NODC ocean heat content update
Posted by Ari Jokimäki on October 6, 2010
Recently I wrote about NODC ocean heat content data pointing out couple of problematic issues there which seemed to indicate that claims of oceans cooling after 2003 (or 2004) based on NODC data are unfounded. The situation seems to be on hold and we better wait for data updates before making any claims on the situation after 2003.
Now NODC has published an update to their data. The resulting situation after early 2000’s doesn’t show cooling unless one wants to cherry-pick the 3-month averages starting from the high peak at about 2004.
Corrections are described here. Also the effects of the corrections to the previous version are shown (see especially page 10 of the PDF). In addition to remarkable difference in 2000’s, the warm episode between 1975 and 1985 has leveled off a bit. The general outlook is that the warming during recent decades seems now to proceed more smoothly than it did before.
On page 11 you can see how the differences arise mostly from the southern hemisphere. Page 13 shows that there has been large addition of temperature profiles in southern hemisphere between 2003 and 2004. There were almost 4000 profiles added back then and practically all of them were Argo floats. Page 14 shows new data additions after the previous data release.
There are three main themes in the corrections:
1. Data additions and quality control changes – including the Argo problems.
2. Changes in base climatology.
3. XBT bias corrections.
I’m not familiar enough with the second item to comment on it. There has been a steady flow of research articles on XBT biases – just this week there was Good (2010) published. Reading Good’s abstract makes me think that there might still be quite a lot to be done on XBT biases.
The item 1 is the one I wrote about previously. They say it includes NODC own corrections and also corrections from originators (including Argo). On Argo, they say:
Substantial quality control has been carried out by the Argo community on the profiling floats, mainly to correct pressure offsets.
Page 15 shows the changed data from Argo and from XBT’s since the last data release. There are quite a lot of the changed temperature profiles. The presentation of corrections ends with the question:
Are quality control differences related to Argo delayed-mode quality control, NODC handling of data, or both?
One thing I wonder is that are all the Argo corrections now done? Last time I checked the Argo website and they said they would be done by the end of 2010. So, does Argo website say anything new now? No, it doesn’t. It seems that either Argo hasn’t informed about their new status yet or they haven’t yet finished the corrections. So, considering that XBT corrections are also still coming in, we’ll just have to wait some more, situation is not yet clear.