the 150khz adcp-equipped oleander measures upper ocean currents to ~250m in gs. new 75khz system...

8
The 150kHz ADCP- equipped Oleander measures upper ocean currents to ~250m in GS. New 75kHz system The Oleander section: 1992 to today...

Upload: sophia-murphy

Post on 27-Mar-2015

213 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: The 150kHz ADCP-equipped Oleander measures upper ocean currents to ~250m in GS. New 75kHz system reaches to ~600m. The Oleander section: 1992 to today

The 150kHz ADCP-equipped Oleander

measures upper ocean currents to ~250m in GS.

New 75kHz system reaches to ~600m.

The Oleander section: 1992 to today...

Page 2: The 150kHz ADCP-equipped Oleander measures upper ocean currents to ~250m in GS. New 75kHz system reaches to ~600m. The Oleander section: 1992 to today

Example of a good weather (no bubbles)

ADCP section. The data are uploaded

when the vessel docks in Port Elisabeth. We are striving to serve

the final product within a few days.

cont. slo

pe

backscatter limit

Direct measurements of currents provide highly accurate information on currents, their structure

and space-time variability. In absolute units, no assumptions about geostrophy, and/or

reference velocities.

Page 3: The 150kHz ADCP-equipped Oleander measures upper ocean currents to ~250m in GS. New 75kHz system reaches to ~600m. The Oleander section: 1992 to today

The non-dimensional Ekman spiral as derived

from Oleander ADCP measurements and NCEP winds and heat fluxes. The numbers indicate

depth normalized by the Price-Sundermeyer

Ekman scale depth. The axes are E-W and N-S velocity normalized by windstress and Ekman layer depth. Given that

the actual Ekman velocities are only a few

cm/s this is an impressive result!

(S. Stoermer, MS-thesis, GSO/URI)

Page 4: The 150kHz ADCP-equipped Oleander measures upper ocean currents to ~250m in GS. New 75kHz system reaches to ~600m. The Oleander section: 1992 to today

Detection and quantification of coherent

vortices in the SS.The main point of this figure is that repeat sampling allows for

quantitative analysis of the mesoscale eddy

velocity field.

Repeat detection of an eddy

Distribution of core vorticity across the SS. (D. Luce, MS-thesis, GSO/URI)

Page 5: The 150kHz ADCP-equipped Oleander measures upper ocean currents to ~250m in GS. New 75kHz system reaches to ~600m. The Oleander section: 1992 to today

The red lines highlight the

GS, Slope and Sargasso Sea

transports over the 11 years.

Integration of velocities across the ship track gives transport in the Slope Sea, Gulf Stream and Sargasso

Seas as a function of time. (Transport in 1 m thick layer at 52 m.) .

Page 6: The 150kHz ADCP-equipped Oleander measures upper ocean currents to ~250m in GS. New 75kHz system reaches to ~600m. The Oleander section: 1992 to today

We have written that the mean path of the

GS may be controlled by Slope Sea transport and

thus has a thermohaline

(Labrador Sea) origin, but is

considered an open question.

-The large variations in the Sargasso Sea very likely have a

wind-driven explanation.

SW

NE

SW

Page 7: The 150kHz ADCP-equipped Oleander measures upper ocean currents to ~250m in GS. New 75kHz system reaches to ~600m. The Oleander section: 1992 to today

Time-varying fluxes from the Labrador region are very likely responsible not only for path (top) but

also for SSS variations. But exactly how is not clear.

data

court

esy

NM

FS

Page 8: The 150kHz ADCP-equipped Oleander measures upper ocean currents to ~250m in GS. New 75kHz system reaches to ~600m. The Oleander section: 1992 to today

At this juncture the main signal to emerge is the constancy of GS transport yet x2 variations in both the Slope and Sargasso Seas. Why is GS transport so stable but not that of surrounding waters which

in part feed it?

By September we will have been in operation for 15 years. This summer we plan to submit a proposal to continue for another five years. The deeper ADCP

reaches well below the main pycnocline in the Slope Sea so we can measure directly upper Labrador Sea water velocities and transport.

Similarly for the Gulf Stream and Sargasso Sea (600+m).

This ongoing effort has been possible by the excellent cooperation of the BCL and with our NMFS colleagues across the street at home. They operate

the XBT and TSG program. have been and are terrific partners.