UAH Global Temperature Update for May 2012: +0.29°C
By Dr. Roy Spencer
The global average lower tropospheric temperature anomaly for May 2012 (+0.29 °C) changed little from April (+0.30 °C), with some warming to near-average temperatures in the tropics being balanced by a little cooling in the Southern Hemisphere (click on the image for the full-size version):
The 4th order polynomial fit to the data (courtesy of Excel) is for entertainment purposes only, and should not be construed as having any predictive value whatsoever.
Here are the monthly stats:
YR MON GLOBAL NH SH TROPICS
2011 01 -0.010 -0.055 +0.036 -0.372
2011 02 -0.020 -0.042 +0.002 -0.348
2011 03 -0.101 -0.073 -0.128 -0.342
2011 04 +0.117 +0.195 +0.039 -0.229
2011 05 +0.133 +0.145 +0.121 -0.043
2011 06 +0.315 +0.379 +0.250 +0.233
2011 07 +0.374 +0.344 +0.404 +0.204
2011 08 +0.327 +0.321 +0.332 +0.155
2011 09 +0.289 +0.304 +0.274 +0.178
2011 10 +0.116 +0.169 +0.062 -0.054
2011 11 +0.123 +0.075 +0.170 +0.024
2011 12 +0.126 +0.197 +0.055 +0.041
2012 1 -0.089 -0.058 -0.120 -0.137
2012 2 -0.111 -0.014 -0.209 -0.276
2012 3 +0.111 +0.129 +0.094 -0.106
2012 4 +0.299 +0.413 +0.185 -0.117
2012 5 +0.289 +0.439 +0.139 +0.033
As a reminder, the most common reason for large month-to-month swings in global average temperature is small fluctuations in the rate of convective overturning of the troposphere, discussed here.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
phlogiston says: “Does this indicate that the Peruvian upwelling might be pulsatile?”
Sorry, phlogiston, I can’t answer your question.
phlogiston says: “Does the persistently very cold north Pacific (according to Ansys) have any significance?”
The North Pacific north of 20N has been cooling since 2005. If you’re looking at anomalies, there’s a large seasonal component to them in the North Pacific, and they’re approaching the annual bottom of that seasonal cycle.
Don’t forget there are two volcanoes (April 1982 and June 1991) which affected the trend over the 2.5 years following by up to -0.3C to -0.5C.
Put those impacts back in, and there is less warming overall and the trend moves toward linear. Pull out the ENSO, AMO, solar influences as well as the volcanoes and it is a linear warming trend of 0.047C per decade.
Why won’t you try linear fit? Oh yeah, that would most likely have some predictive value, althought not the kind you like… (and I base that assumption on the measurements before the start of that graph)
By the way, the more you have free parameters in your fit, the better the fit with ANY data with intrisic variation.
Bill Illis says
Put those impacts back in, and there is less warming overall and the trend moves toward linear. Pull out the ENSO, AMO, solar influences as well as the volcanoes and it is a linear warming trend of 0.047C per decade.
Henry@Bill
Bill, I can give you better than that. And I don’t trust the UAH values anymore.
(precision? accuracy? error?)
I can give you the development of warming/cooling over time.
And it shows that earth has been cooling in the last decade.
See here:
http://www.letterdash.com/henryp/global-cooling-is-here
Note that the sample is well balanced by latitude and 70/30 sea – inland
Personally I think the climate is on this curve:
http://wattsupwiththat.files.wordpress.com/2010/04/orssengo3.png
Study this curve carefully and you will see that around 1994 temps went down (negative/decline) as correctly predicted by me. However, Orssengo thought the max would be at around 2000. That means his curve must be shifted a bit to the left.
Is Orssengo still alive?
Overall, we cooled about 0.2 degrees K since 2000.
That does not yet look like a lot. In fact, showing the difference between the walls inside my house of 0.5 degrees C, my son laughed at me worrying about the 0.2 degree cooling. The problem is that the cooling is not linear. It is on a curve and it is headed further down.
Roy, I see a number of “double” tops and bottoms in the graph seemingly marking a change in direction with a significant swing following.
To laugh at an epileptic seizure is uncouth, but it is all right to be entertained by a polynomial fit.
DMarshall says:
June 4, 2012 at 7:16 pm
Even if the global temp is stable, we’re not done with weird weather for this half of the year.
As a counterpoint to the cold and snow in Sweden and the (well, hardly uncommon) rainy pall on the Royal Diamond Jubilee,
The crowning of the Queen on the 2nd June 1953 saw rain in London with the maximum temperature 12°C.
Yesterday was a wet day in London with the maximum temperature 12°C.
Kelvin Vaughan says
Yesterday was a wet day in London with the maximum temperature 12°C.
(same as the same date in 1953)
Henry says
True.
But do not forget that in the meantime we have had at least 30 years of warming
followed by 18 years of cooling
http://www.letterdash.com/henryp/global-cooling-is-here
My question is:
in which direction do you think we are heading now?
for entertainment purposes only, no value whatsoever…
So why include it at all? It’s misleading, dishonest, and discredits yourself.
[REPLY: Your first comment here and this is your contribution? -REP
“UAH Global Temperature Update for May 2012: +0.29°C
By Dr. Roy Spencer
The global average lower tropospheric temperature anomaly for May 2012 (+0.29 °C) changed little from April (+0.30 °C), with some warming to near-average temperatures in the tropics being balanced by a little cooling in the Southern Hemisphere “
Seems that this stop in increase is in line with the solar tide functions:
http://www.volker-doormann.org/images/uah_temp_4_rs_.gif
V.
Thank you for the reply. Here’s my second comment here:
Anyone who understands trends better than “let’s see which options in Excel get the curve I want” knows that this trend line has LESS than no value whatsoever. It is only useful for those who don’t understand, but are easily swayed by false impressions. Say, that gives me an idea, why don’t you include the valueless “entertainment” trend line graphic in this site’s title banner? That way you can show that the site is about what the “entertainment” says, not what the data say.
Samwell says
That way you can show that the site is about what the “entertainment” says, not what the data say.
Henry says
Why don’t you study some stats (at least first year – it is easy!)
before you start criticizing others because of their use of “excel”
Excel just calculates what you tell it to it do –
do you know what you are doing?
http://www.letterdash.com/henryp/global-cooling-is-here
HenryP says:
“Excel just calculates what you tell it to it do”
So why would you tell it to plot a 4th order polynomial fit? What makes one think that a polynomial fit, specifically a 4th order polynomial, is representative of the actual data? Why not restrain from using *any* fit that requires a “this has no value” disclaimer?
Henry@Samwell
When you do trending in excel, it normally also calculates a correlation coefficient when you ask for it. The nearer that value is to one, the more you know for sure that the trend you find is not random.
In my case the r2 for maxima was 0.997 which is near perfect. For my means it was 0.95. Still not too bad. See:
http://www.letterdash.com/henryp/global-cooling-is-here
I gather that the r2 of the polynominal fit tried here is probably 0.5 or less. So they know that the fit, even though it looks good, is probably random.
In comparing the UAH results to mine you can clearly see that there must be something wrong with them,
or mine, of course.
With the maths on the table I very much doubt that my values are wrong.
Well you seem to know what you’re doing, but I’m not sure that Roy does. If you gather that the fit is probably random, do you think it is because Roy is unable to get a better non-random fit, or because he’s purposefully trying to misrepresent the data?
Samwell says:but I’m not sure that Roy does
I have not seen any comment from Roy here so I doubt if he has even looked in on the comments here.
Otherwise see here:
http://wattsupwiththat.com/2012/06/09/nature-record-heat-tipping-points-hansens-opinion-on-weather-noise-and-all-that/#comment-1005665
I hope my point is made though that the “fit” line is less than no value whatsoever. As you say it is “probably random” and that “they know”. Imagine if they overlaid a plot of some truly random data; it would serve only to distract from the real information on the image. Supposing they know it — never assume stupidity when malice is an adequate explanation and vice versa! — then the goal is not to “entertain” but to mislead. Clearly they’re trying to suggest a cyclical trend. Too bad a 4th order polynomial fit isn’t cyclical! I suppose excel doesn’t have a cyclical fit option?
You predict decreased temperatures over the next few years. I think there will be a hot el nino year in the next few, and it will be a shocking anomaly for skeptics, and then over the following decade the average will become more like the anomaly… just like what happened last time. If it turns out you’re right I’ll remember that you predicted it.
Samwell says:
Anyone who understands trends better than “let’s see which options in Excel get the curve I want” knows that this trend line has LESS than no value whatsoever. It is only useful for those who don’t
understand, but are easily swayed by false impressions.
Can you prove it? How?
Say, that gives me an idea, why don’t you include the valueless “entertainment” trend line graphic in this site’s title banner? That way you can show that the site is about what the “entertainment” says, not what the data say.
Well, in some cases of multi-frequent curves a cubic fit can show low frequency trends.
There is a low frequency decreasing trend since 2009 in the UAH global temperature.
Just for entertainment: Summing up with Excel 11 solar tide functions –
http://www.volker-doormann.org/images/uah_temp_gl_ghi11.gif
– it correlates strong with the UAH global temperature.
BTW. It is senseless to argue on i.) nothing, ii.) no value, iii.) less than no value.
V.
Volker Doorman says:
http://wattsupwiththat.com/2012/06/04/may-uah-global-temperature-unchanged/#comment-1008536
Volker, you should read what the argument was about.
Otherwise, your BTW was spot on:
don’t argue with anyone, if you, yourself, have no figures to show.
I hope you are watching the soccer tonight
it is Holland (me) against Germany (you)
HenryP says:
June 13, 2012 at 11:27 am
Volker, you should read what the argument was about.
Otherwise, your BTW was spot on:
don’t argue with anyone, if you, yourself, have no figures to show.
After the monthly updates of Dr. Roy Spencer ‘s UAH temperatures, I do show my same figure of a simulation of the global temperatures out of real astronomic synodic functions, with no comment or question.
But you know ‘a prophet has no value in his own country’.
I hope you are watching the soccer tonight. it is Holland (me) against Germany (you)
That are all nice people, who bring real fun to the world.
I was 38 years doing physics with Philips research lab.
V.
Henry@Volker
The problem is that UAH shows warming of ca. 0.2 degree K since 2000 whereas my results show that it has cooled by 0.2 degree K since 2000.
UAH is checking the tropics only
I have balanced my sample by latitude and 70/30 sea – inland,
so I would say that my results are more representative.
I also query the accuracy and precision of UAH.
The cooling is not a lot yet, but my maths shows that it will get worse……
see here:
http://www.letterdash.com/henryp/global-cooling-is-here
check my answers to the questions as well.
Imagine, I worked almost 25 years for Siemens in South Africa as a laboratory manager /chemist.
Die Material Pruefstelle…
Germany was better than Holland, (soccer), I think they deserved to win.
Looks like Holland is out of the Euro cup…..
HenryP says:
Henry@Volker
The problem is that UAH shows warming of ca. 0.2 degree K since 2000 whereas my results show that it has cooled by 0.2 degree K since 2000.
UAH is checking the tropics only
I have balanced my sample by latitude and 70/30 sea – inland,
so I would say that my results are more representative.
I also query the accuracy and precision of UAH.
The cooling is not a lot yet, but my maths shows that it will get worse……
see here:
http://www.letterdash.com/henryp/global-cooling-is-here
check my answers to the questions as well.
@HenryP
It may be that you are right. I do not know.
I have taken the monthly global UAH data and have analysed the frequencies of the temperature spectra. Additional I have analysed the main frequency of the sea level oscillation (seasonal data removed) by subtracting the linear increase function. I have shown that the main frequency oft the sea level oscillation have the same frequency as one of the frequencies of the UAH temperature spectra, but a better S/N ratio as the temperature spectra.
Mostly all of the analysed frequencies can be connected to twice the synodic frequencies of planet couples like Mercury/Earth. By fitting empirical the magnitude of all (11) synodic functions with the help of the UAH data, it does result in a function, which correlates with the global UAH temperature function.
http://www.volker-doormann.org/images/uah_temp_gl_ghi11.gif
But moreover, because the synodic functions are well known from astronomic data, the extend of the present data can be calculated to the future, with the help of other reconstructed temperatures like A. Moberg et. al. of 2 ky back in time, in general for the next 1000 years.
The idea to analyse more the frequencies instead of the amplitude is very different to your investigations, and as one could notice here over several month, the monthly updates by Dr. Roy Spencer fit well the astronomical functions.
Do you can extend your temperature function to the future?
Imagine, I worked almost 25 years for Siemens in South Africa as a laboratory manager /chemist.
Die Material Pruefstelle…
Fine to hear. Experts in techniques have a fundament in science.
Some people not really:
http://www.volker-doormann.org/images/banner11.jpg
V.
Volker Doorman says:
Do you extend your temperature function to the future?
Henry says
If we plot the global measurements for the change in Maxima: 0.037, 0.029, 0.015, -0.015 against the relevant time periods, i.e. 37, 32, 22, 12 years respectively, it can be shown that the best fit for the curve is given by
y= 0.0455 ln(x)-0.1273 (r2=correlation coefficient= 0.997).
That is is a very good correlation. If we take that to the present, it shows Maxima are dropping now at a rate of 0.2 degrees per annum. The good news is that that rate is apparently also the maximum rate of that curve – so it stays there for some time, dropping at the same rate, but I cannot say yet until when. I have no idea when it will change signal again (to positive = warming)
If you try to put the means values 0.015, 0.013, 0.014, and -0.019 into a formula, against the relevant time periods, you get
-0.0001xsquare+0.0067x-0.0824
with a correlation coefficient of 0.95.
Not so bad either, but I would be a little less confident making predictions on that.
If you take that to the present it shows -0.08.
If you take that further to the future, say to -14 on my scale (2025), you find a value of -0.2 degrees K. Scary stuff. Let us hope I made a mistake?
Henry@Volker
Henry says:
I have no idea when it will change signal again (to positive = warming)
Henry says
Truly, I do hope that the climate is on this curve:
http://wattsupwiththat.files.wordpress.com/2010/04/orssengo3.png
Study this curve carefully and you will see that around 1994 temps went down (negative/decline) as correctly predicted by me here whereas the green line from the IPCC still wants us to believe that it goes the other way (positive/incline). If the Orssengo curve is correct we will drop a total of about 0.3 or 0.4 degrees C before things turn up again beyond 2030.
\However, Orssengo saw the max. of warming at around 2000, I see it at 1994/5
That means the Orssengo graph must be moved a little to the left?
Henry@Anyone here:
Is Orssengo still alive?>
Henry says
If you take that further to the future, say to -14 on my scale (2025), you find a value of -0.2 degrees K. Scary stuff. Let us hope I made a mistake?
Henry says
I’ve looked at that again, carefully.
It seems likely I did make a mistake./\
we should put it there also on a hyperbolic curve
namely,
-0.0001xsquare + 0.0046x-0.0624
it gives me a correlation coefficient of 0.995
which is amazing.
if we take that to 2025, it shows me earth cooling by about 0.12 degrees K per annum at that time.
Still a bit scary, but the scale of cooling is a bit less now.
All we can do now is hope and pray that Orssengo is right, meaning that earth will start warming up again, beyond 2030.
Is Orssengo still alive?
Is he an alias for somebody else?