2737: Weather Station

Explain xkcd: It's 'cause you're dumb.
(Redirected from 2737)
Jump to: navigation, search
Weather Station
'Pour one out for precipitation data integrity,' I say, solemnly upending the glass into the rain gauge.
Title text: 'Pour one out for precipitation data integrity,' I say, solemnly upending the glass into the rain gauge.


Weather stations are usually equipped to measure atmospheric conditions for weather forecasts and studies on the weather and climate. An anemometer is a device often found in such stations that measures wind speed and direction. The anemometer shown in the comic is a typical three-cup anemometer, whose spin rate is proportional to the wind speed. Thus spinning the cups quickly by hand can create an impression of fast winds and thereby excite any computers that happen to be monitoring the data and enjoy extreme weather.

Combining the speed of winds in a F1 tornado (73~112 mph; 117~180 km/h) and the speed ratio of typical three-cup anemometers (1/2~1/3) gives a cup speed of ~16m/s, comparing the diameter of the anemometer to that of Cueball's arm gives an upper estimate of 0.5 meters (~two feet), topping out at roughly 10 turns per second, which is within a reasonable range for a human[actual citation needed].

The title text continues the trend of messing with weather equipment, this time by messing with rain gauges. "Pouring one out" usually refers to the act of pouring a liquid, usually alcohol, on the ground as a symbol of reverence for a deceased friend or relative. Thus Cueball is mourning the now destroyed (aka "deceased") integrity of the precipitation data by pouring a glass of liquid into the gauge. However, it is this very act that ruined the integrity of the data.


[Cueball is standing halfway up a stepladder, with his back to the ladder, on the top of a building. He is leaning over a tall device on a pole with a large rectangular box with a stripe. On top is an anemometer, which he is spinning quickly. Behind him, there is a hatch leading out onto the roof with a small antenna, and two boxy roof vents facing towards him.]
[In the lower-right corner of the panel, a medium sized circle shows a different scene. An array of computers are shown on server racks, connected to each other by cables. A sound is coming from one of the computers:]
[Caption below the panel:]
Whenever I see one of those little weather stations, I have to fight the urge to climb up and spin the anemometer real fast to make a computer somewhere think it's in a tornado.

comment.png add a comment! ⋅ comment.png add a topic (use sparingly)! ⋅ Icons-mini-action refresh blue.gif refresh comments!


I'm not sure if the comment about the spinning speed in the explanation section is needed, still, added it for clarity. Feel free to remove it if it seems out of place. Xkcdjerry (talk) 07:13, 14 February 2023 (UTC)

Totally needed. -- Hkmaly (talk) 22:14, 14 February 2023 (UTC)
It's what Randall would have done Boatster (talk) 22:28, 14 February 2023 (UTC)

Wheeeeeeeeeeeee! Mushrooms (talk) 10:57, 14 February 2023 (UTC)

It would be great if the explanation also offered a comparison between Cueball's arm speed and the wind speed of most tornadoes, as knowing the speed that Cueball's spinning it at in a bid to make it think there's a tornado doesn't mean much if one doesn't know how fast tornadoes also go. -- 13:46, 14 February 2023 (UTC)

FYI the rain gauge in Borrowdale, Cumbria, England - reputedly the wettest place in England - is set into a wall at thigh-height, with a collecting spout. A local farmer told me that as a boy, the gauge was between the school bus drop-off and the farm, so he and his brothers used to regularly top it up by taking a p1ss....distorting weather records has a long history! 15:34, 14 February 2023 (UTC) 15:34, 14 February 2023 (UTC)

And wait until chinese weather balloons come into play ;-) -- 17:59, 14 February 2023 (UTC)
They don't. That's the point of shooting them down. -- Hkmaly (talk) 22:14, 14 February 2023 (UTC)
I haven't been watching Chinese weather reports, but have they been reporting that the weather over North America is explosive? Barmar (talk) 22:17, 14 February 2023 (UTC)

No WAY "ten turns per second" is in a reasonable limit for a person. That's insanely fast! --Mushrooms (talk) 07:56, 15 February 2023 (UTC)

Sorry, my bad :( should have made it clear that it meant something along hitting it with a hand to get it turning and not the hand rotating with it (which I agree would be ridiculous), any idea how to clarify that without making the paragraph too long? Xkcdjerry (talk) 08:33, 15 February 2023 (UTC)
Afterthought: I tested the 10 r/s with a ruler, a aero-whatsit would probably have less friction from the bearings but much more air resistance so it might not be easy to get it up to speed. However I do not have the means to test *that* scenario. And I'll be ever so grateful if someone with the means would check if it can be done. If it can't the last sentence would be better off as something like "This speed is not acheveable by hand but can be reached by whatever" Xkcdjerry (talk) 08:51, 15 February 2023 (UTC)
For the record, the top speed reached by a ping-pong ball (which should be a bit above the top speed of a human arm dut to the bit of length added by the racket) is 112.6km/h. So a bit less than what's needed for the minimum speed quoted. 11:46, 15 February 2023 (UTC)
Turning manually runs quickly into drag — cube of speed — so it is unlikely that high rotational speeds can be achieved. 12:27, 15 February 2023 (UTC)

Using the drag equasion from google, a speed of 16m/s per the article, a drag coefficant of 0.38 when the air is blowing at the rounded side and human power 600W (10W/kg times an average-if-a-bit-low 60 kg) gives an estmate of max r of ~43cm (which means it measures ~3 foot across), it's a rough calcuation but it should suggest that power itself shouldn't be an issue. The problem is how fast we can get the power flowing from the arm into the object (which is going to be the real pain). Xkcdjerry (talk) 13:13, 15 February 2023 (UTC)
OK this is in no way revalent but when I calcuated the above figures I decided to plug in a cup that's 20cm across to see how fast we can make it go if we managed to dump all 600W into it and found we can make it spin at 43m/s (96mph) and make the computer think wind speeds are at a ridiculous 128m/s (289mph) which according to this chart is a F5 "incredible tornado" capable of, cite: "Strong frame houses lifted off foundations and carried considerable distances to disintegrate; automobile sized missiles fly through the air in excess of 100 meters; trees debarked; steel re-inforced concrete structures badly damaged." (Which somehow this feels oddly satisfiying) Xkcdjerry (talk) 13:13, 15 February 2023 (UTC)
Bear in mind that we're twiddling the 'cups' with the nominal wind, not against the (much more) static air. The force needed to rotate the cups by air is necessarily more than the force needed to send the cups rotating through air. (In normal operation, the cups on the 'unreceiving' side of the spin are moving into the wind that is forcing the 'receiving' cups to move. If wind is W and Creceive is ⅔rds that, as is suggested, then Copposing is going at (5/3)W, relatively. Which is probably why it actually reflects that ½ to ⅔ the wind, proportionately.) And the hand does not need to go the speed of the cup, it might only need to twiddle an index finger around near the base of a cup-arm. For 10rpm, it just needs to sustain a revolution every six seconds, and I can make a good slow circle at maybe 15cm radius with my finger, by flexing my wrist alone, if I need to do that to overcome rotational friction that twiddling nearer the axis would be difficult to do.
(Noting that I'm sure that most anemometers are smaller than described, above, meaning they need to spin faster to give a similar windspeed. But also of negligable intrinsic friction (beyond the half 'counter-flow' resistance) so that almost the lightest breeze still conveys some decently proportional measurement of rotation. And I think I could easily 'force' such a device round at 60rpm or more for enough time to emulate the core of a passing tornado. In fact... I just now hastilly constructed a Lego 'spinner' with two large, flat plates that are fixed flat on to the rotational direction and, despite being omni-resistant to any airflow, I can rotate the pivoting beam around quite happily at a generous speed, without any of the streamlining of the 'non-cupped' sides of the cups. It generates quite the localised breeze, that I know isn't exactly full-strength tornado but then a full-strength tornado wouldn't turn this particular device very fast until it was basically at (or in) the vortex edge, for the necessary airflow differentials across the vanes, and I'm not sure even Lego would withstand those conditions.)
Bearing in mind the uncertainties of the particars of the device, the calculations are good but not necessarily the final word. I'm sure that instantaneous 'twister'-level rotations could be managed, perhaps even sustained, without too much difficulty. 15:11, 15 February 2023 (UTC)
Addendum (from '') - Whoops, I completely misread r/s as rpm. So less possible, but I still think I could twiddle something at nigh on 600rpm from a twiddle of a finger. (With an unloaded finger I can get 160-odd 'small circles' in 20 seconds, perhaps suitable for rotating the very centre of a rotor-bar of the right sort. The finger actually being slightly quicker than I can reliably tally the revolutions whilst keeping an eye on the stopwatch - but the muscles begin to burn a bit if I try to go for much longer than that without a rest.) Full on practical experiments with 'roadside' anemometers will have to await some chance I don't know if I'll ever get, though. (The nearest "container-based" environmental monitoring station is in the middle of a large roundabout near me, with near constant traffic, and I seem to think has been made vandal-proof (and thus "human tornado"-proof) around the top, to make it even more of a challenge. Until then, I'd have to consider some ersatz self-built rig of my own, for which I'd need to hook up a good counter/frequency-meter as well. :P 15:54, 15 February 2023 (UTC)