I’ve just finished the first year of my PhD! As well as being pretty exciting, this means I got to present some of my work so far at the annual Atmospheric, Oceanic & Planetary Physics retreat.

Read on for a whirlwind tour of what I’ve been up to for the last year.

I’ve been investigating variability of the Atlantic Meridional Overturning Circulation. Meridional means along meridians, which are the lines running north-south on maps

Why is this interesting?

The global overturning circulation, of which the Atlantic Meridional Overturning Circulation is a substantial component, moves large quantities of heat and water around the world. The heat and water have substantial impacts on regional and global climate. The movie “The Day After Tomorrow” was a Hollywood alarmist vision what could happen if this circulation stopped. There aren’t any scientists predicting such sudden changes. But, changes in the Atlantic Meridional Overturning Circulation lead to changes in sea surface temperature in the North Atlantic, which in turn are strongly linked to climate variations in the Northern Hemisphere.

Measurements of the Atlantic Meridional Overturning Circulation

In 2005 a paper was published in Nature suggesting that the Atlantic Meridional Overturning Circulation had weakened by 30% over the preceding 50 years. This was a big deal.

The research cruise that took the final measurement they based their paper on also put down an array of moorings to continuously monitor the Atlantic Meridional Circulation. I think my next point is best made graphically. The hollow circles are the measurements used for the 2005 paper. The solid circles are the minimum, mean and maximum values measured during the first year that the array was in operation. The measurements are given in Sverdrups (Sv). One Sverdrup is one million cubic metres of water per second.

All of the previous measurements lie within one standard deviation (5.6Sv) of the mean (18.7Sv) reported by Cunningham et al. (2007). This means that there is very little chance of the observed 30% decrease being anything other than a side-effect of the small number of measurements.

Modelling the Atlantic Overturning Circulation

We’ve just seen that there is a lot of short term variability in the circulation, which means we need a lot of data to analyse if we want meaningful results. Unfortunately, collecting a couple of centuries of data is going to take us a couple of centuries. The short-term solution is to use numerical models. This is exactly the approach that a number of people have taken recently.

One paper, Bingham et al. (2007), examined the meridional coherence of the circulation. This means they looked at the strength of the circulation at each latitude and time and related that to the strength at every other latitude and time. The result that is relevant here is that they always found a reduction in the coherence near the Gulf Stream. This suggests that the circulation north of the Gulf Stream didn’t correspond with the circulation to the south (I mean “correspond” in terms of have roughly the same strength, not in some strange anthropomorphising way).

Following this Zhang (2010) used another global circulation model to look at the movement of anomalies in the circulation. This is particularly important under a changing climate scenario. What Zhang found is that the changes in the circulation’s strength move south slowly from the Arctic until they reach the latitude of the Gulf Stream, at which point their speed changes and they move south much faster.

My work

I’ve been using a simplified model to investigate the Atlantic Meridional Overturning Circulation. The model has a rectangular, infinitely deep ocean with one, two or three active layers. To make the ocean move I have wind blowing over it, but the “wind” is very idealised; it’s speed depends only on the latitude and does not change with time.

Obviously the Atlantic is neither a rectangle nor infinitely deep. Those are simplifications that will (hopefully) allow me to do more in depth analysis of the results.

The results are very promising; this simple model produces a propagation pattern that is very similar to the one computed by Zhang (2010).

This means that my model is complex enough to reproduce the pattern seen in much more complex models. This in itself is interesting, but the plan is to work out exactly what determines the speed at which these anomalies move and the mechanism by which they change speed so abruptly.

P.S.

The actual presentation can be found here. But, it is a collection of slides to talk with, not a standalone document, so it might not make much sense on its own.