Is Spencer Hiding the Increase? We Report, You Decide
by Roy W. Spencer, Ph. D.
One of the great things about the internet is people can post anything they want, no matter how stupid, and lots of people who are incapable of critical thought will simply accept it.
I’m getting emails from people who have read blog postings accusing me of “hiding the increase” in global temperatures when I posted our most recent (Dec. 2009) global temperature update. In addition to the usual monthly temperature anomalies on the graph, for many months I have also been plotting a smoothed version, with a running 13 month average. The purpose of such smoothing is to better reveal longer-term variations, which is how “global warming” is manifested.
But on the latest update, I switched from 13 months to a running 25 month average instead. It is this last change which has led to accusations that I am hiding the increase in global temperatures. Well, here’s a plot with both running averages in addition to the monthly data. I’ll let you decide whether I have been hiding anything:
Note how the new 25-month smoother minimizes the warm 1998 temperature spike, which is the main reason why I switched to the longer averaging time. If anything, this ‘hides the decline’ since 1998…something I feared I would be accused of for sure after I posted the December update.
But just the opposite has happened, with accusations I have hidden the increase. Go figure.
Dr A Burns (15:18:59) :
“What does it matter whether there’s warming or cooling ? The issue is whether man can influence it any more than King Canute was able to influence the tides.”
This indeed the crux of the matter. The warmists with all their billions of research money appear to have avoided this problem like the plague.
Its just so much easier for them to write a program and manipulate some data. And since most folks think these programs are capable of foretelling the future, why bother doing any basic research, when that runs the risk of not supporting their story.
Wondering Aloud (18:15:06) :
“……It looks like a .3 C step change in 1998 and nothing else. I wonder why there is a step change there? Are we certain we don’t have a sensor or program reason for the step change?…..”
That step change….
http://www.woodfortrees.org/plot/uah/plot/uah/from:1979/to:1997/trend/plot/uah/from:2000/trend
….. is not compatible with CO2 increasing at a continuous, uniform rate of about 15 ppm/decade since 1979 being the overwhelming cause of the temperature increase 1979-2010.
From page 8: ‘Who is Maurice Strong?’ National Review Sept 1, 1997 by Ronald Bailey (worth reading for an insight into the institutional origins of it all),
“…..Furthermore, more accurate satellite measurements show no increase in the average global temperature over the last two decades….”.
http://findarticles.com/p/articles/mi_m1282/is_n16_v49/ai_19722906/?tag=content;col1
It was that following El Niño year, together with Mann’s ‘hockey stick’, which gave the AGW hypothesis the necessary boost to last this long.
nanuuq (20:34:25) :
like dude, i was just askin
whats causing it to get warmer?
i just want to know
Well, a degree is not exactly a perceptible difference. So, we trust the thermometers, but given the analysis done showing that high latitudes, where it is colder, are dropping out of the record, and being filled in with jungles (see chiefio.wordpress.com – The Bolivia Effect), I would say that whether it is actually warmer or not is not exactly a settled matter. I mean, thinking about the amount of work that goes into averaging the temperature of the Earth, and the amount of any possible errors resulting from improper sampling, you know, like, it is totally not farfetched that warming is a local event to some areas.
But, dude, even if it were, that doesn’t mean the lack of explanation means the only explanation is CO2. In fact, it seems the natural variations of climate are very poorly understood. If you wish, ask someone how it is possible for the Earth to plummet into glacial ages lasting 100,000 years with CO2 levels similar to what they are now. Or how Earth warms up before CO2 starts outgassing from the ocean (and why they expect that effect would disappear after CO2 levels rise).
nanuuq (18:04:35) : | Reply w/ Link
“So like the above graph shows that there has been global warming.
You dudes here are totally hell bent against any possiblity that humans have in any way contributed to this.”
Wrong impression. A lot of things humans do change climate. Cultivation, forest fires, turning into asphalt and cement swaths of land, changing rivers, damning staits etc etc. and way downthe list, a small contribution from human induced CO2.
SO
what has caused this warming? eh?
can you provide quantitative analysis that show where this warming has come from?
It is not an easy problem, but we are not the ones who are trying to stampede the western world to do hara kiri, and the AGW CO2 camp which should provide the solid science, has only handwaving arguments from faulty computer outputs that have been already falsified. That is the problem.
all I see from you dudes are comments like “oh its natural” but to me thats like telling me that God exists. ie a viewpoint biased by preexisting prejudices.
I admit I have a preexisting prejudice that the sun exists and that the earth with its atmosphere turning around it gets what is described as weather and climate.
have any of you dudes calculated how many joules of energy have been inserted into the biosphere by the burning of “buried sunshine”? HM? like maybe that can cause some heat eh?
Have you? Lets see your numbers.
People are measuring it and part of it is called “the urban heat island effect.
nanuuq (20:34:25) :
like dude, i was just askin
whats causing it to get warmer?
i just want to know
Well, Kevin Trenberth “just wants to know” much more than you do. The great ipcc Climate Scientist wants to know why it isn’t getting warmer. He’s called the divergence of “global mean” temp. from CO2 concentrations a “travesty”, and is quite upset about it.
So, we’re sorry, but he comes first.
Why does the 25 month running average stop at the beginning of 2009?
Shouldn’t it be the ‘last 25 months’ which would therefore end at the last complete month — December 2009.
Norm in Calgary (22:01:36) :
Shouldn’t it be the ‘last 25 months’ which would therefore end at the last complete month — December 2009.
It is correct to center the 25-month curve on a time 12.5 months ago.
[quote Norm in Calgary (22:01:36) :]
Why does the 25 month running average stop at the beginning of 2009?
Shouldn’t it be the ‘last 25 months’ which would therefore end at the last complete month — December 2009.
[/quote]
The smoothed value is usually the midpoint of the interval being smoothed, so the first and last time periods are usually cut off.
You can read a nice article about data smoothing here:
http://www.climate4you.com/DataSmoothing.htm
nanuuq (20:34:25) :
Oh, why didn’t you say so? It’s simple – the science isn’t settled and no on knows for sure. If we did there would be a lot few blogs on the subject.
BTW, it’s not getting warmer, that pretty much ended a decade or so ago and lately it’s been getting cooler. No big deal, it’s happened many times through history.
Consider buying a copy of Strunk & White. It would point out that your “Like, dude, ” is unnecessary.
Norm in Calgary (22:01:36) :
It covers the 12 months before a data point and the 21 months after, e.g. the Dec. 2008 data point covers Dec 2007 to Dec 2009. Any months later than Dec 2008 don’t have 12 months data going forward. I.e. Jan 2009 data is not yet available.
A picture is worth a thousand words, and a graph is worth a thousand pictures.
nanuuq (18:04:35) :
inserted into the biosphere by the burning of “buried sunshine”?
I saw that show on Science Channel too. Catchy little phrase ‘buried solar energy’, or whatever they called ‘fossil fuels’. They didn’t give any data for that hypothesis. But it sounds right to some people—I guess—burning buried solar energy thus adding to the solar energy already in the atmosphere and thus warming the planet. They didn’t say anything about the cooling of the last 5 years though. Alex Filippenko had a tone of voice that didn’t seem like he subscribed to it completely.
They must have been talking about coal when they said ‘fossil fuels’ since it is possible oil and natural gas may not be fossil fuels, or, ‘buried solar energy’.
nanuuq (20:34:25) :
like dude, i was just askin
whats causing it to get warmer?
You can read up on ocean currents, the sun, galactic cosmic rays, UHI, and volcanoes. But it hasn’t been warming for about 15 years running now. So quit thinking global warming is happening.
And for the last 5 years there has been cooling.
J.Peden (21:58:38) :
The great ipcc Climate Scientist wants to know why it isn’t getting warmer. He’s called the divergence of “global mean” temp. from CO2 concentrations a “travesty”, and is quite upset about it.
Alas, poor Trenberth. He doesn’t recognize an object in motion when he sees it. It’s really quite simple: The warming caused a delayed release of trapped C02, and when the warming stopped, the C02 is the object in motion.
Throw in increased Volcanic Activity associated with a long Solar Minimum and it’s liable to keep right on rising.
Even Gore almost got it in his movie when he said “C02 rises with temperature as you can see from this graph…well, not really, it’s complicated” wnen watching the Ice Core data line drawn.
Silly Gore, tricks are for kids.
Dear magicjava, yes, you gave good arguments why such errors are unlikely. So it may have something to do with the actual surface vs. above surface difference.
BTW I realized that the reflected IR radiation from the Sun is really negligible relatively to the thermal one from the Earth, so this can’t be the issue even if there were something wrong with its calculation.
If somebody posts their data and their smoothing algorithm ( I prefer the algorithm to the verbal description of the algorithm), then they can’t
“hide a decline” or “hide an increase” at least not for long. Any “tricks” they
pull will be readily apparent as people will dissect the data and the code to
to show how the choices made in the filter create a different visual message..
OR NOT. The only time we have issues is when the chart maker hides his data and his code and people have to guess ( like St Mc and roman and UC and jeanS had to guess) how the smooth curve came out of the spiky data.
Norm in Calgary (22:01:36) :
My guess would be because a finite impulse response (FIR) filter, which is what a moving average amounts to, has a delay of half its length.
photon without a Higgs (20:11:54) :
This is not universally true. It depends upon what you are doing and why. You also need to take into account any artifacts that result from your smoothing operation should you choose to use the results in further processing (FIR smoothing is acausal, for example, which can be an issue).
I agree, however, that there’s no real reason a 25 month smooth should come under any particular scrutiny since he clearly pointed out what he did, why, and what the result was. For display purposes, who really cares as long as we have the explanation?
Mark
@ur momisugly Louis Hissink (12:31:56) :
yes, there is an impossible political goal to ‘produce’ more ‘scientists’ by filling up the universities with people who are best in learning something by heart and regurgitating it when asked. For years I attended at interviews on people to be hired. If the vacant job needed someone who was able to apply his knowledge in new situations we checked accordingly by stating nonsense once in a while. If there was no opposition, guess what. Precarious is that ‘filling up’ appears to work in the wrong direction because the ‘good’ people who treat learning as some non-formal activity are graded out of the universities.
nanuuq (20:34:25) :
like dude, i was just askin
whats causing it to get warmer?
i just want to know
Hmmm, maybe we should start a fund dedicated to providing people like nanuuq with a copy of Glenn Beck’s book, “How to Argue with Idiots”, so they could have an authoritative source of answers to their questions right at the bedside, and wouldn’t have to get up and waste their time coming over here.
Looking at the UAH data (http://discover.itsc.uah.edu/amsutemps/) I find that the lower atmosphere at 4,4 km (ch05), is extremely hot while the upper part at 17 km (ch09) is extremely cold now compared to the 20 year averages.
Does somebody have any theory on what is going on?
Thanks for the various thoughtful replies to my post – the problem with the deductive sciences is that they tend to fall into intellectual cul de sacs and it’s these that I think Kuhn was thinking of when he introduced the idea of paradigm shifts – one dogma replacing another one not my evidence but from the older guard dying out and being replaced by the young turks with the replacement theory. That can only happen with theories that aren’t anchored in physical reality and it is one big problem – especially in geology.
It is telling that the Lyell changed geology by artful rhetoric, one of his supporters complimenting him on the fact that he could write a text on geology without once mentioning “stratigraphy”.
So also with climate science – where the rhetoric is now couched in maths and stats and complex computer models – it’s technically sophisticated rhetoric that we are confronted with, not evidence per se. And this approach has been with us since Plato.
How the heck do you counter an entrenched belief system that is AGW?
I suspect if the Almighty returned they would interpet it as another denialist plot and thus to be automatically rejected.
So while it is excellent that we are now discovering the real extent of “creative science” (as in creative accounting), that discovery may not necessarily have any effect on the climatically devout if past experience is any indication.
Do we then have to come to a Mexican stand-off and wait for them to lose interest in AGW to inflict us with another bout of intellectual pestilence, or is it possible that public opinion, if swayed, will cause them to blink.
That Roy can receive emails of the type he described above in spite of the facts suggests this battle is far from over. They are Strongly led, you know.
Cement a friend (17:27:59) : I have suggested before that “forcing” is not a sensible term. The driver of heat transfer is temperature difference.
You know, I’ve repeatedly asked the folks who use the term “forcing” to tell me what the SI units are for it, and answer came there none…
Until a reputed phsical driver of the climate has a real physical lable with a real phsical unit of measure (be it S.I. or Imperial or even that strange not quite Imperial “customary” that we use in the USA) it is just so much phlogiston explaining why we are going to all burn up…
Per the graph:
Stock traders look at a 100 of these kinds of charts a day. Price with 2 moving averages or more. Personally, my basic chart uses price and 3 moving avereage ( typically 20, 40, 60 period or 25, 50, 75 period, but sometimes especially on hourly charts, 12, 24, 36 period. Those numbers are not randomly picked. With about 100 – 200 data points available in most graphs with reasonable length, they give the best indications.)
Now stock traders have spent untold millions of hours finding the fastest way to get meaning out of those charts (some would say ‘fabricating meaning’ 😉 that can be used for trades that determine if they eat or not this week. (for some of us, more literally than for others…)
What they tend to do is look at moving average crossovers as signals of change of short term direction. They also look at a kind of “connect the dots” for context. This is all called “technical analysis” (as opposed ot “fundamental analysis” that looks at income statements…)
Using those tools, I would look at the chart above and see a big spike in 1998 (potentially a “news driven event” – i.e. flash in the pan and gone) but often indicative of a “blow off top” – whatever was driving upward dissipated in a final “POOF!” kind of like a pile of flamables burns slowly then picks up speed just before running out of fuel to drive it…
But the interesting one I see is when you connect the tops of the temperature excursions with each other (fitting a trend line to the tops, not the mean) and do the same for the bottoms. We find a nice upward trend line to the tops in the early 1/2 of the chart, that is violated to the upside in 1998, then we have a new trend line to the tops that does not advance. Lay a line on the tops of the 13 or the 25 SMA lines after ’98 and in the first case it slopes down, in the second it is flat. Now look at the bottoms. Individual years have a big spike to the downside in temperatures after ’98. (Like ’04 and ’08) In years before the spike the downside excursions are small and narrow in comparison with only ’89 looking similar (though with a fast recovery). A “connect the dots’ on the bottoms also shows two segments with a kink. Upsloping before 1998, but after ’98 we have a downward slope on the direct tempertature bottoms, an almost flat slope on the 13 SMA line, and a still upward slope on the 25 SMA line (Simple Moving Average). THAT is an inflection. The prior trend is broken.
“Failure to advance” to the upside. Stronger excursion on “down days” to the downside. Top trend dead flat to downward sloping. Bottom trends working off energy and with temp data moving downward first, pulling the SMA’s in order after it. This puppy is in rollover mode after a strong move up.
Can you apply these “stock trader techniques” to temperature data? I have no basis to assert you can. (Heck, a lot of folks assert you can’t apply them to stocks! ;-0 ) But I think they reflect the underlying energy of a complex chaotic system and do let you see what is happening to trends.
In some ways, they seem to me to be a way to get a first and second derivative out of really messy data without any formula. But they also are helpful in “betting the right way”. So I’d look at this and ‘revisualize’ those trend lines (top, bottome, etc.) as a smoother version of this data along with a derivative that would show an inflection. But with the time scales on this graph, it is looking like one that will take 20 years to run and might be 5 years from now before it was clearly in evidence (by which time ‘the trade is over’; at least for stocks, you must trade on too little evidence if you are to make the trade in time… )
What I would do here is put on a 13, 26, 39 SMA stack (well, really, I’d use a 12, 24, 36 for stocks, but I understand the desire to avoid harmonics of the seasonal calendar with the 13…) and then look at the “roll over” of the stack. We have a nice “weave” of the temp with the 13 and the 24. Adding a longer term SMA like 39 as well would let us see if the two faster ones had “crossed over” it to the downside ( I think they have, but the spike makes my ‘visual integration’ a bit higher error band… real numbers would work better). When the two shorter faster SMAs have crossed the faster one to the downside, and the trend lines have gone flat and kinked downward, that’s a classic top inflection.
The driver of mass transfer is concentration difference which can be expressed by partial pressure differences. The term “Albedo” is meaningless in considerations of heat and mass transfer. Clouds of water and ice molecules will absorb radiation both from the sun and the surface and will re-emit the energy in all direction including space. The density of clouds varies. Some “black” clouds will even absorb most visible light.
Hurrah, hurray!!! Another person who wants to see the real physics involved, not just one side effect (temperature). It is energy and mass transfers that matter, not the temperature! (In stock trading “volume” is the “mass” equivalent and it is the “voume” that tells the real story…)
Until we start looking at quantity of energy moved in the mass of water that just turned to snow in the N. Hemisphere and how much of THAT got dumped out the upper atmosphere to space, we are just counting how many coins we have in our purse and not looking at the denominations…
J.Peden (00:53:20) :
Hmmm, maybe we should start a fund dedicated to providing people like nanuuq with a copy of Glenn Beck’s book, “How to Argue with Idiots”, so they could have an authoritative source of answers to their questions right at the bedside, and wouldn’t have to get up and waste their time coming over here.
Assuming his plea is genuine, then he has taken first step on the road to learning. He is probably young, so perhaps a little understanding is called for.
After all adults and scientists often think they know everything and aren’t afraid of telling young folk so.
Dotto (00:54:40) : Looking at the UAH data (http://discover.itsc.uah.edu/amsutemps/) I find that the lower atmosphere at 4,4 km (ch05), is extremely hot while the upper part at 17 km (ch09) is extremely cold now compared to the 20 year averages.
Does somebody have any theory on what is going on?
My theory, that is admittedly “idiot stupid simple” is that the top is radiating the energy away to space, then can sink, while the bottom is still picking up heat from the ocean to transfer to space eventually (i.e. rise). We’re seeing a giant heat transfer from the ocean to space (temperature gradient).
In other words: This is what it looks like in the early stages of getting darned cold starting from pretty warm. And it’s a decade scale process, so “patience grasshopper”…
david elder (19:41:52) : (a) why does Dr Spencer’s 13 month running average seem to start six months earlier than the 25 month running average?
(b) why does the 13 month running average also finish six months later than the 25 month running average?
You need 13 periods of data to calculate a 13 long Simple Moving Average. You need 25 periods of data to calculate a 25 long SMA. 25-13= 12. The 13 can start at 13, the 25 must wait until time 25 for the first data point.
Now I’d *guess* that the 25 was offset to align the peaks ( I find it a bit disconcerting, since in stock trading we depend on the time offset to be part of the signal, but i digress) so take that 12 and split it between each end, you get 6 periods of ‘extra’ on each end of the 13 SMA that is not available in the 25 SMA due to the number of data points needed to calculate each.
Also is there any significance in the fact that the running average values in months, 13 and 25, are of the form 12n + 1 where 12 = number of months per year?
I would *assume* that it is to avoid exact harmonics of the seasons, but it is also possible that the desire was to avoid ‘edge effects’ where your last period is contaminated by things like “updates” that come in later, so you “go one further” then ignore the last period. Both of these are speculation on my part as to motive.