How bad is the global temperature data?

on the “number of stations” over time and by category graph – what is the y axis? just curious on the actual numbers.

First time I’ve done this so if I get something wrong/miss something obvious feel free to flame.
Right, had a look at my patch, the south of England. As far as I can see, there is just ONE station still operating in the southern half of the country (south of Lincoln.)
Now this station is named ‘Bournemouth A’ and looking at the co-ordinates given on the map, it looks like it is in Bournemouth Airport. So the only data station still in the set (that I can see), in the entire southern half of England (on the edge of the Gulfstream) is in an airport (possibly).
Despite this, the last full temp reading (2008), was just 0.01 degree C above the first reading at this station (1961).
Now I know you’re doing an audit of these stations, so a couple of questions:
1) How do I find out *exactly* where it is to do an audit (unless it has already been done).
2) If I do go and have a look, what exactly am I looking for.
Apols for the rambling post.

Well why do we need so many stations anyway. We can tell the climate for 800,000 years just from one ice core drilled at Vostok Station (maybe with extrapolations), so who need all those other readings anyway.
Why not make just one more good climate station, and put it on The White House lawn; and have the President decide what the climate is; after all he’s a nobel prize winner. That would save a lot of money taking care of all those other stations.
Well they don’t hang on to the data anyway; they just annualize it, and then five year running average it, and then just average it in with all those other annualized five yearized numbers toi get the world climate anyway. Let Obama decide.
Do any of these “Climatology or Climate Science” courses at any of the Major Universities ever mention the Nyquist Sampling Theorem; or teach a class in basic sampled data sytem theory?
Would that be an idea insteaqd of just teaching Stat maths.
You can’t generate meaningful statistics if you have data containing no information.

Based on UAH data I calculate that the Hadcrut data has a warming bias of about .07 degrees per decade over the period that the two records overlap. I suspect that this grows smaller as you go further back but it is quite large. The main cause, I think, is the lack of data in Africa, where there has been little warming. I note that Christy et al., studying East Africa, found that the data from Hadley in the study area greatly exaggerates the trend- and that there is evidence for landuse effects there.
Within the US, GISS’s Al-Gore-Rhythm seems to work OK, but in the ROW it’s not clear anyone has a method that doesn’t suck.
Which reminds me. Joe, do you know what kind of seasonality UHI etc. would have? I’m trying to figure out why there seems to be so much winter warming in many data sets and very little summer warming.

I believe Steve McItyre has posted about this problem several times even pointing out that in many cases the “missing” data are available for download on the public Internet. For example, a site might not for some reason report a value for a month to NOAA but a quick search shows the value for that station from another source. The bottom line is that the amount of QA that goes into this data set is something close to zero.
This article highlights the very reason that data such as GISS simply isn’t believed by many people. Many of the numbers are little more than rectal extractions.

“For many, many stations, you will see the data set in a monthly tabular form has many missing data months mostly after 1990”
The post cold war peace dividend

The “missing” temperatures are, in many cases, due to a single missing day’s entry!
Ran across a fellow who’s disentangling the GISS data & Fortran programs, and odd as it may seem, GISS drops the entire month’s data if any daily entry is missing.
http://chiefio.wordpress.com/gistemp/
Why, would any sane scientist drop 27 to 30 valid data points because 1 is missing? Simple interpolation would make far more sense than filling in the month from neighboring stations (up to 1000km away). Here in the Central US, the temperature often varies by 10 F within 100 mile radius.

RE:Gulf Stream
Yeah, there have been many articles posted recently that it has slowed. It isn’t the first time. There were in 1907 about the same thing.

“Mike D. (09:46:59) :
Excellent synopsis. Kudos to Joe D’Aleo.
“The emperor has no clothes.”
If I may suggest a major modifcation to your comment:
The “CLOTHES” have no “EMPEROR”.

Was it Leif’s paper here about the solar magnetic stuff, with a black and a red graph that showed quite good correlation; and as I recall, it said somewhere there that a year wasn’t plotted if they had a missing month of data.
That makes no sense to me. If you have 11 months of data; which you are homogenising into a single annual number, it would seem to me that some interpolation from the 11 existing points; or even as simple as the average of the two adjacent points would give a better representation of what happened that year; than simply no value whatsoever; that is the ultimate in data distortion to have one year’s value be infinitely indeterminate.
As hokey as I believe this whole “science” is (climatology; not meteorology) I see no validity to just discarding a whole bunch of data for a single missing value that is going to get homogenised anyway; specially after a five year running average.

“Hi all, what’s happening with the gulf stream? Is it slowing down? Look at the map in this site!
http://polar.ncep.noaa.gov/ofs/viewer.shtml?-natl-cur-0-small-rundate=latest
Is NOAA having issues with the model or has the gulf stream really stopped?”
Yeah, that looks really strange indeed. Something happened September 18 on all maps.

Has anybody figured out what global temperature is? It seems to me like it’s a myth, perpetrated by the hoaxers. Probably based on the same math as the computer models.
How could you possibly take the global temperature and have it mean anything.

George E. Smith (09:44:23) :
Do any of these “Climatology or Climate Science” courses at any of the Major Universities ever mention the Nyquist Sampling Theorem;
You can’t generate meaningful statistics if you have data containing no information.
Funny you should mention Nyquist, after looking at thier output, they have incredibly poor spatial resolution. If they were photo journalists, the editor would have them canned.
Oh, but they do generate statistics, all the time, it just doesn’t have any value unless one is out to create art by grinding dirt into a Rembrandt.

The idea that there is something wrong with our climate is preposterous,
the global plot to regulate and curb CO2 emissions plainly criminal.
This is a crime against humanity as is Chapter 21.
Both are related.
http://canadafreepress.com/index.php/article/15713

astateofdenmark (09:33:01) :
First time I’ve done this so if I get something wrong/miss something obvious feel free to flame.
___________________________________
We don’t flame at WUWT.

Rhys Jaggar (11:33:20) :
If you believe a certain thing to be true and if in 1990 there was not much difference in temperature reported from, lets say, 8000 observations and 1000 observations and since 1000 is easier to manage, then why not drop the other 7000. That might be their reasoning. And … since the remaining observations support what they believe to be happening, then they have no reason to doubt the result.
People tend to weight information that validates their beliefs higher than data that invalidates their beliefs.
If the data comes out that shows what they expected, it (the result) is left as is without further checking. If the data results in an outcome counter to what is expected, then it is checked and rechecked and checked again.
This is why errors have been found most often when they produced a cold anomaly than when they produced a warm anomaly.
If you expect temperatures to warm, a hot spot generates an alarm for some kind of action to mitigate it. A cold spot generates and alarm to inspect the data.
Steve McIntyre is the only one I am aware of that inspects the data when unusual warming is seen.

This, plus the big holes in World cover and obviously a lack of reliable temporal History for the precision needed, is what makes me say that we don’t know what is going on. We might even be getting warmer. But the corrupt AGW “science” destroys everything.

My New Video’s Made it on Prison Planet
http://www.prisonplanet.com/lou-dobbs-al-gore-not-evil-just-wrong-global-warming-debate.html

It’s really sad. Just look at that big dark red blotch in the south of Norway where I live: It’s probably mostly due to “rural” Gardermoen: http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=634013840003&data_set=1&num_neighbors=1
(which I fear is used to correct a lot of stations in Sweden too…)
Only problem: Gardermoen isn’t rural anymore, since 1998 it’s home to Norway’s main Airport. I don’t know how close to the runways they measure temperature, but the whole area is obviously not the same as it used to be.

Climate Myths and National Security
By Viscount Monckton of Brenchley
http://www.americanthinker.com/printpage/?url=http://www.americanthinker.com/2009/10/climate_myths_and_national_sec.html
The President of the United States recently told the United Nations that “global warming” poses a threat to national security and may engender conflicts as populations are displaced by rising sea levels, droughts, floods, storms etc. etc. etc. However, it is now clear that there is no basis for the notion that the barely-detectable human influence on the climate is likely to prove a threat to climate, still less to national security.
The first principle to which any national security advisor must adhere is that of objective truth. Though he must have an understanding of politics, he is not a politician: he is a truth-bearer. Therefore, he begins by narrowing down the issue to a single, central question whose answer determines whether the suggested threat is real. He then tries to find the truthful answer to that question, and draws his conclusion from that.
Quid enim est veritas? What, then, is the truth? The single question whose answer gives us the truth about the climate question is this: By how much will any given proportionate increase in CO2 concentration warm the world? We now know the answer. The oceans, which must store 80-90% of all heat-energy accumulated in the atmosphere as a result of the radiative imbalance caused by greater greenhouse-gas concentration, have shown no net accumulation of heat for almost 70 years, implying a very small influence of CO2 on temperature (Douglass & Knox, 2009). The devastating analysis of cloud-albedo effects shortly to be published by Dr. Roy Spencer of the University of Alabama at Huntsville will show that the UN has wrongly decided that cloud changes reinforce greenhouse warming, when in fact they substantially offset it. Repeated studies of the tropical upper troposphere (e.g. Douglass et al., 2008) show that it is failing to warm at thrice the surface rate as required by all of the UN’s models, again implying very low climate sensitivity. The clincher is Professor Richard Lindzen’s meticulous recent paper demonstrating – by direct measurement – that the amount of radiation escaping from the Earth’s atmosphere to space is many times greater than the UN’s models are all told to believe. From this, the world’s most formidable atmospheric physicist has calculated that a doubling of CO2 concentration, expected over the next 150 years, would cause 0.75 C (1.5 F) of warming, at most: not the 3.4 C (6 F) that the UN takes as its central estimate.
Most analysts would stop there. Yet some might ask, “Suppose that the single satellite on which Lindzen’s results depend is defective. What then?” They might consider the economic cost of attempting to mitigate the “global warming” which, as our Monthly Reports demonstrate, is not actually happening. The figures turn out to be startlingly simple. To mitigate just 1 C (2 F) of warming, one must forego the emission of 2 trillion tons of CO2. The world emits just 30 billion tons a year. So the analyst, as a thought-experiment, would shut down the entire world economy, emitting no CO2 at all. Even then, and even on the incorrect assumption that the UN’s exaggerated projections of the effect of CO2 on temperature are correct, it would take 67 years to mitigate 1 C warming. Preventing the 3.4 C (6 F) warming that the UN’s climate panel thinks would occur in 100 years would take 225 years without any transportation, and with practically no electrical energy. The national security advisor would at that point advise his head of government that there has never been any security threat less grave, or more expensive to prevent, than the non-problem that is “global warming”. It is the fearmongers that are the real national security threat.

I’ve been looking into the discrepancy between the stations that have been retained and the stations that have been dropped over the past few years.
I’ve looked at a selection of stations in Alaska and California. In both areas the dropped stations showed the 1930’s to be as war, or warmer than the late 1990’s to early 2000’s and the retained stations have been mostly in urban areas and show the 1930’s to be much cooler than the late 1990’s to early 2000’s. This has had the effect of cooling off the 1930’s. It’s a neat trick to “cool off” the 1930’s from a vantage point in the early 2000’s.
This is a graph of Alaska stations that were dropped after 1990 compared to the stations retained in 2009…
Alaska
The dropped stations show the 1930’s to be about 0.5C warmer than the retained stations.
Here’s a comparison of California stations dropped in 2007 to stations retained in 2009…
California
The California “2009” curve is dominated by Los Angeles and it shows about four times as much 20th century warming as the “Rural” curve does (1.85 C vs 0.45 C).
I’m sure it’s just a coincidence, but the station dumping started about the same time that James Hansen was promoted from NASA-GISS climate science manager to NASA-GISS Director in 1981… And that the station dumping accelerated after he published his 1988 model.
NASA-GISS Stations

I like many others who post here, am very concerned regarding the accuracy, reliability and consistency of the data which underpins the concept of runaway global warming. But although these concerns are dealt with from time to time by WUWT in posts such as this, what impact is any of this having on MSM and the majority perception of global warming? The mantra of peer review is constantly invoked even though recent revelations suggest this process is a lot less stringent than it used to be. What I’m really trying to say is that the questioning of global warming is not reaching all that many people. I truly despair at the bastardisation of science that is occurring with climate change and which is shown so clearly in this article and I fear the sceptics have lost the war. Sorry for the negativity but I feel so frustrated that fair debate in the MSM is not permitted.

“”” Dr A Burns (13:15:05) :
These “mean” monthly temperatures are the average of the daily means of the maximum and minimum temperatures aren’t they ? How well does the average of two daily points represent changes in daily temperature ? “””
Dr Burns, I just got through that above.
If the temperature variation thorughout the day is a pure sinusoidal waveform with a period of 24 hours; then the Nyquist Sampling Theorem says you must sample it at least at 12 hour intervals, in order for those two samples to allow you to reconstruct the continuous function and hence properly compute the average.
Actually at just twice the signal frequency sampling rate; you have a degenerate case, and you can’t reconstruct the waveform (what if your two samples came at the 0 and 180 degree zero crossing points). But fortunately you can still get the correct average as the average of your two readings.
But daily temperature graphs are not sinusoids, as the heating and cooling parts of the daily temperature are not symmetrical. So that means that at the very least your temperature function must have at least a second harmonic signal with a 12 hour period; and that means you need to get samples at 6 hour intervals. Now throw variable cloud cover into the daily picture, and now you have temperature change signals at much higher frequencies, with periods of just a few minutes.
So the answer is NO, twice a dat temperature sampling will not give you the correct daily average temperature; it will be corrupted by aliassing noise; since the reconstructed signal spectrum is filded back all the way to zero frequency and beyond; so even the average is not recoverable.
But consider the long wave infra-red radiation emitted from the surface. That varies about as the 4th power of the temperature. And if you take the 4th power of a sinusoidal function, oand integrate it over a full cycle, you will get an always positive offset, that depends on the amplitude of the cyclic variation, relative to the average or DC temperature. For typical daily temperature ranges from max-min, that offset is significant; and when you consider the cyclic temperature range for the annual cycle through four seasons, the underestimate of total emitted LWIR is quite significant; and failure to properly calculate it (by using average temperatures), always underestimate the global emittance, which leads to an over estimate of the mean global temperature required to balance the total solar insolation.
The central limit theorem does absolutely nothing for you in buying a reprieve from a Nyquist violation; the aliassing noise becomes a permanent inherently unremoveable part of the reconstructed temperature funtion; because that out of band signal is convolved into an in band noise, whcih can’t be removed without removing real signal as well.
But Statistical mathematics is so much easier to teach than Sampled data system theory, and signal recovery.

PPS
I am also planning a variation of Woods experiment using double glazed units, (which I have made), one of which is plain glass, the other low E.
I did plan on using low metal glass but to be fair I had to use normal glass as there is currently no low metal low E glass available.
DaveE.

“This is something I intend to check as soon as I have enough funds to build my continuous temp monitor.”
Anthony sells a relatively cheap ($60) device to do what you want to do.
See the sidebar.

Much has been observed and extensive analysis has taken place on land surface temperature records. However, the seas cover 70% of the earth. I would suggest that those who follow the AGW debate would be very interested in understanding how HadCru and GISS calculates ocean surface temperatures. And please do not say that GISS uses satellite data — that is a very misleading claim even though there is some truth in it. Also, does Phil Jones claim that his agreements with foreign countries prevent from sharing his data on sea surface temparatures?

Yawn, yawn, yawn,
temperatures’ boredom,
yawn, yawn, yawn,
they have it all wrong.

Apologies if this question has already been answered; I don’t think I missed a mention of this, but it’s possible. What, exactly, HAPPENED in 1988-1992? What was going on there? The slow drop leading up to that fall is also interesting, too, but…Really, now. Why were so many weather stations allowed to go offline? Anyone know?

Mizen2, I’m on the case, Ellie from Belfast put me on to it and I might have done a post that would have beat Joe dAleo here but I wanted to do more on Yamal before getting involved in what could be a big and much-needed project we could do for the UK. I want to go and hunt for UK records that have really long histories. And I want help with this. All the amateurs we can get on the case. You see, GISS castrated all the records at 1880 at one end, and often in modern times I’ve found hints that a station with someone who might actually care, still keeps records, even when the station stops in the GISS records.
UK Met Office was under the RAF originally which is why so many UK records are RAF, and perhaps also why so many UK records are at current airports.
I’ve done a whole load of blink comparators. Yes, THE PATTERN IS DIFFERENT TO USA. All the end temperatures are left alone and the start temperatures are depressed. Four crimes in one! Pre-1880 castrated. UHI ignored. Integrity of early recorders poo-poohed. Trend “worse than we thought”.
Mizen2, anyone else in the UK, please get in touch!

Patrick
So I guessed right, the only station still operating in the southern half of England, is on the site of an airport. Unbelievable.

Washington is cold:
http://www.wenatcheeworld.com/news/2009/oct/12/record-cold-hits-region-snow-on-the-way/

Current trends for the temperature records based on thermometers from 1979 to the last data point:
(degrees/decade)
HADCRUT: 0.157
GISS: 0.159
And the satellite records which do not rely on thermometers:
UAH: 0.125
RSS: 0.153

Nick Stokes (15:15:20) : It’s true that some data comes in eventually which would register as missing at first, however there are still large portions of the world which have had and still have very little data. Africa has the most severe problem as far as I can tell. And that’s important because the evidence seems to me to indicate very little warming there.
What’s striking is how little influence all that data coming in seems to have. Should not the additions make some difference? Apparently not. That’s probably because the areas where data tends to be the most sparse never come in.

I don’t know if anyone mentioned it yet, but there was also another problem with some stations. The previous month’s data would be re-reported as the current month’s data. That can go unnoticed in tropical areas and even in higher latitude regions in summer or winter but it happened in the fall at some locations in Finland and Russia and at least one other place and was noticed. Going back through the record it was discovered that it had happened several times in the past.
I believe there was a post here on that as well but it has been a long time ago.

TonyB: Hm. You’re saying that it’s due to taking stations offline in Eastern Europe and elsewhere in ’93? But there’s only a small drop in the number of stations between ’92 and ’93; the main plunge takes place from ’89 to ’92. THAT’S what I’m curious about, not the much smaller dip you’re referencing. Does anyone know what caused that?

Pofarmer (18:51:27) : Satellites generally refer to the LT data from UAH and RSS. GISS does not use those, because it is trying to get a “surface temperature” metric. However, GISS uses
Reynolds, R.W., N.A. Rayner, T.M. Smith, D.C. Stokes, and W. Wang 2002. An improved in situ and satellite SST analysis for climate. J. Climate 15, 1609-1625, doi:10.1175/1520-0442(2002)0152.0.CO;2.
For Sea Surface Temperature, using sattelite based estimates of SST in the later years of the record. These are, again, distinct from the LT data, which are bulk atmospheric.

PoFarmer:
“So, where does the satelite data fit in?”
I think you already know the answer, but unlike the surface stations, the satelite data is scrutinized in every detail for any possible error. That is how data should be analyzed and we welcome it.
I honestly do not care one way or the other if the Earth is warming or cooling.
All I care about is the integrity of the raw data and honest analysis, no matter where it may lead.
But from an “old fart” that has been working with weather satellites since 1972, I kinda trust that data when properly processed.

TerryBixler (09:55:43) :
It seems that the goal is not to measure temperature but to stop measuring temperature.
They certainly have some explaining to do.

Tracking satellites since 1972?
Yup, weather satellites did not suddenly appear around 1979 like some people would like you to believe. That is the time when satellite data analysis became somewhat reliable and was archived in an official way.
Back at the Fernbank Science Center in Atlanta George, I was tracking and obtaining weather satellite images in 1972. Rude and crude, since you had to use a circular slide rule to compute the location of the satellite at one minute intervals as it passed over your location, but it was possible to obtain a decent APT image.
Scientific Atlanta donated the satellite equipment that I was using and it was based upon a Polaroid camera and a Tektronic oscilloscope to create the image, one scan line at a time. The operator had to track the antenna manually, according to the locations as calculated with the circular slide rule.
Using the Emory university IBM 360 computer, I soon learned how to program computers to calculate the satellite positions. The rest is a rather interesting personal history.
Do I trust the satellite derived temperature data? ABSOLUTLY!

let’s connect a couple of dots here…
From Monckton’s piece linked by Ed Scott (13:23:32) there is re Lindzen…
“The clincher is Professor Richard Lindzen’s meticulous recent paper demonstrating – by direct measurement – that the amount of radiation escaping from the Earth’s atmosphere to space is many times greater than the UN’s models are all told to believe. From this, the world’s most formidable atmospheric physicist has calculated that a doubling of CO2 concentration, expected over the next 150 years, would cause 0.75 C (1.5 F) of warming, at most: not the 3.4 C (6 F) that the UN takes as its central estimate.”
So the planet is radiating more longwave radiation to space than the models calculate.
George E. Smith (13:47:20) may have the reason why. From his comment…
“But consider the long wave infra-red radiation emitted from the surface. That varies about as the 4th power of the temperature. And if you take the 4th power of a sinusoidal function, oand integrate it over a full cycle, you will get an always positive offset, that depends on the amplitude of the cyclic variation, relative to the average or DC temperature. For typical daily temperature ranges from max-min, that offset is significant; and when you consider the cyclic temperature range for the annual cycle through four seasons, the underestimate of total emitted LWIR is quite significant; and failure to properly calculate it (by using average temperatures), always underestimate the global emittance, which leads to an over estimate of the mean global temperature required to balance the total solar insolation.”
So the temperature series are not reflective of true average temperature, which is more serious than mere weather noise, and thus the calculation of radiative emission is underestimated.
Or am I looking in the wrong wheelhouse?

Steve Huntwork (19:47:51) : According to:
http://www.uah.edu/News/climatebackground.php
The MSU’s started in November 1978-with TIROS-N

Correct with the Microwave Sounding Unit data starting in 1978. However, when it comes to polar ice or cloud induced albedo, that data has been available over a longer period of time.
I got involved with the MSU using the DMSP satellites prior to the first Gulf War with Iraq. By the time that war started, the software had been perfected and was able to obtain accurate upper-air weather conditions behind enemy lines.
This is when I learned to respect the MSU satellite products, because they work.

UAH v Hadley since 1992
http://www.woodfortrees.org/plot/hadcrut3vgl/from:1992/offset:-.2/plot/uah/from:1992/plot/hadcrut3vgl/from:1992/offset:-.2/trend/plot/uah/from:1992/trend

Jari: This is the list of stations actually used: http://data.giss.nasa.gov/gistemp/station_data/station_list.txt.
But that list includes a lot of stations not reporting anymore, you’ll have to look them up here: http://data.giss.nasa.gov/gistemp/station_data/ to find out when they stopped reporting.

“The scientists of the world are conspiring to fool us about global warming!”
Quite a hypothesis (again), Joe.
http://ossfoundation.us/projects/environment/global-warming/myths/station-drop-out
Read about GISTemp here:
http://data.giss.nasa.gov/gistemp/
I guess station dropouts have caused rapid warming in the Arctic and a tremendous buildup of heat in the oceans since 1950? There does seem to be a strong correlation.
See: http://www2.sunysuffolk.edu/mandias/global_warming/modern_day_climate_change.html
Finally, the oceans cover 70% of the surface. How much urbanization on the ocean surfaces has caused their temperatures to be artificially heated? 🙂
A good article about how surface temps are measured can be found here:
http://www.yaleclimatemediaforum.org/2008/04/common-climate-misconceptions-global-temperature-records/

Scott Mandia,
in one of your links Gavin says: “station dropping out doesn’t affect the trends at all (you can do the same analysis with only stations that remained and it makes no difference)”. Gavin is confident that he can get the temperature correct even if there is 1200 km between stations.
I do not quite understand this statement. How many more stations can be dropped out before the trends will be affected? How it is possible that September 2009 GISSTemp can show almost all Greenland 0.5 to 4 degrees hotter than normal (with the 1200 km smoothing radius) when the 250 smoothing radius shows that there was NOT ONE single station used for the whole Greenland in September 2009 (in GISSTemp)?
I would think that Greenland temperature trends should be very important for global temperature trends and monitoring climate change in general.

John Finn (00:32:02) : Again, this is 1. Highly misleading and 2. Meaningless.
OF COURSE starting in 1992, right after Pinatubo, will elevate the trends, and because the response of the UAH data is greater, it will be elevated more.
But “agreement” means nothing, sense it is quite clear that the LT should warm more.
So let me explain what I did. First, I took the UAH and Hadley data, put them on the same baseline, detrended them, calculated the coefficient of a OLS best fit of the residuals of one to the other, then used that to adjust the original UAH data to the implied surface data. This resulted in a trend of about .09 degrees per decade, while the trend in Hadley was a little more than .15 degrees per decade. I would say that it is probably more accurate to say that the bias was something like .06 degrees per decade.
Scott Mandia (04:07:16) : Did Joe ever say that the bias in the record caused all the warming? No, he didn’t. So why are you devoting energy to tearing down that strawman.

@ timetochooseagain (07:07:22)
Joe claims:
Numerous peer review papers suggest an exaggeration of the warming by 30%, 50% or even more. without citing them, of course.
He also posted awhile back about UHI and applied his land-based UHI correction to the oceans. Huh?
Joe may not have explicitly stated how much warming he thinks UHI has caused but he is strongly implying that it is significant.
The problem with his arguments is that UHI is, for the most part, insignificant.
RSS (0.18 C per decade) and UAH (0.14 C per decade) satellite-only derived temperatures show about 0.16 C per decade of warming since 1979, and GISS and HadCRU which use both surface data and satellite data show 0.16 C per decade. Where is the UHI?????
Urban heat island effects are real but local, and have not biased the large-scale trends. A number of recent studies (see IPCC 2007 Chapter 3) indicate that effects of urbanization and land use change on the land-based temperature record are negligible (0.006ºC per decade) as far as hemispheric- and continental-scale averages are concerned because the very real but local effects are avoided or accounted for in the data sets used.
Joe has a conclusion and is looking for data to support it – not scientific method at all.

[youtube=http://www.youtube.com/watch?v=RVT6o6LUxmw&hl=en&fs=1&]

“”” bob paglee (14:36:52) :
George E. Smith 13:26;58 states:
The drop out of rural data points is critical here.
For me, the most accurate way of measuring global temperature, free of UHI effects, is to SOLELY use rural stations.
This article shows either a cunning plan since 1990 to artificially drive the ‘global warming’ agenda, or a rather fortuitous series of events which happened to make such an agenda more likely to apparently fit the hypothesis. “””
Bob, Absolutely nothing contained in the “”” — “”” above is anything I said. Those are simply cut and pastes from womeone else I was citing.
Virtually everything I post here is either something I clipped from someone else, or is my own words (including typos); and it is easy to separate them. I always use “”” — “”” as a format to refer to someone elses work. thats three double quotes followed by three spaces; then the cited text, then three spaces, and the closing three double quotes.
I do that so there is no question as to what I say (which is in plain text without quotes), and what is something I am lifting from another author.
For a large fraction of the postings at WUWT, I cannot distinguish who said what. So use Italics or some other font trick to delineate cited information; but there’s no guarantee that formatting survives everybody’s browsers.
So bottom line; if you do in fact agree with that first sentence as you say; you are agreeing with what somebody else said; not something that I said. My position on what otherws say should be apparent from what I add to their stuff that I lifted.
You can’t delineate citings from self comment by simply doing a paragraph spacing. Good writing would do that just for clarity anyway.
George

“”” George E. Smith (10:42:57) :
“”” bob paglee (14:36:52) :
George E. Smith 13:26;58 states:
The drop out of rural data points is critical here. “””
Re above.
Bob, there isn’t; and there cannot be, any “optimum” mixture of “urban” versus “rural” temperature measuring stations.
First of all remember that what GISS and Hadcrut, UAH and RSS report are NOT global temperatures; they are anomalies, which means some machinations have already been performed on whatever the thermometer siad at the time it was read.
BUT! If you wanted to measure the true average global surface Temperature; it is a simple 8th grade science project to figure out how to do that. By “surface” Temperature, I mean the Temperature at the boundary between the contiguous atmosphere, and that which is not atmosphere; usually either oceans (73%) or land (27%).
1) Divide the surface up into cells; each with some cell area.
2) Read the Temperature from a thermometer in the center of each cell.
3) Multiply the Temperature by the cell area.
4) Sum all the Temperature*Area products.
5) Divide the total of those products by the total surface area of the Earth.
The result is the true global average temperature, at that instant when all the thermometers were simultaneously read.
6) Repeat 1-5 periodically for say a whole year.
7) Take the average of all of the results from 6.
The result is the annual mean global surface temperature.
PROVIDED; that the size of the cells is such that the single temperature reading from the cell center, is truly representative of the temperature over that cell area; and provided that the process is repeated often enough to follow any cyclic variations in those temperatures.
The strict rules for adequacy of either the cell size or the reading frequency are set down in the general principles of Sampled Data System theory, and in particular the Nyquist Sampling Theorem.
Now notice I said this was an 8th grade science project; but if you read what I actually wrote; I said the project was to figure out HOW to do that; certainly NOT to actually try doing it; because the resources required to do it are actually beyond our capabilities; we don’t have enough thermometers on eqarth to actually carry out the task.
Well we don’t; but Gaia does. Every single atom or molecule, is one of Gaia’s thermometers. Each registers its local temperature directly in the form of the average energy per degree of freedom, that that atom or molecule has.
Gaia takes everyone of those into account in deciding what the climate should be; and she always gets the right answer because her sampling strategy is impeccable.
There’s no chance we can equal her performance; but we sure should be able to doa heck of a lot better than we do.
Where UHIs create problems is not that they give an incorrect temperature reading; but that the reading obtained is erroneously applied to adjacent areas for which it is not a valid temperature reading.
The fallacy that seems to go unrecognized, is that for a given set of actual real thermometers that are used ; let’s say to obtain GISStemp results; the prevailing process (AlGorythm) that is used seems to record a relatively continuous, and well behaved function; and the values cnage relatively smoothly from time to time. That lulls us into believing those results are real.
Well they actually are real results for GISStemp; whatever that is (Hansen knows).
What they are not, is a real representation of the temperature of the earth; just the values of GISStemp.
Hansen; might just as well be averaging the phone numbers in the Manhattan Telephone directory, and reporting the changes, as they from time to time republish that.
An exercise he gets paid to do; but which has no underlying scientific basis for doing.
George

John Finn (12:22:40) : “There’s nothing misleading about it. I chose 1992 because that’s round about the time of the reduced number of stations.”
Bull. It is absolutely misleading.
“There’s no “OF COURSE” about it. I accept the effect of Pinatubo will increase the trend but it’s not obvious that the UAH trend has been more affected than the Hadley trend.”
B.S. again, the response to Pinatubo is much greater in the UAH record:
http://www.woodfortrees.org/plot/hadcrut3gl/mean:12/offset:-.15/from:1991/to:1996/plot/uah/mean:12/from:1991/to:1996
“Separate argument.”
You are trying to separate the arguments so that no one can criticize your sacred cow.
Like it or not, using non cherry picked and manipulative comparisons, you are just wrong, the surface clearly has a warm bias.

When a new religion takes over, the most important thing to do is preempt or hide (destroy) all old records that might be contrary to their teachings. Is this happening to the published temperature records? I believe many of the stations, not published, are still being recorded. ( see mizen2 (03:51:19) :above) especialy in rural areas. Thank god for inertia in record keeping. Maybe these need to be privately collected and published, as they are no longer wanted by the powers that be.
Airport temperature recordings are about worthless for weather or climate record purposes.

Andrew (12:42:45) :
John Finn (12:22:40) : “There’s nothing misleading about it. I chose 1992 because that’s round about the time of the reduced number of stations.”
Bull. It is absolutely misleading.
The fact that you keep sayiing it’s misleading doesn’t make it so.
There’s no “OF COURSE” about it. I accept the effect of Pinatubo will increase the trend but it’s not obvious that the UAH trend has been more affected than the Hadley trend.”
B.S. again, the response to Pinatubo is much greater in the UAH record:
Ok let’s take it from 1994 …..
http://www.woodfortrees.org/plot/hadcrut3vgl/from:1994/offset:-.2/plot/uah/from:1994/plot/hadcrut3vgl/from:1994/offset:-.2/trend/plot/uah/from:1994/trend
Big difference – instead of the UAH trend being a couple of hundredths of a degree (per decade) warmer Hadley’s now a couple of hundredth’s warmer. Both are well within the stated error bars of each other whichever start point you use.

Scott Mandia, you post here so much in disagreement with the blog in general without bad-mouthing, which is good. But I think you need to go a lot further back to basics, further than I’ve seen you go so far, and you need to check yourself more carefully lest you merely support strawman arguments.
Re your 5 points (@ 13:31:03):
I don’t think anyone is disputing or ignoring the general agreement of the 4 temperature records that show recent rises in temperature, at least up to 2000. Concern about inadequate allowance for UHI, or about other problems with surface data, is not “ignoring surface data”. Now this “plethora of actual observed effects of rapid climate change” that you mention – I have not come across a single such harmful effect that is both relevant to today and true. May be true but irrelevant, or relevant but untrue, or relevant but harmless or beneficial, or relevant but misleading, or would be relevant only if at least one part of the basic AGW thesis of (a) unparalleled warming (b) CO2 as cause were substantially, transparently proven true. But perhaps you can enlighten me. Perhaps I’ve missed those plethora somehow. I would not like to be accused of ignoring important data.

I could not believe how awful the GISS records might be until I started working with them, doing UK blink comparators for Ellie. I think my response reactions are still too much in the expletives stage, but that will pass… I shall channel it into better science…

Scott A. Mandia (13:31:03) :
“1) All 4 data sets show a warming climate in the past three decades.”
Yes, although this really doesn’t say much, sense the 2 non satellite products are essentially analyses of the same data.
“2) They are fairly close in agreement with warming rates”
Not when one takes into account expected-and empirical-amplification
“(one cannot just throw out RSS because they do not like the results).”
No, one can however cite serious arguments that RSS is not a good data set. which I would indeed:
Christy, J.R. and W.B. Norris, 2006: Satellite and VIZ-Radiosonde intercomparisons for
diagnosis on non-climatic influences. J. Atmos. Oc. Tech., 23, 1181 – 1194.
Christy, J. R., W. B. Norris, R. W. Spencer, and J. J. Hnilo, 2007: Tropospheric
temperature change since 1979 from tropical radiosonde and satellite measurements,
J. Geophys. Res., 112, D06102, doi:10.1029/2005JD006881.
Christy, J.R. and W.B. Norris, 2009: Discontinuity issues with radiosondes and satellite
temperatures in the Australian region 1979-2006. J. Atmos. Oc. Tech., 26, 508-522,
DOI: 10.1175/2008JTECHA1126.1
Sakamoto, M. and J.R. Christy, 2009: The influences of TOVS radiance assimilation on
temperature and moisture tendencies in JRA-25 and ERA-40. J. Atmos. Oc. Tech.,
doi:10.1175/2009JTECHA1193.1.
Randall, R. M., and B. M. Herman (2007), Using Limited Time Period Trends as a Means to Determine Attribution of Discrepancies in Microwave Sounding Unit Derived Tropospheric Temperature Time Series, J. Geophys. Res., doi:10.1029/2007JD008864, in press
All of which find specific issues with RSS and similar data, and find UAH to be the more consistent data set, both internally, and with the best radiosonde data sets.
“3) Joe suggests that there is a serious enough UHI bias that we cannot use this data for policy decisions.”
It’s more than UHI but “use for policy decisions” does not have a clear meaning and I don’t find it particularly relevant. What is clear is that the surface data exaggerates the warming and therefore exaggerates the “problem” motivating policy makers.
“4) There is literature that shows UHI and no UHI but the data from point #1 doesn’t show UHI.”
See my response to point one. And again, it’s more than just UHI
“5) There is a plethora of actual observed effects of rapid climate change if we choose to ignore surface data. Shall we ignore those also?”
You miss the point. Again. Yes, there’s warming. I don’t know what the qualifier “rapid” is there for, but magnitude of the “problem” matters.
“He who refuses to do arithmetic is doomed to talk nonsense” and I tire of the qualitative defense of warming. Yawn.
John Finn (15:23:03) : You really deserve to be ignored for being incredibly thick. Look at the fluctuations. It’s clear that UAH goes up and down more. So why is the trend even slightly less?
If you misdirect with nonsense about “separate issues” you have lost and I will ignore you.
“The fact that you keep sayiing it’s misleading doesn’t make it so.”
Except that I give specific reasons. [snip ~ E].

Scott Mandia
I enjoy your comments, you are a thoughtful and courteous debater and yes I have visited your site.
Several times I have pointed out here about historic precedents and you have said that was something you didn’t know about it and would go away and have a think about that aspect. (arctic ice variations, glaciers being higher in MWP and Roman times etc) it. I never get any feedback on that and the belief that all this is happening for the first time remains unchallenged on sites such as yours.
Do any of these well documented cases of cyclic variations in our climate that predate satellite data ever give you pause for thought?
best regards
Tonyb

Scott Mandia: “Actually, the surface data does not even begin to show the warming that is here and will be coming. The oceans have taken in almost half the CO2 and about 95% of the excess heat.”
yes, and the ocean heat content has been sinking for a while now. Now SST is also going down. I wouldn’t be surprised if we’ll see a drop in land temperatures over the next few months. OHC seems to be the only reasonable measure of a warming earth. Unfortunately, records go no longer back than 1955, which means that they start at a local temperature minimum. We know that at least the Barents sea was just as warm around 1940 as it is now, so what if the global OHC was comparable to today’s values around that time? Even if it was lower: by how much? And would it mean that the difference between 1940 and today is due to AGW, or is it just a natural recovery from the LIA?

timetochooseagain (16:56:46) :
Take a look at this Lucia link and scroll down to the first graph.
http://rankexploits.com/musings/2009/adding-apples-and-oranges-to-cherry-picking/
It shows the trends from Sept of each year (1989-2004) until August 2009. Note all 5 records are pretty close over the longer term (i.e. 14+ years). Hadley and the satellites (UAH/RSS) remain in lock step right up until ~2000. There is divergence after that. This is where the magnitude of the fluctuations becomes more influential, i.e. over the very short term. Since 2004 we’ve moved from predominantly El Nino-type conditions to the La Nina cooling of 2008. If UAH and RSS measure a greater response than Hadley, the trends will differ over this period. As the period lengthens the difference will reduce.
Also I reckon there’s reasonable evidence that the earlier GISS divergence is related to the GISS arctic extrapolation.

Lucy,
Don’t you ever wonder why there are no peer-reviewed journals articles that can explain the warming of the modern climate while also showing why rapidly increasing GHGs have had little influence on recent climate change?
There are certainly plenty that show how CO2 is responsible.
What saddens me the most about posts here at WUWT is that folks here seem to believe one or more of the following so they choose to ignore the current state of climate science:
1) The majority of scientists around the world are incompetent (they do not interpret data correctly, for example), but folks here at WUWT can intepret the data better than these experts.
2) The majority of scientists around the world are afraid to publish alternate viewpoints to an AGW-forced climate despite the fact that they would become famous if they could refute AGW.
3) The majority of scientists around the world are conspiring in order to secure funding.
4) The majority of scientists around the world are just using group speak which implies that they do not think for themselves.
Instead, there are a small handful of scientists that do not fit any of the above points, rarely, if ever, publish their findings in a respected journal, and are the ones that most here are listening to.
Seems very foolish to me.

Scott Mandia: “Recovery from the LIA is not a forcing mechanism. Something has to warm up the planet just like a weak sun and volcanic eruptions likely cooled the planet during the LIA. What warmed up the climate since the LIA and did so at a rapid pace? It was not the sun, BTW.”
Well, except for the probably minuscule contribution from geothermal energy, and the actual heat release from human activity, nothing else than the sun can warm up the planet! The system of ocean currents, cloud cover and precipitation is vastly more complex than what it might seem that you suggest here. And in any case, given your simple observation of warming since LIA and “rapid pace”, how do you explain (I’m looking at the HADCRUT temperature graph now):
1) that the 1910-1940 warming was very similar in its rapid pace to the 1975-2005 warming?
2) the cooling from 1940 to 1975?
…and also:
3) the first part of the LIA recovery?
4) the onset of the medieval warm period?

Scott A. Mandia (07:13:47) :
“Perhaps similar but 1975-2005 was a steeper slope and the past three decades have been warmer than almost all of the decades since 1850. The rate of warming is increasing each decade.”
This is really stupid and I would have hope you would not be so daft, Scott.
Using the HadCrut annual data:
http://i23.photobucket.com/albums/b370/gatemaster99/warmtwice.png
that 1911-1941 and 1978-2008, thirty one year periods with essentially identical rates of change. There is no basis what so ever for you claim of either “acceleration” or the claim that the first warming of the twentieth century occurred slower than the recent warming.
Add to that the lack of warming for twelve years now, and the claim of ever faster warming is really outrageous.

Scott Mandia in reply to Andrew: “…does 12 years change the past 160 regarding trend?”
Scott, I think that after all it all boils down to whether one thinks the hockey stick is broken or not. Even if it should eventually turn out that global temperature data actually isn’t as bad as the article we’re commenting on will have it, and that Hadcrut data really tell a pretty good story, it still remains to be decided if the last 160 years represent something unusual or not. My take on this is that ice cores show that warming usually happens fast, while cooling takes a long time.
So, suppose that the hockey stick is broken and that those who claim that the MWP was at least as warm as the current warm period are right. Further, suppose that the LIA was a pretty unique event for the last couple of thousand years. Then maybe the current warming isn’t unusual at all – it’s just the sign of the planet’s oceans returning to their normal heat content after an unusually long cold spell. It’s the return to normal, to equilibrium. And it happens fast because warming does happen fast, while cooling takes a long time.

Timetochooseagain:
I just realized that you are an 18-19 year old freshman college student. I am very impressed with your intelligence. I had always assumed I was speaking to a degree-holding scientists of some sort. Wow, I am being smacked down by somebody who could be sitting in my class. I wish all of my students could be this well-read, bright, and energetic.
I wish to retract my statement “The rate of warming is increasing each decade.” As I said before, I was wrong, dumb, and daft. I admit it.
However, the data shows the past three decades have been warmer than almost all of the decades since 1850 (and if one believes the hockey sticks – the past 2,000 years) and the underlying trend has been up, up, up. There is no “blah, blah, blah” there.
I also think Figure 5.8 of the Wegman Report does NOT disqualify the various proxy records.
Some advice for you now:
According to your blog, you calim to be libertarian/conservative. Don’t you wonder that your libertarian/conservative belief system is clouding your judgement? Those of your political persuasion do not like any type of regulation by the government or other organization of personal or societal activities. Certainly if there is AGW, there will have to be quite a bit of regulation. Isn’t it possible that this idea abhors you so much that you are trying to find a way out?
When you and I chatted about Zhang and the NAO/AGW influence on global precip, you were fairly courteous in your responses. Your latest posts to me and to Finn, although including valuable information, have a very derogatory and condescending tone. People will judge you on both what you say and HOW you say it. Be careful or you may lose credibility. I will chalk this up to you having a bad week and venting on me. No problem – we all do it occasionally.
You are bright but I still think those profesional researchers in the field who are strong supporters of AGW are at least a tiny bit smarter than you. One day you may put them all to shame but it is not today.
Good luck in your college career and always be sure to learn much but also to have much fun doing so.

timetochooseagain (14:53:32) :
Finn, stuff it, your “right” in a pointless, stupid way, that means nothing. Drop this crap, because I won’t respond again.
Scott Mandia (17:25:15) :
Timetochooseagain:
I just realized that you are an 18-19 year old freshman college student. I am very impressed with your intelligence.
18-19? That surprises me – I’d have said more like 8 or 9.

I’ve often wondered if other weather stations use the same (sloppy) method for calculating the daily average temperature as the one located near me. Several years ago, after becoming suspicious of the accuracy of their temperature reports, I emailed the Ithaca Climate Report folks at Cornell University in Ithaca, NY, and asked them how they calculated their daily average temperature. I was expecting that they sampled the temperature at least once every hour. Much to my surprise, in their reply they said that the daily average is the sum of the daily high plus the daily low divided by two! In addition, all reported temperatures (raw data) are rounded off to the nearest degree F. It is easy to realize that with only 2 daily data points, that on some days their calculated daily average can be in error by several degrees. Garbage in, Garbage Out. B.T.W., Ithaca, NY has many adherents to the church of Al Gore.
http://www.nrcc.cornell.edu/climate/ithaca/