On the 1998 Apparent Step-Up in UAH Land-minus-Ocean Lower Tropospheric Temperatures

Reposted from Dr Roy Spencer’s Blog

November 7th, 2019 by Roy W. Spencer, Ph. D.

A follower of our UAH global lower tropospheric temperature (LT) dataset named “JJ” emailed me asking about what might be considered a spurious feature in the dataset.

The feature is most easily seen if you plot the monthly global time series of Land-minus-Ocean (hereafter “L-O”) temperature anomalies. The result seems to show a step-up of about 0.16 deg. C in May of 1998.

Fig. 1. Difference between the UAH lower tropospheric (LT) land and ocean temperature anomalies between January 1979 and August 2019, showing an apparent step-up in the difference occurring in May 1998. The dashed lines show the average values before and after that date, while the curve is a 5th order polynomial fit to the data.

The year 1998 is key for our dataset because that is when the first (NOAA-15) Advanced Microwave Sounding Unit (AMSU) came online, which initiated the transition from the older Microwave Sounding Units (MSU, the last of which was on the NOAA-14 satellite).

AMSU did not have exactly the same channel frequency selection as the MSU, so the nominal layers of the atmosphere sensed were slightly different. Most importantly, the AMSU channel 5 has a weighting function that senses somewhat more of the surface and lower troposphere than MSU channel 2. If one did not account for this fact, the AMSU’s greater surface sensitivity would produce higher temperatures over land and lower temperatures over the ocean (after a global-average intercalibration between MSU and AMSU was performed). [The reason why is that these channel frequencies are not sensitive to changes in sea surface temperature, because the microwave emissivity decreases as SST increases. The effect is small, but measurable.]

But since these are through-nadir scanners, each view angle relative to the local vertical measures a slightly different layer anyway, which allows us to match the AMSU and MSU measurements. When we developed Version 6 of the dataset, we found that the 50-60 GHz oxygen absorption theory used to find the view angle from AMSU5 that best matches MSU2, the resulting temperature anomalies over land were still too warm relative to the oceans. This meant that we had to perform an empirical (data-dependent) rather than theoretical matching of the AMSU and MSU view angles.

The way we gauged the match between MSU and AMSU is how the temperature anomaly patterns transition across coastlines: we required that there should be little discernible change in that pattern. Before our optimized matching, the land anomalies were noticeable warmer than the ocean anomalies as features crossed coastlines. But after optimization in our Version 6 dataset, here’s the LT anomaly map for last month (October 2019), which shows no evidence for land-vs-ocean artifacts.

UAH-global-LT-land-minus-ocean-2Fig. 2. October 2019 LT temperature anomalies relative to the 1981-2010 average annual cycle. Note the anomalies have a smooth transition between land and ocean, as would be expected for deep-layer tropospheric temperatures (but not necessarily surface temperatures).

Nevertheless, adjustments like these are never perfect. So, the question remains: Is there a spurious change in the L-O temperature difference occurring in 1998?

Evidence that the L-O change in 1998 is real

There are a few lines of evidence that suggest the May 1998 step-up in L-O temperatures is real.

First, if the effect was due to the introduction of AMSU in 1998, it would have occurred in August, not in May (3 months earler). Also, the effect should have been gradual since for almost 4 years after August 1998 the LT dataset is half MSU (NOAA-14) and half AMSU (NOAA-15), after which it becaume 100% AMSU.

But a more important piece of evidence is the effect of El Nino and La Nina on L-O. During El Nino, the ocean airmasses warm more than the land airmasses (especially in the tropics), so that L-O tends to be more negative. Up until the 1997-98 super El Nino a period of greater El Nino activity existed, after which a shift to more La Nina activity occurred. (This is probably also what caused the extended global warming ‘hiatus’ after that El Nino event.)

I statistically regressed the L-O values in Fig. 1 against 3-month running averages of the Multivariate ENSO Index (MEI), and removed that estimate of the ENSO influence from the data. The resulting ENSO-adjusted time series in shown in Fig. 3.

Fig. 3. As in Fig. 1, but with the average influence of El Nino and La Nina (ENSO) subtracted out. Note the evidence for a “break” in 1998 is much weaker.

Note the step-up in mid-1998 is much less evident, and the 5th order polynomial fit to the data is smoother with a more gradual transition in L-O over the 41-year satellite record.

But that’s not the only thing going on during this period that affects the L-O values. There were two major volcanic eruptions (El Chichon in early 1982, and especially Pinatubo in mid-1991) that caused more cooling over land than ocean, causing temporarily enhanced negative values in L-O. Since these events are not as easily correlated with an index like MEI is with ENSO, I simply removed the data from 1982-83 and 1992-93 in Fig. 3 and replotted the results in Fig. 4.

Fig. 4. As in Fig. 3, but with the data influenced by major volcanoes El Chichon and Pinatubo removed.

Now we see that the 5th order polynomial fit to the data comes quite close to the linear trend (dashed gray line), which suggests that the step-up in 1998 in L-O was real, and related mostly to a change in ENSO activity before versus after the 1997-98 super El Nino, and with the major volcanic eruptions in 1982 and 1991 contributing to the seemingly spurious feature.

The remaining upward trend in L-O is simply the land airmasses warming faster than the ocean, as would be expected for any warming trend, whether natural or human-caused.

There remains what might be a spurious feature during 1980-81 in Fig. 4, which would most likely be related to our ad hoc correction for MSU channel 3 drift during that time. This, however, should have little influence on the land and ocean trends as evidenced by the trend line fit (dashed gray line) in Fig. 4.

0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
November 7, 2019 10:43 pm

I just watched a 4 minute video of John Stossel trashing the Paris Accord.

Sounds good, except that Stossel concludes “the world IS [Stossel’s emphasis] warming, and humans may be playing a role in that.”

I don’t believe the world is warming.

And we should stop simply accepting that premise from the leftists. Remember 1999 NASA data [before it was “adjusted”] showed the US to be warmer in the 1930s. NASA’s Chief Scientist confirmed that with this statement:

In the U.S. the warmest decade was the 1930s and the warmest year was 1934.” -James Hansen, NASA, 1999, quote at this link: https://www.giss.nasa.gov/research/briefs/hansen_07/

The US had in the 1930s and beyond a much more extensive and reliable set of thermometer stations compared to VERY spotty global record, with Africa, Antarctica, Asia, and the Arctic having sparse records, but the Chicken Littles have reconstructed the global temps based on that weak record.

With their adjustment they’ve also heated the present and cooled the past to no end. In addition, and the urban heat effect is real, and makes current global temperatures substantially hotter.

The 1930s were probably hotter than today. Globally. There is no warming imo.

Joel O’Bryan
Reply to  Eric Simpson
November 7, 2019 11:15 pm

Even if there is warming ongoing to today since satellite MSU data started in ‘79, the 1910-1945 warming was certainly 100% pure nature, as the deltaCO2 was too small to have mattered. But according to the IPCC this natural variation ceased after 1979 onwards and only human’s CO2 did everything/anything since.

The IPCC also must believe in magic and fairies.

R Stevenson
Reply to  Joel O’Bryan
November 8, 2019 6:09 am

Of course the 1910-1945 warming may have included a modicum of warming by virtue of the second world war not forgetting the 14-18 conflict- quite a lot of BTUs were released then.

Reply to  Joel O’Bryan
November 8, 2019 8:49 am

The global temp is always warming or cooling — never static like Mann’s mann-o-matic hockey-schtick. OK, it’s warming now. Good — the further we get away from the coming glaciation, the better.

Reply to  Eric Simpson
November 7, 2019 11:18 pm

I’m sorry but on re-examination I realize I failed to directly state my main point, which is:

The extensive US temperature record from the 1930s (and beyond) probably reflects global temperatures better than the global record itself, because the global record is so incredibly spotty, being virtually non-existent for most of the world.

Adam Gallon
Reply to  Eric Simpson
November 8, 2019 2:14 am

Very good records for Europe. CET, for example. http://www.climate4you.com/images/CentralEnglandTempSince1659%201100pixel.gif
Kills your idea of the 1930s being hotter.

Reply to  Adam Gallon
November 8, 2019 8:28 am


No, the CET data is a statement about Central England in the 1930’s

Reply to  JaneHM
November 8, 2019 1:28 pm


Many authorities including the British and Dutch met office consider CET to be a good if not perfect proxy for the northern hemisphere.

Mind you, the 1920’s and 1930’s were very warm generally in the arctic


Reply to  JaneHM
November 8, 2019 4:17 pm

Don’t be silly Jane. Many countries with excellent records. Last two decades in the US easily warmer in the US than in the 30s. And still increasing.

Reply to  Adam Gallon
November 8, 2019 9:33 pm

My understanding is that Phil ‘adjusted’ CET and lost the original data.
That was his statement when asked about the original unadjusted data.

Reply to  Eric Simpson
November 8, 2019 2:28 am

As far as I know every official global temperature data set has found a spurious warming bias in the early part of the US record. This is very well documented and is due mostly to time of observation bias (TOB); the switch in the early 1960s from volunteers reading temperatures between 5 and 7 PM to between 7 and 9 AM. The later readings were on average ~0.5C warmer than the morning readings at the same stations. You can easily see the step changes caused by TOB in the raw US station temperature records as published by BEST, for e.g.


The temperature in St Joseph (in this example) didn’t suddenly fall by 0.5C in the early 1960s. They know that because nearby stations not affected by TOB show no such fall. There is no getting away from the fact that TOB in raw US station data up to ~1960 is a real phenomenon that has to be accounted for in any fair comparison between older and newer temperature records. You have to remove this discrepancy somehow, either by adding ~0.5C to the more recent temperatures or subtracting it from the older temperatures. If you don’t remove such an obvious disparity then any trend you generate from the data will be unreliable.

Gordon Dressler
Reply to  TheFinalNail
November 8, 2019 7:09 am

TFN, you concluded with “If you don’t remove such an obvious disparity then any trend you generate from the data will be unreliable.”

I think clarification of the term “any trend” is needed. If the trend covers a period without a shift in time-of-observation (case 1, all volunteer readings are between 5 and 7 PM; case 2, all volunteer readings are between 7 and 9 AM), the associated TRENDS should be reliable. In fact, both case 1 and case 2 trends could be considered “accurate” as long as their datasets note the associated time-of-observation of the volunteer data relative to stations that, say, average data obtained hourly between 7 AM and 7 PM.

It is only trends over periods that include the switch in sampling time-of-observation that will have significantly less reliability (without any TOB corrections such as you noted).

You are obviously aware of this, with your statement “. . . TOB . . . is a real phenomenon that has to be accounted for in any fair comparison between older and newer temperature records.”

Thank you for pointing out this shift in average US temperatures due to TOB that needs to be considered when looking at data prior to the early-1960’s . . . it invites the question of to what extent TOB permeates historical datasets of other nations that are used in calculating “global temperatures” and, most importantly, to what extent this may have created “artificial” global warming.

Kerry Eubanks
Reply to  TheFinalNail
November 8, 2019 8:27 am

Tony Heller has extensively documented, with the very simple “adjusted minus raw” temperature records, the fact that these adjustments are more or less linear, cooling the past the most early in the record, warming the adjusted temperatures the most in the most recent years. No step-wise adjustment for TOB, and the exact wrong direction of adjustment considering the number of stations subject to UHI affects.

This “tampering” has been demonstrated for scores of individual reporting stations, never mind the averaged data set. It is wrong to “adjust” the actual record as appears to be happening with the Australian MET. Show the historical data as it was actually recorded, explain discontinuities such as TOB if they exist, and let people calculate trends whichever way they (one would hope transparently) choose.

James R Clarke
Reply to  TheFinalNail
November 8, 2019 10:56 am

I would argue that the TOB correction should be far less than 0.5 C, and the UHI correction should be far greater than it is.

The argument from the warmest is that corrections are necessary, which is true, but they use that argument to justify their values, which is false. The adjustments are performed by people with a warning agenda, and the values they ‘calculate’ reflect their bias.

A C Osborn
Reply to  TheFinalNail
November 8, 2019 1:20 pm

Sorry, you lost all credibility quoting and showing BEST data.
Their difference from regional average is total bullsh!t for making adjustments or decisions.
They used it on long running stations and then superimposed the regional trend (Final product) on them totally destroying the original information.
Let me quote Mr Mosher at Climate etc on what BEST does.

Steven Mosher | July 2, 2014 at 11:59 am |

“However, after adjustments done by BEST Amundsen shows a rising trend of 0.1C/decade.

Amundsen is a smoking gun as far as I’m concerned. Follow the satellite data and eschew the non-satellite instrument record before 1979.”

BEST does no ADJUSTMENT to the data.

All the data is used to create an ESTIMATE, a PREDICTION

“At the end of the analysis process,
the “adjusted” data is created as an estimate of what the weather at
this location might have looked like after removing apparent biases.
This “adjusted” data will generally to be free from quality control
issues and be regionally homogeneous. Some users may find this
“adjusted” data that attempts to remove apparent biases more
suitable for their needs, while other users may prefer to work
with raw values.”

With Amundsen if your interest is looking at the exact conditions recorded, USE THE RAW DATA.
If your interest is creating the best PREDICTION for that site given ALL the data and the given model of climate, then use “adjusted” data.

See the scare quotes?

The approach is fundamentally different that adjusting series and then calculating an average of adjusted series.

in stead we use all raw data. And then we we build a model to predict
the temperature.

At the local level this PREDICTION will deviate from the local raw values.
it has to.

Tom Abbott
Reply to  Eric Simpson
November 8, 2019 5:50 am

“The extensive US temperature record from the 1930s (and beyond) probably reflects global temperatures better than the global record itself, because the global record is so incredibly spotty, being virtually non-existent for most of the world.”

The global temperature records *are* spotty, but one interesting aspect is all those unmodified temperature charts resemble the temperature profile of the U.S. surface temperature chart, circa 1999 (Hansen 1999), which shows that the 1930’s was as warm or warmer than today. So all the unmodified surface temperature charts from all over the world show the same thing. They show that the Earth is NOT experiencing unprecedented warming today, it was just as warm in the recent past.

No unprecedented warming means CO2 is not a major player in the dynamics of the Earth’s atmosphere. There is no CAGW (Catastrophic Anthropogenic Global Warming). Humanity can relax and go on about their business minus the windmills.

A lot of Sekptical people wanted the temperature record explored using a Red Team/Blue Team sort of debate. That didn’t pan out for political reasons. Apparently, some Republicans think they need to pander to what the brainwashed young are thinking about the climate, rather than the science of the climate, so they nixed the Red Team/Blue Team discussion.

I would like to suggest another way to get at the same problem, and I think it is perfectly reasonable, to the point that even clueless Republicans in the White House might sign off on it. How about us encouraging President Trump to require NASA Climate to explain the changes they have made to the U.S. surface temperature chart. That’s all. That doesn’t seem like an unreasonable request. Why shouldn’t NASA Climate be required to show their work? Then, after they show their reasoning for the changes they made, the discussion can begin.

Here’s what I’m talking about. NASA Climate has bastardized the global surface temperature charts for years and now they have started bastardizing the well-documented U.S. surface temperature chart and they are even getting into bastardizing U.S. State surface temperatures.

NASA Climate changed the Unmodified U.S. surface temperture chart (James Hansen’s baby) into a bogus, bastardized Hockey Stick “hotter and hotter” chart. President Trump should require that NASA Climate explain these changes to the American public.

NASA Climate should not have a problem with this request, unless they are dishonest, no-good, so-and-so’s who are manipulating the temperature records for political purposes. NASA Climate should be required to make all their reasoning available to their bosses, the American People.

Who could reasonably object to this? You mean to say NASA Climate cannot or will not explain why they do what they do? Well, that would be unacceptable, wouldn’t it? They work on the taxpayer’s dime and they owe the taxpayers an explanation of anything and everything they do on our time.

Unmodified US surface temperature chart:

comment image

Bastardized US surface temperature chart:

comment image

NASA Climate has taken the unmodified US surface temperature chart from showing the 1930’s to be as warm as today, to showing the 1930’s as being much cooler than today, and today appears to be the hottest temperatures in recorded history.

The first unmodified version of the US surface temperature chart shows that CO2 is not a significant player in the Earth’s atmosphere, whereas, the bastardized US chart shows the temperatures getting hotter and hotter for decade after decade and showing the temperatures now at the hottest temperatures in history.

The first chart requires no action on humanity’s part because CO2 is not a problem. The second chart screams CAGW!, we must DO something!

The first chart is the real world. The second chart is the Big Lie, that the Alarmists are using to frighten the world out of their senses.

President Trump should require NASA Climate to explain themselves and their work. We want to know why NASA Climate gave the traditional, very well documented US surface temperature record a completely different look. A scary, end-of-the-world look.

Expain yourselves, NASA Climate.

Reply to  Tom Abbott
November 8, 2019 1:08 pm

Would NASA Climate be compelled to respond to a FOI request?

Tom Abbott
Reply to  fish
November 11, 2019 6:30 am

Bureaucracies can draw out Freedom of Information requests for a very long time, if that is their motivation.

A presidential order would cut through all that and force NASA Climate to comply. And as I say, all the presidential requirement would be asking for is that NASA Climate explain how they got from one benign temperature chart profile to a very scary temperature chart profile when none of the historic data had changed. The only changes took place in NASA Climate computers.

NASA should explain why their computer models override the actual historic temperature measurements. A perfectly reasonable request. All scientists should be required to show their work, and most of them do, except for NASA Climate, who is in charge of one of the most important data bases in the world. A database that tells us whether we have nothing to worry about concerning CO2 or one that tells us we are in big trouble because of CO2. The American people deserve the truth of this matter and the president should demand that NASA Climate provide the explanation for the steps they have taken.

John Endicott
Reply to  Eric Simpson
November 8, 2019 7:07 am

Sounds good, except that Stossel concludes “the world IS [Stossel’s emphasis] warming, and humans may be playing a role in that.”

I don’t believe the world is warming.

Warming is a relative term, as such it depends on what it’s being related to. The World *has* warmed (and generally was warming up until the pause. whether warming has continued or we’re heading into a cooling period is something that can only be determined after the fact) since the depths of the little ice age. That’s simply a historical fact. Similarly, it’s correct to say the world has warmed since the end of the 1940s to 1970s cooling period (during which it was, obviously, cooling down from the just as warm as today, or even slightly warmer, 1930s).

As for Man’s contribution? there’s no evidence that man’s CO2 is anything more than a negligible factor at best. I say man’s land-use changes (Damning rivers, clear-cutting forests, strip mining, etc) has a much greater impact on the climate (at least in so far as the areas around the land-use are concerned) then the additional plant food we’ve added to the atmosphere.

Reply to  Eric Simpson
November 9, 2019 9:39 am

Define warming are we talking energy or temperature?
My conclusion from this article is once again which direction the wind blows sets the temperature.

Joel O’Bryan
November 7, 2019 10:47 pm

Figure 2 looks qualitatively like how I would expect the next 110kyr glaciation to start. A warm Alaska, Bering and Chukchi Seas providing moisture to a cold Canada, Northern US interior beginning in the Rockies, spreading eastward to the GLakes and up to Hudson Bay. Also the glaciation in a cooling Scandinavia. And a deepening cold around WAIS.

As far as the ‘98 step change goes… NASA/GISS, NOAA/NCDC, and Hadley have so bastardized the surface data and ERSST data with cooling the past warming the present adjustments – insufficient UHI adjustments, loss of rural stations, ship intake vs buoy data chicanery, and other jiggery (BOMs Darwin station) no one really knows what is going on, except fraud.

Reply to  Joel O’Bryan
November 7, 2019 11:53 pm

Tell it like it is. The temp record has been damaged beyond all recognition. It’s criminal.

The AGW scare story will have to get buried in meters of snow to finally disappear,
but still people will say that’s a sign of warming. Madness rules the world at the moment.

Ben Vorlich
Reply to  Scarface
November 8, 2019 12:40 am

And it’s all our own fault whatever.

Randy A Bork
Reply to  Joel O’Bryan
November 8, 2019 4:48 am

This case is not relevant to any of those datasets, it is UAH set only under discussion, correct?

Joel O’Bryan
Reply to  Randy A Bork
November 8, 2019 6:09 am

The ‘98 step change is also in the adjusted surface data sets. That is until it gets adjusted out in the future to remove the blip.

November 8, 2019 12:52 am

Eric Simpson
“I don’t believe the world is warming.”
I am happy to take the satellite records of Roy which show slight decadal warming 0.13C per decade perhaps .
I am happy to believe the world can warm or cool, and on balance is probably a little warmer.
I agree that many aspects of the artificially modified surface temperatures are suspect.
You might be better saying any effect is slight.
“The extensive US temperature record from the 1930s probably reflects global temperatures better than the global record itself”.
The extensive US temperature record from the 1930s show the US temperature records. Any global record by definition shows the global temp better than one isolated area of the globe.
The US records do have a much better number of observations.

Reply to  angech
November 8, 2019 1:46 am

I appreciate your points, angech. Remember, huge chunks of the global record (maybe over 85%) is totally missing (ie nearly all of Africa, the Arctic etc). I don’t know how then you get a valid global temperature from the 1930s. Further, adding to that all the “extrapolation” by the alarmists to fill in the missing 85% of data (or whatever it is) makes the global record little more than a joke.

But tons of US stations existed in the ’30s, an without the warmist manipulations those stations show that we were warmer in the ’30s then now.

Or go with tree ring data. Isn’t the documented decline in the temps in the US after the ’30s also seen in the global tree ring data?? Isn’t that what Mann was hiding?

Remember also the record for the hottest day on the globe was set in 1913. If there has been warming since then it’s been minuscule because that record stands, 106 years later.

Gary Mount
Reply to  Eric Simpson
November 8, 2019 2:02 am

The decline is in reference to proxy temperatures that declined after the 1960s. For more details see :


Reply to  Gary Mount
November 8, 2019 3:01 am

From the graph as your wuwt link it looks like the decline started before the ’60s, no?

Regardless, is that decline consistent with the thermometer documented decline in US temps after the ’30s. I mean, should it override the worthless global temp record from that time, and give us a good indication that the US decline in temps paralleled a global decline?

Perhaps it’s time again to dust off the old video and rewatch:

Gary Mount
Reply to  Eric Simpson
November 8, 2019 3:38 am

The decline meant that the proxy data wasn’t matching the temperature gauges data. They were going in opposite directions.

Don K
Reply to  Eric Simpson
November 8, 2019 4:22 am

If you believe tree ring data, planetary temperatures peaked in the 1940s (latewood density) or 1950s(ring width). Which is why no one except Michael Mann has much faith in it. And Mann only believes the parts that he likes. Apparently only the part he likes is “Science”. The rest isn’t.

The technical term is the “Divergence Problem” https://en.wikipedia.org/wiki/Divergence_problem.

Jim Gorman
Reply to  Eric Simpson
November 8, 2019 7:33 am

Logic would dictate that if “global temperatures” are increasing, and that well mixed CO2 is causing it, then temperatures most everywhere must be rising also. This includes the U.S.

Reply to  Eric Simpson
November 8, 2019 3:24 pm

Good points Eric. Did not know about 1913.
The change in tree rings proved that Mann should not have used them.
He should be ashamed.

John Endicott
Reply to  Eric Simpson
November 11, 2019 9:22 am

Remember also the record for the hottest day on the globe was set in 1913

Yes and no.

On Sept. 13, 1922, El Azizia, Libya made history after a weather station there recorded the highest temperature ever directly measured on Earth: a blistering 136.4 degrees Fahrenheit However, 90 years later the WMO concluded that the El Azizia record measurement could be inaccurate by as much as 7°C due to a combination of factors including the asphalt-like surface over which the measurement was taken, which is not a fair representation of the native desert soil. with the elimination of El Azizia record, that left the 10 July 1913 record of 56.7°C (134°F) from Greenland Ranch, Death Valley, California as the official hottest.

Either way, the hottest record is from the early 20th century and still stands (at least until the alarmist manage to cool the past sufficiently to let a less hot modern temp take the prize)

Tom Abbott
Reply to  angech
November 8, 2019 6:10 am

“The extensive US temperature record from the 1930s show the US temperature records. Any global record by definition shows the global temp better than one isolated area of the globe.”

Well, there are no “global” temperature records before about 1979. What we do have before 1979 is unmodified regional and local temperature records and all of them, at least all the ones I have seen, from all over the world, resemble the temperature profile of the U.S. surface temperature chart, where the 1930’s show to be just as warm as it is today.

So the real global surface temperature profile is the one represented by the unmodified U.S. surface temperature chart. The US surface temperature chart shows it was just as warm in the 1930’s as it is today. CO2 was not a significant factor in the 1930’s, even according to the IPCC, so if it is no warmer today than in the 1930’s, then that means that all the CO2 humans have pumped into the atmosphere since the 1930’s has not increased the Earth’s surface temperature beyond what was reached in the 1930’s. This means CAGW is dead. This is why the Climategate Charlatans and other Data Manipulators set out on a course to change all the surface temperature charts into “hotter and hotter” Hockey Stick charts in an effort to promote a political agenda and scare all of us to death over CO2. It’s criminal, what they did, and are doing..

Reply to  Tom Abbott
November 8, 2019 12:02 pm

“So the real global surface temperature profile is the one represented by the unmodified U.S. surface temperature chart.”

YES, that’s what I was saying is probably the case!

And it just so happen that the tree data aligns with this.

And with US state temperature records, with the vast majority of the hottest day records for states set decades ago like in the ’30s and few set recently: https://wattsupwiththat.com/2012/10/24/climate-and-state-high-temperature-records-wheres-the-beef/

AND .. it aligns with the record for the hottest day ever on this planet. If the globe really had been warming precipitously for over a century then that record should have been broken time and time again. BUT NO. The 1913 record for the hottest day ever still stands. That doesn’t jive with the warmist picture, and all their “adjusted” temperature charts.

Reply to  Tom Abbott
November 8, 2019 1:48 pm
Coeur de Lion
November 8, 2019 12:59 am

If it gets as warm as the Medieval Warm Period then we should start to panic, move to Green Land and also build a few magnificent cathedrals. Oh, and sell wines to the French ho ho.

November 8, 2019 1:16 am

“ Is there a spurious change in the L-O temperature difference occurring in 1998?”
A funny question.
There is a change.
It could be due to the “The year 1998 is key for our dataset because that is when the first (NOAA-15) Advanced Microwave Sounding Unit (AMSU) came online”.
This would not be spurious. It would be due to either
1. A real upward trend in temperatures at this time or
2. A mistake in the empirical adjustment of 0.16 C
Spurious would be an accidental mistake of 0.16 C in the code entered by a technician by accident or design.
Finding it would involve complicated rechecking of the code used at that time, not the algorithm.

Perhaps the problem arise through this,
“The way we gauged the match between MSU and AMSU is how the temperature anomaly patterns transition across coastlines: we required that there should be little discernible change in that pattern.”
My expectation of coastline anomaly patterns is that they should be quite anomalous. Borders are tiger country. No reason for smooth transition at all. In smoothing out the expected sudden changes, particularly warming lines from sea to land, you may easily incorporate a 0.16C error.
Perhaps you need to add an area of rapidly changing anomaly at the borders.

Reply to  angech
November 8, 2019 9:43 am

The question was about the jump-up being “spurious”, as in not real.

And, as mentioned in the article, while *surface* temperature anomalies can change across coastlines (and even be of opposite signs), they do in in deep-layer tropospheric temperatures. The resulting pressure differences would eliminate the contrast through baroclinic instability.

Reply to  Roy W. Spencer
November 8, 2019 3:55 pm

Thanks for the reply.
I appreciate that you and Christie have put years of work, knowledge, blood, sweat and tears in your efforts to generate as accurate and reliable weather and temperature data from the satellites as possible.
Also that the scientific knowledge of yourselves and your team and the people supplying the instruments to you and the analysing and reanalysing that you do is world class and supported by other world class scientists in this and related fields.
It is a shame that some mean spirited and mean minded commentators blinded by their noble cause corruption choose to denigrate the very science they otherwise promulgate.

The point was that the way the smoothing was chosen, and all graphs do need to be “smoothed” to create a useable function, may have inadvertently created a spurious difference in 1998.
Although resulting pressure differences will lead to a real world smoothing (eliminating the contrast) there are known differences which the science does not adequately explain hence the empirical adjustments at this point in time may be “right” for merging the data sets but “wrong” in the new level the data sets are set at. It is easy to assume (hence obviously wrong on my part) that adjusting between a current and a new warmer in areas set will lead to a spurious warmer combined set compared to the original.
The comment about it happening 3 months? before the adjustments are made is provocative.
Did the adjustments flow over backwards into the measurements to help smooth the transmission at this point?
Zeke does this automatically all the time with his data adjustments.*
Did the other surface and ocean data sets show a real surge at this time and if so could the data just have been set at this real for the time but spurious for the data set base?
Finally seeing that the other data sets (Zeke)* are constantly changing and that you incorporate some of this data to balance and calibrate your instruments how do you manage and merge the continually changing past data sets from other sources?
Do you use the original data out had unchanged,the only way to keep your calibrations pristine or does your data output change to agree with the backward changes generated by the other data sets?

Reply to  angech
November 9, 2019 1:32 am

It’s nothing but noise anyway. Real resolution is 1K. Should be reported as 0 plu-minus 1K.

Ron Long
November 8, 2019 2:11 am

Nice reporting, Dr. Roy. I am happy to see you discuss the importance of El Niño and La Niña. If you are in the agriculture or ocean fishing business in South America you look at ENSO conditions and forecasts to plan your season. Basically, and it involves willing to accept some hail damage to be guaranteed summer thunderstorms delivering rain, El Niño is a good boy and La Niña is a bad girl

Geoff Sherrington
November 8, 2019 3:44 am

Slightly OT, but the Connollys are postulating (speculatively) oxygen dimers and trimers above the tropopause to explain a line of fit divergence of some functions derived from the ideal gas laws and seen in lapse rate type explanations.
Question is, would these polymers be able to emit oxygen brightness microwaves at all; or ones that could be confused with those used by UAH? Would there be a divergence of thermometer and microwave temperatures in horizontal slices of the atmosphere with and without these postulated oxygen polymers?

Reply to  Geoff Sherrington
November 8, 2019 10:56 am

I really cannot tell what they are talking about. Any imperfections in our understanding of what happens around the tropopause are small and inconsequential to global warming theory. The 50-60 GHz oxygen absorption theory is pretty good (as evidenced by our local comparisons between radiosonde measurements and the satellite data, in monthly averages). Any errors in the theory will be more in the form of biases or offsets, which we take out of the data anyway.

Geoff Sherrington
Reply to  Roy W. Spencer
November 8, 2019 4:50 pm

Thank you, Roy, including for your patience with this OT question.

Thomas Homer
November 8, 2019 6:04 am

Instead of using spotty temperature records to conclude that back radiation has increased, imagine if we could measure back radiation.

If we had been measuring back radiation we could derive patterns and answer simple questions like:
– how does back radiation behave in the equator’s atmosphere relative to the pole’s atmosphere given the atmosphere is twice as thick at the equator
– how do atmospheric tides, i.e. changes in atmospheric thickness due to the moon’s phase, affect back radiation
– how does back radiation behave on Mars with 95% CO2?
– how does altitude affect the behavior of back radiation, for example Miami, FL (sea level) has a mile more of the most dense atmosphere than does Denver, CO (mile high) yet Denver typically reaches over 100F several times a year while Miami has never recorded a 100F temperature
– how does day/night affect back radiation
– how does humidity affect back radiation

We don’t know any of these answers because we’re not measuring back radiation.
– we could be measuring back radiation at locations of various elevations
– we could be measuring back radiation on planes
– we could be measuring back radiation on trains
– we could be measuring back radiation during an eclipse
– we could be measuring back radiation in the proximity of forest fires and volcano eruptions
– we could be charting back radiation measurements and derive patterns that lead to tools of reason like Laws, Axioms, Postulates and formulae and extend our scientific knowledge

Jim Gorman
Reply to  Thomas Homer
November 8, 2019 8:03 am

I have often wondered this myself. Satellites should be able to measure outgoing radiation very well. It should be quite possible to measure LWIR coming back toward the earth. We should then be able to track how that radiation difference affects temperature.

Reply to  Thomas Homer
November 8, 2019 9:55 am

such measurements ARE made, here in the U.S. (link below) and some in a few other countries. The equipment is a little pricey, though. The CERES satellite also has “retrievals” of downwelling radiative flux at the surface, but that’s not the same as measurements.


It should be that these SURFRAD stations measure the downwelling IR as a residual, the difference between the net IR at the surface and the surface-emitted (upwelling) IR based upon the instrument’s surface temperature and emissivity. A cooled-detector spectrometer can measure the downwelling IR more directly, but it’s much more expensive.

Johann Wundersamer
Reply to  Thomas Homer
November 20, 2019 9:37 am
Carlo, Monte
November 8, 2019 7:10 am

I would be very hesitant to call Fig. 3 a “linear” trend inside all the noise.

Carlo, Monte
Reply to  Carlo, Monte
November 8, 2019 8:40 am

Error: Fig. 4

November 8, 2019 9:33 am

This is not unique in data sets, accuracy of none should be taken for granted.
I suspect the step of about 0.5C. (see link) might be rare, but I suspect it might have political background to it.

Reply to  Vuk
November 8, 2019 11:07 am

Since beginning of the month my Sunspot count page SSN.htm has had dozens of hits, I hope this is not some kind of malicious bot.
If it is a genuine interest in the October count, I have failed to update it since I am far away from my desktop.
According to SIDC-SISLO October SSN was about 0.3 on the old classic count or 0.4 on the new revised method of counting.

Dave Fair
November 8, 2019 10:49 am

I believe that Dr. Spencer makes the case that, when calculating multi-decadal tropospheric temperature trends, one need remove major stratospheric-penetrating volcano effects. I think we need a period longer than 40 years to tease out ENSO effects.

November 9, 2019 6:20 am

How do they factor out the UHI Effect in the step up? We had a huge U.S. Government-controlled construction boom thanks to an influx of government-regulated spending/financing of the BIG banks. Lots of concrete and asphalt poured/laid during that time period all across the planet (China and India).

%d bloggers like this:
Verified by MonsterInsights