C. Heuzé and M. Årthun (2019) The Atlantic inflow across the Greenland-Scotland ridge in climate models (CMIP5), Elementa Science of the Anthropocene, 7(1), p.16, doi: 10.1525/elementa.354.

The motivation for this paper was simple:

• Climate models poorly represent the Arctic sea ice extent;
• This sea ice is heavily controlled by the oceanic heat that enters the Arctic;
• Most of the heat comes from the North Atlantic, via the Nordic seas;
• …no one has looked at how the Nordic seas are represented in climate models?!

So we did exactly this, eventually narrowing it down to the representation of the oceanic heat flow from the North Atlantic into the Nordic seas. We used 23 CMIP5 models – I’ll be honest, these are the exact same ones as in Heuzé (2017) in order to make the most of the Terabytes of data I had to download over the past years.

Main result: poor representation of the oceanic heat inflow, with most models underestimating it while a few heavily overestimating it, and even inaccurate seasonal cycles.
Reason: a mix of poor bathymetry and inaccurate large scale oceanic and atmospheric circulations, which result in models importing waters from the ‘wrong’ part of the Atlantic into the Nordic seas.
Why should people care? Sure, lots of transformations occur once in the Nordic seas before the water goes into the Arctic, but that most models already have too little heat coming in the Nordic seas is consistent with most models also simulating more sea ice than we have. Fixing the bathymetry is not trivial, but it is worth a try to improve sea ice projections.

Mean heat inflow through Denmark Strait in CMIP5 models (after Heuzé and Årthun, 2019)

Download the full-text here.

You’re still here? Then I’ll tell you why this paper means a lot to me on the personal level.
Marius and I came up with this idea at EGU2017. The previous few months had been chaotic on the personal level, but finally life was sorting itself out and I was optimistic again. So I was full of energy and we submitted the first draft already in September. Then came October.  The same week this paper was rejected, my mother was diagnosed with cancer. As it became clear that the treatment was not working, I lost interest in the paper (among other things). Marius and I agreed on corrections and large improvements at Ocean Sciences in February 2018, but it took time, so that I ended up writing this new version very soon after my mother had died. When the reviewers’ comments eventually came six months later (!), I could not work on them; re-reading the manuscript threw me back to the early months of grieving. I had to force myself, but that meant being again a wreck mentally. Yet we did it. And it is quite fitting that the paper was published so close to the first anniversary of her death – I have made peace with both, and can now move on.
Worry not, the other paper comments will be less personal.