Guest Post by Willis Eschenbach
I’ve been thinking about the Argo floats and the data they’ve collected. There are about 4,000 Argo floats in the ocean. Most of the time they are asleep, a thousand metres below the surface. Every 10 days they wake up and slowly rise to the surface, taking temperature measurements as they go. When they reach the surface, they radio their data back to headquarters, slip beneath the waves, sink down to a thousand metres and go back to sleep …
At this point, we have decent Argo data since about 2005. I’m using the Argo dataset 2005-2012, which has been gridded. Here, to open the bidding, are the ocean surface temperatures for the period.
Figure 1. Oceanic surface temperatures, 2005-2012. Argo data.
Dang, I like that … so what else can the Argo data show us?
Well, it can show us the changes in the average temperature down to 2000 metres. Figure 2 shows that result:
Figure 2. Average temperature, surface down to 2,000 metres depth. Temperatures are volume-weighted.
The average temperature of the top 2000 metres is six degrees C (43°F). Chilly.
We can also take a look at how much the ocean has warmed and cooled, and where. Here are the trends in the surface temperature:
Figure 3. Decadal change in ocean surface temperatures.
Once again we see the surprising stability of the system. Some areas of the ocean have warmed at 2° per decade, some have cooled at -1.5° per decade. But overall? The warming is trivially small, 0.03°C per decade.
Next, here is the corresponding map for the average temperatures down to 2,000 metres:
Figure 4. Decadal change in average temperatures 0—2000 metres. Temperatures are volume-averaged.
Note that although the amounts of the changes are smaller, the trends at the surface are geographically similar to the trends down to 2000 metres.
Figure 5 shows the global average trends in the top 2,000 metres of the ocean. I have expressed the changes in another unit, 10^22 joules, rather than in °C, to show it as variations in ocean heat content.
Figure 5. Global ocean heat content anomaly (10^22 joules). Same data as in Figure 4, expressed in different units.
The trend in this data (6.9 ± 0.6 e+22 joules per decade) agrees quite well with the trend in the Levitus OHC data, which is about 7.4 ± 0.8 e+22 joules per decade.
Anyhow, that’s the state of play so far. The top two kilometers of the ocean are warming at 0.02°C per decade … can’t say I’m worried by that. More to come, unless I get distracted by … oooh, shiny!
Regards,
w.
SAME OLD: If you disagree with something I or anyone said, please quote it exactly, so we can all be clear on exactly what you object to.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Willis Eschenbach says:
March 2, 2014 at 11:48 am
Egads, ferd, don’t make that kind of claim without a citation. What is your basis for the claim?
=========
The original Argo data showed cooling. The problem was attributed to a problem with some of the floats and these were removed. After the data no longer showed cooling. I believe the source for this was a published interview with Dr Willis, but it is too long ago to be sure. As far as I’m aware it was fairly well publicized at the time that the Argo record was initially showing cooling, which was seen as a problem because it was not what was expected. So a correction was done because the data did not match expectation. This points to an experimenter expectation problem in the design of the experiment.
Ulric.
In my original post I made no comment about strength or weakness of the trade winds so why did you raise the issue ?
I see that a negative AO gives weaker trade winds and a positive AO gives stronger trade winds but that is not inconsistent with my original post
However it is inconsistent with the other WUWT thread but that is not my position.
My position is that it could go either way depending on the relative strengths of the top down solar effect and the bottom up oceanic effect.
If both are in phase (inactive sun and cool oceans or active sun and warm oceans) the trade winds would be weaker but if they are out of phase (inactive sun and warm oceans or active sun and cool oceans) the trade winds would be stronger.
A the moment we have a weakening top down solar effect allowing the global air circulation to become more meridional.
At the same time we still have relatively warm oceans in historical terms so they are pushing poleward against the solar effect pushing equatorward.
Thus stronger trade winds whilst sun and oceans are in opposing phases.
george e. smith at 11:41 am
WHAT remarkable precision, are YOU talking about ???
From 6.9 +/- 0.6 E +100! Amps per kelvin, I get about 8.7% precision !!
What on earth do the units have to do with precision ??
I honestly don’t understand your point.
Willis reported:
Personally, I dislike the practice of measure OHC in units where you must watch the exponent and prefer to convert to Zeta Joules (10^21 Joules), so I rewrote it in units I prefer without changing the percentage precision.
69 ± 6 ZJ.
and then converted to the units of original measurement:
0.022 ± 0.002 deg C / decade
(at 27.5 ZJ per 0.01 deg C for 0-2000 meters)
[Willis, thank you for the charts converted to deg C in your 11:43 am reply to rgb.]
I said: 0.002 deg C / decade ” is remarkable precision, especially when the average temperature of the entire 0-2000 meter column runs from 0.000 – 10.000 deg C [one part in 5,000] and the surface runs from 0.000 to 30.000+ deg C “, and there is one Argo per 200,000 km^3 of ocean. Furthermore the Argo’s have a measurement precision of 0.005 deg C, a recorded drift in some tested bouy’s of 0.003 deg C in four years.
Not to mention what processing, cleansing, and adjusting might go on behind the scenes.
ferdberple at 9:34 am raises some concerns there, but citations are needed.
Maybe I didn’t express clearly enough that an uncertainty of ± 0.002 degC / decade is below our ability to measure it and is therefore somehow too small.
Rgbatduke made a related comment:
Like Willis, I don’t think the uncertainty is as bad as ± 1.0 C/decade, but I don’t think it can as good as ± 0.002 C/decade.
[correction to 12:55 pm]
units I prefer without changing the percentage precision.
69 ± 6 ZJ. per decade.
george e. smith says:
March 2, 2014 at 12:36 pm
Why all this sudden interest in the Nyquist sampling theorem ?
____________________________________________________
George,
My interest in adhering to the Nyquist-Shannon sampling theorem requirements is not driven by what clownish so-called climate scientists do rather, is driven by the discipline of proper scientific behavior. Also, it isn’t all of a sudden for me. When I leaned about confidence levels, sampling sizes, s/N, and error analysis, in my graduate work I became more fastidious about how I represented my work and more critical about “scientific vs artistic” renderings without the appropriate disclaimers. If the rendering is 10% data and 90% artistic, then it should be watermarked or annotated as such in my opinon. Sloppy science by Michael Manna and Phil Jones does not mean we ought not abide by scientific principles ourselves.
Willis,
in your Fig. 3 you give the change in ocean surface temperatures in the years 2005 to 2012 as +0.03 K/decade (global). There are quite a few SST data around. I took ERSST from KNMI:
http://climexp.knmi.nl/selectfield_obs2.cgi?someone@somewhere
and find (globally) quite a different number, on a rough estimate something like -0.05 K/decade for those years.
So I wonder about the ARGO data quality for SST data.
Sincerely, Werner Weber
physics, TU Dortmund University, Germany
RE: George E Smith: 12:36 pm.
George E. Smith brilliantly introduced Nyquist into the ARGO precision debate Jan 27, 2012 8:16 pm in the Decimals of Precision thread.
Wow, people are concerned about only 4,000 Argo buoys. Darn side better than one Yamal Charlie Brown Christmas tree ! +10
“Planet earth (Mother Gaia) does not compute averages; she integrates everything”
A running average IS an integral (of sorts). Calculates the area under a curve over a given time span. 🙂
Nice work Willis!
Stephen Wilde says:
“In my original post I made no comment about strength or weakness of the trade winds so why did you raise the issue ?”
Because stronger trade winds means La Nina, and that is during positive AO, which is when the jet is more poleward.
Willis Eschenbach says: “Egads, ferd, don’t make that kind of claim without a citation. What is your basis for the claim?”
Thanks, Willis. (Not the Argo Willis, the other Willis!)
ferdberple says: “The original Argo data showed cooling….”
That is 100% correct, ferd. The collective trend was downward.
“…The problem was attributed to a problem with some of the floats and these were removed….”
You may be right, but I don’t recall their being removed.
“…After[wards] the data no longer showed cooling….”
That is also correct, by my recollection.
“…This points to an experimenter expectation problem in the design of the experiment.”
As I remember it, Dr. Willis had a (heh-heh) “come-to-Jesus-moment” one night and realized that there must be something wrong with the numbers. The next day, he made some sort of “adjustment” and (vwalah!) the problem disappeared.
Another, more complete viewpoint is here: http://tallbloke.wordpress.com/2012/02/27/argo-the-mystery-of-global-warmings-missing-heat/
And, better yet: http://earthobservatory.nasa.gov/Features/OceanCooling/page1.php
From the latter: “Basically, I used the sea level data as a bridge to the in situ [ocean-based] data,” explains Willis, comparing them to one another figuring out where they didn’t agree. “First, I identified some new Argo floats that were giving bad data; they were too cool compared to other sources of data during the time period. It wasn’t a large number of floats, but the data were bad enough, so that when I tossed them, most of the cooling went away. But there was still a little bit, so I kept digging and digging.”
There’s more, regarding the XBT drop devices. See the last link.
ferdberple says:
March 2, 2014 at 12:40 pm
So based on nothing more reliable than a vague memory, you accuse Josh Willis of an “experimenter expectation problem”?
A bit of digging finds this. It discusses the underlying issue, which was that the ocean heat content data didn’t fit with the sea level data. They found two problems.
One was a subset of Argo floats that were reading too cool. The other was a problem with the calculated falling speed of the XBTs. The second of these is of no interest to this discussion.
The part that doesn’t fit your preconceptions is that Josh Willis had already written about the recent ocean cooling … so he was surprised and had to retract some claims when he noted that the purported cooling didn’t agree with the sea level data he had accepted as valid.
Near as I can tell, the 2005 inaccurate floats were removed, the data was corrected, and since then there haven’t been issues. Since the data I used starts in 2005, seems like none of this applies.
In any case, I fear that based only on your memory, you’ve made an unsubstantiated, unpleasant, and untrue accusation of scientific malfeasance against Josh Willis …
w.
jeffguenther8,
Re: ARGO cooling before ‘adjustments’…
That is also correct, by my recollection.
That’s my recollection, too. In fact, here is a pre-adjustment ARGO chart.
It’s funny how every government ‘adjustment’ of the data ends up showing either faster warming, or turns raw cooling data into warming. I am at the point where I don’t believe any of it.
The government needs to show us the raw, unadjusted data as a starting point. Then show how and why it was adjusted. We can figure out the rest.
Willis, many thanks for creating the animations. Having to work on something else this weekend, I found myself drifting back regularly to watch the “breathing”, “gurgling” and other movements. Even while I reminded myself that these are just computerized images and simplified depictions of two variables yet another level removed from reality, I kept having the same impression of a living organism that others here described.
It is fascinating how a spinning wet rocky spheroid orbiting a ball of plasma can create something so unlike Newton’s “universe as a machine”.
We are in debt to you for sharing not just your scientific inquiries but also the beauty you know how to find in a pile of data tables.
Stephen Rasey says:
March 2, 2014 at 12:55 pm
I just report’em as I calculate them. I gave the standard deviations and values for N above in the response to Robert Brown (rgb) above.
w.
and if the weight of a man were concentrated onto a shovel, the the blade would pierce 6 inches deep into the soil, the weight of a man normally piercing him little therein.
iron brian
<soarergtl says:
<March 2, 2014 at 2:26 am
<One of the responders to my Groan comments quoted Levitus at me:
<“Levitus (2012) thankfully puts it in context for the measurement-challenged:
<If this heat were instantly transferred to the lower 10 km of the global atmosphere it would result <in a volume mean warming of this atmospheric layer by approximately 36°C”
OK, I read the link Willis provided above, and it sounds reasonable. However, their chart shows the discrepancy happened well before the time frame in this link.
Like most, I want to believe the data. It’s OK by me if there is the slight ocean warming shown by the current ARGO data. In fact, it really doesn’t matter to me if there is runaway global warming. I just want to know what’s happening, and why. But NASA/GISS and others have been so manipulative with their data that it is very easy to assume the same thing is happening here.
Anyway, those are some excellent graphics, which show the various temperature changes. Very interesting. Also, not very alarming at all.
Paul Westhaver says:
March 2, 2014 at 12:26 pm
You’re looking at Nyquist backwards. Since it’s physically sampled at ~30000 samples per frame, we can’t detect anything changing on a smaller plysical scale than that. In fact, we can’t really detect anything changing on a smaller scale than twice that size …
But all that does is limit the resolution of what we can detect in the final image.
w.
Regarding the question of errors, I had an interesting thought …
There are ~ 3,500 Argo floats. They are on a ten-day cycle. That means that there are no more than 10,500 samples taken at each level per month.
Now, before I said that the standard error of the mean on the monthly averages was about a tenth of a degree … but in fact, that was assuming that because there are 29,404 gridcells in the ocean measured by Argo, that the number of samples was N = 29,404 … where in fact N is not more than 10,500, only about a third of that.
Hang on … let me recalc … OK, the errors need to increase by 1.7 [ which is the sqrt(29,404/10,500) ]. So the standard error of the mean of the monthly average surface temperature increases to 0.17°C, and for the average to depth increases to .02°C …
Note again that these are estimates of the MINIMUM uncertainty, because it is purely the statistical uncertainty. It does not allow for uneven coverage, particularly at depth. It does not measure the accuracy of the averages.
The same is true of the trends, that the stated uncertainty is the statistical uncertainty.
w.
Werner Weber says:
March 2, 2014 at 1:12 pm
I’m surprised by that level of agreement, actually. I just found the numbers you cited and analyzed them. They show a slight cooling, not statistically significant, of -0.03° ± 0.03°C per decade. The Argo data says it’s 0.01° ± 0.02°C per decade. Neither one is statistically significant, nor are they statistically different from each other.
w.
“””””…..RichardLH says:
March 2, 2014 at 1:24 pm
“Planet earth (Mother Gaia) does not compute averages; she integrates everything”
A running average IS an integral (of sorts). Calculates the area under a curve over a given time span. :-)…..”””””
You clearly do not understand what the integral of a function (sequence of events) is.
It is the sum total of the consequences of everything that happened (in that sequence); which doesn’t need to be a continuous or closed form function, it can be discrete events.
If I light a candle once a day, and measure the flame height, and do that for a million years, The effect of that on the table I put the candle on, will not be very significant (on average).
But if next 4th of July, instead of a candle, I set off a one megaton bomb on my table, the result on average will not change much. The integral of all those events will be quite different.
That difference is exactly what is wrong with climate “science”.
Earth’s weather and climate does NOT result from a continuous input of 342 Watts per square meter, on each and every single square meter of the earth surface 24 hours per day and 365 days per year, which is exactly what Kevin Trenberth claims it does; even in the dead of the Antarctic winter midnight. Earth is NEVER in a state of thermal equilibrium.
Watts is A RATE OF ENERGY FLOW / input /output / conversion / whatever ; INSTANTANEOUS !!
A megawatt for one second, does NOT produce the same consequences as one Watt for a million seconds.
The TSI is 1366 W/m^2, not 342 W/m^2, and as a consequence, points on the earth surface, reach Temperatures, that they could never even approach, at 342 W/m^2, so THINGS HAPPEN that absolutely never happen under Trenberth’s view of the earth.
On average, NOTHING happens.
The climate is NOT a linear system.
Many buoys read cold, but none read warm? Uh, yeah, OK. Not.
Willis,
A really illustrative presentation of the ARGO data! I’m mesmerized by Figure 2, having ‘zoomed’ it to max magnification to watch the 2D+time macroscopic circulation of ‘heat’ in the respective ocean basins and between them. The flow patterns I observed as a boy laying on creek banks in Wisconsin and watching small scale flow around obstructions, eddy currents, standing waves and more are all represented on the global scale in Figure 2.
Fascinating!
Mac
Interesting because it’s not so much the averages but the differential ocean warming that causes changes in local climate. Writing this from Western Australia, it’s easy to see why the last few years have been so warm during summer, whilst the rest of the planet isn’t.
George E. Smith says:
March 2, 2014 at 3:25 pm
Actually, a definite integral and an average are very, very closely related measures.
and
In other words, the average is the integral divided by n … it’s not crucial, just sayin’ …
w.