This is interesting, and of course it goes hand-in-hand with what I have been saying for years.
Scott Sistek, of KOMO News/Weather reports:
=============================================================
For several years the thermometer at SeaTac airport has been reporting temperatures 1-3 degrees above surrounding areas.
Instead, it seems the thermometer at Sea-Tac is finally back on track, reporting temperatures more realistic with respect to other nearby thermometers. It’s been a long suspicion among some local meteorologists that the thermometer at the airport been running a bit warm over the past few years, frequently reporting temperatures 1-3 degrees warmer than surrounding sites. (Both UW professor Cliff Mass and I have done blogs on this apparent warming in the past.)
Here is just one example from July 16 last year when Sea-Tac reported a high of 88 degrees but everyone else around the Sound was closer to 83-86. (KSEA is Sea-Tac, the numbers on the far right are the preliminary highs for the day. This link will help decode the other cities listed here.)
UW research meteorologist Mark Albright has been tracking this anomaly for the past couple of years and has been among the most vocal in this apparent discrepancy. As just one example, he found for those first two weeks last July that the Sea-Tac gauge ran an average 2.3 degrees warmer than four other neighborhood thermometers placed within a couple miles of the airport.
=============================================================
Apparently, they fixed the ASOS thermometer, and the problem went away.
Read the whole story here: http://www.komonews.com/weather/blogs/scott/Did-Seattles-climate-cool-15-degrees-in-a-blink-of-an-eye-254419491.html
h/t to reader Steve Z.
And as I’ve documented before, such errors remain in the climate record. Like the malfunctioning airport sensor in Honolulu, where the skewed temperature set new high temperature records. See this interview with a NOAA/NWS meteorologist on the issue:
But, even after knowing they were caused by a malfunction, the NOAA/NWS leaves the bogus high temperature records intact. Only government could screw up data this badly.
Here is where the sensor is:
The SeaTac ASOS, according to NCDC HOMR, is located below.
SeaTac is part of the GHCN network used for climate. But was it surrounded on three sides by heat holding asphalt in 1948 when the weather records began there?
Doubtful.
First Sea-Tac Airport Terminal, ca. 1946
I wonder what this revelation will do to this study, which used Sea-Tac and other GHCN stations as the basis for the claim?
New study finds “nighttime heat waves” increasing in Pacific Northwest
UPDATE: more on why these sorts of failures tend to be mostly hot failures:
If the recordings were abnormally low, the records would have been booted.
That’s called “fraud”.
I know this is Anthony’s big deal (kudos to the studies and work on the data integrity). However, unless thermometers tend to read higher when broken, then it makes sense to keep all the data, rather than finding only the anomalously hot ones, unless the entire record is purged for both high and low data.
@Ed Barbar
In the case of the airport ASOS HO83 thermometer, the way they fail makes them almost always read higher. In fact many of the ASOS thermometers have been replaced with one less prone to failing this way.
So no, it doesn’t make sense to keep these records. They are false and don’t “even out”.
Would it be possible to correct the error, note an “*” with the original data and the error correction method?
Why did it read high?
That study is invalidated if it depended on this temp record.
Is there a pattern? Do ASOS sensors always fail high? Are partial failures detected? Is there a range of failure from 1 degree to ?? Until we know the answers to these questions all ASOS temperature data should be viewed with skepticism.
@edcaryl ASOS sensors are aspirated (fan driven airflow) they almost always fail high from these sorts of failures:
Fan slowdown/failure (slowdown due to dust grime most common) When the fan slows, incoming solar and nighttime LWIR have more effect on instrument
Blockage: birds nests, wasp nests, etc impede air flow.
more here: http://wattsupwiththat.com/2008/01/10/inside-the-asos-ho83-tempdewpoint-sensor/
[SNIP waaayyyy waaaay WAAAAAAAY off topic, and a rant – don’t clutter up my technical thread with this junk – Anthony
Well, isn’t that precious of Mr. Albright. NOW…. that he and his fellow AGWers are desperate for temperatures to warm…… they correct the record DOWN.
Still, a good thing, though. The TRUTH is what matters.
Thanks for sharing, Anthony. Heh, heh, Western Washington (with its Envirostalinism) may not be so hot… but, even so, we’re stll pretty cool.
GREAT WORK ON GETTING TH TRUTH OUT ABOUT TEMPERATURE MEASURING ACCURACY, ANTHONY!
“But was it surrounded on three sides by heat holding asphalt in 1948 when the weather records began there?”
Not to mention surrounded on all sides by 60,000 lb thrust jet engines producing massive hot air streams?
Always some amount of wiggle room with meteorological data,,,,, does that help getting a 97% consensus? Is there a time frame for inspection and calibration at these sites?
Why is it that when one of these thermometers “goes bonkers” they only ever report higher temperatures, never lower? Hey?
Using temperature measurements at airport to cover wider areas is only slightly less dum than using ones at steel-mills. These weather stations are designed to offer information for incoming and out going flights , that is it . They have become used for other things not because they data they offer gives information that relates well to the greater area , in fact airports can be an oddity not a norm for there area. But becasue their there and its better than nothing and cheaper than having another one.
Anyone that has worked on the big open concrete pans of an airport can tell you its gets warming there than in other locations. In the past this did not matter has weather was so hard to predict and so verbal even quite large errors were OKish, now we seen silly claims of multi- decimal accuracy and it not OK any more.
Would anyway think that the weather you get at the top of 20,000 foot mountain can be used to cover the valleys below at sea level becasue its within a few miles or would they accept that given that is such a big difference between the two you really should not ?
The Cult saw the temperature, and it was good. So sayth the Cult.
brians356 says:
Why is it that when one of these thermometers “goes bonkers” they only ever report higher temperatures, never lower?
Anthony gave some technical reasons. But one reason didn’t make the list: human manipulation of the temperature record.
As the article shows, the high T problem was the subject of repeated complaints. But no one ever fixed it, until they were forced to replace the thermometer. I suspect that someone preferred the artificially high temps.
on one side you have corrections based on assumptions in front of computers miles away from sites.. on the other side you have people checking each measuring site .. who do you trust?
And here I thought it was MY (Seattle) thermometers that were “off.”
I wonder how much of a play this will get in the Seattle Times (I don’t read it anymore).
Janice Moore, please don’t accuse me of being an AGWer. I follow the objective truth wherever it may lead.
“Janice Moore says:
Well, isn’t that precious of Mr. Albright. NOW…. that he and his fellow AGWers are desperate for temperatures to warm…… they correct the record DOWN.”
Ed Barbar says:
April 14, 2014 at 8:48 am
This is one huge Red Herring. The thermometer was discovered to be reading high because it was discovered to be broken. The proper action is to lose the high readings for this thermometer. The question that you raise has to do with systematic investigation of all thermometers and, though an interesting question, bears only67 superficial resemblance to the question at hand.
Pardon the “only67.” I did not type that.
can’t provoke the second coming if the world ain’t hot enough for satan yet
The airport diagram for KSEA shows the ASOS alongside taxiway N between 16R/34L and 16C/34C. The possibility of extra dust or grime hosing down the site would be a certainty. A quick spike in temp during a winter day? Maybe a 737 holding short of 16C on the taxiway?
Large swaths of residential areas were acquired in the 1990’s for the construction of the third (westernmost) runway at SeaTac. There is plenty of unused airport property on the west side of SeaTac that would be a much better location for instruments than adjacent to a taxiway.
Mod: o/t here. @ur momisugly Janice Moore. I believe you are about 15 miles NW of Oso. Do you have a rain gauge? If so, were you showing extreme precip during March? The Finney Creek RAWS had around 20″ prior to March 22.
Anthony Watts says:
“So no, it doesn’t make sense to keep these records. They are false and don’t “even out”.”
If the erroneously high records are kept then the temperatures registered by the new more accurate instruments will indicate cooling compared to historical data. So the truth will win out in the long run, aside from the fact that they are poorly placed in the first place. If all of the UHI record temps were kept and the locations corrected to eliminate the UHI effects we would see some erroneous, but major, cooling I suspect, relative to those erroneous historical records.
“If the recordings were abnormally low, the records would have been booted.”
No, if the recordings were “abnormally” low, 2 degrees would have been added to the temperature records of every thermometer in the country to make up for the supposed error.
or maybe that already happened….
Why is there only one thermometer at each station? Five or six thermometers simultaneously recording temperatures would show up any inaccurate temperatures like a sore thumb. They cannot be that expensive — hey, the whole weather station can’t be that expensive. Why not have several many-yards-apart duplicates of them also?