Where The Warmth Is

Guest Post by Willis Eschenbach

I got to thinking about the “hiatus” in warming in the 21st Century, and I realized that the CERES satellite dataset covers the period since the year 2000. So I’ve graphed up a few views of the temperature changes over the period of the CERES record, which at present is May 2000 to February 2017. No great insights, just a good overview and some interesting findings.

First, here are the raw CERES global average surface temperature data, the seasonal variations, and the anomaly that remains after removing the seasonal variations.

ceres plotdecomp surface temp 2017.png

Figure 1. Seasonal decomposition of the CERES surface temperature data. Statistical results (bottom line) are adjusted for autocorrelation using the method of Koutsoyiannis.

So … what are we looking at? The top panel shows the raw data, the actual temperature variations. The middle panel shows the repeating seasonal variations. The bottom panel shows the “residual anomaly”, the variations that remain once we’ve removed the repeating seasonal component of the signal.

The bottom panel, the residual anomaly, is the panel of interest. You can see how little the temperature has varied over the seventeen years of record. The El Nino of 2016-2017 is quite visible … but other than that there isn’t much happening.

There is one thing that is interesting about the residual … other than warming as a result of the 2016-2017 El Nino, the temperature anomaly only varied by about ± 0.2°C. Among other places, I’ve discussed what I see as the reason for this amazing stability in a post called Emergent Climate Phenomena.

The next question of interest to me is, where is the temperature changing, and by how much? Here is a Pacific and an Atlantic centered view of the warming trends recorded by CERES, in degrees C per decade.

CERES Surface Trends 2017 Atlantic.png

CERES Surface Trends 2017 Pacific.png

Figure 2. Temperature trends around the globe.

So … what is of note in these global maps? Well, both the poles are unusual. The area around Antarctica is cooling strongly, and the Arctic is warming. Presumably, this is why we’re getting less sea ice in the North and more sea ice in the South. It also affects the hemispheric averages, with the Northern Hemisphere warming and the Southern Hemisphere basically unchanging. Figure 3 shows the average decadal temperature trends by latitude band.

ceres decadal temperature trends 2017.png

Figure 3. Average decadal temperature trends by latitude band.

As you can see, the only parts of the planet where the temperature is changing much are the Southern Ocean around Antarctica, and the area above the Arctic Circle.

Next, in Figure 2 you can see that the North Atlantic is generally cooling. On the other hand, the Pacific is mixed, with areas of slight cooling and other areas of slight warming. Go figure.

On land, northern Russia, parts of the Sahel, the Gobi, and western Australia are warming. On the other hand, the upper Amazon is cooling strongly. So it looks like some (but not all) deserts are warming, and some (but not all) tropical forests are cooling … why?

I haven’t a clue. In my opinion, the most important words that anyone studying the climate can learn to say are “I don’t know.”

At the end of the story, I’m left with my usual amazement at the stability of the system. Despite being controlled by things as evanescent as winds, waves, and clouds, the temperature anomaly doesn’t vary more than about two-tenths of a degree. Nor is this due to “thermal inertia” as many people claim. Look again at Figure 1—the temperature changes by four degrees C peak to peak in the course of a single year, and changes by a degree and a quarter C in a single month, but the anomaly barely budges. To me, this is clear evidence of strong thermoregulatory systems, but of course, YMMV …

Sunshine today after rain, the Pacific ocean glitters in the far distance, the earth abides …

Regards to all,


PS—As always, my polite request is that you QUOTE THE EXACT WORDS THAT YOU ARE DISCUSSING, so we can all be clear about what you are referring to. Please be aware that while my request is polite, if you ignore the request I may say unflattering things about your ancestry, commenting habits, or cranial capacity … be warned.

DATA: For the temperature data I have used a straight Stefan-Boltzmann conversion of the CERES EBAF Edition 4.0 datafile showing upwelling longwave radiation. The dataset is available here. I have checked and compared this temperature dataset to a variety of other temperature datasets (HadCRUT4, Reynolds SST, HadISST, TAO buoy data) and found very little difference.


newest oldest most voted
Notify of
Peter Sable

I’m curious why this analysis requires the CERES data. Hasn’t RSS and friends been around since 1977?


I was going to ask the same about the UAH data.


RSS and UAH don’t cover latitudes above +85 degrees, below -85 degrees and, in the cases of TLT and TMT (lower and mid-troposphere), some areas with land above 1500 m altitude.

Peter Langlee

Ceres covers the poles?

Roy W. Spencer

As far as I know, CERES does not measure surface temperature. Could be estimates from the MODIS imager that the CERES team passes through… but I wouldn’t trust MODIS infrared for surface temperature because it’s too prone to cloud contamination, and so they only have temperature info on clear days… an obvious source of bias.

Walter Sobchak

Willis you should have put the information in the last graf up towards the top. I won’t speak for Dr. Spencer, but when I hit the P.S. line, If figure the post is over and that I can move on.


Regardless, CERES data seems to agree more closely to with UAH than RSS. I suspect that remain the case and RSS will continue to diverge over time.


Without checking the orbital details, I think both platforms probably have similar orbital inclination: near pole but not actually over the poles.
The reason that UAH and RSS don’t cover higher latitudes is because of gaps in the data. There is some data but are too many holes to get a representative coverage.
Coverage may come down to viewing angle.of the instruments used and the swath overlap on the ground.


Roy, you really should read to the end of the post before commenting …

Willis, something as fundamental as that should be in the article, probably near the top, not in a PS after the usual boilerplate “please quote me” spiel.

PS—As always, my polite request is that you

When I got that far, I said, yeah OK, the usual ( perfectly reasonable ) Willis closing paragraph and stopped reading because I’ve read it a hundred times. I did not expect crucial scientific information relating to the scientific content of the post at that stage. So I also missed the vital explanation of how you got temperature from CERES and was under the mistaken impression that was a data field provided by the CERES folks, not a home spun derivation. I will now delete the graph I just saved.
Please be clearer, preferably in the intro when doing that kind of thing.
Many thanks to Dr Spencer for drawing attention to that.

DC Cowboy

AFAIK neither do any of the other temperature data sets. I think what Willis meant to show was that the temps represented in those data sets didn’t vary much from the CERES data, not that they were accurate measurements of temps.

Please, clearly indicate on the Home page that an article is by Willis, so it could be safely skipped without looking inside. Thank you.


Aw, you no get along with W? Opening the article and checking the top line for author too much effort?

Wow. Two wasted clicks. You must be spent.


LMAO. Good one Max.


How ludicrous to assume you can never learn anything, even from somebody who might be wrong. But that’s Alarmists for you, closed minds, arrogant attitudes.


The first line of the summary on the home page clearly indicates the author.

Should we also make the same adjustment for your comments, so people can skip over them?

NW sage

Perhaps Alexander’s comments should be printed in invisible ink?

Walter Sobchak

Back in the paleolithic days of blogging, there was a blogger who disemvowled obnoxious commenters. She had a script that would remove all of the vowels from their comments, rendering the comments unintelligible. Just saying.


I have a Greasemonkey script on my laptop that changes the text color of any poster I add to its blacklist to Red. That way I can still read them if I want to, but I’ll know ahead that it probably won’t be worth my time.
Of course, I then read most WUWT articles on my phone.

Crispin in Waterloo but really in Ulaanbaatar

Anthony, that is exactly what I was thinking by the time I got to the 5th ad hom by Bradley. I started skipping because of my experience with his earlier junk comments.
Fortunately each comment has a bolded “Rob Bradley” at the top left informing me that it is time to flick the wheel (I have one of those mice with a no-click centre wheel – makes for faster fly-bys). It is the ultimate in efficient no-click browsing.
Willis: Thanks for the charts. I want to cite them in Atlanta in August. At best, CO2 might promote some regional warming, but it is certainly not global. It can be argued that is it sub-regional at best. It is also sub-annual because it is only in winter that one can trace some upward evidence.
Catastrophic anthropogenic sub-regional polar warming in winter? CASRPWIW? Not enough there to frighten the kittens.


What’s worse, Rob Bradley, is that you’re wasting everybody’s time. On and on and on it goes.
And you complain that Willis’ prose isn’t worth anybody’s time??
How can I print “CATASTROPHIC HYPOCRISY” in letters 10 feet high?

Mr. Watts, my name always appears above my comments, and anybody who doesn’t like them, can skip them. We are talking about articles, not comments here. I don’t want even to look inside any articles posted by Willis, and I sincerely hope that he would never forget to put his name on the home page again. Willis knows very well, what he is guilty of, and I will never forget it.


I agree with Alexander that a guest post should be clearly labelled. That omission has been promptly correct and hopefully more attention will be paid to this.
I don’t know what he thinks W. is “guilty” of and don’t really give a damn. But thanks for flagging the omission.


Jeeze, Rob…. The message went over your head by 10 feet, didn’t it?
Your stupidity… it burns.
But you’re not alone–you’re in good company with Mr. Feht.
(BTW, I can easily print 10-ft high letters on a Calcomp drum plotter, which has been available since the 80’s, and link them together in one big sign, proving you’re as ignorant as you are stupid.)
Again, you’re just wasting everybody’s time.


When I was in high school, I went on a field trip to a company that printed bill boards.
They could easily print 10′ tall letters.

Robert from oz

Reminds me of a fake test given to first year plumbing apprentices, the header stated read all questions before starting test ! Last line on the second page said ” to complete test only answer question 1 on page 1 ” , only one apprentice passed the test .
And the purpose of the test was just a lesson in reading paperwork from start to finish .

donald penman

I thought I was the only one Alexander , I don’t like being critical but people who have a high opinion of themselves only want to talk about themselves and I get bored.

Bruce Cobb

What’s hilarious is that the Warmunists have now changed their tune to “we should be cooling now”, implying that cooling would be good, but also deftly moving the goal posts in one fell swoop. Impressive.


“On land, northern Russia, parts of the Sahel, the Gobi, and western Australia are warming. On the other hand, the upper Amazon is cooling strongly. So it looks like some (but not all) deserts are warming, and some (but not all) tropical forests are cooling … why?”
Cloud cover springs to mind?


Human activity might actually be having an effect in the Amazon. But not much.

Richard M

B.j., I agree that it has to somewhat to do with clouds. Dr. William Gray believed that adding more CO2 will lead to higher evaporation (caused by increased downwelling IR) which will work as a negative feedback. This would occur over rain-forests but not over deserts. Hence, the deserts would warm while the rain-forests could very well cool due to a strong negative feedback (clouds and reduced high altitude water vapor).


And as Dr Spencer pointed out the way W. has derived the “temperature” data probably is being biased by cloud cover problems with the coverage in CERES data. This is not CERES temperature data produced and endorsed by those running the project, it is Willis’ homespun, non PR “temperature” data.
If CERES allowed a reliable temperature field to be calculated I would expect the CERES guys to be marketing it.

J. Seifert

The answer is provided within a 10,000 year context in paper Holocene
part 8, AD 1600 to AD 2050, at
This study starts with paper part 1, at 8,500 BC.
Willis, let Anthony publish it, run it through your computer analysis.
The given figures are correctly calculated, feed them into your system,
and a wonderful job is awaiting you to shred a paper series into pieces –
let Anthony give you the possibility to show that the “Californian tiger”
has strong teeth to bite…. Cheers from the author, JS.

Robert of Texas

It looks like using a bunch of sin waves to match a curve… I don’t understand how that explains the climate changes, but instead seems to just model them.
You prediction is that it should continue to warm, or at least not to cool until 2049 (+/- some error margin?). The problem with these kind of predictions is that I will be dead before I know the answer… 🙁 Its easy to make claims that are far into the future – no one will remember if you are wrong – this is a common tactic used by the AGW crowd (or whatever they call their religion these days). To have any immediate value you need to make some predictions that can be measured in a few years (10 or less would be nice).
I have been following some people’s prediction that due to the solar activity, we will be entering a period of cooling somewhere in the early 2020’s. If it does show any degree of cooling for several years, this would be proof that your hypothesis failed?

Can such a short time series have any significance whatsoever?

O ye of little patience and too much science!
All will be revealed in full technicolor next year, as you’d know had you taken the time to read The Future History of The Climate Debate:
Trenberth Travesty seen from space

The centrepiece of Nature’s April cover story is a stunning panorama of the Trenberth Travesty, stitched together from satellite imagery of the famous “missing” thermal energy.
“Using the hermeneutics of quantum gravity,” report the authors, “we are at last able to visualize this tricksy, mercurial zone of heat exchange whose 20,000-km front stretches from Cape Illusionment in autumn to The Isle of Mann in fall.”
The latest scientists believe the Travesty acts by thermal subterfuge to “launder” Kelvins from the atmosphere all the way down to the bathyclimatic ecosystem of the ocean floor, converting them to Hiroshimas en route.


How many Hiroshimas in a Travesty? Or is the unit of heat missing in the deep sea a Trenberth?

I’m a science communicator, so, much like the authors of the Future History of the Climate Debate, I am far too busy to hand-hold everybody who doesn’t understand things.
Do your own homework: convert Kelvins to Nagasakis at room temperature and 1 atm using Clausius-Capeyron, then Steffen-Boltzmann / Navier-Nick Stokes should get you the rest of the way from mNgs to microTrvs via deciHss.
It should take you all of 2 minutes, assuming you’ve put in the several years of college-level reading required to opine competently in a forum like this, where real scientists occasionally stoop to lurk.
You have, haven’t you?


I’ll freely admit that my Loopy Transform Functions are a bit rusty, which is why I asked for help.

if I explained it to you I’d have to explain it to everybody. And that, of course, is precisely what denihilists want: to bog real scientists down in the futile attempt to teach elephants to dance instead of spending time in the lab, advancing our fundamental understanding of the big settled questions.

Mary Brown

Despite having many years of college in “climate science”, I have no idea WTH Chimps and Keyes are talking about. Maybe it is the Sat night wine. Maybe it is them. 🙂

Mary, you might want to rethink the first word of your comment (despite?)!


Chimp, don’t hold your breath, he’ll always obfuscate, so no wonder Mary is a bit confused.
He never quotes the lapse rate in degrees Fahrenheit per fkn furlong like normal scientists do.
Not that it matters any more though because we’ve been educated recently by Navier-NStokes about CO2 not having any effect where the CO2 actually is, so fk the lapse rate.
Climate communicate that Keyes


/ Navier-Nick Stokes …

I also wondered if they are related.


Can if find the missing hot spot(s) in the upper troposphere over the equatorial band as the physics upon which the models demand?

Missing hot spots? oh dear.
Had you taken the time to sit some basic Theology or Divinity 101, as all serious climate communicators from President Gore to Reichsführer Dr Cook have done, you’d understand the cliché that absence of evidence is not evidence of absence.

Robert of Texas

Oh gawd… ROFL You are killing me… Plz stop! 8-D


rah, did you not look at the maps provided, the tropospheric hot spot and the missing heat is hiding in plain sight in the arctic 😉


… absence of evidence is not evidence of absence.

There is some evidence that you have been drinking absinthe.


good grief..we better do something….it’s getting incredibly hot ….where they have no thermometers


Why has northern Russia warmed in recent years.?
The sharp decline in spring snow extent is likely to have had a large part to play in that.
As land surface free of snow cover warms up itself and the air lot quicker then snowfields with the coming of spring.


would that be a positive feedback that the warmists are always warning people about?


Yes l agree its a positive feedback.
l just don’t agree with the warmists claim that its been due to CO2 levels. Far more likely its been due to changes in wind patterns. With increases in warmer southern winds and a decrease of Polar air flowing over the area.


Nice thought. But wrong.


That chart is for February snow cover not for the spring.
lf you check the Rutgers spring snow cover chart for Eurasia. You will see there has been a sharp decline in spring snow cover over the last 50 years.

Richard M

Russia sits right next to the Arctic with the areas of missing sea ice close by. The ocean heat released has a short hop to venture into Russian air space.


A plausible explanation as good as any. Thanks.


Hi Willis,
When do you expect your thermoregulatory system to start working? The top figure gives a temperature increase of 0.1 degrees per decade. Which given a few years will exceed your stated limits of +/- 0.2 degrees? Or is your claim that over any 20 year period the temperature will remain within a 0.2 degree limit
– a position completely consistent with global warming allowing you be right while raising CO2 levels continue to cause the temperature to increase.
Do you want to claim that over a 40 year period the trend will be zero? Or is there a measureable trend over a 40 year period?


Since Earth has probably warmed only about 0.5 degree C since the end of the LIA (rather than the cooked book figure of 0.8 degree often cited), the trend for that period is around 0.03 degree per decade.
As with all previous warm intervals during the Holocene, the Modern Warm Period has contained cooling cycles as well as warming. Each lasts about 30 years. The warming from c. 1977, when the PDO flipped, until the past decade has been no different from prior instances, the early 20th century and mid-19th century warmings, separated by cooling cycles. The world is already cooling again, although the El Niño of 2016 obscured its effect.
Arctic sea ice, for instance, has been growing since 2012. When the AMO enters its cool phase in a few years, the effect should become more pronounced.


Hi Chimp,
The world has warmed significantly more than 0.5 degrees since the little ice age (although I have no idea
when you think it ended). And I would like to see your evidence that the world is cooling. Willis analysis above suggests that it is currently warming at 0.1 degree per decade since 2000.

“Germonio March 24, 2018 at 2:45 pm
Hi Chimp,
The world has warmed significantly more than 0.5 degrees since the little ice age (although I have no idea
when you think it ended). And I would like to see your evidence that the world is cooling. Willis analysis above suggests that it is currently warming at 0.1 degree per decade since 2000”

Warming periods in the 1880s, 1930s-1950, and 1990s-2017 that are interspersed with cooling periods.
Your claim for greater warmth is evanescent.
Willis has demonstrated CERES data for the period from 2000-2017, which everyone basically agrees was a warming period.
Trillions spent and CO₂ effect is still unproven.
Willis’s Earth Temperature control processes are ever active.
Willis has published several times here explaining Earth’s temperature control method explicitly.
Even you are capable of searching for and reading those articles.


No, it hasn’t. In 1995, even alarmists agreed on 0.5 degree C. Since there has been essentially no warming since then, and that figure was probably high, the real number is around that level.
The traditional end of the LIA is given as 1850, but dates before and after that year have been suggested for the first warming cycle of the Modern WP. Earth cooled dramatically from the 1940s to late ’70s, as it also did in the decades before WWI.


Oops. Forgot to reply about start of global cooling. That would be around 2007, so if it has still warmed since 2000, that should change after the effects of the recent ENSO swing wear off.


As nearly as I can reconstruct them from spotty raw data, here is my take on the alternating warm and cool cycles of the secular Modern Warming interval:
W: 1857-87
C: 1888-1917
W: 1918-44
C: 1945-76
W: 1977-2006
C: 2007-
The early 20th century warming was probably hotter than the late 20th to early 21st century interval. So far the Modern Warming has been cooler than the Medieval, which was cooler than the Roman, which was cooler than the Minoan, which was probably cooler than Holocene Climatic Optimum peak warmth.
Previous secular warm and cool periods of the Holocene and prior interglacials show similar countertrend cycles.

“Chimp March 24, 2018 at 4:37 pm

Your numbers are much more detailed and accurate than my rough statements! 🙂
Thank you!
You’ve well demonstrated, Chimp, that warming cycles are followed, unfortunately, by cooling cycles. While the overall trend may be higher since the Little Ice Age, there are concerns that cooling is masked by adjustments and UHI.
From a Spring post:comment image
And “Plummeting March 24 Temperaturescomment image


Even the IPCC only claims 0.8C, Germino, are you calling the IPCC a bunch of liars?


“I haven’t a clue. In my opinion, the most important words that anyone studying the climate can learn to say are “I don’t know.””
Well said. All the best to you.
Total intensity as at 2015:
Predicted annual rate of change of total intensity for 2015 – 2020:
Blue rags to cans of red bull …
http://www.geomag.bgs.ac.uk/images/image022.jpgcomment image


yes it should be I dont know why mars, pluto and jupiter are warming, instead we get some convoluted story about wind and storms, and orbital shifts over 50 year periods..


Or satellites flying through magnetic anomalies.


Willis, the temperature trends at latitude is a chart I’ve not seen before. Very interesting- thanks.




Why has northern Russia warmed in recent years.?
Because more people live there now than ever before. Bigger cities warming the rivers they live next to kept free from ice as long as possible like the sea routes with huge icebreakers to connect those cities.
Looking for power plants?


Robertvd, March 24, 2018 at 3:03 pm
“Why has northern Russia warmed in recent years.?
Because more people live there now than ever before. Bigger cities warming the rivers they live next to kept free from ice as long as possible like the sea routes with huge icebreakers to connect those cities”.
Actually no! More people do not live in Russia than before. I don’t generally trust Wiki but it will suffice in this case.
“Low birth rates and abnormally high death rates caused Russia’s population to decline at a 0.5% annual rate, or about 750,000 to 800,000 people per year from the mid-1990s to the mid-2000s. The UN warned in 2005 that Russia’s then population of about 143 million could fall by a third by 2050, if trends did not improve”.
Demographics of Russia – Wikipedia


Could it be a mini bipolar seesaw?


The AMOC is a major player because it connects both poles.

JJM Gommers

Deserts and forest are subject to prevailing weather patterns and can be different in cooling/warming.


The Great Pacific Garbage Patch looks really hot.

“On land, northern Russia, parts of the Sahel, the Gobi, and western Australia are warming. On the other hand, the upper Amazon is cooling strongly.”
I have a gadget here which will show trends in surface measurement (either unadjusted or homogenised).It shows various time periods; the nearest to this is 1997-2016. It shows measurement points and the triangular mesh used to interpolate. The original gadget is a trackball like Google Earth. Here is a snapshot of unadjusted data showing N Atlantic, poles, Sahara etccomment image
The N Atlantic cooling does show out. The Sahara warmed, but is part of a band of warming extending from the Arctic through Eastern Europe. The US is a patchwork, mostly showing the variable effect of unadjusted TOBS.

Your image is much warmer than the images Willis displays.
Willis’s graphics show the Northern hemisphere as warming 0.22°C per decade. Your graphic Nick show a temperature over 2.5°C.
Apparently, GHCN V3 and ERSST includes infilling.

“Apparently, GHCN V3 and ERSST includes infilling.”
No, I am showing the trends of the stations as marked – ERSST are grid centres. The shading is linear in triangles between correct colors at the nodes.
The trends, as marked, are in °C/Century. So they aren’t that different.

Your legend “Deg C/Cent” translates as “Degree Celsius/Centrigrade”.
Not that one can tell exactly what “red” equals, since at 2.5°C the color is yellowish orange. Plus, it looks like the scale is increasing since the distance from 2.0°C to 2.5°C is much smaller than the scale at lower temperatures.
CERES during a warming period runs at 1.2°C.
Once several periods of cooling and warming are properly tracked, there may finally be an accurate rate °C per century.
Way over 2.5°C per Century? Extremely doubtful.
Your graphic states GHCN V3 and ERSST, I simply copied what you listed.
Nor does your answer unequically state that there is not any infilling.
Does that claim of yours “the trends of the stations as marked”, mean you downloaded the station data from the stations?
Or is it the data as entered into the GHCN V3 system by NOAA/NASA

Methods for removing inhomogeneities from the data record associated with non-climatic influences such as changes in instrumentation, station environment, and observing practices that occur over time were also included in the version 2 release (Peterson and Easterling, 1994; Easterling and Peterson 1995). Since that time efforts have focused on continued improvements in dataset development methods including new quality control processes and advanced techniques for removing data inhomogeneities (Menne and Williams, 2009)”

Both historical and near-real-time GHCN data undergo rigorous quality assurance reviews. These reviews include preprocessing checks on source data, time series checks that identify spurious changes in the mean and variance, spatial comparisons that verify the accuracy of the climatological mean and the seasonal cycle, and neighbor checks that identify outliers from both a serial and a spatial perspective”

It does not appear that “unadjusted” means what you think it means.

” mean you downloaded the station data from the stations?”
I used the GHCN V3 tabulation. For these years, that comes directly from the CLIMAT forms, as submitted by the met offices and displayed here.
“It does not appear that “unadjusted” means what you think it means.”
I know very well what it means. I have extensive experience with the set and its errors, as described here. It is unadjusted. Values thought doubtful are flagged, not removed or altered, no matter how ridiculous. Errors are often dramatic, like a sign or decimal point error. In my own work, as here, I removed all quality flagged readings. That is my choice. They are a very small percentage. Details for 2010-2014 with all flagged stations are in that link. There are in total 1101 flagged data (listed in detail) in 53 months, out of about 2000 stations per month.
“Not that one can tell exactly what “red” equals”
If you go to the app, you can click on any station to get name and trend, in numbers. The color scale has to be non-linear, otherwise most of the color range goes to separating the extremes.

Sorry to inform you but CERES also infills.

Reg Nelson

Of course, there are no surface temperature measurements for 75% of the Earth’s surface, so your graphs and analysis are totally worthless and meaningless.

Of course, you always say no-one knows anything about anything. I use the ERSST data for oceans, used by most climate indices.

Steve Fraser

Does CERES always take 7 months to get a month’s dataset out there? According to the site, the last Version 4 set is from last June.

NW sage

In your 3rd chart most of the variance is contained in a band plus or minus 0.2 deg C around nominal. Can this variance band be considered the sum of all errors and uncertainties in the instrumentation and measurement systems used to acquire the original data? It seems possible to me. [But what do I know, I’m just a metallurgist with a failure analysis background].
Good work


Deserts warming more? Isn’t that what you would expect from increasing CO2 in areas where the water vapour isn’t already soaking up all the OLR?
Willis – do you have the tech to check daytime and nighttime temperature trends from CERES data? It would certainly be interesting if all the trend growth was because if slightly less cold nights..

Thomas C Bakewell

Hi honoured sir! Is there any way you could display your findings on polar stereographic projections for both poles?
Tom Bakewell

“At the end of the story, I’m left with my usual amazement at the stability of the system”
Exactly. A system of finite volume with finite heat flow from a source, will experience a steady state firmly anchored around the mean values. And gh-gases have nothing to do with the temperatures in such a steady state.
The laws of thermodynamcs determines the state, and gh-gases are not included, because they don’t add any energy to temperature.

“At the end of the story, I’m left with my usual amazement at the stability of the system”
Exactly. A system of finite volume with finite heat flow from a source, will experience a steady state firmly anchored around the mean values. And gh-gases have nothing to do with the temperatures in such a steady state.
The laws of thermodynamcs determines the state, and gh-gases are not included, because they don’t add any energy to temperature.


Once you start averaging tgen detaiks are easily lost. Willis says source is ” pole warming”. Herecomment image decadal changes are confined to seasons. The chart is pretty much why CO2 is an undetectable impact. But perhaps ozone can be. Erl has lots more to say here. https://reality348.files.wordpress.com

There’s only so many Joules to go around and for one region to cool, another must warm. As for why the N preferentially warms while the S preferentially cools, this could be an artifact of the S receiving more solar energy than the N owing to the alignment of precession with the seasons and/or the relative ratio of land to water between hemispheres.


The SH is 81% seawater; the NH 61%, for the well known global average of 71%.

CERES seems to be responding to moisture. Higher moisture, the cooler.

Richard M

This could very well be the negative feedback from increasing CO2 predicted by Dr. William Gray.

Warren Blair

Wait I’ve just got to move a set of goal posts . . . ok now 400 ppm is preventing global cooling!

Michael Carter

I very recently came across a most interesting historical BBC documentary on YouTube. It is about one hour long and covers the 1963 snow blizzard in the UK. It was compiled during the event.
The impact on transportation, food and water distribution, sewage/wastewater and fuel supplies was crippling after but a few weeks. At one stage the commentary makes the point that a Russian weather station above the arctic circle had registered higher temperatures than some recorded in the UK at that time. Interesting.
I don’t know how to find a link for this documentary as it has been pulled from YouTube, possibly due to copyright.


Climate Alarmist is Playing San Francisco Judge as a Complete Fool
Dr. Myles Allen must think that the San Francisco Judge is a complete fool. I just finished a post refuting many of his claims, but one example needed to be singled out. In his presentation, Dr. Myles Allen replaced the poster child Mt. Kilimanjaro, which was exposed as a fraud in the Climategate emails, with the Glacier National Park Glacier. He claimed that man-made global warming is the cause of the decline of the glacier.

So upward radiation increased near the N pole and decreased around the southern ocean, perhaps because of changes in the frequency and duration of cloud cover? Perhaps because of changes in heat carried to and above the Cloud Condensation Level? Something else? That’s something, but I don’t see it having any straightforward implication. It reads like the start of an investigation, where you confirm that you can download, manipulate, and graph data. If it’s real, and your analysis suggests that it is, then it is worth pursuing.

Willis writes, “what is of note in these global maps? Well, both the poles are unusual.”
I think an important part of the difference between the two poles is because of the north polar and south polar snow and ice albedo differences. The south polar snow and ice is pristine. The northern is covered with a very fine layer of soot, sometimes more.
The melting from the soot albedo change causes additional albedo change and warming, a positive feedback. Look at the photographs of Greenland dirty ice and snow.-

from the datasets data quality manual
“Because of the degradation of Terra water vapor chanles that affects cloud retrievals
starting around 2008, downward longwave flux anomalies over polar regions shows
downward trend (see Section 4.5). Therfore, trend analyses with surface fluxes over
polare regions from Ed4 EBAF-surface should be avoided.”

How does this jive with temp profiles for the Arctic which show virtually all warming during the winter season, when there is no insolation to be absorbed by any soot. Might melt some snow/ice during high sun, but obviously has little effect on summer temps.

Glad to see you using that data.
when you use data you implicitly accept the physics used to create the data
so, ya that means accepting radiative transfer theory.
What’s the purpose of this adjusted dataset?
“EBAF-Surface is for climate model evaluation, estimating the Earth’s global mean energy budget and to infer meridional heat transport.”
Looking at the processing steps is always important
and its always good toi note the uncertainties in the fundamental data that the modelled surface data derives from
Usually I have found when using most satellite products for surface data that the devil is in the details.
and in the details its models and assumptions and adjustments all the way down.

Richard M

If we look closely at the data you will it is almost a perfect match to the UAH satellite data product. Different sources giving similar results.

John Robertson

Interesting Map.
Stability of this water world is indeed amazing, funny how well water in its various states does such a beautiful job of keeping this planet..green.
Are these trends,warming or cooling actually significant?
What are the error bars, of CERES?
I did follow the link to their web site,however this information eluded my weak search.
So do we know it is warming for sure,in these places no one lives or no one cares?
How do we establish a trend line where data is so slim?
I live at 62 degrees North.
The Canadian Government has relocated nearly all Arctic temperature sensors to Airports since the 1990s, the manned remote stations are all closed.
These airstrips are all now paved, most were still gravel not too long ago.
Yellowknife Airport just relocated the weather station to the wind shadow of the terminal building and surrounded it with black dirt.
I am sure this station will show unusual warming, especially last summer when the dirt had no vegetation.
But being now sheltered from the prevailing wind, I expect to see another “unprecedented” Arctic temperature increase recorded by this station.
How far south do we have to look before ;”Don’t know,couldn’t say for sure” does not apply?

CERES clear sky TOA data shows no reduction in radiation to space, either LW or net.comment image
CO2 remains when clouds are gone. Not looking good for CO2.

As always RTFM
“Because of the degradation of Terra water vapor [sic] chanles that affects cloud retrievals
starting around 2008, downward longwave flux anomalies over polar regions shows
downward trend (see Section 4.5). Therfore, trend analyses with surface fluxes over
polare regions from Ed4 EBAF-surface should be avoided.”


In the experience of this truck driver, at least 30% of drivers don’t know how to even engage cruise control let alone use it properly.

bill hunter

“To me, this is clear evidence of strong thermoregulatory systems”. Yep its name is SUN. Its not clear to me if you are seeing anything else. The concept of blackbody equilibrium suggests a natural oscillation around that point due to both diurnal and seasonal variations. Pretty much the same likely for ocean oscillations like ENSO and multi-decadal ones if indeed they are oscillations that over time establish their own natural time delays, for instance living near the Pacific Ocean the coldest month is February about 2 months beyond the least sun. About the same thing at the other end in June max sun, August max temperature. As regular as those changes are why would we expect a bump in the anomaly that we don’t already see?

Stephen Richards

Nor is this due to “thermal inertia” as many people claim.
The one thing that has troubled me for decade. It is obvious to any observer that there is very little inertia. Yes, the ENSO leaves a ‘taste in the mouth’ for more than a year and an SSW appears to create an atmospheric image for a few months. so What causes ice ages? What is it that changes the global temperature by 3 to 5 °C and holds it while glaciers SLOWLY build.
Thanks Willis. Great post again.

bill hunter

Its pretty clear to me that you could probably tune an ice and ocean inertia model to explain the approximate 2 to 3 degrees warming since the end of the Maunder Minimum in approximately 1700 using global glacial ice change and an ocean uptake (inertia) model. Some of the longterm instrument records show a rapid warming occurring in a few decades after the Maunder Minimum, this could be a rapid warming of the surface ocean that reduced low latitude and low elevation ice. The warming then flattened until interrupted temporarily for a short period by the Dalton Minimum. After brief recovery post 1815 temp increases from the LIA (500 years of cooling) then another flattening to the mid 19th century during which period global glaciers continued to increase and increase surface albedo. Then in the mid-19th century glacial advance reversed into retreat allowing climate warming to dominate from the underlying ocean inertia. The 500 years of cooling from 1200 to 1700 could take as long to reverse back out with the same intensity things cooled and drove the Vikings out of Greenland from glacial advance as far back as 1300 to 1400 from the more distant settlements. That would put on a model the warmed as fast as it cooled warming from ocean inertia until 2200. Its hard to put values on it lacking early temperature records from the Medieval Warm Period but today they are assumed to be comparable by some studies to day plus or minus a degree or so. The only criticism of the idea of an extended LIA recovery was the flat period from the mid part of the first half of the 18th century until the mid 19th century. So an explanation is needed for that and glacial advance and retreat data overlies it very nicely chronologically. All thats left to do is argue the actual degree effects which can probably be tuned in quite nicely. Of course even today we are not measuring what it happening in the lower 3/4ths of the ocean so as one study put it we are probably today a 1,000 years or more away from a data record that would demonstrate that.

The ocean is the tortoise and the atmosphere the hare.

Non Nomen

A bit off-topic, probably: In 1950, world population was about 2.5 bn. In 2018 the population has risen to 7.5 bn. The average body temperature is 36.5 to 37.2° C. Today, there are 5 bn people more on this little planet that heat it up. “AGW” is said to be on the rise. Is there a correlation or even a causation?

Stephen Richards

Population numbers are socialism scare stories. Have you noticed by how many billions it has risen in the last 5 years?
It’s a finger in the air number.

Non Nomen

I completely agree with your scare story. But IF there were numbers that might show that a good part of the alleged AGW is due to humans heating up the atmosphere, then Co2 alone can”t be blamed any longer.

7.5 billion people would generate about 750 GW body heat. Surface area of Earth is about 5e14 m2. So that is about 1.5 milliwatts/sq m. Not much compared with radiative forcing of 2-3 watts/m2.

Non Nomen



Which is undetectable from 50W/m2 from the surface, which changes with moisture content. Forget CO2.

And that 1.5 mW/m2 is retarded energy from the Sun. Also not much compared with the average global heat flux from the Earth’s interior of 91 mW/m2.

oebele bruinsma

Dear Willis, “Despite being controlled by things as evanescent as winds, waves, and clouds, the temperature anomaly doesn’t vary more than about two-tenths of a degree.” Given the stable climate conditions, it may be better to state: Because being controlled by things…… As nature is regulated by chaotic processes from molecular level up, averaging each other in impact, the end result. the macro world as we experience it, must nearly by definition be rather stable; variations with 0.2 – 0.3 degrees.

As I’ve been saying, Tmin is regulated on the cooling side at night to limit how cold it gets by water vapor, just as W found a relationship with max ssts, and day time T max.


As usual, the strongest variations happen in places where the data is lacking the most. Heavy interpolation going on. Still, the cooling shown for the southern oceans contradicts the rapid melting of ice caps shown by satellite pictures for the kerguelen islands for example. But the rapid melting can also be explained by weather factors.


You are 60 years behind the research team visiting Svalbard:
The noted a substantial change in temperatures from 1900 to 1930. 6 degrees Celcius at Svalbard.
Very similar as today.
They even tried to make a global map where to find this change.
The current knowledge has to upgrade the knowledge of currents.

Crispin in Waterloo but really in Ulaanbaatar

Richard M
Don’t forget to add that increasing the CO2 concentration increases the emissivity of the atmosphere.
It is interesting that back radiation gets frequent mention but not the ‘greyness’ of the body of the atmosphere. Without GHG’s, the atmosphere couldn’t cool by radiation at all. It would be heated by the surface and stay that way.
A desert that has higher temperatures could just have lower humidity and the same energy input. The enthalpy (total energy) in the atmosphere above the desert is the correct number to watch. I realise it is fashionable to talk of temperatures only, but that is a technically not how to think about energy gain and retention (etc). Equally, the Amazon could be cooler simply by having a higher average humidity, not a lower energy input or higher emittance.

Vic Pearson

As you all know a little knowledge is dangerous. From the discussion between what it seems to knowledgeable scientists the opposite may also be true leading to a battle of egos rather than an objective debate ?

Area Man

“the Earth abides…”.
Love it.

Polar see-saw
From Wikipedia, the free encyclopedia
The polar see-saw (also: Bipolar seesaw) is the phenomenon that temperature changes in the northern and southern hemispheres may be out of phase. The hypothesis states that large changes, for example when the glaciers are intensely growing or depleting, in the formation of ocean bottom water in both poles take a long time to exert their effect in the other hemisphere. Estimates of the period of delay vary, one typical estimate is 1500 years. This is usually studied in the context of ice-cores taken from Antarctica and Greenland.

Barry Hoffman

Did anyone notice the Himalayas, average height 20,000+/- feet, are getting colder? Latitude 28*N.


And glaciers in that region are growing, both from more precipitation and cooler temperatures. So much for the IPCC’s ill-founded prediction of glacier disappearance.
On balance, the whole world has been cooling for about a decade, a fact obscured by the big ENSO swing of 2016, which blew off a lot of oceanic heat, thus contributing to further future cooling.


Some say two decades, since the late ’90s.