What is the red dot?

A simple question; what is that red dot on the map? I was looking at the CONUS map browser depicting the 2008 temperature departure from normal provided by NOAA’s High Plains Regional Climate Center and noticed something odd:

last12mtdeptus-shaded-520

Click for a larger image

Note the red dot in Arizona, which is the only one in the USA. Truly an anomaly. At first I thought it might be University of Arizona Tucson and its famous parking lot station, but that is further southeast.

The other map depiction HPRCC offers also shows it, and narrows it to a single data point:

last12mtdeptusd-520

Click for a larger image

HPRCC allows us to zoom in to the regional level to get a better handle on the location:

last12mtdeptwrcc-swd-520

Unfortunately, I have not found any tools on the HPRCC website that will identify this station ID. I can narrow down the location to Pinal County Arizona, and using some crude graphical tools I can approximate the lat/lon of the red dot to be : 32.9, -111.4. This puts it near the town of Florence, AZ.

Doing a search in NCDC’s MMS database for all stations in Pinal county, I find that there is indeed a COOP station #23027  in Florence, and more importantly, it is part of the “A” sub-network, which makes it a climate reporting station.

According the NCDC MMS database the lat lon for Florence COOP station is 33.0363,-111.388 so it is not very far away from my crude lat/lon estimate as seen in this Google Earth view:

florence-az-google-earth

Further searching the NCDC MMS database tells me that the station is “current” and that the station has an MMTS temperature unit equipped with a newer NIMBUS LCD display, and a standard rain gauge.

Using the Location tab of the NCDC MMS database I find the station is located at:

Location Description: 1206 MAIN STREET WITHIN AND 0.1 MI NW OF PO AT FLORENCE, AZ

Prior to that it was located at: 1707 S WILLOW ST, WITHIN AND 0.4 MI SW OF PO AT FLORENCE, AZ

So, I put that address into Google Web Search and found this in the FCC database for a tower registration:

1206 Main St (Lat: 33.020056 Lon: -111.384000), Call Sign: WRA544

Assigned Frequencies : 155.475 MHz

Grant Date: 04/19/1999, Expiration Date: 07/05/2004, Cancelation Date: 09/26/2004

Registrant: Florence, Town Of, 130 Main St, Florence, AZ 85232, Phone: (602) 868-5873

So it appears the location is some city owned property, which makes sense, since COOP stations are often located at places that are staffed 24/7 (so somebody can take a reading once a day) and many city offices are. The lat/lon is fairly close to what the COOP coordinates are, but not quite close enough.  The street address is about a half mile south of the lat/lon listed in the NCDC database:

florence-az-coop-city-map

The new location is at about 500 Main Street, rather than the 1206 Main Street listed in the NCDC MMS database. Perhaps it has been moved to a new location and NCDC has not caught up with the street address change.  Perhaps the lat/lon is off. Anything is possible as I and the surfacestation volunteers constantly find discrepancies and errors in the database.

So I decided to use the new Google Street Level View feature to snoop around a bit at the two locations. I found nothing at 1206 S. Main Street except a lot of grass and buildings. It looks like perhaps a community college:

florence-az-1200-s-main

Click for an interactive view from Google Maps

But when I went looking around 500 Main Street – BINGO! I can spot both the MMTS sensor unit and the standard rain gauge to the west of the street:

florence-az-500-n-main-street

Click for an interactive view from Google Maps

Looking at an aerial view using NCDC’s most current coordinates of 33.0363,-111.388 and Microsoft Live Search Maps, we can see what surrounds the sensor:

florence-az-coop-aerial-view-520

Click for a larger image

Click for an MS Live Maps interactive view

You can just barely make out the MMTS in the aerial view. In the street level view, it looks as if some crushed rock has been laid down near the sensor and it is fairly fresh. But more importantly, look at what surrounds the sensor:

  • Main Street with it’s traffic.

  • Buildings North, South, and West within about 10-30 yards

  • Parking lots West and East. The one East has quite an albedo. In the Arizona sun I’m sure it gets quite toasty in full sun.

It is possible this station was recently moved from the south Main Street location to the North Main Street location, which may be a warmer location, I don’t know for certain because I can’t locate any imagery of the sensor at 1206 South Main Street. Further research is needed to pin that down.

This is neither a USHCN station nor a GISS station. It is also not the only possibility for the station that produced the red dot in the HPRCC map

There is another nearby COOP “A” sub-network station at the Casa Grande National Monument run by the US Park Service, COOP station #21314:

Location Description: CASA GRANDE RUINS NATL MON OUTSIDE AND 1.7 MILES NW OF PO AT COOLIDGE AZ

Its lat/lon of 32.9947,-111.5367 is also close to my original crude estimate of 32.9, -111.4

You can see the red dot is question and it’s nearest neighbor here in this closeup of the HPRCC southwest US dot map:

pinal-co-az-station-closeup1

When I plot both stations in Google Earth and compare to the HPRCC map above, it appears that the yellow dot lines up with Casa Grande, AZ and the red dot lines up with Florence, AZ. My original lat/lon estimate is the white marker:

florence-and-casa-grande-coop-ge-520

Click for a larger image

The Casa Grande COOP station also has some interesting issues that could be responsible for a temperature rise there. Comparing aerial images on the Google Earth and Microsoft Live Search maps, which are taken at different times by different vendors, show us that it appears the parking lot for the visitor center has recently been resurfaced:

casa-grande-ls-aerial-520

Above: Casa Grande National Monument via Microsoft Live Search Maps – Click for a larger image

casa-grande-ge-aerial-520

Above: Casa Grande National Monument via Google Earth – Click for a larger image

Since I have no time reference for the photos, it is also entirely possible that I have the sequence reversed and the parking lot has faded with time. But since I don’t see any significant vegetation changes nor other changes in the landscape between the two photos, and since fading usually takes a couple of years, I’m betting that we are seeing a resurface job, which can appear in a couple of days. I would expect more differences in vegetation or other changes if the pictures were taken years apart.

I think I can make out the Cotton region Shelter on the Google Earth image, just south of the visitor center. There is a street level view of the visitor center parking lot, which you can examine for yourself, but there are no weather instruments visible.

But there is another twist, according to the NCDC database, the station has recently been converted from a Cotton Region Shelter with max-min thermometers to the MMTS system, with the CRS maintained as backup instrument, So the MMTS may be closer to the building and/or the parking lot:

casa-grande-equipment-list-520

Click for a larger image

But on 10-18-2007 it appears the CRS was removed as a backup instrument. The picture above may be the only photographic record of it’s placement. As we have seen time and time again, the MMTS often gets closer to buildings due to trenching issues and cabling, so it may have introduced a bias in this station due to the placement change. It may not, I don’t know for certain since I can’t spot the MMTS at Casa Grande.

I also thought perhaps there may be a large amount of missing data in the observer B91 forms that could account for the anomaly. I checked both Florence and Casa Grande B91 observer forms at NCDC and they both appear current and well populated with data in the last year, you can search for B91 forms yourself here:

http://www7.ncdc.noaa.gov/IPS/coop/coop.html

I did note though that the Florence form changed in appearance from May of this year to November. It went from  hand written to typed, which suggests an observer/location change:

Casa Grande B91 11-2008 Florence B91 Form 05-2008 Florence B91 Form 11-2008 (PDF files)

So we have two possible candidates for the station that made the red dot. Both have potential placement issues. It makes you wonder how many more of the dots in the HPRCC map have issue like this. I only spotted this one because it was such a large singular anomaly. I’ll check with HPRCC on Monday to see if they can identify the dot’s data origin for me. In the meantime I need help from our readers and volunteers.

Can anyone living in Arizona get photographs of these stations for me?

Advertisements

  Subscribe  
newest oldest most voted
Notify of
Al Pratt

I live in Tempe and can easily get to both Casa Grande and Florence. Be glad to do that, unless somebody else has already volunteered.
Al Pratt
REPLY: You are the first, thank you! – Anthony

Dave Andrews

Anthony,
As a UK citizen I can’t help with photos but I am very struck by the visual representation given by the two HPRCC maps. ( I have noted on previous threads the distortions introduced by the projections used by the reporting Agencies)
To my mind the first, colour, map presents a totally different picture ( especially to someone who may only be browsing ) to the second map. For example, the red area in the first map appears to cover a large area whereas the second map shows it originates in a single station.
What is the justification for publishing the first map?
REPLY: I think it is the perception that since we are plotting the atmosphere, which is a fluid, dots don’t do the job but gradients appear more “atmosphere” like. Other than that, I have no idea. – Anthony

It would seem prudent to hold off on immediately pulling the plug on an old station when a new replacement station is fired up. Comparing the old/new readings for a month or so would reveal problems like this.

Bill Illis

Man, the CIA or one of the Centcoms could certainly use you Anthony.
Ever consider trying to get the $50M reward for Bin Laden?

crosspatch

What strikes me is the magnitude of the anomaly for that one station. It appears that all of nearest stations around that one have a 0-2 degree anomaly and that one has a 6-8 degree anomaly. That is a huge difference. A real 6 degree difference in annual temperature average at that one location should make for a fairly strong “heat low” in that location with all winds spiraling in to Florence. In other words, for that much of a heat anomaly, if it is real, there should also be a wind anomaly to go with it.
Elevation and surrounding topology could also make a difference too because if there is a lower lying area nearby, heat could rise up the hill as the lower area cools off. From what desert camping I have done, I know that it is generally more comfortable at night if you are not in a low area (besides not having to worry about getting drowned in a flash flood). The desert is sort of counter-intuitive in that respect as a little more elevation gives you a warmer night.
I also wonder about those two air conditioning units and from which direction the prevailing breeze comes in that area.
REPLY: Those appear to be A/C ground level transformers or power interconnects for underground cable. – Anthony

MC

Anthony,
Keep on keepin on Watts. Hard for me to suggest you do anything because you do a great job of this whole station monitoring thing but I really believe you need to print you out a copy and send it certified mail reciept to Hansen and Gore. This way they gonna know they’re had. To me this is the most important piece of information you’ve posted.
Hat’s off to ya.

Keith J

Anthony,
You may not be surprised that the 2-4C anomalies in Texas correspond to urban heat islands. DFW can be seen in the north central, Austin in the south central and San Antonia just southwest of Austin. Even Duluth shows up well at the tip of Lake Superior. And to think the IPCC says this is a 0.05C effect.

DJ

So you find one red dot in the whole USA. Just shows how good the data really are. Those who read the literature know that urban heat island affects are on the whole slight.
BTW any comments on the blue dots in Colorado and Arkansas?

Will

Keith J,
UHI are only significant when they need to be…like when it is on the ‘data in’ side instead of the ‘data out’. Here’s an example re another catastrophe that strikes fear into the hearts of AGWists, that is, melting permafrost and rising (what methane?).
http://www.geography.uc.edu/~kenhinke/uhi/HinkelEA-IJOC-03.pdf

Mike Bryant

It still seems odd to call this product “Departure from Normal Temperature”. Why the word “departure” and why the word “normal”? Departure means “a deviation from normal”. A few degrees one way or the other is certainly not a deviation or a departure from any normal. Every day has a greater change in temperatures from sunup to sundown. This is just more political correctness disguised as science.
Why not call it what it is, 2008 Average Temperatures Compared to XXXX-XXXX?
By their definition of this graphic, every single year is abnormal.

Leon Brozyna

This is a most impressive piece of investigative science. Not everything done in science is mathematically or statistically based. Made for quite compelling reading. Almost like a murder mystery.
I trust everyone will notice that you didn’t grab at the first easy to pick station.

Alex

it is absolutely amazing what satellite technology can do! Brilliant, Keep up the good work!

Ellie in Belfast

Mmm…….nice work.
Here’s a naive question that drifted into my mind last night (and please bear in mind that I live in a region where air conditioning is not that common – might be needed only a few days in summer if we are lucky). All those badly sited stations…are the biases always going to be up (warming)? Aren’t some only seasonal? Would some be cooling also in some seasons?
OK, a non-pristine station/record is a problem, I accept that. I’m just curious. For example in Ireland recent temperatures have been less extreme than historically, but the average temperature is going up. Is that happening in other places? I would not expect it naturally in more continental climates, but it might happen due to anthropogenic influences.
So….
– AirCon – warming in summer – more likely to be used in daytime (?) Warm temperatures => higher. Off in winter? What about if it is an HVAC – will that have heat output all year?
– BBQs – Heat, but perhaps not always at hotest part of day, although could exceed day’s Tmax if siting is very poor. In Ireland we’re crazy enough to BBQ all year (“hey – it’s not raining – let’s have a BBQ”)
– Asphalt – warmer day and warmer night depending on strength of sun, wind etc. Cooler in winter (? – depending on sun)
– Building – distance, angle, paint or other materials (a la Anthony’s test)
http://wattsupwiththat.com/2008/01/14/a-typical-day-in-the-stevenson-screen-paint-test/
I could go on, but thoughts anyone else?
Is this type of information available anywhere in one place? It might be useful to anyone starting to investigate station siting – knowing what to look for and get the right information and photos.

Don Keiller

Even the BBC admits it on a regular basis (UHI that is) in their weather forecasts-
Quote “These are temperatures in towns and cities, rural temperatures will be 1-2 degrees colder”

Go to the map http://www.wrcc.dri.edu/summary/Climsmaz.html which has pop-up stations. There are two stations nuzzling each other in about the right spot. I think that your station might be PICACHO RESERVOIR, ARIZONA (026506) . Worth looking at its history. But just a guess.

Maybe the data of the surrounding stations just needs to be homogenized.

Mike Young

I notice another possible oddity. Look at the brown dot in the southwestern corner of New Mexico. The two nearest dots are both green, which would suggest quite a large gradient. There are some sizable mountain ranges in the area, which could account for large differences, but it might be interesting to take a look nonetheless.
Perhaps there are other paired dots with implicitly large gradients that could give clues to errors or distortions in the reported data. Just a thought.

All that heat must be migrating into Arizona from somewhere. My guess is that it’s coming from new sea ice.

It would appear the observer doesn’t believe the instuments either. There are several comments in the record about the “Actual PM Max” being different from the recorded temperature. For example, in Florence on 11/04/08, the observer points out that the PM Max was 76°F, even though they recorded 84°F in the temperature column. There are several entries like that, though the 8°F difference is the largest recorded in comments. It would be interesting to know why and what the total impact is…
http://www1.ncdc.noaa.gov/pub/orders/E8D7DDB1-2D06-6A27-630B-D360CF1A34A5.PDF

DJ (13:43:31) said: “So you find one red dot in the whole USA. Just shows how good the data really are. Those who read the literature know that urban heat island affects are on the whole slight.”
No, DJ, only those who read the “literature” and are naive enough to believe it think that “UHI affects (sic) are…slight.” And the data are obviously NOT validated by a low number of red dots: with a little thought you’d realize that half of the orange dots might be off, too. Maybe they should be blue.

henry

Search Results for pinal county
Station ID Network Name Lat Lon Begin End Data Available
021314 coop CASA GRANDE NM 32.99 -111.54 2000/1/1 Current
023027 coop FLORENCE 33.04 -111.39 2008/11/8 Current
026513 coop PICACHO 8 SE 32.65 -111.40 2000/1/1 Current
This all from:
http://www.hprcc.unl.edu/stations/index.php?address=pinal+county&distance=20&station_network=1&begin_Year=2009&end_Year=2009&action=search
When I click on the station ID, I get this:
http://www.hprcc.unl.edu/stations/index.php?action=metadata&network_station_id=026513
Shows last years temp data, up to current.
REPLY: Great find, thanks! Oddly one agency (WRCC) shows it closed, with data up to 1983 and another shows current. I’ll look into this more. – Anthony
REPLY2: figured it out, the station “was” closed but apparently reopened. I found this in the popup balloon for the station:
Begin: 2000/1/1
End: Current

Brian D

This seems to happen from time to time. I remember Erie,PA area having one of these hot spots for a couple months last summer (Jul-Aug). I remember it being brought up in a weather forum.
This area shows this hot spot problem in Mar-Jun last year.
Also seems to be one for SW MN this month in a rural area to the east of Marshall. We’ll see if it continues after this arctic outbreak.
Missing data problems?

henry

Those coordinates (32.65 -111.40) put it in Picacho Peak State Park…

In the entire northern U.S., there appears to be only 2 light orange spots:
1 in ND and 1 in upstate NY (Albany?).
In fact, other than CA, TX, and NM, there are very few orange spots.

H.R.

@Mike Bryant (14:00:52) :
“It still seems odd to call this product “Departure from Normal Temperature”. Why the word “departure” and why the word “normal”?…
By their definition of this graphic, every single year is abnormal.”
Well said! WHO is it that decides what “NORMAL’ is?
Anybody? Anybody? Class? Class?

Gary A.

“Picacho Peak State Park is open. Please note that our new Visitor Center is still under construction, but the park is open for public use. The new Visitor Center is scheduled to be completed by the end of February 2009. Thank you for your understanding.”

Alan Wilkinson

You have to wonder if NOAA’s team of data auditors are all occupied investigating and recalibrating the anomalous green dots covering the map and therefore have no spare capacity to check out that single red dot.

Paul Penrose

DJ,
Just because there is only one red dot on this particular map does not say anything about the quality of the entire network. Since this is a map of differences, if there were stations out there that were way off, but consistently off, then there would be no differences to see.

Wyatt A

I really hate that graphic scale. Why not put one of the buckets centered on zero? In other words, make the center ±1, the adjacent bins -3 to -1 and 1 to 3, and so on and so on.

John W.

The thread title is: “What is the red dot?” To answer the question, it’s evidence of global warming. All the green stuff on the map is just weather.

Mike Bryant

Wyatt,
If they want to keep calling it “Departure from Normal”, perhaps the center bucket should be designated + or – 5.

David Smith

Anthony, while perusing the co-op reports for Florence I noted some odd observer comments on the form
http://davidsmith1.files.wordpress.com/2009/01/0110097.jpg
(I inserted the comments from the left side of the B-91 form onto the right side, for viewing.)
Why wouldn’t the recorded max for the day also be what the observer considered the “true” max (Watts up with that?0
The “PM max” notes are also a bit of a mystery.

Alan S. Blue

DJ,
A fair number of us have, indeed, read the relevant publications. The interesting thing is how the IPCC relies upon the studies by Jones et al. – which have made some odd choices for defining ‘urban’ and ‘rural.’ The main thrust is “Big cities and small cities both see roughly equivalent temperature growth from development.” That’s great. But that’s not really a study of how large a heat bloom individual cities have. You’d want data gathered from outside -any- city for that. And there are very few such sites. But the vast majority of land in the US is outside of any city.
Fields other than climatology use direct satellite data – where the heat bloom for even small cities is evident. These are “rural” under Jones. It isn’t shocking that small cities and large cities both see development issues.
Additionally, microsite issues compound the unsuitability of a long list of sites. Putting a “rural site’s” temperature monitor within mere yards a the one whole-building heat pump isn’t sane. But it means that it isn’t shocking that it gets a similar reading to one put on a fire station’s roof in a big city – next to their own heat pump.

Have you ever seen as I have, also, small red spots on the sea, in NOAA maps?

Morris

I love that there is a “normal” temperature, as if the climate were something that we could adjust like a thermostat.
Next time one of you is talking to global climate change promoter, ask them this question;
“Assume that we could create a bug that ate CO2 and could set the CO2 at an optimum level. What would that level be? What should we set the global thermostat to, to prevent catastrophic storms that have happened throughout Earth’s history?”

Squidly

OT: Wanted to post this and didn’t really know how or where. But I found this interesting, as it is suggesting that:

Decline Of Carbon-dioxide-gobbling Plankton Coincided With Ancient Global Cooling
link to article

To this I say, what the hell? I thought AGW was based upon the premise that CO2 drives temperature higher! According to this research, the lack of CO2 gobbling Plankton lead to a rise in CO2 and thus created global cooling.
Can anyone explain this one to me? Am I reading this wrong, mis-interpreting, or just plain nuts?
: Looks to me like WUWT will win the 2008 Blog Awards! To that I say, AWESOME! and, great work people of the best blog on the planet!

Squidly, don’t count Anthony’s chickens…
A deluge from an unfriendly place could alter the standings.

henry

Since that “red dot” MAY be at the state park, maybe it’s near the ranger station or residence. Whas there any other data showing better location than the [8 SE]?
In the past hasn’t that meant a distance and direction from the last placement, or from the post office? I think there was a standard used for the measurement…

ken the weather guy

Are you guys serious? This is weather. Meteorology is a science, but it aint an exact science. Anomalies will pervade logic. sh*t happens.
On the other hand, it’s nice to know that the geeks are watching. Keep up the good paranoia.

henry

Also, it’s too bad that we can’t do a “month-by-month” diagram. That red spot is for the entire year – 1/08 to 12/08 (an average, right?)
The whole year had to run hot in just that one spot.
A few months of missed data?

Super sleuthing!!! Sherlock would be proud. Can’t wait to hear from the local Baker Street irregulars.
And five stars to Google Earth. Wonderful tool for wildfire tracking, too.

Arizona has a zit? Or a cancerous mole… caused by Global Warming!

Leslie

Someone made a comment regarding the graphic scale where there is no bucket centered on 0. I have a similar concern. In the first graphic, where does zero belong, 0 to 2 or -2 to 0?

Larry T

I think it is definitely missing data problem from fill of the august thru november data from FLORENCE.

anubisxiii

Just wanted to get on the skeptic train…
Love the site, one of my daily stops….

Scott Gibson

@David Smith and others:
I suspect the notations about true daily maximum temperature are because they end their “max-min temperature day” at 5:00 PM. Hence, the max of 90 on day 2 was actually recorded on day 1 after 5:00 PM and the 88 was the highest actual high on day 2. It’s possible also that it is the other way around, and that they report the 88 because it is the highest temperature recorded before 5:00 but report an actual high of 90 that happened after 5:00. Perhaps someone could post a report on how these things are determined and whether there is a standard way to report this data throughout the world.

Joel Shore

Keith J says:

You may not be surprised that the 2-4C anomalies in Texas correspond to urban heat islands. DFW can be seen in the north central, Austin in the south central and San Antonia just southwest of Austin. Even Duluth shows up well at the tip of Lake Superior. And to think the IPCC says this is a 0.05C effect.

First of all, the temperature anomalies plot is degrees fahrenheit, not degrees celsius, so to compare apples to apples, you would compare 2-4 F to 0.09 F. Second of all, you have to compare to 2-4 F you see to the 0-2 F of the background in the region to get a good measure of what is likely due to any urban heat island effect. Third of all, the IPCC doesn’t claim that the effect of UHI is only 0.05 C at the places that it occurs…What that number would be is the amount by which it contaminates the (global?) surface temperature record. And, indeed, note that these regions that you pointed out represent a very tiny fraction of the total area of the U.S. so they would affect the nationwide average a lot less.
Mind you, what I have done here is not any sort of careful analysis but just pointing out to you that your simple logic by which you are trying to discredit the IPCC’s estimate doesn’t really stand up to any sort of serious scrutiny.

Brian

Friend works at Casa Grande. Will send him message!

Mike Bryant

OT…
Wondering if the Mauna Loa CO2 annual mean growth rates correspond to Pacific Ocean temperatures. Maybe someone better at plotting graphs has some idea…
Thanks,
Mike Bryant