuarine and coastal processes are those that control using a hydrologic and geomorphic model. When calibrated, the
the form, composition, size and existence of the beach berm, and model is able to reproduce the initial seasonal breaching, seasonal
the functioning of the channel inlet. The construction of dams and closure, intermittent closures and breaches, and the low-streamow
beach groins reduces littoral sediment supply, causing beach estuary hydrology. For three years when the estuary inlet drains
narrowing (Willis and Griggs, 2003) and potentially reduction of directly across the beach berm, and during the portion of the
beach-berm heights. Dams also alter the caliber of sediment a river year when the inlet is actively opening and closing, the root mean
transports (Kondolf, 1997), potentially impacting the composition square errors of calibrated versus observed 48-h estuary stage
(e.g. hydraulic conductivity) of the beach barrier. Changes in grain amplitude are 0.65 m, 0.50 m and 0.64 m. For the entirety of the
size also impacts surfzone sediment transport direction (Dean, three years, model performance improves to 0.45 m, 0.39 m and
1973) and therefore processes affecting estuary closure (Ranasinghe 0.42 m. Model performance deteriorates when the channel inlet
and Pattiaratchi, 2003). migrates from the direct conguration to one ~ 10 longer. Under
calibrated conditions, the probability that the Carmel Lagoon inlet
5.1. Model uncertainty is closed is strongly linked to inow and seasonal patterns of inlet
state are a reection of streamow seasonality. The probability
The results indicate that the model works best when the channel of the inlet being closed decreases 10-fold as mean daily inow
inlet ows directly from the estuary to the beachface, in accordance increases from 0.2 to 1.0 m 3/s, whereas 98.5% of open days occur
with model assumptions. There does not exist a functional predictive when estuary inow is greater than 0.5 m 3/s. The estuary may
model of channel migration, and pursuing such a model is beyond the breach from overtopping alone, but such breaches are short-lived.
scope of this work. Nonetheless, the model is still able to accurately Wave overtopping does not play a large role in determining inlet
predict the seasonal closure timing within 10.5 days when this state under calibrated conditions.
assumption is invalid. The inlet conguration impacts the hydraulic In conjunction with beach processes, the hydrologic water balance
efciency of the inlet and erosive ability of the ow itself (van de controls the functioning of these systems. Therefore, factors that
Kreeke, 1985), and is therefore important in the closure process alter the storage, transmission, and inow control estuary function.
(Behrens, 2012). Changes to estuary hypsometry or beach-berm height increase stor-
The absence of a nearshore wave transformation model introduces age volumes; increasing berm heights diminishes the overtopping
uncertainty into the model and results. This uncertainty is small for ux, rendering streamow the important control on breaching,
Carmel Lagoon (Laudier et al., 2011), though if not, the heuristic whereas a reduced berm height allows wave overtopping to inuence
value of this work is still valid. For longer model runs, the uncertainty breaching. By excavating an estuary it will breach less frequently and
in other model parameters likely produces a greater amount of uncer- store water for longer periods. The transmission of inow through the
tainty than the lack of a wave model. The beachface slope, which at berm is controlled by the hydrologic properties of the berm itself, in-
Carmel has been observed to vary from 0.08 to 0.22 (Laudier et al., cluding the beach width and length, and its hydrologic transmissivity.
2011), introduces a large uncertainty into the wave run-up height, The effect of altered beach conditions is not always linear, however.
which varies linearly with the beachface slope. For a given offshore The tendency for the inlet to be closed increases with wave energy
wave conditions, wave overtopping is a nonlinear function of berm and wave run-up, and the availability of beach sand is relatively less
height, the reduction factor (r), and the beachface slope (Laudier important in the closure process. Lastly, if beach berms remain in
et al., 2011), all of which are not likely constant for > 1-month their current locations, but aggrade vertically under predicted sea
periods. level rise conditions, sea level rise will cause coastal lagoons to reduce
74 A. Rich, E.A. Keller / Geomorphology 191 (2013) 6474

the amount of time they are open by up to 44%. This prediction is an James, G., 2005. Surface Water Dynamics at the Carmel Lagoon Water Years 1991
Through 2005. Monterey Peninsula Water Management Agency, Monterey, Ca.
end-member and does not account for beach-berm migration. Janssen, P.H.M., Heuberger, P.S.C., 1995. Calibration of process-oriented models. Eco-
Finally, watershed and coastal processes that affect estuary func- logical Modelling 83, 5566.
tion can be grouped as follows: 1) water and sediment discharge, Jarrett, J.T., 1976. Tidal prisminlet area relationships. US Army Coastal Engr. Res. Cent,
p. 55. G.I.T.I. Report 3.
2) land use, 3) climate and sea level, 4) estuary geomorphology, and Kirk, R.M., 1991. River-beach interaction on mixed sand and gravel coasts: a geomor-
5) coastal and barrier beach change. Restoration and management phic model for water resource planning. Applied Geography 11 (4), 267287.
of these systems should account for the above factors, especially for Kjerfve, B., Magill, K.E., 1989. Geographic and hydrodynamic characteristics of shallow
coastal lagoons. Marine Geology 88 (34), 187199.
changes in streamow. Kondolf, G.M., 1997. PROFILE: hungry water: effects of dams and gravel mining on river
channels. Environmental Management 21 (4), 533551.
References Kondolf, G.M., Maloney, L.M., Williams, J.G., 1987. Effects of bank storage and well
pumping on base ow, Carmel River, Monterey County, California. Journal of
Allan, J.C., Komar, P.D., 2006. Climate controls on US west coast erosion processes. Journal Hydrology 91, 351369.
of Coastal Research 511529. Kraus, N., Munger, S., Patsch, K., 2008. Barrier beach breaching from the lagoon side,
Baldock, T.E., Weir, F., Hughes, M.G., 2008. Morphodynamic evolution of a coastal with reference to Northern California. Shore & Beach 76 (2), 3343.
lagoon entrance during swash overwash. Geomorphology 95 (34), 398411. Laudier, N.A., Thornton, E.B., MacMahan, J., 2011. Measured and modeled wave
Bascom, W.N., 1951. The relationship between sand size and beach face slope. Transac- overtopping on a natural beach. Coastal Engineering 58 (9), 815825.
tions of the American Geophysical Union 32, 866874. Lee, J.-S., Julien, P.Y., 2006. Downstream hydraulic geometry of alluvial channels.
Behrens, D.K., 2012. The Russian River Estuary: Inlet Morphology, Management, and Journal of Hydraulic Engineering 132 (12), 13471352.
Estuarine Scalar Field Response. Dissertation, Univ. of California, Davis, 340 pp. Meyer-Peter, E., Mller, R., 1948. Formulas for bed-load transport. Proceedings of the
Behrens, D.K., Bombardelli, F.A., Largier, J.L., Twohy, E., 2009. Characterization of time 2nd Meeting of the International Association for Hydraulic Structures Research,
and spatial scales of a migrating rivermouth. Geophysical Research Letters 36 (9), Delft, Netherlands.
L09402. Miller, M.C., McCave, I.N., Komar, P.D., 1977. Threshold of sediment motion under uni-
Beighley, R.E., Melack, J.M., Dunne, T., 2003. Impacts of California's climatic regimes directional currents. Sedimentology 24 (4), 507527.
and coastal land use change on streamow characteristics. Journal of the American MPWMD, 2008. California American Water Production by Source for Customers in its
Water Resources Association 39 (6), 14191433. Mains System; Water Years 1996Present. Monterey Peninsula Water Management
Carter, R.W.G., Forbes, D.L., Jennings, S.C., Orford, J.D., Shaw, J., Taylor, R.B., 1989. Barrier District, Monterey.
and lagoon coast evolution under differing relative sea-level regimes: examples O'Brien, M.P., 1931. Estuary tidal prisms related to entrance areas. Journal of Civil Engi-
from Ireland and Nova Scotia. Marine Geology 88, 221242. neering 738739.
Cayan, D.R., Redmond, K.T., Riddle, L.G., 1999. ENSO and hydrologic extremes in the Pierce, J.W., 1970. Tidal inlets and washover fans. Journal of Geology 78 (2), 230234.
Western United States. Journal of Climate 12 (9), 28812893. Ranasinghe, R., Pattiaratchi, C., 1999. The seasonal closure of tidal inlets: Wilson inlet
Collins, M., 2005. El Nio- or La Nia-like climate change? Climate Dynamics 24 (1), a case study. Coastal Engineering 37 (1), 3756.
89104. Ranasinghe, R., Pattiaratchi, C., 2003. The seasonal closure of tidal inlets: causes and
Cooper, J.A.G., 2001. Geomorphological variability among microtidal estuaries from the effects. Coastal Engineering Journal 45 (4), 601627.
wave-dominated South African coast. Geomorphology 40 (12), 99122. Revell, D., Williams, P., Rich, A., Bozkurt, S., Collison, A., Ginney, E., Kunz, D., Vandever, J.,
Costanza, R., d'Arge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Gamble, J., 2010. An Assessment of Potential Restoration Actions To Enhance
Naeem, S., O'Neill, R.V., Paruelo, J., Raskin, R.G., Sutton, P., van den Belt, M., 1997. The Ecological Functions Of The Lower Santa Ynez River Estuary. ESA-PWA, San
The value of the world's ecosystem services and natural capital. Nature 387 (6630), Francisco.
253260. Rich, A., 2012. Restoration of Devereux Slough: Effects on Estuary Breaching and Closing,
Dalrymple, R.A., Breaker, L.C., Brooks, B.A., Cayan, D.R., Griggs, G.B., Han, W., Horton, Cheadle Center for Biodiversity and Ecological Restoration. Univ. of California Santa
B.P., Hulbe, C.L., Mcwilliams, J.C., Mote, P.W., Pfeffer, W.T., Reed, D.J., Shum, C.K., Barbara, Santa Barbara, Ca.
Holman, R.A., Linn, A.M., Mcconnell, M., Gibbs, C.R., Ortego, J.R., 2012. Sea-level Rich, A., Keller, E., 2012. Watershed controls on the geomorphology of small coastal
rise for the coasts of California, Oregon, and Washington: past, present, and future. lagoons in an active tectonic environment. Estuaries and Coasts 35 (1), 183189.
National Research, Council.The National Academies Press, Washington DC. Rustomji, P., 2007. Flood and drought impacts on the opening regime of a wave-
Dean, R.G., 1973. Heuristic Models of Sand Transport in the Surf Zone, Australian dominated estuary. Marine and Freshwater Research 58 (12), 11081119.
Conference on Coastal Engineering (1st: 1973: Sydney, N.S.W.). Institution of Engi- Seager, R., Vecchi, G.A., 2010. Greenhouse warming and the 21st century hydroclimate
neers, Australia, Sydney, N.S.W. of southwestern North America. Proceedings of the National Academy of Sciences
Donnelly, C., Kraus, N., Larson, M., 2006. State of knowledge on measurement and 107 (50), 2127721282.
modeling of coastal overwash. Journal of Coastal Research 22 (4), 965991. Storlazzi, C.D., Field, M.E., 2000. Sediment distribution and transport along a rocky,
DWR, 1974. Vegetative Water Use in California. State of California, The Resources embayed coast: Monterey Peninsula and Carmel Bay, California. Marine Geology
Agency, Department of Water Resources, Sacramento. 170 (34), 289316.
Elwany, M.H.S., Flick, R.E., Aijaz, S., 1998. Opening and closure of a marginal Southern Syvitski, J.P.M., Vrsmarty, C.J., Kettner, A.J., Green, P., 2005. Impact of humans on the
California lagoon inlet. Estuaries 21 (2), 246254. ux of terrestrial sediment to the Global Coastal Ocean. Science 308 (5720),
Escofer, F.F., 1940. The stability of tidal inlets. Shore & Beach 8, 114115. 376380.
Ferguson, B.K., Suckling, P.W., 1990. Changing rainfallrunoff relationships in the Takeda, I., Sunamura, T., 1982. Formation and height of berms. Transactions, Japanese
urbanizing Peachtree Creek Watershed, Atlanta, Georgia. Journal of the American Geomorphological Union 3 (2), 145157.
Water Research Association 26 (2), 313322. Turner, I., 1993. Water table outcropping on macro-tidal beaches: a simulation model.
Hart, D.E., 2007. River-mouth lagoon dynamics on mixed sand and gravel barrier Marine Geology 115, 227238.
coasts, 9th International Coastal Symposium. Journal of Coastal Research, Gold van de Kreeke, J., 1985. Stability of tidal inlets, Pass Cavallo, Texas. Estuarine, Coastal
Coast, Australia, pp. 927931. and Shelf Science 21 (1), 3343.
Hope, A., 2007. Carmel river lagoon: hydrographic survey and stagevolume relationship. Van der Meer, J.W., Janssen, W., 1995. Wave run-up and wave overtopping at dikes.
Prepared for: Monterey Peninsula Water Management District. In: Kabayashi, D. (Ed.), Wave Forces on Inclined and Vertical Wall Structures.
Howard, A.D., McLane III, C.F., 1988. Erosion of cohesionless sediment by groundwater American Society of Civil Engineers, New York, pp. 127.
seepage. Water Resources Research 24 (10), 16591674. Willis, C.M., Griggs, G.B., 2003. Reductions in uvial sediment discharge by Coastal
Howell, B.F., 1972. Sand movement along Carmel River State Beach, Carmel, California Dams in California and implications for beach sustainability. Journal of Geology
(71 pp.). 111 (2), 167182.
Hughes, S.A., 2002. Equilibrium cross sectional area at tidal inlets. Journal of Coastal Winter, T.C., Harvey, J.W., Franke, O.L., Alley, W.M., 1998. Ground water and surface
Research 18 (1), 160174. water a single resource. United States Geological Survey Circular 1139.
Hunt, I.A., 1959. Design of seawalls and breakwaters. Proceedings of the American
Society of Civil Engineers 85, 123152.