An Irminger Ring Mooring in the Labrador SeaAn Irminger Ring Mooring in the Labrador SeaA S Bower, H H Furey, and T McKee / Woods Hole Oceanographic Institution, Woods Hole MA 02534 / abower@whoi.edu

Introduction: Deep ocean convection is limited to a small number of isolated regions worldwide, including the Labrador Sea, but it has a profound impact on the ocean’s ther-mohaline circulation and climate. While the convec-tion process itself has been studied intensively over the last decade, the restrati�cation of the water column after convection, which will directly impact convection during subsequent winters, is not as well-studied. It has recently been suggested that the decay of coherent, long-lived, anticyclonic eddies shed from the warm, salty Irminger Current west of Greenland are potentially important in restratifying the interior Labrador Sea. The goal of this research is to collect new information on the initial structure of Irminger Rings and on the evolution of their core properties as they propagate across the Labrador Sea. We deployed one densely instrumented mooring in the northeastern Labrador Sea near the eddy forma-tion site in September 2007 to document the full water column hydrographic and velocity structure of new rings where they detach from the boundary. The mooring also serves as the “launch pad” for the auto-matic release of a pro�ling �oat into an eddy core each time an eddy sweeps by the mooring.

Summary: The sensor-driven SALP has released �oats using temperature and pressure criteria as expected, a testament to the usefulness of the SALP technology. Out of the eleven �oats that were on the original carousel, two remain to launch, and will be triggered as the next warm temperature anomaly passes by the mooring. A preliminary look at the pro�ling �oat data suggests that new anticyclonic eddies do contain a relatively freshwater cap. The mooring and its instruments will be recovered in September 2009, and the �oats will continue to circulate in the Labrador Sea for up to two more years. The complete data set will give us a new look at the charac-teristics of recently formed Irminger Rings, including the evolution of the Irminger Rings over time.

Reference: Lilly, J., P. Rhines, F. Schott, K. Lavender, J. Lazier, U. Send, and E. D’Asaro, 2003. Observations of the Labrador Sea Eddy �eld. Progress in Oceanography, 59(1):75-176.

Acknowledgements: The authors wish to thank Steve Riser and Dana Swift for the data acquition software, WHOI engineers Jim Valdes for the SALP instrument design, Rick Trask for mooring design, and Will Ostrom and the crew of the R/V Knorr for mooring deployment. This work was funded by the National Science Foundation Grant OCE-0623192.

1. Mooring DesignThe densely instrumented mooring (Figure 1) is designed to resolve the vertical structure of Irminger Rings. Seabird Microcats and Aan-deraa RCM-11 instruments are distributed over the 3000 meter mooring length, and measure salinity, temperature, and pressure, as well as current speed and direction.

WHOI engineers Jim Valdes and Dave Wellwood load the Sub-merged Autonomous Launch Platform (SALP) with APEX �oats (Figure 2). The prototype SALP is designed to launch �oats when triggered by passing temperature and pressure anomalies, which indicate that an Irminger Ring is passing by the mooring. After re-lease, the �oats pro�le once every �ve days, and drift at a nominal 300 meter ‘park depth’.

2. SALP AlgorithmThe SALP controllers are pre-programmed to release a �oat whenever a ring is detected at the mooring. Since the rings have a distinct warm core that is 1-1.5°C warmer than the background, temperature at the depth of the warm core (e.g., 500 m) is a robust indicator of a ring. But since the water property anomalies extend radially beyond the velocity core, there is some risk of a �oat being released on the eddy �ank and quickly separated from the eddy, or in a warm �lament. A more robust test that depends on both temperature and velocity ensure that �oats are released within the trapped �uid of a warm eddy core. The method takes advantage of the very strong relationship between vertically averaged ve-locity and mooring tow down. Figure 3 illustrates the ap-proximate vertical displacement of a mooring as a Gaussian-shaped eddy with a maximum vertically averaged azimuthal velocity of 50 cm/s at a core radius of 20 km passes the mooring at four di�erent radii. The two slices through the eddy’s velocity core (at Y = 0 and 10 km, heavy solid and dashed lines) result in two ~150-m tow-downs of the moor-ing, Figure 3c, whereas the two slices at the edge of and out-side the core (thin solid and dashed lines, Y = 20 and 30 km) produce a single tow-down. This tow-down is represented schematically in Figure 4, where the mooring tilts in the eddy velocity maxima, measuring a higher pressure, and the mooring is again upright in the quiet eddy interior.

Figure 5 shows results from a test of the �oat release algo-rithm on the Bravo mooring data from the central Labrador Sea: (a) potential temperature and (b) pressure at the 500-m instrument for one year during 1998-1999. Numbered hori-zontal bars indicate Irminger Ring events as identi�ed by Lilly et al. (2003). Open circles show hypothetical �oat re-leases based on our algorithm. Figure 5 shows that both single- and double-peaked pressure signatures were pres-ent in the Bravo mooring data when Irminger Rings passed by. Events 1, 5 and 6 have the predicted signature of an iso-lated Irminger Ring core: a double peak in pressure accom-panied by a single temperature maximum. Using our algo-rithm, �oats would have been correctly deployed within the cores of eddies 1, 5 and 6, as well as in the more complex events 2 and 3 (which represent two rings interacting).

3. Mooring PlacementIn addition to �netuning the mooring algorithm to release when an Irminger Ring passed over it, the mooring had to be placed in a location that would have the best chance of a Ring passing over it. Lilly et al. (2003) (Figure 6) identi�ed about 2000 eddy-like along-track SSH anomalies, with amplitude > 10 cm, in TOPEX/Poseidon altimetric data from the Labrador Sea during 1994-2000. A localized maximum in eddy observations is located in the northeastern Labrador Sea where the 3000 m isobath diverges from the west coast of Greenland, near 61.5°N, 52°W (Figure 6a). It is presumed to be the most common site of eddy for-mation, which is thought to be triggered by destabili-zation of the boundary current as it �ows over the steep continental slope there. Figure 6a further shows an extension of the highest anomaly concen-tration southward along a relatively narrow corridor toward the Bravo site.

We extended the analysis of the altimetric data for the 2001-2006 time period to document any changes in eddy pathways since the Lilly study. Figure 7 shows the two dominant pathways of anticyclones that origi-nated along the west coast of Greenland for the 2001-2006 time period. These anticyclonic anomalies were identi�ed manually from daily satellite images to create continuous eddy trajectories. The eddies that left the boundary current and travelled into the inte-rior Labrador Sea (red) further strengthened our deci-sion to place the mooring at its current site.

4. SALP and APEX Float PerformanceEight APEX �oats released from the mooring, and one from the ship during the de-ployment cruise, are shown in Figure 8, which illustrates the �oat launch date, dura-tion and mode of release. All �oats but one have been released as programmed and are still pro�ling. One �oat failed to launch for unknown reasons. Initially, �oats were programmed to release when both temperature and pressure criteria were met. After 9 months, at the start of the second fall season (2008), the SALP was pro-grammed to relax criteria to a simple temperature criterion. Two �oats remain to launch, and will release when the SALP next registers a temperature of over 5˚C.

Trajectories of the nine �oats launched to date (Figure 9). The longest trajectory (�rst launched) is shown in upper left, and the most recent release in the lower right. The trajectory has been color-coded by temperature measured by the �oat at its park depth, nominally 300 meters. Pro�le locations are shown as open black circles under-neath the color trajectory. The mooring site/ launch location is shown as a circle-x, and release date and �oat number in the lower right corner of each subplot. Three �oats (5261 5266 5271) have boxes that designate the location of eddy observation shown in Figures 10 and 11.

5. Freshwater cap on anticyclonesInitial results show that anticyclones do trap relatively fresher water in their surface layer. An anticyclone tagged by APEX �oat 5266 (Figure 10) in the northeast section of the Labrador Sea between the 1000 and 2000 meter isobath can be seen in a series of sea level anomaly (SLA) images from April through June 2008. The �oat remained with the eddy for approximately 25 days, from 4 May 2008 through 29 May 2008. The chronological series of plots shows the decrease in upper layer (0-100 meter) mean salinity as the �oat moves into the anticyclone, makes one loop, and then the subse-quent increase in salinity as the �oat leaves the anticyclone on 8 June. This decrease of upper layer salinity in an eddy suggests the generation of a freshwater capped an-ticyclone. This eddy may eventually bring its cap of relatively fresh water into the in-terior Labrador Sea, although the �oat did not “stick” with the eddy.

A similar event (Figure 11), but near the mooring site, with two �oats entrained. In this series of SLA images, only the current positions of the �oats and 10 day �oat tails are depicted, for clarity. At least one �oat (5271) is entrained in the outer �ank of an anticyclone as it passes over the mooring site. The salinity thresholds are higher, but the trend is similar: The upper 100 meters of the water pro�le con-tains lower average salinity than outside the eddy.

The �oats released from the mooring are programmed to pro�le every �ve days and park at 300 m between pro-�les, whereas ARGO �oats in the area pro�le every 15 days and are parked at 1000 m. The di�erent sampling scheme was chosen to ensure that the APEX �oats are en-gaged with mesoscale activ-ity of the upper ocean.

Figure 1

X (km)

Y (k

m)

−75 −50 −25 0 25 50 75−75

−50

−25

0

25

50

75

−75 −50 −25 0 25 50 750

10

20

30

40

50

X (km) (or time)

spee

d (c

m/s

)

−75 −50 −25 0 25 50 750

50

100

150

X (km) (or time)

moo

ring

disp

l. (m

)

Figure 3

3

3.5

4

4.5

thet

a (5

00 m

)

1 2 3 4 5 6

0 50 100 150 200 250 300 350 400400

600

800

1000

pres

sure

(500

m)

time (days)

1 2 3 4 5 6

Figure 5

Figure 2

×

×

×

50 m150 m

600 m500 m

3000 m

Figure 4

−60 −58 −56 −54 −52 −50 −48 −46 −44 −60 −58 −56 −54 −52 −50 −48 −46 −44

54

56

58

60

62

54

56

58

60

62

Dis

tanc

e no

rth

(100

km

)

−6

−5

−4

−3

−2

−1

0

1

2

3

4

5

6

Distance east (100 km)

Dis

tanc

e no

rth

(100

km

)

−5 −4 −3 −2 −1 0 1 2 3 4 −6

−5

−4

−3

−2

−1

0

1

2

3

4

5

Distance east (100 km)−5 −4 −3 −2 −1 0 1 2 3 4 5

Perc

ent t

ime

insi

de e

vent

cor

e (×

2 fo

r a)

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

All Eddies Dipoles

Anticyclones Cyclones

(a) (b)

(c) (d)

Figure 6

64N

62N

60N

58N

56N

54N

50W54W 48W52W 46W

GREENLAND

NFLD

BRAVO

AR7W

Mooring Site

1000 m20

00 m

3000 m

Figure 7

52635266

52725264

527152615269

52655267

Irminger Rings Float Duration Chart

Jul Sep Nov Jan Mar May Jul Sep Nov Jan Mar May 2007 2008 2009

Deployed from ship.

T/P criteria release.

T/P criteria release.

Timed release, 6 months.

T criterion release.

}Release criteria relaxed from temperature (T) and pressure (P), to temperature only.

Release criteria for carousels modified to allow for independent launch.

T criterion release.

Figure 8

56 oN

58 oN

60 oN

62 oN

64 oN

66 oN

52632007−09−26

52662007−12−08

52722008−02−04

56 oN

58 oN

60 oN

62 oN

64 oN

66 oN

52642008−03−22

52712008−06−03

52612008−12−28

60 oW 56 oW 52 oW 48 oW 44 oW

56 oN

58 oN

60 oN

62 oN

64 oN

66 oN

52692008−12−28

60 oW 56 oW 52 oW 48 oW 44 oW

52652009−03−07

60 oW 56 oW 52 oW 48 oW 44 oW

52672009−03−07

2 4 60

1000 m

3000

m

2000 m

Temperature (C)

Figure 9

−20 −15 −10 −5 0 5 10 15 20sea level anomaly (centimeters)

Float location on date of SLA image

Averaged Salinity of upper 100 meters

S >= 33.5 & S < 34.0

S >= 34.0

Salinity (S) < 33.5

5266

57 oW 56 oW 55 oW 54 oW 53 oW 52 oW 51 oW 50 oW

62 oN

63 oN

64 oN

65 oN

5266 / 2008.04.24

1000 m

2000 m

a)

57 oW 56 oW 55 oW 54 oW 53 oW 52 oW 51 oW 50 oW

62 oN

63 oN

64 oN

65 oN

5266 / 2008.05.04 b)

57 oW 56 oW 55 oW 54 oW 53 oW 52 oW 51 oW 50 oW

62 oN

63 oN

64 oN

65 oN

5266 / 2008.05.14 c)

57 oW 56 oW 55 oW 54 oW 53 oW 52 oW 51 oW 50 oW

62 oN

63 oN

64 oN

65 oN

5266 / 2008.05.24 d)

57 oW 56 oW 55 oW 54 oW 53 oW 52 oW 51 oW 50 oW

62 oN

63 oN

64 oN

65 oN

5266 / 2008.05.29 e)

57 oW 56 oW 55 oW 54 oW 53 oW 52 oW 51 oW 50 oW

62 oN

63 oN

64 oN

65 oN

5266 / 2008.06.08 f )

57 oW 56 oW 55 oW 54 oW 53 oW 52 oW 51 oW 50 oW

62 oN

63 oN

64 oN

65 oN

5266 / 2008.06.19 g)

55 oW 54 oW 53 oW 52 oW 51 oW 50 oW

60 oN

61 oN

62 oN

5271 and 5261 / 2009.02.19

55 oW 54 oW 53 oW 52 oW 51 oW 50 oW

60 oN

61 oN

62 oN

5271 and 5261 / 2009.02.14

55 oW 54 oW 53 oW 52 oW 51 oW 50 oW

60 oN

61 oN

62 oN

5271 and 5261 / 2009 .02.09

55 oW 54 oW 53 oW 52 oW 51 oW 50 oW

60 oN

61 oN

62 oN

5271 and 5261 / 2009.02.04

55 oW 54 oW 53 oW 52 oW 51 oW 50 oW

60 oN

61 oN

62 oN

5271 and 5261 / 2009.01.29

55 oW 54 oW 53 oW 52 oW 51 oW 50 oW

60 oN

61 oN

62 oN

5271 and 5261 / 2009.01.24

−20 −15 −10 −5 0 5 10 15 20sea level anomaly (centimeters)

Float location on date of SLA image

Averaged Salinity of upper 100 meters

S >= 34.3 & S < 34.5

S >= 34.5

Salinity (S) < 34.3

Float 5261Float 5271

a)

b)

c)

d)

e)

f )

Figure 11

Figure 10

April 2009, HHF.