FL47CH23-Horner-Devine ARI 21 November 2014 8:2

ANNUAL
REVIEWS Further Mixing and Transport in
Click here for quick links to
Annual Reviews content online,
including:
Coastal River Plumes
• Other articles in this volume
• Top cited articles Alexander R. Horner-Devine,1 Robert D. Hetland,2
• Top downloaded articles
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

• Our comprehensive search and Daniel G. MacDonald3

1
Department of Civil and Environmental Engineering, University of Washington, Seattle,
Access provided by [Link] on 04/04/23. For personal use only.

Washington 98195; email: arhd@[Link]

2
Department of Oceanography, Texas A&M University, College Station, Texas 77843
3
Department of Civil and Environmental Engineering, University of Massachusetts Dartmouth,
Dartmouth, Massachusetts 02747

Annu. Rev. Fluid Mech. 2015. 47:569–94 Keywords

First published online as a Review in Advance on coastal ocean circulation, stratified flows, turbulence, geophysical flows
October 1, 2014

The Annual Review of Fluid Mechanics is online at Abstract

fl[Link]
River plumes are generated by the flow of buoyant river water into the coastal
This article’s doi: ocean, where they significantly influence water properties and circulation.
10.1146/annurev-fluid-010313-141408
They comprise dynamically distinct regions spanning a large range of spa-
Copyright  c 2015 by Annual Reviews. tial and temporal scales, each contributing to the dilution and transport of
All rights reserved
freshwater as it is carried away from the source. River plume structure varies
greatly among different plume systems, depending on the forcing and ge-
ometry of each system. Individual systems may also exhibit markedly differ-
ent characteristics under varied forcing conditions. Research over the past
decade, including a series of major observational efforts, has significantly
improved our understanding of the dynamics and mixing processes in these
regions. Although these studies have clarified many individual processes, a
holistic description of the interaction and relative importance of different
mixing and transport processes in river plumes has not yet been realized.

569
 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

1. INTRODUCTION
Rivers carry more than one-third of land-based precipitation to the ocean (Trenberth et al. 2007),
focusing large freshwater fluxes through narrow outlets along the coast (Milliman & Farnsworth
2011). For example, the Mississippi River drains approximately 40% of the continental United
States to two distributaries, each of which is less than 1 km wide (Milliman & Meade 1983). The
impact of the terrigenous material carried by the river water into these ecologically sensitive
coastal waters depends strongly on the physical processes that transport and transform the
buoyant freshwater in the region around the river mouth as it merges with deeper, salty ocean
waters. In particular, the dilution rates and along-coast transport rates of the river-borne material
are determined by a suite of processes, including stratified-shear mixing, frontal processes,
geostrophic transport, and wind forcing. In the coastal ocean, these discharges form river plumes,
which are distinct regions where water properties and dynamics are significantly influenced by the
riverine freshwater. River plumes are also sometimes referred to as regions of freshwater influence
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

(ROFIs), although this term is more commonly reserved for systems with multiple freshwater
sources or shallow frictional shelves. The distinguishing dynamical feature of a river plume is the
Access provided by [Link] on 04/04/23. For personal use only.

horizontal advection of freshwater from the river mouth that defines the shape and character of the
plume.
River plumes are fundamentally multiscale flow structures, which occur in a range of sizes and
shapes. The primary independent parameters that determine the structure and scale of individual
plumes are freshwater discharge, tidal amplitude, coastline bathymetry/geometry, ambient ocean
currents and water properties, wind stress, and Earth’s rotation. Of these parameters, freshwater
discharge typically has the most variability within a system, often changing over an order of
magnitude within a single system, and among different systems, varying by over four orders of
magnitude. This large variability results in significant structural and dynamical differences among
systems. For example, the River Teign has a small discharge (5 m3 s−1 ) and forms a buoyant layer a
few meters thick seaward of the estuary mouth after each ebb tide (Pritchard & Huntley 2006); this
plume dissipates in a matter of hours and forms anew with each tide. The Merrimack River plume
is of moderate size (300 m3 s−1 ) and occasionally persists between tides, depending on the wind
(MacDonald et al. 2007). The Columbia River plume is large (10,000 m3 s−1 ), and it is strongly
affected by the tide near the estuary mouth but is persistent and governed by other processes far
beyond this tidal region (Horner-Devine et al. 2009, Hickey et al. 2010). The Mississippi River
plume is very large (30,000 m3 s−1 ) and is in a region with very small tides (Wright & Coleman
1971). It forms a huge, persistent plume that dominates the coastal circulation along most of the
Louisiana and Texas shelves (Cochrane & Kelly 1986, Murray 1998, Zhang & Hetland 2012,
Zhang et al. 2012).
The impact of a plume on shelf circulation and ecosystem health depends on the dilution
rate and transport processes within the plume. Dilution is controlled primarily by vertical mixing,
which increases the salinity of the plume and decreases the concentration of river-borne materials.
Transport in the plume is controlled primarily by horizontal advection, which is driven in large
part by plume buoyancy, and thus by mixing. Here, we focus specifically on the processes that
contribute to mixing (Section 3) and transport (Section 4) in river plumes. Most analyses and
observations consider processes active in specific regions of the plume in isolation. We begin,
therefore, by introducing the basic dynamical regions of the plume. The objective of this article is
not to provide a comprehensive review of previous research on river plumes (see O’Donnell 2010,
Chant 2011, Hetland & Hsu 2013) but rather to discuss and synthesize the processes relating to
mixing and transport to gain a more holistic picture of how the different processes acting at a wide
range of spatial and temporal scales interact.

570 Horner-Devine · Hetland · MacDonald

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

2. THE DYNAMICAL REGIONS OF RIVER PLUMES

The dominant dynamical balance differs significantly among each region of a river plume, mo-
tivating a description of the river plume structure in terms of four dynamical regions (Garvine Source region: the
1984, Hetland 2005, Horner-Devine et al. 2009). estuarine region,
First, in the source region, the buoyancy and momentum that initiate a river plume are de- inside the river mouth,
where the initial
termined by estuarine processes, which are responsible for the initial transformation of river
buoyancy and
discharge. Nash et al. (2009) show that the median salinity, thickness, and turbulent mixing rates momentum of the
of the near-field Columbia River plume are set by the competition between the stratifying influ- river plume are set
ence of river discharge and the mixing provided by tidal energy within the estuary, as expressed Liftoff: the location
by the estuarine Richardson number, of the bottom attached
Qf salt front, where the
Ri E = g f . (1) buoyant outflow loses
W E u 3tidal contact with the
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

The reduced gravity is defined in general as g  = g(ρ/ρo ), where g is the gravitational accelera- bottom
tion, and ρ is the density difference relative to background ocean water ρo . Here g f represents Fronts: sharp lateral
Access provided by [Link] on 04/04/23. For personal use only.

the density difference between fresh river water and salty ocean water, Q f is the river discharge, gradients in fluid
density that form
WE is the estuary width, and u tidal is the tidal velocity. More information on estuaries is presented
internally and, most
in recent reviews by MacCready & Geyer (2010) and Geyer & MacCready (2014). commonly, on the
In strongly forced systems, the estuarine discharge separates from the bottom near the river boundaries of the
mouth and forms the buoyant layer (Figure 1). The liftoff point is the location of the bottom- plume
attached salt front where the upper layer loses contact with the bottom and the upper-layer Froude Near-field region:
number Fr1 = 1. Here, Fr1 = u(g  h)−1/2 , and u and h are the upper-layer velocity and depth, the jet-like region of
respectively. The dynamics of the liftoff process are explained by inviscid two-layer hydraulic initial plume expansion
where the momentum
theory and are analogous to the case of strong barotropic forcing (Armi & Farmer 1986, Farmer
of the plume layer
& Armi 1986), in which the point of hydraulic control (Fr1 = 1) is forced seaward from the point dominates over its
of the smallest plume cross section and flow in the lower layer is arrested. Internal hydraulic theory buoyancy, resulting in
predicts that liftoff may result from either a deepening (Farmer & Armi 1986) or widening (Armi intense mixing
& Farmer 1986) of the flow. In the deepening case, the interface remains horizontal seaward of
the liftoff point, whereas it shoals when the flow widens. Liftoff is often observed at or near a
sill, where depths are beginning to increase; however, shoaling of the surface layer is consistently
observed, indicative of the importance of lateral flow expansion. Finally, the liftoff front may be
maintained at an angle to the oncoming flow such that local Froude numbers are substantially
supercritical (MacDonald & Geyer 2005). In these cases, the liftoff front must be anchored at a
position in the channel where local Froude numbers are critical, with the remainder of the front
defined by a Froude angle ψ = sin−1 Fr−1 1 , such that Fr1 is critical with respect to flow normal to
the front.
Second, the jet-like near-field region begins at liftoff, or at the river mouth if liftoff occurs in
the estuary, and is generally considered to be the region where the river momentum exceeds the
buoyancy of the plume layer (i.e., Fr1 > 1). Thus, the near-field region is generally considered
to end where the flow ceases to be supercritical (Fr1 = 1; Hetland 2005, 2010), although this
point can be difficult to identify because Fr1 gradually transitions from supercritical to subcritical
over a significant fraction of the near-field area. The near field has much in common with an
engineering jet (e.g., Jirka et al. 1981, Jones et al. 2007), although its large aspect ratio isolates it
from significant lateral mixing (Chen & MacDonald 2006).
The near-field momentum balance is dominated by barotropic and baroclinic pressure gradi-
ents, turbulent interfacial stress between the discharging plume layer and the underlying ambient
water, and flow acceleration (McCabe et al. 2009), with the interfacial stress as the primary driver
of plume deceleration under low discharge conditions and peak ebb tide (Kilcher et al. 2012). In

[Link] • Mixing and Transport in Coastal River Plumes 571

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

a
I NE
TL
AS
CO O (10 3 m)

Sea surface
gradients
u u
ρ
PLUME
SIDECUT O (10 0–101 m)
RIVER
FLOW
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

Flow
acceleration
Access provided by [Link] on 04/04/23. For personal use only.

and Maximum
intense mixing Fr1 Sustained mixing and spreading

O (10 3 m )
b Liftoff point
Fr1 = u(g'h) –1/2 = 1

SE
AB
ED
Large-scale
mixing
structures
c Shear
instabilities

O (10 0 –10 1)

Fr1
Supercritical
10 0
0
Distance from coast

Figure 1
(a) Schematic representation of the plume liftoff process and the near-field plume region. (b) Large-scale mixing structures in the flow
acceleration region. Panel b adapted from Geyer et al. (2010). (c) Shear instabilities comprising the large-scale mixing structures.

cases in which the river mouth is wide compared to the Rossby radius, L R = (gh)1/2/ f , a near-field
region will not exist, and the discharge will exit the estuary as a far-field plume, attached to the
right bank (in the Northern Hemisphere). In regions with large tides, the dynamics in the near-
Mid-field region: the
field region are strongly time dependent, and the near-field plume is also referred to as the tidal
region where Earth’s plume (Horner-Devine et al. 2009, Kilcher & Nash 2010, Kilcher et al. 2012).
rotation begins to Third, the mid-field is the region where inflowing river water transitions from the inertial
dominate, arresting near-field jet into a geostrophic or wind-dominated far-field plume. Lateral spreading is arrested,
plume spreading and inflow momentum is lost, and plume dynamics are increasingly dominated by Earth’s rotation. In
turning the plume in
the downcoast
most systems, the plume is initially directed offshore and turns in the mid-field region to form a
direction shore-parallel coastal current. This process is driven by Earth’s rotation and is associated with a
corresponding increase in centripetal acceleration (Garvine 1987, McCabe et al. 2009), although

572 Horner-Devine · Hetland · MacDonald

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

ambient alongshore currents may also contribute to turning the plume (Fong & Geyer 2002).
Under sufficiently low wind conditions, some outflows can form a recirculating bulge region in
the mid-field, as discussed in Section 4.3. The momentum balance in the mid-field is dominated
Bulge: an unsteady,
by the Coriolis force, centripetal acceleration, and the cross-stream internal pressure gradient anticyclonic eddy
(Garvine 1987, Yankovsky & Chapman 1997, Horner-Devine 2009, McCabe et al. 2009). circulation that can
Finally, the far field is the region beyond the mid-field in which the plume no longer has a form offshore of the
memory of the initial momentum of the river discharge, and the dynamics are primarily governed river mouth under
certain conditions,
by Earth’s rotation, buoyancy, wind stress, and sometimes bottom stress. This region may extend
expand continuously,
hundreds of kilometers from the mouth. When winds and ambient currents are not sufficient to and accumulate a
force the plume offshore, the far-field plume forms a geostrophic coastal current that transports fraction of the river
diluted river water in the direction of Kelvin wave propagation (hereafter referred to as downcoast). discharge
The structure, dynamics, and propagation speed of the coastal current depend on the shelf slope Far-field region: the
(Avicola & Huq 2002, Lentz & Helfrich 2002) and whether the plume is in contact with the region beyond the
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

bottom (Yankovsky & Chapman 1997). mid-field where the

plume water has lost
The anatomy of the prototypical large-scale plume, which is composed of all these regions,
all memory of the
Access provided by [Link] on 04/04/23. For personal use only.

is presented schematically in Figure 2. Numerical and laboratory model studies with simplified inflow momentum but
geometry generate plumes similar to the prototypical plume (e.g., Fong & Geyer 2002, Avicola is still distinct from the
& Huq 2003a, Horner-Devine et al. 2006), and field observations from a number of large-scale ambient receiving
plume systems indicate that they have a similar structure during low-wind conditions (e.g., Chant water
et al. 2008, Hickey et al. 2010). However, it important to note that some of these regions will
be absent in many plumes, and thus the structure of the plume may be quite different than the
prototypical plume. This variability in plume structure and behavior is discussed in Section 5.

3. MIXING

3.1. Local Mixing: Regions and Processes Within the Plume

Mixing in river plumes may be defined as the transport of buoyancy and momentum across isopy-
cnal surfaces (e.g., Sherman et al. 1978, Ivey et al. 2008, Stacey et al. 2011). Because of the very

Mid-field plume:
bulge

Tidal
plume
front
Far field:
coastal current

Near field
Liftoff
Source

River/
estuary

Figure 2
Schematic representation of the prototypical plume comprising all dynamical regions. Other plume
morphologies are discussed in Section 5 and shown in Figure 5.

[Link] • Mixing and Transport in Coastal River Plumes 573

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

small aspect ratio of most river plumes, O(10−3 ), the vertical turbulent flux of density, w ρ  , is
generally considered to be dominant over the horizontal turbulent fluxes.
In practice, the intensity of turbulence is often quantified based on the turbulent kinetic energy
dissipation rate , which can be estimated from field measurements using microstructure tech-
niques (e.g., Moum et al. 1995) or overturn scales (Orton & Jay 2005). Turbulent mixing is typically
quantified in terms of the turbulent vertical salt flux w S or the buoyancy flux B = (g/ρ)w ρ  .
The buoyancy flux in river plumes has been estimated directly using control volume techniques
(MacDonald & Geyer 2004, McCabe et al. 2008) or dye releases (Houghton et al. 2004, 2009).
In energetic mixing regions, the plume maintains a gradient Richardson number, Rig , of order
0.25, and an order unity turbulent Froude number, Frt , resulting in maximal mixing efficiencies
  0.2 (e.g., Ivey & Imberger 1991) and allowing buoyancy flux to be estimated from dissipation
measurements as B = . Here, the gradient Richardson and turbulent Froude numbers are
defined according to Rig ≡ N 2 (∂z u)−2 , where N 2 = −g∂z uρ0−1 , and Frt = urms N −1 L−1 , where
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

L is the scale of the large eddies.

The net impact of mixing in a specific region of the plume depends not only on the intensity
Access provided by [Link] on 04/04/23. For personal use only.

of the turbulence and the density gradient available for mixing, but also on the spatial extent of
the active mixing area and the duration of the mixing event. Mixing throughout the plume can
be related to the discharge by considering the plume in isohaline coordinates, as described by
Hetland (2005). The volume of plume water contained within an isohaline, with salinity SA , can
be related to the discharge and the average mixing across that isohaline surface as

∂
SA V f A = SA Q f + w S dA, (2)
∂t SA

where VfA is the volume of freshwater contained within the isohaline SA , and w S is the vertical
salt flux across the isohaline surface. In the limit of no mixing, w S = 0, the plume expands
like a balloon. In the steady-state limit, the input of freshwater is balanced by mixing across the
isohaline surface. Hetland (2005) continues this analysis by differentiating Equation 2 with respect
to salinity so that mixing and discharge can be related in specific regions of the plume, as defined
by particular salinity ranges. Typically, lower salinity classes of the plume near the source are
associated with stronger mixing and are smaller; higher salinity classes have weaker mixing and
are larger.

3.1.1. Mixing in the near-field region. Some of the most intense turbulence and mixing in the
plume occur in the near-field region, where  values are observed to be as high as 10−4 to 10−3 m2
s−3 (MacDonald & Geyer 2004, MacDonald et al. 2007, McCabe et al. 2008, Kilcher et al. 2012).
As described in Section 2, plume expansion seaward of the liftoff point (Fr1 = 1) causes the plume
interface to shoal rapidly, following internal hydraulic theory (Armi & Farmer 1986). Shoaling
of the interface in turn causes the upper layer to accelerate according to the inviscid upper-layer
momentum equation, u ∂x u = −g  ∂x h, where h is the interface depth. Acceleration in the upper
layer and shoaling isopycnals increase Fr1 and drive local Rig values down, resulting in intense
mixing. Turbulence is most intense in the shoaling region immediately seaward of the liftoff point
and subsequently decreases throughout the near field (Luketina & Imberger 1987, MacDonald
et al. 2007, McCabe et al. 2008).
The decrease in mixing is a consequence of the entrainment of low-momentum, salty ambient
water into the plume, which will decelerate the plume and decrease the Froude number according
to
∂W
∂Fr1 hu(Fr2 + 2) − 3we W Fr2
= ∂x , (3)
∂x 2W hFr(Fr2 − 1)

574 Horner-Devine · Hetland · MacDonald

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

where W is the plume width, we is an entrainment velocity of ambient fluid across the base of the
plume, and x is in the streamwise direction (Hetland 2010). Thus, the seaward evolution of Fr1
results from competition between the acceleration associated with spreading and the deceleration
associated with mixing, the left and right terms in the numerator, respectively.
Plume spreading enhances mixing in the near-field region by increasing shear as described
above and also by increasing the planform area over which mixing occurs (Yuan & Horner-
Devine 2013). The expansion is driven by lateral density gradients and the resulting stream-normal
barotropic pressure gradient, which is particularly strong in the immediate vicinity of the river
mouth, as discussed by McCabe et al. (2009). The lateral fronts expand away from the axis of the
plume as freely propagating gravity current fronts, resulting in a spreading rate that is proportional
to the local internal gravity wave speed, c = (g  h)1/2 (Wright & Coleman 1971, Hetland &
MacDonald 2008, Chen et al. 2009, Hetland 2010, Yuan & Horner-Devine 2013). In addition
to these mechanisms, recent observations by MacDonald & Chen (2012) provide evidence that
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

spreading may locally enhance turbulence as the divergence of the plume stretches the turbulent
eddies. This effect was not observed in subsequent laboratory experiments of spreading plumes
Access provided by [Link] on 04/04/23. For personal use only.

(Yuan & Horner-Devine 2013), however, and the impact of spreading on local turbulent mixing
processes remains uncertain. McCabe et al. (2009) discuss the relative role of rotation in plume
spreading. They argue that the important quantity for plume spreading is the sum of the stream-
normal pressure gradient and the Coriolis term. Spreading will vary across the plume; the pressure
and Coriolis terms are in opposition on the upcoast edge of the plume but are in concert on the
downcoast edge.
Mixing in river plumes is dominated by stratified-shear flow instabilities (Stacey et al. 2011,
Yuan & Horner-Devine 2013), particularly Kelvin-Helmholtz (KH) instabilities (Thorpe 1969,
Kantha et al. 1977, Trowbridge 1992, Smyth & Moum 2000). In the near-field plume, strong
shear, on the order of 0.5–1 s−1 , is often present at the plume base, with density differences of
15–20 kg m−3 between the plume and ambient waters. Corresponding gradient Richardson num-
bers are maintained at approximately 0.25, providing sufficient conditions for the development
of KH billows (e.g., Thorpe 1973, Smyth et al. 2001) and subsequent turbulent mixing. KH-like
structures have been observed acoustically in estuarine and plume flows (Tedford et al. 2009,
Geyer et al. 2010) at larger scales, although Geyer et al. (2010) indicate that these are self-similar
structures at scales an order of magnitude larger than the KH billows actually responsible for
mixing (Figure 1b). The size of KH billows is set by the Ozmidov scale, Lo = ε1/2 N −3/2 , which is
O(10 cm) in strongly stratified plume regions. Less than 10% of the interfacial volume is occupied
by coherent overturn structures capable of initiating significant mixing (MacDonald et al. 2013).
Turbulent dissipation rates and turbulent buoyancy fluxes resulting from stratified-shear mixing
processes are best nondimensionalized by the quantity g  U (Imberger & Ivey 1991, MacDonald
& Geyer 2004).

3.1.2. Wind and wave mixing. Wind stress generally results in lower turbulence levels than
are observed in more energetic areas such as the near field or plume front. Houghton et al.
(2009) use dye release measurements in the far-field plume and show that the salt fluxes vary from
approximately 5 × 10−5 kg m s−1 during low wind conditions to 3 × 10−4 kg m s−1 for high winds
(12 m s−1 ). Assuming  = 0.2, these are equivalent to average dissipation rates of 2 × 10−6 m2
s−3 and 1 × 10−5 m2 s−3 , which are one to two orders of magnitude lower than rates observed in
the front or near field. However, wind typically acts over the entire plume, and the spatial area of
active mixing may be large enough to make the net effect substantial.
Mixing of the plume by the wind relies primarily on shear in the surface Ekman layer. Ekman
transport in the surface layer will be constant for a given wind stress, but the flow speed is inversely

[Link] • Mixing and Transport in Coastal River Plumes 575

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

proportional to the plume depth. A model of plume mixing by the wind was presented by Fong
& Geyer (2001) and later modified by Lentz (2004) and Hetland (2005). The wind drives Ekman
transport in the plume layer, which drives vertical shear and shear mixing. As described by Fong &
Geyer (2001), the influence of wind on the far-field plume is modified by upwelling or downwelling
winds, which tend to thin or thicken the plume, respectively, thus modifying the mixing response.
η
It is convenient to define a freshwater thickness, h f = −H (S0 − S)/S0 d z, which is the vertical
integral of the salinity anomaly, S0 − S, relative to the ambient water salinity, S0 . An upper limit to
g is defined as g f = gρ f ρ0−1 , where ρ f is the density difference between purely freshwater and
ambient water. By combining conservation of mass, g  h p = g f h f , the Ekman transport equation,
u E h p = τ f −1 , and the definition of the critical bulk Richardson number, Ric = g  h p U E−2 , one can
calculate the critical plume thickness for a given wind stress as

2τ Ric
hp = . (4)
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

ρ0 f g f h f
Access provided by [Link] on 04/04/23. For personal use only.

The plume will mix to the critical depth, which depends on the wind stress and freshwater thickness,
and then mixing will be suppressed. If the plume has already mixed to this point, or past it, no
further mixing will occur. Thus, in this case, the wind supplies the kinetic energy for mixing, but it
is modulated by rotation in a way that prevents the wind from continually accelerating the fluid. In
a nonrotating case, a continuously applied wind would keep accelerating the fluid, and the plume
would continually mix. Thus, rotation limits the amount of mixing that can occur for a given wind
stress.
Wave breaking generates intense turbulence at the ocean surface, with observed dissipation
rates reaching values of 10−3 –10−1 m2 s−3 in typical wind waves (Agrawal et al. 1992, Gemmrich
2010, Thomson 2012). The wave height is the primary scale that determines the penetration
depth of breaking wave–enhanced turbulence (Terray et al. 1996), suggesting that plumes may be
strongly influenced by wave breaking when their thickness is of the same order as the wave height,
although this has yet to be tested in the field. Using a numerical model, Gerbi et al. (2013) find
that wave breaking has a substantial effect on the structure and dynamics of a river plume but find
little impact of wave breaking on mixing, presumably because the plume they investigate is too
thick to feel the effect of intensified surface turbulence at the density interface. Wave breaking is
parameterized in their model as a surface source of turbulent kinetic energy that diffuses down
into the water column, following Craig & Banner (1994).

3.1.3. Frontal mixing. Intense turbulence is often observed in plume fronts generated by ener-
getic tidal plumes, with dissipation rates on the order of 10−4 –10−3 m2 s−3 (Orton & Jay 2005,
O’Donnell et al. 2008, Horner-Devine et al. 2013), which are of a similar magnitude to those ob-
served in the near-field plume region. However, the contribution of the front to the mixing budget
of the plume and the magnitude of this contribution relative to mixing in the near- or far-field
plume regions are still uncertain. Current estimates of the frontal mixing contribution to the total
mixing range from approximately 100% (Pritchard & Huntley 2006) to 20% (Orton & Jay 2005)
and possibly much less (Cole 2014). Whereas turbulence is very high in the plume front, it decays
exponentially behind the front (O’Donnell et al. 2008, Horner-Devine et al. 2013), resulting in a
narrow band of active mixing. O’Donnell et al. (2008) estimate the size of the frontal mixing region
based on laboratory grid turbulence experiments (Itsweire et al. 1993) as L I = 5 u fr N −1 , resulting
in estimates ranging from 15 m in the Connecticut River plume to 60 m in the Columbia River
plume. Here u fr is the frontal propagation speed. The total area of frontal mixing is a function of L I
and the plume circumference, both of which may change as the plume front propagates offshore.

576 Horner-Devine · Hetland · MacDonald

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

Kilcher & Nash (2010) show that the frontal Froude number Fr = u fr (g  h)−1/2 in the Columbia
River plume is relatively constant in the tidal plume front during ebb tides, in large part because
of increasing freshwater flux from the estuary. This suggests that mixing may be maintained as the
frontal circumference increases, at least in a large tidal plume such as the Columbia River plume.
As the ebb tide ends and the estuary discharge decreases, the front speed drops below the ambient
internal wave speed, and it releases internal solitons, losing further energy to the wave field (Nash
& Moum 2005; Jay et al. 2009, 2010).

3.1.4. Mixing in the bottom boundary layer. When a plume interacts with the seafloor, there
may be convective adjustment owing to Ekman transport in the bottom boundary layer (see
Section 4.6). Although it is widely accepted that this process influences the plume frontal position,
it is not well understood how the associated mixing alters plume water masses. In shallow shelves
with strong tides, bottom stress associated with the tidal flow may mix the plume from the bottom.
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

This is more common in the type of broad, shallow seas that are found in northern Europe and
are referred to as ROFIs (see Simpson 1997). In these regions, the interaction of the rotational,
Access provided by [Link] on 04/04/23. For personal use only.

oscillatory bottom boundary layer with the plume’s stratification results in counter-rotating tidal
ellipses in the surface and bottom layers (Visser et al. 1994, Souza & Simpson 1996). This form
of tidal straining can periodically increase and decrease the vertical stratification (De Boer et al.
2006, 2008), can cause the plume to periodically separate from the coast (De Boer et al. 2009),
and may drive mixing (Fisher et al. 2002).

3.2. Integrated Plume Mixing: Quantifying and Comparing Mixing

Within and Across Plumes
The net dilution of freshwater as it traverses through a plume is determined by the sum of the
mixing processes and is thus essential in a complete description of plume mixing and transport.
To understand how the different mixing processes described in Section 3.1 combine to produce
the total dilution of the plume and to compare the relative contributions of each, one needs to
develop a framework for comparison.

3.2.1. Mixing budget. Many studies investigate the role of mixing in the context of an energy
balance (e.g., Winters et al. 1995, Wunsch & Ferrari 2004, MacCready et al. 2009). The most
relevant to the present review is that by MacCready et al. (2009), who investigate mixing in the
Columbia River plume using an energy balance for the estuary and plume regions. MacCready
et al. show how three types of energy—kinetic energy associated with the flow field, barotropic
energy associated with sea surface height displacements, and baroclinic energy associated with
density differences—change in response to various forcing terms. The energy balances in the
Columbia River estuary and plume show that pressure work dominates the balance in the estuary,
whereas either winds or tides were most important for driving mixing in the plume, depending on
the relative strength of the winds and the magnitude of the tides. Although mixing is observed to
be weaker, more heterogeneous, and more episodic in the wind-driven plume than in the estuary,
the net mixing is a significant component of the total energy budget in both. The energy balance
approach is most effective when hydrodynamic model results are used that capture the full spatial
and temporal variability of a system; it is virtually impossible to generate sufficient observational
data to adequately evaluate the mixing budget as described by MacCready et al. (2009).

3.2.2. Comparison of plume mixing estimates. Although considerable work has been done to
estimate mixing rates in different regions of the plume, we are still unable to quantify the relative

[Link] • Mixing and Transport in Coastal River Plumes 577

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

importance of each process or region in the net mixing in a plume. A notable exception is work by
Pritchard & Huntley (2006), who compare wind, tidal, and frontal mixing within the River Teign
plume. They argue that these three processes can explain all the mixing required to completely
disperse the small, tidally pulsed plume. Pritchard & Huntley model mixing in this system based on
in situ observations. Using a potential energy budget that is much simpler than that in MacCready
et al. (2009), they estimate the magnitudes of net wind, tidal, and frontal mixing and compare
them to the total potential energy supplied by the buoyant outflow. Their energy budget is based
on
 η
1
φ= gz(ρ̄ − ρ) d z, (5)
η + H −H
where ρ̄ is the vertical mean density. The term φ represents the mechanical energy required to
vertically homogenize the water column and is used as a metric to quantify stratification (e.g.,
Simpson et al. 1981, Burchard & Hofmeister 2008).
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

The River Teign forms a very small plume with Q f ∼ 5 m3 s−1 ; its plume is mixed out entirely
on each tide and forms anew on the following ebb. Pritchard & Huntley (2006) find that frontal
Access provided by [Link] on 04/04/23. For personal use only.

mixing dominates in this plume and is capable of mixing 100% of the buoyancy input on each
tide. Wind mixing is only comparable to frontal mixing with winds between 15 and 20 m s−1 , and
tidal mixing is always negligible. This result is in sharp contrast to the results from the Columbia
River plume (MacCready et al. 2009), suggesting that the relative contributions of each mixing
process depend on the magnitude of the river discharge and plume.
The normalization presented by Pritchard & Huntley (2006) provides a framework for compar-
ing the relative importance of different mixing mechanisms in the dilution of a plume. It considers
the rate of energy converted to mixing due to a specific process relative to the total energy required
to mix the freshwater discharge completely into the ambient ocean. We express this ratio here
as the ratio of the turbulent buoyancy flux B for an individual mixing process to the available
potential energy from the estuary:
BA
M =2 γ T ∗, (6)
g f Q
where γ = g M (g f − g M )−1 , and A is the horizontal area over which the buoyancy flux acts. The
mixing normalization M differs from that of Pritchard & Huntley (2006) in two ways. First, we
consider the amount of energy required to mix the discharge from its initial g 0 to a final g M ,
instead of complete mixing over the entire water depth. Second, we introduce a nondimensional
time variable T∗ , which represents the fraction of time for which the process is active. For example,
T∗ is between 0.3 and 0.5 for a process that occurs only on the ebb tide. In theory, individual M
terms representing the various components of a single plume realization should sum approximately
to 1.
Measurements of  or B from observations in various regions (e.g., Luketina & Imberger 1987,
Houghton et al. 2004, MacDonald & Geyer 2004, Orton & Jay 2005, MacDonald et al. 2007,
McCabe et al. 2008, O’Donnell et al. 2008, Kilcher et al. 2012, Horner-Devine et al. 2013) could be
used to estimate M, but these estimates are typically point measurements, which must be assumed
representative of processes over some unknown volume and time fraction. Approximations of the
spatial and temporal extent of mixing can be made, in some cases more reliably than others, but
in any case, they generally inject significant uncertainty into estimates of M from available field
data.
A comparison of values of M estimated based on plume turbulence measurements summarized
in Section 3.1 illustrates the challenges in constructing a meaningful mixing budget. Near-field

578 Horner-Devine · Hetland · MacDonald

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

M values based on observations from the Columbia and Merrimack Rivers (e.g., MacDonald et al.
2007, Kilcher et al. 2012) appear reasonable and suggest that near-field mixing might be responsi-
ble for 10–30% of the total required mixing. However, estimates based on observations from the
far-field regions of the Delaware Bay outflow and Hudson River (Houghton et al. 2004, 2009);
frontal regions in the Columbia River (Orton & Jay 2005), Merrimack River (Horner-Devine
et al. 2013), and Connecticut River (O’Donnell et al. 2008); and the mid-field plume region
of the Columbia River (McCabe et al. 2008) appear far less robust, with values ranging from
less than 0.2% to nearly 400%. Clearly, these observational estimates are significantly impacted
by errors in estimating the volumetric extent of mixing and variability of mixing across plume
regions. This highlights an important gap in our current understanding of plume dynamics, for
which most studies to date are process oriented and focused on point measurements, which may
or may not be representative of the entire plume over larger spatial and temporal scales.
The range of estimates above also highlights the difficulty in using temporally and spatially
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

isolated measurements from different regions to draw broad conclusions about plume mixing.
As conditions may vary from ebb to ebb, enhanced mixing in one region might decrease the
Access provided by [Link] on 04/04/23. For personal use only.

energy available to mix in a subsequent region, thus altering the chain of M values leading to
complete dilution. This highlights the fact that many different pathways to dilution may exist for a
single system, depending on a variety of external forcing mechanisms. Hetland (2005) describes a
plume as a series of connected regions, with each region setting the water properties that enter the
next region. Although this mixing sequence has been well characterized using numerical models,
these models necessarily involve parameterizations of the mixing processes. It is a tremendous
observational challenge to resolve the mixing processes across the temporal and spatial scales of the
plume necessary to accurately compare interconnected mixing processes acting in different regions
of the plume. For example, using multiyear ferry-based measurements of the Fraser River plume,
Halverson & Palowicz (2008) show that increased river discharge decreases the salinity of the water
exiting the near field, thus quantifying the net mixing across the near-field plume under different
forcing. Recent in situ measurements of the Columbia and Merrimack River plumes using multiple
concurrent platforms may bring us closer to the goal of actively comparing interconnected mixing
processes occurring in different regions of a plume. Also, recent advances in remote sensing may
offer solutions to estimating mixing across the entire plume synoptically (Chickadel et al. 2011).

4. TRANSPORT

4.1. Plume Transport

Transport of river water away from the mouth is driven primarily by the plume’s buoyancy,
wind stress, and ambient coastal currents and is strongly mediated by Earth’s rotation. Mixing
processes near the river mouth or in the core of the near-field plume play an important role in
plume transport because baroclinic pressure gradients, proportional to internal density gradients,
drive plume transport. In the absence of significant external forcing, transport away from the river
mouth may be dominated by the retention of river water in a persistent eddy feature referred to
as the bulge. In this section, we describe the dominant parameters that modify the alongshore
transport. Most alongshore transport occurs in the coastal current region of the plume, which
is typically in geostrophic balance. However, winds, tides, and bottom slope will influence the
structure and transport in the coastal current; this is described in order to present a full picture of
the transport processes.

[Link] • Mixing and Transport in Coastal River Plumes 579

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

4.2. Control Volume Approach

Nof & Pichevin (2001) provide a theoretical framework for understanding the alongshore transport
in a simplified river inflow. Their approach is based on the layer-averaged alongshore momentum
equation, integrated over a control volume that bounds the river mouth region. Equation 7 is
a generalized expression for the control volume–integrated alongshore momentum for a river
entering at angle θ i :
 
∂ UQ 1 g h2
− (uh)d xd y + f (vh)d xd y + cosi 2 + 2 +Fe = u2 h + d y . (7)
∂t 2 Fri 2
 A
   A
     c  
Mt MC M cc
Mi

Mt and MC are the momentum contributions due to the unsteadiness of the plume and Cori-
olis force resulting from the net cross-shore velocity in the control volume, respectively. Mi is
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

the inflow contribution of the alongshore momentum as well as the alongshore component of
the baroclinic pressure force acting on the control volume at the inflow. This term is zero if the
Access provided by [Link] on 04/04/23. For personal use only.

river inflow is perpendicular to the coast, θi = 90◦ , and provides a contribution in the positive
x direction if θi < 90◦ . Fri is the inflow Froude number, based on the average inflow velocity,
depth, and density anomaly. Fe represents the external forces acting on the plume, including winds,
bottom drag, and the force imparted by ambient currents. Mcc is the momentum flux out of the
control volume in the coastal current plus the baroclinic pressure force back on the control volume.
Because it is positive definite, this term provides a force on the control volume in the negative x
direction. It is important to note that Equation 7 is a layer model and does not explicitly include
mixing processes. This is a reasonable approximation in systems in which winds and tidal mixing
are low and much of the plume mixing occurs in the estuary and the near-field plume.
Pichevin & Nof (1997) consider the simplified case in which the inflow is perpendicular to the
coast (M i = 0), the flow is steady (M t = 0), there is no net across-shore flow (M C = 0), and
there are no external forces (Fe = 0). In this case, no force exists to balance the flux of momentum
into the coastal current Mcc , resulting in what they refer to as the momentum imbalance paradox.
Because this is not possible, one of the neglected terms in Equation 7 must be nonzero.

4.3. The Bulge and the Reduction of Alongshore Flow

Nof & Pichevin (2001) conclude that the momentum equation can be satisfied if an unsteady
growing recirculation region, referred to as the bulge, exists. The bulge, which is shown in
Figure 2, is a persistent feature in most simplified numerical models (e.g., Oey & Mellor 1993,
Kourafalou et al. 1996, Garvine 2001, Fong & Geyer 2002) and laboratory experiments (Avicola
& Huq 2003a,b; Horner-Devine et al. 2006) simulating river plumes, when there is no ambient
current and the wind stress is low or moderate. Bulge formation requires that the estuary width
is narrow relative to the Rossby deformation radius [i.e., the Kelvin number K = W f (g  h)−1 is
less than 1] (Garvine 1995, Huq 2009). When present, the bulge is a subtidal feature that grows
continuously in time, accumulating between 25% and 75% of the river’s freshwater discharge
and causing a corresponding reduction in the transport of freshwater away from the river mouth
region in the coastal current (Fong & Geyer 2002, Horner-Devine et al. 2006). The mechanism
leading to retention in the bulge is associated with splitting of the buoyant current as it impinges
on the coast at a stagnation point downstream of the river mouth. Based on the momentum
balance presented by Whitehead (1985), the fraction that returns to the bulge rather than flowing
into the coastal current depends on the incidence angle of the impinging bulge current (Avicola
& Huq 2003a, Horner-Devine et al. 2006). Recently, a well-defined bulge circulation and

580 Horner-Devine · Hetland · MacDonald

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

corresponding reduction in alongshore freshwater transport have been observed in large-scale

field sampling efforts (e.g., Chant et al. 2008, Horner-Devine 2009). The unsteadiness of the
bulge flow is documented in satellite observations (Horner-Devine et al. 2008), and its impacts
on the coastal ecosystem have been observed (Chant et al. 2008, Kudela et al. 2010).
The bulge flow consists primarily of a large eddy rotating anticyclonically in which the momen-
tum is in gradient-wind balance (Yankovsky & Chapman 1997; Nof & Pichevin 2001; Horner-
Devine et al. 2006, 2009),
u 2θ ∂h
+ f uθ = g  . (8)
r ∂r
This balance is typically expressed in a cylindrical coordinate system in which uθ is the azimuthal
velocity and r is a radial coordinate with its origin at the center of the bulge. There is an initial setup
period, during which the unsteady acceleration term ∂ur /∂t is also important in the momentum
balance. After this 1–2-day period, the bulge is in gradient-wind balance, despite the fact that
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

it is continually expanding (Horner-Devine 2009). Nof & Pichevin (2001) assume M t  0 and
model the offshore expansion as a constant net cross-shore migration speed Cy , and the resulting
Access provided by [Link] on 04/04/23. For personal use only.

alongshore momentum balance is between MC and Mcc . If the bulge is in gradient-wind balance,
this results in predictions for the central depth Hb , radius Rb , and migration speed Cy for the bulge
(see Nof & Pichevin 2001). In their formulation, the alongshore transport in the coastal current
is given by
Q cc 1
= , (9)
Q 1 + 2α
where α = −2u θ / f r is the nondimensional bulge vorticity. When the potential vorticity is zero,
corresponding to a relatively intense eddy with α = 1, Equation 9 predicts that 33% of the river
discharge will be transported away from the river mouth in the coastal current, and the remaining
67% will accumulate in the bulge. When the eddy is weaker (α  1), the coastal current transport
is predicted to increase (Figure 3). Few field data exist to test Equation 9, but estimates from both
the Columbia River (Horner-Devine 2009) and Hudson River (Chant et al. 2008) plumes suggest
that it provides a reasonable prediction (Figure 3).
The drawback of Equation 9 is that it cannot provide an a priori estimate of the alongshore
transport because it depends on the observed bulge vorticity α. Efforts to predict Q cc Q −1 a priori
do so based on the inflow Rossby number Roi = U f −1 W −1 (Fong & Geyer 2002, Horner-
Devine et al. 2006), which can be interpreted as the inflow vorticity. Here, U and W are the inflow
velocity and width, respectively. Q cc Q −1 is less sensitive to Roi compared with the prediction for
α (Figure 3). This may be related to the amount of mixing occurring in the near-field plume,
which will determine the degree of modification of vorticity between the inflow and the bulge.
A good understanding of how inflow conditions set α may provide the information necessary to
generate an a priori prediction of alongshore plume transport for this simplified case.

4.4. The Role of Wind

In addition to mixing via Ekman shear (Section 3.1.2), wind may also translate a plume and enhance
or limit transport. Wind is often the dominant driver of transport in a plume. In the case of a
thin plume that is not in contact with the seafloor, this transport is straightforward and should
correspond to Fe ∼ M cc . In general, a downwelling wind will enhance alongshore transport,
whereas an upwelling wind will diminish it (Fong & Geyer 2001). An upwelling wind is an along-
shore wind in the opposite direction to the coastal current that creates a divergent, offshore flow
due to Ekman transport; a downwelling wind blows in the same direction as the coastal current

[Link] • Mixing and Transport in Coastal River Plumes 581

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

1.0

0.9

0.8

0.7

0.6
Q cc Q –1
0.5

0.4
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

Roi : Numerical (Fong & Geyer 2002)

0.3 Roi : Laboratory (Horner-Devine et al. 2006)
Access provided by [Link] on 04/04/23. For personal use only.

α: Inviscid theory (Nof & Pichevin 2001)

α: Columbia River (Horner-Devine 2009)
0.2
α: Hudson River (Chant et al. 2008)

0.1
10 –2 10 –1 10 0
Roi , α

Figure 3
Comparison of theoretical predictions for the coastal current transport from Equation 9 (Nof & Pichevin
2001) with numerical (Fong & Geyer 2002), experimental (Horner-Devine et al. 2006), and field
observational (Chant et al. 2008, Horner-Devine 2009) estimates. Note that the abscissa is Ro for the
laboratory and model data and α for the theory and field data. These correspond to estimates of the
normalized vorticity at the inflow and in the bulge, respectively.

and creates a convergent, onshore flow. However, for a plume in which the pycnocline intersects
a significant stretch of the seafloor, wind forcing is more complicated. For an unstratified shelf,
the convergence or divergence of the Ekman transport at the coast will cause a setup or setdown
of the sea level, which in turn will drive geostrophic alongshore currents in the same direction as
the alongshore wind stress (e.g., Csanady 1977). Buoyancy-driven coastal currents will respond
to the wind in the same way, with two notable modifications. First, the buoyancy-driven flow will
be superimposed on the wind-driven currents, such that an upcoast wind will only drive upcoast
currents if it can overpower the buoyancy-driven flow (Zhang & Hetland 2012). Second, the
cross-shore density structure must evolve in response to the wind stress such that it is also in
geostrophic balance with the wind-driven currents (Whitney & Garvine 2005, Zhang 2013).

4.5. Modification of Transport by Other Processes

Although the bulge flow described in Section 4.3 provides a useful baseline for understanding
plume transport and presents a good target for analytical modeling, it is applicable only a fraction
of the time in real plumes due to coastline geometry, ambient currents, tides, and winds that often
dominate plume transport (e.g., Kudela et al. 2010).
When the inflow angle θi is reduced below 90◦ , the inflow contributes momentum in the
alongshore direction that can act to balance the flux of momentum into the coastal current (M i ∼
M cc ). Thus, the offshore expansion and Coriolis term decrease, and more fluid is transported away

582 Horner-Devine · Hetland · MacDonald

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

from the river mouth in the coastal current. Almost all the river discharge is carried downstream in
the coastal current when θ i is less than a critical angle between 40◦ and 60◦ (Garvine 2001, Avicola
& Huq 2003b, Horner-Devine 2004). The propensity to generate a bulge is related to the degree of
flow separation from the coast (Bormans & Garrett 1989) and is explained in terms of a separation
ratio d f/u, where d is the maximum offshore distance of the separated flow (Avicola & Huq 2003b).
Ambient along-coast currents can also significantly influence plume transport, augmenting
transport if the current is in the same direction or decreasing transport if the current opposes
the plume. Fong & Geyer (2002) show that the barotropic transport component augments the
baroclinic transport that is present in the plume in the absence of an ambient current. Even a
relatively small ambient current of 0.2 m s−1 is sufficient to increase the alongshore transport to
match the river discharge.
Isobe (2005) reports that tidal currents stabilize the bulge and increase alongshore transport.
His simulations suggest that after approximately 60 days, the coastal current transport is equal to
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

the river discharge in small- to medium-sized discharges with tidal currents of 0.3 m s−1 .
Access provided by [Link] on 04/04/23. For personal use only.

4.6. Coastal Currents over Sloping Bathymetry

Near the source, there can be significant cross-shore currents and substantial mixing. Eventually,
however, in the absence of prominent headlands or bays, the flow will evolve such that the currents
are very nearly in geostrophic balance, oriented along shore, with density gradients primarily
oriented offshore. The flow within the coastal current is in thermal wind balance such that the
vertical shear is in balance with the horizontal density gradients. However, this balance does not
set the transport of the coastal current, which is specified upstream by the value of Qcc .
If a buoyancy-driven current interacts with the bottom, it will adjust such that the bottom
flow is zero. Such balance is similar to the buoyancy arrest in the bottom boundary layer during
upwelling conditions described by MacCready & Rhines (1993) but occurs through the entire water
column. If the bottom currents are initially downcoast in the same sense as the surface buoyant
flow, this will drive an offshore Ekman transport within the bottom boundary layer in which light
coastal current water is pushed beneath heavier shelf water. The unstable water column mixes and
reduces the integrated salinity at a particular depth due to the offshore flux of freshwater in the
bottom boundary layer. The net effect of this process is to drive the plume front offshore, into
deeper waters. Eventually the geostrophic transport within the plume front comes into balance
with the thermal wind shear such that the flow at the bottom is approximately zero, thus halting
further offshore migration of the plume front. This process is described in detail by Chapman &
Lentz (1994) and Yankovsky & Chapman (1997) in a series of idealized numerical experiments.
Curiously, these numerical experiments suggest that the cross-shore density gradients within the
plume front remain roughly constant through the offshore migration process, such that the plume
front retains its structure as it is pushed offshore. Observations (e.g., Kirincich & Hebert 2005) of
buoyancy-driven flow over the continental shelf support the idea of buoyancy arrest; in fact, it is
a very reasonable approximation to assume quiescent bottom currents for buoyancy-driven flows
over the continental shelf at timescales of more than a few rotational periods.
Lentz & Helfrich (2002) and Avicola & Huq (2002) investigate the influence of a sloping bottom
on the propagation of a plume. When the plume propagates along a vertical wall, the propagating
nose is fed by currents along the wall, with a return flow back toward the source, relative to the

translating nose. The nose propagation speed is given by c w = g  h p = (2Qg  f )1/4 , where hp is
the maximum thickness of the plume, and Q is the net transport carried within the plume. If the
plume propagates over a very gently sloping bottom, with slope α, with the propagation speed
entirely determined by the bottom slope, the propagation speed reduces to c α = αg  f −1 . For

[Link] • Mixing and Transport in Coastal River Plumes 583

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

intermediate cases, the propagation speed is

cw
cp = . (10)
1 + c w /c α
This theory is able to explain the arrival time of pulses of freshwater from the Chesapeake River
plume observed at Duck, North Carolina (Rennie et al. 1999).

5. DISCUSSION

5.1. Pathways of Dilution

Individual mixing and transport processes associated with freshwater discharge to the coastal
ocean are the core components of any river plume system, and the isolation and investigation
of these processes are the focus of most river plume studies to date. However, the integration
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

of these individual processes creates the resulting plume structure and defines the pathway to
dilution for terrestrially derived freshwater. As described in Section 2, freshwater dilution begins
Access provided by [Link] on 04/04/23. For personal use only.

in the estuary, where a variety of mixing processes driven by tides and river discharge can serve to
increase the discharge salinity before it reaches the coastline, and it continues offshore through a
series of plume regions before its ultimate dissolution into the adjacent ocean waters. There are
several resulting plume structures, depending on the relative importance of the various mixing
and transport processes, resulting in a variety of dilution pathways. The nature of these pathways
is critically important for identifying the overall fate and transport of the freshwater discharge, as
well as any nutrients, pollutants, sediments, or other admixtures within the plume.
Owing to the addition of salt into the plume by mixing, the plume will become increasingly
diluted as it evolves, characterized by an ever-decreasing density difference between the plume
and ambient ocean waters. Thus, depending on the nature and strength of the mixing processes
relative to the initial magnitude of the discharge, the dilution pathway may terminate at any stage,
such that the far-field plume may be observed only for a subset of river discharges or may contain
only a fraction of the total discharge.
In some cases, limited initial discharge or enhanced mixing may result in the dilution of the
entire freshwater signal before the plume has time to pass through all the potential stages of its
evolution (i.e., near field, mid-field/bulge, far field/coastal current). This is certainly the case for
the River Teign plume (Pritchard & Huntley 2006) and other very small discharge plumes, but
it is likely to be the case even for larger plumes, at least during specific times of the year. Other
plumes, such as the Merrimack or Columbia River plumes (Hickey et al. 2010), may be dissolved
within the mid-field region, depending on forcing conditions. Even plumes that lack a near-field
region, such as the Delaware Bay plume, may be swept offshore and mixed away prior to the
formation of a coherent coastal current, depending on wind conditions. Thus, a myriad of plume
structures, and resulting pathways to dilution, can exist. Figure 4 illustrates the possible dilution
pathways that could exist, depending on various physical considerations at several critical points
in the evolution of a plume. We note that in a single system, the dilution pathway and associated
plume structure may shift as a function of river discharge, tidal range, winds, and so on, such
that it is possible that a single system could at different times display many of the possible plume
structures.
Most river discharges are associated with an estuary, where dilution begins in the laterally
constrained lower reaches of the river channel. If freshwater discharge is high enough to maintain
a supercritical value of Fr1 at the river mouth through all phases of the tidal cycle, then no estuary
exists, and all dilution processes are pushed into the unconfined coastal ocean. In very limited cases,
with sufficiently small discharges, complete dilution could occur within the limits of the estuary,

584 Horner-Devine · Hetland · MacDonald

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

Is there Is there
Qf an estuary? a near-field Near-field mixing mechanisms
YES plume? YES Spreading/stratified-shear flow
uf (g’f H)–1/2 < 1 u(g’ h)–1/2 > 1 Frontal
Wind/waves Potential for
plume to disperse
at any stage
NO NO Mid-field mixing mechanisms
Frontal
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

Rotation arrest
Wind/waves
Access provided by [Link] on 04/04/23. For personal use only.

Retention YES
for >1 day Is there
a bulge?
NO
Low wind
K≤1
θi > 60

Far-field mixing mechanisms

Ekman shear
Wind/waves
Bottom interaction* if
(2Qf/g’)1/2 > H

* Yankovsky & Chapman (1997) C A D

Figure 4
A schematic representation of mixing pathways in river plumes based on the presence or absence of different plume regions. Pathways
A–D correspond to plumes A–D in Figure 5.

such that no plume is formed (see Geyer & MacCready 2014 for a detailed review of estuarine
processes). The vast majority of discharge conditions are characterized by both an estuary and a
plume.
At its simplest, the plume structure can be reduced to two defining aspects: (a) Does a super-
critical near-field region exist, and (b) does a recirculating bulge exist? A near-field region exists
if conditions at the mouth are supercritical with respect to an internal Froude number. If this
condition is not met, then the plume will form a far-field plume or coastal current upon discharg-
ing from the estuary, with the dynamics of the forming plume strongly influenced by rotation.

[Link] • Mixing and Transport in Coastal River Plumes 585

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

If the supercritical condition is met, then a near-field region will be present until the outflow
conditions are reduced to subcritical with respect to the internal Froude number at some distance
(typically on the order of kilometers) from the river mouth. Beyond the near-field region lies a
transition region, or mid-field plume, where the processes active in the near field transition to the
geostrophic dynamics active in the far field. Ultimately, far enough downstream, the plume will
become geostrophic, entering the final stage of plume dilution. All these regions may be influ-
enced by winds, waves, background currents, and tidal currents. The relative importance of wind
is typically considered to be weakest in the near field, where the inertia of the river discharge is
dominant.

5.2. A Menagerie of Plumes

The variety of dilution pathways identified in Figure 4 emphasizes that river plumes are multi-
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

scale flow structures comprising a number of distinct processes and regions. Plume dynamics are
determined by stratified mixing processes operating at scales of 10−2 –10 m, large-scale transport
Access provided by [Link] on 04/04/23. For personal use only.

processes at scales of 103 –105 m, and a host of other processes that span the range between these
scales. In an individual river plume, the processes that are most important are determined by the
scale, geometry, and forcing of the plume. Figure 5 shows a menagerie of plume morphologies,
which represent most river plumes observed in the field. The imprint of some of these observed
plume types is shown by the color-coded pathways in Figure 4. These plume types represent
characteristic morphologies based largely on the geometry of the coast and the magnitude of the
discharge. However, external forcing, such as wind in particular, will often alter the morphology
of the plume such that the shape of the plume and the dilution pathways may be very different
under different forcing conditions.
Prototypical plumes (plume A), comprising all dynamical regions, have a relatively high dis-
charge and narrow mouth, and their dynamics are strongly modified by Earth’s rotation. These
plumes have the morphology observed in laboratory and numerical model experiments for a sim-
plified coastal inflow. Examples exhibiting this morphology are the Columbia River plume (Hickey
et al. 2010, 1998), the Merrimack River plume (MacDonald et al. 2007), the Niagara River plume
(Masse & Murthy 1992, Horner-Devine et al. 2008), and the Hudson River plume (Chant et al.
2008). Important dynamical constraints in this type of plume include the internal Froude number,
Fr1 , which sets the scale and intensity of the near field, and the Rossby number, which sets the
scale for the downstream transition to a coastal current (Yankovsky & Chapman 1997) and the
alongshore transport (Fong & Geyer 2002, Horner-Devine et al. 2006).
Nonrotational plumes (plume B), which have no mid-field or geostrophic coastal current, feel
little effect of Earth’s rotation, either because they are too small, as in the case of the River Teign
plume (Pritchard & Huntley 2006), or because the river mouth is close to the equator, as in the case
of the Amazon River plume (Lentz & Limeburner 1995). Their behavior is primarily characterized
by an internal Froude number, Fr1 , resulting in various outflow conditions (Kashiwamura &
Yoshida 1978), and can often be considered analogous to an engineering jet (e.g., Jirka et al. 1981,
Atkinson 1993, Chen & MacDonald 2006, Jones et al. 2007). Many plumes in laboratory studies
(e.g., Yuan & Horner-Devine 2013) are nonrotational and thus fall into this category.
Wide estuary plumes (plume C), which have no near field, a weak mid-field, and no bulge,
have a sufficiently wide mouth that the freshwater separates from one side due to Earth’s rotation
inside the estuary. These plumes are similar to a geostrophic coastal current that adjusts as it
rounds the corner into the coastal ocean, and as such comprise primarily far-field plumes. Classic
examples include the Chesapeake Bay plume (e.g., Valle Levinson et al. 1998, Johnson et al.
2001, Dzwonkowski & Yan 2005) and the Delaware Bay plume (e.g., Whitney & Garvine 2006,

586 Horner-Devine · Hetland · MacDonald

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

20 km 2 km
Plume A : Plume B :
prototypical nonrotational
Mid-field

River Teign, England

(X-band radar image)
Columbia River, Washington
(satellite imagery)

Near Far field

field
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

River
Access provided by [Link] on 04/04/23. For personal use only.

10 km 10 km
Plume C : Plume D :
wide estuary angled inflow

Chesapeake Bay, Virginia (SAR image) Eel River, California (rendering)

30 km 20 km
Plume E: Plume F:
delta region of
freshwater
influence

Mekong Delta, Vietnam Rhine River, Netherlands

(satellite-derived turbidity) (AVHRR SST)

Figure 5
River plume morphologies: prototypical (plume A), nonrotational (plume B), wide estuary (plume C), angled inflow (plume D), delta
plume (plume E), and region of freshwater influence (plume F). Inset images show examples of each plume type: (plume A) the
Columbia River (SeaWiFS), (plume B) River Teign (Pritchard & Huntley 2006), (plume C) Chesapeake Bay (Donato & Marmorino
2002), (plume D) Eel River (Geyer et al. 2000), (plume E) Mekong River (Wisdom Project), and (plume F) Rhine River (De Boer et al.
2009).
[Link] • Mixing and Transport in Coastal River Plumes 587
 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

Tilburg et al. 2007) outflows. In these plumes, rotation is of primary importance, and they can be
characterized by the Ekman and Rossby numbers.
In angled inflow plumes (plume D), which typically have no bulge, the inflow is at a small angle
to the coast, so it has a significant component of alongshore momentum initially. An example is the
Eel River plume along the northern California coast (Geyer et al. 2000). This plume orientation
can significantly affect plume transport, depending, in particular, on whether the inflow is aligned
with or opposed to the direction of the coastal current. Depending on its size, this type of plume
may be characterized by the internal Froude number or the rotational influences defined by the
Ekman and Rossby numbers.
Delta plumes (plume E), which have mutually interacting regions, enter the ocean through
a delta system, delivering freshwater via a number of river channels. Plumes formed by each
of the channel outlets will interact with each other, depending on the spacing of the channels
(Yuan et al. 2011). This formation is characteristic of many of the world’s largest river systems
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

(Wiseman & Garvine 1995), including the Mississippi (e.g., Wright & Coleman 1971), the
Changjiang (Yangtze) (Beardsley et al. 1985), the Nile, and the Mekong Deltas. In this case,
Access provided by [Link] on 04/04/23. For personal use only.

plume characterization is highly dependent on the overall size of the plume and also the nature
and spacing of the individual distributaries.
In ROFIs (plume F), which have a strong interaction with the bottom, bottom friction is
important and controls the cross-shore scale of the plume. Shallow marginal seas contribute to
this type of plume being more prevalent along European coasts, with the Rhine (Simpson et al.
1993) and Liverpool Bay (Verspecht et al. 2009) as classic examples. The critical balance in these
plumes exists between the destratifying effects of mixing due to bottom friction and the stratifying
effects of sheared flow in the presence of lateral density gradients, characterized by the Simpson
number,
g ∂ρ
H
ρ0 ∂ x
Si = 2
. (11)
cDU

5.3. Future Research Needs

Research on river plumes over the past several decades has resulted in the development of a frame-
work for understanding the transformation of river discharge and its dissolution into the ambient
ocean, including an idealized view of the plume structure and a detailed understanding of many
mechanisms responsible for plume mixing. Modern numerical models have shown significant skill
in predicting plume water properties, positions, and circulation patterns (MacDonald et al. 2007,
Chant et al. 2008, Liu et al. 2009, Hetland & DiMarco 2012). However, significant gaps still exist
in our understanding, including a mechanistic understanding of the role of winds and waves on
plumes, the evolution of frontal mixing as plumes evolve, and the ultimate fate of river plumes as
they merge into and alter ambient shelf and ocean waters. Perhaps more importantly, our under-
standing of plume mixing mechanisms focuses on specific locations within a plume. The significant
spatial and temporal variability in plume structure and mixing processes precludes meaningful
comparisons of these mixing estimates between plumes and even within the same plume system.
Furthermore, no coherent scaling framework exists that sufficiently explains the relationships be-
tween plume forcing and mixing in various plume regions and that would enable intercomparisons
among plume regions, among different forcing conditions for the same plume, and among different
plume systems. Finally, no good methodology exists for making synoptic measurements of plume
mixing. Thus, although significant advances have been made in understanding a few specific plume
types and mixing process, we still lack a general framework in which to compare and contrast

588 Horner-Devine · Hetland · MacDonald

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

all plume types and predict the dominant mixing mechanisms and circulation patterns within a
plume.

DISCLOSURE STATEMENT
The authors are not aware of any biases that might be perceived as affecting the objectivity of this
review.

ACKNOWLEDGMENTS
The authors would like to acknowledge support from National Science Foundation grants OCE-
0550096, OCE-0850948, OCE-1233068, and OCE-0648655. A.R.H.-D. would also like to ac-
knowledge support from the Allan and Inger Osberg Professorship.
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]
Access provided by [Link] on 04/04/23. For personal use only.

LITERATURE CITED
Agrawal Y, Terray EA, Donelan MA, Hwang PA, Williams AJ III, et al. 1992. Enhanced dissipation of kinetic
energy beneath surface waves. Nature 359:219–20
Armi L, Farmer DM. 1986. Maximal two-layer exchange through a contraction with barotropic net flow.
J. Fluid Mech. 164:27–51
Atkinson JF. 1993. Detachment of buoyant surface jets discharged on a slope. J. Hydraul. Eng. 119:878–94
Avicola G, Huq P. 2002. Scaling analysis for the interaction between a buoyant coastal current and the
continental shelf: experiments and observations. J. Phys. Oceanogr. 32:3233–48
Avicola G, Huq P. 2003a. The characteristics of the recirculating bulge region in coastal buoyant outflows.
J. Mar. Res. 61:435–63
Avicola G, Huq P. 2003b. The role of outflow geometry in the formation of the recirculating bulge region in
coastal buoyant outflows. J. Mar. Res. 61:411–34
Beardsley RC, Limeburner R, Yu H, Cannon GA. 1985. Discharge of the Changjiang (Yangtze River) into
the East China Sea. Cont. Shelf Res. 4:57–76
Bormans M, Garrett C. 1989. The effect of rotation on the surface inflow through the Strait of Gibraltar.
J. Phys. Oceanogr. 19:1535–42
Burchard H, Hofmeister R. 2008. A dynamic equation for the potential energy anomaly for analysing mixing
and stratification in estuaries and coastal seas. Estuar. Coast. Shelf Sci. 77:679–87
Chant RJ. 2011. Interactions between estuaries and coasts: river plumes—their formation transport and dis-
persal. In Treatise on Estuarine and Coastal Science, Vol. 2, ed. E Wolanski, D McLusky, pp. 213–35.
Amsterdam: Elsevier
Chant RJ, Wilkin J, Zhang W, Choi B, Hunter E, et al. 2008. Dispersal of the Hudson River Plume in theNew
York Bight: synthesis of observational and numerical studies during LaTTE. Oceanography 21:148–61
Chapman DC, Lentz SJ. 1994. Trapping of a coastal density front by the bottom boundary layer. J. Phys.
Oceanogr. 24:1464–79
Chen F, MacDonald DG. 2006. Role of mixing in the structure and evolution of a buoyant discharge plume.
J. Geophys. Res. 111:C11002
Chen F, MacDonald DG, Hetland RD. 2009. Lateral spreading of a near-field river plume: observations and
numerical simulations. J. Geophys. Res. 114:C07013
Chickadel CC, Talke SA, Horner-Devine AR, Jessup AT. 2011. Infrared-based measurements of velocity,
turbulent kinetic energy, and dissipation at the water surface in a tidal river. Geosci. Remote Sens. Lett.
IEEE 8:849–53
Cochrane JD, Kelly FJ. 1986. Low-frequency circulation on the Texas-Louisiana continental shelf. J. Geophys.
Res. 91:10645–59
Cole KL. 2014. A numerical study of the mid-field river plume. PhD Thesis, Texas A&M Univ., College Station

[Link] • Mixing and Transport in Coastal River Plumes 589

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

Craig PD, Banner ML. 1994. Modeling wave-enhanced turbulence in the ocean surface layer. J. Phys. Oceanogr.
24:2546–59
Csanady GT. 1977. The coastal jet conceptual model in the dynamics of shallow seas. In The Sea, ed. I
Goldberg, N O’Brian, JH Steele, pp. 117–44. New York: Wiley
De Boer GJ, Pietrzak JD, Winterwerp JC. 2006. On the vertical structure of the Rhine region of freshwater
influence. Ocean Dyn. 56:198–216
De Boer GJ, Pietrzak JD, Winterwerp JC. 2008. Using the potential energy anomaly equation to investigate
tidal straining and advection of stratification in a region of freshwater influence. Ocean Model. 22:1–11
De Boer GJ, Pietrzak JD, Winterwerp JC. 2009. SST observations of upwelling induced by tidal straining in
the Rhine ROFI. Cont. Shelf Res. 29:263–77
Donato TF, Marmorino GO. 2002. The surface morphology of a coastal gravity current. Cont. Shelf Res.
22:141–46
Dzwonkowski B, Yan X-H. 2005. Tracking of a Chesapeake Bay estuarine outflow plume with satellite-based
ocean color data. Cont. Shelf Res. 25:1942–58
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

Farmer DM, Armi L. 1986. Maximal two-layer exchange over a sill and through the combination of a sill and
contraction with barotropic flow. J. Fluid Mech. 164:53–76
Access provided by [Link] on 04/04/23. For personal use only.

Fisher NR, Simpson JH, Howarth MJ. 2002. Turbulent dissipation in the Rhine ROFI forced by tidal flow
and wind stress. J. Sea Res. 8:249–58
Fong DA, Geyer WR. 2001. Response of a river plume during an upwelling favorable wind event. J. Geophys.
Res. 106:1067–84
Fong DA, Geyer WR. 2002. The alongshore transport of freshwater in a surface-trapped river plume. J. Phys.
Oceanogr. 32:957–72
Garvine RW. 1984. Radial spreading of buoyant, surface plumes in coastal waters. J. Geophys. Res. 89:1989–96
Garvine RW. 1987. Estuary plumes and fronts in shelf waters: a layer model. J. Phys. Oceanogr. 17:1877–96
Garvine RW. 1995. A dynamical system for classifying buoyant coastal discharges. Cont. Shelf Res. 15:1585–
96
Garvine RW. 2001. The impact of model configuration in studies of buoyant coastal discharge. J. Mar. Res.
59:193–225
Gemmrich J. 2010. Strong turbulence in the wave crest region. J. Phys. Oceanogr. 40:583–95
Gerbi GP, Chant RJ, Wilkin JL. 2013. Breaking surface wave effects on river plume dynamics during upwelling-
favorable winds. J. Phys. Oceanogr. 43:1959–80
Geyer WR, Hill P, Milligan T, Traykovski P. 2000. The structure of the Eel River plume during floods. Cont.
Shelf Res. 20:2067–93
Geyer WR, Lavery AC, Scully ME, Trowbridge JH. 2010. Mixing by shear instability at high Reynolds
number. Geophys. Res. Lett. 37:L22607
Geyer WR, MacCready P. 2014. The estuarine circulation. Annu. Rev. Fluid Mech. 46:175–97
Halverson MJ, Palowicz R. 2008. Estuarine forcing of a river plume by river flow and tides. J. Geophys. Res.
113:C09033
Hetland RD. 2005. Relating river plume structure to vertical mixing. J. Phys. Oceanogr. 35:1667–88
Hetland RD. 2010. The effects of mixing and spreading on density in near-field river plumes. Dyn. Atmos.
Oceans 49:37–53
Hetland RD, DiMarco SF. 2012. Skill assessment of a hydrodynamic model of circulation over the Texas-
Louisiana continental shelf. Ocean Model. 43–44:64–76
Hetland RD, Hsu T-J. 2013. Freshwater and sediment dispersal in large river plumes. In Biogeochemical
Dynamics at Large River-Coastal Interfaces: Linkages with Global Climate Change, ed. TS Bianchi, MA
Allison, W-J Cai, pp. 55–85. New York: Springer
Hetland RD, MacDonald DG. 2008. Spreading in the near-field Merrimack River plume. Ocean Model. 21:12–
21
Hickey BM, Kudela RM, Nash JD, Bruland KW, Peterson WT, et al. 2010. River influences on shelf ecosys-
tems: introduction and synthesis. J. Geophys. Res. 115:C00B17
Hickey BM, Pietrafesa LJ, Jay DA, Boicourt WC. 1998. The Columbia River plume study: subtidal variability
in the velocity and salinity fields. J. Geophys. Res. 103:10339–68

590 Horner-Devine · Hetland · MacDonald

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

Horner-Devine AR. 2004. The dynamics of buoyant, rotational river plumes. PhD Thesis, Stanford Univ., Stan-
ford, CA
Horner-Devine AR. 2009. The bulge circulation in the Columbia River plume. Cont. Shelf Res. 29:234–51
Horner-Devine AR, Chickadel CC, MacDonald DG. 2013. Coherent structures and mixing at a river plume
front. In Coherent Flow Structures in Geophysical Flows at the Earth’s Surface, ed. J Venditti, J Best, M
Church, R Hardy, pp. 359–69. New York: Wiley
Horner-Devine AR, Fong DA, Monismith SG. 2008. Evidence for the inherent unsteadiness of a river plume:
satellite observations of the Niagara River discharge. Limnol. Oceanogr. 53:2731–37
Horner-Devine AR, Fong DA, Monismith SG, Maxworthy T. 2006. Laboratory experiments simulating a
coastal river discharge. J. Fluid Mech. 555:203–32
Horner-Devine AR, Jay DA, Orton PM, Spahn EY. 2009. A conceptual model of the strongly tidal Columbia
River plume. J. Mar. Sys. 78:460–75
Houghton RW, Chant RJ, Rice A, Tilburg C. 2009. Salt flux into coastal river plumes: dye studies in the
Delaware and Hudson River outflows. J. Mar. Res. 67:731–56
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

Houghton RW, Tilburg CE, Garvine RW, Fong A. 2004. Delaware River plume response to a strong
upwelling-favorable wind event. Geophys. Res. Lett. 31:L07302
Access provided by [Link] on 04/04/23. For personal use only.

Huq P. 2009. The role of Kelvin number on bulge formation from estuarine buoyant outflows. Estuaries Coasts
32:709–19
Imberger J, Ivey GN. 1991. On the nature of turbulence in a stratified fluid. Part II: Application to lakes.
J. Phys. Oceanogr. 21:659–80
Isobe A. 2005. Ballooning of river-plume bulge and its stabilization by tidal currents. J. Phys. Oceanogr.
35:2337–51
Itsweire EC, Koseff JR, Briggs DA, Ferziger JH. 1993. Turbulence in stratified shear flows: implications for
interpreting shear-induced mixing in the ocean. J. Phys. Oceanogr. 23:1508–22
Ivey GN, Imberger J. 1991. On the nature of turbulence in a stratified fluid. I: The energetics of mixing.
J. Phys. Oceanogr. 21:650–58
Ivey GN, Winters KB, Koseff JR. 2008. Density stratification, turbulence, but how much mixing? Annu. Rev.
Fluid Mech. 40:169–84
Jay DA, Pan J, Orton PM, Horner-Devine AR. 2009. Asymmetry of tidal plume fronts in an eastern boundary
current regime. J. Mar. Sys. 78:442–59
Jay DA, Zaron ED, Pan J. 2010. Initial expansion of the Columbia River tidal plume: theory and remote
sensing observations. J. Geophys. Res. 115:C00B15
Jirka GH, Adams EE, Stolzenbach KD. 1981. Buoyant surface jets. J. Hydraul. Div. Proc. ASCE
107(HY11):1467–87
Johnson DR, Weidemann A, Arnone R, Davis CO. 2001. Chesapeake Bay outflow plume and coastal upwelling
events: physical and optical properties. J. Geophys. Res. 106:11613–22
Jones GR, Nash JD, Doneker RL, Jirka GH. 2007. Buoyant surface discharges into water bodies. I: Flow
classification and prediction methodology. J. Hydraul. Eng. 133:1010–20
Kantha L, Phillips O, Azad R. 1977. On turbulent entrainment at a stable density interface. J. Fluid Mech.
79:753–68
Kashiwamura M, Yoshida S. 1978. Outflow dynamics at a river mouth. Proc. 16th Coast. Eng. Conf., pp. 2925–44.
Reston, VA: Am. Soc. Civil Eng.
Kilcher L, Nash JD. 2010. Structure and dynamics of the Columbia River tidal plume front. J. Geophys. Res.
115:C05S90
Kilcher L, Nash JD, Moum JN. 2012. The role of turbulence stress divergence in decelerating a river plume.
J. Geophys. Res. 117:C05032
Kirincich AR, Hebert D. 2005. The structure of the coastal density front at the outflow of Long Island Sound
during spring 2002. Cont. Shelf Res. 25:1097–114
Kourafalou VH, Oey L-Y, Want JD, Lee TN. 1996. The fate of river discharge on the continental shelf: 1.
Modeling the river plume and the inner shelf coastal current. J. Geophys. Res. 101:3415–34
Kudela RM, Horner-Devine AR, Banas NS, Hickey BM, Peterson TD, et al. 2010. Multiple trophic levels
fueled by recirculation in the Columbia River plume. Geophys. Res. Lett. 37:L18607

[Link] • Mixing and Transport in Coastal River Plumes 591

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

Lentz SJ. 2004. The response of buoyant coastal plumes to upwelling-favorable winds. J. Phys. Oceanogr.
34:2458–67
Lentz SJ, Helfrich KR. 2002. Buoyant gravity currents along a sloping bottom in a rotating fluid. J. Fluid
Mech. 464:251–78
Lentz SJ, Limeburner R. 1995. The Amazon River plume during AMASSEDS: spatial characteristics and
salinity variability. J. Geophys. Res. 100:2355–75
Liu Y, MacCready P, Hickey BM, Dever EP, Kosro PM, Banas NS. 2009. Evaluation of a coastal ocean
circulation model for the Columbia River plume in summer 2004. J. Geophys. Res. 114:C00B04
Luketina DA, Imberger J. 1987. Characteristics of a surface buoyant jet. J. Geophys. Res. 92:5435–47
MacCready P, Banas NS, Hickey BM, Dever EP, Liu Y. 2009. A model study of tide- and wind-induced
mixing in the Columbia River estuary and plume. Cont. Shelf Res. 29:278–91
MacCready P, Geyer WR. 2010. Advances in estuarine physics. Annu. Rev. Mar. Sci. 2:35–58
MacCready P, Rhines PB. 1993. Slippery bottom boundary layers on a slope. J. Phys. Oceanogr. 23:5–22
MacDonald DG, Carlson J, Goodman L. 2013. On the heterogeneity of stratified-shear turbulence: observa-
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

tions from a near-field river plume. J. Geophys. Res. 118:6223–37

MacDonald DG, Chen F. 2012. Enhancement of turbulence through lateral spreading in a stratified-shear
flow: development and assessment of a conceptual model. J. Geophys. Res. 117:C05025
Access provided by [Link] on 04/04/23. For personal use only.

MacDonald DG, Geyer WR. 2004. Turbulent energy production and entrainment at a highly stratified
estuarine front. J. Geophys. Res. 109:C05004
MacDonald DG, Geyer WR. 2005. Hydraulic control of a highly stratified estuarine front. J. Phys. Oceanogr.
35:374–87
MacDonald DG, Goodman L, Hetland RD. 2007. Turbulent dissipation in a near-field river plume: a com-
parison of control volume and microstructure observations with a numerical model. J. Geophys. Res.
112:C07026
Masse AK, Murthy CR. 1992. Analysis of the Niagara River plume dynamics. J. Geophys. Res. 97:2403–20
McCabe RM, Hickey BM, MacCready P. 2008. Observational estimates of entrainment and vertical salt flux
in the interior of a spreading river plume. J. Geophys. Res. 113:C08027
McCabe RM, MacCready P, Hickey BM. 2009. Ebb tide dynamics and spreading of a large river plume.
J. Phys. Oceanogr. 39:2839–56
Milliman JD, Farnsworth KL. 2011. River Discharge to the Coastal Ocean: A Global Synthesis. Cambridge, UK:
Cambridge Univ. Press
Milliman JD, Meade RH. 1983. World-wide delivery of river sediment to the ocean. J. Geol. 91:1–21
Moum JN, Gregg MC, Lien RC, Carr ME. 1995. Comparison of turbulence kinetic energy dissipation rate
estimates from two ocean microstructure profilers. J. Atmos. Ocean. Technol. 12:346–66
Murray SP. 1998. An observational study of the Mississippi-Atchafalaya coastal plume, final report. Rep., US Dep.
Inter., Miner. Manag. Serv., Gulf of Mexico OCS Reg.
Nash JD, Kilcher LF, Moum JN. 2009. Structure and composition of a strongly stratified, tidally pulsed river
plume. J. Geophys. Res. 114:C00B12
Nash JD, Moum JN. 2005. River plumes as a source of large-amplitude internal waves in the coastal ocean.
Nature 437:400–3
Nof D, Pichevin T. 2001. The ballooning of outflows. J. Phys. Oceanogr. 31:3045–58
O’Donnell J. 2010. The dynamics of estuary plumes and fronts. In Contemporary Issues in Estuarine Physics, ed.
A Valle Levinson, pp. 186–246. Cambridge, UK: Cambridge Univ. Press
O’Donnell J, Ackleson SG, Levine ER. 2008. On the spatial scales of a river plume. J. Geophys. Res. 113:C04017
Oey L-Y, Mellor GL. 1993. Subtidal variability of estuarine outflow, plume, and coastal current: a model
study. J. Phys. Oceanogr. 23:164–71
Orton PM, Jay DA. 2005. Observations at the tidal plume front of a high-volume river outflow. Geophys. Res.
Lett. 32:L11605
Pichevin T, Nof D. 1997. The momentum imbalance paradox. Tellus A 49:298–319
Pritchard M, Huntley DA. 2006. A simplified energy and mixing budget for a small river plume discharge.
J. Geophys. Res. 111:C03019
Rennie SE, Largier JL, Lentz SJ. 1999. Observations of a pulsed buoyancy current downstream of Chesapeake
Bay. J. Geophys. Res. 104:18227–40

592 Horner-Devine · Hetland · MacDonald

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

Sherman F, Imberger J, Corcos G. 1978. Turbulence and mixing in stably stratified waters. Annu. Rev. Fluid
Mech. 10:267–88
Simpson JH. 1997. Physical processes in the ROFI regime. J. Mar. Syst. 12:3–15
Simpson JH, Bos WG, Schirmer F, Souza AJ, Rippeth TP, et al. 1993. Periodic stratification in the Rhine
ROFI in the North Sea. Oceanol. Acta 16:23–32
Simpson JH, Crisp DJ, Hearn C. 1981. The shelf-sea fronts: implications of their existence and behaviour.
Philos. Trans. R. Soc. Lond. A 302:531–46
Smyth W, Moum JN. 2000. Length scales of turbulence in stably stratified mixing layers. Phys. Fluids 12:1327–
42
Smyth W, Moum JN, Caldwell DR. 2001. The efficiency of mixing in turbulent patches: inferences from
direct simulations and microstructure observations. J. Phys. Oceanogr. 31:1969–92
Souza AJ, Simpson JH. 1996. The modification of tidal ellipses by stratification in the Rhine ROFI. Cont. Shelf
Res. 16:997–1007
Stacey MT, Rippeth TP, Nash JD. 2011. Turbulence and stratification in estuaries and coastal seas. In Treatise
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

on Estuarine and Coastal Science, Vol. 2, ed. E Wolanski, D McLusky, pp. 9–35. Amsterdam: Elsevier
Tedford EW, Carpenter JR, Pawlowicz R, Pieters R, Lawrence GA. 2009. Observation and analysis of shear
Access provided by [Link] on 04/04/23. For personal use only.

instability in the Fraser River estuary. J. Geophys. Res. 114:C11006

Terray EA, Donelan MA, Agrawal YC, Drennan WM, Kahma KK, et al. 1996. Estimates of kinetic energy
dissipation under breaking waves. J. Phys. Oceanogr. 26:792–807
Thomson J. 2012. Wave breaking dissipation observed with SWIFT drifters. J. Atmos. Oceanic Technol.
29:1866–82
Thorpe SA. 1969. Experiments on the stability of stratified shear flows. J. Radio Sci. 4:1327–31
Thorpe SA. 1973. Experiments on instability and turbulence in a stratified shear flow. J. Fluid Mech. 61:731–
51
Tilburg C, Houghton RW, Garvine RW. 2007. Mixing of a dye tracer in the Delaware plume: comparison of
observations and simulations. J. Geophys. Res. 112:C12004
Trenberth KE, Smith L, Qian T, Dai A, Fasullo J. 2007. Estimates of the global water budget and its annual
cycle using observational and model data. J. Hydrometeorol. 8:758–69
Trowbridge JH. 1992. A simple description of the deepening and structure of a stably stratified flow driven
by a surface stress. J. Geophys. Res. 97:15529–43
Valle Levinson A, Li C, Royer TC, Atkinson LP. 1998. Flow patterns at the Chesapeake Bay entrance. Cont.
Shelf Res. 18:1157–77
Verspecht F, Rippeth TP, Howarth MJ, Souza AJ, Simpson JH, Burchard H. 2009. Processes impacting on
stratification in a region of freshwater influence: application to Liverpool Bay. J. Geophys. Res. 114:C11022
Visser AW, Souza AJ, Hessner K, Simpson JH. 1994. The effect of stratification on tidal current profiles in a
region of freshwater influence. Oceanol. Acta 17:369–81
Whitehead JA. 1985. The deflection of a baroclinic jet by a wall in a rotating fluid. J. Fluid Mech. 157:79–93
Whitney MM, Garvine RW. 2005. Wind influence on a coastal buoyant outflow. J. Geophys. Res. 110:C03014
Whitney MM, Garvine RW. 2006. Simulating the Delaware Bay buoyant outflow: comparison with observa-
tions. J. Phys. Oceanogr. 36:3–21
Winters KB, Lombard PN, Riley JJ, D’Asaro EA. 1995. Available potential energy and mixing in density-
stratified fluids. J. Fluid Mech. 289:115–28
Wiseman WJ Jr, Garvine RW. 1995. Plumes and coastal currents near large river mouths. Estuaries 18:509–
17
Wright LD, Coleman JM. 1971. Effluent expansion and interfacial mixing in the presence of a salt wedge,
Mississippi River delta. J. Geophys. Res. 76:8649–61
Wunsch C, Ferrari R. 2004. Vertical mixing, energy, and the general circulation of the oceans. Annu. Rev.
Fluid Mech. 36:281–314
Yankovsky AE, Chapman DC. 1997. A simple theory for the fate of buoyant coastal discharges. J. Phys.
Oceanogr. 27:1386–401
Yuan Y, Averner ME, Horner-Devine AR. 2011. A two-color optical method for determining layer thickness
in two interacting buoyant plumes. Exp. Fluids 50:1235–45

[Link] • Mixing and Transport in Coastal River Plumes 593

 FL47CH23-Horner-Devine ARI 21 November 2014 8:2

Yuan Y, Horner-Devine AR. 2013. Laboratory investigation of the impact of lateral spreading on buoyancy
flux in a river plume. J. Phys. Oceanogr. 43:2588–610
Zhang X, Hetland RD, Marta-Almeida M, DiMarco SF. 2012. A numerical investigation of the Mississippi
and Atchafalaya freshwater transport, filling and flushing times on the Texas-Louisiana Shelf. J. Geophys.
Res. 117:C11009
Zhang Z. 2013. Wind- and buoyancy-modulated along-shore circulation over the Texas-Louisiana shelf. PhD Thesis,
Texas A&M Univ., College Station
Zhang Z, Hetland RD. 2012. A numerical study on convergence of alongshore flows over the Texas-Louisiana
shelf. J. Geophys. Res. 117:C11010
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]
Access provided by [Link] on 04/04/23. For personal use only.

594 Horner-Devine · Hetland · MacDonald

 FL47-FrontMatter ARI 22 November 2014 11:57

Annual Review of
Fluid Mechanics

Contents Volume 47, 2015

Fluid Mechanics in Sommerfeld’s School

Michael Eckert p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 1
Discrete Element Method Simulations for Complex Granular Flows
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

Yu Guo and Jennifer Sinclair Curtis p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p21

Access provided by [Link] on 04/04/23. For personal use only.

Modeling the Rheology of Polymer Melts and Solutions

R.G. Larson and Priyanka S. Desai p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p47
Liquid Transfer in Printing Processes: Liquid Bridges with Moving
Contact Lines
Satish Kumar p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p67
Dissipation in Turbulent Flows
J. Christos Vassilicos p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p95
Floating Versus Sinking
Dominic Vella p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 115
Langrangian Coherent Structures
George Haller p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 137
Flows Driven by Libration, Precession, and Tides
Michael Le Bars, David Cébron, and Patrice Le Gal p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 163
Fountains in Industry and Nature
G.R. Hunt and H.C. Burridge p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 195
Acoustic Remote Sensing
David R. Dowling and Karim G. Sabra p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 221
Coalescence of Drops
H. Pirouz Kavehpour p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 245
Pilot-Wave Hydrodynamics
John W.M. Bush p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 269
Ignition, Liftoff, and Extinction of Gaseous Diffusion Flames
Amable Liñán, Marcos Vera, and Antonio L. Sánchez p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 293
The Clinical Assessment of Intraventricular Flows
Javier Bermejo, Pablo Martı́nez-Legazpi, and Juan C. del Álamo p p p p p p p p p p p p p p p p p p p p p p 315

v
 FL47-FrontMatter ARI 22 November 2014 11:57

Green Algae as Model Organisms for Biological Fluid Dynamics

Raymond E. Goldstein p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 343
Fluid Mechanics of Blood Clot Formation
Aaron L. Fogelson and Keith B. Neeves p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 377
Generation of Microbubbles with Applications to Industry and Medicine
Javier Rodrı́guez-Rodrı́guez, Alejandro Sevilla, Carlos Martı́nez-Bazán,
and José Manuel Gordillo p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 405
Beneath Our Feet: Strategies for Locomotion in Granular Media
A.E. Hosoi and Daniel I. Goldman p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 431
Sports Ballistics
Annu. Rev. Fluid Mech. 2015.47:569-594. Downloaded from [Link]

Christophe Clanet p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 455

Access provided by [Link] on 04/04/23. For personal use only.

Dynamic Stall in Pitching Airfoils: Aerodynamic Damping and

Compressibility Effects
Thomas C. Corke and Flint O. Thomas p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 479
Ocean Spray
Fabrice Veron p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 507
Stability of Constrained Capillary Surfaces
J.B. Bostwick and P.H. Steen p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 539
Mixing and Transport in Coastal River Plumes
Alexander R. Horner-Devine, Robert D. Hetland, and Daniel G. MacDonald p p p p p p p p p 569

Indexes

Cumulative Index of Contributing Authors, Volumes 1–47 p p p p p p p p p p p p p p p p p p p p p p p p p p p p 595

Cumulative Index of Article Titles, Volumes 1–47 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 605

Errata

An online log of corrections to Annual Review of Fluid Mechanics articles may be found
at [Link]

vi Contents