Which mercury reading in Mercury is the right one?

How can two stations less than 1/2 mile apart be so different? How can one set a new record and best it by two degrees while the other doesn’t?


I was intrigued by this Record Event Report out of Las Vegas today because of the mercury reading, specifically in the “town” of Mercury:


Hmm, smashing. 113°F seemed high to me, since I was monitoring the state of the art NOAA Climate Reference Network Station in Mercury, NV earlier in the day I was surprised to see that value. I looked up the Mercury Desert Rock airport, and sure enough, there was 113°F. But, there was also another value.


Source: http://www.wrh.noaa.gov/mesowest/getobext.php?wfo=vef&sid=KDRA&num=72&raw=0

Upon checking the data from the nearby Climate Reference Network station, I discovered that the highest temperature recorded there was only 110°F


Source: http://www.wrh.noaa.gov/mesowest/getobext.php?wfo=vef&sid=MRYN2&num=72&raw=0

The CRN monthly report says 111°F, so the high must have been very brief to not be captured in the 5 minute CRN data above.


Source: https://www.ncdc.noaa.gov/crn/newmonthsummary?station_id=1049&yyyymm=201306&format=web

When I say it is a nearby station, I really mean it. It is only 2164 feet (0.41 mile) away as I demonstrate on this Google Earth image:


I double checked the image by zooming in to be sure both stations are actually there, and they are. The signature of the ASOS aviation weather station is clearly visible as is the signature of the CRN station.

So what do we have here?

  • Two stations separated by 2164 feet, both are “official” stations operated by NOAA.
  • One station, the KDRA ASOS is designated for aviation purposes, but used for daily climate records. It has a single sensor, with a known history of problems. It is also near a big chunk of asphalt tarmac with an albedo different from the rest of the desert floor.
  • The other station, the MRYN2 CRN is a Climate Reference network station, has state of the art sensors, including triple redundant temperature sensors, with automatic quality control, is placed specifically to be away from influences such as asphalt tarmac, and is surrounded by albedo matching the desert floor.
  • The elevations are nearly identical with the airport ASOS station at 3238 feet and the CRN station at 3295 feet. That difference of 57 feet won’t amount to any appreciable delta in temperature due to the lapse rate.
  • The temperature difference at the time the ‘record temperature” of 113°F is recorded at KDRA is two degrees. The CRN station is at 111°F.

So which one is the real temperature? I suppose it is a case of which one gets used for the “official” official reading, in which case is the warmer KDRA ASOS station.

The nearby CRN station does call the KDRA station data into question, but the horse has already left the barn now that NWS has released the numbers to the media. I will lodge a request with NWS Las Vegas to have them look at Mercury ASOS system though in case it is on the fritz as so many often do.

Oddly, after spending millions of dollars on the Climate Reference Network, the old COOP and ASOS network still gets used for high and low climate records.

Preserved data: (PDF files)



0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
June 30, 2013 10:15 pm

“Records for Desert Rock date back to 1978.”
Maybe, but then they might be stretching their claim:

Ian George
June 30, 2013 10:15 pm

Your post is similar to one I found for Sydney.
On 18th Jan, 2013, Sydney had its highest temp of 45.8C.
It reached that temp at approximately 2:54pm. I followed the AWS record that day and it was showing temps at 10 min intervals (usually it’s 30min).
At 2:49 the temp was 49.9C – the temp at 2:59 was 49.7C.
So in that 10min period, the temp jumped 0.9C and then dropped 1.1C.
Obviously there was some form of ‘spike’.

June 30, 2013 10:18 pm

Is Nevada a desert region anyway?

June 30, 2013 10:21 pm

Sure a new record high would have climatic implications but record lows are only weather.
Here in Canada, David Phillips senior climatologist at Environment Canada was predicting record hell on Western Canada… but since at least on Sunday it did not materialize more than a usual hot summer day, the press, form CBC to Postmedia went on to blabbing on Death Valley albeit mentioning that the 1913 record was not broken. Comments asking about the CO2 concentration back then and suggesting that the media were trying to deflect the fact Canada did not have a record day, were censored by the CBC team of moderators. Inconvenient truths are not welcomed in the network that employs David Suzuki… So much for the $1.1 billion of taxpayer’s money these clowns receive yearly!
Here is what the moderators censored:
“In Death Valley, temps reached about 52 C, according to the National Weather Service. Death Valley’s record high of 56.6 C, set nearly a century ago on July 10, 1913, stands…”
“… will hit the interior B.C. and parts of Alberta on Monday, boosting temperatures in those areas for two days.”
And what was the CO2 concentration in 1913? An inconvenient truth.
As for TWO DAYS of high temps in the BC interior, it must really be unprecedented… journalistic hype! A simple perusing EC’s map of Canada’s temperatures today shows normal summer conditions. I guess when the heat is not here, the press can always agitate the specter of Death Valley with a good chance it’ll be hotter than anywhere in Canada…
Content disabled”

June 30, 2013 10:35 pm

So, Anthony, what are you disputing here, or are you simply trying to call into question NOAA’s credibility? Let’s think for a moment… what were other recording stations saying about Las Vegas today?
Let’s consider Weather Underground, a money making entity rather than a governmental one. Let’s consider a smattering of highs for the day registered in the area, and compare them to what was showing up in LV.
Odd, how going through them, even throwing out ones which anecdotally don’t make sense (e.g. 112 degrees at 10PM…) as a body they do seem to indicate rather high temperatures, generally in agreement with what the weather service was reporting. Is it possible, LV might, just might, be unusually hot, perhaps even historically hot, even without the affirmation of NOAA?
Riddle me this, Mr. Watts; how is this weather normal? Not just in LV, but across the region?
REPLY: First you have to start making sense. This post is about Mercury, Nevada, not Las Vegas, and NOAA, not Weather Underground. It is about two official NOAA stations separated by less than 1/2 mile varying two degrees, one makes a record, the other doesn’t. That’s all there is to it. I call the accuracy of the NOAA ASOS station calibration into question, since it (the design) has a long history of problems as I document here: http://wattsupwiththat.com/2008/01/10/inside-the-asos-ho83-tempdewpoint-sensor/
and here: http://wattsupwiththat.com/2008/01/21/guest-weblog-by-professor-ben-herman-of-the-university-of-arizona-maximum-temperature-trends/

June 30, 2013 11:01 pm

This type of article is what I love about this site. An easy to understand laying out of the facts for the non-scientist like myself.
Bravo Anthony & Crew!

June 30, 2013 11:04 pm

Ian George,
I noticed exactly the same thing. I believe the 2:49 & 2:59 were 44.9C & 44.7C (typos). What is interesting is that the wind direction had already swung from a weak westerly (hot) to a weak easterly (slightly cooler). Maybe the gentle wind shift flushed out a pool of hot air from the nearby Sydney CBD only a few hundred metres away causing a very short term transient peak. It is unlikely that a transient lasting only a minute or two would have been picked up by the old style glass and mercury min-max thermometer which would have been used in 1937 (or was it 1939) when the previous record of 45.3C was set. In other words had the modern electronic thermometer with a much lower thermal mass been used in the 1930s, would the 45.3C recorded it hen have been higher? In addition Sydney’s population would have more than doubled since then. Other factors would be large scale air conditioning plants (unknown in the 1930s) and the huge increase in traffic on the 8 lane Sydney Harbour Bridge the southern approach of which is only 50m away from Observatory Hill.

george e. smith
June 30, 2013 11:27 pm

Well I was watching TV news weather report,yesterday, which was being reported by a woman weather person and a guy, from the Las Vegas strip, which evidently failed miserably to reach the forecasted 119 deg F or thereabouts. I think they had 112, and Death Valley had not got near the forecasted 127 deg F.
So this lady weather geek, was standing on the side walk, and talking about the “unprecedented “heat”. She had a remote IR thermometer, which she pointed at the blacktop paving on the road surface on the strip.
The reading: 173 deg F; that’s 80 deg C. Not as high as the 90 deg CI have heard can be measured on such surfaces.
Well just imagine how fast that paving is radiating, and also the conductive/convective transport of heat energy to the upper atmosphere.
These urban heat islands, are wonderful for cooling the planet; much better than the polar ice.

John F. Hultquist
June 30, 2013 11:28 pm

Under the road sign for Mercury there is a second sign saying ‘no services’. It should say “government services only’ so no one would even bother checking.
Seriously, though, this shows that when a temperature record is proclaimed and there isn’t independent confirmation that record should not be believed. Having readings every minute that fit the “proper” pattern using regularly tested equipment with 360 degree video would instill some trust. Money for that apparently went to alternative energy subsidies. The real alternative would be to not base all of one’s world view on fractions of degrees.

June 30, 2013 11:40 pm

Anyone who has driven from Vegas to LA knows that Needles California is always where it’s REALLY HOT. Googling shows it was 117 today, but was supposed to be 120, neither of which beats the record high for June (121°), let alone July (125°).
“Normal” summer high temps are routinely from 106° to 120°. I’ve driven through there many times and usually expect it to be about 112+.
That place routinely has asphalt that flows under your feet like mud…and right over that asphalt feels like it’s about 130… Crazy hot. I imagine that slab of asphalt in Mercury could easily effect adjacent thermometers.

Steinar Midtskogen
June 30, 2013 11:47 pm

The daily extremes depend a lot on siting and microclimate. So a 2F difference is not suprising for sensors more than 100 meters apart. Both readings may be quite accurate. And by accurate I mean within half a degree F. If their monthly averages differed by 2F, I would be more concerned.
REPLY: If we had varied landscape in this instance that could result in a microclimate, you’d certainly have a point. But this is flat desert, with virtually no variance except the tarmac. – Anthony

July 1, 2013 12:05 am

Now we know why the temps of the past always have to be adjusted down 😉

July 1, 2013 12:09 am

Actually, NOAA claims initial numbers 122° for Needles. http://www.nws.noaa.gov/view/validProds.php?prod=rtp&node=kvef
Surprising because the temperature curves on Google were consistently lower than that.

Brian H
July 1, 2013 12:17 am

The CRN is giving poor “references” to the NOAA, it seems.
BTW, zooming in on Google Earth proves nothing about the real time situation, only that at the time the photo was taken.

July 1, 2013 12:59 am

The CRN station has 5 min readings. But it seems to have a gap between 3.10pm and 4.15pm on the relevant afternoon. That seems to be a period when a max might have occurred.
REPLY: That’s probably a satellite uplink outage/deferral. CRN is low priority in GOES traffic CRN’s do store and forward of high temps though, as shown in the monthly report. Temps in that area tend to peak between 4-6PM BTW – Anthony

July 1, 2013 1:03 am

It seems Badwater can be 3°–5°F (2°–3°C)
higher than at Furnace Creek:
“For visitors seeking the hottest temperatures,
Badwater, the lowest point on the salt flat in Death
Valley, is likely hotter than Furnace Creek. However,
official temperature records are not measured at
Badwater. Robinson and Hunt (1961) report that
Badwater temperature maximums were often 3°F
(ª2°C) warmer than those at Furnace Creek. This
conclusion was probably based on an unpublished
study by a Park Service ranger (Wauer 1959) that reports
temperatures at Badwater were 3°–5°F (2°–3°C)
higher than at Furnace Creek during July and August
1959 (Wauer used NWS quality recording thermographs
in a standard shelter, but he emphasizes the
thermographs were not calibrated daily and thus are
not recognized as official readings).”

July 1, 2013 1:12 am

In the greater scheme of things, none of this would be important. A degree or two when it is hot? No importance. But when international social and economic policy is determined by differences from a long-term mean, 2 degrees becomes a multiple of “normal” variation and becomes important.
When portions count, accuracy to portions is required. Laypeople – and possibly politicians – think that scientists are accurate in all ways in all times to the accuracy and precision of atomic clocks. There is one answer and a Mann in a white-coat can be relied upon to determine it so that a chubby man in a suit can report it to the world.

Stephen Skinner
July 1, 2013 1:24 am

jdallen says:
“June 30, 2013 at 10:35 pmRiddle me this, Mr. Watts; how is this weather normal? Not just in LV, but across the region?
What would you consider normal weather for a desert region situated around the equator?

July 1, 2013 1:31 am

Why would NOAA position two weather stations so close together unless it had good reason for expecting them to show slightly different results?

Steve Richards
July 1, 2013 1:33 am

Surely we all know that temperature becomes more ‘spikey’ with global warming?

Peter Miller
July 1, 2013 1:43 am

Within six weeks of either the summer or winter solstice occurring, there will always be a record high or low temperature occurring somewhere on our planet.
Has anyone looked in the southern hemisphere for a record, or near record, low somewhere?

Ian George
July 1, 2013 2:22 am

Yes. Casino in NSW, Australia had its coldest day on Jun 21st (winter solstice) – well, since 1965 as there are no daily temperatures on the BOM’s website before that year.

July 1, 2013 2:28 am

If you have weather records over a period of a hundred years, the odds that any given year would be a record would be one in a hundred. One percent. That may seem like small odds, but, with 365 days in a year, the odds ought be that you would get between 3 and 4 record highs each year. (Also between 3 and 4 record lows.) Conclusion? “To set records is normal.”
Of course, saying anything is “normal” attracts fewer viewers than appearing on the air and speaking like a person who has consumed too much coffee.

July 1, 2013 2:47 am

temperature measurement is fraught with problems. I am carrying out a small experiment with two digital thermometers, Oregon Instruments, taking max/min readings over periods of 24 hours. The sensors are 20 yards apart one near the house the other down the garden and the difference between max readings can be up to 7C. So this post is not a surprise.

Luther Wu
July 1, 2013 3:46 am

Does anyone really believe that an agency of the United States government would purposely “misinform” us in order to comply with a political agenda?

Billy Liar
July 1, 2013 4:31 am

Nick Stokes says:
July 1, 2013 at 12:59 am
Nick Stokes has a point. Why is there a gap in the CRN data? Where is the missing data?
REPLY: Not concerned. That’s probably a satellite uplink outage/deferral. CRN is low priority in GOES traffic CRN’s do store and forward of high temps though, as shown in the monthly report. Temps in that area tend to peak between 4-6PM BTW – Anthony

more soylent green
July 1, 2013 4:37 am

bushbunny says:
June 30, 2013 at 10:18 pm
Is Nevada a desert region anyway?

They don’t call the place “Mercury” for nothing.

more soylent green
July 1, 2013 4:39 am

dwr54 says:
July 1, 2013 at 1:31 am
Why would NOAA position two weather stations so close together unless it had good reason for expecting them to show slightly different results?

Are you familiar with the American government? Why have one when you can get two for twice the price?

July 1, 2013 5:23 am

At Stokes and Liar,
The station near the airport set its record close to 5 PM. So the missing data is not relevant in this case. (But I do wonder where it is)

July 1, 2013 5:29 am

Bill_W says: July 1, 2013 at 5:23 am
“The station near the airport set its record close to 5 PM. So the missing data is not relevant in this case.”

No, that only says that the record was set in the 6 hrs before 4.52pm. The highest hourly reading shown was at 3.52pm – right in the missing section at CRN.

Bob Koss
July 1, 2013 5:33 am

Nick Stokes,
It appears you may be correct about the highest 5 min reading not being shown above.
I used the actual CRN files found linked below here. http://www.ncdc.noaa.gov/crn/qcdatasets.html
Looking at the hourly station file shows a Tmax of 44C(111.2F) for the 1500 hour. This can only be if they are using 5 minute data no longer shown.
I checked all the Nevada CRN 5 minute station files. Beginning on June 6 all of them started showing only one reading 10 minutes after the hour. Maybe they are intending to reduce the file sizes in the future if no one complains. They did split the 5 minute station files into individual years a few months ago to reduce size due to the file size becoming unwieldy.

July 1, 2013 6:01 am

LOL. The CRN station actually had a warmer daily mean by 0.1F Anthony! Try harder next time.
CRN: 111.2/75.0 mean: 93.1F
ASOS: 113/73 mean: 93.0F
REPLY: Are you dense? I’m calling the daily high record into question, not the daily mean. I’m concerned the ASOS might be out of calibration. The HO83 thermometer on ASOS has a tendency to do that and rather quickly. The way these fail tends to be in the Tmax they report, as a string of false ASOS HO83 high temperature records in Tucson that still remain uncorrected in the 1980’s can attest, along with subsequent peer reviewed papers studying the problem. Can you say with certainty that the record high in Mercury is accurate? Try harder to understand the instrument next time. – Anthony

July 1, 2013 6:30 am

Confucius say,” Man with two watches never know what time it is”.

July 1, 2013 6:54 am

It’s worse than we thought — a hot desert in summer — and all caused by man’s sins & evil CO2.
Time to sacrifice a virgin, dunk a witch, or turn off a light.

Gary Pearse
July 1, 2013 7:00 am

Can we get some explanations from NOAA to be put into another post. I think all this should be reported to the House committee dealing with this kind of stuff. Is this “false advertising” or worse malfeasance? We now helplessly wait for the certain new world ‘record high’ from Bad Water in Death Valley which was set up at the focus of a natural parabola of landscape facing west just to get the new high. Anthony, this is the flip side of your investigation in the Surfacestations Project. Now they are using the elements of bad siting to get what they want. Also, it represents a brand of cynicism on the part of taxpayers’ activist employees that only comes from ‘elites’. They count on ignorance and trust of the majority of citizens who are kept in the dark by an adoring and uncritical press. Someone handy to the Bad Water unit should paint it black so we will have a humdinger of a record.

July 1, 2013 7:14 am

I agree with Gary. This post is an example of why I come here first every morning. It’s accurate and fair.
Then I go to Tamino’s “Open Mind” for the other kind.

July 1, 2013 7:23 am

Fan aspirated radiation shield Met One used in USCRN is known to be better than the model in ASOS (a few too hot with the strong solar radiation)
The instantaneous temperature to calculate the maximum temperature (Tx) is not the same in the USCRN, because the fan in the Met One change the time constant and would show too strong Tx.
An instantaneous measurement every 10 seconds, is the average / 5 mn of all samples (the standard in WMO station is the average / 1 min of all samples). See note for the temperature in this page :
You can see the problem when we use a ventilated radiation shield without a good average to define the instantaneous temperatures ( to calculate Tn and Tx) here in my weather stations, with the modern stevenson screen (in ABS) used at Météo France (MF) and the ventilated Young 43502 (Yng) in CRN 2.
The WMO instantaneous temperature (avg/1 mn all samples 10 s, in °C) :
The fan vents at 6 m/s in the Young 43502, it makes a strong noise in the temperature (with the modification of the constant time. The sensor in the shield, captures all the small and too short variations of the micro scale)
The max of each minute with the too short average, gives too strong Tx with a good ventilated radiation shield : http://meteo.besse83.free.fr/imfix/relev%e9svtgagro_fichiers/image003.gif
The average / 5 mn gives good Tx in the ventilated radiation shield ( too hot if you use the WMO standard / 1 mn in these radiation shields).

Jeff Alberts
July 1, 2013 7:42 am

How can they say “all-time” high when the “official” records only go back to 1937? More fear mongering.

Ian W
July 1, 2013 10:10 am

NOAA has at considerable expense created a Climate Reference Network, yet when creating climate metrics it uses other observation sites and does not provide the CRN data to the media. I would have thought that NOA needs to explain its expenditure to the GAO. NOAA justified the CRN on the grounds it MUST have a valid WMO ISO compliant observation network yet it continues using the older network invalidating its justification for funding.

Mark Albright
July 1, 2013 10:56 am

An ASOS warm bias can also be found at Kingman AZ. The Kingman ASOS is showing a +2.4 F warm bias relative to the regional CRN site co-located at the Kingman airport. Here are the 3 readings for the past 3 days:
Kingman AZ
Date RCRN ASOS Difference
28 June 108.5 110 -2.5
29 June 109.2 112 -2.8
30 June 108.0 110 -2.0

July 1, 2013 11:46 am

As I pointed out yesterday, an AP article stated that the high in Death Valley National Park was 128, but included a line regarding a LA Times article stating that the high was 129.9 F at a station 200 yards away.

Gail Combs
July 1, 2013 4:49 pm

jdallen says:
June 30, 2013 at 10:35 pm
Let’s consider Weather Underground, a money making entity rather than a governmental one.
Weather Underground is run by warmists as I noted a few days ago. I watch my local weather very closely and Weather Underground revises the daily high temperate up by 2 to 3 degrees F a day later for our rural station.

Wunderground’s Climate Change Position
Earth’s climate is warming. This time, humans are mostly responsible, and the overwhelming majority of climate scientists agree. Climate change is already causing significant impacts to people and ecosystems, and these impacts will grow much more severe in the coming years. We can choose to take economically sensible steps to lessen the damage of climate change, and the cost of inaction is much higher than the cost of action.

July 1, 2013 6:16 pm

As I type this the (weather.com) weather widget in my Mac is displaying the Sydney temperature as 16C while the BOM has it at 13.3C

July 1, 2013 8:56 pm

I can’t believe this report. Look at the minimum temps too. They drop alarmingly at night.

Nigel Harris
July 2, 2013 2:09 am

You dismiss Nick Stoke’s comment pointing out that there is a gap in the data from 3.10pm to 4.15pm with a casual “Temps in that area tend to peak between 4-6PM BTW”, and later comment that you are “Not concerned.” and repeat the assertion that temps peak between 4 and 6pm.
Well, a quick look at the 7 days of easily available data for the station say otherwise:
June 25 – high of 92F first reached at 3.10pm and last recorded at 4.05pm
June 26 – high of 99F first reached at 3.10pm and last recorded at 4.10pm
June 27 – high of 104F first reached at 3.10pm and last recorded at 3:50pm
June 28 – high of 107F first reached at 4.00pm and last recorded at 5.45pm
June 29 – high of 110F recorded at 4.00pm
June 30 – data missing from 3.10pm to 4.15pm; highest reported of 110F first reported at 1.55pm and last at 5.10pm
July 1 – high of 110F recorded at 3.45pm and 4.15pm
It seems to me that a gap between 3:10 and 4:15 is actually rather crucial. It would have missed the highs recorded on five out of the six days with complete records.

Mark Albright
July 3, 2013 12:52 pm

Since the Desert Rock CRN site was installed 10 years ago the temperature has exceeded 109.0 F seven times. On these 7 days the ASOS maximum temperature has averaged +1.6 F warmer than the CRN site located nearby.
18 July 2005 109.0 F
6 July 2007 109.6 F
1 July 2013 109.8 F
29 June 2013 109.9 F
5 July 2007 110.3 F
4 July 2007 110.7 F
30 June 2013 111.2 F
18 July 2005 112 F
6 July 2007 111 F
1 July 2013 112 F
29 June 2013 111 F
5 July 2007 112 F
4 July 2007 111 F
30 June 2013 113 F

July 3, 2013 3:42 pm

Anthony, you said : I call the accuracy of the NOAA ASOS station calibration into question, since it (the design) has a long history of problems as I document here: http://wattsupwiththat.com/2008/01/10/inside-the-asos-ho83-tempdewpoint-sensor/
and here: http://wattsupwiththat.com/2008/01/21/guest-weblog-by-professor-ben-herman-of-the-university-of-arizona-maximum-temperature-trends/
So … if it’s a known issue already documented, then what’s the deal to bring that subject again ?
A long story that could be resumed with .. The station at Mercury is still showing innacuracy. When will NOAA fix it ? and wait for the response.
Have a nice summer.

Grant W. Goodge
July 3, 2013 8:23 pm

Thanks for the post Mr. Watts: If you check the 5 minute data for the Mercury site you will find that the 111.2F max at Mercury on June 30, 2013 did come from a 5 minute observation. I was going to send you a screen capture of the temp plot but I don’t see any provision for an attachment

%d bloggers like this: