This is why you don't put an official NOAA temperature sensor over concrete

You’d think the answer would be obvious, but here we have a NOAA operated USHCN climate station of record providing a live experiment. It always helps to illustrate with photos. Today I surveyed a sewage treatment plant, one of 4 stations surveyed today (though I tried for 5) and found that for convenience, they had made a nice concrete walkway to allow servicing the Fisher-Porter rain gauge, which needs a paper punch tape replaced one a month.

Here is what you see in visible light:


Here is what the infrared camera sees:

Note that the concrete surface is around 22-24°C, while the grassy areas are between 12-19°C

This station will be rated a CRN5 by this definition from the NOAA Climate Reference Network handbook, section 2.2.1:

Class 5 (error >~= 5C) – Temperature sensor located next to/above an artificial heating source, such a building, roof top, parking lot, or concrete surface.”

Now a caveat: There had just been a light rain, and skies had been overcast, it had just started to clear and you can see some light shadows in the visible image. Had this rainfall and overcast not occurred, the differences between grass and concrete temperatures would likely be greater. Unfortunately I was unable to wait around for full sun conditions. The air temperature was 58°F (14.4°C) according to my thermometer at the time.

Here is another view which shows the NOAA sensor array, the sky, and the evidence of recent rainfall as evidenced by the wet parking lot:

Why NOAA allows installations like this I’ll never understand. And this station is a USHCN climate station of record, used in who knows how many climate studies.

I’ll tell you more on this station and others I surveyed tomorrow.



newest oldest most voted
Notify of

Well I think that adds some additional buckshot to what the Rev (and i) have been banging on about concerning the LeRoy CRN “estimates”.
Not to mention the Yilmaz (2008) and LaDochy (Dec. 2007).
And the Rev’s earlier accounting of the “Baltimore rooftop” comparisons.
A picture is worth a thousand words! (And I will indeed be adding a few thou on this particular subject!)

The color on the MMTS shield is a shade of purple that, from eyeballing the color scale, leads me to think it is 1/3 of the way from 10 to 24, or about 14.5C. When you say that “The air temperature was 58°F “, does that mean the air temperature at the MMTS, or is that a representative average of the surrounding area?
If it’s the former you could have measured the microsite bias at this station if you wanted to, but you didn’t. Take the temperatures over the concrete, and then over the grass and show us how much they changed.
If the latter, then we can reasonably conclude that the microclimate issues with this station are small.
REPLY: I don’t think you can conclude that at all and I think you miss the point. There’s a standard, (CRN1-5 or COOP 100 foot rule, take your pick) that has been ignored. There’s good reasons for such standards. Though given your defense last year of your UofA USHCN station in the parking lot, I suppose I can see why you don’t see much issue here. Your defense of the UofA station suggested you don’t think siting is important.
Unfortunately, I didn’t have a portable calibrated thermometer and portable aspirated IR shield with me, which would be required to do a proper job of air temperature measurement. The air temperature value I gave came from the vehicle thermometer, as I drove up. I’d call it anecdotal.
You aren’t the first person to ask why comparative temperatures aren’t being taken at these sites at time of surveys. The answer is that I (and volunteers) can’t do the job of a complete study when often there are only a few minutes of time at these stations. A couple of spot readings isn’t enough.
But the new datalogger and shield (see previous blog post) I’ve designed has just completed its first parallel trial at a COOP site, so we’ll see what it says when I get back and get the data from the COOP.
This really should be NOAA’s job of understanding microsite bias issues. And it wouldn’t be needed at all if exposure/placement standards were adhered to. But they aren’t paying attention or even enforcing their own 100 foot rule.
The key here is not any instantaneous air temperature, but the average of Tmax/Tmin for the day. What does that heat sink of concrete (and the environment of the sewage treatment plant) do for the Tmin which is much more capable of being affected than Tmax and how does that affect the daily average?
Its all about artificial heat sources. That concrete mass releases heat slowly. The IR shows that even the light rain didn’t do much for cooling it similarly to nearby things. We don’t know what the picture looked like before the rain (which would affect the low mass gill shield more than the concrete) so how do you know it wasn’t reading higher before? How do you know that the air temperature will read accurately in full sun with no rain? I don’t and you don’t. But the siting standard suggests there is an issue. The IR picture shows the heat sink effect clearly.
The point here is siting: don’t put a thermometer next to a heat sink like a walkway, driveway, active parking lot, or building and you don’t have to do the disentanglments in the first place.

Jennifer Marohasy



Note how weathered the concrete is. Its albedo will have changed substantially from when it was laid, introducing a spurious cooling trend to daytime temperatures and probably a warming trend to nighttime temperatures.
That picture goes a long way toward explaining the long term decrease in diurnal range seen in the temperature record.
(read it for a description of long term changes in diurnal range)




The White Knight disarmed.


You betcha there had already been a lot of cooling of the sidewalk from energy of evaporation.


Look at the temp of the galvanized post! If this is typical of most installations, a calm wind situation would have heat rising into the MMTS on any site not in the shade. That post would be warm for hours after sunset. Those posts should be white fiberglass.
What a mess!


Do you think that Al Gore, when he was Vice President, knowingly had the sidewalks installed, resulting in higher recorded temperatures in order to bolster his “Global Warming” farce.

Mike Bryant

Mike Bryant

Wondering Aloud

What I think… is that anyone who uses any of these grade 4 and 5 sites, and pretends they represent anything other than distortion and bias, is attempting to commit fraud.

Leave it to the scientists to form a consensus based on garbage. I’m sure more than one PoorHaplessDope reviews the data.

Bill in Vigo

I may be a bit confused here. I keep hearing (from some of the warmers) that just because the station has a CRN of 3,4,or 5 doesn’t really mean anything because the error would have been over a long term and wouldn’t effect the trend. I have some thoughts on this an may be confused.
1. If the average reading is elevated 3f degrees above a station with out micro-site bias and this happens at enough stations would that increase the average regional/global average? (3f is an arbitrary picked our of the air number)
2. If the average regional/global average is increased due to micro-site bias introduced in the last 40/50 years it would appear to make readings increase over that time frame.
3. If more and more sites are being upgraded with new equipment but also for convenience of operation micro-site bias is being unintentionally introduced to the sites would tend to make the regional/global average increase even more again due to micro-site bias.
My thoughts would be that the USHCN record has been compromised and I doubt that any algorithm would be able to completely remove all micro-site bias. With the introduction of more and more sites with micro-site bias that the trend would artificially be upward and would be man made due to the introduction of the bias.
Let me state that I agree that the climate has been warming for the last 100+ years. I do believe that we (humans) have caused a small part of that warming by producing heat to produce energy and using energy which produces more heat. I also believe that most of the warming is part of a cyclical rhythm that nature is in control of and that we will not be able to any time in the near future be able to change to any significant amount.
I do believe that elimination of pollution should be one of our priorities. Declaring a naturally produced gas (CO2) that is required to sustain life on this planet makes no sense to me. There are many more harmful pollutants that need to be controlled and that is where we should be putting our resources. I do believe in conservation and have changed many of my daily activities to conserve. I have changed over to mostly fluorescent lighting, I have reduced daily travel, I grow much of my food at home in the garden. I do believe that we need to develop alternate energy sources but not because of CO2 levels but for national security and economic reasons.
Perhaps I am confused but to me common sense is needed in this area and open not closed scientific study with no replication possible. Our controlling authority (NOAA) should enforce its own rules controlling the placement of weather stations. They should also insure that stations that have problems are either
A. repaired
B. relocated
C. removed from service if A and B can not be accomplished due to
urban encroachment or commercial site development.
We just need to get it right before we make a big mistake and cause more harm than the natural progression of climate would.
Just some thoughts from some one not a scientist.
Bill Derryberry

Bob_L, the metal post would not stay warm for hours after sunset. Metal’s effect is more short term. That is why it makes a poor insulator. — John M Reynolds

RE the comment by ATMOZ above, am I correct in concluding (he she) they don’t know the difference between surface temps of an object vs ambient air temps, and why the two may not be the same?

John B

I believe in the work as it’s been an eye opener for me. I agree with the concept that many of these sites are contaminated (badly) and that they may not be providing accurate results. I also agree that without visual evidence and exhaustive study how would someone be able to adjust a station back to where it should be (70% + of the US stations receive adjustments and who knows if they aren’t just adjusting the few good stations up to match the contaminated stations). Some of these stations have seen a great deal of change around them (Baltimore rooftop, U of A parking lot), but my question is about those that are contaminated that haven’t had a lot change around them.
I look at the CRN Rating Pie Chart and you can see the percentage of stations suffering from 1-5+ degree differentials, but is this differential a constant or is it a delta over the course of a century? In other words, a station like what we see above, is it 5 degrees off of average every year or is it a site that would suffer from a 5 degree temperature shift over a century? If it is the former, then it doesn’t matter as the delta over time would not change. I have had someone ask this in a conversation I had while explaining your project.

Harold Vance

The response by Atmoz immediately popped this question into my head: Where have all the skeptics gone?

steven mosher

Faced with the evidence of siting violations you have THREE options.
1. Ignore them, wave your arms, say the warming bias will be balanced by the cooling bias. This is Jones. This is Hansen.
2. Quantify them and adjust. In the same way Karl quantified the TOBS bias, the microsite bias could be evaluated. This is hard work fellows, but a person could get a PHD proposing an approach.
3. Eliminate the class5 from official datasets. Do this at the Quality control
level of data processing.
I prefer #3. I prefer it because #1 is pure bullshit and #2 will take time.
If your standard says, dont collect temps over concrete, then………DUHHHHH
dont collect temps over concrete.


we can reasonably conclude that the microclimate issues with this station are small.
Yilmaz et alia (2008) would argue otherwise.
Jennifer Marohasy: I confess that have been a bit of a fan ever since that writeup in the Australian. I have tried to locate the negative feedback info from the AquaSat that you cite but I have come p sompletely empty. In fact the sites covering Aqua seem to avoid listing results entirely (a cynical individual might almost be led to wonder why this should be the case). I have seen many references to the negative feedback/homeostasis conclusions, but they all lead back to that same article–no independent confirmation whatever. So if you could put up any links that would help us here out in this regard, we’d be very happy! Keep up the good work (you have the support of at least one liberal).
What I think… is that anyone who uses any of these grade 4 and 5 sites, and pretends they represent anything other than distortion and bias, is attempting to commit fraud.
Think what you like. Who Decides? It is we, the laymen who decide. The expert witnesses merely inform; it is the laymen jurors who determine the policy verdict.
I will further observe that if the “experts” (supported by our taxes) had been doing their jobs properly in the first place, microsite issues would never have arisen, and if they had, they would be scrupulously adjusted for, and the adjustments would be well studied/calculated/confirmed, not merely “estimated”. Hint: “Yilmaz”–a name to remember.


Note how weathered the concrete is.
However, the darkness of the concrete isn’t from weathering. It’s wet! That’s a cooling bias.
According to the Yilmaz measurements, tempertures 2 meters over concrete or asphalt are radically higher. This would confirm the the LeRoy “estimates”.


(Energizer Bunny)Hrumph, hrumph … a minor effect … washed out by the law of large numbers … hrumph, hrumph. (/Energizer Bunny)


ATMOZ tries to conclude that the sensor shield temperature is representative of the air temperature. The sensor shield is doing its job; it is not changing the temperature of the air flowing across it and onto the sensor inside. The concrete walkway, on the other hand is warming the air that is flowing across the sensor.
Anthony, I disagree with you that ATMOZ defense of the U of A site in the parking lot shows he believes siting is not important. Instead, it shows he is willing to go to great lengths to argue rhetorically to diminish the importance and relevance of your work on the surface stations project. It is consistent with his little April Fools joke that was intended to harass you by drawing you into a situation where his friends such as TCO would then extend you written abuse. It is also consistent with when he was masquerading about the internet as a climate scientist when in fact it turns out he is just a student. He is obviously polarized and has no credibility with me, which is why I no longer visit his blog.

Bill P

The pictures do speak volumes. I agree with the reader who notes the post temperature problem. Something needs to be addressed there. Has there been a gradual “creep” of these UHI / methods / materials problems, or was the system poorly-designed from the beginning? Regardless, it looks like it’ll be hard to do patchwork fixes after the visibility of your study.

Jerker Andersson

Interesting and very revealing infra red picture.
Now, how much does it actually raise temperature?
It would be intersting to actually have two temperature sensors for experimental use and then put one in similar conditions as above and the other over grass at a safe distance. When both show stable temperature the concrete piece is moved away and the sensors are read again to see what actual impact it has.
REPLY: I agree and I hope to accomplish such an experiment.

J. Peden

Have I finally gone over the edge in thinking that a site involving manicured, green grass can’t be very desireable either?


My question is why is the media not reporting this .There is odviously a problem with the way these temps are recorded. It’s as though they want the temps to be warm.We already know the media is bias on this issue.Maybe in a couple years we could start having cold day,like earth day,Oh wait that would mean the earth may accualty be cooling and that could spell the end the global warming .

Pierre Gosselin

We’ve got 2 sunspots!
Cycle 24, as they are not near the equator?

Pierre Gosselin

John B, I had the same thoughts, but I have since realized that there are other factors. I will use the site in this post as an example. On a year that had more clouds during the day, the bias of a nearby slab of concrete would be less. Similarly, if a high pressure system stalls over the area for months, then the bias would be much more due to the lack of clouds. The bias is just not realiable. That is the reason the standards were created. — John M Reynolds


My thoughts would be that the USHCN record has been compromised and I doubt that any algorithm would be able to completely remove all micro-site bias.
Considering the deplorable state of the station histories, I would be forced to agree. It’s obvious that if the microsite violations are not being adjusted for, they never would have been recorded in the first place. (Giving NOAA the benefit of the doubt.)

Pierre Gosselin, they have the magnetism of solar cycle 23:
John M Reynolds


Has there been a gradual “creep” of these UHI / methods / materials problems, or was the system poorly-designed from the beginning?
If UHI is calculated from surrounding CRN4 & 5rural stations, it is patently obvious that the UHI is woefully lowballed. To me that’s a “D-uh” conclusion.


It would be intersting to actually have two temperature sensors for experimental use and then put one in similar conditions as above and the other over grass at a safe distance.
See Yilmaz et al. (2008 ).


However, the darkness of the concrete isn’t from weathering. It’s wet
Admitedly it’s hard to tell from a picture, but the albedo of concrete is known to change over time as it weathers (from higher to lower albedo). This would introduce a trend to temperature measurement nearby. How large a trend, is anyone’s guess. says the news sunspot(s) have cycle 23 polarity.

Pierre Gosselin

Is CO2 headed down?
Here’s an interesting report from the European Institute for Climate and Energy in Jena, Germany
Quick translation into English:
Is the trend of constantly rising CO2 now ending?
Scientists at the European Institute for Climate and Energy in Jena have recently observed signals of a CO2 trend change, delayed, but similar to the current methane trend. Methane hasn’t risen in a long time. E.G. Beck writes about this in his paper (see file): “During the last two years, CO2 data from the measuring stations of the WDCGG (World Data Center of Greenhouse Gases) and the NOAA now show a change in trends, especially for the northern hemisphere.”
Atmospheric physicist Dr. Borchert, who has worked closely for years on the Svensmark Effect; the influence of cosmic rays on cloud formation, has reached the same conclusion. Taken by aback by the stagnating trend in methane concentrations seen for years, he has also studied recorded data from the major CO2 measuring stations worldwide. He has reached identical results.
It matches up excellently with the report from the AWI, which states the oceans have again started to cool down. Should we be indeed witnessing a change in trends, then temperatures begin to fall, then methane, and then the CO2? The near future will be suspenseful.
Edited by Michael Limburg

Pierre Gosselin

And the sunspots have suddenly returned – cycle 24


I just went to ,sun cycle 24, web site. A new sunspot has formed just north of the equator but they say it ‘t with cycle 23 magnestism.


Hmm… What about the temperature of the pole holding the MMTS? I notice it’s a bit warmer than the MMTS itself and of course is directly below it. Has anyone examined measured temperature of an MMTS with a pole vs without? Stevenson Screens didn’t have such a heat source directly below them. Could this be a reason why a switch from Stevenson Screen to MMTS might show a slight jump in average temperature?

Maynard Lay

Thank you, Thank you for the work you are doing. I have a comment along the line of measurement integrity of instrumentation aside from the terrible conditions in which the instruments are placed. Most of my professional life work was in the field of Metrology (science of Measurement) – NOT METEOROLOGY. In the last years of my employment, I was the chief scientist of a large military organization responsible for the calibration and repair of test and measurement equipment used throughout the US Air Force. (Something like 1.2 million instruments) We had responsibility for the oversight of over 150 calibration labs worldwide, and the Primary Standards Laboratory. When measurement audits were conducted of the labs, I was always amazed how far off of the true value some of the measurements were. That was in spite of fairly uniform training of personnel, calibrations standards used, and a most benign environment in which to make their measurements. Further, their traceability was checked to make sure that all certificates of measurements for their standards and cal lab instruments were on hand and current.
Further, the act of making scientifically valid measurements requires a generally accepted process or procedure, properly calibrated and certified equipment and standards, qualified personnel, an error analysis of the process, and proper statistical treatment, handling, and interpretation of the data.
That being said, I see no way on this God’s Green Earth that the temperature measurements going back over 100 years could be expected to have any validity. The measurements from NOAA and other sources fail almost every criteria for accurate measurements that I have mentioned above. It looks to me like most of the temperature measurements made are pure garbage considering the conditions under which they are made, both from the standpoint of environment and instrumentation integrity.
I believe that the Global Warming theory will turn out to be far more hoax than fact as time passes, and the truth comes to light.
That will be extremely unfortunate as I believe we need to conserve carbon resources for future generations. We may have already passed the peak of the ability of the world to produce petroleum. We have a longer lasting supply of coal, but those reserves will diminish all too quickly. The US and the world definitely need to find alternative sources of energy. Ethanol, wind power, sea waves, geothermal, and solar panels just will not get the job done. Nuclear energy is some help, but not the total answer.
We need to conserve petroleum for the things it alone can provide such as chemicals, medicines, lubricants, plastics, and a host of other products.
Hope you find these comments of interest. I have not found anyone yet who commented on the total lack of measurement integrity of NOAA and other temperature measurements made around the world. Everyone knows the old cliche: Garbage In – Garbage Out. And that is what we have. I just do not think there is any way anyone can go back and retrieve measurements that mean anything with one possible exception.
I do believe the Satellite temperature and radiation measurement intruments are more trustworthy than almost anything else for their purpose. From my understanding of their design, they are almost continuously compared with a stable (once calibrated) on-board reference standard. And from what I have read, their results almost always refute global warming claims.
Big M
REPLY: An error here, and error there and pretty soon you’re talking real garbage. 😉
Sat measurements show a warming trend, a bit less than that of the surface trend. But the question is, what is the on-board reference calibrated to? What is the science and error band behind it?


Admitedly it’s hard to tell from a picture, but the albedo of concrete is known to change over time as it weathers (from higher to lower albedo). This would introduce a trend to temperature measurement nearby. How large a trend, is anyone’s guess.
True enough. But that would be a warming trend, not cooling, as asphalt has an even worse effect than concrete. (I think.)
La Dochy (Dec. 2007) says that not only offset, but also trends [SIC] are increased by UHI heat sink effect!
And Yilmaz rousingly vindicates the LaRoy “estimates” (as does our humble Rev with his “HOT-L Baltimore” observations).

Have you located an example of an “ideal” weather monitoring station and published a picture of that?
REPLY: Corvallis, OR, operated by the soon to be retired George Taylor, the OFFICIAL STATE CLIMATOLOGIST OF OREGON, is a good example of a CRN1 station.
Karma is going to come back and bite that idiot Governor Ted Kulongoski of Oregon on the ass.

I wonder if the bias of small items like the pole would be small enough not to register. I thought the thermometers had a single decimal place as its last significant, and thus most error prone, digit. i do get a chuckle when GISS tries to claim that the temp anomoly has changed by 0.67 C. Perhaps this pole test would be another good test. Does anyone have 3 MMTS sensors? And do the specs have anything to say about the material of the mounting pole?
John M Reynolds



Notice the concrete is still wet and it was 22-24c. Had it been dry it would be interesting to see how much of a difference there would have been between the concrete and the grass. Why would they change the way they have recorded temps for years at there stations unless they’re wanting to get a different reading.If you’ll notice concrete turns darker as it ages too.

David S

On a slightly different subject, The International Climate Science Coalition has released the list of signatories to the “Manhattan Declaration on Climate Change”
The text of the declaration is at;
I was happy to see one Anthony Watts among the signatories. Although I was a bit disappointed that Prof Lindzen was not among them.


Jennifer: RE: Australian article: Could be a watershed.

Jerker Andersson

J. Peden
Indeed, grass is probably not the best reference either, but some kind of reference area is needed still.
Temperature sensors meassures local temperatures. With local i mean it often meassures the area close to the sensor, not what it is 10-1000m away.
What would an optimal senosor position be? It very much depends on where it is.
In a small city that is surronded by a forrest the sensor should be in the forrest. In agricultural areas it should be placed out in the fields.
Now considering how much of the earths surface that is covered with forests and deserts, how many sensors are placed in the middle of those? Not many I would guess.
Also as jmrSudbury mentioned above, even the material for the mounting pole could affect the readings if it is not well defined.

It is so rare that information on the Internet is not derivative. Here is an original story, with insight and interpretation (not to mention legwork.) My highest regard for some push-back against the corporate media who abuse science to advance taxation agendas.


True enough. But that would be a warming trend
I’m no expert on concrete, but I understand that the albedo of most concrete will decrease over time. Decreasing albedo will turn a heat reflector into a heat sink. So the daytime trend (as albedo decreases) would be cooling and the nighttime trend probably warming. Of course there would be a warming step change when the concrete is first laid.

David S

Anthony may I ask what type of equipment you are using to take those infra red photos? I have a use for something like that (not climate related).
REPLY: See this