Skip to main content

Advertisement

Log in

Mechanisms of interannual- to decadal-scale winter Labrador Sea ice variability

Climate Dynamics Aims and scope Submit manuscript

Abstract

The variability of the winter sea ice cover of the Labrador Sea region and its links to atmospheric and oceanic forcing are investigated using observational data, a coupled ocean–sea ice model and a fully-coupled model simulation drawn from the CMIP5 archive. A consistent series of mechanisms associated with high sea ice cover are found amongst the various data sets. The highest values of sea ice area occur when the northern Labrador Sea is ice covered. This region is found to be primarily thermodynamically forced, contrasting with the dominance of mechanical forcing along the eastern coast of Baffin Island and Labrador, and the growth of sea ice is associated with anomalously fresh local ocean surface conditions. Positive fresh water anomalies are found to propagate to the region from a source area off the southeast Greenland coast with a 1 month transit time. These anomalies are associated with sea ice melt, driven by the enhanced offshore transport of sea ice in the source region, and its subsequent westward transport in the Irminger Current system. By combining sea ice transport through the Denmark Strait in the preceding autumn with the Greenland Blocking Index and the Atlantic Multidecadal Oscillation Index, strong correlation with the Labrador Sea ice area of the following winter is obtained. This relationship represents a dependence on the availability of sea ice to be melted in the source region, the necessary atmospheric forcing to transport this offshore, and a further multidecadal-scale link with the large-scale sea surface temperature conditions.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13

Notes

  1. Further poles of variability are also evident in the Gulf of St. Lawrence, west of Newfoundland, and in the Greenland Sea, in the north east extreme of the domain shown in Fig. 1. However, these two poles will not be discussed further in this work, as we will focus on the Labrador Sea region.

  2. Note that whilst high and low SIA quartile years corresponding to the full Labrador Sea region are used throughout in the construction of the composites, the years representing the upper and lower SIA quartiles are, with the exception of 1 year in both quartiles, the same as for the northern subregion and yield essentially identical results.

  3. Note that SSS is not available for MIROC4h in the CMIP5 archive, hence the use of the 1 m depth level

References

Download references

Acknowledgements

We acknowledge the World Climate Research Programme’s Working Group on Coupled Modelling, which is responsible for CMIP, and we thank the Atmosphere and Ocean Research Institute (The University of Tokyo), National Institute for Environmental Studies, and Japan Agency for Marine-Earth Science and Technology for producing and making available their model output. For CMIP the U.S. Department of Energy’s Program for Climate Model Diagnosis and Intercomparison provides coordinating support and led development of software infrastructure in partnership with the Global Organization for Earth System Science Portals. We are grateful to the three anonymous reviewers for their helpful comments on the manuscript. The research leading to these results has received funding from the European Union 7th Framework Programme (FP7 2007-2013) under Grant agreement 308299, NACLIM Project.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to S. Close.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (PDF 1450 KB)

Appendix A1: Estimations of freshwater input required to drive salinity changes

Appendix A1: Estimations of freshwater input required to drive salinity changes

Consider the equation of conservation of salt:

$$\begin{aligned} S_{r}=\frac{{\rho }_{w}V_{w}S_{w}+{\rho }_{f}V_{f}S_{f}}{{\rho }_{w}(V_{w}+V_{f})}, \end{aligned}$$
(1)

where \(\rho\), V and S denote density, volume and salinity respectively, and the subscripts w, f and r are used to denote the properties of the source water, the freshwater input and the resultant admixture respectively. We wish to solve for \(V_f\) and thus rearrange (1) to give:

$$\begin{aligned} V_{f}=\frac{{\rho }_{w}V_{w}(S_{w}-S_{r})}{{\rho }_{w}S_{r}-{\rho }_{f}S_{f}}. \end{aligned}$$
(2)

We consider the full northern subregion as the source water, and calculate the mean density over this region over the full period to obtain \(\rho _w=1027.3\) kg m\(^{-3}\) (with a range of \(\pm \,0.2\) kg m\(^{-3}\) about this value). \(V_w\) is estimated by multiplying the surface area of the northern subregion by a characteristic layer depth. This depth is found by evaluating the correlation between the time series of the JFM surface salinity of the region and that of the underlying depth levels. Decorrelation occurs at a depth of approximately 50 m, yielding a value of \(V_w=2.565\times 10^4\,\hbox {km}^3\) for the source water volume. We then take the mean salinity of the northern subregion in October as \(S_w\) (the month after which the correlation between the summer/autumn and winter SSS in the region breaks down), and the mean salinity over the region in January (the time at which ice growth is underway) as the target admixture, \(S_r\). In a first step, we consider the freshwater input necessary to effect the observed salinity change, and thus use S f  = 0. From (2), we estimate the freshwater volume \(V_f\), required to drive the salinity change in each year. The average freshwater input required for the highest quartile of SIA years is thus \(\sim \,130\,\hbox {km}^3/\hbox {year}\) (with a maximum required input of \(390\,\hbox {km}^3/\hbox {year}\)).

In a second step, we take into account the increased salinity of the ice meltwater. Since the salinity data are model outputs, we use the values of \(\rho _f\) and \(S_f\) employed by the ice model to preserve a consistent framework, thus taking \(\rho _f=900\) kg m\(^{-3}\) and \(S_f=6\). With these values, a mean required input of \(\sim \,150\,\hbox {km}^3/\hbox {year}\) is obtained (with a maximum of \(460\,\hbox {km}^3/\hbox {year}\)).

1.1 Appendix A2: Estimates of influence of glacial meltwater

Drawing on Fig. SI1 of Bamber et al. (2012), the difference between highest and lowest glacial meltwater fluxes from the Greenland ice sheet over the 1979–1998 period is approximately \(40\,\hbox {km}^3/\hbox {year}\). Using this value as \(V_f\) in (1) above, and using the mean surface salinity over the study period, \(S_w\)=33.9 (again, using the October mean SSS value and model output, with all other parameters as above), this would imply a change of 0.04 in salinity were the full volume of meltwater to be mixed over the northern subregion. For comparison, both the model and EN4 objective analysis suggest that the large freshening episodes corresponding to the high ice cover states were associated with a decrease in salinity of order 0.5.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Close, S., Herbaut, C., Houssais, MN. et al. Mechanisms of interannual- to decadal-scale winter Labrador Sea ice variability. Clim Dyn 51, 2485–2508 (2018). https://doi.org/10.1007/s00382-017-4024-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00382-017-4024-z

Keywords

Navigation