Global Temperature Report: November 2017

From UAH and Dr. John Christy

Global temperatures drop; November still warm

Global climate trend since Nov. 16, 1978: +0.13 C per decade

November temperatures (preliminary)

Global composite temp.: +0.36 C (about 0.65 degrees Fahrenheit) above 30-year average for November.

Northern Hemisphere: +0.33 C (about 0.59 degrees Fahrenheit) above 30-year average for November.

Southern Hemisphere: +0.38 C (about 0.68 degrees Fahrenheit) above 30-year average for November.

Tropics: +0.26 C (about 0.47 degrees Fahrenheit) above 30-year average for November.

October temperatures (revised):

Global Composite: +0.63 C above 30-year average

Northern Hemisphere: +0.67 C above 30-year average

Southern Hemisphere: +0.59 C above 30-year average

Tropics: +0.47 C above 30-year average

(All temperature anomalies are based on a 30-year average (1981-2010) for the month reported.)

Notes on data released Dec. 4, 2017:

The average global temperature drop between October and November, 2017, tied for the fifth largest one-month-to-the-next drop in the 39-year satellite temperature record, according to Dr. John Christy, director of the Earth System Science Center (ESSC) at The University of Alabama in Huntsville. Compared to seasonal norms, the average temperature around the globe fell 0.27 C (almost 0.49 degrees F) between October and November. (The largest drop was from January to February 2013, when the global average temperature fell 0.32 C.)

Despite that temperature drop, however, November 2017 was still the second warmest November in the 39-year satellite temperature record for both the globe and the southern hemisphere. In both cases, the warmest November on record was in 2016.

Compared to seasonal norms, the coldest spot on the globe in November was in Hudson Bay, near Fort Severn, Ontario. Temperatures there were 2.94 C (about 5.29 degrees Fahrenheit) cooler than seasonal norms.

Compared to seasonal norms, the warmest place on Earth in November was over the Bering Sea near the island of St. George, Alaska. Temperatures there averaged 6.47 C (about 11.65 degrees Fahrenheit) warmer than seasonal norms.

Christy and Dr. Richard McNider, a professor emeritus at UAH, recently published in the Asia-Pacific Journal of Atmospheric Sciences a study that mathematically removed from the satellite temperature record the effects of volcanic eruptions and of El Nino and La Nina Pacific Ocean heating and cooling events. This was done in an attempt to identify that part of the overall warming during the 39-year period that might be attributed to human influences. The 0.155 C per decade trend reported in that study differs from the 0.13 C per decade trend reported here in the Global Temperature Report. That is because this most recent research in the APJAS was done using an earlier version of the satellite microwave sounding unit dataset. That dataset was revised and updated, and the revisions published (Spencer et al., APJAS 2017) while the research looking at the effects of natural climatic events was under peer review.

As part of an ongoing joint project between UAH, NOAA and NASA, Christy and Dr. Roy Spencer, an ESSC principal scientist, use data gathered by advanced microwave sounding units on NOAA and NASA satellites to get accurate temperature readings for almost all regions of the Earth. This includes remote desert, ocean and rain forest areas where reliable climate data are not otherwise available.

The satellite-based instruments measure the temperature of the atmosphere from the surface up to an altitude of about eight kilometers above sea level. Once the monthly temperature data are collected and processed, they are placed in a “public” computer file for immediate access by atmospheric scientists in the U.S. and abroad.

The complete version 6 lower troposphere dataset is available here:

Archived color maps of local temperature anomalies are available on-line at:

Neither Christy nor Spencer receives any research support or funding from oil, coal or industrial companies or organizations, or from any private or special interest groups. All of their climate research funding comes from federal and state grants or contracts.

— 30 —


newest oldest most voted
Notify of

This is all within the range of natural variability. A half a degree warmer, a half a degree cooler…what were we expecting, that everything would stay exactly the same? The one thing we know for 100% certainly is that over time, the weather/climate is and always has been changing, within fairly well understood thresholds.

Note that this Christy/McNider 2017 paper ASSUMES “that the net impact of this decadal scale natural variability has been near zero since 1979 and that the underlying trend due to the net impact of human influences is +0.096 ± 0.012 K dec−1”, and STILL concludes that “the tropospheric transient climate response (TTCR) is 1.10 ± 0.26 K.”/(2xCO2).

This ~1.1K is a conservative (upper bound) estimate of TTCR, which suggests that there is NO REAL GLOBAL WARMING CRISIS, and that any warming due to increasing atmospheric CO2 will be minor – and either benign or beneficial to humanity and the environment.

This is similar to the conclusion that we reached in our article written in 2002:
PEGG, reprinted in edited form at their request by several other professional journals , The Globe and Mail and La Presse in translation, by Baliunas, Patterson and MacRae.

Reference – 2017 paper:
Asia-Pac. J. Atmos. Sci., 53(4), 511-518, 2017 pISSN 1976-7633 / eISSN 1976-7951
“Satellite Bulk Tropospheric Temperatures as a Metric for Climate Sensitivity”
John R. Christy and Richard T. McNider
Earth System Science Center, The University of Alabama in Huntsville, Alabama, USA
(Manuscript received 9 June 2017; accepted 14 September 2017)
© The Korean Meteorological Society and Springer 2017


Abstract: We identify and remove the main natural perturbations
(e.g. volcanic activity, ENSOs) from the global mean lower
tropospheric temperatures (TLT) over January 1979 – June 2017 to
estimate the underlying, potentially human-forced trend. The unaltered
value is +0.155 K dec−1 while the adjusted trend is +0.096 K dec−1,
related primarily to the removal of volcanic cooling in the early part
of the record. This is essentially the same value we determined in
1994 (+0.09 K dec−1, Christy and McNider, 1994) using only 15
years of data. If the warming rate of +0.096 K dec−1 represents the
net TLT response to increasing greenhouse radiative forcings, this
implies that the TLT tropospheric transient climate response (ΔTLT at
the time CO2 doubles) is +1.10 ± 0.26 K which is about half of the
average of the IPCC AR5 climate models of 2.31 ± 0.20 K. Assuming
that the net remaining unknown internal and external natural forcing
over this period is near zero, the mismatch since 1979 between
observations and CMIP-5 model values suggests that excessive
sensitivity to enhanced radiative forcing in the models can be
appreciable. The tropical region is mainly responsible for this
discrepancy suggesting processes that are the likely sources of the
extra sensitivity are (a) the parameterized hydrology of the deep
atmosphere, (b) the parameterized heat-partitioning at the oceanatmosphere
interface and/or (c) unknown natural variations.
However, here we shall assume that the net
impact of this decadal scale natural variability has been near
zero since 1979 and that the underlying trend due to the net
impact of human influences is +0.096 ± 0.012 K dec−1, where
the error range represents the spread of the various simulations.


We are not defending this rather bold assumption regarding
natural variability, but simply stating it as a basis for going
forward to derive climate sensitivity estimates, acknowledging
the strong dependence on this assumption to what follows.




The assessment of tropospheric climate sensitivity from the
calculation of the underlying trend above requires significant
assumptions. If we assume, among other things, that the
impact of the net of natural external and internal forcing
variations has not influenced the observed trend and that
anthropogenic forcing as depicted in the average of the IPCC
AR5 models is similar to that experienced by the Earth, then
observations suggest the tropospheric transient climate response
(TTCR) is 1.10 ± 0.26 K. This central estimate is likely less
than half that of the average of the 102 simulations of the
CMIP-5 RCP4.5 model runs also examined here (2.31 ± 0.20).


Richard M

Keep in mind the .096 C / decade is UAH 5.6. If you apply a similar correction to UAH 6.0 you end up with .078 C / decade.

Extreme Hiatus

This is the real BEST data and I appreciate how quickly and broadly it is made available to the public.

Next year will be 40 years of data; only 20 to go to see most of a cycle maybe.


A full AMO/PDO cycle with accurate temperature recording for most of the Earth’s surface. That will be nice indeed. We seem to be at the 1950 stage where the PDO has gone negative but the AMO has reached a top but stays there longer. Once both drop it gets cold quick. Those of us old enough to remember the 60’s & 70’s know cold.

Tom Halla

It does not look as if there is a La Nina yet, and ENSO seems to account for most of the variations in temperature.


I posted this under ‘Weekly News Roundup’, where it was roundly ignored. Perhaps it may elicit a response here. I’m keen to know why even warmists seem to accept ENSO as something separated from the main debate –
I’ll go along with Christy and McNider’s report and it is certainly a useful contribution.
What isn’t at all clear to me is why ENSO events are seen as something entirely separate to global warming. Given the vast heat capacity of the oceans, it seems reasonable to me that any warming trend – or a cooling one for that matter- might present as periods of stasis punctuated by ENSO-mediated jumps. Not saying it must be so, just that it is a fair working hypothesis. We may not know the details of how it works, but what are the reasons why it can’t be so?………………..”

DC Cowboy

The reason is that there is no link between ENSO events and rising concentrations of CO2 in the atmosphere and ‘Global Warming’ is generally linked to increased temps caused by the increasing concentration of CO2 in the atmosphere. As far as we know, ENSO is a ‘natural’ event and is not an effect of CO2.


sorry, DC –
didn’t understand a word of that.
Anyone else? Willis? Nick?

Richard M

Think of ENSO events like hills and valleys along a road. They are temporary ups and downs and don’t change the ongoing elevation of the road itself.

Now, if you start to compute a trend while on the top of a big hill or the depths of a valley it would certainly affect it.


Richard – okay, except if, for example, the El Nino leaves the base a step higher, as in 1997-1998. Is everbody missing the point of the question I’m asking? Sea temperatures rule the roost, and they may have influence on Mean GST on long to very long timescales, therefore controlling the way in which warming appears. Warming – whether GHG, Solar, or other in origin – gets stored in a very large battery (the ocean), which periodically discharges to restore an equilibrium. El Nino is a symptom of this process. Not possible?

I loved the “-30-” at the end, reminded me of when I used to work for the newspaper.

You remember “newspapers”, right? …



Please explain willis ! I could not find an explanation on Google. I assume it’s some kind of marker that signifies the end of a story – but what does it have to do with newspapers ?

Steve Fraser

News wires.


It’s decimal for XXX, which used to be put at the end of a story.

When I worked for the paper, it was how the AP signified the end of one story and the start of another on the teletype … from there, it went to the Linotype machine and was cast in hot lead, with a distinctive smell … for the most part, gone now.



My father was a linotype operator for years before they switched machinery; he was still employed by the paper when he died. They used “-30-” to honor his passing.

RSS4 dropped a similar amount: from 0.805 to 0.549, or a drop of 0.256. But it was still the third warmest November for RSS4.

I found a drop of 0.12°C in surface temperature, using NCEP/NCAR reanalysis. But again, October was warm, so November wasn’t very cold. Second coolest for 2017.

Steve Fraser

Dig that…the Northern Pacific was warm.


10 degrees colder here than last year for the same time period. Just sayin’ – the gas company here keeps track of such things as heating days and temps.


A come on. The gas company? What do they know? Propane truck drivers? Do they care if their customers run out at 1AM? Accuracy is just not important to the underclass, you know. Only to the really important scientist type.



Roger Knights

“the gas company here keeps track of such things as heating days and temps.”

Has anyone written a paper using gas company statistics to check the reliability of nearby official weather stations? Thy might be a valuable corrective.


They probably use the local airport’s weather data instead of capturing their own.

Roger Knights

“They probably use the local airport’s weather data instead of capturing their own.”

Probably that is the case now, but before the war (I’ve read) newspapers commonly had weather stations on their roofs and I suspect that local gas companies used those in some cases. I seem to recall that some gas companies or newspapers computed “degree-days,” which may have involved obtaining hourly temperature numbers.

Anyway, even a spotty coverage of the country might yield important results, if it undermined the reliability of official stations. Some ambitious young scientist (or skeptic!) should look into it.


I doubt it, but it might be interesting to use it to keep NASA honest. We should all be checking our heating/cooling bills.

Here, we had a very cold first half of November, then a very warm second half. Not too unusual as Chinooks tend to do that to us.

That global temperature graph seems to reflect the solar cycle activity over that period.


(Yes it does!)…


Where is it hot?
Where heat is no problem!

tony mcleod

Except for October…did you bother to look any further?


Look? I was in one of the bubbles in northern Spain+3,5 Degrees. No problem.
India and Pakistan suffers but their hot season is in the summer before monsoon rain.

Stevan Reddish

(All temperature anomalies are based on a 30-year average (1981-2010) for the month reported.)

Just some questions about why certain comparisons are made? It is clear to me that when calculating a particular month’s anomaly, it is appropriate to use data compiled for the same month in previous years.

When it comes to making comparisons, it makes sense to compare this November with October so that we can see what changes are happening now. It also makes sense to compare this November with last November so we can see any change that has happened over the last year. If we want to see rolling decadal changes, it makes sense to compare this month with Novembers of 10 years ago, 20 years ago, etc. Finally, if we want to know how the anomaly has changed since satellite measurements began, it makes since to compare with the very 1st reliable month in the record.

But why make comparisons like:”warmest November on record was in 2016″. We are talking anomalies, yes? Why would it matter if last November happened to be the warmest November if there were other months with greater anomalies? Don’t we want to know how the current month compares with the months with the greatest anomalies, with the months with the lowest anomalies, and with the average anomalies of different periods, such as the current rolling decade?


Stevan Reddish

OOPS! 1st sentence was supposed to be in quotes.



Why, too, do the anomalies seem to remain relatively level, even though the base period of 1981-2010 is supposedly warmer than the previous base periods. How can anomalies based on averages not decrease as the average increases? 1981 to 2010 is warmer than 1961 to 1990, but the anomalies always seem to be about the same. As the world warms and we move the base period forward, the anomalies should decrease.

w p

Such absolute global temperature ‘anomaly’ statements, nota bene to hundreths of a degree Celsius, amaze me.
Unless I missed it, is there any precision and accuracy statement, repeat- and reproducibility information? A footnote, link?
As a layperson, living comfortably, with temperature variances of 50 full degrees C over the year, I just cannot fathom the significance of these minuscule fractions of a degree C, ‘valid’ for the entire globe?
Still, for argument’s sake, taking these temp ‘anomalies’ for granted, looking at their values and distribution on the globe, then why is it, that apparently the ‘coldest’ minus variances are along the most heavily industrialized longitude? Per generally mediatized wisdom, that s where most human made CO2 is produced, hence AW should be the highest, not minus ‘anomalies’?
But then, what do I know. Simply confused.

tony mcleod

2 or 3 degrees is the difference between glaciation and inter-glacial. Either a sunny day or a kilometer of ice in northern parts.
The temperature anomaly areas shift around from month to month. This is October’s:
comment image?w=1200

The gasses are rapidly mixed, you are not looking at localised co2 effects. Nothing mediatized about it.


Two or three degrees? Evidence?


Shifts around, like WEATHER you mean. its anomalous alright, anomalous rubbish.

The “Well Mixed” theory turned out to be wrong. There was an article here showing a CO2 concentration map. Once they put up a satellite to measure CO2 globally, they found out it wasn’t as well mixed as they thought.

Bob boder

tony mcleod

Why would anyone listen to you, you are a proven dishonest person.

I ran monthly anomalies for several GHCN stations according to Nick’s description of the process. I used TMAX because TAVG is made just by averaging TMAX and TMIN for a day, and I wanted to introduce as little error as possible. I averaged each month’s TMAXs over the period 1981-2010, and then used that baseline to get the monthly anomaly for each month. It was normal to get ranges of +6 to -7 deg. C, and standard deviations of over a degree.

That this kind of data can be published with precision to 0.01 deg C. with no error bars is a travesty of science. I’d have flunked Physics 1 if I reported numbers like that for a calculation.


James, none of these modelers show error bars because they know the bars would completely overwhelm the so-called trends. Christy might be doing it ‘better’ than most, but his stuff is still a model. None of this stuff is better than a ‘random walk’ of numbers.
As you say , the ‘accuracy’ quoted is a joke. Also in the ‘real’ world, we are interested in temperatures, not ‘anomalies’. Average temperature increases of one tenth of a degree is meaningless. Has the maximum gone up? no. There has been a very slight increase in nighttime temperatures and winters have become very slightly less cold. There has been an outbreak of alarming ‘mildness’ ( albeit only very slightly).
The whole thing is absolute rubbish.

michael hart

+3 jim.
The whole thing is rubbish, as judged by common standards in most of science and technology.


w p , you do not need to know any of that. all you need to know is we are all gonna die due to the warming trend/sarc


The 2017/18 La Niña cycle will continue to cool global temps for the next 12 months.

By 2019, the 2015/16 Super El Niño warming spike should be completely negated, and perhaps we’ll again see a 0.00C/decade trend from mid-1996, making 23 years years without a global warming trend.

CAGW advocates will have a very hard time justifying a 23-year flat global temp trend and an even harder time explaining why the disparity between CAGW global temp projections vs. reaity exceed 3 standard deviation for over two decades…

The silky CAGW hypothesis is quickly becoming laughable.

I agree Samurai – please see my post below, originally posted on Sunday Dec3


The 2014/16 El Niño will always affect the trend. It will make the trend higher for as long as it is in the right half of the period considered, and lower if it is in the left half of the period considered.

If we go back to the temperature of 2003 no net warming would have taken place in 15 years, yet the warming trend for that period would still be positive.

Over a long enough period, it is generally considered that Niños and Niñas compensate their effect on the trend. I do not think that is necessarily true. Paleo studies seem to support that there were no Niños during the Holocene Climate Optimum, when the planet was warmer.

Richard M

First time I’ve heard about this. Very interesting. This could support the idea that El Nino does cause step ups in temperatures but only up to a point which was reached during the Holocene Optimum.


By 2019, the 2015/16 Super El Niño warming spike should be completely negated, and perhaps we’ll again see a 0.00C/decade trend from mid-1996, making 23 years years without a global warming trend.

I wouldn’t get your hopes up.

By my reckoning, even if UAH has the next two years average an anomaly of 0, (ie similar to the two years after the 1998 El Nino), you still won’t have a zero trend from 1996 before 2020.

Richard M

Bellman, that is correct and expected as the +AMO should cause a warming trend over that period.


This strong decline in UAH LT temperatures was predicted by me several times, most recently circa Nov.4, 2017 on wattsup.

I expect global cooling to continue to decline for several months more, possibly with a few bumps along the way, to reach ~0.0C for the UAH LT global anomaly

The basis for this prediction is that the Nino3.4 temperature anomaly provides a good 3-month predictor of the UAH LT Tropical temperature anomaly, and a good 4-month predictor of the UAH LT Global temperature anomaly.

Similarly, the East Equatorial Upper Ocean temperature anomaly provides a good 5-month predictor of the UAH LT Tropical temperature anomaly, and a good 6-month predictor of the UAH LT Global temperature anomaly. [H/T to Bill Illis.]

All this is from memory, so please don’t get excited if I messed up a minor point. No time to re-check the details now – maybe later.

And for all you warmist trolls, “I TOLD YOU SO!”

And NONE of the very-scary predictions by the IPCC and its minions have ever proven correct. They have a perfectly NEGATIVE predictive track record, and thus NEGATIVE CREDIBILITY.

Time to play the BIG SHORT on global warming alarmism…

Regards, Allan 🙂

This is the plot of the UAH LT TROPICAL Anomaly vs the East Equatorial Upper Ocean Temperature Anomaly, lagged 6-months to show the ~5-month lag of UAH Tropical LT after the East Equatorial Upper Ocean Temperature Anomaly. UAH Global LT temperature follows UAH Tropical LT ~1 month later.
There is a typical pattern after major El Nino’s, in which atmospheric (LT) temperature diverges above the level predicted by the long term relationship between LT temperature and the East Equatorial Upper Ocean Temperature Anomaly. I suggest that the relationship will converge again soon, as atmospheric cooling resumes.

The temporary divergence is caused by the time it takes this excess heat to dissipate from the LT into the upper atmosphere and then into space.

Based on this generally-robust long-term relationship, I expect that within ~6 months, the UAH LT Global anomaly will decline to about 0.0C. I have no time to work on this further now, my apologies,

See also my related correspondence with John Christy, posted on this page.

From: John Christy
Date: November 4, 2017 at 7:28:22 PM GMT+7
To: Allan MacRae
Cc: Anthony Watts, Roy Spencer, John Christy, Joe D’Aleo, Joe Bastardi

Subject: Re: Sorted – atmospheric cooling will resume soon

Yes. We’ve seen this correlation since our first paper about it in Nature back in 1994. The Pacific gave up a lot of heat between July and October – and some of it is making its way through the atmosphere. We think the anomalies will drop soon too.

John C.
Sent from my iPhone
On Nov 4, 2017, at 6:03 AM, Allan MacRae wrote:

FYI friends.

I was curious why the UAH LT was diverging above its predicted value based on the East Equatorial Upper Ocean Temperature Anomaly – I re-plotted the data vs UAH LT Tropics (instead of UAH LT Global) and the situation became more clear- atmospheric cooling will resume soon, imo.

Best, Allan

Sorted – atmospheric cooling will resume soon. See the plot below of the UAH LT TROPICAL Anomaly vs the East Equatorial Upper Ocean Temperature Anomaly and the situation becomes more clear.

This is a typical pattern after major El Nino’s, in which atmospheric (LT) temperature diverges above the level predicted by the long term relationship with the East Equatorial Upper Ocean Temperature Anomaly. The pattern will converge again soon, and atmospheric cooling will resume. WHY this happens after major El Nino’s is still to be explained. Perhaps the El Nino heat in the atmosphere just needs time to dissipate.

– Allan M R MacRae

Roger Knights

“I expect global cooling anomaly to continue to decline for several months more, possibly with a few bumps along the way, to reach ~0.0C for the UAH LT global anomaly”

Thank you Roger – for better wording.
““I expect global cooling to continue for several more months, possibly with a few bumps along the way, to reach ~0.0C for the UAH LT global anomaly”

This is probably just short-term temperature variation – the typical cooling that follows the warming of a major El Nino,

In the bigger picture and the longer term:
I published in 2002 that natural global cooling (similar or more severe than the cooling from ~1940-1975) would recommence by 2020-2030. I am now leaning towards the start of such cooling by 2020 or sooner.

In 2002, Solar Cycle 24 was generally predicted to be strong, and now SC24 has proven to be the weakest in a century. We also expect SC25 to be weak. so our 2002 global cooling prediction is looking more probable.

John M. Ware

I’m still uneasy about the use of the term “norms” in the article. Weather and climate don’t have norms; they have means and averages. A norm is something that should occur or be present, and it is arrived at by observation and/or experimentation. Normal human body temperature is 98.6; significant deviation from that can kill. Normal eyesight is 20/20; too great a deviation means being unable to drive. George Bernard Shaw was persuaded to visit an oculist (optometrist, we call them now); that professional, having examined Shaw thoroughly, said, “Your eyesight is normal; that condition is very rare.”

Normal weather is a false term. Is a rainy day normal? Not here–we haven’t seen one in close to a month. I wouldn’t call rain abnormal, though; I would rather say that some rain would be both typical and welcome. At my central Virginia house, November averaged a bit more than 5 degrees F below the long-term mean for Richmond (as far as I know, there is no long-term mean recorded for Mechanicsville, 12 miles from the Richmond airport). I have always been suspicious of the anomalies reported for the weather; the reference period is far too short, and itself already has an anomaly from the true long-term figures.

John M. Ware: George Bernard Shaw was persuaded to visit an oculist (optometrist, we call them now); that professional, having examined Shaw thoroughly, said, “Your eyesight is normal; that condition is very rare.”

That reminds me of the old joke about the man visiting his doctor and being told he had Tom Jones Syndrome. The man asked, “I’ve never heard of that. Is it common?”

His doctor replied, “It’s not unusual.”

“The satellite-based instruments measure the temperature of the atmosphere from the surface up to an altitude of about eight kilometers above sea level. ”

err no.
they measure brightness at the sensor. Actually they record a digital count. Then using physics models and a variety of assumptions they estimate the temperature.
then they adjust the data using stastics.They then infill and extrapolate and finally produce a gridded set of anomalies.

You cant review the code that does this.


For once I totally agree with you. However you could apply the same inferences to your work and those of all the other modelers. ‘Bull[snip] baffles brains every time’. You all peddle in pseudo-science, numerology wrapped up as ‘science’.
In the real world experienced by people, temperatures have hardly moved a jot since records begun. The only slight changes experienced by real people are very slightly warmer nights and very slightly less cold winters, balanced by very slightly less warm summer maximums. A terrifying outbreak of ‘mildness’ ( albeit very slightly).
This whole , very expensive, exercise is a load of rubbish.


This is like the ground temps.. they measure the real number, add in all the close ones, average it out then multiply it by 1.1 to get the new real number.

Juan Slayton

You cant review the code that does this.

Puzzled by this. If All of their climate research funding comes from federal and state grants or contracts, is it not subject to FOIA?

Taylor Ponlman

And temperature stations measure at a nearly infinitesimal radius ( relative to the surface area of the earth) using an extremely small number of stations essentially arbitrarily located, and then use computer code to extrapolate from those points, when it is demonstrated that the temps can vary more than a degree within a km radius of any given station. The entire process of anomalies depends on those variances being uniform day to day, month to month, to make the math models work. So what’s your point? That the process I just described is better/ more accurate than what UAH does?

Taylor Ponlman

Mosh, you forgot to mention that they use radiosonde to also adjust and compare, in addition to ‘stastics’ by which I assume you meant statistical analysis. In other words, the models for UAH are informed by measurements. By contrast, the surface temp measurements are informed by models – it’s the wrong way round, IMHO.


That’s a wake up call map with the EU and Australia looking blue.

Coach Springer

So. Holding steady above 1978 when we were afraid of an ice age. Cool.


These are not temperatures. They are estimates of CHANGES in temperatures and subject to the intrinsic accuracy of the instruments used and statistical/model methods used as well as the base period selected for the comparison. Seems this is only pointed out by some when “cooling” is involved. There is no global average temperature that is worthy of note on the time scale being considered. Geologic time scales make more sense but are themselves subject to the proxies used to estimate them. “Much ado about nothing.”

Coeur de Lion

Any bets on next month?

michael hart

Lucia used to run bets at The Blackboard, but I haven’t visited much recently.

comment image

… to say nothing of the plus-or-minus instrumental error, “adjustments”, etc. — all considered within the vast, vast zone of 1.3 degrees.

… much cause for alarm, I’d say.

A C Osborn

Robert, also remember it is not the actual surface measurement, but thr measurement of the heat leaving the surface.
If it leaves the surface quicker than the Sun replaces it is getting or will get cooler.

Great Minds… I quickly looked at the graph as well and noticed -0.4 to +0.4 in 40 years and thought, Hmmm, why that’s just 2 deg/cent — Thats OK.

that’s just embarrassing. you do realize that, don’t you? yeah, no

If 0.36°C is the 5th warmest November, I don’t think there is too much to worry about.


True, it should only ever mention from the 3rd to 1st warmest ever for any concern or never mention 5th warmest.

Unless you’re only concerned with burning energy watching your electronic bank account online grow.


I know that I have no qualifications amongst the wonderful scientist that report on WUWT. Bu I suggested some time age that all of us talked about “ever changing climate”. So simple, so true. It disarms any alarmist.


Typo. “Ago” not “age” sorry.


At least Paris is in the blue again.

“(The largest drop was from January to February 2013, when the global average temperature fell 0.32 C.)”

The global average anomaly fell by 0.32C. The global average temperature fell by even more. It’s important to recognize this distinction.

Slight correction. The global average is at the sinusoidal minimum between Jan and Feb, so the anomaly is approximately equal to the change in temperature. Between March and June, the global temperature increases by about 0.5C per month while between September and November, the global average temperature decreases by about 0.5C per month. This average month to month change is what gets cancelled out when anomalies are calculated.


If one looks at the Danish Met chart of the Daily Mean Temp for North of 80, it spent the “melt season” under the mean. For the last 40 days (implying 40 nights as well) it has been in 5-degree “trading range” well above the mean, which line is in steep decline.
Is there still a little El Nino hanging around?



Anecdotal, I know, but the sea temperature here is perfect for swimming, at least month earlier than normal. Looking good for a belter summer. Nelson, New Zealand.


The renewed drop in NINO3.4 presages well for further big drops in global temps for the next 4 months. Will be a change to get a cool start to the new year and as someone said 7 years ago a cool first 3 months sets the tone for a very cool year. Come on 2018 [cool].

With thanks! This a awesome internet site.