Vous êtes dans :

North-Atlantic Temperature and salinity

A summary of recent conditions in the North Atlantic is updated annually. It ihas been established using the ARGO global observing system based on profiling floats, complemented by CTDs and mooring data available at Coriolis datacenter.
Near surface, the year 2012 appears as an extreme in the 2002-2012 decade for the cold winter observed in the Labrador Sea and the Irminger Sea where temperatures went well below the climatological mean. The warm summer that extends over most of the basin except in the South-East part is clearly seen in the surface cycles but remains within the variability of this period. A negative salinity anomaly is building up west of 20°W. It started in 2009 in the centre of the basin, in 2012, a large fresh anomaly, 0.5 PSS below the ‘normal’ WOA05 conditions, is observed (see illustration).
At depth (1000 m), the main points already noted in the previous years are confirmed:
The Greenland Sea, the Labrador Sea and the Irminger Sea are warmer than normal and this is a clear tendency since 2002.
The Mediterranean Outflow water seems warmer and saltier west of the Iberian peninsula.

ISAS: gridded temperature and salinity fields

The ARGO network of profiling floats has been set up to monitor the large-scale global ocean variability (http://www.argo.ucsd.edu/ ) . Argo data are transmitted in real time and hastily made available by the two Global Data Assembly Centres (Argo-GDAC). Delayed mode data undergo expert calibration processes and are delivered later.  In the North Atlantic, the temperature and salinity conditions of the upper 2000 m are adequately described since 2002. This dataset is thus suitable for an overview of the oceanographic conditions in this basin, giving the general context for the repeat stations and sections collected mostly at the periphery of the basin by the partners of the ICES Working Group on Ocean Hydrography (WGOH).

Temperature and salinity fields are estimated on a regular half degrees (Mercator scale) grid using the In Situ Analysis System (ISAS), (Gaillard et al., 2009). The dataset is downloaded from the Coriolis Argo GDAC (http://www.coriolis.eu.org/). It should be noted that Coriolis assembles many types of data transmitted in real time, merging the ARGO data set with data collected by the GTS such as mooring data, marine animals, gliders, CTDs. However, the ARGO dataset remains the main contributor in the open ocean. The last year of the analyzed series uses the Near Real Time dataset prepared by Coriolis at the end of each month from real time data. Delayed mode data are progressively taken into account for the previous years, replacing the NRT data.

Data are pre-processed before entering the analysis. First we perform a climatological test to detect outliers then we vertically interpolate the profiles on 152 standard levels between the surface and 2000m. The analysis to produce gridded fields is performed at each standard level independently. The method is based on optimal estimation principles and includes a horizontal smoothing through specified covariance scales.  The results presented here were produced with version 6 of ISAS (Gaillard, 2012). The reference state was computed as the mean of a 2004-2010 analysis (D2CA1S2) and the a priori variances were computed from the same dataset.  The period 2002-2012 was fully reprocessed to take into account new delayed mode data and flags.

Near surface seasonal cycle

During winter, the near surface waters were anomalously cold and fresh in the North-West Labrador Sea and over a large area south of Greenland. The cold temperatures in the Labrador Sea were associated with a strengthening of North-easterly winds. Further South, waters were extremely warm and salty in the western basin along 40°N, indicating a northward shift of the Gulf Stream. The rest of the basin was still slightly warmer and saltier than normal.

Summer 2012 has been very warm over most of the basin, west of a line joining Spitzbergen to 40N-40W. It was only moderately warm South-East of the Iberian peninsula and slightly colder than normal in the central Atlantic and off Ireland.  Unlike winter, summer salinity anomalies were not correlated to temperature anomalies. Waters were very salty in the Greenland Sea/Norvegian Sea and along the East Greenland coast. They were fresh along the western boundary: starting from the West Greenland coast, following the North American coast and from there, extending toward the west.

ICES_seasanoTEMP_2012

ICES_seasanoPSAL_2012

The year 2012 appears as an extreme in the 2002-2012 decade for the cold winter observed in the Labrador Sea and the Irminger Sea, (Figure 2) where temperatures went well below the climatological mean (nearly 2° lower in the Labrador Sea) and the warm temperature in the south-west part of the basin. The warm summer that extends over most of the basin except in the South-East part is clearly seen in the surface cycles but remains within the variability of this period.

cycle_Labrador-N cycle_Irminger cycle_W60N40 cycle_W20N50

Winter mixed Layer

Winter surface conditions determine the mixed layer properties.  In order to compare all areas over the decade, we adopt a simple definition for the mixed layer depth, using the level at which temperature changes by more than 0.5°C with respect to the 10 meter depth. The month of February is selected as the common period for maximum mixed layer depth. This is not perfectly true since the time of the deepest mixed layer may vary from year to year at a single location and does not occur at the same time over the whole basin. During the year 2012 the area covered by a deep mixed layer (deeper than 600 m) is more extended than usual in the North  of the basin (even more than during the winter of 2008), starting from the Labrador Sea, it includes nearly all the Irminger Sea and progresses southward along the coast of America (Figure 3).  In the South-East of the basin, the deep mixed layer extension stops around 48/50°N such that only moderate mixed layer depths are observed along the shelf in the Bay of Biscay contrary to the 2009, 2010 and 2011 winters.

Mixed layer 2007-2012

Annual mean conditions

Near surface

The most salient feature of the 2012 annual mean temperature is an intense warm anomaly over the western basin from the tip of Greenland to 40°N and the persistence of a moderately warm anomaly over the Greenland Sea and along the East Greenland coast (Figure 4).

The structure of the salinity anomaly is not correlated with the temperature anomalies. While the warm anomaly over the Greenland Sea is associated with saltier waters, a negative salinity anomaly is building up west of 20°W. It started in 2009 in the centre of the basin (50°N - 30°W) and has since then gradually increased both in size and intensity.  In 2012, a large fresh anomaly, 0.5 PSS below the ‘normal’ WOA05 conditions, is observed.

Anomalie de température à 10 m Anomalie de salinité à 10 m

Deep layers

At 1000 m (Figure 5), the main points already noted in the previous years are confirmed:

  • The Greenland Sea, the Labrador Sea and the Irminger Sea are warmer than normal and this is a clear tendency since 2002 as seen in the time series (Figure 6).
  • The Mediterranean Outflow water seems warmer and saltier west of the Iberian peninsula and along the eastern boundary. The salt increase seems to extend over the basin.
  • A cold and fresh anomaly stands from the South of Iceland down to Rockall trough

A cold and fresh anomaly is observed south of the Gulf-Stream and Azores current (sub-tropical gyre).

Anomale de température à 1000 m Anomale de salinité à 1000 m

References

Gaillard, F.,  E. Autret, V.Thierry, P. Galaup, C. Coatanoan, and T. Loubrieu , 2009 : Quality control of large Argo data sets. JAOT, Vol. 26, No. 2.  337–351.

Gaillard, F.,  2012. ISAS-Tool Version 6: Method and configuration.  Rapport LPO-12-02, http://archimer.ifremer.fr/doc/00115/22583/

Gaillard Fabienne, Reverdin Gilles, Morin Pascal (2013). Contribution to the ICES Working Group on Ocean Hydrography. National report: France. 2013. http://archimer.ifremer.fr/doc/00134/24509/