One of the great things about running this blog is that people send me things to look at. Sometimes I see connections between two things that were initially unrelated by the original messages. This is one of those cases.
Dr. Roger Pielke Sr. suggested back in 2003 in a peer reviewed BAMS paper, that “…it is the change in ocean heat content that provides the most effective diagnostic of global warming and cooling.” Recently at ICCC 2009, Dr. Craig Loehle did a presentation titled “1,500-Year Climate Cycles, Broken Hockey Sticks, and Ocean Cooling” (PowerPoint) which talked about the ocean heat content.
I was reminded of one of his graphs from that presentation by a recent post on Jennifer Marohasy’s blog. For your viewing pleasure, using graphic editing tools, I created a slightly larger and annotated version, shown below:
The next day, on an email list I subscribe to, Alan Siddons sent along this graph with this note:
“Thought you’d like to see the Mauna Loa rate of CO2 change up to now. Kind of odd these recent years.”
I didn’t think much of Siddons’ graph initially, but as luck would have it, I happened to have Loehle’s graph open in a desktop window from Jennifer’s blog. I noticed something interesting and unexpected looking at the two.
Here is Alan Siddons’ graph of recent MLO CO2 data that shows the changes in the rate of CO2 with their measurements. I added some annotation and a title to make it clearer as to what this graph is:
What interested me about Alan’s MLO CO2 rate of change graph was the period from 2004 to the present. There’s a noticeable downturn in the peaks. I’ve bracketed the area of interest below and added an eyeball trend line for the peaks:
When you take the bracketed period from Alan Siddon’s MLO CO2 rate of change graph, and compare it (again using graphical editing tools) to Loehle’s Ocean Heat content graph, there appears to be some correlation:
It makes sense, as the heat content of the oceans drops, CO2 solubility in seawater increases, and thus we see an absorption of CO2 and dampening of the annual peaks in the rate of change. Obviously this is just a simple visual analysis, and I don’t pretend to know everything there is to know about either of these subjects or datasets, but I thought the serendipity of these two pieces of initially independent and unrelated graphs of data was interesting and worth discussing.
Of course there will be those that argue that “the oceans have not cooled” and cite the work by Josh Willis on catching some errors in the ARGO floater data. I won’t dispute his work here since I’m not an expert on the ARGO project. I’ll leave that to Dr. Roger Pielke Sr., as he wrote in this post on his Climate Science blog:
Josh Willis is a well respected scientist and his view merit consideration. In this case, however, Climate Science concludes that he is misinterpreting the significance of his data analysis. He agrees that
“Indeed, Argo data show no warming in the upper ocean over the past four years”.
He dismisses this though by claiming that
“…but this does not contradict the climate models. In fact, many climate models simulate four to five year periods with no warming in the upper ocean from time to time. “
Where are these model results that show lack of upper ocean warming in recent years? There is an example of a model prediction of upper (3km) ocean heat content for decadal averages in Figure 1 of
Barnett, T.P., D.W. Pierce, and R. Schnur, 2001: Detection of anthropogenic climate change in the world’s oceans. Science, 292, 270-274,
but they did not present shorter time periods. Nonetheless, since Figure 1 is presumably a running 10 year average, the steady monotonic increase in the model prediction of upper ocean heat content (the grey shading) suggests that no several years (or even one year) of zero heating occurred in the model results. The layer they analyzed in the figure is also for the upper 3 km but in Figure 2 the Barnett et al study showed that most of this heating was in the uppermost levels.
Thus the lack of heating in the upper 700m over the last 4 years does conflict with at least the Barnett et al model results!
What the upper ocean data (and lack of warming) actually tells us is that if global warming occurred over the last 4 years, it was in the deeper ocean and is thus not available in the short term to the atmosphere.
Indeed, if it is in the deeper ocean, it likely more diffused and therefore could only enter the atmosphere slowly if at all. This heat could also have exited into space, although the continuation of global ocean sea level rise suggests that this is less likely unless this sea level rise can be otherwise explained.
The other heat stores in the climate system are too small (and the atmosphere has clearly not warmed over the last few years). Global sea ice cover is actually above average at present (the Antarctic sea ice is at a near record level). The continued sea level rise indicates that the heat is in the deeper ocean (which is not predicted by the models).
Josh Willis too easily dismisses the significance of his research findings.
The interesting thing about what I’ve pointed out above is that we have two independently analyzed datasets (Oceanic heat content and MLO CO2 rate of change) that appear to demonstrate the same thing: the oceans appear to have cooled in the past 5 years. That is also partially consistent with a third dataset, the RSS global temperature anomaly (or fourth if you want to count UAH same data, different method) which shows there has been a flat trend in the past few years. The graph below is both for land and ocean data:
RSS Data Source is here
Even Josh Willis’ own graph of corrected -vs- uncorrected ARGO data illustrating sea level change due to thermal expansion shows a flat trend during this period:
Clearly something is happening to heat content within our oceans, whether it is a flat trend or yet unrecognized loss of heat, remains to be hashed out. The year 2008 was a cooler year globally, and there is quite a bit of measured as well as anecdotal (weather event) data to support that. Our oceans are in fact the planet’s largest heat sink, and it has been routinely demonstrated that changes in that heat sink status (AMO, PDO, El Nino and La Nina) do in fact affect our weather and climate.
So to paraphrase Josh Willis in his rebuttal of his own data: “Is it me, or did the oceans cool”?
UPDATE 4:45 PM 3/21: Allan Siddons has provided two additional graphs. The first being an overlay of MLO monthly data on MSU oceans data
The second is a 12 month average of MLO CO2 rate overlaid on my RSS MSU land and ocean graph posted originally. It seems clear that there is a CO2 rate of change response that mirrors global temperature.
Bob Tisdale has also provided some similar graphs via many links made in the comments. Be sure to have a look. – Anthony