Components
of the Gravity Current
Entrainment CPT
Summary: Table of Observations
Observationalists' Table of Plumes: Compiled by Arnold Gordon, Jim
Price, Hartmut Peters and James Girton
For excel version click
here.
References:
[1] Hartmut Peters: The outflow from
the Red Sea has a strong seasonal cycle with a summer minimum (at
times a complete shut-down) and a winter maximum.
The Red Sea waters flows down to depth in 2 channels, the Northern
Channel being `130 km long and typically only 5 km wide at the level
of the Red Sea water. It produces the densest water. The Souther
Channel is wider and shorter and produces less dense waters than
the Northern Channel.
The data provided are for the NORTHERN CHANNEL in WINTER from the
2001 REDSOX Experiment. The comments also provide data from the REDSOX
SUMMER cruise. The observations are insufficient to iallow ncluding
ranges,
[2] Arnold L. Gordon: the water just
before descent [that is: at the sill]
[3] Hartmut Peters: The source region,
the Strait of Bab el Mandeb (BAM), stretches some 120 km from the
Hanish Sill (shallowest, 150 m depth) to the Perim narrows, a constriction.
BAM has substantial longstream salinity gradients. There is mixing
in BAM owing to strong tides.
[4] Girton: 3 water types with distinct
T-S properties are present in the dense layer. I've given mean values
for the layer as a whole as well as ranges spanning all 3 types.
[5] Hartmut Peters: SUMMER: 20.7 (20.2-21.0)
[6] Hartmut Peters: SUMMER: 39.1 (38.8-39.3)
The maximum salinity at the Perim Narrows apparently
can at times increase to thevalue of 40.6 of red Sea Deep Water.
[7] Arnold L. Gordon: not sure what density to use: sigma-0, sigma-p,
where p = sill depth; or some sort of neutral density?
[8] girton: sigma-0
[9] Hartmut Peters: SUMMER: 27.44 (27.19-27.73)
If pure Red Sea Deep Water appeared, its pot. density
would be 28.6.
[10] Arnold L. Gordon: this may duplicate 'sill depth'
entry below
[11] Girton: center of mass depth at sill (i.e.,
100m above the bottom)
[12] Hartmut Peters: The Hanish Sill, the shallowest part of Bab
el Madeb, has a depth of 150 m.
[13] Arnold L. Gordon: at the point that entrainment
ceases [plume becomes a 'normal' quasi-isobath following flow].
[14] Hartmut Peters: The outflow has a distinct vertical
structure with a mixed bottom layer which does not get much thicker
along the outflow and a stratified and sheared interfacial layer,
which does become thicker along the plume path, which is diluted
along the path much more than the bottom layer, and which carries
the bulk of the plume transport toward the bottom end of the two
outflow channels.
The table lists the respective mean values
in the bottom layer and the whole plume as "BL / WPL".
The data are from (only) 4 casts from the NORTHERN
CHANNEL just BEFORE the final equilibration. The indicated range
occurred over 6 days.
[15] Hartmut Peters: I do not know how to quantify
the highly variable, sputtering SUMMER outflow. SOUTHERN CHANNEL:
21.5-22.5
[16] Hartmut Peters: SOUTHERN CHANNEL: 38.5-39.1
[17] Hartmut Peters: SOUTHERN CHANNEL: 26.9-27.25
[18]Hartmut Peters: SOUTHERN CHANNEL: 400-600 m
[19] Arnold L. Gordon: the water just above the cap
of the plume…i.e what is being
entrained.
[20] Girton: Here the range is over the depth interval
where entrainment is occurring, rather than temporal variability
[21] Girton: Note that there is also a thin layer of low salinity
water (down to 34.5) at the plume interface which may be receiving
lateral input from the Greenland shelf, so representative salinity
numbers are tricky. For the most part, numbers here represent the
background Atlantic water.
[22]Girton: Really we need a profile of each of these properties.
[23] Hartmut Peters: The data given are from the 2001 REDSOX experiment.
The long-term mean transport at the Perim narrows is about 0.4 Sv,
with winter values of about 0.3-0.7 Sv and small to vanishing summer
flow.
[24] Girton: Denser than 27.8 sigma-0
[25] Girton: These are mean velocities over the plume cross-section.
Within a profile values can be higher. The range represents variability
in time and along the overflow path. 0.7 is a decent average in the
entrainment region but not at the sill, where the value is more like
0.3.
[26] Hartmut Peters: The velocity is the mean
for the mixed bottom layer.
[27] Hartmut Peters: The plume appears to seize
almost immediately after exiting the channels, at which point the
plume is equilibrating.
[28] Arnold L. Gordon: from sea floor to mid-way
in the uppermost cap [where the T/S properties begin to deviate
from the ambient stratification]
[29] Hartmut Peters: The thickness is bottom
layer + 1/2 interfacial layer. (BL + 1/2 IL)
As the IL is much thicker than the BL in the Red
Sea outflow, and as even the top of the IL is flowing and contributing
to the plume transport, the above is not the most appropriate definition.
[30] Girton: Hmm, this is tough… Lots of variation
both in time and across the plume. We need a way to standardize.
A related and possibly more robust quantity is Cross-sectional area:
4 km^2 (range 2-15)
[31] Hartmut Peters: This is from a single cast. No tidal average
is avialable from the REDSOX winter cruise.
[32] Girton: "source" and "product" values
presented here are at the sill and after entrainment has more-or-less
ceased, respectively. In between, there is another region in which
the overflow has thinned due to acceleration and a mean thickness
is closer to 120m. Once the major entrainment starts, mean thickness
is back to 200m. These along-path changes are likely due to a combination
of hydraulics, entrainment, and (perhaps most importantly) cross-stream
slope changes.
[33] Hartmut Peters: Note: The plume really is much thicker than
this.
[34] James Girton: Lots of variability, but often
about half of the overflow (defined by property anomalies) is strongly
stratified
[35] Hartmut Peters: This is from a single cast.
No tidal average available for winter
conditions from REDSOX.
[36] Hartmut Peters: 120 m is typical for winter/REDSOX.
The great spread is due to a slow-down of the plume over the 8 days
of observations. The bottom layer thickness for the "product" is
70 (40-90)
[37] Arnold L. Gordon: I asked Mile McPhee for his
thoughts of how to express the importnace of the thermobaric effect
[compessibility dependence on temp] to the gravity current in question.
[38] Arnold L. Gordon: the top of the plume [gravity
current] is mid-way in the uppermost cap that marked a deviation
from the ambient stratification
[39] Girton: This value is in the main entrainment region. Sill
value is 4 x 10-3, value in the "thin" region is is 3 x
10-3, product value is 1 x 10-3.
[40] Hartmut Peters: This is for the source density, which has an
excess of 1.5 kg/m^3 over the background density profile at the source
water depth.
[41] Girton: The minimum is at the sill and the maximum
value is in the subsequent "thin" regime
just before the main entrainment region. After major entrainment
begins, thickening plus dilution drops F slightly to 1.1. The product
water value after some deceleration and more dilution is more like
0.9.
Note that these values come from plume averaged velocities and have
not had an upper layer velocity subtracted. This should probably
be done, which will likely reduce the Froude numbers due to the often
substantial barotropic component.
[42] Hartmut Peters: This computed with the plume depth as BL+1/2
IL, with the avg plume density (over BL+IL), and the density in a
30-m thick layer just above the plume.
[43] Hartmut Peters: The plume has many occurrences
of Ri<1/4, but Ri varies a lot.
[44] Girton: Downslope angle in radians changes with
slope (0.01-0.03) while rate of descent (slope along overflow path)
is pretty constant at 0.006.
[45] Girton: More like 6 x 10-5 near sill.
[46] Hartmut Peters: This is computed from the transports.
[47] Girton: 1.5 layer model (sqrt(g'h)/f)
[48] Arnold L. Gordon: Rd of the benthic layer [plume]
[49] Hartmut Peters: Computed as NH/f with N and H=120 m from the
product conditions. N=1e-5s^-2
[50] Hartmut Peters: The average slope of the channels
is 1/3 deg. The length of the Northern Canyon is ~130 km.
[51] Hartmut Peters: Hanish Sill - 120 km upstream from the Perim
narrows.
[52] Arnold L. Gordon: depth below regions sea floor
[53] Hartmut Peters: Perim Narrows
[54] Arnold L. Gordon: width of crests of the canyon
walls
[55]Girton: 0.25 in upstream basinl
[56] Hartmut Peters: The typical tidal currents in the Perim Narrows
are ~0.8 m/s during spring tide, much less during neap tides.
From Bab el Mandeb out into the outflow region the tidal currents
drop to a few cm/s owing to the funnel-shaped topography of the area.
|
More Announcements