More on the Santa Ana Rooftop Weather Station: comparison stations also problematic

You may recall a couple of weeks ago I did an investigation of a COOP weather station in Santa Ana, CA prompted by front page article in the Orange County Register titled Urbanization Raises The Heat in Orange County. It was front page news that day, on Friday, August 8th.

The story focused on the 7.5 degree average temperature increase in Santa Ana in the past century. Most of the focus was on UHI, some was on AGW, but one thing that wasn’t mentioned was station siting.

Warming trend

A little bit of investigation on my part showed that the official climate station for Santa Ana was not only on the roof of the fire station there, but also near several A/C unit exhausts, and the station shelter door was facing the wrong way, west instead of north which would allow sunlight into the shelter, potentially biasing the thermometer at the times readings were taken. NOAA and the World Meteorological Organization both have specs that the shelter does must face North.

Santa Ana Station looking North.  Click for a larger image

And as we see from this photo, direct sunlight does indeed enter the screen when the door is opened. At certain times of the day and year it likely strikes the thermometer directly.:

Click for a larger image

The reporter that wrote the original story, Gary Robbins, also has a blog called “sciencedude”. I asked him to review the Santa Ana COOP station, and he reported again on it on Friday September 12th in a story titled:

Has Santa Ana been reporting wrong temperatures for years?

In that story, he enlists the help of the local NWS office to check out the issues I’ve uncovered. As a result, there is a quote from Noel Isla, from the NWS in San Diego, who is responsible for the Santa Ana COOP station:

“The weather station was moved from the ground to the roof,” Isla later told me. “And for an undetermined period, the door of the weather station wasn’t facing in the right direction. But I compared the Santa Ana data with other nearby stations (Fullerton, Yorba Linda and John Wayne Airport) and the data is consistent. The affect of the air vents and the relocation of the station appears to be negligible, if they had any affect at all. The temperature readings are good.”

I was curious about the three used for comparison, so I did a cursory check of the three stations Fullerton, Yorba Linda and John Wayne Airport using the tools available to me today.

First I logged into the station database at the National Climatic Data Center using the “Guest Login” button.

Then I got the latitude, longitude, and equipment type for each station, so I knew what to look for.

Then I did Google Earth/Maps searches and Microsoft Live Maps (closeup aerial views) to actually locate the stations and their surroundings.The airports were easy to spot, because they use the ASOS system, and that is easily visible from the air by the presence of the anemometer and wind vane towers, which are painted FAA regulation hazard red/white stripes.

Here is a ground level ASOS photo from NOAA for comparison:

Now lets look at the three stations cited by Mr. Isla.


John Wayne Airport ASOS

You can spot the ASOS station on this live Google Maps view

And here is a closeup aerial view from Microsoft Live Maps:

John Wayne Airport ASOS click for a larger interactive image.

Note that this ASOS station is not only surrounded by asphalt runways and tarmac, it’s actually ON the asphalt access road to it. That would introduce a positive bias to the temperature. What is puzzling is why they did not place it in one of the grassier areas at the airport. Perhaps it is due to the primary mission being aviation, and not climate monitoring.


Fullerton Airport ASOS

You can spot the ASOS station on this live Google Earth View

Fullerton Airport ASOS click for a larger interactive image.

This station is just a few feet away from a large airport tarmac, which will of course add heat during the day to nearby air and retain heat at night, resulting in a higher than average temperature than if it were in a grassfield for example.


Yorba Linda (Nixon Presidential Library)

This one is a but more difficult, because the temperature sensor is much smaller than an ASOS. It is listed in the NCDC database as a “NIMBUS” which is the newer model of the MMTS (Max-Min Temperature System). The MMTS looks like a beehive mounted on a pole about 5.5 feet high. Here is a photo from NOAA:

I believe I can see the MMTS unit in the NW corner of the photo, near the A/C plant, look for the circle driveway with a grass center in this Google Maps View, there is a white dot on the grass, which appears to be the MMTS unit.

Here is a close up aerial view, looking south, from Microsoft Live Maps:

Click for a live interactive view

While it is over grass, it’s still within 100 feet of asphalt and the building and the HVAC plant. Since this aerial photo lacked the closeup detail needed to absolutely identify the MMTS, I called upon volunteer Frank Perdicaro, who lives in Orange County. He was able to get photos today. Thanks Frank!

Click for a very large image

Additional views:

Click for large images

Note that in addition to the MMTS proximity to the HVAC plant and the parking lot, as shown in the additional photos, construction is going on, altering the albedo and eventually the character of the area around the temperature sensor.


So with all the three stations the NWS used for comparison (to Santa Ana Fire Station on the roof) also being out of compliance for station siting, is it any wonder that they’d be in agreement since they all have similar sorts of positive biases?

The NWS staffer, Mr. Noel Isla, probably isn’t aware of the placement issues with these stations and the potential for bias effects.

To get an idea of how these effects manifest themselves, here is a peer reviewed paper on the subject of placing temperature sensors over concrete and other surfaces:

Yilmaz et al (2008 ) Heat over grass/soil/concrete

Since we can’t rule out positive bias in the three stations cited by Mr. Isla, they may not be useful for comparison. So the issue remains, is Santa Ana reading higher because of it’s rooftop placement near A/C exhaust vents? Only a nearby parallel measurement in a proper siting venue over time will tell for sure.

In the meantime, how much of Santa Ana’s temperature increase is from UHI, AGW, station siting or natural causes? Before this investigation, station siting wasn’t considered, now I think in light of what is presented above, siting should be given strong consideration as a factor.

0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
September 14, 2008 2:07 am

It is obvious the siting problems is the cause of the increase in temperature and not the global warming theory! Unfortunately, some in the scientific community just will not accept that.

September 14, 2008 4:51 am

Just wondering if the following news has any impact on this:
If it does, it should be historical. But changing the orientation of the shelter door, could have had some impact with the winds mentioned by JPL, when that orientation was changed.

September 14, 2008 6:34 am

I guess we’ll have to wait and see what the next excuse is from the NWS if Mr. Robbins decides to hold their feet to the fire (or hot tarmac) in yet another follow-up article.
It’s a crime that the vast majority of the Lamestream Midiots™>/b> are wholly-owned subsidiaries of Great Goreacle Glow Bull Worming, Inc. and don’t have the journalistic integrity (I know, that’s an oxymoron these days.) to investigate the entire carbon trading scam with the same vim and vigor that they do for sex scandals, hurricanes heading towards New Orleans or Paris Hilton forgetting to put on a pair of panties before going shopping in a mini skirt.

Chris D.
September 14, 2008 6:44 am

Anthony, the MMS shows the location being 120 W. Walnut St. from at least 1992-1997. I sure can’t spot a fire station at that address on Google maps. Note the warmest year being 1997, with a precipitous drop after that. Looks like there was a station move in ’97. If that’s the case, it would be very interesting to see the siting conditions during that period, as well. Then, there were all of the “AD HOC” updates:
It would also be interesting to see what method they used to calibrate those ASOS.
Anyway, I was glad to see that the reporter was at least receptive to further investigation. I’m sure he got a bit of an education doing that series.
REPLY: Chris: you can see the fire station from ground level here:
– Anthony

M. Jeff
September 14, 2008 7:04 am

The referenced paper, “Yilmaz et al (2008 ) Heat over grass/soil/concrete”, shows a 7.54 C mean difference between asphalt/concrete and grass at a height of 2 meters during the month of August. Any thoughts on what the temperature difference might be for an entire year. Any approximation might be enlightening.

Brent Matich
September 14, 2008 7:18 am

Hey, where’s the propane torch! Santa Ana get your station together. LOL! Mr. Isla, have you ever been on a rooftop with multiple A/C units going? Geez!

Bill Illis
September 14, 2008 7:33 am

Considering the temp increase in Santa Ana would be one of the highest measured anywhere, the area might represent a great example of the impacts of global warming. I don’t think the climate models predict 7.5C by 2100 for any particular region so this would be the most extreme example of global warming’s impact there is.
I imagine very little has changed in Santa Ana except population and metropolitan growth.
Going by the pics, it looks like grass still grows and trees still grow.
From Wikipedia, it is one of the most densely populated areas in the United States so there hasn’t been massive migration of population away from the warming.
Also from Wiki – “Santa Ana’s climate can best be called mild. Winter daytime temperatures average in the high sixties and low seventies (degrees Fahrenheit), and summer daytime temperatures average in the mid eighties.”
Going by Google Maps, Santa Ana is just part of greater Los Angeles since there is no green space in between. Urban Heat Island it is then. 49,000 people in 1950. Part of the city of 12,000,000 in 2008.

Mike Bryant
September 14, 2008 7:38 am

This is all very interesting, however, how does it refute the feeling that the sciencedud gets as he sits in a traffic jam?

Chris D.
September 14, 2008 7:48 am

I have to say, though, it was very disappointing to see Robbins wrap up his post by turning the readers’ attention to an Ad Hom attack on Anthony’s credibility. Nowhere did see him address the issue of site moves other than what Mr. Isla commented on. And Mr. Robbin’s journalistic spidey sense should have really started tingling when Mr. Isla admitted that “And for an undetermined period, the door of the weather station wasn’t facing in the right direction”.
For an undetermined period?! Huh?
So much for accurate record keeping. Mr. Robbins, if you really want to do your readers a service, have yet another look at this. Try to pin Mr. Isla down exactly when station moves took place, exactly where it was sited and when.

Chris D.
September 14, 2008 8:17 am

Ok, I spotted it – your overhead view was looking south which threw me off a bit.

Leon Brozyna
September 14, 2008 8:19 am

My reaction to this portion of the quote from Noel Isla, from the NWS in San Diego:
But I compared the Santa Ana data with other nearby stations (Fullerton, Yorba Linda and John Wayne Airport) and the data is consistent.
is that he doesn’t get out very much. Considering the placement of the slats on the Stevenson screen, I would imagine that it’s even hotter inside than the roof itself as all the rising heated air would quickly build up inside the box. So it’s ‘consistent’ with other stations? ‘Consistent’ in that the temperatures are the same or in that temperatures vary in the same manner, i.e. showing rising/falling temperatures in a way that matches?
The siting of these stations may be indicative of what’s happening in the microclimate – UHI {though even at that, the numbers would seem to be skewed upwards}. I just hope that the new climate monitoring stations resolve these siting problems and don’t just replace old equipment with new equipment at the same old poor sites.

September 14, 2008 2:41 pm

I posted on Anthony’s field trip thread about airport weather station sites. Won’t repeat it here, I’ll just refer folks to it, because I think it raises a valid concern.

Bobby Lane
September 14, 2008 2:42 pm

When you get back you should contact that reporter again and explain your findings to him. I am sure it would be a great ‘scoop’ for that newspaper to report that due to biases the weather reports for the area regarding temperatures have been off for the last X number of years; in fact, higher than the actual temperatures due to UHI. And, if you include your surface station project results thus far, it will give him more than enough information for him to report how the entire nation’s temperature measurements may be off and why GISS and NASA’s James Hansen might be wrong about AGW. You can use the rules of siting as an objective witness of sorts to show that you are not just sharing an opinion.
It’s my opinion now, with nearly half the sites in the US surveyed, that the entire system needs a major overhaul. I don’t know exactly what kind, but I am convinced that it does.

Aussie John
September 14, 2008 2:46 pm

Is there a list of the UHI ‘corrections’ for individual sites when these recordings are used for determining nationwide or worldwide ‘average temperatures”?
It would be interesting to see what the corrections are for this particular site.
AC units discharge cold air to the atmosphere in the winter and warm air in the summer (assuming these are reverse cycle units) so any adjustments would have to take the seasons into account, as well as the occupancy of the building, hence the use of the AC units, which it is doubtful would be known.

Bob B
September 14, 2008 3:00 pm

Anthony, thank you for following up on this, I think Science Dude really wants to do the right thing except he feels as if he has to be an AGW advocate. Thank you for all you do Anthony.

September 14, 2008 3:15 pm

So the reporter bought the idea that asphalt, buildings, urban heat, and nearby a/c units do not cause any rise in the temperatures recorded. Hmmmm.
Somebody ask him to walk barefoot over the parking lot for an afternoon in July.
I wonder if it ever crossed his mind to look at the “science”? He calls himself science dude, right? Perhaps he could even have asked Anthony for some citations.
If he thinks that his cursory check of 3 other stations constitues a scientific refutation of Watts’ points, he should change his name.

Frank Perdicaro
September 14, 2008 4:37 pm

Please note there is a difference between the shot Anthony provided and
the shots I took. Yes, there is construction today at the Nixon library
and there was NOT construction going on time the aerial shot was taken.
The point to be illustrated is not invalidated by taking photographs at
different times. In all cases the temperature guage is located next to
the AC exhaust for a large building.

September 14, 2008 6:03 pm

Absolutely amazing, I’m sure the US is the only country with problems in their ground stations. Really though, can you imagine what the south America, or Chinese installations look like!
With all the corrections GISS employs to correct these things, they don’t take into account heat exhaust or the fact that roads are almost always nearby.
With all the corrections to the satellite measurements and GISS measurements the only thing they have going is that they correlate well. If we fixed this ridiculous problem, I wonder what other corrections would have to be invented!

Aussie John
September 14, 2008 7:01 pm

According to the ‘Quality Control, Homogeneity Testing, and Adjustment Procedures’ section for the USHCN ( their last ‘adjustment’ of 6 is:
“The final adjustment is for an urban warming bias which uses the regression approach outlined in Karl, et al. (1988).”
That reference states in its abstract:
“The average heat island impact during the period 1901-84 for the HCN is largest for the daily minima (0.13°C) and the temperature range (0.14°C), while the impact on the daily maxima (0.01°C) is an order of magnitude smaller.”
These people are fooling themselves if they think the impact on this site is as small as they think.

September 14, 2008 8:16 pm

The NOAA has a positive adjustment for Station History Adjusment Procedure (SHAP).
Yes, I said “positive”.
No, I didn’t make a typo and really mean “negative”.
I actually meant “positive”.
No, you don’t have sand in your eyes . . .

September 14, 2008 9:53 pm

Interesting discussion.
I’ve spent considering whether to recalibrate my weather instrument based on NOAA’s idea of data quality based on surrounding instruments. When there are no comparable instruments, it is impossible to come up with absolute values that track “nearby” areas. Comparing Santa Ana to Yorba Linda? It will never work.
Comparing long-term trends between these instruments is unconvincing. There are no controls in place to prevent changes to the instrument surrounding, or even the instrument itself. Too many variables. It is mentioned that the Santa Ana instrument was “moved to the roof”. Doesn’t that make comparison of long-term measurements, before and after the move, impossible?
And then, then there are all the other assumptions such as asphalt, lawns and “air conditioners”. I don’t know if the picture of the Santa Ana station was intended to show the offending “air conditioning vents”, but the vents shown in the picture are most certainly fresh air intakes on input plenum of forced-air heater. The influence (if there is one) would be to make the instrument read warmer on cold days instead of warmer on hot days as would be the case if the vent were attached to an air conditioner condenser.
Which is to say, it seems like there are a lot of assumptions in your observations — all difficult to verify in the instance of the Santa Ana instrument.

Mike C
September 14, 2008 10:19 pm

A good KBSF article on the science dude here:

Chris D.
September 14, 2008 11:35 pm

What’s really interesting now is that I can’t even pull up the data for this station.
Using this:
I tried “santa”, “ana”, and “santa ana”. Maybe I should try searching under “Waldo”?

Chris D.
September 14, 2008 11:54 pm

I’d love to check the claims that have been made – want to confirm the actual rise he’s reported as well as look at the data from the other sites referenced above, but not finding data for any of them right now. Is something amiss?

September 15, 2008 8:29 am

Jim Reardon,
excellent observation.
I would add that it STILL increases the vaunted AVERAGE TEMP!!
Also, Anthony keeps talking about a definitive study to validate the readings and adjustments made to account for site variability. This is being bypassed, which will leave an erroneous history, to go directly to a new system. How the new system’s readings will be integrated with the questionable old systems readings leaves me in awe of these people.

September 15, 2008 11:05 am

What’s really interesting now is that I can’t even pull up the data for this station.
Using this:

I don’t think we’re discussing this as a GISS station; the newspaper is what drew attention to this station.
It’s a NOAA CO-OP station. Login as Guest then search for SANTA ANA
SANTA ANA FIRE STN is at 33.7442,-117.8667 and if on the GISS map you click on the south California coastline (just north of Mexican border) you can sort by location (click on “*”) to find stations near 33.7,-117.8. None of those have the Santa Ana Fire Station name.
REPLY: It is a COOP-A station, not a GISS or USHCN station. Different network, so GISS won’t have it for the same reason they don’t have RAWS or Agrimet stations. – Anthony

Chance Metz
September 17, 2008 10:17 am

Why do we put these weather stations in the worst places? Is it to make false proof of global warming or do they have no clue on how and where to set them up?

September 26, 2008 3:49 pm

[…] And my complete write-up on it is available here […]

Verified by MonsterInsights