Before the ARGO floats were deployed, there were so few temperature and salinity observations at depths below 700 meters that the NODC does not present ocean heat content data during that period for depths of 0-2000 meters on an annual basis. That is, the NODC only presents its annual ocean heat content data for the depths of 0-2000 meters starting in 2005. The NODC’s annual ocean heat content data (0-2000 meters) for global oceans are here; for the Indian Ocean, see here; for the Atlantic Ocean, they’re here; and for the Pacific Ocean, data can be found here. The NODC now only presents its long-term data (1955 to present) for depths of 0-2000 meters in pentadal form. They had provided the long-term annual data for the depths of 0-2000 meters for a very short time period, but removed it from their website as soon as they released the pentadal data. There are two problems with the pentadal data. It mysteriously adds more than 30% to the long-term trend when compared to the formerly available annual data for those depths. (See the post here.) And the 5-year “averaging” of the pentadal data makes the dataset useless in attribution studies. 5-year filters have been used by the climate science community for years to mask responses to El Niño and La Niña events, which also have significant impacts on ocean heat content data.
So that leaves us with only 8 years of annual ocean heat content data to examine for the depths of 0-2000 meters. That’s okay. We can learn something from the 8 years of data.
Keep in mind, ocean heat content data is unlike surface temperature and lower troposphere temperature data. With ocean heat content data, the heat is either there or it’s not. There is no unrealized heat. Roger Pielke Sr. provided a more detailed explanation in his blog post here.
Also keep in mind that Balmaseda et al (2013) Distinctive climate signals in reanalysis of global ocean heat content (paywalled), of which Kevin Trenberth was one of the authors, began their abstract with:
The elusive nature of the post-2004 upper ocean warming has exposed uncertainties in the ocean’s role in the Earth’s energy budget and transient climate sensitivity.
That falls right in line with the period from 2005 to 2012, for which we have annual data. And as we’ll see, the warming is still elusive.
Ever since the NODC released their ocean heat content data for the depths of 0-2000 meters and published Levitus et al (2012), it seems that each time a skeptic writes a blog post or answers a question in an interview, in which he or she states that global surface temperatures haven’t warmed in “X” years, a global warming enthusiast will counter with something to the effect of: global warming hasn’t slowed because ocean heat content continues to show warming at depths of 0-2000 meters. Recently, those same people are linking Balmaseda et al (2013) and claiming the warming of ocean heat content data continues.
It is true that the NODC’s ARGO-era ocean heat content (0-2000 meters) continues to warm globally, but always recall that the ARGO data had to be adjusted, modified, tweaked, corrected, whatever, in order to create that warming. That is, the “raw” ocean heat content data for 0-2000 meters shows the decreased rate of warming after the ARGO floats were deployed. (See the post here.) Also, while the much-revised NODC ocean heat content data for 0-2000 meters might show warming globally, it shows very little warming for the Northern Hemisphere oceans since 2005. See Figure 1. Only about 7% of the warming of ocean heat content for the depths of 0-2000 meters occurred in the Northern Hemisphere from 2005 to 2012, yet the surface area of the Northern Hemisphere oceans represents about 43% of the surface of the global oceans.
Can well-mixed human-created greenhouse gases pick and choose between the hemispheres, warming one but not the other? One might think that’s very unlikely.
Something else to consider: the Northern Hemisphere warming of ocean heat content for depths of 0-2000 meters occurs in only one ocean basin, and it’s not one of the big ones.
NORTHERN HEMISPHERE OCEAN HEAT CONTENT – 0-700 AND 0-2000 METERS
Before we begin with the individual ocean basins, let’s take a look at the change in the ocean heat content of the Northern Hemisphere from 2005 to 2012, for the depth ranges of 0-700 meters and 0-2000 meters. See Figure 2. There was a comparatively minor warming in the Northern Hemisphere at depths of 0-2000 meters from 2005 to 2012. But the upper 700 meters in the Northern Hemisphere cooled. The difference is provided to show the additional warming that occurred at depths of 700 to 2000 meters.
NORTH INDIAN OCEAN HEAT CONTENT – 0-700 AND 0-2000 METERS
Before we begin, recall that the North Indian Ocean is not like the North Atlantic and North Pacific. The North Indian Ocean basically ends in the tropics with the Arabian Sea and the Bay of Bengal, both bordering Asia, while the North Atlantic and North Pacific have large extratropical portions. The North Indian Ocean is comparatively small for that reason.
If manmade greenhouse gases were capable of warming the global oceans, one would have to assume the warming would be similar to what’s taking place in the North Indian Ocean. As shown in Figure 3, the ocean heat content for the North Indian Ocean is warming at depths of 0-700 meters and 0-2000 meters—and at 700-2000 meters.
On the other hand, the ocean heat content data for the North Atlantic and North Pacific are not cooperating with the hypothetical warming of the oceans by manmade greenhouse gases.
NORTH ATLANTIC OCEAN HEAT CONTENT – 0-700 AND 0-2000 METERS
RealClimate recently published a post titled “The Answer is Blowing in the Wind: the Warming Went into the Deep End”. It’s about the Balmaseda et al (2013) paper. There Rasmus Benestad writes about the North Atlantic:
A weakening of the Atlantic meridional overturning circulation (MOC) may have played a role in the deep ocean warming.
So let’s take a look at the North Atlantic ocean heat content data. Based on the linear trend, the ocean heat content data of the North Atlantic for the depths of 0-2000 meters haven’t warmed from 2005 to 2012. See Figure 4. And the data for depths of 0-700 meters show cooling in the North Atlantic. The additional warming at the depths of 700-2000 meters (illustrated by the “difference”) was comparable to the cooling at 0-700 meters, inferring there might simply have been an exchange of heat between two depth ranges, but there is no evidence of manmade greenhouse gas-driven warming in the North Atlantic from 2005-2012.
Also recall that Mauritzen et al (2012) Importance of density-compensated temperature change for deep North Atlantic Ocean heat uptake (paywalled) found that while the upper 2000 meters of the North Atlantic warmed since the 1950s, the deep ocean below 2000 meters cooled, suggesting an exchange of heat between the deep ocean and the depths above 2000 meters. That cooling below 2000 meters is obviously not considered in the NODC ocean heat content data. Mauritzen et al (2012) was discussed in my post Is Ocean Heat Content All It’s Stacked up to Be? under the heading of SPEAKING OF STILL-TO-BE-DISCOVERED SUBSURFACE OCEAN PROCESSES.
NORTH PACIFIC OCEAN HEAT CONTENT – 0-700 AND 0-2000 METERS
There have been a number of blog posts about Balmaseda et al (2013) by proponents of the hypothesis of manmade global warming, including the RealClimate post linked above and at least one post at SkepticalScience. From what I can gather from those posts, Balmaseda et al (2013) are basically saying this: when La Niña events dominate, warm water is forced to depths greater than 300 meters, driven there by the stronger-than-normal trade winds associated with the La Niñas. According to the new and improved Oceanic NINO Index, starting in 2005, there have been 5 La Niña events and only 2 El Niños. Clearly, La Niña events dominated the last 8 years. Therefore, we would expect the ocean heat content of the North Pacific to be showing all sorts of excessive warming.
But the NODC’s annual ocean heat content data for the North Pacific shows cooling, Figure 5, not warming, for the depths of 0-2000 meters. And the depths of 0-700 meters have cooled at an even more drastic rate. If the stronger trade winds associated with the La Niña events have contributed to the warming of the North Pacific ocean heat content from 700-2000 meters (shown as the “difference”), then the only thing those stronger trade winds did was prevent the 0-2000 meter data from cooling as quickly as the 0-700 meter data.
I linked this post earlier, but I’ll link it again. For a detailed discussion of the problems with ocean heat content data refer to Is Ocean Heat Content All It’s Stacked up to Be?
For introductory discussions of how the ocean heat content data for the depths of 0-700 meters indicate the oceans warmed naturally, and discussions of the natural processes that caused the warming, refer to my illustrated essay “The Manmade Global Warming Challenge” [42MB]. And if you’re looking for a much more-detailed discussion of the natural warming of the global oceans, refer to my book Who Turned on the Heat? It was introduced in the post “Everything You Ever Wanted to Know About El Niño and La Niña”. It’s for sale in pdf form at US$8.00.
With their continuing failed attempts, the alarmist wing of the climate science community still has a lot of work to do to explain the warming of the global oceans, or lack thereof. If they can’t explain why the ocean heat content data of the North Atlantic and North Pacific have not warmed, they can’t claim greenhouse gases are responsible for the warming of the ocean heat content in the Southern Hemisphere. They also have problems with the long-term data, as we’ve been presenting and discussing for a few years. But first…
SkepticalScience’s Rob Painting provides a reasonable explanation of the hypothetical cause of greenhouse gas-driven warming of the global oceans in the post Observed Warming in Ocean and Atmosphere is Incompatible with Natural Variation. Painting writes (my boldface):
Arguably the most significant climate-related impact of increased concentrations of greenhouse gases in the atmosphere, is that they trap more heat in the ocean. Over the last half-century around 93% of global warming has actually gone into heating the ocean. A little-known fact is that the oceans are almost exclusively heated by sunlight (shortwave radiation) entering the surface layers. Geothermal heat emanating from deeper layers of the Earth does contribute a small amount to ocean warming but, on the time scale of decades & centuries, this contribution is effectively constant as it is determined by the rate of radioactive decay.
So how do greenhouse gases accomplish this ocean heating? This is discussed in this SkS post, but briefly; greenhouse gases radiate heat (longwave radiation) back toward the surface and, although they cannot penetrate into the ocean itself, they warm the uppermost surface of the thin cool-skin layer. The thermal gradient through this layer dictates the rate of heat loss from the (typically) warmer ocean surface, to the cooler atmosphere above. When greenhouse gases increase, more longwave radiation is directed back at the ocean surface, which warms the cool-skin layer, lowers the thermal gradient, and consequently reduces the rate of heat loss. The sum effect is that the oceans trap more of the sun’s energy and therefore warm over time.
While Painting does note that longwave radiation from greenhouse gases only impacts the top few millimeters (the skin) of the surface, he fails to consider that evaporation occurs at the surface, he fails to consider that the ocean release heat primarily through evaporation, and he fails to consider that when the ocean heat content data of the global oceans are divided into logical subsets, they indicate the oceans warmed naturally.
For almost 4 years, I’ve been illustrating how (and presenting the natural processes through which) specific La Niña events are responsible for the warming of ocean heat content in the tropical Pacific. More recently, I’ve shown that without the 1973-76 and 1995/96 La Niña events, Figure 6, the ocean heat content of the tropical Pacific would have cooled since the 1950s. It’s tough to claim manmade greenhouse gases are responsible for the long-term warming when that warming relies on only 4 years of data.
And I’ve been showing that the warming of the ocean heat content (0-700 meters) for the extratropical North Pacific is dependent on a 2-year shift and without that upward shift, which is likely caused by a change in wind patterns there, the ocean heat content of the North Pacific north of the tropics would also cool since the 1950s. Again, it’s tough to claim anthropogenic greenhouse gases caused the warming in this portion of the North Pacific when the ocean heat content data there would cool if you remove just 2 years of data.
(Note: Almost 2 decades ago Kevin Trenberth discussed the same wind pattern-based upward shift in the sea level pressure of the extratropical North Pacific. Refer to Trenberth and Hurrell (1995) Decadal coupled atmosphere-ocean variations in the North Pacific Ocean. See their Figure 6, which shows the sea level pressure of a portion of the extratropical North Pacific. One would think Trenberth should also know that same shift exists in extratropical North Pacific sea surface temperature and ocean heat content data.)
Curiously, when you combine those ocean heat content subsets for the Pacific Ocean with their different natural warming and cooling periods, Figure 8, they mimic the long-term warming of the global ocean heat content data, which alarmists happily but falsely claim is caused by manmade greenhouse gases.
To conclude this post, I’ll paraphrase the opening sentence of Balmaseda et al (2013):
The elusive nature of the post-2004 upper ocean warming continues to expose uncertainties in the ocean’s role in the Earth’s energy budget and transient climate sensitivity.
The annual ocean heat content data (0-700 meters and 0-2000 meters) presented in Figures 1 through 5 are available from the NODC webpage here. The NODC ocean heat content data for Figures 6, 7 and 8 are available through the KNMI Climate Explorer.