From the “dashed hopes for the warmest year evar!” department comes this update from Dr. Ryan Maue on the global surface temperature:
Via Twitter:
Global temperatures have generally settled to +0.26°C compared to 1981-2010 climatology continuing downward glide thru 2017 (black line)
He adds:
Tropical vs. non-tropical temperature anomalies have balanced out mostly for the past few months. No El Niño suggests continued T levels:
In a nutshell what Dr. Maue is saying is that without a strong El Nino event to boost temperature, global temperatures are stabilizing around +0.26°C. FYI, NCEP data used in these plots is from the NOAA National Centers for Environmental Prediction. The data is available here: http://cfs.ncep.noaa.gov/cfsr/downloads/
Recently, we covered a story from the Australian Bureau of Meteorology (BoM) that called off their El Niño watch. BoM says:
All eight international models surveyed by the Bureau of Meteorology now suggest tropical Pacific Ocean temperatures are likely to remain ENSO-neutral for the second half of 2017.
It’s a tough business to be in when your CO2 driven “climate change” can’t get there unless a natural ENSO event pushes up the temperature for you. Meanwhile, Justin Gillis at the New York Times claims “Earth Scorching CO2” is higher than ever while temperatures stabilize at a value that is the same as about 1980 (0.27°C), according NASA’s GISTEMP:
Just look at the sea surface temperatures, there’s not a lot of warm water:
We live in interesting times.
NOTE: I expected some complaints about comparing GISS and NCEP graphs, and there were plenty. I did it to illustrate a point.
Which one is the RIGHT temperature anomaly? Anomalies are all products of their baselines, and baselines are a choice of the publisher.
If NASA GISS is to be believed as the world’s most cited source for global temperature, then 0.27C is correct for 1980.
Unfortunately, they have been living in the past, and refuse to update their baseline. UAH did it, RSS did it, NOAA/NCEP did it….why not GISS? The answer: Gavin Schmidt.
Not sure about BEST: They don’t list their baseline period in their graph: http://berkeleyearth.org/wp-content/uploads/2015/03/land-and-ocean-summary-large.png
This is why absolute temperatures don’t suffer from the choices made by the researchers for the anomaly baseline. There’s no musical chairs with anomaly baselines.
It would be nice if GISS got with the program used the 1981-2010 baseline like other data sets, or all the climate data publishers agreed on using one baseline. For example, here’s a BEST plot with all the baselines adjusted to NASA GISS 1951-1980.
The general public really doesn’t care or know about anomaly baselines – they just want to know what today’s temperature is relative to the past.
Standardizing on one baseline for all climate data sets would make that easier for the public consumption. I’m sure that call for standardization of baselines will fall on deaf ears at NASA GISS, where their lead researcher, Gavin Schmidt, is so petty he can’t even appear on the same TV set with another researcher.
Let the squawking begin.
UPDATE: To further illustrate the point about different baselines giving different results to the public, here is the HadCRUT4 data, which uses a 1961 to 1990 baseline:
According to their data (which is mostly the same raw GHCN data used by NASA GISS, plus some others) their 1980 temperature anomaly was somewhere around 0.1°C (see green lines intersection), where GISS says 0.27°C
Again, why can’t climate science do a simple thing like standardize on a baseline period ?
I’ve added a caveat in the title to reflect this: (depending on who you ask)
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
FERDBERPLE
there is no international standard to calculate the average temperature, and depending upon the algorithm you choose, it is possible to show the earth on average is both warming and/or cooling at the same time.
HENRY
you are right. my results show it is already cooling, if you look at a globally balanced sample
maxima or minima
in degrees C/annum
Forrest Gardener
Because it fits 100%? To define a function you need at least 4 points.
Admittedly, the period I looked at (1973-2015) is approximately half a Gleissberg cycle.
So, the whole wave is a sine wave, wavelength 87 years.
Still, I think for the half GB the parabola proves my pint, i.e. all warming and cooling is natural.
Man made warming either does not exist or is too small to make even make a dent in what nature gives us,
hence my correlation of 100% for the speed of warming/cooling.
they had that already more or less figured out before they started with the CO2 nonsense:
http://www.cyclesresearchinstitute.org/cycles-astronomy/arnold_theory_order.pdf
Rick Perry, Sec. of Energy, is discussing the need for nuclear energy and environmental issues right now on todays White House Press briefing.
Rick Perry: “Climate changes, always has. Mankind is contributing to it. The question is how much. Let’s have a conversation about that.”
Humans contribute to local climate change, but it doesn’t add up to enough noticeably to affect the global average. Unless your only data come from urban heal islands which used to be cool, dark forest.
I was just looking at these National Weather Service guidelines:
1. Place the thermometer 5 feet above the ground (+/- 1 ft.). A thermometer too low will pick up excess heat from the ground and a thermometer too high will likely have too cool of a temperature due to natural cooling aloft. 5 ft. is just right.
2. The thermometer must be placed in the shade. If you put your thermometer in full sunlight, direct radiation from the sun is going to result in a temperature higher than what it should be.
3. Have good air flow for your thermometer. This keeps air circulating around the thermometer, maintaining a balance with the surrounding environment. Therefore, it is important to make sure there are no obstructions blocking your thermometer such as trees or buildings. The more open, the better.
4. Place the thermometer over a grassy or dirt surface. Concrete and pavement attract much more heat than grass. That is why cities are often warmer compared to suburbs. It is recommended to keep the thermometer at least 100 ft. from any paved or concrete surfaces to prevent an erroneously high temperature measurement.
5. Keep the thermometer covered. When precipitation falls, you do not want your thermometer to get wet as that could permanently damage it. A Stevenson screen is a great place to store thermometers and other instruments as they provide cover as well as adequate ventilation. If you can’t get one, a simple solar radiation shield is adequate.
… and wondering how a person places a thermometer in the shade and out of direct sunlight AND keeps it in the open and positioned for good air circulation at the same time?
I still see the perfect temperature-measuring spot as elusive. How does anybody agree that they are even measuring the same thing consistently from one location on Earth to the next ?
Forget whether experience agrees with a scientific guess or not, Mr. Feynman (yeah, THAT Feynman) — just tell me how the heck do I even take a blasted temperature measurement to help confirm a scientific guess or not !
Measuring in precisely the same way, at the same times of day at exactly the same spot will tell you about the changes there over time. But how many such good locations are there? And how can they represent the planet?
Hence, satellites and balloons are the only even remotely good enough data for scientific purposes. Floating temperature gauges in the oceans move around too much. Even balloons aren’t sampling exactly the same volumes of the atmosphere.
Probably not many. But as long as they in general do the same bad thing, the day to day change will be as good as it can get.
This is one of the reasons I follow the day to day change in a single station, subtract yesterday’s Tmin from today’s Tmin as difference, intra-day change Tmax-Tmin, and average Tmin and Tmax.
You can look at all three, and get a good idea of what’s happening where we have surface stations, and they they change from day to for long periods
NASA (the National Adjusting the Science Association) CAN’T use 1980-2010 as a baseline period, because they haven’t yet decided how how or cold that period was. Cooling the past can’t begin while youre still using part of that period in your hottest evah decade claims.
Thanks for the CFS data link.
The low solar activity cooling regime is in place & will continue until SC25 starts, 2019-20.
Using my tried and true F10.7-TSI-SST model, which is based an intimate perfect working knowledge of the temporal relationship between these three measures, I’ve estimated Had3SST will drop a further 0.27C to 2020 from the Dec 2016 Had3SST value of 0.447C, to 0.178C (+.05/-.1), putting the end of solar cycle value (if it were to end at the year-end) of somewhere between just above to just below the cycle SST yearly starting value of 0.141C in 2008.
CFSv2 2m will drop along with it, possibly going to “zero” or negative by 2019-20.
This estimate will be updated in January 2018 after the 2017 numbers are all in. It could go lower.
The very best way to keep up on ocean warming/cooling is the daily 7-day SSTa change:
http://www.tropicaltidbits.com/analysis/ocean/cdas-sflux_ssta7diff_global_1.png
Using my tried and true F10.7-TSI-SST model, which is based an intimate perfect working knowledge of the temporal relationship between these three measures
Only in religion and cults does one find ‘intimate perfect working knowledge’. Not in science…
The perfect knowledge of these relationships and principles is based on science, research, & work.
?dl=0
Cult? At least I’m not involved in the continual promotion of your cult of personality. 😉
You have no idea what I’ve done, so the attitude towards me is wholly unwarranted.
A few days ago you were trying to tell me there is no such F10.7-TSI-SST relationship. Instead you presented the rather false and pathetic formula as you did here today that you use to describe the sun-earth temperature relationship to support your theory that the sun only warms by 0.1C. The formula this mathematical theory is based on has no predictive power through a solar cycle like my work.
My work is powerfully predictive and had immediate application to my knowing the timing of the 2015-16 ENSO. Three years ago I said on an ENSO blog post here,
“…Climate change comes from solar changes. Solar activity ramped up late last year and has since tapered off. The “recharge” of the oceans from that rampup is now dissipating. If and only if there is another spike in solar activity this year will there be an El Nino.
I said that then because I knew at that time that all the ENSOs at the top of the solar cycle occured above the 120 sfu level AND are delayed due to the temporal relationship of F10.7cm to TSI.
The green arrow in the image below signifies the time when I first plotted and realized this relationship.
The rest of my model involved first smoothing the daily data and finding more confirmation of the temporal relationship, and creating a very nice low error TSI predictor based on the SWPC monthly SSN/F10.7cm forecast. I used the SWPC 2016 F10.7 forecast in late 2015 to forecast the Had3SST change over the year based on a second empirically derived regression formula of TSI-SST. I was less than 3% off.
The SC24 TSI rise & maximum drove the whole 0.6C 2008-2016 SST spike, and is now cooling us off.
The solar cycle influence for SC24 was 0.6C, not 0.1C. You’re way off Leif. My stuff works.
The SC24 solar cycle influence driving the 2009-10 and 2015-16 ENSO is very apparent.
Anthony,
I thought you would know this. The WMO defines climatological periods of reference ending in the last complete decade, while for comparison purposes also establishes a fixed reference period. Then it is up to research institutions to adhere to this standard or not.
http://www.wmo.int/pages/prog/wcp/ccl/guide/documents/Normals-Guide-to-Climate-190116_en.pdf
So now you know. Some are using the climatologial standard normal 1981-2010, that will be changed to 1991-2020 in less than 3 years, while others are using the fixed reference period 1961-1990. Both are doing it in accordance to WMO guidelines.
But since past temperatures keep changing, the references aren’t actually fixed.
“that will be changed to 1991-2020 in less than 3 years”
There are simple practical considerations that cause people to make different choices, which the WMO recognises here. GISS uses 1951-1980 because that was the most recent 30 year period when they started. And they now have a large base of published numerical data. If they changed, then whenever you saw GISS data you’d have to look up which anomaly they were using.
On the other hand, the anomaly base temperature is supposed to be your best estimator of present data. That ensures that you don’t have to worry about whether the sample for a given month includes the right balance of warm and cold places, because you have subtracted out the difference. If there has been significant drift since the base period, this works less well. That is why, when HADCRU (using 1961-90)T included more Arctic stations in V4, the anomaly (and trend) went up.
Then there is the issue that the anomaly base is first used for individual stations, so you have some work to do if they don’t have data in the period. That is another reason why 1961-90 is popular with people using GHCN; they can include more stations. It’s also why they don’t use more than 30 years. However, the issue is manageable – BEST uses the same least squares system I do, which doesn’t need a fixed period at station level. But you do need eventually to decide on a reference period.
So NOAA uses 1961-90 for its initial average calculation. Once the data has been aggregated, you can convert to any other base just by subtracting the average for that period. So NOAA converts to 20th century for a lot of reporting.
Satellite data, of course, has to use some period since 1979.
The 30 year periods of which you write were decided on back in the mid-1930s.
That was a time of printed materials, before computers (as we know them), and before the United Nations. Moreover, the issue of interest was more about meteorology, and less about using scary climate to justify “social justice” — or whatever it is that is going on.
As he who is about to go on vacation says “why can’t climate science do a simple thing like standardize on a baseline period ?”
But I would add, it doesn’t have to be 30 years.
I vote for 73, it is a nice prime number.
Javier,
Rather than a step up, as during the switch from 1971-2000 to 1981-2010, Warmunistas are liable to get a nasty surprise after the switch to 1991-2020. For that matter, if the rest of this year, 2018 and 2019 drop back under the present baseline, as after the last super El Nino, the dreaded Pause will be on again.
“to get a nasty surprise after the switch to 1991-2020”
No-one gets a surprise. It is an elementary calculation, which does not cause difficulty in the real world.
Nick,
Not sure you’re familiar with the real world, but the surprise I have in mind is that the nest 30 years are liable to be cooler than the past 30 years, as demonstrated by the past 300, 3000, 30,000, 300,000, 3 million, 30 million, 300 million and three billion years of climate history.
I am thinking we might need to pull out our thermometers again……….
http://suyts.files.wordpress.com/2013/02/image266.png
Mac
Is that graph in Farenheit?
And, yes, that presentation makes it look as though there is very little change, but sometimes even small changes can have important consequences. Going from -0.5C to 0.5C – just one degree – means that my ice lolly drops off the stick and turns to slush. Probably ruins my trousers. This is the sort of horror that Margaret Thatcher was warning us about, and why we have to hang everyone who breathes out.
So perhaps a rethink on the presentation.
This post should be deleted and rewritten. It should be argueing temps are increasing at similar rates looking 37 years apart. In particular, looking at 2 neutral El Nino years we have:
1980 +0.27°C warmer than the 1950-1980 baseline
2017 +0.26°C warmer compared to 1981-2010 baseline
And the later baseline was also warmer.
I hate to say it but that sounds like an argument that temperatures are continuing to increase with a near linear trend. That’s not what most WattsUpWithThat readers, including myself, were expecting 3 years ago.
I hate to say it but …
http://scienceandpublicpolicy.org/wp-content/uploads/2011/02/uncertainty_global_avg.pdf
So, yeah, THIS graph:
http://suyts.files.wordpress.com/2013/02/image266.png
The step up in baseline translates to about 0.5 degrees C per century, assuming the next 63 years gain as much as the past 37 years, which is highly unlikely.
Not the least bit scary.
Gabro,
Then, that’s what the article should say. At present its “false news”. Anthony definitely needs to take his vacation and do a better job of quality control when he gets back.
“I’ve added a caveat in the title to reflect this: (depending on who you ask)”
It doesn’t reflect it very well. There isn’t any data source saying that the temperature now is similar to 1980. If you don’t take account of base differences, you can mix them to get any number you like. GISS then with NCEP now, it’s a small difference. If you compare GISS now with NCEP then, the difference is huge.
Nick,
Only on Planet Stokes is a 0.2 degree C difference in 37 years huge.
Average UAH anomaly for first five months of 2017: +0.31 (thanks to high May figure)
Average UAH anomaly for first five months of 1980: +0.01
So I was off by a tenth of a degree. The anomaly gain from the first five months of 1980 to the same months of 2017 is 0.3 degrees C, not 0.2. That equates to 0.8 degrees per century.
http://www.nsstc.uah.edu/data/msu/v6.0/tlt/uahncdc_lt_6.0.txt
Still not very scary, half a degree more warming in the next 63 years. And the preliminary May figure might be revised downward.
“a 0.2 degree C difference in 37 years huge”
ERA-Interim, using a 1981-2010 base, says the temperature in 1980 was negative. GISS says the 2016 average was about 1°C. That shows boost you can get with base period fiddling.
Nick,
As you are well aware, 2016 was a super El Nino year.
If you use the same base, as I did above, for UAH, you get a gain of only 0.3 degree C for 2017 over the same part of 1980, for which year the average monthly anomaly was indeed very slightly negative.
Gabro,
You’re missing my point, which is the difference you can make using different base periods in a comparison. As Joe Bastardi points out below, a reasonable estimate of warming from 1980 to now is about 0.45C. As I point out above, a reasonable estimate of the difference between a 1951-80 and a 1981-2010 base is 0.42. So if you use the 1959 base for 1980, and the 1981 base for now, you get something like .45-.42 = not much change. But if you do it the other way, you get .45+..42=0.87C, which would be a large change in 37 years.
Nick,
I get your point.
Mine is that in UAH, 2017 has been only 0.3 degrees C warmer than the first five months of 1980. Hardly anything to get worked up about.
Of course you’re right that the 1950s, ’60s and ’70s were cooler, so you’d get a bigger difference with an earlier baseline. That is, until you go back to the 1920s, ’30s and ’40s.
If you compared the past 30 years with the 30 years around 1690, we would indeed be importantly warmer. But the same interval would be cooler than the 30 years from 1181 to 1210, or many other such intervals in the Medieval WP.
The warming since 1979 is trivial.
I think Anthony needs a holiday. Maybe someone in New Zealand can show him how to convert temperatures from one baseline to another. Ask a high school student, it’s part of the curriculum.
We are not talking about a single thermometer, we are talking about a homogenised average. It’s simple algebra. I note Anthony has now totally rewritten this article to cover his tracks 😉
Ditto.
See Mosher’s comment below about converting between absolute temperatures and anomalies. This is not rocket or climate science.
I much prefer to see the actual temperatures, rather than contemplate anomalies to the second decimal.
NOAA reported: “The combined global average temperature over the land and ocean surfaces for April 2017 was 0.90°C (1.62°F) above the 20th century average of 13.7°C (56.7°F)—the second highest April temperature since global records began in 1880, trailing 2016 by 0.17°C (0.31°F) and ahead of 2010 by 0.0.7°C (0.13°F).”
Mmm, so approaching a searing 15°C, boy the world is steaming /sarc.
And for May, NOAA reported: “Averaged as a whole, the global land and ocean temperature for May 2017 was 0.83°C (1.49°F) above the 20th century average of 14.8°C (58.6°F) and the third highest May in the 138-year global records, behind 2016 (+0.89°C / +1.60°F) and 2015 (+0.86°C / +1.55°F). ”
But to keep up their hottest “evar” meme, they also noted: ‘May 2017 was characterized by warmer- to much-warmer-than-average conditions across most of the world’s land and ocean surfaces. However, near- to cooler-than-average conditions were present across the eastern half of the contiguous U.S., eastern Europe, western and north-central Russia, as well as parts of the northern and southern Atlantic Ocean, northern and southern Pacific Ocean, and the tropical Indian Ocean.”
“The global land-only surface temperature was the coolest May land temperature since 2011 and the seventh highest since global records began in 1880 at 1.15°C (2.07°F) above the 20th century average 11.1°C (52.0°F).” Wow, the land temperature averaged only 12°C in May, no wonder I have the heater on.
“Wow, the land temperature averaged only 12°C in May, no wonder I have the heater on.”
That is why it is very foolish of NOAA to quote such an average in these reports (they explain here, S 7). Not only is it almost impossible to measure properly, but it is meaningless. Most places were nowhere near 12°C in May. But if you say that the average anomaly was 1° (it wasn’t, in May), then there is a reasonable chance that it was warmer than usual where you are, whatever usual is.
Buuuu …buuu …. buuuu …. buuuuuut … I just read in my MSM feed …. Seaaaaaaaaaa levelllllllllllllll ….. FIFFFFFFFFFtyyyyyy …. perCENNNNNNT …. FAAAAASterrrrrrrrrr ………
…. OVErrrrrrrrrrr … the .,… PAST …. TooooooWENNNNNty YEEEEEAAAaaaarrrrrrrrrrrrrrs!!!!
Extra … Extra …. read all about it!
Gavin says that he can accurately and precisely take Earth’s surface temperature with 50 stations. If they all have continuous records since 1880, stations uniformly maintained during that time, with no switch to electronic thermometers, and cover the land surface uniformly, to include elevation differences and are all in areas which have been rural all that time, away from pavement, then, yes, maybe, theoretically, but for the land only. But there are few if any such sites.
Antarctica’s fringes only started getting measured continuously in the 20th century, and at the South Pole only since 1957, IIRC. There has been no warming at the SP, which is precisely where it should be most evident, according to AGW theory.
Oops, wrong Paul in previous comment. Anyway, The woodfortrees blog has a lengthy discussion of baselines, including BEST’s, and shows all the different data set trends baseline shifted to the UAH baseline.
http://www.woodfortrees.org/notes
THATS ABSURD We are no where near 1980 temps
http://www.drroyspencer.com/wp-content/uploads/UAH_LT_1979_thru_May_2017_v6-1.jpg
Joe,
Anth@ny has clarified that the point he was making was precisely about the problems with using T anomalies from different baseline periods. He maybe should have made the point more explicitly rather than relying upon sub rosa satire.
PS: Some would say that 0.3 degrees C warmer for the first five months of 2017 over 1980 is actually fairly near.
We’ve had a cool summer. The pool is staying at 82 degrees. It is definitely cooler than last year which was a constant 86 about this time.
Simply no place left globally(the oceans) for warmth to come from.
Few of we Meteorologist put any real credence in a trace gas(vital to life).
We calculate absolute temperature.
You can choose any base peroid you like.
For display purposes and consistency with the most widely used series (hadcrut) we use 1951-80.
U can play with periods to your hearts content if you first do the real series in absolute as we do.
It’s not an issue worth discussing.
Waste of time.
Not scientifically relevant.
…You use the same time period (51 – 80) when “scientists” like yourself were claiming the next “Ice Age” was just around the the corner…..Things that make you go…hmmmmmm……
If u use the baselines of different time periods, you get nonsense. We should be more patient with the climate to change. Climate is slow, like evolution. Its no use to watch evolution from one month to the other.
Except that new species do spring into existence in a single generation.
New climates, not so much.
Well there certainly has been lots of squawking, both here and elsewhere. Some the elsewhere squawking was downright mean, but that’s OK, it’s part of the rigid mindset those people have, they fear change, and they fear different ways of looking at things.
Joe Bastardi and Leif Svalgaard (among others) point out that the comparison is ridiculous, and that is indeed the point; It is. But how is the public supposed to be able to interpret these differing surface temperature presentations done on different baselines? The answer is: they can’t, unless there is a standardized baseline.
All the squawking elsewhere has shown me that there’s really no interest in the climate science community in coming up with a standardized baseline for public surface temperature presentations, they’d rather defend their own work and declare “Watts is an idiot for saying so”. It’s pretty typical, and exactly what I expected. They don’t like change, they really don’t like anyone else suggesting that they way they present temperature data might not be in the best public interest, because after all, they are saviors of the planet and who are you to question us.
As Steve Mosher likes to say: Too Funny!
On the plus side, I’ve been given a marvelous gift from all this squawking. Watch this space after I return. For now I’m closing the thread, as I’m heading out, but there will be a new post sometime after I return on this very topic.