Wikipedia:Reference desk/Archives/Science/2016 January 11

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

January 11

Wind direction confusion

I always see sat maps showing clouds/wind moving one way and weather reports showing it moving the other way. What's going on here? [1] [2] [3] Thanks. Anna Frodesiak (talk) 01:22, 11 January 2016 (UTC)

I remember learning that an ocean current flowing from the north towards the south (for example) is called a southerly current because it is heading towards the south. Conversely, when the wind is blowing from the north towards the south (for example) it is called a northerly wind because it is coming from the north. So when you see clouds and weather systems moving south, the weather reports will talk about northerly winds. Dolphin (t) 01:29, 11 January 2016 (UTC)

Yes, actually, I agree. What the heck is a north wind? Does it come from to or head to the north?. μηδείς (talk) 01:29, 11 January 2016 (UTC)

See the first sentence in each of north wind, south wind, east wind and west wind. Dolphin (t) 01:36, 11 January 2016 (UTC)

East wind comes from the east, etc. Anna Frodesiak (talk) 01:56, 11 January 2016 (UTC)

Ah, but is that direction referenced to magnetic north or true north? The answer depends! METAR winds are true, and ATIS winds are magnetic, except Digital ATIS; and winds aloft are either true or magnetic depending on your software; and if you go to official sources, wind direction is coded... "as a number between 51 and 86 (vice 01 to 36) when the wind speed is 100 knots or greater. To derive the actual wind direction, subtract 50 from the first pair of numbers. To derive wind speed, add 100 to the second pair of numbers. For example, a forecast at 39,000 feet of "731960" shows a wind direction from 230 degrees (73-50=23) with a wind speed of 119 knots (100+19=119). Above 24,000 feet the temperature is assumed to be negative, therefore the third pair of numbers indicate a temperature of minus 60 degrees Celsius." Simple, right? Nimur (talk) 02:58, 11 January 2016 (UTC)

Well, describing a wind as "north" or "east" is comparable to giving just one significant digit of its direction. In most of the world, it would make no difference if you did mean magnetic north. --76.69.45.64 (talk) 10:08, 11 January 2016 (UTC)

Yeah - that's rather extreme pedantry. Winds don't even travel in a straight line, they follow the local pressure gradient and are affected by the coriolis force - so specifying the wind direction to the kinds of precision where you'd notice the difference between true and magnetic north is kinda silly. The exceptions to that are:

1. when the prediction is for the very immediate future - and at a single point on the ground (such as an airport) - which is why those various flight reporting services actually do care which they are talking about.
2. when you are actually near to the poles, where the difference due to those two descriptions is large.

But for general weather reporting, north, north-east, east, south-east, etc is sufficient precision. For maritime weather, they may talk about north-north-east, and east-north-east and such to get a little more precision.

Of course, if you actually are at the North pole then all winds are southerly (or, arguably, northerly) anyway - so you'd really need some kind of alternative description!

SteveBaker (talk) 13:59, 11 January 2016 (UTC)

And the weather reports are right as far as the ground-level wind direction. If it says north wind and you go to the park, the wind is coming from the north. That is always so. The odd thing is that the sat maps always show the coulds going the other way. Is this some sort of illusion where the air is squeezed and makes clouds appear so the wind is going from right to left and giving the appearance of the clouds going the other way? Anna Frodesiak (talk) 01:58, 11 January 2016 (UTC)

The wind direction often changes with height -- see e.g., wind shear and thermal wind. But if satellite imagery always shows clouds moving in the exact opposite direction from the reported winds, this suggests either a misunderstanding of how wind direction is interpreted or an error in the satellite animation loop (with the images sequenced backward in time). Shock Brigade Harvester Boris (talk) 02:05, 11 January 2016 (UTC)

It seems to b always with all weather reports and all sat loops from different sites. Anna Frodesiak (talk) 02:35, 11 January 2016 (UTC)

OK, something is messed up. If you can happen to have a current example I'll take a look. Shock Brigade Harvester Boris (talk) 02:44, 11 January 2016 (UTC)

Most winds reported for normal "retail consumers" - at popular weather websites, and television or radio news services - report only surface winds. Those can be completely different from the prevailing winds at altitudes where the clouds are. If you want to see winds aloft, you need to use a more sophisticated weather service, like the Aviation Weather service winds forecast. World Area Forecast grids are available for most of the volume of planet Earth's atmosphere, including some pretty sparsely populated areas like the northern polar regions. It takes a little skill to use those data products, because they're detailing information about a 3D volume of air movements. Sometimes, you can get a better idea of bulk air mass movement from the satellite or RADAR graphical products, or the prog charts, instead of trying to interpret wind forecasts and observations.

I also use ForeFlight, a commercial software, for supplemental weather information. It has a great user interface for looking at winds, and it provides a global wind database.

Right now, at Haikou, Hainan, China, (where our original questioner was asking about) I see northerly winds at the surface, up to about 3,000 feet AGL, and at 6000 feet AGL the winds shift toward the prevailing trend - winds out of the west - with speeds increasing all the way up to the flight levels. (Note that ForeFlight and its data are not free software: good quality, reliable weather is worth paying for!)

Down at the surface, you're probably seeing a sea breeze; it blows inland. It's reported out of the north because the weather comes from ZJHK, on the north of the island. On the other side of the island, at ZJSY, the current weather report indicates wind out of the south - in other words, the surface wind points inland all around the island!

Nimur (talk) 03:05, 11 January 2016 (UTC)

Very interesting and informative. I actually have been using http://aviationweather.gov/obs/sat/intl/ for years. Good site. So this is no illusion, right? I mean, clouds are actually moving from left to right while the reports say right to left? Anna Frodesiak (talk) 03:25, 11 January 2016 (UTC)

Correct, cloud tops over Hainan right now are well above the altitude where prevailing winds would move them "left to right" (winds out of the west, blowing clouds toward the east). Meanwhile, surface winds all around the island are in various other directions. |f you're looking at infrared or visible satellite imagery, you're usually looking at cloud tops - so the satellite loops indicate winds at higher altitudes. When you watch RADAR, it's more complex - you might be looking at cloud bases, composite reflectivity, precipitation, or some other advanced RADAR product. Nimur (talk) 03:31, 11 January 2016 (UTC)

Ouch, stupid me didn't notice the links in Anna's original post. Sorry about that. Anyway, you can find upper-level charts from various sites. Here is one example. You can click "Southeast Asia" for the map that includes Hainan, and the choose whatever pressure level you like: 850 hPa is around 1.5 km altitude; 700 hPa is around 3 km; 500 hPa is around 5.5 km; and 300/250/200 hPa are around 10/11/12 km respectively. If you look for example at the 500 hPa chart (a reasonable level to choose if you don't know what else to pick) you'll see that winds are out of the southwest, whereas the surface wind reports at Haikou that Anna linked are out of the northeast. Shock Brigade Harvester Boris (talk) 03:37, 11 January 2016 (UTC)

(Indeed, one thing to be wary of: if you're using American data sources, like National Weather Service or the links posted above - and you're looking at international forecasts - get ready to dance around a mish-mash of confusing units. You'll sometimes see feet and hectoPascals (!), meters-per-second and knots, and various "standard" barometric conventions, all mushed together in the same data report. Watch out for unit- and convention- changes, especially if the correct answers actually matter)! Nimur (talk) 03:43, 11 January 2016 (UTC)

Don't get me started on "hectopascals"... Shock Brigade Harvester Boris (talk) 03:58, 11 January 2016 (UTC)

IMO, there is no reason to use a convention, such as "East wind" meaning "coming out of the East", when one can so easily just say that. After all, for any convention there is sure to be somebody who doesn't know it or who uses a different one. StuRat (talk) 04:10, 11 January 2016 (UTC)

StuRat, nearly everything pertaining to modern, systematic meteorology has its roots in practical scientific advances that were designed for aviation applications. For example, you can read our article on the history of surface weather analysis. There is a reason we have procedure and terminology and conventions - lots of good reasons, actually.

Wind direction has a well-defined correct standard phraseology. Unlike a few of the minor details we described above - like metric and standard units, or the true- versus magnetic- direction details, deviating from this convention just means that you are reporting the wind incorrectly.

Brevity has value: to quote the AIM, "Brevity is important, and contacts should be kept as brief as possible." If it's a windy day and I'm landing an aircraft, I can call "wind check" over my radio, and get an immediate response, exactly two numbers that I can exactly interpret in an automaton-like fashion. As much as I enjoy verbose discussions about the philosophy of semantics - from the comfort of my home! - I don't want to have that conversation when I'm on short final and I can't see the wind sock.

If I want a long discussion about wind direction, ... I can read the Area Forecast Discussion, which is written in plain English. Even in these products, meteorologists still use phrases like "east wind" or "southerly wind." If you are unfamiliar with weather terminology, you can use National Weather Service's web-based, HTML web format with Glossary hyperlinks right inside in the AFD.

Non-standard phraseology contributes to accidents and incidents and generally increases workload for everybody. Standardized weather reports mean that every place we go, every local language we speak or use, every unit we use to measure, our interpretation of wind is consistent. Even the reporting-order for the weather information is standardized. Weather observers have to study standardized reporting methods and pass a certification before they can disseminate the weather report. This is why, for example, I can look at a weather brief from China and make sense of the winds.

Standard reporting also simplifies data aggregation for commercial users, weather and climate researchers, and so on.

If you're a casual consumer of weather information, none of this "convention" matters to you - but that doesn't change the value or importance of standardization for those who depend on it.

Nimur (talk) 05:56, 11 January 2016 (UTC)

Are you certain that there is no place in the world which uses the reverse convention ? StuRat (talk) 06:05, 11 January 2016 (UTC)

There is no place that correctly uses the reverse convention - at least, not while using the English language. Again, this is covered in the article, wind direction. Nimur (talk) 06:10, 11 January 2016 (UTC)

That sounds a lot like the No true Scotsman fallacy. BTW, why not concern ourselves with people who speak other languages ? StuRat (talk) 07:03, 11 January 2016 (UTC)

Science and engineering fields standardize terminology because it's extremely helpful to avoid having to constantly define terms. Similarly, as a programmer, I know that at least in programming (not always in digital signals and other EE stuff, but that's all magic anyway) a byte is eight bits, unless I know we're talking about old systems that predate the the eight-bit byte becoming universal, so I don't have to continually look up the byte size of whatever I'm working with. The no true Scotsman fallacy is when a speaker continually redefines a term that already has an accepted definition, or equivocates between multiple definitions, to a definition that is friendlier to their argument. The proverbial argument from whence the fallacy gets is name is someone saying "No Scotsman would commit a crime", being confronted with evidence of a Scotsman doing just that, and asserting the criminal isn't a true Scotsman. This is fallacious because the accepted definition of "Scotsman" is "a man from Scotland", not "a man from Scotland who has never committed a crime"; the speaker is attempting to save their false statement by redefining the word. --71.119.131.184 (talk) 10:13, 11 January 2016 (UTC)

You are using "correctly" just as NTS uses "true". That is, no matter how much evidence was presented that others use the reverse convention, or none at all, you could argue that they aren't using it correctly, just like arguing that the Scotsman isn't a true Scotsman. Also, I don't believe the 8-bit byte is as universal as you think. Some counter-examples are at Byte#Common_uses. StuRat (talk) 04:40, 12 January 2016 (UTC)

The ancient Greeks personified their winds - see Anemoi, still with the same convention we use. Mikenorton (talk) 11:48, 11 January 2016 (UTC)

Also see Zephyr. ←Baseball Bugs ^{What's up, Doc?} carrots→ 12:37, 11 January 2016 (UTC)

That's probably where our convention comes from, which in turn depends on the wind Gods blowing to produce wind. If the myths had them inhaling instead, then we would have the reverse convention. StuRat (talk) 04:48, 12 January 2016 (UTC)

I believe the logic behind specifying the direction where the wind came from (versus where it's going to) relates to the fact that the weather that is brought to you comes from a place in that relative direction. So, if you live in the northern hemisphere, a north wind is more likely to bring cold weather, a south wind, warmer. When the wind direction is from the same direction as the ocean, you'll probably get more humidity. As a practical matter, back in history when these terms first came into being - they probably didn't care much which direction the airflow was moving so much as the effect it would most likely have on the temperature and precipitation. So that convention did make a degree of sense back then, and now we're stuck with it. The convention does seem a little counter-intuitive - and we do go to some lengths to disambiguate matters in other situations. For example, we don't hear "There is a fender-bender on highway 99 and south traffic is moving slow" - we use words like "southbound" to make it clear what we mean. SteveBaker (talk) 13:59, 11 January 2016 (UTC)

Simple logic can explain it. If you're in a fixed location, you don't care where the wind is going, you care where it's coming from. The opposite would be true of ocean currents: If you're using the current to move through the water, you don't care where it's coming from, you care where it's going. ←Baseball Bugs ^{What's up, Doc?} carrots→ 02:32, 12 January 2016 (UTC)

Nimur, it is not just wind measured at the airport showing east wind while the entire province has upper-level west winds. See this. Clicking a bar near the top shows locations all over the province (Dinganxian, Wenchang, Qionghai, Wanning). From Dongfang to Sanya, all show east winds. I still don't get how this is possible. Anna Frodesiak (talk) 23:08, 11 January 2016 (UTC)

In the tropics, climatological average surface winds usually are easterly near the surface (northeasterly in the Northern Hemisphere, southeasterly in the Southern Hemisphere) while upper-level winds are westerly. This has to do with a large-scale circulation cell that dominates the equator and tropical regions. If you want more details Hadley_cell#Mechanism is not bad, or just ask. Shock Brigade Harvester Boris (talk) 01:26, 12 January 2016 (UTC)

Wow, that is bizarre. I can't make heads or tails of the article, but I will take your word for it. Next time I'm out, I will watch the clouds to see them going over in the opposite direction to the wind. Oh, and I am pretty sure that when typhoons come, the winds on the ground and clouds moving over are in the same direction. And then there's the eye, so no wind, and then it all goes backward, and then an upside down cow goes by. Many, many thanks. You've solved one of the big mysteries on my list. :) Anna Frodesiak (talk) 09:08, 12 January 2016 (UTC)

Winds at the surface and at altitude are unlikely to be in opposite directions; but their directions could easily differ by 30 degrees. When the difference is in one sense it is called "backing" and in the other sense it is called "veering". See HERE. Dolphin (t) 03:29, 13 January 2016 (UTC)

• Can I simply assume that a northerly wind means the opposite from a wind from the north? Those two terms seem to be used in opposition, but it is not always clear what is meant by northerly on forecasts, while "out of the north" is clearly unambiguous. Evene a north wind (if we look at Boreas) should be a wind blowing to the south. But I don't think newscasters are always consistent on this. μηδείς (talk) 21:15, 12 January 2016 (UTC)

No. A northerly wind is a wind from the north. You can safely assume a northerly wind is the opposite of a wind blowing towards the north. Dolphin (t) 03:32, 13 January 2016 (UTC)

For example, the term prevailing westerlies. ←Baseball Bugs ^{What's up, Doc?} carrots→ 05:02, 13 January 2016 (UTC)

Wood's metal

Why does Wood's metal melt at a temperature so far below the melting point of all its component metals? If it's answered in one of the linked articles, I failed to find it; in particular, I wasn't able to understand Eutectic system or Fusible alloy very well. Nyttend (talk) 03:10, 11 January 2016 (UTC)

Wood's metal, for those playing along at home. The reason the melting temperature is low is an idea related to the Phase diagram#Binary phase diagrams, and the Melting point#Thermodynamics of a mixed substance. Essentially, the presence of "some substance" as an impurity in some other substance typically makes the melting point lower than if that other substance were pure(r). So an alloy of two metals is "impure" with respect to either one, and the melting point is thus expectedly lower than either one alone. Wood's metal is just one specific ratio of one specific set of chemicals that happens to have a well-defined behavior of this nature at a useful temperature. Being a "eutectic" just means it's a special ratio for the given set of chemicals that gives it additional useful melting-point properties (a narrow range of temperature for the solid/liquid transition, for example). The real questions of "why this ratio for these chemicals" and "why so much lower than the components" are much more complicated (and possibly not known in some cases beyond "because that's how they are" based on other variable that are they way they are "because that's how they are"...empirical rationalization/parameterization, etc.). DMacks (talk) 04:23, 11 January 2016 (UTC)

Nicely put. Another example is salt and water. perhaps https://en.wikipedia.org/wiki/Freezing-point_depression is applicable. Greglocock (talk) 05:19, 11 January 2016 (UTC)

The melting point of a material is determined by the forces between its molecules, and that involves a lot of complex interactions. One thing that's probably relevant is that cadmium is a group 12 element, right on the edge of the transition metals (some people even argue that they aren't true transition metals) and in the same group at mercury. These are notoriously volatile (see Metallic bonding#Strength of the bond for the details) compared to other metals, and adding cadmium to alloys often lowers their melting point because it disrupts metallic bonding. Smurrayinchester 10:58, 11 January 2016 (UTC)

See Eutectic system and freezing-point depression from the Wood's metal article. LongHairedFop (talk) 19:19, 12 January 2016 (UTC)

synthesis of tetrabutylammonium borohydride versus a PTC reaction of sodium borohydride + tetrabutylammonium bromide

I have a large amount of tetrabutylammonium bromide (although the iodide form is also available for about 130-140% the cost). While I'm waiting for another reagent to arrive, I've wondered about using and synthesising tetrabutylammonium borohydride (TBA-BH4) directly. One advantage would be that maybe I would need to use less water, or maybe not even perform a PTC reduction at all. A hypothetical procedure is as follows:

1. Dissolve TBAB in acetone
2. Dissolve sodium borohydride in acetone**
3. sodium bromide precipitates
4. Decant acetone from sodium bromide
5. Quickly evaporate (rotovap/vacuum) acetone before it has a chance to significantly react with the borohydride**
6. TBA-BH4 remains

** half-life of sodium borohydride in acetone is about 13 minutes, or 90% rxn in 40 minutes [4]  

If I'm able to evaporate the acetone quickly, I think I could get a decent yield. Of course, acetone is still moderately reactive, and I'm not sure if I want significant alkoxide counterion impurities. Is there another choice of solvent to use a Finkelstein-ish type metathesis (benzophenone?)

Partially this is due to my concerns about exposure of newly-formed imines to water even in a PTC rxn. I need the borohydride to react with the imine, but I also don't want my water to react with my imine. Is there a way to promote the lipohilicity of the borohydride ion ? Yanping Nora Soong (talk) 00:23, 11 January 2016 (UTC)

In water is used to get yields in 90% range for this sort of reaction. But the sodium bromide would dissolve in the water, so you may need to partition your TBABH4 into a more polar liquid phase. In your product dissolves nicely in dichloromethane, so perhaps this would be a better solvent. Graeme Bartlett (talk) 05:21, 11 January 2016 (UTC)

References

This chemistry is beyond my level really, but a quick search turns up chemistry forums like this one that discuss solvent considerations with this compound. While the emphasis of such forums is more recreational, the chemistry and practical limitations of independent therapeutic research are really quite similar.

I should say once again, I think you should get better answers here if you would provide details of the complete synthesis, what you're trying to make and how. Since you're not doing something illegal, why not take advantage of the ability to collaborate more openly? Wnt (talk) 15:10, 11 January 2016 (UTC)

How does this atom have such high valence?

Uranocene. How does the uranium atom have 16 different bonds sticking out of it? Also, is it wrong to call this uranium hexadecamethine or U(CH)₁₆ since the methine groups are bonded to each other? Sagittarian Milky Way (talk) 15:17, 11 January 2016 (UTC)

As I understand it (which is not very much to be honest), these are delta bonds, see this book. Mikenorton (talk) 15:27, 11 January 2016 (UTC)

Note that uranium is 5f3 6d1 7s2. Remember, s p d f hold 2, 6, 10, 14 electrons respectively, so that leaves room to add 11 f, 9 d, perhaps 6 p electrons. Of course, these will really interact as hybrid orbitals of some sort or other - what sort, I certainly can't guess. Assuming uranium contributes each of its electrons to a bond, that's six bonds, and then it can accept seven more dative bonds, plus three more if the p's are in the mix ... hmmm, that adds up to 16 bonds, coincidentally enough. But this would suppose that 26 electrons are donated into uranium orbitals, but I see only 16 electrons in the carbon pi bonds. No, maybe what goes on is that each double bond donates one "bond" to the uranium, two electrons, making it only eight bonds, a total of 16+6 = 22 electrons sitting in s2 f14 d6 maybe? Alright, now it's time to actually look it up. And [5] tells me there are six bonds from each 8-carbon ring. That means the p orbitals aren't filled ... I seem to be two electrons short even with the others. Hmmm.... Wnt (talk) 15:45, 11 January 2016 (UTC)

Wnt, you're correct that pi-bond pair is involved in bonding to the central metal atom, as usual for bonding "on the face" of pi systems such as the rings in sandwich compounds. In your "No, maybe..." analysis, you missed four electrons: the rings are formally cyclooctatetradienyl, giving ten pi electrons as a double-anion not just eight as a neutral cyclooctatetraene bond diagram contains. See Hapticity# Electrons donated by "π- ligands" vs. hapticity for some other ideas related to this aspect of counting. DMacks (talk) 17:24, 11 January 2016 (UTC)

Good point. cyclooctatetraenide has a 2- charge to make it aromatic, so there are ten electrons running around in each ring. In uranocene they aren't shown as charged, since they are covalently associated by dative bond to the uranium. But though they donate 20 electrons, they have to get their four electrons from the charge somewhere, i.e. the uranium which means it would be 20 added to 2 (in U4+) instead of 16 added to 6. I'm still counting five bonds there rather than six though, so no matter how I do the analysis in that sentence it still disagrees with the observation given in the next. Wnt (talk) 19:00, 11 January 2016 (UTC)

Ah true, didn't notice you also included U as neutral, rather than +4 based on the ionic synthesis of this compound. 2•8+6==2•10+2 indeed. DMacks (talk) 19:31, 11 January 2016 (UTC)

Well, you still should have been right that there would be five bonds' worth of pi electrons ... except that still doesn't gibe with the reported six. 8 atoms, 10 electrons, 20 "holes" to complete s+f+d, and ... six bonds per ligand. Hmmmmmmm. Now going back to basic chemistry, I know that 8 p orbitals are combined to get various degrees of overlap. psi1 = ++++++++, I think psi2 = ++++---- and psi3 = --++++--, nonbonding psi4 and psi5 = ++--++-- and +--++--+ maybe (as much bonding as antibonding contact, roughly). I suppose psi6 and psi7 might be ++-+--+- and +-++-+-- (rotated 90 degrees, four same in each), and psi8 should be +-+-+-+-. Now it's pretty clear that, despite what my source there says, the ligand is aromatic with the nonbonding orbitals full ... but if I try to push in two more electrons to fix the count, it shouldn't be aromatic anymore. So... Wnt (talk) 01:06, 12 January 2016 (UTC)

To go further in on this one: the source I cited before describes QTAIM analysis, which appears to be more theoretical than I'd supposed from context, and ELF analysis, which I take to be electron localization function analysis. The paper it cites isn't trivial to access [6] but appears to be just a crystal structure. The images ... well, the one with six bonds probably wants the 3D animation, but I've reached the point of not keeping Java on the computer, so I haven't examined it precisely. So I'm a bit out to sea here. Searching QTAIM and uranocene does turn up someone's doctoral thesis (same university as the last) which provides a lengthy introduction to some of the relevant issues. (For example, page 20 demonstrates one of the 5f orbitals en flagrante delicto with an 8-lobed 5f orbital...) [7] Wnt (talk) 13:04, 12 January 2016 (UTC)

This is really overthinking it. This is a sandwich compound. It has high hapticity. The classic case is ferrocene, which was discovered decades ago. Large metal ions are able to bond to several atoms at once, and many metallic species bind to delocalized bonds as a whole, not localised atoms. Heavy metals have many d and f orbitals, which nicely overlap with the 6 or 8 pi orbitals present in aromatic 6-8 membered rings (in this case cycloctratetraenide is aromatic, compared to its antiaromatic parent compound), just like in ferrocene, in which a much less heavy Fe(II) atom is bonded to over twelve atoms. Yanping Nora Soong (talk) 21:30, 13 January 2016 (UTC)