WUWT readers may recall last summer when the ASOS weather station at the Honolulu Airport malfunctioned, giving a whole series of shonky readings that resulted in a string of new record high temperatures being set. What was even worse, is that NOAA, knowing the records were based on faulty equipment, let them stand anyway even when a nearby NOAA station demonstrated the records were false.
WUWT reader “Geo” alerted me to the issue at DCA in “tips and notes”, and I made some immediate screencaps of the data which I have below. The DCA ASOS station at Washington National/Reagan Airport is part of the COOP A network which makes climate observations. First, a look at the station itself.
By siting standards, it at first glance appears to be well sited, being over 100 meters from the nearest runway, in the grass, and on the bayside riverside:
But on closer inspection it appears to be sited over asphalt:
From the ground, NOAA has this photo and more:
Here’s the hourly data in question, note the big jump at midnight and the 87.1F recorded at the 6 hour preceding mark. What’s odd about it is that its a big jump compared to the readings in the hour before and after, plus it occurred during thunderstorms and light rain. It seems unlikely.
Nearby Andrews AFB, just a few miles away, doesn’t show a similar jump in readings. It shows 64F and light rain at midnight, and a 6 hour max of 73F.
The high 87.1F reading at DCA made it into the official climatological report for Friday May 28th, which places it at 12:25AM, in the middle of rain and thunderstorms. It seems pretty clear that the reading is erroneous.
http://www.weather.gov/climate/getclimate.php?wfo=lwx
I recorded all the screen caps above last Friday, May 28th. I figured I’d wait to see if the NWS staff caught this and corrected it.
Sure enough, they did:
That’s the right way to do it. But is the 76F reading at 127AM valid? With a malfunctioning sensor, who knows for sure?
Now the question is: why does one NWS office fix an obviously faulty ASOS reading in the climate record while another ignores it and leaves it as a new record? Are there not standards for handling such things organization wide?
Another question is: how many events like this go undetected due to a lesser, non obvious magnitude, and remain in the climatic record?
Invariably, such events almost always (though there are rare exceptions) seem to embrace a warm error, be it electronic or human error, or even transcription error, such as “Dial M for METAR“.
And yet, there has been a large migration of GHCN to airport systems both in the US and globally.
At ICCC4, Apollo 17 astronaut and geologist Harrison Schmitt came up to me after my talk to tell me that I was “spot on” with criticizing the use of airport sensors for climate, because they were designed for a different mission. It was a proud moment for me, having watched this man walk the moon as a boy in high school. He said to me (and I’m going on recollection):
“The only purpose of the ASOS system is to measure the runway conditions for flight safety. I’ve seen runway temperatures vary as much as ten degrees from other airport locations. These shouldn’t be used for climate.”
The data at Reno, NV airport, showed not just siting differences, but a UHI factor also.
I agree with Dr. Schmitt’s assertion, and given that ASOS continues to produce faulty data, perhaps it is time to look at ASOS data issues on a broader scale.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Anthony, I am really looking forward to your article.
I’ve been scanning wunderground.com for a while now, looking at various major conurbations around the world to see the range of temperatures being recorded at the regional locations surrounding the target city. Most large/capital cities have at least 20-30 stations within a smallish radius – 50km or thereabouts? – of the main entry, so they provide a good basis for comparison and cross-checking; these readings are all shown at the bottom of the page. They are readings from automatic online sensors that are refreshed at regular intervals, therefore at any given moment they are largely aligned to within a few minutes or seconds of one another.
What I find interesting is that every time I look, wherever I look, there is a significant range of readings – typically between 2-5 degrees C – from the lowest to the highest. Now, if readings vary so much over a relatively small area, how can the “powers that be” justify extrapolating a single remote location’s temperature readings across an entire 1200km grid section and beyond, and then claiming that the single averaged number they have used is a fair representation of the monthly heat flow characteristics of a huge swathe of the earth? How do they sleep at night, knowing that they have used a handful of small anomalies for a given month (which are grossly inadequate averages given the 43,200 dynamic weather minutes in a month), and which may be less than than the differences between adjacent suburban stations, to daub the entire Arctic region red? Sure, a method is required to deal with the difficulty of reconciling sparse measurements into a global picture, but the results should be regarded as, at best, mildly indicative, and certainly a million miles away from the gospel truth which we must insist upon before venturing to disrupt the world economy.
And why, given these substantial local variations in the places we live in and the demonstrated instrumental glitches that can skew the results without sensible intervention, oh why are those people getting so dramatically worked up about less than a degree of apparent (but still disputable) global rise in a whole century?
Global warming? Possible, but minuscule. Less than the difference between moving from one room in a house to another. And we’re supposed to treat this as catastrophic?
Clearly, one thing that does not get handed out with a PhD degree in climatology is a sense of proportion.
To all those who would have us believe that disaster is nigh, I say, “Forget the fancy maths and contrived logical sophistry; just look at the data, stupid.”
Just a little more information regarding the HO-1088 temperature system…
First, the system is checked out every 3 months by the ASOS Electronics Technician.
He cleans the system, and checks the components, then compares the temperature readings against a secondary standard, which is checked by a Claibration Lab contractor once a year. The claimed accuracy of the system for the basic system is +/- 0.5 degree celsius. However, in the S-100 book, (the ‘bible’ on how to calibrate the system) it is stated that the system is within calibration if the readings are within +/- 5 degrees celsius!(could this be a typo, or a failure to print the decimal place, and not inserting a zero in front of the .5?)) No technician would let the system drift out this far, I will assure you.
Secondary problem… The ventilation fans which are running 24 hours a day are beginning to wear out. (about 8 to 10 years usage). The only way that this is caught,
unless the 3 month maintenance catches it, is the high temperature readings which look really out of sync with surrounding systems. So, there you are.
I hope this clarifies the situation. Also, the system measures temperature in degrees celsius, then converts it to degrees fahrenheit. Remember, if the temperature is negative, minus 0.1 through -0.5 is added upward (add 0.5 to the reading) , rather than taking the absolute number, and adding 0.5, then restoring the sign of the number.
Just a few thoughts.
But I presume that NONE of the stations used to try to guess at climate change are sited to give a guide to ‘world’ conditions.
They have been put up (I only presume) to measure local conditions. They have grown up by accident and convenience, and as you have shown they have not been adapted.
There is no central controlling standardising body for these stations.
My presumption may be bollocks of course, so my apologies.
PS – I live next to an RAF base in England and I am told that our area will always pick up record cold temps due to the nature of the soils around us. This may be rubbish but it points to another non climate reason for temperature anomalies.
Isn’t this a bit of a leap Gail? Wouldn’t the use of Occam’s Razor which applies the meta-theoretical principle that “entities must not be multiplied beyond necessity” (entia non sunt multiplicanda praeter necessitatem) and the conclusion thereof, that the simplest solution is usually the correct one would lead one to a more concrete, logical, and dare I say it, rational reason without having to reach for the world-wide, eternal conspiratorial reasons?
Suffice it to say, bad/easy and ‘convenient’ siting of these stations is sufficient reason enough to see why the data shows warming, reason enough without involving the CFR, the Trilateral Commission or the NWO bogey men …
.
.
All temperature is local.