July 1912 GISS Anomaly (-0.47) Was Warmer Than January 2007 GISS Anomaly (+0.94) – (Now Includes February Data)

Excellent in pointing out the weakness of using anomalies both scientificly and in the sense of how they exagerate temperature changes over time. Using anomalies is similar to graphically fudging the x or y coordinate scales to make changes look more or less extreme. If people were regularly exposed to actual temperature graphs, over time they would be much more cognizant of reality and less likely to buy the coolaide being served to them.

Eliza says:
April 3, 2014 at 9:35 am
“If earth was closer to sun in July we would certainly have a problem as 9/10 of land is in the NH!!!. It is precisely that is farther away in july that possibly life is possible on earth.”
Life is much too hardy to be impressed by that, even though the Green Mob always tries to make it sound like it would perish the next moment.
Caveat: once created. Which is still a mystery for science.

Col Mosby says:
April 3, 2014 at 10:08 am
Besides, the Earth’s orbit appears almost circular
True. For the exact numbers, see: http://spaceweather.com/glossary/aphelion.html
“In January when we’re closest to the Sun (perihelion), the distance is 147.5 million km. In July we’re 152.6 million km away–a five million kilometer difference.”

There are even greater anomalies when you figure in that the temperature isn’t exactly the same everywhere on earth. For a blackbody
Watts/square meter = (5.67*/ 10^8) * T(K)^4
Plug in an average temperature of 15C = 288K , and you get
390.08 watts/square meter for a black body radiating at 15 C.
Suppose half the earth is at 25 C, and half at 5 C (Still a 15C average),
and for a blackbody you get half the earth radiating at 447.14 watts, half radiating at
338.66 watts, for an average of 392.9 watts, increased radiation of 2.82 watts with NO increase in average temperature anomaly. And with a DOUBLING of CO2 you’re looking at an increase of only 3.8 watts. Looking at anomalies is meaningless without knowing changes in VARIANCE.
To give credit where credit is due, I picked this up reading Lubos Motl’s website.

Alan McIntire says:
April 3, 2014 at 11:25 am
Looking at anomalies is meaningless without knowing changes in VARIANCE.
True enough. As well, when most of the warming is in the northern Arctic in the winter, then it is obvious that there is no 1:1 correlation with anomaly increase and energy increase.

rgbatduke says:
April 3, 2014 at 11:33 am
“The next graph shows just the lines to illustrate the above. Think of it as a sideways bar graph where the lengths of the lines indicate the relative times where the slope is 0. In addition, the sloped wiggly line shows how CO2 has increased over this period.”
No, it doesn’t. It is really difficult to convey this properly, impossible on the same scale
I will be the first to admit that my computer skills leave lots to be desired, so I have to use the tools I have. As well, I feel if it is good enough for Lord Monckton, then it is certainly good enough for me!
It was as a result of your comment several months ago that I revised my comment about the significance of the CO2 line. And I thank you for it. I obviously did not hit the nail on the head as I thought I did.
I said: “In addition, the sloped wiggly line shows how CO2 has increased over this period.”
You said: “No, it doesn’t.”
Would it improve things if I changed “how” to “that” so it reads:
“In addition, the sloped wiggly line shows
that CO2 has increased over this period.”

Nick Stokes says:
April 3, 2014 at 2:28 pm
Thank you!
Your “active plot” looks cool! Can it be used to show RSS from September 1996 to February 2014 and also show the slope is flat for this period of 17 years and 6 months? Thanks!

Nick Stokes says:
April 3, 2014 at 4:08 pm
Thank you. I guess at this point I need to wait for the new version to be ready. I basically need something like WFT, but that additionally allows both right and left y axis to be adjusted. In addition, it needs to be updated as soon as the data is released. WFT for example updates automatically at 8:00 PM mountain time any new data that appeared in the past 24 hours. Please let me know when you have what I need. Thanks!

“””””…..Steven Mosher says:
April 3, 2014 at 10:50 am
“If you use a thermometer, it will give you a temperature. It will not give an anomaly.”
not really.
For example. bulb thermometer actually gives you a measure of the expansion of a liquid……”””””
Well that depends on where you buy your thermometers. If you buy good mercury in glass thermometers from a reputable industry standard supply house for such lab equipment, you can get thermometers that are individually calibrated. That is, the markings on the thermometer are individually applied at the Temperature they are reading. I have a very good -10 – +50 deg. C thermometer, that was made that way (got it for doing color film processing.)
Such calibration accounts for any variation in the capillary diameter along the scale, and also for any non uniformity of temperature along the scale. The calibration is specified for an immersion length (in water) that as I recall is 76 mm, which is about three inches, and is measured from the bottom of the mercury bulb. The higher the temperature, the longer the mercury column extends, and the more the extended part cools, and all that is allowed for in the marking of the scale on the glass. As I recall, this one is marked in 0.1 deg. C increments, but maybe it is 0.2, I’ll have to go look at it.
But I agree with your general premise; you can’t be sure just what it is whose temperature you are assuming corresponds to the reading..
My thermometer, would be no good for reading the air Temperature, because the immersion length would be wrong.

Reblogged this on The GOLDEN RULE and commented:
This is not for the non-academic readers but included as additional evidence that claims about trends in the so-called average global temperature are best left to the mathematical experts.
However, two things are clear –
It is virtually impossible to compute a meaningful global average temperature.
It is impossible to define a temperature trend that can be agreed to by a genuine spread of experts. Only those with a barrow to push can expect the world to accept a particular unsupported conclusion and unfortunately, much of the world does. This is not science, as it should be!
WUWT, with its many expert submissions, shows this quite clearly.
Then to use a conclusion that is impossible to ratify to create major global financial and political upheavals is, again, clearly unjustified.
So, dear readers, please think carefully about the science aspect when listening to the political jargon.
Not just regarding temperatures but also about the incredible claims that are being propagated about extreme weather events and the need for carbon controls.
The risk of unwittingly becoming a willing participant in an extensive (and expensive) deception is real.

“Richard Barraclough says:
April 4, 2014 at 3:19 am
Werner,
Your final chart shows the fallacy of trying to read too much into fluctuations from month to month. You adjusted all the anomalies to the same level in January 2013. By November there is a difference of about 0.5 deg C between UAH and GISS, with the other datasets scattered in between. Perhaps one of them is correct – and perhaps none of them?”
GiVEN a period of global warming due to changes in greenhouse gases, the COLDER months’ temperatures will increase faster than warmer months’ temperatures.
Likewise, with a period of global COOLING, the warmer months’ temperatures will cool faster than warmer months’ temperatures. So the answer to your question is not” perhaps none of them” ,
but ” temperature anomalies for warmer and cooler months AUTOMATICALLY change at different rates thanks to that 4th power black body rule and the fact that increased global warming tends to
BALANCE temperatures between colder and warmer regions- there IS no scale for which anomalies will change at the same rate for all seasons.

Would it improve things if I changed “how” to “that” so it reads:
“In addition, the sloped wiggly line shows
that CO2 has increased over this period.”
But this information is not useful. The stock market increased over this period too.
If you are going to use the W4T interface, I’d strongly suggest just building two figures (or better yet, three figures, as it is always good to blow up the “temperature anomaly” right after showing people the actual absolute temperature to scale). Sadly, W4T kinda sucks for building arbitrary graphs (or else, its interface is non-intuitive and is missing things I’d like to be able to do such as define my own axis limits, apply filters that are better than the simple ones they provide, fit curves, generate Kolmogorov-Smirnov or Pearson statistics and so on). R is much better suited for doing real graphs with real statistical tools (and is free and pretty easy to learn — lots of books out there to help you as well as ample online resources and examples).
Or, if you want to do it in text why not simply state what I just stated: The total temperature change from 1950 to 2014 of roughly 0.5 C represents a variation of 0.17% of the absolute temperature (that is “seventeen hundredths of one percent”) which is itself only known within about 1 C. The total CO_2 concentration change from 1950 to 2014 of roughly 100 ppm represents a variation of 33% (that is “one third”) of the initial absolute concentration.
You can never go wrong stating simple fact.

New RSS data now shows a 212 month cooling trend. This is more than half of the entire record (423 months).

The Section 1 analysis for slopes that are “at least very slightly negative” does not make much sense to me. First, all the slopes that turn negative turn positive again. When they do turn positive again, they peak at higher positive values than the lowest negative values reached (absolute values, of course). They tend to be trendless (slope=0) for significant periods around the period they turn negative.
UAH is probably the worse case example. It’s slope remains near zero for four to five years. It turns negative October 2004. However, it is only negative for four months (Oct 2004, Dec 2004, Jan 2005, Mar 2005). These four months have an average slope of -0.00019. This is less than 1% of the positive slope of 0.020 reached later in Nov 2007. After turning positive in Apr 2005, UAH doesn’t turn negative until over three years later (June 2008).
When the slopes of these data series are graphed, they all share something in common. The positives slopes have a peak in 2007 and the negative slopes have a peak in 2009.
A better method for analyzing the slopes might be to look at where the slopes peak rather than where they cross zero. Also, might want to look at where the R^2 values peak. An indication of robustness would be if all data series peak at the same time (rather than varying from 1996 to 2004). A good indication of robustness is if the same dates can be obtained by going in the opposite direction. In the backward direction (as in this blog), we start with the PRESENT date and go to the furthest month in the PAST where the slope peaks (NEGATIVE). In the forward direction, we would start with a PAST date (Jan 1976) and go to the furthest month in the FUTURE where the slope peaks (POSITIVE).
What we find in all the data series is that positive peaks occur in 2007 both forward and backward. In the forward analysis we see a dip after 2007 decreasing to a trough in 2009. These peaks and troughs are at the same time as in the backward analysis. Also, peaks in R^2 occur around these same 2007 and 2009 points.
It should also be noted that the backward analysis in this time period is more volatile than the forward analysis.
What’s interesting is if we compare all the previous positive and negative cycles going back to the 19th century. Here’s what we find:
cooling trend: 1878-1910 (32 yrs), slope = -0.0070
warming trend: 1910-1943 (33 yrs), slope = 0.0142
cooling trend: 1943-1976 (33 yrs), slope = -0.0028
warming trend: 1976-2008 (32 yrs), slope = 0.0193
[Note: 2008 cutoff could be in range from 2007 to 2009.]
Looks like an average cycle of approximately 32.5 years. If these cycles continue, we should be in a cooling trend that lasts to approximately 2040. How much cooling? If the cycles continue, it might be less than 0.1 degrees Celsius (-0.003*32).

In reply to “Werner Brozek says: April 5, 2014 at 6:11 pm”
Yes, I know what you did. There was no need to repeat a description of it. I made no claims that you did not know that there were positive slopes after your dates or noise producing positive slopes in 2014 (in other words, I knew that the slope ALWAYS is positive before your calculated date; and I assumed you knew it too.). I did not see the point in your analysis because it did not provide any useful information in my opinion. Rather than just saying you were “not interested,” It would have been helpful if you explained why. That’s just the inquisitive scientist in me.
If I was to guess why, I believe you were attempting to prove that this is all chaos and it should all be taken with a grain of salt. Temperature (not just anomalies) is a meaningless property because it does not measure a change in energy. Average temperatures are meaningless because some places on Earth can vary by 60 degrees C or more. Temperature anomalies are meaningless because what is “normal.” And look at these slopes of the various temperature anomaly time series, one turns positive in 1996 and another in 2004, so those are meaningless too. And on and on. [Note: I know you did not use the word meaningless, but that seemed to be the message.]
As you said, “it is hard to imagine that an increase of 2 C over 100 years, even if it occurred, would be catastrophic.” It would have been a better article if you would have just explained why sea levels will not rise significantly, agriculture will not shift across boarders, extreme storms will not be more frequent, etc. Or why that won’t matter. Instead you disguise politics as science by attempting to bring a sense of confusion to where there should not be confusion. Concepts like 3.8 degrees C seasonality are simple fundamentals that should not be a source of contention.
Obviously I am not a “warmist” because I just made a case in my last comment why there might be global cooling until 2040. I came to wattsupwiththat.com because through word-of-mouth I heard there was excellent science here.