Sometimes journalists are so focused on a particular story that they ‘hear what they want to hear and disregard the rest’. There was a perfect example of this last week in the Guardian reporting from the RAPID Climate Change conference in Birmingham (UK) which I was attending. The conference, whose theme was observations, modelling and paleo-climate related to the Thermohaline and Meridional overturning circulation (MOC) in the North Atlantic, could have been expected to attract media attention (particularly in the Europe) and indeed it did. However, the Guardian story, which started “Scientists have uncovered more evidence for a dramatic weakening in the vast ocean current that gives Britain its relatively balmy climate” was in complete opposition to the actual evidence presented and I wasn’t the only person to notice. How could the reporting be so wrong?
First, a bit of background: RAPID is a focused research program being run mainly out of the UK, but with contributions from Norway, the Netherlands and from the US. One of their main achievments has been to set up a mooring array (which consists of a dozen or so permanently attached monitors of temperature, salinity and pressure) that can continuously monitor the circulation in the North Atlantic across a section at 26°N. Measurements taken as the moorings were first installed were highlighted in the Bryden et al paper last year. As readers will no doubt recall, that publication, suggesting that a long term decrease in the MOC was underway, was greeted by a media storm. We cautioned at the time that the results were preliminary and, specifically, that the internal variability was probably high enough to make it unlikely that the changes had risen above the noise.
At the meeting this week, Bryden and colleagues gave an update of the work, specifically focusing on the first year of data from the moored array. This is the first time that there has ever been such a continuous set of estimates across the whole Atlantic and so reports of the size and nature of the variability were eagerly anticipated. And they did not disappoint! There were two key observations: first, that the approximations that had been used in the Bryden et al study were actually valid, and secondly, that the variations day by day varied by around 5 Sv (1 Sv is about 10 times the flow of the Amazon). The mean over the year for the MOC was 18 Sv – very close to what was expected and in the middle of recent estimates – and significantly, larger than the value seen in the 2004 snapshot. Given that degree of ‘noise’, this implies that no conclusions about trends over recent decades can be supported.
Other results presented supported this basic picture: transport estimates at different latitudes were not coherent with the initial results, model variability in the best ocean models was large (suggesting that detectability of a MOC slowdown before 2030-2050 was unlikely), and temperature, salinity and velocity changes in the overflow waters between Greenland and Europe showed significant connections to the North Atlantic Oscillation but no obvious trends. A number of records that had seemed to be trending strongly when first looked at, now seem to be simply more variable than first thought. This was something of a theme at the conference – the closer we look at the ocean, the more dynamic it appears.
So why was the Guardian story so wrong? Well, the nature of variability invariably implies that there are periods when the values are above the mean, and periods when it is below the mean. The minimum values appeared to be during a 10 day interval in November 2004 when the inferred deep western boundary current appeared to be very weak indeed. But then it came back. Now, recall that we have never seen this quality of data before and explanations for the variability (deep eddies? waves?) are not yet available. Thus, no-one has any clue whether this is normal or unusual – right now it’s simply an interesting phenomenon. Picking this out of the results is therefore a little perverse. The big story should have been the phenomenal effort that has gone into exploring this important issue, the much improved context for previous measurements and a welcome reassessment of the significance of previous results. It’s a shame the Guardian missed it.