Deep Ocean Warming in Degrees C

In recent weeks and months, there’s been a run of papers—with typical mainstream media misinforming embellishments—about improved estimates of the warming of the oceans…and blog posts about them. One thing common to the papers is the use of the metric zettajoules for ocean heat content—a metric that’s meaningless to most people.

The former National Oceanographic Data Center (NODC) portion of what is now contained under the umbrella of the National Centers for Environmental Information (NCEI) produces a dataset of depth-averaged temperatures for depths of 0-2000 meters, upon which, in part, their Ocean Heat Content data are based, and it’s presented in degrees C. To put deep-ocean warming in perspective, I’m presenting that data in this post.

Sadly, with the government shutdown, the NCEI website is offline, so I can’t provide you with links to additional information about the dataset. That is, we can’t confirm during the shutdown that this dataset is made up primarily of data from ARGO Floats. I did, however, present this same dataset and discuss its foundation in an October 2014 post There is a Wide Range in the Argo-Era Warming (And Cooling) Rates of the Oceans to Depths of 2000 Meters. The WattsUpWithThat cross post is here. Of course, today, the links to the NODC website there will bring you to reminders of the government shutdown.

The NCEI’s subsurface ocean temperature anomaly data to depths of 2000 meters are available through the KNMI Climate Explorer, but there it’s still identified as NODC data. Regardless, with the KNMI Climate Explorer, we can examine the data regionally—for individual ocean basins and parts of them. The KNMI Climate Explorer also allows users to create trend maps, so we can see where the oceans have warmed to depths of 2000 meters, where they have not warmed, and where they’ve cooled…yes cooled. The data start in 2005, so it’s not a long term dataset, but the intent of this post is to put deep-ocean warming in perspective for readers so the short time span is not a concern. And to simplify matters, I’m going to call it NODC data, as it is identified at the KNMI Climate Explorer.

Note: For those readers who want to investigate further, the KNMI Climate Explorer also has the NODC depth-averaged temperature data for the depths of 0-100 meters and 0-700 meters beginning in 1955. It is unfortunate that the data aren’t subdivided by depths…that is 0-100 meters, and 100-700 meters, and 700-2000 meters. [End note.]

Keep in mind the average depth of Earth’s oceans is about 3,788 meters, so we’re examining a little more than the top half.


Figure 1 presents the NODC depth-averaged temperature anomaly data for the global oceans, from pole to pole, from February 2005 to September 2018. Based on the values of the linear trend line, the temperature of the global oceans for depths of 0-2000 meters has risen about 0.065 deg C since 2005. (As a reminder, 0.065 deg C is read sixty-five one-thousandths of a degree Celsius.) One thing you will note in Figure 1 is all the tiny steps in the data. The reason: The NODC produces that data on a quarterly basis, while KNMI presents it monthly.

Figure 1

Instead of whining about how much the Earth’s oceans have warmed, assuming that increasing levels of CO2 are solely responsible for that warming as is commonly believed, people should be thankful we live on a planet where the oceans can absorb the vast majority of global warming (90%, they say), with so little effect.

Figure 2 presents the data for the Northern and Southern Hemispheres. The data indicate the oceans to depths of 2000 meters in the two hemispheres warmed at similar rates.

Figure 2


BUT, big but, the similarities do not necessarily extend to the individual ocean basins. In the Atlantic, as shown in Figure 3, the South Atlantic warmed at a rate that was noticeably higher than the global rate and 50% higher than the rate of the North Atlantic.

Figure 3

The warming rates are closer together in the Pacific, though, with the South Pacific warming at a rate that was only about 20% faster than the North Pacific, as shown in Figure 4. In this basin, the North Pacific warmed at a rate that was a bit below the global rate.

Figure 4

The North Indian Ocean is so small I didn’t bother to divide that basin into hemispheric subsets. The Indian Ocean, overall, warmed at a rate that is comparable to and a little faster than the global rate of 48 one-thousandths of a degree Celsius per decade. See Figure 5.

Figure 5

And that brings us to the Arctic and Southern Oceans. Yes, there are parts of both polar ocean basins that are ice free even in their respective winters. And as you can see in Figure 6, the data for the Arctic Ocean can have large variations from year to year. Not surprisingly, the warming rate of the Southern Ocean data is less than half the rate for the global oceans.

Figure 6


Figure 7 is a map of the North Pacific Ocean that displays the trends in the temperature anomalies for the depths of 0-2000 meters for the period of 2006 to 2017. The scaling of the color-coding ranges from -0.02 to +0.02 Deg C/Year.

According to the NOAA webpage here, (link not working due to government shutdown) the North Pacific covers about 21% of the surface of the global oceans, or about 77 million sq. km. To put that in perspective, the North Pacific covers almost as much of the globe as Asia, Europe and North America combined (78.5 million sq. km.), so the North Pacific is a huge chunk of wet real estate.

I was interested to see if The Blob had an impact on the depth-averaged temperature data there. Lo and behold, it appears it has.

Figure 7

Note: The Blob was the name given to the area of naturally (weather) warmed ocean (at the surface and to depth) in the eastern extratropical North Pacific. It formed in 2013, coupling with a resilient ridge of high pressure that impacted weather patterns across North America well into 2016. The Blob dissipated in October 2016, as discussed in the post here. (The WattsUpWithThat cross post is here.) Also see the series of posts about The Blob for additional general information.

The impacts of The Blob makes one wonder how often a persistent ridge of high pressure causes a portion the ocean to warm at the surface and to depth, only to go unnoticed because it’s happening in a remote part of the oceans. The primary reason The Blob in the Eastern North Pacific was studied was it was close to the west coast of the U.S. and Canada and was impacting regional weather patterns.

And as you might recall, a NOAA GODAS monthly briefing revealed that the naturally-occurring ocean-warming effects of The Blob reached depths of more than 300 meters. That aspect of The Blob was discussed in the post The BLOB Seems to Be Disappearing at the Surface – But Will It Re-emerge? (WUWT cross post is here.) and further shown with my Figure 2 from that post, which is from page 25 of that NOAA GODAS report. Unfortunately, I can’t find an archived version of that GODAS monthly briefing. Good thing I discussed it in a post and posted a copy of their illustration. [End note.]

To remove any thoughts you might be having that the warming in the Eastern North Pacific occurred gradually with time, see Figure 8, which presents the depth-average temperature anomalies to 2000 meters for that region as bordered by the coordinates of 10N-65N, 150W-100W, for the period of February 2005 to September 2018. It appears as though the combined effects of The Blob and the 2014/15/16 El Niño put an end to the cooling taking place in the Eastern North Pacific from 2005 through 2013.

Figure 8

Does the curve in Figure 8 remind you of another dataset? More on that later in the post.

To capture the warming in the West-Central High-Latitude North Pacific seen in Figure 7, I used the coordinates of 35N-65N, 150E-170W for Figure 9. The warming there more than tripled the global rate of 48 one-thousandths of a deg C per decade. The gradual warming there was interrupted by a noticeable dip and rebound in 2012 and 2013.

Figure 9

Next is the area of extreme cooling east of the Philippines. Normally, in satellite-based sea level trend maps we see very high sea level trends in that part of the western Tropical Pacific—high trends that are associated with leftover subsurface warm water from El Niño events being returned to the western tropical Pacific via off-equatorial Rossby Waves. One might then suspect the high negative trend there in the depth-averaged temperatures to 2000 meters is also a result of a very strong El Niño, specifically the 2014/15/16 El Nino. See Figure 10, in which I’ve used the coordinates of 10N-20N, 120E-165E to capture that cooling area. It sure looks like the high rate of cooling was caused by the dip and rebound related to the 2014/15/16 El Niño as the West Pacific Warm Pool was “drained”.

Figure 10

Figure 11 presents a trend map, globally, for the trends in the temperature anomalies for the depths of 0-2000 meters for the period of 2006 to 2017. The scaling is the same as Figure 7.

Figure 11

We’ll discuss a couple of those areas in the second part of this post, especially the three obvious regions in the North Atlantic.

I’m also curious about the lines of high subsurface temperature trends south and southwest of Australia, and South of the South Atlantic in the Southern Ocean. Are they gradual high rates of warming like Figure 9 or a period of minor cooling followed by a sudden substantial warming like Figure 8?

We’ll find out next time.


The following is a curiosity. I found it interesting. I’ll let you speculate about it in the comments.

Did the curve in Figure 8 of the depth-averaged temperature anomalies (0-2000 meters) for the Eastern North Pacific remind you of another dataset? The timing of the big wiggles and big spike in response to The Blob and the 2014/15/16 El Niño? Well, it reminded me of another dataset, that’s for sure. So, for this part of the post, I won’t name the dataset I’m referring to and call it the “Mystery Dataset”. That way you can try to figure out what it is. I downloaded that data for the Mystery Dataset through the KNMI Climate Explorer so that I could set the base years to 2005-2017, the same as the depth-averaged temperature anomalies (0-2000 meters) for the Eastern North Pacific.

For the comparison in Figure 12, I’ve standardized the depth-averaged temperature anomalies (0-2000 meters) for the Eastern North Pacific and the Mystery Dataset. Have you figure out the source of the Mystery Dataset. That is, to standardize, I divided each dataset by its respective standard deviation.

Figure 12

No cheating and looking ahead.

The Mystery Dataset is noisier than the depth-averaged temperature anomalies (0-2000 meters) for the Eastern North Pacific, so I’ve smoothed both with 12-month running mean filters, centered on the 6-month, in Figure 13.

Figure 13

Obviously, The Blob and the 2014/15/16 El Niño had major impacts on both datasets.

Have you guess the dataset presented as the Mystery Dataset? The answer is here.

That’s it for this post.

Have fun in the comments and enjoy the rest of your day.


Please purchase my recently published ebooks. As many of you know, this year I published 2 ebooks that are available through Amazon in Kindle format:

And please purchase Anthony Watts’s et al. Climate Change: The Facts – 2017.

To those of you who have purchased them, thank you. To those of you who will purchase them, thank you, too.


Bob Tisdale

0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
nw sage
January 23, 2019 6:08 pm

When you state the average depth is 3788 meters, is that a numerical average of the highest and lowest or is the average based on the area above/below that depth?

Alan Tomalty
Reply to  Ashok Patel
January 23, 2019 9:56 pm

Could the wayback machine be used to check on murderers in Europe? THE EU decided 5 years ago to require web sites and search engines to delete old curriculum vitae information from the web so as to please people that didn’t want negative info on them lasting forever.

nw sage
Reply to  Ashok Patel
January 25, 2019 7:43 pm

I glanced at the archive page (thanks) and, at first glance the method seems to be something similar to looking at all available nautical charts, converting all depths in fathoms (ie 6 ft) to meters and running an average for each ocean basin. Obviously some very old data with virtually unknowable precision was used. and the intrinsic error bars are pretty big.
This method also has little to do with the volumes of seawater at any particular temperature yet the relative volume of each portion has a huge influence on the amount of energy exchanged to make a meaningful temperature change.

Tom Halla
January 23, 2019 6:12 pm

I wonder how much minor changes in temperature matter. Even if one can measure to a .01 degrees C. does it mean anything?

David Chappell
Reply to  Tom Halla
January 23, 2019 9:25 pm

Have any of the experts ever talked to submariners? My military career involved the atmosphere rather than under the surface, but I’m led to believe these guys know quite a lot about sea temperature profiles which probably makes a nonsense of the concept of a 2km average.

Ben Vorlich
Reply to  David Chappell
January 24, 2019 3:23 am

Yes from what I’ve read experienced U-boat commanders were extremely good at hiding from ASDIC/Sonar under different temperature layers. As I understand it the Sonar pulses would reflect/diffract at the layer junctions, either confusing the operator or hiding the Submarine.

Paul Linsay
Reply to  David Chappell
January 24, 2019 8:08 am

A friend of mine signed up with the Navy to aviod combat in Viet Nam. His job was finding undersea currents where submarines could hide because of the temperature differences. There were lots of them, everywhere.

Alan Tomalty
Reply to  Tom Halla
January 23, 2019 10:09 pm

PM Justin Trudeau in Canada is trying to measure the effect on temperature decrease from his carbon tax. I calculated that by 2100, Canada’s contribution will be 0.0015 C reduction in global temperature. So after 24 days of carbon taxes assuming the top 620 emitters in Canada have been able to switch over to a non CO2 generating fuel ( of course impossible, but anyway) ; the temperature decrease so far (assuming linear response, I know the climate sensitivity isn’t linear but you get the idea) that Canada has caused in global temperature is 0.0000012 C. I am wondering if NASA has any thermometers that can measure to that accuracy?

Caligula Jones
Reply to  Alan Tomalty
January 24, 2019 7:09 am

And as I point out: even if we can magically measure the split between man-made CO2 and natural (say 60/40 worldwide), we have to be very, very sure we are only scrubbing the man-made CANADIAN stuff.

I mean, how terrible it would be if we somehow culturually exproriated the CO2 of some jungle somewhere causing undue hardship to an uncontacted Amazon tribe by making things colder for them?

Louis Hooffstetter
January 23, 2019 6:20 pm

What’s the conversion factor from Metric Zettajoules to Hiroshimas?

R Shearer
Reply to  Louis Hooffstetter
January 23, 2019 6:34 pm

Multiply by a Hiroshima over a Zettajoule.

Dave Miller
Reply to  R Shearer
January 24, 2019 8:18 am

No, in order for the units to work out, you need to multiply by an Olympic Swimming Pool.

Reply to  Louis Hooffstetter
January 23, 2019 7:55 pm

I want to know:
Is a Zettajoule bigger than a Gazillion?

I never use Zettajoules, but I use Gazillions a lot – a Gazillion is much bigger than a breadbox.

Bjarne Bisballe
Reply to  Louis Hooffstetter
January 24, 2019 12:24 am

63 teraJoule per Hiroshima bomb, I have been told

Caligula Jones
Reply to  Bjarne Bisballe
January 24, 2019 7:11 am

What’s that in baby polar bear deaths?

Michael S. Kelly, LS, BSA, Ret.
Reply to  Louis Hooffstetter
January 24, 2019 2:06 am

A kiloton of TNT is 4.184E12 joules (4.184E-9 zettajoules). “Little Boy,” the Hiroshima bomb, was estimated to have produced anywhere from 13 to 18 kilotons. You can do the arithmetic.

Louis Hooffstetter
Reply to  Michael S. Kelly, LS, BSA, Ret.
January 24, 2019 5:55 am

So at 15 kilotons, one Hiroshima (“Little Boy” bomb) equals 0.00000006276 zettajoules.
Or 1 zettajoule equals ~15,933,716 (or 16 million) Hiroshimas.

DD More
Reply to  Louis Hooffstetter
January 25, 2019 2:47 pm

Louis – Did some converting in the past back when the heat content was a claimed 2.5 EE23 Joules rise from 1970 to 2014. For comparison to a KM^3 of Glacial Ice vs the well known ‘Hiroshima bomb’ measurement.

By today’s standards the two bombs dropped on a Japan were small — equivalent to 15,000 tons of TNT in the case of the Hiroshima bomb and 20,000 tons in the case of the Nagasaki bomb. (Encyclopedia Americana. Danbury, CT: Grolier, 1995: 532.)
In international standard units (SI), one ton of TNT is equal to 4.184E+09 joule (J)

Hiroshima bomb TNT 15000 x TNT to Joules 4.18E+09 = Joules total 6.276E+13

To raise 1 KM^3 of ice from the average -30 °C to 0 °C then melt it, then 0 °C water to 4 °C average sea temperature => 413420 Joules/KG x 1000 KG/t x 1,000,000,000 t/KM^3 = 4.1342E+17 Joules / KM^3

So we can add another metric KM^3 of Glacial Ice. That conversion is
(4.1342E+17 / 6.276E+13) = # HiroBmb per Km^3 = 6587
That is correct. Place one Hiroshima bomb in a grid every 54 meters apart to melt the ice.

January 23, 2019 6:30 pm

A Zettajoule is a lot of heat. But the oceans are big and can take the heat. 🙂

Alan Tomalty
Reply to  kmann
January 23, 2019 10:31 pm

We all can take the heat. OOOps I forgot. The lefties and alarmists and alarmist climate scientists can’t take the heat but they can take 1 way trips to Mars on Elon Musk’s regularly scheduled flights. The flights will be only 1 way because Elon has to save money on ticket prices, by not having separate exit vehicles from Mars.

Nick Schroeder
January 23, 2019 6:33 pm

The average depth of the oceans is about 4,000 meters.
So, what’s happening in the bottom half?

paul courtney
Reply to  Nick Schroeder
January 23, 2019 7:42 pm

Nick: One version is, there’s a cat there. It could be both dead and alive.

My understanding of quantum physics.

Pillage Idiot
Reply to  paul courtney
January 24, 2019 8:19 am

Excellent point, Paul.

However, since it is the deep oceans – I think you should refer to Schrodinger’s tardigrade.

John F. Hultquist
Reply to  Nick Schroeder
January 23, 2019 8:05 pm

The bottom half is cold and changes slowly.
In the deepest parts, water temperature ranges from 34-39 degrees F (1-4 degrees C).
In a general sense, the most dense water sinks, and where surface waters cool more it stays near the surface, where ice forms. Because ocean water is not pure H2O, and most published tables assume that is what is of interest, one cannot take a number from a table and apply it to ocean water.

Caligula Jones
Reply to  Nick Schroeder
January 24, 2019 7:15 am

I saw a great documentary called “The Abyss” by reknown explorer James Cameron. Explains a lot…

Gary from Chicagoland
January 23, 2019 6:36 pm

What is the uncertainty of these ocean temperature measurements?

Patrick B
Reply to  Gary from Chicagoland
January 23, 2019 8:45 pm

The entire discussion is useless without proper error analysis.

Reply to  Patrick B
January 24, 2019 12:36 am

So true. What would those graphs look like with error bars?

Caligula Jones
Reply to  Graemethecat
January 24, 2019 7:17 am

A CNN headline with five alarms?

Just kidding, they will never show error bars. They have nobody on staff that would know what they are.

Reply to  Gary from Chicagoland
January 23, 2019 8:50 pm

The temperature is linearly related to the total stored energy, so at a 280K average temperature, a 1% error in the total stored energy represents a temperature uncertainty of about 2.8C. If we are generous and say that the relative error from year to year is only +/- 0.01%, this represents an error bar on each yearly change of about +/- 0.028C which is about 5 times the reported average change in the anomaly from year to year.

With the kinds of numbers involved, a trend of 0.05C per decade can arise from a bias of less than .002% per year.

Reply to  co2isnotevil
January 24, 2019 12:33 am

” If we are generous and say that the relative error from year to year is only +/- 0.01%”

You would be hard pressed to get that kind of accuracy from a highly sophisticated lab experiment. It is downright dishonest to pretend that you measure something as large and variable as the total majority of the water on Earth to that kind of accuracy.

David Ross Harmon
Reply to  Greg
January 24, 2019 4:33 pm

Even if you could, look at the time frame. That’s not even a blip on the larger scale. It’s a mild uptick in temperature range which may or may not be proceeded or followed by an even greater amount of downtick.

Reply to  Gary from Chicagoland
January 24, 2019 10:28 am

How many sensors and where?

The figures will, as usual, be extrapolated such that the average sensor covers 1,000,000 sq km (guessing?)

Dave Fair
January 23, 2019 6:39 pm

Thanks for your good and entertaining work, Bob. WUWT readers should go over to your blog, Climate Observations, and learn a lot about ocean dynamics.

Joel O'Bryan
January 23, 2019 6:42 pm

The conundrum the GISS-ters will have with their hand-massaging of LOTI to NODC anomalies is, “What happens when the oceans keep cooling under several weaker than average solar cycles?”

January 23, 2019 6:43 pm

Thanks Bob.

January 23, 2019 6:43 pm

The sun controls the oceans and the oceans control the land and the air. CO2 can come bubbling out of the oceans when they warm but that is all it gets to control.

James Clarke
January 23, 2019 6:57 pm

“Instead of whining about how much the Earth’s oceans have warmed, assuming that increasing levels of CO2 are solely responsible for that warming as is commonly believed…”

Why is it commonly believed that the increase in atmospheric CO2 is warming the oceans to the depth of 2,000 meters, when it isn’t even warming the atmosphere? Two thirds of the alleged ocean warming (0.041C) occurred from 2005 to 2014, a period of time when the mystery data set reveals no atmospheric warming at all!

If it is commonly assumed that increasing atmospheric CO2 is warming the oceans, while it is not even warming the atmosphere, then someone is going to have to explain how this miracle of physics is happening! The argument that the increase in atmospheric CO2 is just reducing the the amount of heat escaping from the oceans (hence, causing the oceans to warm) makes no sense if the air is not warming in the first place.

If the ARGO data is accurate and truly representative, then there must be other unknown factors causing the observed changes. The assumption that CO2 is solely to blame has no more validity than a claim that the oceans are warming due to invisible unicorn submarines.

Alan Tomalty
Reply to  James Clarke
January 23, 2019 10:50 pm

The supposed tiny amount of warming in the oceans is dwarfed by the error bars which alarmist climate scientists are always loathe to publish. Such a farce.

Anthony Banton
Reply to  Alan Tomalty
January 24, 2019 2:09 am

Wrong on both counts.
Try converting that “tiny amount of warming” into Celsius equivalent for the atmosphere.
Hint: you need to multiply by 1000.
So, 0.065C is 65C.
Tiny eh?
Another hint.
Temperature is NOT heat.

And as regards error bars….

Anthony Banton
Reply to  Anthony Banton
January 24, 2019 2:20 am
Dale S
Reply to  Anthony Banton
January 24, 2019 6:56 am

ARGO measures *temperatures* in the ocean. And the temperature changes are, as Alan says, tiny. The fact that it takes a massive amount of heat to change the vast oceans by a tiny amount of temperature doesn’t make that tiny change more noticeable to ocean organisms. The fact that that amount of heat would warm something else by a large amount is also irrelevant, whether you are talking about the atmosphere or the city of Hiroshima. Without a mechanism for that heat to magically decide to jump to the atmosphere (or Hiroshima) instead, all you are doing is trying to make a trivial change sound scary by presenting it in a false context.

Loren Wilson
Reply to  Anthony Banton
January 24, 2019 10:19 am

Temperature is heat when we are looking at heat stored in water. The enthalpy (heat) of water is a nearly linear function of temperature, since the heat capacity of water changes only slowly from 0 to 30°C (4.23 to 4.18 J/g-K, ignoring the PV work done on the fluid at depth).

mike macray
Reply to  Anthony Banton
January 25, 2019 8:29 am

Anthony Barton
Thank you!
Living for decades in coastal environments all over the world it has been my experience that the ambient air temperature seldom deviates more than 10º F either side of the water temperature which fluctuates seasonally lagging the sun by about three months. This would suggest a pretty efficient heat transfer between atmosphere and ocean. Given three orders of magnitude difference between their respective thermal capacities, a degree or two increase, or even six or ten, in the air would disappear virtually without trace into the ocean. What am I missing?

Robert of Texas
January 23, 2019 7:05 pm

I am looking at the temperature maps and wondering…

How the H*LL are they measuring temperature changes of less than 20 thousandths of a degree (C)? Those must be some really good thermometers and getting them all calibrated… Wow. (and keeping them all calibrated under the field conditions…)

Wait a minute…my skeptical alarm just went off. It is telling me they used the scientific method of “Wishful Thinking” to get and keep these miraculous devices calibrated. Between 2012 and 2014, it looks to me like the way data is collected changed to make for such a vertical jump.

Seriously, can someone answer me how they do this? Technology must be advancing faster than I ever imagined.

Robert of Texas
Reply to  Robert of Texas
January 23, 2019 7:15 pm

Oh, here is part of the answer…

“Further climate science analysis purportedly showed FS1 temperature profiles in a few hundred of the early ARGO floats were probably erroneous. Those floats were taken out of service, leaving just SBE sensors. All five ARGO float designs use current model SBE38 from 2015.”

Robert of Texas
Reply to  Robert of Texas
January 23, 2019 7:31 pm

And still more questions…

The instruments are supposed to be +/- 0.002 degrees C accuracy of temperature
+/- 0.005 psu using Salinity as a proxy (Red alert, proxy used)

Wouldn’t that be about +/- 0.007 total error since they are measuring temperature at specific depths? Which would be about +/-26% of the total measured difference (from zero).

This assumes that salinity is a PERFECT proxy? What is the ocean decides not to cooperate? What if there are areas where salinity is lightly different than expected? Like different by 20/1000?

I am not buying ANY of this. Too small a measurement versus too much that we don’t know and just assume.

January 23, 2019 7:08 pm

Hi Bob – This is excellent! Also, do you have access to the CMIP5 results to see how the models did on a regional basis? The 2018 Hausfather et al paper did not discuss their regional skill. The Argo time period is the robust time to do such a comparison as the data are temporally more homogenous and the spatial coverage is better. This would be a peer reviewed paper by itself.

Also, it would be useful to present not only in degree C, but as an average flux in Watts m-2 so as to compare with model predictions of TOA radiative imbalance, as I proposed in

Pielke Sr., R.A., 2003: Heat storage within the Earth system. Bull. Amer. Meteor. Soc., 84, 331-335.

Thanks for doing this very informative much needed analysis.

Roger Sr.

P.S. I tried to post a comment earlier from my Ipad but it disappeared. If it reappears as a comment, it is just an abbreviated version of this comment. 🙂

[both were held in automatic moderation, this one was approved since it was the more detailed of the two – Anthony]

Alan Tomalty
Reply to  Roger A. Pielke Sr
January 23, 2019 10:53 pm

Nice to see you back Anthony.

Dr. Strangelove
Reply to  Roger A. Pielke Sr
January 25, 2019 7:18 am

Dr. Pielke,

Based on my computation of deep ocean warming in W/m^2, only 32% can be attributed to CO2
world’s ocean area = 3.619 E+11 m^2
depth = 2000 m
volume = 7.238 E+14 m^3
mass = 7.238 E+17 kg
warming trend = 0.048 K/decade
change in temperature (2005-2018) = 0.048 (1.3 decades) = 0.0624 K
specific heat water = 4191 J/kg-K
heat energy = 1.89 E+20 J
time = 13 yrs (365 x 24 x 3600) = 4.1 E+8 s
power = energy/time = 4.62 E+11 W
flux = power/area = 1.28 W/m^2

CO2 forcing:
2005 = 380 ppm
2018 = 410 ppm
5.35 ln (410/380) = 0.41 W/m^2

Warming attributed to CO2
0.41/1.28 = 0.32

Dr. Strangelove
Reply to  Roger A. Pielke Sr
January 25, 2019 8:02 pm

Dr. Pielke,

I used this deep ocean temperature data to calculate thermal expansion of seawater, which is the main cause of sea level rise. My calculated sea level rise of 0.84 mm/yr is much lower than 2.9 mm/yr from satellite altimetry.

ocean area = A = 3.619 E+11 m^2
volume = V = 7.238 E+14 m^3
coefficient of themal expansion at 10 C = o = 0.88 E-4 K^-1
change in temperature = dT = 0.0624 K
time interval = dt = 13 yrs

change in volume = dV = o dT V = 3.97 E+9 m^3
change in depth = dx = dV/A = 0.011 m = 11 mm
trend = dx/dt = 0.84 mm/yr

I think the altimeter data are wrong. It is physically impossible unless we attribute sea level rise to melting glaciers.

Dr. Strangelove
Reply to  Dr. Strangelove
January 25, 2019 9:30 pm

The original Envisat altimeter data had a trend of 0.76 mm/yr before it was ALTERED. The original trend is close to my calculated 0.84 mm/yr

Anthony Watts talked about it back in 2012

Jo Nova too in 2012

I think we have a case of data manipulation

Chris Hanley
January 23, 2019 7:32 pm

According to Argo data since 2004 the oceans seem to be warming from the bottom up:

Chris Hanley
Reply to  Chris Hanley
January 23, 2019 7:54 pm

I see what I’ve done, this is the warming at different depths: 65N-65S.gif

Dr. Strangelove
Reply to  Chris Hanley
January 25, 2019 6:14 pm

All the depths are warming but it’s warming faster at greater depths. It doesn’t mean heat is flowing from bottom up. The bottom is colder than surface. It violates the 2nd law of thermo

January 23, 2019 7:35 pm

The data on graphs look fake after very intense massaging. If we had their raw data and plotted them there would be no curve and no trend.

January 23, 2019 7:47 pm

Based on the values of the linear trend line, the temperature of the global oceans for depths of 0-2000 meters has risen about 0.065 deg C since 2005.

0.065 deg C for 0-2000 meters would be very amazing accuracy considering for example the uncertainty in HadSST3, which is 0.03 deg C for only the sea surface.

It appears as though the combined effects of The Blob and the 2014/15/16 El Niño put an end to the cooling taking place in the Eastern North Pacific from 2005 through 2013.

Both events were solar forced via TSI when SC24 TSI was higher than my solar-ocean warming threshold, until TSI fell below the warming threshold in 2016, ie there was enough solar energy for net warming, until there wasn’t, and things turned the other way.

Bob your last 3 figures 11-13 are especially revealing, nice job.

Steven Mosher
January 23, 2019 8:10 pm

“Instead of whining about how much the Earth’s oceans have warmed, assuming that increasing levels of CO2 are solely responsible for that warming as is commonly believed.”


I wish you would not spoil your data presentations with this nonsense.

1. Scientists do not believe all the warming is caused by c02.
2. If you are referring (commonly) to other folks, then who cares what dopes think?

Chris Hanley
Reply to  Steven Mosher
January 23, 2019 9:08 pm

Berkeley Earth seem to think so:
” Human Effect: Many of the changes in land-surface temperature can be explained by a combination of volcanoes and a proxy for human greenhouse gas emissions. Solar variation does not seem to impact the temperature trend “.
That statement is backed up by charts that imply most if not all the warming is due to CO2 and by extension human emissions.

Jan Kjetil Andersen
Reply to  Steven Mosher
January 24, 2019 5:32 am

You frequently show that you have good insight in this field Steven, but you do not elaborate much in your comments.

One has to guess what you really mean.

Here I guess you mean that no scientist think that CO2 is the only climate gas; and you are of course right. We have CH4, N2O and many fluorinated gases as well.

Gary Ashe
January 23, 2019 8:13 pm

Whats the point of these articles ?.

The oceans cover 74% of the ears surface, does any sane person actually believe the average temperature of them all can be measured ?, its almost comical.
Like Monty python sketch.
And now the av temps at 2 kilometres deep, even if you can swallow that much b/s, how any rational person could swallow enough is beyond me.
Then they claim to be measuring the vast deep depths to 1000th of a degree C.

They should be ridiculed and mocked for the mentally unstable arse-wipes they are.

Tim Beatty
January 23, 2019 8:40 pm

I am puzzled. The land + ocean temp datasets are repreented as 0.1C per decade increase. But these deep ocean datasets are a tenth of that. What volume percentage is warming slowly and doesn’t that dampen sea level rise estimates substantially? Secondly, this also implies the gradient through depth is increasing. Don’t they measure the 0-2000 incrementally and are they seeing a gradient increase (and the thermal circulation and salinity changes a thermal gradient should create?)

Alan Tomalty
Reply to  Tim Beatty
January 23, 2019 11:02 pm

The Land Ocean temperature index measures water temperature at the surface.

Tim Beatty
Reply to  Alan Tomalty
January 26, 2019 3:33 am

Which is why I mentioned the incresing gradient. Do the current models support ocean wrming at a substantially slower rate thn the atmosphere?

Gary Ashe
January 23, 2019 8:51 pm

Tim they couldnt measure the av temp of gold fish bowl of water in laboratory to 2/1000ths of degree C, The argo floats are just prop’s in a magic show.

January 23, 2019 9:15 pm

A really great post in such detail and so well explained and documented. It was a pleasure to read. Thank you. Mr Tisdale is and has been for some time an expert on the ocean heat content topic.

January 23, 2019 10:05 pm

There are two reasons they use the meaningless zettajoules. If they expressed it in the commonly-understood Kelvin or Celsius, it would be so fractionally small everyone would wonder what the big deal was, which points us to the second reason. The Big Deal is demonstrating that the “missing heat” of the last 2 decades of essentially no atmospheric warming really exists and it’s in the oceans where they claimed it was in order to explain the embarassing Global Warming Hiatus that shows in all the un-tortured temperature recods.

January 23, 2019 11:14 pm

❶①❶① . . . An open letter to . . .

Gavin Schmidt from is a wise person.

He gives good advice.

If Gavin gives you some advice, then I suggest that you take it.

Gavin recently gave all Deniers some good advice.

If Deniers don’t want to be rightly accused of climate denial, they don’t go around denying climate science.

That bit of advice is beautifully simple. It can’t be argued with. It is logically sound. It is a statement that Yogi Berra would be proud of.

But Gavin is a busy person. He doesn’t have time to give everybody the advice that they deserve.

To solve this problem, I have managed to “get inside Gavin’s head”, so that everybody can benefit from Gavin’s wisdom.

First, some advice for Alarmists.

Alarmists, if you don’t want to be called a stupid arrogant jerk, then don’t act like a stupid arrogant jerk.

Now some advice for climate scientists.

Climate scientists, if you don’t want to be called an undemocratic third world dictator, then don’t act like an undemocratic third world dictator.

Even Gavin could benefit from some “Gavin” type advice.

Gavin, if you don’t want to be called an obnoxious Tamino-like character, then don’t act like an obnoxious Tamino-like character.

I think that I am getting the hang of giving out wise advice.

I think that I will quit while I am ahead, with one last piece of advice for Gavin.

Gavin, if you want to “weasel” out of answering any difficult climate questions, then call the questioner a Denier. Because nobody is expected to answer questions from Deniers.

Sorry, Gavin. I can see that you already knew that last piece of advice.

Keep up the good work !!!


An open letter to

Anybody who would like to read the conversation that I DIDN’T have with Gavin, should click this link:

Reply to  Sheldon Walker
January 24, 2019 12:21 am

Maybe this is why.


Reply to  SMS
January 24, 2019 10:31 am

That’s an excellent article. They measured how likely liberals and conservatives were to unfriend folks they disagree with. The beauty of that approach is that the subjects didn’t know they were being evaluated. They couldn’t pretend to be what they weren’t, which is what happens in surveys all the time.

Reply to  Sheldon Walker
January 24, 2019 1:13 am

Gavin is too busy to discuss climate with another qualified climatologist and prefers to walk of the TV set and refuse to be on the set at the same time as Dr. Roy Spencer.

This little display tells us all we need to know about the level of “wisdom” and the scientific objectivity of Gavin Schmidt.

Thanks for reminding us.

Geoff Sherrington
January 23, 2019 11:29 pm

Figures 7 and 11 above show evidence of data being used when it suffers from too much variability. A common cause of this is working too close to the detection limits of the systems. A demonstration of the effect is isolated bulls eyes and smooth contours breaking up into blobs. If people maintain the art of hand countouring and close examination of their data, they are more likely to understand this problem, which is a downside of computing taking inspection and relevance away from the researcher.
The Argo authorities claim an accuracy of 1 milliK for the platinum resistance thermometers and low milliK errors for the whole system at temperatures away from calibration triple points. Same for drift over 3 years, low milliK. It is much harder to assess accuracy under actual working conditions and conditions where there is difficulty separating random variability from systematic and attribution of causes. Often, the ocean variability factor requires arrays of instruments comparing each other over long periods and in different settings. If there has not been enough of this type of comparative accuracy study, then expect to see these telltale signals following data processing.
In several more years it will become clearer if the Argo design is found wanting, like XBT was before Argo. It might happen, happily, that Argo were good enough for purpose, which is good. In the meantime, however, it should not be concluded that they are good enough for important outcomes like global policies of mitigation.

Peter Foster
January 23, 2019 11:59 pm

Is this data raw or has it been adjusted ? Willis wrote several papers in the late naughties showing the oceans, and particularly the south Pacific, were cooling. Then he got leaned on by the warmists and had a eureka moment where he confessed his sins and found some very convoluted justification for adjusting his data to match Syd Levitas’s ocean heat content graph. Amazing.
So how kosher is this data ?

January 24, 2019 12:29 am

I think the following is close:

1 Hiroshima = 4.2 E9 Joules/ton * 20000 ton = 84E12 Joules
1 Zettajoule = 1E21 Joules / 84E12 Joles = 11.9E6 or about 12 million Hiroshima Bombs

And in solar input to Earth (No Atmosphere):
Earth Area (12.7E6 m)^2 x PI / 4 = 1.267E14 m^2
Tota Solar input to Earth 1.267E14 m^2 x 1.36E3 J/s = 1.768E17 J/s
1 Zettajoule = 1E21 J / 1.768E17 J/s = 5656 s or a little more than one and a half hour of solar input to Earth

Please check 🙂

Greg Goodman
January 24, 2019 1:18 am

It is unfortunate that the data aren’t subdivided by depths…that is 0-100 meters, and 100-700 meters, and 700-2000 meters.

It should be quite simple to get that by subtraction. Even your limited skills with a spreadsheet should enable you do get the 10-700 and 700-2000 datasets without anything too complicated. That would be worth doing.

Specifically look at where the heat came from and went to before and during the 1998 El Nino.

Reply to  Greg Goodman
January 24, 2019 5:41 am

NB, since temperatures are not fungible you would need to scale down the 0-100m values by a factor of 7 to more correctly represent the relative energy content OHC. You can subtract energies.

January 24, 2019 1:21 am

0.048 per decade? That’s about as doomed as we can get!
You probably won’t survive long enough to read this comment.

January 24, 2019 1:24 am

Three zones in N Atl looks a lot like a standing wave within that basin. Currently cold in the middle hotter at ends.

Pacific looks similar cooler around WPWP hotter nearer the poles.

A C Osborn
January 24, 2019 2:00 am

Take a look at the Argos data for the north it Chefio’s forum here

It shows more cooling over the same period, so the heat goes North and disappears into space via the atmosphere.

Anthony Banton
January 24, 2019 2:18 am

” To put deep-ocean warming in perspective, I’m presenting that data in this post.”

Except it’s not Bob.
That is nothing like the correct perception.
The vast science ignorant masses will convert *your* temperature to that they will experience outside in their gardens.
Temperature is not energy/heat.
Mass and SH need to be applied to make an equivalence.
So what would 0.065C convert to if the energy used were to be applied to the atmosphere (obs without the fact that the atmosphere is able to radiate to space)?

Greg Goodman
January 24, 2019 2:59 am

Here is the 0-700m vs 0-100m for tropical Pacific. 0-700m scaled up for comparison on the basis of uniform relationship between temp and OHC, which is I suppose the whole point of this dataset.

comment image

Reply to  Greg Goodman
January 24, 2019 3:19 am

It is interesting to see where the heat increases and decreases in relation to Nino/Nina years. 1998 was very different from the 20116 event. The huge build up of OHC occured from 1995-1997 and dumped out two years later.

All other El Ninos seem to be a mostly synchronous move of heat from deep to shallower waters.

Richard M
January 24, 2019 7:05 am

It is interesting to compare this data to SSTs and surface temperatures. When we look at the surface data there was no warming at all from 2005-2014. All the warming has occurred as part of the super El Nino since then.

The warming shown in the 0-2000 meter chart shows pretty consistent warming over the entire period. So, if AGW was the cause, how did it bypass both the atmosphere and the oceans?

Dave Fair
Reply to  Richard M
January 24, 2019 10:48 am

I’m too lazy to do the research, but where are the data for 700 m to 2,000 m depths? Since 0-700 m overlaps with 0-2,000 m, it would seem that 0-700 m data would “pollute” the 0-2,000 m data.

Reply to  Dave Fair
January 24, 2019 2:04 pm

No pollution. It’s like the GDP of Wales, UK and Europe. If you want UK without Wales you subtract 😉

Dave Fair
Reply to  Greg
January 24, 2019 3:19 pm

Only if the GDP of Wales is separately identified and not amalgamated within the UK’s. Since temperatures are not differentiated, you can’t just subtract 0-700 from 0-2,000 to get anything intelligible about 700-2,000.

January 24, 2019 7:40 am

The Mystery Dataset correlates pretty well with the S&P500 index over the same period. May we conclude that El Ninos (i.e., global warming) heat up the stock market?

January 24, 2019 8:53 am

It reminds me of Dr Spencer’s monthly UAH graph, in fact almost identical.

January 24, 2019 9:58 am

Two questions:

What does “depth-averaged” mean?

Why would Global Warming Climate Change affect one part of the Oceans differently than another part? Global means the whole thing, I would have thought…

Gary Hudson
January 24, 2019 11:31 am

Whenever Pacific ocean temps, and The Blob are discussed the presentation by Professor Wyss Yim: Geothermal Heat and Climate Variability should be considered. The link here: The Blob is discussed at about 24:30 of the video. Professor Yim makes a persuasive case that geothermal heat caused the Blob, and the subsequent warming in the western arctic.

January 24, 2019 2:14 pm

“One thing common to the papers is the use of the metric zettajoules for ocean heat content”

ANthropogenic CO2 provides 0.94W/m^2, a W * Second is a Joule, and water has the highest specific heat of all common materials at 4.186 joule/gram °C. In other words, CO2 ain’t warming the oceans, simply do the math. Not in a billion years…literally.

January 24, 2019 2:20 pm

Before Bob answered his own question I was thinking that the temperature profile looked a lot like the inverse of cosmic rays reaching the ground station:

Kevin A
January 24, 2019 7:05 pm

I have hunted around for why they have a temperature sensor capable of 0.002 measurements and came to the conclusion it must be for the CTD to work at all. Pressure changes with temperature, salinity must be really sensitive but the sources of this information I’ve found are behind paywalls. Currently my temperature sensors I use are NIST certified, 16-bit temperature result with a resolution of 0.0078°C LSB giving it a -20° to 50°C 5 times more accurate then a Vantage Pro 2 Plus.

%d bloggers like this:
Verified by MonsterInsights