Model-observation and reanalyses comparison at key locations for heat transport to the Arctic (D2.1)

Assessment of key lower latitude influences on the Arctic and their simulation

Summary

Blue-Action Work Package 2 (WP2) focuses on lower latitude drivers of Arctic change, with a focus on
the influence of the Atlantic Ocean and atmosphere on the Arctic. In particular, warm water travels from
the Atlantic, across the Greenland-Scotland ridge, through the Norwegian Sea towards the Arctic. A
large proportion of the heat transported northwards by the ocean is released to the atmosphere and
carried eastward towards Europe by the prevailing westerly winds. This is an important contribution to
northwestern Europe's mild climate. The remaining heat travels north into the Arctic. Variations in the
amount of heat transported into the Arctic will influence the long term climate of the Northern
Hemisphere. Here we assess how well the state of the art coupled climate models estimate this
northwards transport of heat in the ocean, and how the atmospheric heat transport varies with changes
in the ocean heat transport. We seek to improve the ocean monitoring systems that are in place by
introducing measurements from ocean gliders, Argo floats and satellites.
These state of the art computer simulations are evaluated by comparison with key trans-Atlantic
observations. In addition to the coupled models ‘ocean-only’ evaluations are made. In general the
coupled model simulations have too much heat going into the Arctic region and the transports have too
much variability. The models generally reproduce the variability of the Atlantic Meridional Ocean
Circulation (AMOC) well. All models in this study have a too strong southwards transport of freshwater
at 26°N in the North Atlantic, but the divergence between 26°N and Bering Straits is generally
reproduced really well in all the models.

Altimetry from satellites have been used to reconstruct the ocean circulation 26°N in the Atlantic, over
the Greenland Scotland Ridge and alongside ship based observations along the GO-SHIP OVIDE Section.
Although it is still a challenge to estimate the ocean circulation at 26°N without using the RAPID 26°N
array, satellites can be used to reconstruct the longer term ocean signal. The OSNAP project measures
the oceanic transport of heat across a section which stretches from Canada to the UK, via Greenland.
The project has used ocean gliders to great success to measure the transport on the eastern side of the
array. Every 10 days up to 4000 Argo floats measure temperature and salinity in the top 2000m of the
ocean, away from ocean boundaries, and report back the measurements via satellite. These data are
employed at 26°N in the Atlantic to enable the calculation of the heat and freshwater transports.
As explained above, both ocean and atmosphere carry vast amounts of heat poleward in the Atlantic. In
the long term average the Atlantic ocean releases large amounts of heat to the atmosphere between
the subtropical and subpolar regions, heat which is then carried by the atmosphere to western Europe
and the Arctic. On shorter timescales, interannual to decadal, the amounts of heat carried by ocean and
atmosphere vary considerably. An important question is whether the total amount of heat transported,
atmosphere plus ocean, remains roughly constant, whether significant amounts of heat are gained or
lost from space and how the relative amount transported by the atmosphere and ocean change with
time. This is an important distinction because the same amount of anomalous heat transport will have

very different effects depending on whether it is transported by ocean or the atmosphere. For example
the effects on Arctic sea ice will depend very much on whether the surface of the ice experiences
anomalous warming by the atmosphere versus the base of the ice experiencing anomalous warming
from the ocean. In Blue-Action we investigated the relationship between atmospheric and oceanic heat
transports at key locations corresponding to the positions of observational arrays (RAPID at 26°N,
OSNAP at ~55N, and the Denmark Strait, Iceland-Scotland Ridge and Davis Strait at ~67N) in a number of
cutting edge high resolution coupled ocean-atmosphere simulations. We split the analysis into two
different timescales, interannual to decadal (1-10 years) and multidecadal (greater than 10 years). In the
1-10 year case, the relationship between ocean and atmosphere transports is complex, but a robust
result is that although there is little local correlation between oceanic and atmospheric heat transports,
Correlations do occur at different latitudes. Thus increased oceanic heat transport at 26°N is
accompanied by reduced heat transport at ~50N and a longitudinal shift in the location of atmospheric
flow of heat into the Arctic. Conversely, on longer timescales, there appears to be a much stronger local
compensation between oceanic and atmospheric heat transport i.e. Bjerknes compensation.