Warren Meyer, one of the first surfacestations.org volunteers, delivered Tucson today. You may remember from the “Red and Blue States” thread on CA that Tucson had the greatest positive temperature trend for any USHCN station after TOBS adjustment. I’ve been trying to find the words to describe this station, but the pictures really do a better job than I ever could.
They are so proud of this station they even had a sign made for it to hang on the chain link fence enclosure:
The complete photo essay is available at the Tucson album at http://www.surfacestations.org The satellite and aerial photo images there are telling of the environment being measured.
Besides the obvious questions like “why is it in the middle of a parking lot?” and “why would scientists who should know better allow such a bizarre siting for a USHCN climate station of record?” Then there is this burning question: “Why did they go to the trouble of installing a precision aspirated temperature sensor and then not even bother to place it at the standard observing height?”.
It appears that the Stevenson Screen serves no other purpose except as an equipment holder, as Warren reports the Stevenson Screen to be empty. Originally the standard mounting boards for the mercury max/min thermometers were mounted about 1.5 foot higher than the air inlet of the precision aspirated temperature sensor. So the lower mounting height for the precision sensor adds a positive bias.
Is there no diligence left in basic measurement?
Climate Audit 
81 Comments
Why yes. The USHCN is diligently recording warming as its boss expects.
It would be very interesting to see what
i) station moves have occurred since 1867, and
ii) how the local landscape surrounding the landscape has changed through time (eg.,how old are those building around the station?)
The second question is very relevant due to night-time heat release from the underlying surface (which is drastically different than, say, what is observed at the Central Park Station – the issue of network inhomogeneity comes in to the picture) and the warming influence of the reduced sky view factor.
Testing….I think the spam filter is eating EVERYTHING.
At least there’s no BBQ in sight. Unless they have a gathering in the parking lot.
But don’t worry, this can all be corrected by applying correction factors found by examing tea leaves.
Didn’t find a survey report among the thumbnails – the grass is watered via automatic sprinkler system?
Nothing is surprising anymore. It is fascinating that the station is located in the middle of a parking lot when the aerial photos show a large grassy area less than 100 feet NW of the site. While far from perfect, it has to be better than the current location. It’s not as if this particular location has some special historical significance. According to NCDC records the station has moved seven times since 1956. Looks like it moved to the current location in 2001 having moved previously in 1998.
The GISS plot of this site is interesting in that the mean annual tempererature dropped fully 3 deg C from 1997 to 1998. The trend since 1998 seems to have settled into a fairly stable range about 1.8 degrees below the high range seen in the early to mid 1990s. Station records show that 1997 was the last year using max-min thermometers with 1998 using “other temperature equipment” (whatever that means). It’s clear that something changed significantly in 1998 and it probably wasn’t the weather.
Are you sure it is in the parking lot – or is it Mistaken Assumption #13?
Is that true? I’m more and more depressed with almost each station record.
Anthony, did you get already some GHCN photos?
My experiments show that a concrete or tarmac car park can be
as much 1k warmer than a nieghbouring grassy area.This however is
eliminated during and after rainfall when everything is wet.
Is possible to do study of rainy day temperatures.I suspect
that they would give better picture of GW.
That’s the first thing that popped into my head: “It’s in the middle of a parking lot!!!”
If they measure pollutant levels they will no doubt notice curiously high CO levels at certain times of the day. What a climate!
Is it possible that you are on Mistaken Assumption #13?
Tucson sprints ahead in the race to the bottom of the World Temperature Measurement Tourney. Surely you must be kidding! Like Gavin, I have no concern whatsoever about composite grid cell accuracy. This station is offset by the one behind the ice rink where the Zamboni dumps all its ice shavings. Sheesh.
since the sensor has always been over pavement the anomaly method cures everything
#10. Does it rain much in Tucson?
Gerald, are you unable to see the pictures that are posted above? The station is in a fenced enclosure right in between two asphalt areas where vehicles are clearly parked. And on the fenced enclosure is a sign clearly identifying that the equipment inside the enclosure is in fact the USHCN station in question. So are you questioning whether those are actually vehicles parked next to the station? Or are you questioning whether that is actually the USHCN station?
I disagree with Jeff C. in comment 7. A gravel surface next to the parking lot, albeit less than ideal, is more representative of the downtown Tucson urban environment than a heavily irrigated grassy surface.
re 7, 17 This is what I got from the Tucson site curator via email: “I can tell you that the gravel under the instrument was placed there purposely to more accurately simulate a desert location. Grass would have seemed to be out of place for a desert.”
Even so, with all the other positive bias around, and the problem with the aspirated sensor being 1.5′ below standard observing height, and closer to the surface, the gravel probably doesn’t make much difference.
#17
Good point, I didn’t see the gravel under the station and just assumed it was a continuation of the asphalt surface surrounding the site. My mistake. That being said, it would be nice if the gravel extended more than 6 feet beyond the instrument.
Tucson: raw Giss looks like it warms at about 1.7C century. GISS adjustment is a linear
cooling from 1890 to present. Site is cooled to 1.1C/ century, slighly above the global trend of .8C/ century.
What is the date of this photo showing cars. Atmoz is claiming with an old sat image that the cars he is seeing out his window “may” be the same ones out there now. Well I want to work for the university, if they can afford to throw away or abandon $150,000+ dollars, they afford to pay me an exorbidant salary. After all if Atmoz has been looking out his window all these years and doesn’t know if any one uses them, or not and they are beside the station he is watching and you would have to drive over sidewalks to use them, how rigourous could their expectations be? LoL.
Seriuosly, when was the photo taken and is their any reason to believe that someone would abondon automobiles in such a way? I would suggest Atmoz walk by one and see if they have current license tags. That way the validity of any assumptions would be more likely.
Atmoz of the University of Arizona writes:
http://atmoz.org/blog/2007/07/24/tucson-ushcn-weather-station/
“This is not a public parking lot. The lot is hardly used. The cars and trucks seen in the (at least 3 year old) satellite image are quite possibly the same as the ones that I can see right now.”
This is pretty funny even for a climate scientist. To suggest that the cars have been in the same location for 3 years – LOL.
If it’s true (and I don’t doubt it, but anything’s possible) then “But notice that the parking lot does not connect to any of the streets! If you want to actually park in this lot, you need to drive on the sidewalk.” seems reasonable. However I do find it hard to believe that anyone leaves that many vehicles just sitting on a closed street for 3 years. And they’re not using the Stevenson Screen.
While I doubt that concrete or asphault with or without gravel is the same as hard AZ dirt (or rocky surfaces out in the ‘wilderness’) I’ll agree that we are measuring the change more so and therefore I think that’s a reasonable explanation. All in all, I’d say this isn’t a ‘bad’ location for it.
At least there’s no exhaust on it, not close to buildings, no garbage cans, people can’t touch it, etc etc.
On the other hand I don’t know if it matters, as I said earlier, “It is calming to know that we are to some unknown degree of accuracy tracking the changes in microsite air temperature due to urbanization and population growth.”
The Cars are the same cars for the past three years. And the difference
that would make would be? Its ok, because the same old cars have been
here for three years and the lot is public.. so its a university lot and
we all know that university cars.. well, they are different. I can see
my house from here!
The cars are not the issue. Wind shelter and the lack of Evapotranspriration ( spell check on isle 3 ) are the issue.
Question to a silly climate scientist. Have you ever been to a community pool in the summer time where concrete surrounds the pool and then you have a nice grass covered park outside the concrete patch?
Ever see a kid WALK on the concrete and RUN on the grass because the latter was too hot? Somebody else sell these guys a vowel cause I’m fresh out.
Actually, if you examine the photos in detail, you can see they are not the same. In the old sat picture they are all white or whitish except for one at the end that is red, and there are no cars towards N Park Avenue. In the picture above, there are cars on the N Park Avenue side, and there are silver and dark vehicles besides the white. As Steve hinted, this is pretty bad, for anybody, much less someone trying scoff at some one else’s work. A reasonable examination of the photographs would force a reasonable person to conclude that the vehicles are being used. Since Atmoz couldn’t be bothered to actually look at pictures that he choose to comment on, could not walk so far to check for current license tags (he must help with updating tree-rings), I would not stake much hope on him finding a road into the parking lot. Considering the problem that may be at Pike’s Peak in terms of actually locating the correct area, I would not hold out much hope for that group finding it either. I certainly hope Steve was exagerating or assuming that Atmoz is some sort of climatologist – comment #23.
Re #23 Atomz stated that there is nothing wrong per se about a climatology-grade site being located in a parking lot, that it doesn’t matter whether the surface is asphalt or native Arizona rock/soil – a hard hot surface is a hard hot surface. (That’s my recollection, as Atomz’ post has been removed at least for the moment.)
I struggle mightily with that line of thought. Climatology-grade sites should be designed to detect very small changes in temperature over long time frames. The typical change is about 0.15C over ten years, which is tiny.
Locating a climatology-grade site in a parking lot puts it into close proximity to human activity, and human activity generally changes on decadal time scales. Parking lots are resurfaced, nearby sidewalks are added or modified, concrete and asphalt surfaces discolor over time, cars spill oil which changes albedo of the surface, nearly trees and shrubs (planted and tended by humans) change in size, car characteristics change, parking habits change, school hours change, tall buildings are built, reflective glass windows change, grass irrigation practices change. These are ones that come to mind right away – there are undoubtedly others which I’ve missed or which involve things that won’t be invented or practiced until decades from today.
None of these makes a big difference individually, but each could modify parking lot temperature in magnitudes similar to the decadal changes being sought. It’s quite risky to assume that the sum of the variability is zero.
A climatology-grade site should have few variables, and those variables should be similar across all such climatology-grade sites, so that random variability cancels and systematic variability can be removed. A university parking lot doesn’t cut it. What’s needed are sites which are distant from human activity and distant from tree, shrub and grass changes – in short, sites which conform to the standards.
RE 27.
In one regard he is right. As long as the site was always similiarly situated, the higher
temperature would disapper in anomaly maps. The parking lot may be hotter than “normal”
but AGW folk concentrate on the TREND in warming. Simply, the hot parking lot got scrching hot.
The issue is these changes, subtle changes, new pavment here, a new building there, an AC unit installed… these undocumented changes can corrupt the record in a way that may not be detectable.
One way to test this is to look at the raw data for the best sites and the raw data for the worst sites.
That will fun.
Re #27 Indeed. So long as the site remained atop asphalt, and nothing else changed, then a long-term trend may have some meaning. The problem is that “nothing else changed” is an improbable assumption for a college parking lot surrounded by development, especially on decadal timescales.
And there are other issues, such as the effect of pavement on diurnal temperature range, which make it tricky to use the data in any comparison with other sites.
The best practice is for all sites to follow the siting standards.
What I find to tbe the msot unexpected passage from Atoms’s blog is
He spends a great dal of effort in the rest of the posting arguing that the siteing of the station will not affect the trend in temeprature data. He then shows data that demonstrates that the is a difference in trend between this site and a nearby one. However creates arguments that it doesn’t matter or that there may be different trends at different elevations or that this one is accurate and the other site trend isn’t (take your pick – one from column A and two from column B).
The most amazing thing about this is that he does not have an accurate answer at his finger tips.
Re:#30; Hmm, Atomz seems to be explaining away warming trends as being due to natural causes.
Anyway, there is vehicle access to the parking lot via the southeast corner of the lot visible via Google Earth. One can even see evidence on the concrete of their passing. It appears to be a small driveway of sorts that runs south from that corner of the lot and connects to the street.
#27 There used to be a : at the end of the address that messed up the link. The site works.
Anyway, besides finding it hard to beleive that many cars wouldn’t get used for three years… What, nobody’s ever driven over a sidewalk before? Or put in some asphault or concrete “ramps” for maint. vehicles etc. They look too new, also. Although it’s probably not a high volume lot, no. It’s not on a main street. However….
re 31
Atomz claims that the parking lot will not likely affect a trend in the measurement. However, there are different trends in the parking lot and at a nearby airport with the parking lot trend being higher. This is rather hard to reconcile with a hypothesis that the local environment has not effect on trends. Atomz has no answer for this that I see so he starts waving his arms and writing condescending prose.
Re 32
Most parking lots are entered by driving over a sidewalk. That is why it is called off-street parking.
I remember seeing this station during my days as an undergraduate at UofA (Class of 2002) and wondering as well about its placement.
When founded in 1885, the university had a single building and was located well outside of town. Since then Tucson has sprawled substantially and engulfed the entirety of the surrounding regions. As I recall, the buildings nearby are mostly 2-4 decades old, but the tall glass building to the north (thinking now of reflected sunlight) is probably the newest of the adjacent buildings.
Since the vehicles in the restricted lot are still shiny from their last wash (no desert dust build up) so I can say with confidence that they are being actively used.
re 34.
pull building permits for the area. Permits are public record. ( most on on line)
Permits will probably show Sq. ft of the building aditions.
could make a nice plot….look at what percent of the area has been permitted for new
building, over time.
Use Building Sq footage as a proxy for Urbanization.
Must read:
Click to access sf_heatwave.pdf
Note in particular the comparison of urban sites with Diablo Junction, last few pages.
#33 OOooops, Doh! I meant to say “over the curb” as in what he said, “drive on the sidewalk”, (rather than over it). Regardless, those vehicles look newer, used, and are certainly capable of going over curbs. And sidewalks 🙂
He’s clarified it in a post on his blog that he should take out the part about them not moving in 3 years and that his point was this “misc” parking lot didn’t get a lot of use. Rather immaterial, but at least he recognizes he was clear.
Anyway, I think something’s been missed here. What is the purpose of the land stations? What are we supposed to be measuring? The given purpose is to track what “the climate” is doing, correct? So let’s see if we can get there from here.
In order to track a change from normal in an area in the tenths of a degree over decades requires the entire time being tracked:
1. Instruments that are capable of doing so.
2. Instruments that are accurate and remain so.
3. Instruments located in a stable environment.
4. Instruments located where outside influences are known and accounted for.
Answers to that?
1. Not not the entire time there hasn’t.
2. Not the entire time, maybe not now.
3. Not the entire time, no; not now; climate, stable?
4. Hardly seems possible.
For the sake of argument, let’s say we do, though.
What are we trying to measure at the end of it all?
What the overall temperature of the earth is doing, increasing or decreasing or remaining stable, I believe. How are we doing it?
There is an average of 1 anomaly station every 1000 kilometers of area. Obviously, if you have 5 in some specific 1000 kilometers, 4 others don’t get measured. That may or may not be important. The point is, there are at most 1 per every 1000 km.
But what are we measuring when we do? The temperature of the air 5 feet off the ground. (Let’s not get into how many actually are, or how pristine the temp is; I’ve got a point here.) So if every 1000 kilometers we were sampling 1 point of air 5 feet off the ground in an area free from other influences with an accurate measurement, and extrapolating that was indicative of the temperature for the entire area (and that was in fact true), what would that mean? Anything?
Probably not even with 1 station, and especially not combined with every other one of the measurements (how that’s done, and how it’s combined with the SST is another issue that pollutes the meaning of any of it, even if everything else wasn’t jacked up, and of course it is, see numbers 1-4 above).
But what we’re missing? That we’re paying attention to any of that in the first place, or any of the side issues. I think those are all immaterial, and I’ll tell you why.
#35 Steve Sadlov…As, where and how you build you harvest …extreme
heatwaves…The foehn effects are enhanced by the inner and south
Californian urban aridification…Ie today come across the
phenomen of heat-bursts…A Wiki source mentions Abadan Iran June
1967 reaching 87 C!! …Checked TuTiempo which has the longest day by
day temp stats Ie seen but some days are missing in June 1967…Any
how a similar event in Portugal 1949 but only around 70 C …Here is
another straw for AGW crowd…Don tell “el Al”(Gore)…
If any one is interested: The location is
32°13’45.76″N 110°57’15.34″W
Google map
If any one is interested: The location is
32°13’45.76″N 110°57’15.34″W
That should have said “not clear” and I’m just guessing it’s ‘he’ no first name or gender-specific language has been used.
Back to my rant about what we’re missing.
We still don’t know what it is we’re actually measuring, do we? Regardless of its dubious accuracy or not. I said earlier we were measuring the temperature of “the air” 5 feet off the ground in order to track the anomalies. We’re not really. We are measuring how much heat the ground is absorbing and how much heat the ground is releasing, and how that mixes with air 5 feet up. We are basically tracking the anomalies of the ground. That is, if there are no other sources absorbing and releasing heat nearby. And if there are no other influences affecting the air itself. Pretty tall order to fill.
In the case of this station, we are really tracking how much heat the fence, asphault, and gravel absorb and give off and how that mixes with the air at the thermometer. In the case of one where the thermometer is in grass, we are measuring the ground and the grass and how that mixes with the air at the thermometer. You get the point.
We are in large part not tracking anomalies of climate, but tracking the heat absorbtion abilities of asphault, and how it mixes with air at a thermometer location. How many of the stations are at airports on asphault? Interesting question, no?
Of course, it is true that anywhere we measure we’re measuring that same thing, and it is the temperature, and it is indicative of the temperature variations around that area of the datapoint. So that’s rather obvious, but perhaps not. But it’s not just “air” free from influences we’re measuring, it’s what the surface 5 feet under the air is doing and how that’s mixing with the air.
But what we’re really missing?
Even if the air at 5 feet was not influenced at all by what’s under (and/or around) it, which is clearly impossible, it doesn’t matter: We are working with models. Think about what was said.
The location was not picked to measure the air, or measure the climate, or to get the anomalies of such things. The location was picked to have the readings not be like the climate at the measurement location, it was picked to give the readings of someplace else. What possible meaning is there to the measurements if we’re artifically trying to mimic “what it’s really like”?
So rather than getting the readings of the area we’re in, we purposely are trying to skew the measurements so they are like another location. Why not just put the damn thermometer in the freezer, or on a roof, or inside a car trunk, or bury it ten feet underground? Why not mail it to somebody in Hawaii to take the measurements? Why not put the sensor on a 100 foot pole, or put it into the gulf stream, or throw it in the ocean? Why bother taking them at all?
We’re guessing the asphault is more like what Tucson (or the desert or the mountains or some native shrubby grass) is actually “like”, so we purposely bias the thermometer to make it look more like that. So now we have another reading of what the surface of a runway is like and how it behaves.
What we have a bunch of data on how asphault covers more of the surface of the earth and what its thermal properties are in various strengths and amounts of sunlight, wind, humidity, rainfall, etc, for different locations. On purpose.
If you can change the measurements of the air by changing the underlying surface, you are measuring the surface and what it does. There is no way the surface doesn’t have its own anomalies, so that’s what we get.
42.
The anomaly method fixes everything.
Re 36 and 39
In light of record low winter temps in the Southern Hemisphere: in Brisbane, AUS, lowest low since establishment; record lows and first snow in Buenos Aires since 1918, record lows and snow twice in the past weeks in Santiago de Chile, record lows and deaths due to severe cold in Peru, etc., etc, is there anything comparable on record re: deep cold “waves/snaps” for either NH or SH?
Re 43 Steve
Please tell us that you are being sarcastic. Geoff.
I think steve mosher has in implicit sarcasm atg. The trick is guessing when he isn’t being sarcastic!
/humor
Re #46
s/atg/tag/
Why is it you don’t see the errors until 0.1 seconds after you press Submit?
Sorry, sometimes I can’t even tell when I’m being sarcastic.
There’s a difference between hard packed dirt and asphault. That is, it’s ability to store heat.
Individual grains of sand have an irregular shape. When they are pressed against each other, even in hardpack, the grains only touch each other at the ends of each grains protrusions.
The tar in asphault fills in the spaces between individual pieces of the aggregate. As a result, asphault is much more efficient at transporting heat from the surface, where the sun is heating it.
50: Also, I think the specific heat of asphalt is very high.
28: Hmmm, I’m not so sure that some heat source like asphalt or nearby buildings do not influence the anomaly. If the true temperature increases, because of, say, increased TSI, the Tmax of the asphalt probably increases more than that. If so, that could create an upward trend that is greater than the upward trend of the true temperature. The AC units are even more of a problem, since they run more the higher the temperature, and this would defintely exaggerate any true temperature effects.
These pictures raise the question of whether the warming bias in the temperature record may be intentional. I would normally assume stupidity rather than malice, but when the university Department of Atmospheric Science is responsible for recording data – how could they not know this station is not a good quality station? These people get their funding by scaring people with alarming scenarios of impending doom from global warming.
This is a major embarrassment to the University of Arizona and to the Department of Atmospheric Science. If I was a prof there, I would have been jumping up and down years ago to get this changed. This is not simple incompetence.
No, it’s publicly funded incompetence.
Steve Sadlov,
I am pretty surprised that the author of that heat wave report never even glanced down the urban/rural path in looking at factors affecting the heatwave. Even if he only brought the issue up in order to summarily dismiss it, the absence of any discussion at all is pretty surprising. Given similar findings are both high altitude rural stations and low altitude costal rural stations, isn’t there any innate curiousity that would cause him to question whether the urban influence had any factor in the changes? Especially since high nighttime temperatures and elevated dewpoints are hallmarks of urban heat island effects?
I think some of the numbering is off…. 44 talks about 42 but it’s 43. 46 talks about 43 but it’s 44
You can usually tell when Steve Mosher is being sarcastic. He’s typing. 😀 I almost always agree with him, tho. Just FYI.
As I said:
I wasn’t going to out the Atmoz person specifically, with specifics, (notice my wording in my rant) but now that I see others are picking up on it, and the person is posting on other blogs with the same info, it’s appropriate I believe.
The blog states that the person, N. Johnson from the reply, is a graduate student in the atmospheric sciences department whose current area of study is the impact of aerosols on stratocumulus microphysics. So a cloud focused climate scientist type. Who’s going to bring up the “but you don’t work in this particular field” card first? I’d hope it would be the AGW folk, since they love to do that it seems.
Of course it will introduce a change in the trend, the material is different. But nice of you to admit that the surface influences the readings, and that some sites are being purposly chosen to bias the readings a certain way. So they can then be corrected? This makes absolutely no sense.
And how about this:
Have you had time yet? And if so, did you find 1 answer?
Some numbers would be nice. But let’s see…. “I’m guessing dry dirt is like asphault” Um, no, it’s not.
Nope, we’re interested in accurate change in natural temperature over time that is free from adjustments and tweaks.
“Anyone reading this blog ever been to the Southwest?” Um, I doubt that’s possible at all.
Of course some of us have been there. Know where Sierra Vista is?
No matter. If you want to measure the “native rocky terrain of the Southwest” put the thermometer on the “native rocky terrain of the Southwest”. And here’s an interesting point; possibly the “native rocky terrain of the Southwest” in point A, may react differently than the “native rocky terrain of the Southwest” of point B or C or D or ZZ?
Re 57.
Like you Sam I almost agree with everything I say. The arguments our friend in AZ is using
are ” Anomaly method fixes everything” and ” we detect trend”
Now, there is a certain logic to it. but it breaks down for me in a few ways
1. We would want to be able to test whether variability over grass were any different than variability
over Asphalt. is siting over asphalt varience reducing ( like a noise filter)
2. We would want to know if the site had changed over time. (Chances are that a site in the middle of a grass
feild has NOT had asphalt removed, but the opposite seems less true)
3. Absolute temperature matters not just trend. TREND doesnt melt ice or expand oceans.
I don’t knw why these guys just don’t take the approach gavin did and suggest that we calculate trend using only the good sites.
Thanks. As I said in my post at Asphalt:
Putting them [the thermometers] all above the same type of surface may seem like an interesting way to make the data the same, but you’re still just measuring how the thermal characterists of the substance acts in mixing with the air 5 feet up. You’re measuring asphalt and air. Or whatever and air.
Take for example a few different locations. They aren’t all going to have the same amount of sunlight (how much the material warms, how much heat will be released), the same age or condition (and if not planned correctly or using the same material there versus new, thickness) of the material (rate of absorbtion and release, which would change depending on the amount of sunlight – say 10 hours a day versus 2, the material ages differently), the type of soil underneath and the amount of moisture in that soil (also affecting the material’s condition and properties), and the temperature of the air and the amount of water vapour in it (which is in turn affected by air windspeed and direction).
The thermometer needs to be located away from human and environmental influences, and in a place indicitive of the area.
The USHCN station and the sign have been removed. I went by the location and took a couple of pictures this morning June 25th 2008.
The unit was mounted in a location that obviously biased the readings to the high side. A couple of the vehicles do look like they have been parked there for maybe a year. There was not nearly enough dust to demonstrate a three year stay. But none of the vehicles in the picture are there.
The pan directly underneath the unit was metal filled with sand. Fully 10% of the pan was bare and none of the sand covered more than 1/4″ deep. The gravel surround was just a hastily shoveled covering with no flora. The surrounding buildings reflect a large amount of solar radiation from one side in the morning and then the other side in the afternoon. The buildings also block the natural air currents.
All in all I would say that one of three things possibly transpired here.
1. The job was doled out to some maintenance department and they were just told to move and re-install the station.
2. One of the typically well-indoctrinated but poorly educated graduate students was in charge of the relocation and installation.
or
3. One of the typically liberal and agenda driven professors did the job this way on purpose.
HAve you posted the pictures at Anthony’s website? They sound interesting.
Being in a bit of a hurry this morning I only took two photos of the missing station. The area is a short drive for me so if anyone would like a more complete photo essay of the site and surroundings please let me know. Also where do I find “Anthony’s website”?
#62
Surface Stations
Yes, a full set of photos of the “empty” site would be nice, as well as a survey of the new site, if you can find it.
I was a student there in the late 80’s, I don’t remember the weather station, I have a recollection of stuff on the roof of the Physics and Atmospheric Sciences building, but that might of been different instruments. The Gould-Simpson building to the west went in the late 80’s, and it would of made the heat worse, I do recall that building causing an oven like effect between it and the PAS building.
About the parking lot, some time in the past, that would of been a working parking lot, (you could park near your building!) Heat stroke has became a desirable property during the time I was there and parking has been consolidate towards large structures at the edges of campus. I doubt the station was in the parking lot during the 70’s.
To be fair to the people at the U of A there is no good place to get a consistent temperature measurement on campus, the place seems to exist for the benefit of building contractors.
USHCN offers “Average Mean Temperature” as well as “Average Urban Heat-Adjusted Mean Temperature” monthly data for many of its sites. They also offer the same analysis for a site’s minimum and maximum temperatures. See here , bottom portion of page, for access to such data for Tucson, AZ.
I plotted the difference between the adjusted and unadjusted data for mean, min and max at Tucson:
This plot prompts several comments as well as a simple question.
I can rationalize the strong UHI adjustments to the minimum temperature as well as the weaker, UHI opposite-sign adjustments to max temperatures. I have difficulty accepting, at first glance, the early-50s sudden increase in UHI adjustments and the circa 1990 leveling of the adjustments but perhaps there are good reasons for those. What I struggle with is the difference between “average mean” temperature and the corresponding “(min +max)/2” temperature. It seems like they should be the same. My tired brain is undoubtedly overlooking the obvious answer to this elementary question so, if anyone would kindly explain the difference, I’ll greatly appreciate it.
Re: David Smith (#66),
David, are not the mean temperatures adjusted for MMTS and TOBS while the Tmin and Tmax are not.
When I plotted the listed mean (average temperatures) for this station versus the calculated (Tmin + Tmax)/2, I got good agreement in recent times with poorer agreement going back in time.
Re: David Smith (#66),
I’m confused. The chart seems to say that UHI currently increases Tucson minimum daily temps by about 3 degrees, but also DECREASES maximum daily temp by about 1/3 degree. Is that what they really mean?
[And if so, hooray for UHI!]
If more time is spent below the (min+max)/2 value than above, this would explain it. Also other odd daily curve shapes. This can be worked out if hourly readings are recorded as well as the daily max/min.
RE #67 thanks, pjm. Perhaps that’s the case, though I’d be surprised to learn that Tucson has hourly data back to 1907.
My understanding is that USHCN uses only two values, daily min and daily max, in their temperature calculations, even if hourly data is available.
Tucson sits in the middle of the Sonora desert. I have visited Tucson since 1968 and was spending a week a year there from 1985 til 2003. One should consider the amount of vegitation that has been planted there over the years as the town grew. So depending on sta. siting direction of wind etc. this could explain the difference. If this is the station at the university there are no adjustments that could take into account all of the problems at that site.
Re # 69 Kenneth, that may be it. I suspect some kind of adjustment(s) is contained in the mean data (green curve) which is absent from the min and max data (blue and red curves). However, in the website readme I see this quote:
That seems to indicate that the adjusted data exists for min and max. If so, then why not present that data at the website.
A related puzzle for me is that the difference between the green curve and the dashed curve grows over time rather steadily. That doesn’t match how I would expect MMTS and TOBS to behave.
I may be turned the wrong way on this and thereby missing the obvious, simple explanation.
Re #70 jc, I think your interpretation is correct – the graphic suggests that UHI raises the nighttime temperature and (slightly) lowers the daytime temperature.
Re: David Smith (#72),
My calculations showed just the opposite – the difference diminishes going forward in time.
David, as I recall, the Filenet includes all the corrections save UHI, i.e. Areal, SHAP, TOBS and MMT. I do not believe the OBS and probably the MMT (not so sure here) corrections are conveniently or meaningfully applied to the maxT and minT. As I also recall, USHCN has a data set series called calculated Mean which is, I think, (Tmax +Tmin)/2 and a second series called mean temperature which I assume has the corrections noted above.
Re: Kenneth Fritsch (#73),
David, I calculated the differences for both Tucson, AZ filenet and urban adjusted temperature series for Tave – (Tmin+Tmax)/2 and I obtained, in both cases, a diminishing difference going forward in time.
The differences for (Tmin+Tmax)/2 between the filenet and urban adjusted were small compared to the differences between these two series for Tave. This result indicates to me that the Tmin and Tmax temperatures for the USHCN urban adjustment do not receive any or very little of the urban adjustment given the Tave temperature.
Re #74 Kenneth, thanks for the check on Tucson. I’ve explained my puzzle poorly and, as a result, we may be talking past each other.
I’ll try a graph:
The graph shows the difference between mean and (max+min)/2 for the UHI-adjusted data (red line) and also for the non-UHI adjusted data (blue line). As you note, both the red line and the blue line move towards zero as time progresses.
My question has to do with the difference between the blue line and the red line. That difference (blue minus red) becomes greater as time progresses. I don’t understand why. There is probably a simple and obvious explanation, but it escapes me at the moment.
Increasing urbanization creates more and more urban heat over time, which increases how much it’s removed. It’s only a few tenths of a degree, and only for the mean of the high and low hour anyway. And it’s originally measured in whole degrees anyway. Don’t worry about it. It’s not really there.
David, the differences that you see between the T ave – (Tmin + Tmin)/2 and UTave – UTmin + UTmax)/2 over time can be shown to be entirely due to the differences between (Tmin + Tmax)/2 and (UTmin + UTmax)/2 and goes from 0.02 degrees F at 1904 to 0.17 degrees F present time. Therefore, the difference that is of interest to you in your post is a difference in the minimum and maximum temperatures between the filenet and urban adjusted temperature series. This in turn indicates that an adjustment is made to the minimum and maximum temperature on going from filenet to urban adjusted, but that it is of a magnitude less than that made to the average temperatures going from filenet to urban adjusted (0.2 degrees F at 1904 to 1.4 degrees F at present time).
I need to recheck this observation but as I recall there was an additional adjustment to the urban adjusted temperature series beyond that for the UHI and it involved adjustments for missing and questionable data. That, however, does not make sense of the differences that we see going forward. I do know that when you work with the individual data series using the raw data some of these differences become more apparent. Too bad it is not explained in detail in the readme files.
If you are like me, you will want to attempt to find this difference yourself before going to the source.
Re #77 Ken, thanks. I plan to back to square one and work my way forward from the original data. It’s a learning exercise for me.
While I do that I think I’ll try to develop a step by step flowsheet to illuminate all the steps and decision points. The lowly flowchart can be quite helpful at times.
David, a flow chart would be a great tool for looking at and analyzing some of these puzzles we see with USHCN temperature series.
I think I can rationalize the differences between the (Tmin+Tmax)/2 and Tave temperatures for the progressively more adjusted USHCN series prior to the urban adjusted, but given the readme statements from USHCN below, I cannot do the same for the urban adjustment.
I base that thought on the proposition that the urban adjustments are made to the maximum and minimum temperatures as noted by the evidence shown in the graphs in Re: David Smith (#66), for the Tmin and Tmax between the filenet and urban adjusted data series. I think that the answer may be in the Karl paper on the urban adjustment.
Perhaps an adjustment made to the maximum and minimum temperatures is also applied independently to the average temperatures.
Finally, I think it is worth noting that the urban adjustments appear large for Tucson in light of some of the data sets where these adjustments are minimized in importance. It is also interesting to see the inflection points in the adjustments and the large difference and direction in the adjustment to the minimum versus maximum temperatures – as David has noted in this same graph.
David, the link and excerpt below shows that NOAA/NCDC/USHCN recognize that the calculated (Tmin +Tmax)/2 does not necessarily equal the value for Tave – otherwise why publish a Tave and calculated Tave series of temperatures. The remaining question is why they differ.
So far, I have only been able to find an abstract of the Karl et al. (1988) paper on the urban adjustment. I have also noted that this adjustment leads only to a 0.06 degree change in the US temperature trend over the 20th century.
http://ingrid.ldgo.columbia.edu/(plotaxislength)540+def/SOURCES/.NOAA/.NCDC/.USHCN/detailed_documentation.html
Houston, we have a problem. Actually it is I that has the problem as I shall detail
below.
David, I think we should not bother with the urban adjusted differences between (UTmin+UTmax)/2 and UTave and here is why.
Click to access US_Temperatures_and_Climate_Factors_since_1895.pdf
http://www.ncdc.noaa.gov/oa/climate/research/ushcn/#urbanization
The link below gives the adjusted version 2 average, minimum and maximum temperatures for the 1221 USHCN stations. I extracted these data for Tucson, AZ station number 28815 and compared the Tave to (Tmin + Tmax)/2 and they agree completely. David, they have moved on and I suspect that we should follow suit.
ftp://ftp.ncdc.noaa.gov/pub/data/ushcn/v2/monthly/
Re: Kenneth Fritsch (#81),
Now that I have the latest and greatest version of the Tucson final adjusted temperature (Version 2), I thought it would be appropriate to compare it to the urban and filenet versions from earlier times. The graph below shows that comparison.
Additionally I did regressions for these three temperature series for the time period 1895-2006 and report a summary of those calculations below. Lag 1 r is the lag 1 autocorrelation for the regression residuals. The trends are in degrees C per decade.
Ver2 Adjusted Temperature:
Adj. R^2 = 0.42; Trend = 0.117; Stdev Trend = 0.013; Lag1 r = 0.072
Urban Adjusted Temperature:
Adj. R^2 = 0.43; Trend = 0.144; Stdev Trend = 0.016; Lag1 r = 0.263
Filenet Adjusted Temperature:
Adj. R^2 = 0.64; Trend = 0.221; Stdev Trend = 0.016; Lag1 r = 0.262
Re: Kenneth Fritsch (#82),
David Smith, I would suggest in doing your flow chart as noted in your post above that you review what Steve M has previously provided at Station Data at this link: http://www.climateaudit.org/?page_id=1686
I wanted to comment in more detail in this post on my findings from above with regard to the trends that are calculated from the various USHCN temperature versions and now also including the GISS temperature data for the University of AZ station.
What I find most interesting and worthy of further discussion and analysis is the relatively large adjustment in the USCHN temperature data on going from filenet to urban adjusted. It amounts to approximately a 0.7 degree C per century adjustment or 1/3 of the temperature trend prior to the UHI adjustment. I would think that such large adjustments could be subject to large errors and/or uncertainties. Remembing also that the Karl adjustments for UHI effects for the USHCN temperatures are based on populations and considering that simplistic approach in light of what we have found for the Watts and team CRN quality findings, regardless of population considerations, and the correlations with temperature trends, one has to have some major doubts about the uncertainty of those adjustments over time.
When we compare the two different UHI adjusted USHCN temperature series, i.e. the urban adjusted and Version 2 adjusted, we see further uncertainties in the comparison of the 1.17 degrees C per century for Ver2 and 1.44 degrees C per century for the Urban adjusted. Also note the lower Stdev of the trend for the Ver 2 series and its lower lag 1 auto correlation of the regression residuals when compared to the comparable values for the urban adjusted series.
When the GISS series is thrown into the mix it appears to be approximating the filenet adjusted USHCN series and shows no adjustments for UHI. The Ver 2 USHCN minus GISS series is shown in the plot below. I find this difference in temperature series most curious and worth some investigation with other stations. The GISS series was extracted from this link:
http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=425722740020&data_set=1&num_neighbors=1
Regression values from regressing annual temperatures over the time period 1895-2006:
Ver2 Adjusted Temperature:
Adj. R^2 = 0.42; Trend = 0.117; Stdev Trend = 0.013; Lag1 r = 0.072
Urban Adjusted Temperature:
Adj. R^2 = 0.43; Trend = 0.144; Stdev Trend = 0.016; Lag1 r = 0.263
Filenet Adjusted Temperature:
Adj. R^2 = 0.64; Trend = 0.221; Stdev Trend = 0.016; Lag1 r = 0.262
GISS Temperatures:
Adj. R^2 = 0.62; Trend = 0.220; Stdev Trend = 0.016; Lag1 r = 0.185
Re: Kenneth Fritsch (#83),
I am in the process of comparing by station the USHCN temperature series Version 2 adjusted, Urban adjusted, Filenet adjusted, TOBS adjusted, Areal adjusted and GISS temperature series. I found that the default GISS temperature series that I used in the comparison in the previous post on this thread is comparable to the USHCN Filenet and that another series listed as Homogenous adjusted is probably closer to the USHCN Version 2. To show a more complete comparison I have listed the previous regressions and the new one for the GISS Homogenous adjusted series below. The trends are in degrees C per decade.
I also have included a plot of the difference between the USHCN Version 2 temperature series and the GISS Homogenous for the period 1895-2006. The differences are in degrees F. While the regressions of these two time series would seem to indicate that there are similarities in the trends over time, the difference plots shows some substantial differences from year to year.
Regression values from regressing annual temperatures over the time period 1895-2006:
Ver2 Adjusted Temperature:
Adj. R^2 = 0.42; Trend = 0.117; Stdev Trend = 0.013; Lag1 r = 0.072
Urban Adjusted Temperature:
Adj. R^2 = 0.43; Trend = 0.144; Stdev Trend = 0.016; Lag1 r = 0.263
Filenet Adjusted Temperature:
Adj. R^2 = 0.64; Trend = 0.221; Stdev Trend = 0.016; Lag1 r = 0.262
GISS Temperatures:
Adj. R^2 = 0.62; Trend = 0.220; Stdev Trend = 0.016; Lag1 r = 0.185
GISS Homogenous Adjusted Temperatures:
Adj. R^2 = 0.28; Trend = 0.11; Stdev Trend = 0.016; Lag1 r = 0.180
3 Trackbacks
[…] Efron Link to Article t-pain Tucson’s Amazing USHCN Station » Posted at Climate Audit on Sunday, […]
[…] I was wondering when someone would finally get around to posting pictures of the Tucson USHCN surface station online. That day was yesterday. It was originally posted (I’m guessing) at SurfaceStations.org, and quickly picked up by Climate Audit. […]
[…] has other issues […]