Guest post by Steven Goddard
The GISS website shows the graph below, which indicates a steady, steep warming trend over the last 30 years. The monthly average anomaly for 2008 (0.44) is 0.26 degrees warmer than the monthly average anomaly for 1980 (0.18.) Data obtained from here: http://www.woodfortrees.org/data/gistemp/from:1980/plot/uah/from:1980
By contrast, the UAH monthly average anomaly for 2008 (0.05) is 0.04 degrees cooler than the UAH monthly average anomaly for 1980 (0.09.) Again, data obtained from here: http://www.woodfortrees.org/data/uah/from:1980
This 1980-2008 discrepancy between GISS and UAH is important, as it is nearly equal to the claimed warming trend since 1980.
Taking this one step further, I made a graph of the difference between the GISS and UAH monthly anomalies since 1980.
As you can see below, the discrepancy has increased over time. Using Google’s linest() function, the divergence between GISS and UAH is increasing at a rate of 0.32C/century. (GISS uses a different baseline than UAH, but the slope of the difference should be zero, if the data sets correlated properly.) The slope is not zero, which indicates an inconsistency between the data sets.
Click for larger image
Raw data from here: http://www.woodfortrees.org/data/gistemp/from:1980/plot/uah/from:1980 Calculations done here.
Factoring in the baseline
Some readers will undoubtedly again point out that the GISS baseline (“normal”) temperature is lower than the UAH baseline. This is true, but as I said above does not affect the slope calculation. The difference between the GISS and UAH monthly baselines is a constant, which affects the relative position along the y-axis – but it does not affect the slope. Subtracting a monthly constant from each point in a graph does not alter the slope over a large set of years. It only alters the y-offset.
The equation of a line is y = mx + b, where m is the slope and b is the y-offset. m and b are completely independent. The different baselines affect only b, not m. If the UAH and GISS data were closely tracking each other, the slope (m) would be close to zero. The fact that GISS shows 2008 temperatures much higher than 1980, and UAH shows 2008 temperatures lower than 1980, is also a clear indicator that the two data sets are divergent.
Steve McIntyre has coincidentally just done a similar comparison of NOAA USA yearly data vs. GISS USA yearly data, and came to the conclusion that the NOAA slope is even steeper than GISS, diverging from UAH by 0.39C/century.
This would imply that NOAA is diverging from UAH by an even larger amount than GISS is diverging from UAH.
Clearly, problems exist with both datasets.
Steven Goddard (15:37:12) :
Also consider that during the 1960s and 1970s, temperatures dropped so much that many people were worried about an ice age. Yet CO2 increased steadily during that period.
The simple cause and effect you are looking for is perhaps not so simple.
The Al Gore and MM disciples have a response for that. “A natural cooling trend was masking the catastrophic anthropogenic global warming which was taking place at the time.” 🙂
A unique phenomena which can occur whenever a valid scientific explanation is not available. I fully expect in future years that an ‘adjustment’ to the data will make the inconvenient cooling of the 70s to disappear.
I didn’t read all the comments, but if two data sets are fairly similar and going in the same direction and they both use similar methods, but the third data set that uses a conversion of some other data as a proxy for temperature (from what limited knowledge I have of satellite data) how can we trust the rogue data set over the first two?
Just so nobody is misled, – the graph Smokey linked to uses UAH data, however the graph itself and the fitted curve are the work of Andrew Barr, graphic artist at the National Post, not the University.
http://network.nationalpost.com/np/blogs/fullcomment/archive/2008/10/20/lorne-gunter-thirty-years-of-warmer-temperatures-go-poof.aspx
It appears to be a sixth order polynomial, with the last part hand drawn. I am not aware of any sixth order processes in the climate so the statistical legitimacy seems approximately nil (it may be just a coincidence that a 6th order polynomial is the highest order available in Excel who – knows?) To demonstrate this, here is the same graph with red lines showing what the same fit would have produced had it been performed in 1998 or 2006. Scary!
Here’s the link to the RSS comparison. It seems to have gotten lost in my previous post:
“Comparing RSS and UAH Atmospheric Temperatures to Adjusted Radiosonde Data.”
http://www.remss.com/msu/msu_data_validation.html
First, thanks to Arn Riewe for the link – I’ve been noodling around over there for awhile. Also, thanks to crosspatch, who seems to be saying that politics is playing a very important role in this debate, which, I’m afraid, I already knew but was trying my hardest to deny. Politics I can understand [got “clean for Gene” in ’68 – I would say I’ve grown up, others would say I’ve regressed]. My hopes, dare I say expectations, that science would somehow be above politics have apparently been dashed.
For a totally gross generalization, it seems to me that many fervent believers in AGW come from the left of the political spectrum. With that in mind, I will give you my only possible contribution to this discussion. The new energy czar, Carol Browner, is a former Commissioner of the Socialist International, with the website scrubbing her name. http://www.foxnews.com/story/0,2933,480025,00.html [this is certainly neither the first nor the only internet article on this subject]. I don’t think this bodes well for those, like me, who would hope that this discussion would be non-partisan, or that the effects of any policy would take seriously into consideration the (relatively) free-market economic philosophy that has allowed this Country to feed, clothe and provide medical care for many in this world. While I doubt that I will ever comprehend the “science” of global warming as well as those who “live” here, I think I know politics as well or better than most. And what I know about that does not make me optimistic.
I am interested in why the analysis starts in 1980, given that both datasets have complete figures for 1979?
The 1979(1980) numbers are UAH: -0,08 (+0.08). NASA 0.09(0.18).
So the effect of starting a year after the data starts is to give a larger divergence, start in 1979 and both deltas are positive at 0.16 and 0.35C respectively. I don’t set much store by single year comparisons, but I am just interested why 1980 was chosen – a nice round number?
JP
Alan Wilkinson (16:55:55) :
Mike, if you look at my chart (comment at 15:47:48) you can see it isn’t just the 1998 peak that is lower – all the fluctuations are smaller in GISS than in UAH. So the smoothing effect is not a one-off but consistent.
Looking at the difference chart at the top of this post, 1998 hangs down like an icicle off a melting glacier. The conspiracist would say that’s because the whole global warming movement would be in trouble (no record high healdines) for decades if they didn’t severely damp the 1998 peak. If I may repeat posting this chart –
http://i40.tinypic.com/10p11mt.jpg
the consistent floating of GISS above everyone is apparent, but the damping of 1998 shows up quite clearly.
On your chart,
http://www.iforce.co.nz/i/fc1cef18956ae2a7ac491b97c3409f92.gif
one thing I noticed were the GISS temp spikes recurring every few years after 1990, each just a bit above the previous spike. Were I a conspiracist myself, I’d bet the linear trend of these peaks alone matches the official govt warming rate.
But as I’m more sun guy than conspiracist, I believe Hansen’s next spike might be a very long way off.
Re the UAH/RSS divergence, here is another Excel version:
http://www.iforce.co.nz/i/50e2a7c12e04437074bc86a62b282c82.gif
which shows the data and it is clear that there is a one-off step function between RSS and USS in mid 1993 which is what Douglass and Christy discuss in the paper I referenced:
http://arxiv.org/pdf/0809.0581v1
One problem with GISS is that it is based on data from mostly problematic surface stations.
Another problem is, that nobody really understands what the corrections do with the data.
Another problem is, that nobody can reproduce the data.
Another problem is, that nobody can write down, what the corrections are supposed to do.
It is obvious therefore, that nobody can explain or will be able to explain, why GISS shows an increase of 0.32 deg/century over UAH. GISS is just garbage that is good for nothing and shouldn’t be used anywhere, particularly not in real science.
Just for completeness, here is the chart from 1979 instead of 1980. It doesn’t make much difference except the GISS slope increases a tad above RSS.
http://www.iforce.co.nz/i/4ae531bcb6bb7cef4b8214ec0d8952c5.gif
I did a standard deviation analysis of UAH and GISS which confirms John Christy’s 1.23 times multiplier for GISS to the lower troposphere. The difference is from models and sonde data and from what I can tell it is correct
http://noconsensus.wordpress.com/2009/01/06/more-fun-with-giss-temps/
Before you take the difference you really need to divide the sat data by 1.23. This makes the trend much worse. This is the reason the blue line dipped so low in 1998. I haven’t heard a good explanation for the trend being so different yet except for the obvious one.
I also got a raw slope for 30 year giss of 0.183 and 0.128 for UAH for uncorrected data. No adjustment for days of the month.
JP can you explain your numbers? I don’t see much effect from including 1979 at all.
Also, the author gave the links where he got his numbers from. Neither files don’t extend to prior to 1980, which is obviously why he stopped there.
Also, the RSS publishes the following trends:
Channel TLT 1979 2008-12 29 0.157 K/decade (temp. lower troposphere)
Channel TMT 1979 2008-12 29 0.092 K/decade (temp. middle trop.)
Channel TTS 1987 2008-12 21 -0.029 K/decade (temp upper trop.)
Channel TLS 1979 2008-12 29 -0.334 K/decade (emp lower stratosphere)
Based on this, it would appear that UAH is measuring somewhere towards the uppper end of the TMT. Assuming that what it is measuring is an accurately calibrated temperature of course.
Some of the most questionable trend lines I’ve seen come from Had-Crut
http://www.cru.uea.ac.uk/cru/data/temperature/nhshgl.gif
Look at the Southern Hemisphere for the last six years. Temperatures have been dropping at an accelerating rate, yet the trend currently has an upwards curvature. Similar for the other plots. The southern hemisphere trend in particular should be tailing off fairly steeply now, but it isn’t.
I commiserate with Walter Cronanty, as I am also not a scientist and I too have difficulty understanding some of the presentations here. (However, Al Gore is not a scientist either, so I feel he is no more competent than I.) But I am not surprised that science has not risen above politics in this debate.
I am a retired litigation attorney with 30 years experience. I often sought assistance from scientific experts to introduce evidence and to critique testimony of the other side’s experts. Thousands of lawyer do this every week in the USA. With this assistance as well as their own study, trial lawyers learn to analyze scientific evidence as to whether the material appears logical, complete, unbiased and depends on accurate data.
The first point to keep in mind is that scientists who testify are not super-human and they are subject to human imperfections. Most scientific expert witnesses are competent, thorough, knowledgeable, and straightforward. It is not unusual, however, for some expert scientific witnesses to display stubbornness, arrogance, insecurity, sloppiness, inaccuracy, and vulnerability to financial incentives.
Not long ago, I began paying attention to the AGW issue, and noticed advocates making mutually contradictory claims. The general news media is no help. So to get to the bottom, I decided to review the materials as if preparing for a trial and see if I could reach a conclusion.
As I began reading I started to have the same “cold feeling on the back of neck” that I felt in trials when listening to opposing expert witnesses testify: there is something wrong here. My judgment, after studying hundreds of pages of articles, reports, blog entries and graphs, is that AGW is a hoax. I would love to be able to cross-examine some of these people as if at trial where they would be under oath and could not obfuscate or avoid directly answering the questions.
Those who understand these data know that the trend lines in the UAH, RSS, GISS, UAH, NCDC, and UKMO radiosondes products are statistically indistinguishable and all show a statistically significant positive trend.
Why haven’t we seen a P value?
Re: Walter Cronanty (12:19:58) :
Exactly….. It proves that the Proponents of AGW, champion it’s cause, not from a science perspective, but rather from a political perspective.
The Environmentalists and the Neo Socialists, usually in the guise of Social democrats, are making funding available to the field of Climate science in order to influence the findings and thus, use that to justify their policies.
It has nothing to do with Climate and everything to do with Socialist politics…..
There. I’ve said it…….
Well how else do you explain how an obscure hypothesis, has become a 50 billion dollar industry in twenty years, with designs on changing human society, economics and energy policy….. How? What? Why? Because the science is good, the observations concure and the methodology is impeccable???….. Pffft.
Political meddling coupled with unskilled or politically motivated scientists. That’s how.
Steve uses Woodfortrees as his data source, which does have UAH data from 1979:
http://www.woodfortrees.org/data/uah
so the question of why 1979 was excluded is still of interest.
[ BTW:Woodfortrees is a secondary source, which imports its data from the primary sources which you can find here
UAH http://vortex.nsstc.uah.edu/public/msu/t2lt/uahncdc.lt
NASA: http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts+dSST.txt
I’ve never found an error in the WoodforTrees treatment, but primary data is always best. ]
I am reminded of a piece along the same lines that Mr Goddard wrote for the Register in which he points out that According to Hadley’s data, the earth is not much warmer now than it was than it was in 1878 or 1941. … By contrast, NASA data shows worldwide temperatures increasing at a record pace – and nearly a full degree warmer than 1880.
Do you see what he did there? – different start dates for the different series, It just so happens that 1880 was >0.2C cooler than 1878 in the Hadley data giving him a bigger delta. Most of the rest of the difference was accounted for by the fact that the 2008 figure was a YTD figure which is distorted by the unusually cool (La Nina) Feb 08 – I doubt that a climate auditor would be impressed.
I am not doubting that UAH does indeed diverge from GISS and the other two major global series, and the reasons may well turn out to be interesting. One would not expect an exact correlation between GISS and UAH as they are measuring different quauntites, the surface temperature and the temperature in the lower troposhere (which is more sensitive to ENSO peturbations)respectively. Actually the satellite transponders measure ‘brightness’ which must be then converted to temperature, a non trivial process. RSS and UAH use the same raw data and apply different treatments, so the divergence between them is probably an artifact of the different algorithms. There is an unfortunate precendent, for many years there was a flaw in the UAH analysis (an incorrect sign in an arithmetical term amongst other things), that caused them to under-report the trend substantially. Let us hope all is well now.
JP.
Stephen,
Just to point out you can actually do the whole thing at WFT:
http://www.woodfortrees.org/plot/gistemp/from:1979/offset:-0.24/mean:12/plot/uah/mean:12/plot/gistemp/from:1979/offset:-0.24/trend/plot/uah/trend
or, including all RSS & HADCRUT3 as well:
http://www.woodfortrees.org/plot/hadcrut3vgl/from:1979/offset:-0.15/mean:12/plot/gistemp/from:1979/offset:-0.24/mean:12/plot/uah/mean:12/plot/rss/mean:12/plot/hadcrut3vgl/from:1979/offset:-0.15/trend/plot/gistemp/from:1979/offset:-0.24/trend/plot/uah/trend/plot/rss/trend
trend figures here:
http://www.woodfortrees.org/data/hadcrut3vgl/from:1979/offset:-0.15/trend/plot/gistemp/from:1979/offset:-0.24/trend/plot/uah/trend/plot/rss/trend
What’s remarkable to me is how tightly GISTEMP, HADCRUT3 and RSS all agree, at around 0.16K/decade, with UAH as a noticeable outlier at 0.13K/decade. I’ve never seen any explanation of that…
Walter Cronanty (18:10:02) :
If you think Carol M. Browner’s record is amazing, check out Holdren’s or Chu’s. I personally tipped off Steven Milloy (author of the Foxnews article) to Browner’s socialist leadership position on Jan 1, 09 at JunkScience.com and he broke the story within an hour or so of my email (I’ve never seen the page change during the day like that). He wrote me later saying “thanks to you…” with the link to the Washington Times article…! One small victory for me.
The case for catastrophic AGW is entirely political in my opinion. The scientific basis for it is extremely weak. Even the IPCC has to base their summaries on only a few papers purporting extreme positive feedbacks to support their case, feedbacks which, I might add, have never been observed in nature, and are shown by other studies to be negative. And even their conclusions are written by a small group who overstates the certainty, or have even changed the conclusions in direct conflict with the panel’s opinions (Ben Santer)… This is why I often call the IPCC “a consensus of one”
If you ignore enough variables, or assign them weak enough coefficients, then, I would agree, there isn’t much left but CO2. But why would you do that? Assign them proper coefficients (like those that a statistical analysis reveals), and CO2 might still be a factor, but isn’t even remotely in the same league as ocean cycles, solar, etc. You can leave it out completely and explain upwards of 88% of temperature variation. Adding it doesn’t change that figure by much.
Emergency indeed… for whom? and for what?
JP: Yes, for the record, I would also recommend anyone who wants to do their own analyses to go to the original sources, not re-use WFT data. Not that I think there’s anything wrong with my intepretation, of course, but the fewer stages in the pipeline the better.
BTW, if people are finding it tricky to read the many and various formats, you can just fetch the original files and use the underlying ‘analyse’ tool to parse them (“use the source, Luke”).
http://www.woodfortrees.org/downloads/analyse-0.7.2.tgz
Incidentally, I didn’t mention it above, but the OLS trend differences of 0.16K/decade – 0.13K/decade = 0.03K/decade = 0.3K/century, which corresponds more or less to Stephen’s estimate in the original article.
Data from: http://www.woodfortrees.org/data/gistemp/from:1979/trend/plot/uah/trend
This is quite significant, of course, but not nearly so significant as the difference between either of these and the 3K or 6K/century being bandied about at the more alarmist end of the spectrum. The sign and magnitude of the feedbacks are what’s really critical.
Hi Paul,
Thanks for the info, and obviously I value your site quite a bit! Looking at your examples, I don’t see how to do a difference between two plots, say GISS minus UAH, which was the exercise in this article. Also, I like using your raw data because you record the month as an absolute decimal, which allows a minimum of pain in spreadsheet computations.
DJ,
Please feel free to report on the results of any further any statistical calculations you feel are appropriate. My data sources and calculations are all listed in the article.
I’m doing another analysis of just the last six years – which the divergence between GISS and satellites is quite large.
“I want to see the reaction of a $.50 a gallon tax increase and CO2 taxes on generating plants. It’s been zero here for 2 days and people are laughing at the mention of global warming. Most people look at now, not the past and don’t care about the future.”
One of the issues is that things will be done slowly, and on a well-crafted scale, such that people don’t see the individual changes as being problematic. 10yrs later, people will be looking around saying “How the heck did we get HERE???”, but it will be too late. Look at banking regulations and the economy.
JimB
Alan: UAH vs. RSS. Indeed there appears to be a marked divergence from the mid 1990’s
http://www.woodfortrees.org/plot/uah/mean:36/plot/rss/mean:36
although if you zoom in it’s hard to pinpoint to a particular year:
http://www.woodfortrees.org/plot/uah/from:1990/to:1995/plot/rss/from:1990/to:1995
From above:
“The Al Gore and MM disciples have a response for that. “A natural cooling trend was masking the catastrophic anthropogenic global warming which was taking place at the time.” 🙂
A unique phenomena which can occur whenever a valid scientific explanation is not available. I fully expect in future years that an ‘adjustment’ to the data will make the inconvenient cooling of the 70s to disappear.”
—-
Er, uhm, well, you see, they already made that inconvenient adjustment.
ANOTHER peak decade which has ALREADY conveniently “disappeared” from the GISS data is the 1932 – 1942 peak.
In the “original” GISS graphs in 1998, 1932-1942 peaked at .4 degree above baseline, then temperatures declined over the next 30 years into the artificial “0.0 baseline” of 1972, then began climbing again from 1972 through 1998.
Today’s GISS graphics?
Well, (magically) somehow the temperatures in the 1930’s and 1940’s have decreased significantly – until they are just a bump crushed under the AGW’s political bandwagon.
The difference in the various series from 1979 to 2008 is small but very, very significant when extrapolated over the 250 years we are talking about for global warming’s impact.
The UAH trend indicates global warming will not be a problem at all.
The RSS trend indicates global warming will not be a problem at all.
The GISS and Hadley Centre data since 1979 and since 1880 indicates global warming will only be about half of that predicted.
The satellite-derived temperatures since 1979 also point to a conclusion that the pre-1979 temperatures from GISS and the Hadley Centre were probably manipulated – artificially increasing the trend since 1880 by about 0.3C or so.
When you extrapolate that over the 250 years of global warming in question here – it also indicates that global warming will not be a problem at all.
These small differences are important.