Bush, J.W.M., Stone, H.A., and Bloxham, J., 1992. The Motion
of an Inviscid Drop in a Bounded Rotating Fluid, Phys. Fluids A,
4
(6), 11421147.
The motion of a buoyant inviscid drop rising vertically
along the rotation axis of a rapidly rotating low viscosity fluid bounded
above and below by rigid horizontal boundaries is considered in the case
that the drop is circumscribed by a Taylor column which spans the entire
fluid depth. Both the shape and steady rise speed of the drop are deduced
as a function of its interfacial tension. The analysis demonstrates
that the drop assumes the form of the prolate ellipsoidal figure of revolution
which would obtain in the absence of any relative motion in the surrounding
fluid. The hydrodynamic drag on the drop follows simply from the analysis
of Moore and Saffman, who considered the equivalent motion of a rigid particle.
The rise speed of an inviscid drop is generally half that of an identically
shaped rigid particle; in particular, the rise speed of a spherical
inviscid drop is 0.44 that of a rigid sphere.

Woods, A.W. and Bush, J.W.M., 1999. The dimensions and dynamics
of megaplumes, J. Geophys. Res.,
104, 2049520507.
We investigate the generation of megaplumes by the
release of buoyant hydrothermal fluid from the seafloor. We show that
megaplumes may be generated from various modes of venting,
including both the instantaneous and continuous release of hydrothermal
effluent from either a point or line source. The hydrothermal effluent
forms a buoyant plume which rises through the water column to its neutral
buoyancy height and then intrudes laterally to form a neutral cloud. Owing
to the influence of the Earth's rotation, whose magnitude is omega =
f / 2 , the neutral cloud eventually becomes unstable, giving rise
to geostrophic vortices that propagate away from the source. By combining
the scaling laws governing turbulent plumes and geostrophic vortices, we
establish new relationships between the megaplume geometry and
the source conditions.
We find that megaplumes whose radius greatly exceeds
their height of rise are formed from sources which persist for at least
several days, since, in the deep ocean, the radii of eddies produced by
short lived releases of buoyant fluid are comparable to their rise height.
Our model predicts the total buoyancy, B, of the hydrothermal
effluent released in forming such megaplume structures. For a maintained
source, B ~ N^{2}R^{2}hH ,
where H is the height of the plume in the water column, R
and h are the megaplume radius and halfdepth and N is the
BruntVaisala frequency characterising the oceanic stratification.
Laboratory experiments suggest that the constant of proportionality depends
on the geometry of the source and has a value of 0.8 for a linear source
and 0.5 for a localised source. We also find that for a longlived source
of fluid, there is a maximum megaplume size, R = HN / f , and that
the buoyancy contained within this maximal megaplume scales as B
~ H^{4}N^{4 }/ f ^{2 }.
Finally, we calculate the total megaplume heat content
in terms of the total buoyancy release and the thermal anomaly of the megaplume,
by considering the effects of the ambient stratification in both temperature
and salinity on plume properties. Applying the model to data from three
historic megaplume events at the Juan de Fuca ridge, we estimate that the
EP86
and EP93a,b megaplumes were produced by sources of effluent which
persisted for times in excess of 10^{6}, 10^{5} and 10^{5}
s, and that the total heat released was approximately 4*10^{16},
2*10^{15} and 1*10^{15} J, respectively.
0, owing to the long retention time of
fluid within the bodies. For highly permeable twodimensional bodies,
D_{xx}
= alpha (C_{yy} + 1)
U L, where C_{yy}
is the addedmass coefficient characterising the flow around an impermeable
body moving parallel to the yaxis. Consequently, dispersion
by highly permeable bodies is enhanced when the bodies are slender, in
contrast to the low permeability limit.
The influence of finite tracer diffusivity on longitudinal
dispersion is also considered and demonstrated to make a negligible contribution
when kappa > 0 provided Pe >> max(1, 1 / kappa)$ and
for impermeable bodies provided Pe >> 1 . The coefficient
of longitudinal dispersion for Pe << 1 is calculated using
Maxwell's (1873) method. When the body is impermeable, the longitudinal
dispersivity is D_{xx} = (1  (C_{xx}+1)
alpha
)
D_{2
}, where D_{2} is the molecular diffusivity outside
the body. Here the influence of body bluffness is to reduce longitudinal
dispersion, which is opposite to the high Pe regime.
>

Bush, J.W.M. and Hasha, A.E., 2004. On the collision of
laminar jets: fluid chains and fishbones,
J. Fluid Mech., 511, 285310.
We present the results of a combined experimental and theoretical
investigation of the family of free surface flows
generated by obliquely colliding laminar jets. We present a parameter study
of the flow, and describe the rich variety of forms observed.
When the jet Reynolds number is sufficiently high, the jet collision generates
a thin fluid sheet that evolves under
the combined influence of surface tension and fluid inertia. The resulting
flow may take the form of a fluid chain: a succession of mutually orthogonal
links, each composed of a thin oval film bound by relatively thick fluid rims.
The dependence of the form of the fluid chains on the governing parameters
is examined experimentally. An accompanying theoretical model describing
the form of a fluid sheet bound by stable rims is found
to yield good agreement with the observed chain shapes.
In another parameter regime, the fluid chain structure becomes unstable,
giving rise to a striking new flow structure ressembling fluid fishbones.
The fishbones are demonstrated to be the result of a RayleighPlateau instability
of the sheet's bounding rims amplified by the centripetal force associated with
the flow along the curved rims.

Peacock, T., Blanchette, F. and Bush, J.W.M., 2005. The stratified Boycott effect.
J. Fluid Mech., 529, 3349.
We present the results of an experimental investigation of the
flows generated by monodisperse particles settling in a
stratified ambient with an inclined sidewall. In this configuration, upwelling
beneath the inclined wall associated with the Boycott effect is opposed by
the ambient density stratification. For sufficiently weak stratification, the
Boycott layer transports dense fluid from the bottom to the top
of the system. Subsequently, the upper clear layer is mixed by the convective
motions resulting from the combined influence of the newly emplaced dense saline
fluid and the buoyant fluid released by the particle settling at the top of
the suspension. For sufficiently strong stratification, layering
occurs.
Within each layer, convection erodes the initially linear density gradient, generating a
steplike density profile throughout the system. Particles are transported across
the discrete density jumps between layers by millimetric plumes of particleladen
fluid. Mixing models are developed to describe the evolution of particle concentration
and salinity in both the low and high stratification limits, and geophysical
applications are discussed.
