How NOT to measure Earth's temperature

[Dr Mordrid]

You can't make this stuff up

There is a growing photo database of the US weather stations used to construct their temp records, and WAY too many of them don't come close to making sense or meeting their own guidelines; away from heat sources, heat sinks (expanses of concrete etc.), nearby buildings, vehicles etc. The ideal;



This one in Tahoe, California is a prime example of the opposite;



And this one in Lovelock, Nevada has its own personal tarmac (read: heat sink) 10' away, complete with the exhaust of an old MIG 15 (?) fighter aimed in its general direction;





And Roseburg, OR only 4' above a roof, the hottest part of a building, and just a few feet from an A/C hot air exhaust vent;



Europe & the UK aren't immune. Here is the station at the Royal Observatory at Edinburgh, which sits in the middle of a concrete & stone heat sink;



Might want to check your local station to see how its layout stands up to scrutiny

A ton more here;

http://www.norcalblogs.com/watts/weather_stations/

http://www.surfacestations.org/ has set about documenting them.

[TransformX]

I'll put one in the oven, just in case

[Brian Ellis]

This is a serious problem. I have difficulties myself and I tried at least ten sitings before settling on the best compromise. However, there are many other factors, such as radiation errors. The Stephenson screen (the white boxes in your 2nd and 3rd photos), which is the standard protection, will give different temperature readings according to the wind speed and when they were last painted. There are much better designs possible.

Worse, in modern systems the thermo- and hygro- meters are usually in the same housing. According to NOAA recommendations, the thermo, in its Stephenson screen should be above grass at (I think) 1.2 m height. But the hygro- in the same screen will read much too high because of evapotranspiration, especially if the grass is long. Your first photo is not ideal, either, because the three TH units (unscreened!) are more than the recommended height. Have you thought why they need three?

In my case, I chose my siting as a compromise, by comparison with calibrated lab instruments. The TH unit is at 1m60 above a red concrete paving. The screen is much more efficient than a Stephenson screen, having three sets of louvres between the sensors and where the sun can strike it. There is no apparent drop in temp with high winds on a sunny, summer day (impossible with a Stephenson screen). The increase in temp due to the concrete is measured at ~0.5°C on a very hot sunny day and I consider this as acceptable. The hygro is 5-10% RH less than when I tried it over grass (~20% just after the evening irrigation) and I consider this more important than a small temp error.

Much more prone to errors is wind. The recommended height for anemometers is 10 m above ground level with no obstructions for 360° around it. Most weather stations are on airfields (METAR) and this is usually impossible within easy access, so they stick it on top of the control tower, usually more than 10 m. As wind is braked close to the surface, this means they tend to read much higher than if they were at 10 m. For practical reasons, mine is at 6.5 m and has errors due to some pine trees at ~25 m distance and the crest of the roof of the house at 8 m. These errors are small and I have more or less compensated them at a software level.

Pluviometry is also error-prone. Modern rain gauges are usually of the tipping bucket type, rather than the traditional funnel-in-a-cylinder type. Properly maintained, they are roughly of equal accuracy, although the tipping bucket is more sensitive (mine can measure as little as 0.2 mm of precipitation). The aerodynamics of the TB type is better, too, as the funnel design is more vortex-free (when the rain falls near-horizontally!). However, the downside of the TB type is that it requires maintenance: spiders love to spin their webs around the bucket mechanism! This may reduce their accuracy much more than you would think.

Some comments on your photos.

1. Design of 3 TH sensors is not brilliant, with the parasol screen: OK at high solar angles, but I would expect radiative errors near dawn/dusk.

2. I suggest the site is a rubbish dump rather than owt else. The incinerator is on a wooden palette which shows no sign of scorching; it probably has not been used there. The tennis court is far enough away to avoid radiative errors; those from the design of the screen plus an apparent lack of cleanliness are probably more important. In fact, I suggest the screen has been dumped there as rubbish, as access (essential) is restricted.

3. The biggest errors are certainly due to the state of the Stephenson screen that looks as if it hasn't been repainted for 20 years! Difficult to get an idea of the distances, but the building looks as if it is probably 20 m away: I suggest the hot air from the aircon has dissipated long before it reaches the TH unit. It would seem from the shadows that the anemometer mast is not 10 m and it would be shadowed somewhat by the control tower. The hard path is permissible.

4. No comment!!!

5, I happen to have been there. That is not a weather station in the strict sense of the word. It is a prehistoric Stephenson screen which used to be used for manual readings, housing a barograph, thermograph and hygrograph. It was used for compensating atmospheric refraction for the telescope observations. It is at the top of Calton Hill, right in the city centre, <500 m horizontally from the famed Princes Street. The second one was at Blackford Hill, just 3 km away. You can see the two telescope domes top right, but both the city observatories are now used only for school trips and not for serious astronomy (too much pollution, especially light pollution: they have built a better observatory elsewhere where they do much in the way of cosmic particle research). The thb-graphs are clearly no longer used as there is no path worn in the grass to it!

You can see the proximity of the observatory to Princes Street:


Speaking more generally, I agree that many weather stations leave a lot to be desired, but you have to divide them into categories:

1. Those run by national meteorological services These are generally serious and are homologated by the IMO (headquarters Geneva). These are the ones used mainly for global climate determination. The national services have the power to accept data from other stations, but only if they follow strict guidelines (e.g. NOAA in the USA)

2. METAR stations. These are mainly for aviation and give limited data (e.g., many don't measure rainfall). They are divided into 3 categories, roughly:
a) those at major international airports, often full-scale homologated stations
b) those at smaller airports, usually bare-bones for aviation purposes only
c) those at minor airfields, often limited to wind direction/speed and barometric pressure.

3. Weather ships/buoys, run by the IMO or national services

4. Research institutes, academic stations etc. Some are serious, many are relatively primitive.

5. Other ones, by far the most numerous, often used in industry for various purposes, often with limited data. There are also an estimated 200,000 relatively serious amateur weather stations around the world (not talking about $50 cheapo stuff here). There are three sub-categories here:
a) consumer units, such as La Crosse ones, which are fairly doubtful in quality (I started with one of these, but upgraded after a year) ~$200
b) prosumer units, such as Davis ones (which I now have), in the $500 range. These are robust and accurate, approved by the NOAA for secondary pro use.
c) professional units, costing up to $15,000

You can see the efforts made by several hundred amateurs here. Some of these guys (especially in the USA) have been officially approved by the NOAA and other organisations. Mine is here, if you wish to see it.

[Dr Mordrid]

Brian;

that Tahoe station is the official station of record; GISS ID 425724880010. BTW: I've seen burning barrels on palettes before and they usually don't burn; heat rises and the thermal mass of the barrel is enough to mitigate a few papers getting cooked. Besides that there are usually air inlets cut in the bottom, hence the need for the palette, and the trash placed on an old circular BBQ grill above.

One more for good measure: Santa Rosa's NOAA MMTS is in a sea of A/C units (5 in a wider shot);



and San Diego's at the Natl. Weather Service facility;



The more I see the more I think these are not exceptions.

[Brian Ellis]

Now you're going from the sublime to the ridiculous.

Image 1. The TH unit is on a mast clearly planted in the ground (not on the roof). This mast is a standard design used for pro equipment and is 10 m high for the anemometer. It is nowhere near the exhaust unit, as is implied by the cleverly foreshortened photo, taken just at the height of the flat roof to deliberately eliminate any possibility of scaling.

Image 2. This is a deliberate disinformation photo. Can you really believe that one of the nation's best-equipped (and most important) weather stations looks like this? Remember that the US Navy has an important base on Coronado Island and will rely strongly on data from SD NWS station. In fact, the photo wasn't even taken at San Diego, but at Laguna Beach, which, if my memory is correct, is about 100 km north of SD and not even in SD county. Anyway, it clearly shows a disused Stephenson screen, visibly falling to bits, sitting in a rubbish yard. The fact that it is not even accessible shows that it is not in use, even if it were usable. I would say that your ascribing it to the SD NWS as being current equipment is tantamount to libel and is certainly calumnious. You should be more careful with what you post.

FYI, you should be aware that hundreds of Stephenson screens have been abandoned in recent years, as the old baro-, thermo- and hygro- graphs have been replaced by fully automatic electronic systems which often have more sophisticated radiative screening: even my prosumer equipment is better than those old wooden louvred boxes. That is why you will find many of them dumped in all sorts of rubbish yards.

Not having seen for myself what a US NWS weather station looks like, I find it very hard to believe what you are implying by this thread.

This is a typical regional airport weather station:

and this is a cooperating NOAA/NWS (not owned by them) in Texas:


If these are minor examples, you can imagine the equipment in official NWS sites.

I think you would be wise to stop this campaign of denigration.

[Technoid]

"Any evidence that not comply with the desired results should be disdarded"
The new universal golden rule for science

[TransformX]

Tons of pictures here, pick your favorites: http://www.wrh.noaa.gov/mtr/cpm/stations.php

[Dr Mordrid]

Now you're going from the sublime to the ridiculous.

Image 1. The TH unit is on a mast clearly planted in the ground (not on the roof). This mast is a standard design used for pro equipment and is 10 m high for the anemometer. It is nowhere near the exhaust unit, as is implied by the cleverly foreshortened photo, taken just at the height of the flat roof to deliberately eliminate any possibility of scaling.

USHCN Station 47965

Top view. Note the shadow of the tower, which is on the roof and not unusual on US buildings. Our police HQ has a huge radio/MW mast on its roof. That "cleverly foreshortened" vent is a couple meters left of the towers base at about 7:30;

Image 2. This is a deliberate disinformation photo. Can you really believe that one of the nation's best-equipped (and most important) weather stations looks like this?

Given the incineration barrel at Tahoe and the Roseburg OR site you yourself said "no comment!" to, YES!! Government agency stupidity knows no bounds, which is why so many people are loathe to trust a lot of this "data".

Enough for talk, here's the PROOF that this is a WORKING STATION under NOAA CONTROL;



That image is directly linked to the NOAA page listing their stations in their SAN DIEGO OFFICE catchment. It has been active since 1943. Look under "Laguna Beach", as you should have before spewing.

http://www.wrh.noaa.gov/sgx/cpm/station.php?wfo=sgx

I think you would be wise to stop this campaign of denigration.

Not when denigration is due, or overdue in this case

Well?