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.
This is an interesting post – does anyone know when the sidewalk was poured? Maybe at one time the station conformed to 100 foot rule and, as stated, it appears the sidewalk was poured for convenience of maintenance. This means the station was already there. Reasonable. The person who installed the station may have been aware and the person who installed the sidewalk unaware. Just a thought…
Mars
the daytime trend (as albedo decreases) would be cooling and the nighttime trend probably warming.
According to Yilmaz (and LaDochy) you don’t get cooling at TMax with a heat sink. You get warming. (And even more warming at TMin.) A heat sink does NOT suck up heat, leaving cooler air above it! It retains the heat and bleeds it out joules at all times.
Have I finally gone over the edge in thinking that a site involving manicured, green grass can’t be very desireable either?
Grass definitely absorbs/reflects/exudes less heat than either soil or concrete/asphalt. Both at surface and 2 meters above the surface (Yilmaz, 2008 ). The figures are actually quite staggering: Grass-concrete/asphalt: at surface >11°C, 2 m above surface >7.5°C (Conditions: 24 hour average, during summer, masked by UHI).
is this in Arkansas? I think I’ve been there…? That would also explain the Al gore/clinton reference. Seriously where is this?
REPLY: Fayetteville NC, Wastewater treatment facility
jerker and Evan, thanks for the responses on grass and problems with siting in general concerning just what any sensor is actually measuring. Even apart from the concrete and post, to me that sensor just kinda looks out of place – like it might sometimes be in the midst of people having a picnic or playing golf or football, or something, instead of being a scientific station protected from immediate human influences as much as possible.
Just a thought but it appears from the picture that the sensor is placed on a slight elevation between two parking lots. I can wonder if that might make the site even more biased especially for the Tmin on calm evenings with the rising of the heat from both parking lots following the contour of the site. I believe that the side walk isn’t the only problem but the parking lots on both sides. To me it appears that the sensor might be placed on a containment levy with asphalt to either side and concrete below. Not good.
Bill Derryberry
oh you want them to walk on the grass huh? and maybe get some mud on their shoes if it’s raining. or worst wet grass
you’re a cruel person
discipline yourself
i say they should also build shaded walkways with boze speakers and soothing music with a water cooler, coffee, donuts and a nice leather couch from ikea for days when they need rest
Once again, zero proof of bad data from the station, zero proof of any bias from the data, zero proof of heat effect on the sensor.
Evan, I don’t think we are disagreeing.
My point was that weathering of the concrete over time will introduce a spurious trend to the temperature data.
This trend would be independent of the effect of the concrete relative to grass. So, the concrete probably introduces a warming bias to the station, and in addition the size of that bias will change over time as the concrete weathers resulting in a spurious trend.
You may well be right and the trend is a warming trend. Although, I would add that the effect would vary between day and night, and probably between seasons as well. And only a well structured experiment would tell us the answer with confidence.
BTW, my view is, it’s impossible to separate out these different effects and a station like this should be used. I.e. just use the good stations.
Chas Bee:
If you don’t think asphalt or concrete affects readings in the air well above it, read Yilmaz, et al. (2008). Short, but sweet.
http://www.ejournal.unam.mx/atm/Vol21-2/ATM002100202.pdf
I think you will it will fry your socks off, then boil your head off.
You are right, we do basically agree.
Although, I would add that the effect would vary between day and night, and probably between seasons as well. And only a well structured experiment would tell us the answer with confidence.
Check this out (Yilmaz). it has a 24-hour graph.
http://www.ejournal.unam.mx/atm/Vol21-2/ATM002100202.pdf
There is a big difference at all hours, but especially at TMax (which I did not expect).
Chas Bee: I think you should follow that link. I suspect you will find the results quite surprising.
Jerker Andersson wrote, “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.”
The trouble is with sorting out the real temperature trend. Conditions are too inconsistent in a forest. The sites have been planned for consistent results. As you point out, it will not give an accurate average global temperature, but it was the best plan they had until satellites came around. Really, they don’t need an accurate 1.5m above ground global temperature. What they need is to see how the temperature 1.5m above ground changes over time. Inconsistent shade in a forest would produce confusing and difficult to interpret results.
Then again, enthalpy might be a better measure than temperature anyway.
John M Reynolds
Evan
the study focused on 15 cloudless windless days. The difference seen is not typical of the whole year. Read the last two paragraphs of the study.
That said, we all know concrete is a no-no, but its effect, as the study notes, and as I have repeatedly said, is modulated by other conditions, clouds and wind being two important ones.
So when looking for microsite bias you really have to have some other data,
like wind conditions and cloud conditions to find the bias signal reliably.
What needs to be done is a full year-round study on the issue. I would certainly welcome it.
In the meantime it seems that LeRoy may well have had his head screwed on straight.
And don’t forget the Rev’s “HOT-L Baltimore” comparisons. That seems to support LeRoy’s estimates, as well.
(How do I put this gently?)
Who f*#&ing cares?
The world aint perfect and so long as you understand and control for the variations ….. it doesn’t bl**dy matter.
What do you want? Every measurement to be taken from a some parallel universe where there are no external influences? Are you on crack? Externalities are what you are trying to measure.
Please, get a grip.
JM… I care, and so do many others. Apparently, the NCDC also cares. Besides, this IS about controlling variations. You must have missed something. No one here has used harsh language, why have you? Why won’t you use your real
name? I don’t know where you’re from, but where I live you would be considered very impolite.
Thanks,
Mike Bryant
Mike
Because this whole insane, twisted project is based on the idea that each and every site that isn’t completely perfect against some unspecified standard, can be excluded for totally arbitrary reasons that are different every time.
So long as the physical conditions at a site don’t change, the physical conditions don’t matter. We’re looking at changes, right? So why do have to start with nonsense of “oh, those measurements have to be excluded, there’s concrete there”
It’s rubbish, just absolute rubbish.
REPLY: Well one man’s rubbish is another man’s treasure. The fact that the director of NCDC invited me to present the finding of this project in two days of meetings in Asheville on April 23rd seems to suggest that it may be “treasure” to them.
The metadata we are gathering can help verify a proper adjustment. NCDC wants me to continue the project to collect more metadata so that they can get a better feel about how well the new USHCN2 methodology is working towards removing step changes introduced when a thermometer is placed next to such biases. These aren’t being caught in many cases right now, and the undocumented step change can produce an increased trend.
For example, Lampasas, TX, a move resulted in a warming spike, USHCN1 adjustments have not caught or removed it from the climate record:
http://wattsupwiththat.wordpress.com/2008/02/14/how-not-to-measure-temperature-part-51/
The constant bias can also result in an offset that will boost Tmax and may create situations where long standing records are broken. Baltimore USHCN is a good example. NWS voluntarily closed this station down because it was creating a series of false Tmax readings.
See this:
http://wattsupwiththat.wordpress.com/2008/01/23/how-not-to-measure-temperature-part-48-noaa-admits-to-error-with-baltimores-rooftop-ushcn-station/
The fact that NWS chose to close the Baltimore station due to the errors introduced by it’s siting shows clearly that this study is not rubbish. I think I’ll go with the opinion of NOAA/NWS and NCDC on this issue over yours. But if you wish to continue to defend poor quality control in measurement environments, be my guest.
Oh and by the way.
Could you please answer this question: Why should a measurement station be excluded because there’s a concrete path next to it?
You guys have really jumped the shark this time.
REPLY: Actually I’m sorry to say that it is you that are incorrect sir. There are siting standards for weather stations, the most basic being the “100 foot rule from NOAA”, and then there’s the CRN rating system. The idea here is to keep bias influences away from the thermometer. In this case, the concrete will act as a heat sink and provide a bias to the measurement.
Constant biases will not affect trends if the data is processed properly. A constant bias of this type will affect the offset, and may result in higher Tmax, or Tmin. But, undocumented change points when the sensor and concrete were placed in proximity with each other cause an offset and resultant increase in trend slope. This is what the new USHCN2 methodology is going to to; catch and remove such steps so that the trend is truly correct.
Even so, biases such as concrete can still create false readings, such as assistant in breaking new record highs for a particular day, for example. A site like this can cause readings five degrees or higher warmer due to the concrete influence. For example, see this paper by Yilmaz which shows a 7.5 degree difference between Asphalt/Concrete and grass:
http://www.ejournal.unam.mx/atm/Vol21-2/ATM002100202.pdf
I’ll also point out that I presented this photo of the USHCN station in Fayetteville both in visible and IR exactly as you see it in this blog to the staff at the National Climatic Data Center in Asheville, NC on April 23rd, 2008. In the meeting were Dr. Thomas Karl, director of NCDC, Sharon LeDuc, assistant director, Dr. Gary Baker, chief scientist of the Climate Reference Network (CRN) among others.
Nobody stood up or raised a hand to protest or came up afterwards to say “you’ve jumped the shark” or even suggested “that image exaggerates the problem”. They were in full agreement that the problem I identified was real. The applauded the use of the IR camera and plan to purchase some for their own use.
So yes, this is why you don’t put an official thermometer over concrete. Sorry, but you are the one with the problem not us.
How long’s the path been there then? When was it installed? Was there an earlier version such as ashphelt?
Unless you can answer these questions you don’t have a case.
Another way of doing it is the filters that Andrew stupidly applied to the whole record about a week ago. They’re good for finding this stuff in individual sites, but completely wacky for the combined series.
And what really gets me is that when the GISS guys do apply corrections, Andrew goes nuts and accuses them of manipulating the data.
You can’t have it both ways.
REPLY: My biggest complaint with GISS is that they don’t catch these undocumented change points either, and in some cases, the adjustments they apply (homogenization) make the problem even worse. See the Miami and Lampasas examples on this blog showing how the GISS adjustments used the present spikes as hinge points and made the past cooler, resulting in an increased (and exaggerated) temperature trend for the last century at these stations. The problem is not unique to these two, but is happening at other stations also.
“Another way of doing it is the filters that Andrew stupidly applied to the whole record about a week ago. ” Sorry you’ll have to enlighten me about what flter you are referring to. I don’t recall dong any analysis about a week ago that was presented here. A week ago I was in transit to North Carolina for the NCDC meeting.
Sorry, not Andrew, Anthony
But my main point is still. We’re looking at the delta’s not the absolute values.
REPLY: The deltas are affected by undocumented change points. If you add a change to the environment, or move the sensor, resulting in a new warm or cold bias, then there will be a step offset in the record. That step offset (if not caught and corrected) will change the trend.
You see the problem is that NCDC doesn’t know the environment has changed, they often don’t get detailed information about the measurement environment such as we provide. When I first started this project I thought they’d know, but now it is clear that they don’t.
As far as “throwing out” these sites, NOAA/NCDC has recognized the problem and already decided to do so, and is modernizing the network to remove such biases in the future:
see the press release they provided me on the day of my visit:
http://wattsupwiththat.files.wordpress.com/2008/04/press_release_042408_climatereferencenetwork.pdf
So long as the physical conditions at a site don’t change, the physical conditions don’t matter.
That’s what I used to think. But LaDochy et al (Dec. 2007) finds that the presence of a constant heat sink causes an actual increase in trend (sic).
Could you please answer this question: Why should a measurement station be excluded because there’s a concrete path next to it?
According to Yilmaz, 2008 The mean temperature during summer 2 meters over concrete is 7.5 degrees C warmer than over grass. 8C warmer using Tmax+Timin average, which is the way they measure temperatures. You would be very surprised at the effect of nearby concrete. (If you want to warm a greenhouse, all you have to do is put a big rock in it.)
Also, the NOAA has not adjusted for microsite violations. therefore whatever offsets are introduced usually wind up getting counted as trends.
It’s hard, I don’t know who I’m talking to here, but.
“Even so, biases such as concrete can still create false readings, such as assistant in breaking new record highs for a particular day”
So concrete appears and disappears from day to day does it? Interesting property. I wasn’t aware of it.
REPLY: You’re talking to the blog owner, Anthony. I never claimed that it disappears, but sensors do get moved to/from fairly regularly. For example in my last “doomed journey” I located two stations where the sensors had been moved to/from large concrete and asphalt biases in the last 6 months. One resulted in a cooler environment, the other a warmer.
And you totally miss the point or maybe you just don’t want to see it. A sensor over concrete/asphalt can read 5-7 degrees C warmer than the surrounding area. So let’s say we get a particularly hot day. and the record high for that day is 35C The air temperature reaches a max of 33C but over concrete that sensor reads 33C + 5C = 38C a new record high.
How much do you want to bet on the fact that the new record high makes the newspaper, and gets included in some climate change study looking at frequency of record high temps. yet it is entirely false.
See here’s the thing, I don’t need to convince you, the people that really matter, the director and staff at NCDC, have already been convinced by the work of surfacestations.org. You’re just another bad mannered anonymous angry poster that want to see the project shut down because they see it as threatening, which they mask by calling it “stupid”.
The project will continue, regardless of such negative opinion, since NCDC has shown they want to work with us. If you don’t like it or remain unconvinced, tough noogies.
“Sorry you’ll have to enlighten me about what flter you are referring to. ”
Sorry, I misattributed that. I was talking about Basil Copeland’s post ” To Tell the Truth: Will the Real Global Average Temperature Trend Please Rise? Part 2″ 13/3/2008
In that post he uses the “Chow” test to show several breaks in the global record on a monthly basis since 1978/9.
He says :
‘a “structural break” doesn’t necessarily prove anything. It merely suggests that things were different in some way before the “break” than afterward.’
So something has to be different before and after the break, hey? Good, I agree.
Basil’s analysis shows breaks at about (he’s not very clear) 1997, 1999 and 2002 using monthly data.
So what changed on a global scale (because he’s using global data) on those dates? What physically happened there? What justifies the quite substantial discontinuous breaks Basil finds, literally from one month to the next?
Nothing. This sort of stuff would insult the intelligence of an idiot.
On the other hand, if you, or Basil, can explain Basil’s conclusions in real physical terms rather than the rubbish outcome of poorly applied statistics – I’ll listen to you.
REPLY:
I can’t speak for Basil here, perhaps he’ll chime in.
“The deltas are affected by undocumented change points. ”
Only on an individual station basis, and the change points are very infrequent. Concrete doesn’t come and go like rain.
More to the point, GISS and HAD correct for this stuff. Your doomed journey to look at the details doesn’t undermine the basics.
REPLY: “concrete doesn’t come and go like rain” No it doesn’t need to, only ONE significant undocumented change point can screw up the entire series.
“GISS and HAD correct for this stuff”: I beg to differ, GISS’ own graphs show otherwise that they fail to remove such undocumented change points.
See Miami, AZ GISS graph showing unadjusted and adjusted data overlaid, the present is used for a hingepoint, making the past cooler. Only problem is that the present is terribly biased by the location:
Graph:
http://wattsupwiththat.files.wordpress.com/2008/02/miami_az_giss_raw-homogen520.png
location:
http://wattsupwiththat.wordpress.com/2008/02/17/how-not-to-measure-temperature-part-52-another-ufa-sighted-in-arizona/
The same thing happens in Lampasas, TX
graph:
http://wattsupwiththat.files.wordpress.com/2008/02/lampasas_giss_rawhomogen.png
location:
http://wattsupwiththat.wordpress.com/2008/02/14/how-not-to-measure-temperature-part-51/
“doomed journey” Hmmm…wishful thinking perhaps? I arrived home safely, and NCDC is asking for more metadata (photos) like we’ve been doing so far. I’d say the fact that I completed the journey safely, got 20 more stations surveyed, and got the full attention of NCDC would count as a success.
Why not simply admit you can’t stand the idea of citizen scientists doing the work that professionals should be doing in the first place?