Template talk:Weather box/Archive 4

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 1

Archive 2

Archive 3

→

Needs doing

Subpages as per Infobox weather ideally by a page move and history merge. Rich Farmbrough, 17:06, 22 August 2010 (UTC).

Done any problems, let me know. Rich Farmbrough, 22:15, 26 August 2010 (UTC).

Wind Chill units

The wind Chill section is currently unit-less (°F, °C). It should probably be set up like one of the temperature lines. —MJCdetroit ^(yak) 02:38, 24 August 2010 (UTC)

Sorry I didn't see this earlier. There is a bit of a problem using units for the wind chill as not every country uses them. It would appear that US National Weather Service calls it a wind chill temperature and do include the unit. Because they include the unit it can be converted to C as well. On the other hand Environment Canada does not use a unit because it is an "...indices and not physical quantities.", which means that it can't be converted to F. I tried to see what they use in the UK but the Met Office seems to use a descriptor rather than a figure. The BBC calls it a temperature and uses a unit. They also say that "The JAG/TI algorithm is now used, which measures 'face only wind chill' and is a Canadian method. The Met Office use this method as it has been clinically tested, it is simple to use and based on current research." However, a link at the NWS site for the wind chill brochure says that JAG/TI (Joint Action Group for Temperature Indices) and the current method of calculating wind chill is a cooperative effort of the US and Canada. Enter CBW, waits for audience applause, not a sausage. 14:43, 12 December 2010 (UTC)

Request for Narrower Width

I'm not sure if it's possible, but could the "Weather box" be compressed to fit left of "Infobox settlement" on short city articles? Right now, the articles create a big white space gap so it can fit the Weather under the Infobox. Some ideas: narrow the MONTH column, optionally remove YEAR column. I'm not sure if this is possible. Sbmeirow (talk) 01:38, 20 October 2010 (UTC)

As the documentation explains "use {{Clear}} before the climate section, or <div style="width:50%;"> before the table, and </div> after." 117Avenue (talk) 04:29, 20 October 2010 (UTC)

The clear didn't work, which is what I was using, but the div works now. Sbmeirow (talk) 01:30, 21 October 2010 (UTC)

Personally I like {clear} because it works for all browsers and browser windows. For example, Great Bend, Kansas on my browser had no conflict, the infobox ends two paragraphs higher, but now there is an odd space on the table's right. 117Avenue (talk) 02:53, 21 October 2010 (UTC)

Sbmeirow, bad idea to remove year column. you want users to calculate the annual averages themselves? --HXL 's Roundtable, and Record 23:15, 20 November 2010 (UTC)

There is something called text, which could be added above or below the weather box. This is a moot point, because they already gave me a solution 1 month ago. Sbmeirow (talk) 01:57, 21 November 2010 (UTC)

This is still a problem for places with longish infoboxes, see forex Williams, Arizona, Lake City, Colorado, etc. etc. The weather box overlaps the infobox unless you widen the browser window, at least in Mac/Firefox. Pete Tillman (talk) 21:30, 24 February 2011 (UTC)
Yea, this is still an issue. The template should self limit the width to the size available. There must be some way to do this since other templates like {{Category diffuse}} do. Vegaswikian (talk) 19:25, 16 June 2011 (UTC)
It is possible, using "margin-left" and "margin-right" rather than "width", but then it will be more narrow when there is no right floating element. I have asked for CSS help at VPT. Until then, it does look like there is a "width =" parameter that you can use to override the default of 90%. Making it say "70%" could make it narrow enough allow it to float up in the gap, but then this seems like a hack, since the right floating elements are not based on percentages. This method is not robust if you resize the window. Hopefully some guru will have a suggestion. Frietjes (talk) 20:51, 16 June 2011 (UTC)

Maximum snow depth

I feel that this might be a useful contribution to the weather box. could somebody add it?— Preceding unsigned comment added by OettingerCroat (talk • contribs) 03:44, 12 December 2010 (UTC)

Is there a weather agency that uses it? 117Avenue (talk) 04:33, 12 December 2010 (UTC)

Canada uses it, here, last line in the precip section. I'm wondering if it might not be a good idea to seek wider input as to what people would like to see in the box. Especially as it's getting bigger. Enter CBW, waits for audience applause, not a sausage. 14:05, 12 December 2010 (UTC)

Sunshine hours

After lengthy analysis of many of sources and questions of over a dozen peoples outside users of Wikipedia I consider: the need to change (or add a new) the option "Sunshine hours". This is sample current data in the Weather box:


	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
Sunshine hours	50	74	90	123	178	210	246	298	238	178	112	72

Many people do not know what it is. 50? 123? What it is? People interested in meteorology know what it is, normal people do not know. I propose a change from "all hours of the month" to "daily average per month":


	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
Sunshine hours at day	5	7	8	9	11	12	12	10	9	7	6	5

Many sources give such information (average sunshine hours at day) and many people understand this information. Example: Malta has average 5 hours of sunshine at day in January^, Los Angeles - 7^, Lisbon - 4.5^, Cayenne- 4.6^ etc. It is easy to understand than any number of "48" and other similar. Subtropical-man (talk) 22:38, 16 December 2010 (UTC)

How do you factor cloud cover into average hours per day? See for example Seattle's sunshine hours per month in December where they get 52.7 hours a month...but that's mostly because of cloud cover, not astronomical sunshine hours (in this case I mean average time between sunrise and sunset). If you put it into the format of hours per day, which for Seattle would be ~1.7 hours a day in December, I'm pretty sure some users will think it refers to astronomical sunshine hours per day, get confused, and think the statistic is wrong. If we leave it as average hours per month, there's less confusion on that front. Perhaps we could link the words "sunshine hours" to something explaining what they are? Ks0stm ^(T•C•G) 22:59, 16 December 2010 (UTC)

But see for example Warsaw (I know the climate of Warsaw on my own skin), with 1.3 average sunshine hours per day in January (43 per month) and I agree with this data. Because this is average data. In Warsaw often there are so few sunshine in winter, sometimes 1, 2 and from time to time 5 or 0 sunshine hours per day. Generally, this is an average of 1.3. Similarly, with temperatures: average temperature in January is 0°C during the day, but this is average data. Really data is from -12 to +8°C. Therefore, average data is used. People understand word "average".


	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
Avg. sunshine hours at day	5	7	8	9	11	12	12	10	9	7	6	5

If there are doubts, time to create the article Sunshine hours or "we could link the words "sunshine hours" to something explaining what they are". In addition to the "astronomical sunshine hours" - this names is not exist. There are: "Hours of Sunshine"/"sunshine hours" (for sun duration) and "Hours of Daylight"/"Daylight's hours" (for time between sunrise and sunset) - sample source with "Hours of Sunshine" and "Hours of Daylight" data. These are two separate names and things. PS. Article Daylight already exists. Subtropical-man (talk) 23:45, 16 December 2010 (UTC)

Three points. First it does not matter which is used. If it's not explained then there will always be people that don't understand exactly what is being said. Most people are not going to understand from what is given, using either daily or monthly, that the number refers (usually) to bright sunshine only and not diffuse sunshine. Second, what does the source give. If the source gives hours per month or per day then it can't always be converted to the other. Third, does the World Meteorological Organization have a standard. If so that should be the one used. Enter CBW, waits for audience applause, not a sausage. 10:15, 18 December 2010 (UTC)

Ad. Third: The World Meteorological Organization not have a standard of sunshine hours neither even the number of rainy days (sometimes as 0.1mm, sometimes 1mm). Ad. Second: "If the source gives hours per month or per day then it can't always be converted to the other" - exactly. Many sources gives data as sunshine hours at day (daily average per month), not to change this but used apply the data provided. At the moment we must to convert because there is no option "Avg. sunshine hours at day" in the Weather box. Subtropical-man (talk) 11:51, 18 December 2010 (UTC)

Suggestion

2 problems with current temp setup:

1) blue to orange transition is at a bad spot. 4.5C is arbitrary and 0C makes much more sence.

2) Some places large groups of temps look the exact same. Suggestion: New color scheme for temperature that transitions between these colors (filling in the gaps between with colors too, obviously)

-90	-85	-80	-75	-70	-65	-60	-55	-50	-45	-40	-35	-30	-25	-20	-15	-10	-5	0	5	10	15	20	25	30	35	40	45	50	55	60

The Unforgiven One (talk) 01:47, 26 March 2011 (UTC)

2 new rows

Suggest adding rows for "lowest high" and "highest low". You can see an example of what I mean at [[1]] in the "warmest maximum" and "coldest minimum" columns. The Unforgiven One (talk) 02:04, 26 March 2011 (UTC)

Humidity

I suggest adding a slight purple tinge to higher values that will give the impression of a very wet sky.

0	10	20	30	40	50	60	70	80	90	100

The Unforgiven One | ¿Como estás? 02:49, 26 March 2011 (UTC)

Conversion of weather boxes from other languages to English

Is there any tool which would convert weather boxes from other languages into English? Especially on Russian wikipedia there are so many weatherboxes which could be on en wiki too... The program or macro is the simplest thing a programmer can imagine. I would add some weather boxes from there myself, but it's too time-consuming.--Me ne frego (talk) 06:33, 31 March 2011 (UTC)

Number sign

Hi. We received a comment from User:Lightmouse during a GA review that number signs ('#') are more a local convention and are not suitable for an international resource like Wikipedia. Could you possibly change the source numbering to 'Source No. 1' and so on per MOS:NUMBERSIGN? Thank you. -SusanLesch (talk) 16:27, 3 May 2011 (UTC)

I updated it. Frietjes (talk) 18:06, 3 May 2011 (UTC)

Fabulous. Great response time, too. -SusanLesch (talk) 18:13, 3 May 2011 (UTC)

Specifying periods from which climate averages are derived

In the course of checking some suspect edits to weatherbox data (including a subtle form of vandalism and outright falsification), I've come across a problem in trying to find the information from the original sources cited

An example is the weather information for Mumbai (Bombay), India. The World Meteorological Organization has a limited amount of data categories, based on the 30-year period 1961-1990. The Indian Meteorological Department has a much larger number of data categories, but for the slightly earlier 30-year period 1951-1980. Comparing these two sets of figures, the differences for the monthly minimum and maximum average temperatures can be as much as 0.2-0.3°C.

It seems to me that the weatherbox template data field for the primary citation (Mandatory fields, source) should have a new value added, to specify the period from which the data have been derived. This would help to reduce confusion and improve confidence in WP reliability, as well as helping later editors determine whether the weatherbox needs to be updated as more recent climate data is published. Cheers, Bahudhara (talk) 04:18, 7 June 2011 (UTC)

Well this addition should only be used when all average fields below are from the same time period. That is often not the case for many articles' tables. —HXL's Roundtable and Record 04:14, 8 June 2011 (UTC)

I think that where the information is sourced from a countries met agency that follows the WMO standards then all the averages follow a 30 year period. You can already include that as part of the reference but it is probably not clear to the average reader that there is a 30 year period and that it only applies to certain parts of the box. What you really need is two sources. One for the actual data and the second that explains the weather data. Using Cambridge Bay#Climate as an example; reference 84 is the link to the Canadian Climate Normals 1971-2000. Then reference 85 links to Calculation Information for 1971 to 2000 Canadian Normals Data which shows the periods that the data came from. Of course not every met organisation is going to provide what Environment Canada calls the "metadata". CambridgeBayWeather (talk) 15:01, 8 June 2011 (UTC)

Removal of green colours for precipitation and rain

User:Someone35, recently changed the green display of precipitation/rain in the climate tables on many articles with no explanation other than "no reason to use green". Since this potentially involves reversion on a mass scale, I will allow him a chance to explain himself here. To begin, those who participated in the overhaul of this template more than a year ago would not have added the parameter "|rain colour = " if they felt that green should not be used or if there was no need to use that colour. Secondly, the default blue colouring could cause a 'blending' of colours if snow, humidity, precip/rain/snow days, and/or record temps are filled in. Conduct some test edits yourself to see how this is the case. Thirdly, as I see it, blue, especially for non-arid places, could falsely leave an impression of a chilly climate for places that aren't cold in winter (averages below −3 °C or 0 °C) anyway. Conversely, in places that are cold and very wet in winter, green would not be so appropriate. —HXL's Roundtable and Record 04:11, 8 June 2011 (UTC)

The green colour is redundant. I support the removal of green colour from Weather box. This colour makes much confusions. Subtropical-man (talk) 12:54, 8 June 2011 (UTC)

What confusions? —HXL's Roundtable and Record 13:04, 8 June 2011 (UTC)

I like the green, given that it makes precipitation/rainfall stand out from snowfall, and would actually support making it to where green is always used for rain/precipitation and rain/precipitation days, with blue being left for snowfall and snowfall days (basically green for water-related precipitation amounts and days and blue for snow/ice). Ks0stm ^(T•C•G) 20:59, 13 June 2011 (UTC)

Display of F below C in yearly column partly missing

The header says it all, really. I don't want to break this, not even sure if would understand it if I clicked "edit", but a couple of temperature rows are missing the F figure below the C figure. Can we fix this please? Huw Powell (talk) 03:30, 14 July 2011 (UTC)

Which ones exactly? I'm seeing five rows all with C and F figures. CambridgeBayWeather (talk) 05:18, 14 July 2011 (UTC)

I don't see a conversion in to Foreignheight for the yearly record high in the sample from the doc page. I don't know what was wrong exactly but the version I've got in the sandbox is working okay. JIMp_talk·cont 15:14, 17 October 2011 (UTC)

It was running through too many templates and equations, it seems you have fixed that. 117Avenue (talk) 00:24, 18 October 2011 (UTC)

Yeah, somehow ... (not quite sure what was wrong with the old code). The yearly record low temperature is working too. JIMp_talk·cont 14:34, 18 October 2011 (UTC)

The conversion for the yearly record low was missing too. I've got it working in the sandbox. JIMp_talk·cont 17:02, 20 October 2011 (UTC)

Minus signs again

As mentioned a couple of times above the template really should display true minus signs rather than hyphens. The response given was that it's not possible because of {{#expr:}}. This is not true. It is possible. {{Convert}} can do it: {{convert|-40|C}} gives "−40 °C (−40 °F)" (that's a hyphen in the input & two true minus signs in the output). JIMp_talk·cont 05:22, 17 October 2011 (UTC)

Sandbox

I've been working on a new version. It's currently in the sandbox. I've fixed the two problems mentioned above (missing conversions & minus signs). I've also simplified the code. I'm about to paste it into the main page. JIMp_talk·cont 23:40, 20 October 2011 (UTC)

Just wondering, since the precipitation and rainfall rows can be green could you also make it to where average precipitation days and average rainfall days have a green scale (as is it looks a bit odd with green precipitation and blue precipitation days, see Salina, Kansas#Climate. Ks0stm ^{(T•C•G•E)} 23:48, 20 October 2011 (UTC)

Rainfall conversion

I just noticed, see Rockhampton#Climate, that in some cases the rainfall when given in mm is being converted to 3 decimal places. It should only be to two. CambridgeBayWeather (talk) 16:23, 3 November 2011 (UTC)

Issues with recent overhaul

1) For subfreezing temperature values, inputs such as "−2.0" should not have precision reduced, i.e. displayed as "−2". 2) Any input of annual average temperatures prevents conversion to the other temperature scale. The Tartanator 15:26, 5 November 2011 (UTC)

2) For the "Average high" and "Average low" Year, the conversion is not working. In Detroit and Toronto, the value to be converted (Celsius for Detroit and Fahrenheit for Toronto) is not being displayed. —MJCdetroit ^{(tell me)} 23:25, 5 November 2011 (UTC)

It turns out you have to remove the yearly number, like this. However, when the template then does the calculation to two decimal places instead of one. In the dropping of the trailing zero that The Tartanator notes with record high temperatures the zero is left in place (May and November in the above link). For all other temperatures the trailing zero is dropped but the conversion is always to two places, even if the conversion ends in a zero. CambridgeBayWeather (talk) 21:23, 6 November 2011 (UTC)

The precision needs a fix in general. While most entries are set to one decimal, the rainfall in inches is converted to three decimals from entries in mm. De728631 (talk) 00:11, 5 December 2011 (UTC)

See section above this. The US standard seems to be to two decimal places. CambridgeBayWeather (talk) 12:48, 5 December 2011 (UTC)

It would be best to use the standard that is provided by the source data. If the source rounds to one decimal then we shouldn't display two decimals just because it is another standard. Filling up the source data with trailing zeros would be scientifically incorrect since you don't know that rounding (floor/ceiling) was behind the source data. And not only does it look weird to have a mixed table with accuracies ranging from 0 to 3 decimals but imo for weather data a higher accuracy than .1 is also not needed for encyclopedic use. De728631 (talk) 23:11, 8 December 2011 (UTC)

Option to not collapse in dependent boxes?

Many cities use this template to create their own template for that city. In an article on the city, it could make sense to have the box collapsed -- but in an article on the city's climate it seems it could very often make more sense to start with the box shown. I am not aware of any way to pass a parameter (in, say, Vancouver Weatherbox) that would control whether the box was shown or hidden on a per article basis.--JimWae (talk) 00:14, 10 December 2011 (UTC)

Add the following line to the template:

|collapsed={{#if:{{{collapsed|}}}|{{{collapsed}}}}}

This can also be done to the open parameter. 117Avenue (talk) 04:28, 10 December 2011 (UTC)

Thanks. Suppose we still wanted collapsed to be the default--JimWae (talk) 19:03, 10 December 2011 (UTC)

Then you'd have to create an opt out word:

|collapsed={{#ifeq:{{{collapsed|}}}|no||Y}}

Or create an uncollapsed parameter:

|collapsed={{#if:{{{uncollapsed|}}}||Y}}

Does this help? 117Avenue (talk) 22:43, 10 December 2011 (UTC)

Abbreviations

Abbreviation templates are necessary for accessibility. See this discussion.

Which “errors” about year ? If there are any, they exist both after my last change and before my last change. And speaking of errors, Avg and Dec are not English words. If we keep the abbreviations, they need to be correctly marked up, for accessibility's sake.

--Nnemo (talk) 04:01, 14 December 2011 (UTC)

I've reverted your edits as they introduce errors to the template. In the Year column, Humidex, Record high, Record low and Wind chill all show "Template:Max/27". You need to experiment with the template in a sandbox before adding it to here. CambridgeBayWeather (talk) 16:51, 14 December 2011 (UTC)

My edits did not introduce “errors”. Check. I checked.

--Nnemo (talk) 03:37, 16 December 2011 (UTC)

What did you mean that Dec and Avg are not correct? They are not words, but are correct or commonly used abbreviations. Wiktionary uses Dec and Dec. as does this which also uses avg. This site shows AVG and so does Environment Canada. CambridgeBayWeather (talk) 17:17, 14 December 2011 (UTC)

Avg and Dec might be correct abbreviations, they are not correct English words. They are not accessible. So I mark them as abbreviations, with their meaning. This makes them accessible, this is necessary so that a screen reader read them correctly, this also helps the readers who are not familiar with English abbreviations. For most of the people of the planet, English is a foreign language, or not a language at all. The links you give me show some of the abbreviations, with their meaning.

--Nnemo (talk) 03:43, 16 December 2011 (UTC)

Why only some of the abbreviations? Why not the rest of them? My revert of the page had nothing to do with the abbreviations but the bright red "Template:Max/27" error that was showing up. CambridgeBayWeather (talk) 14:56, 16 December 2011 (UTC)

View, Discuss, Edit Links

I propose to include "v" "d" and "e" at the top of the template so that users have an easy way to link directly to a city-specific weatherbox. Thanks, epicAdam^(talk) 14:56, 18 December 2011 (UTC)

Usually those links lead to the template in question. How will they lead to a city-specific weatherbox? CambridgeBayWeather (talk) 08:12, 19 December 2011 (UTC)

I'm not exactly sure. I was hoping those with more experience in this area than I would have some idea. The city-specific weather boxes for example, Template:New York City weatherbox or Template:Washington, DC weatherbox, use this standard template, so I don't know if including those direct links can be made directly on those pages. Best, epicAdam^(talk) 16:12, 19 December 2011 (UTC)

I added code to the template to allow this to work, like it works with template:navbox. hopefully, this is not controversial and doesn't cause additional problems. Frietjes (talk) 18:17, 19 December 2011 (UTC)

Excellent! Thank you. Best, epicAdam^(talk) 18:29, 19 December 2011 (UTC)

OK now I get it. Looks good. CambridgeBayWeather (talk) 23:15, 19 December 2011 (UTC)

Colour scheme

Is there any recommendation about the usage of temperature colour= "pastel" and rain colour= "green"? Why are there several colour schemes in the first place? Isn't it better if the colours are consistent across articles? Calimo (talk) 09:18, 12 February 2012 (UTC)

Percent sunshine

Would it be possible to perhaps give this the same color scale as sunshine hours? Right now the whole range makes me think cloudy/overcast. Ks0stm ^{(T•C•G•E)} 20:24, 14 March 2012 (UTC)

Sure, I realized the same thing. Maybe black to gray to yellow. If you can do it quickly, please feel free too. I've gotten a lot of "server busy" messages in the last few minutes, so it's hard for me to make the edits. Thanks. Ufwuct (talk) 20:28, 14 March 2012 (UTC)

So long that, when both fields are input, %sunshine does not supplant Sun hrs (for an obvious reason), there should be no problem. GotR ^Talk 21:25, 14 March 2012 (UTC)

IW problems

Re: Cricieth article: I have copied all templates to Wiki-cy (Welsh Wikipedia); took me well over an hour. Still doesn't work. Any ideas, please? I'm not sure why it needs to be done manually! Can we not just transfer all templates into every language servers? If you change ONE of these templates it takes hours for us to find and recopy iw. There's only a handful of us in Wiki-cy, and my time would have been better spent writing much needed articles rather copying and pasting hundreds of templates. Please simplify and help. Llywelyn2000 (talk) 08:06, 21 March 2012 (UTC)

Fixed you were missing two templates and after creating them from here two more were listed as missing. One of those, {{Rnd/c4dec1}}, doesn't exist here either but I finally realised it had been moved and was able to create it. CambridgeBayWeather (talk) 11:09, 21 March 2012 (UTC)

I wondered why the example on the documentation page wasn't working. It needed a bunch more templates, most of which have nothing to indicate that this template needs them. Another problem was this. Translating only the documentation page will cause problems if you haven't done the Weather box template. You need to do both at the same time. I can do some of the template based on the changes you did to the documentation but it would probably be a couple of days and anything you didn't translate would still be in English. CambridgeBayWeather (talk) 17:03, 21 March 2012 (UTC)

Just remembered you should probably create Categori:Weather box for convenience. CambridgeBayWeather (talk) 17:05, 21 March 2012 (UTC)

Many thanks CBW! It's clear as ginn! The dificutly lies in differentiating between what to change / translate and what not to translate: usually it's quite straigntforward, and the difference between template names and what's on the screen is obvious. Here it's very different. If I list the translations here, could you input them please? Many thanks, the storm seems to be coming to an end, due to a glimps of sun shining in the South-East! Llywelyn2000 (talk) 07:35, 22 March 2012 (UTC)

All that needs to be translated here, really, is the names of months, which you have and:

month = mis
Climate data for ... = Hinsawdd ...
Average high = Tymheredd uchel, cyfartalog
Average low = Tymheredd isel, cyfartalog

Thanks.... diolch! Llywelyn2000 (talk) 07:41, 22 March 2012 (UTC)

A few more CBW, and thanks again:

Hinsawdd the Local Airport = Hinsawdd y Maes Awyr Agosaf
Average uchaf = (Tymheredd uchaf (cyfartalog)
Daily mean = Cymedr dyddiol
Average isaf = Tymheredd uchaf (cyfartalog)
Precipitation mm (inches) = dyddodiad mm (modfeddi)
Rainfall mm (inches) = glaw mm (modfeddi)
Snowfall cm (inches) = eira cm (modfeddi)
(lleithder) = no brackets needed
Avg. precipitation days (≥ 0.2 mm) = dyddodiad cyfartalog dyddiol (≥ 0.2 mm)
Avg. (glaw)y days (≥ 0.2 mm) = glaw cyfartalog dyddiol (≥ 0.2 mm)
Avg. (eira)y days (≥ 0.2 cm) = eira cyfartalog dyddiol (≥ 0.2 mm)
Sunshine hours = Sawl awr o heulwen
Percent possible sunshine = Y canran posibl o haul
Souces (in the refs) = Ffynhonnell

The last colmn heading should also be changed from "Ylwyddyn" to "Blwyddyn". CBW: we are in your debt! Many thanks again. Llywelyn2000 (talk) 02:38, 24 March 2012 (UTC)

Last little problem on the Cricieth page. Where it should read:

Average high = Tymheredd uchaf cyfartalog
Average low = Tymheredd isaf cyfartalog

However, both at present read Average High. Thanks. Llywelyn2000 (talk) 03:08, 28 March 2012 (UTC)

Sunshine hours per day

I think that, instead of giving the total number of sunshine hours in a certain month, it would be better to give the number of sunshine hours per day in that month: it would be much more immediate to understand, and would be independent of the month length (think to the 28 days of february). --87.11.208.178 (talk) 19:27, 31 March 2012 (UTC)

Agree . This problem I previously reported, was forgotten. I am glad that the case was reopened. Subtropical-man (talk) 19:43, 31 March 2012 (UTC) My previous post:

Previously reported (from 16 December 2010):
After lengthy analysis of many of sources and questions of over a dozen peoples outside users of Wikipedia I consider: the need to change (or add a new) the option "Sunshine hours". This is sample current data in the Weather box:


	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
Sunshine hours	50	74	90	123	178	210	246	298	238	178	112	72

Many people do not know what it is. 50? 123? What it is? People interested in meteorology know what it is, normal people do not know. I propose a change from "all hours of the month" to "daily average per month":


	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
Avg. sunshine hours at day	5	7	8	9	11	12	12	10	9	7	6	5

Many sources give such information (average sunshine hours at day) and many people understand this information. Example: Malta has average 5 hours of sunshine at day in January and 12 in July (source: weather2travel.com), Lisbon - 4.6 in January and 11.4 in July (source Hong Kong Observatory), Cayenne - 4.6 in January and 8.7 in September (source: climatetemp.info), Rome - 3.6 in December and 10.7 in July (source: Servizio Meteorologico), Sydney - 5.5 in June and 7.8 in November (source: Bureau of Meteorology) etc. It is easy to understand than any number of "46" or "246" and similar other. Subtropical-man (talk) 22:38, 16 December 2010 (UTC)

You would need to have both available as not all sources give daily sunshine. At the same time you don't want editors dividing the monthly sun by the number of days in the month as that could provide an incorrect answer. CambridgeBayWeather (talk) 00:20, 1 April 2012 (UTC)

Perhaps the editors could be asked to continue to put total hours, and the template could automatically divide them by the number of days of each month, giving the result with a fixed (one or two) decimal digits. In this way, the existing tables would be automatically updated. Instead, you cannot continue to ask the reader to divide mentally 246 by 31 or by 28 if he wants to understand whether this means a lot of sunshine or not. --87.6.230.98 (talk) 07:30, 1 April 2012 (UTC)

No you can't do that either. There is no way to tell if it would be anywhere close to accurate. CambridgeBayWeather (talk) 18:49, 1 April 2012 (UTC)

I do not understand your point. In which sense the procedure I propose (that is, the template software dividing the sunshine hours in a month by the days in the same month to get the average number of sunshine hours per day) would not be accurate? --87.17.219.6 (talk) 22:24, 1 April 2012 (UTC)

Sorry that was a poor comment and I hope I can explain it better. Sunrise and sunset are not static they change from day to day and the further away you get from the equator the more pronounced this becomes. Take a look at the following. Other than the first one just north of the equator they are all real places. I used the advanced options and Choose Location by Longitude and Latitude to get the sunrise/sunset. The total hours of sun were obtained from the Wikipedia and for the month of March.


Location	Possible sun 1 March	Possible sun 31 March	Difference	Total sun	Mean sun
Just north of the equator	12:06	12:08	2 mins
Honolulu	11:45	12:21	36 mins	260.4	8.4
Casablanca	11:29	12:31	1 hr 2 min	241.8	7.8
Toronto	11:14	12:44	1 hr 30 min	178.0	5.7
London	10:56	12:56	2 hrs	103.2	3.3
Cambridge Bay	09:45	13:55	4 hrs 10 min	175.8	5.7

So the further away from the equator the more meaningless the average daily sun becomes. Which is why you need both. If the source gives daily sun then the table should accommodate but not convert to a monthly figure. Then if the source gives the monthly figure the table should accommodate but not convert to a daily figure.

First: sunrise/sunset and sunshine duration is two different things. There is no sense does some calculations between sunrise/sunset and sunshine duration because for sunshine duration there are many factors, for example cloudy weather, fog, drizzle, rain, snow etc. Second: numbers of 175.8, 111, 241.2 etc do not say anything for those peoples, who not familiar with the meteorology. However, numbers of 5 or 11 avg. sunshine hours speak for themselves. For example: Malta has average 155 hours of sunshine duration in January and 372 in July. Ok. What are these numbers? You have to somehow share? These are the questions from normal people. Look at this: Malta has average 5 hours of sunshine duration at day in January and 12 in July, that is to say: in Malta in January only a few hours of sun per day and in July even 12. Layman understands everything. Subtropical-man (talk) 19:48, 2 April 2012 (UTC)

I've updated the template to allow both daily or monthly depending on the source, here. As an example see Template:Sydney weatherbox which now uses it, see here. Strangely enough the 200.6 that was given for the month of January is not the 7.1 * 31 as given in the source. The possibility of understanding either the daily or monthly figures has nothing to do with be familiar (or not) with meteorology but more to do with what the person has been provided with. I would guess that in Australia they are more used to daily sun and in Canada the monthly. I do realise the difference between possible sun and actual sun. What I was trying to say was that if at the start of a month the possible sun is 9:45 and at the end it is 13:55 then 5.7 is not representative. CambridgeBayWeather (talk) 23:34, 2 April 2012 (UTC)

You still need to change the colors, now data of Mean daily sunshine hours is black. Subtropical-man (talk) 12:59, 3 April 2012 (UTC)

Ah that's beyond me. I couldn't figure it out. Do you have a suggestion as to what the change over number should be? I'll ask someone for assistance. CambridgeBayWeather (talk) 15:08, 3 April 2012 (UTC)

The option for a multiplication factor is already built in. However, the year column is usually a total, so that still doesn't work. Unfortunately I don't have time anymore to dive into templates. 117Avenue (talk) 07:06, 4 April 2012 (UTC)

Meaning of certain fields is not clear

Average low: is this calculated by averaging overall monthly lows over a number of years, or is it calculated by averaging daily lows through that month?

Average high: same question.

"Rainfall" and "Snowfall" should presumably read "Average rainfall" and "Average snowfall".

It is unclear to me exactly how one would measure a (per-month) "daily mean", and how that would differ from a (per-month) "monthly mean".

It may be that these all have standard definitions, but currently there seems to be no easy way to find this information. I suggest that a page is constructed somewhere with precise scientific definitions of what all the fields mean (e.g., in addition to those I've mentioned, precisely how is "sunshine" defined?). Then a small link to that page can be put on the template. 86.179.112.16 (talk) 11:17, 24 April 2012 (UTC)

Apart from the third bullet, I don't think this is a reasonable request at all since the template doesn't have any space to accommodate such definitions, and there is little chance there will be articles with such definitions. For average highs and lows, you arrive at the same figure through either method of calculation. Daily means are calculated differently from one meteorological organisation to another: NOAA and Environment Canada simply take the arithmetic mean of the high and low, while China Meteorological Administration, for example, uses the Riemann sum method, probably by computing the average of hourly temperatures. GotR ^Talk 15:28, 24 April 2012 (UTC)

I am not asking for these explanations to be put on the template, I am suggesting that a separate page is created, with a small link, as I explained. 86.160.217.111 (talk) 17:07, 24 April 2012 (UTC)

Wind speed

Could we have rows for (a) Max Wind speed (b) Average Wind speed, with Beaufort scale coloration. --Iantresman (talk) 16:38, 31 May 2012 (UTC)

Average low and high

Hi, further to my comment above, once again I have been looking at some weather data, and once again I find myself confused about what "average low" and "average high" mean. For example, does an average high of 20°C in June mean that in an average year the highest June temperature (over all June days) is 20°C? Or does it mean that on an average June day (in any year) the highest temperature is 20°C? The answer given above that "you arrive at the same figure through either method of calculation" is incorrect. Does anyone know which convention is actually used? 86.179.3.247 (talk) 17:21, 24 July 2012 (UTC)

In terms of the arithmetic, so long as rounding doesn't occur, I don't see how the two different methods could arrive at a different figure. The first answer is correct, as few locations in the world are monotonous in June, and "average" is equivalent to the (usually) 30-yr mean. GotR ^Talk 19:07, 24 July 2012 (UTC)

Try doing the calculations with some sample data, and you will see that the two methods generally give quite different results. 86.179.3.247 (talk) 19:27, 24 July 2012 (UTC)

That clearly is original research. Source? GotR ^Talk 19:37, 24 July 2012 (UTC)

What a ridiculous response. Can anyone actually answer the question? 86.179.3.247 (talk) 19:43, 24 July 2012 (UTC)

I did. Rather, you would do this page a favour if you would quit peppering it with the most naive questions; otherwise, I will have to fetch someone to lock it down. GotR ^Talk 19:47, 24 July 2012 (UTC)

Not only do you not know the answer, you do not even understand the question. 86.179.3.247 (talk) 19:50, 24 July 2012 (UTC)

I believe it is the average of the daily highs. Does the meteorology society in your country provide definitions of the terms it uses in its record keeping? 117Avenue (talk) 00:54, 25 July 2012 (UTC)

117Avenue, please do not do that again. None of these sections pertain directly to the template. Of course, it was my mistake in exacerbating the problem. GotR ^Talk 01:08, 25 July 2012 (UTC)

These sections are about how the template is used, and if there is a problem, we need to fix. If you aren't going to read a commenters issues, and understand them, please don't respond. 117Avenue (talk) 01:14, 25 July 2012 (UTC)

Then this is a matter of interpretation. (S)he should know that, in general, we report data in Weatherboxes exactly as our sources put it, regardless of calculation methods. I am disappointed that you seem to buy into the IP's deceitful response: "...you do not even understand..." GotR ^Talk 01:24, 25 July 2012 (UTC)

Please remember that Wikipedia is international. Other regions may not use the same terminology, nor methods of metrology recording. If a template is intended to be used across Wikipedia, we should ensure there is no confusion with the terminology used. This means addressing the issue, not attacking users. 117Avenue (talk) 05:39, 27 July 2012 (UTC)

GotR, I just happened to arrive here by chance and have no vested interest in the discussion, but I have to admit that you clearly did misunderstand the OP's question and were quite rude and dismissive of them. Ðiliff «» (Talk) 11:06, 30 July 2012 (UTC)

Harder to find than I expected. It turns out it is one of the few things Environment Canada does not define but the Atlas of Canada does. The Australian Bureau of Meteorology as has a definition. Other countries may differ but as both use the standard WMO 30 year period it may be the same in other countries. CambridgeBayWeather (talk) 07:49, 27 July 2012 (UTC)

Percent possible sunshine vs Mean Monthly Sunshine hours

Recently, a large amount of weatherboxes has been changed from mean monthly sunshine hour values to percent values. There should be discussion on this before changing it a percentage value and which one is better, the percentage one or the mean monthly one. Ssbbplayer (talk) 05:38, 16 November 2012 (UTC)

Can you give an example of one? CambridgeBayWeather (talk) 12:56, 16 November 2012 (UTC)

See Template:Washington, D.C. weatherbox I also think this needs to be discussed. I'm not sure the percentages are as useful. Thanks, epicAdam^(talk) 13:57, 16 November 2012 (UTC)

Yesterday I asked the user to discuss here, before changes on a large scale. Ssbbplayer is right, there should be discussion on this before changing it a percentage value and which one is better. Subtropical-man (talk) 14:45, 16 November 2012 (UTC)

I would think that a better central place for this discussion would be at Wikipedia talk:WikiProject Meteorology. As before I would say go with whatever is the most common for the particular country. CambridgeBayWeather (talk) 14:48, 16 November 2012 (UTC)

Maybe you can talk about it in the article such as describing how it has sunny winters or summers, despite receiving 200 hours of sunshine for example. I prefer using monthly sunshine hours to determine how much sunshine it gets in absolute terms. The colours are a little bit off (especially the yearly value), and the template doesn't calculate it properly . Although if the source only gives a percentage, or if the reliable source uses percentage values and an unreliable source use monthly values, go with the more reliable one. Ssbbplayer (talk) 21:37, 16 November 2012 (UTC)

However, both of them have disadvantages and advantages in describing the climate. Ssbbplayer (talk) 21:38, 16 November 2012 (UTC)

The yearly colour does need fixing. I wasn't able to figure it out though. CambridgeBayWeather (talk) 10:01, 17 November 2012 (UTC)

Fixed, the daily hours and percent change were being treated like monthly hours. 117Avenue (talk) 07:07, 18 November 2012 (UTC)

Thanks. CambridgeBayWeather (talk) 16:20, 18 November 2012 (UTC)

width issues

The template sets box width to 90%, I think it would be better defaulted to auto, on pages with things like infoboxes the current default width can push the weatherbox to below the box leaving a big blank space, e.g. on Miyako, Iwate most of the article seemed to be blank if the browser window was less than 1368 pixels wide due to the weatherbox being pushed offscreen below the long inforbox. I was able to fix this by setting width=auto on that article's weatherbox template but the same issue probably crops up on other articles so I would be inclined to change the default. The current default introduces a lot of empty space within the weatherbox which may look pretty but not at the expense of screwing up article formatting. Maybe there is some way of getting it to use available (i.e. remaining) space without demanding more than is available? Samatarou (talk) 00:06, 11 December 2012 (UTC)

I support changing the default to auto. I just (unscientifically) randomly sampled five U.S. places in the state of Washington that transclude this template: 3 of the 5 have the described problem (they are Cathlamet, White Salmon, and Naches). —Mrwojo (talk) 01:06, 17 December 2012 (UTC)

Help! Help! The colors are gone

All of a sudden the {{Weather box}} climate boxes arent displaying properly for me on any page. Instead of a rainbow there are only three colors: red, white, and black, and mostly red. This happens only when I'm logged in, for some reason, and it happens on every browser, and every account. Since this template itself has not been edited recently I figure it must be a transclusion of some other template, though there don't seem to have been any recent changes to the templates transcluded onto this template either. Could someone please help figure out what the problem is and if possible fix it? I assume it may have been done on purpose to improve some other thing. {{Weather box/concise C}} still seems to work. ☮Soap☮ 00:14, 18 December 2012 (UTC)

It happens to me when logged in (Firefox) and logged out (Chromium). Note that the colours are fine until the page is purged, at which point they go away for everybody. (The colours on Berkeley, California, for instance, were fine until I purged the page, whereupon they disappeared.) — Wolfgang42 (talk) 01:39, 18 December 2012 (UTC)

Same issue here for Lewisville, Texas, but unfortunately I don't know how to fix it. Doesn't seem to be edit issue, but it looks like someone's a line or two off somewhere and they don't know it yet. -Runfellow (talk) 08:05, 18 December 2012 (UTC)

there were updates made to the "mod" function in #expr in the backend software. I am pretty sure this edit fixed the issue. Frietjes (talk) 18:35, 18 December 2012 (UTC)

Looks like the fix worked, thanks. – Runfellow (talk) 19:16, 18 December 2012 (UTC)

The order of the lines

I suggest, to move line of % humidity below the Avg. precipitation/rainy/snow days' in Weather box. Currently it looks ugly, there are line of Precipitation mm (inches), Avg. rainy days (≥ 1.0 mm) and between this "Precipitation/rainy" lines exist line of % humidity. This line of % humidity between this two lines about "Precipitation/rain" makes unnecessary confusion. The line of % humidity to move below lines about "Precipitation/rain". Should be: lines Precipitation mm (inches), Avg. precipitation days and later % humidity. Subtropical-man (talk) 19:54, 19 January 2013 (UTC)

OK, I fixed it. Example:

Climate data for London (Greenwich)
Month	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec	Year
Mean daily maximum °C (°F)	8.3 (46.9)	8.5 (47.3)	11.4 (52.5)	14.2 (57.6)	17.7 (63.9)	20.7 (69.3)	23.2 (73.8)	22.9 (73.2)	20.1 (68.2)	15.6 (60.1)	11.4 (52.5)	8.6 (47.5)	15.2 (59.4)
Mean daily minimum °C (°F)	2.6 (36.7)	2.4 (36.3)	4.1 (39.4)	5.4 (41.7)	8.4 (47.1)	13.2 (55.8)	14.2 (57.6)	14.0 (57.2)	11.2 (52.2)	8.3 (46.9)	5.1 (41.2)	2.8 (37.0)	7.5 (45.5)
Average precipitation mm (inches)	51.6 (2.03)	38.2 (1.50)	40.5 (1.59)	45.0 (1.77)	46.5 (1.83)	47.3 (1.86)	41.1 (1.62)	51.6 (2.03)	50.4 (1.98)	68.8 (2.71)	58.0 (2.28)	53.0 (2.09)	591.8 (23.30)
Average precipitation days	10.8	8.5	9.6	9.4	9.0	8.3	8.0	7.6	8.5	10.7	10.1	9.9	110.4
Average relative humidity (%)	91	89	91	90	92	92	93	95	96	95	93	91	92
Source: Met Office

Subtropical-man (talk) 20:14, 19 January 2013 (UTC)

Faster Weather_box templates

The current slow processing of Template:Weather_box, often 3-4 seconds with temperatures & rainfall, is one of the primary causes of slow edit-preview of some major city articles. Timing tests have revealed that a climate-table can be formatted within 1/3 second by streamlining the templates to use simpler methods to process the data. If numbers are just calculated with rounding by 1 (or 2 in totals), then amounts can be formatted over 10x faster. Also, when color-tone values are selected by a #switch of 25 branches, then colors can be assigned 9x times faster than using a 3-part hexadecimal encoding algorithm. Compare versions:

Template:Weather_box/colgreen - uses 3-part hexadecimal encoding
Template:Weather_box/colgreen/sandbox - uses #switch of 25 branches, as 9x faster
Template:Weather_box/cold - uses 3-part hexadecimal encoding
Template:Weather_box/cold/sandbox - uses #switch of 25 branches, as 9x faster

I am currently running tests, of alternative methods, to display the basic temperature and rainfall data, but run 11x faster than the original Template:Weather_box from February 2013. This is just a status note, for general consideration, about some ways to make {weather_box} run 11x faster. -Wikid77 (talk) 18:34, 1 March, 22:04, 2 March 2013 (UTC)

I don't like these changes. You've got the same colour for 5°C and 7°C, but 8°C is different. The expressions ensured that every number was different, and it was all at a scale. This is the reason I put so much time into creating the expressions, and moving away from these huge stepped colour changes. 117Avenue (talk) 05:31, 6 March 2013 (UTC)

Default ordering

Since much of the world uses Metric as opposed to Imperial units, shouldn't the former be displayed by fault and the parameter metric_first be changed to imperial_first? Ditto for the separate-line display of conversions. GotR ^Talk 21:28, 4 March 2013 (UTC)

Yes, you're right. In the Weather box, imperial units (not compatible with International System of Units) and Fahrenheit temperature scale are as default. The system of International System of Units has been nearly globally adopted. Only Burma, Liberia and the United States have not adopted SI units. And also Fahrenheit temperature scale, only United States (again), Cayman Islands, Palau, Bahamas and Belize have not adopted Celsius units. About 6.6 billion of the 7.0 billion (94%) of World population uses the metric and Celsius system (compatible with International System of Units), so, there is an error in the template. This should be improved. Subtropical-man (talk) 21:47, 4 March 2013 (UTC)

And neither I nor other users will mind making the changes. It only takes an AWB run to do this, just as occurred with the last major overhaul in August 2010 that removed all the underscores (_). However, I won't have time to undertake this task until UTC 01:00 Wednesday. GotR ^Talk 22:12, 4 March 2013 (UTC)

I just took a quick glance at some articles to do with Liberia, Burma, Cayman Islands, Palau, The Bahamas and Belize. From what I could see all of them are using C and mm. This would mean that only the US is using F and in. So it would seem to me the easiest thing to do would be to have a bot run through all the US articles using the template and add a line "|imperial first = Y" then fix the template and finally remove "|metric first = " from all the other articles. Doing it that way ensures that nothing is broken, even for a short time. See below:

Climate data for the Detroit area
Month	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec	Year
Mean daily maximum °F (°C)	32.0 (0.0)	35.2 (1.8)	45.8 (7.7)	59.1 (15.1)	69.9 (21.1)	79.3 (26.3)	83.4 (28.6)	81.4 (27.4)	74.0 (23.3)	61.6 (16.4)	48.8 (9.3)	36.1 (2.3)	58.9 (14.9)
Mean daily minimum °F (°C)	18.5 (−7.5)	20.5 (−6.4)	28.1 (−2.2)	38.9 (3.8)	49.0 (9.4)	59.1 (15.1)	63.4 (17.4)	62.3 (16.8)	54.3 (12.4)	42.8 (6.0)	33.7 (0.9)	23.6 (−4.7)	41.2 (5.1)
Average precipitation inches (mm)	1.95 (50)	2.01 (51)	2.28 (58)	2.90 (74)	3.38 (86)	3.51 (89)	3.37 (86)	3.00 (76)	3.26 (83)	2.52 (64)	2.78 (71)	2.45 (62)	33.41 (850)
Source 1: NOAA (normals 1981–2010, records 1874–2010)^[1]^[2]
Source 2: HKO (sun, 1961–1990)^[3]

If you fix the template first then all the US articles will have the temperature and precipitation backward and that will generate a lot of complaints. CambridgeBayWeather (talk) 09:57, 5 March 2013 (UTC)

Since no one else has commented within 60 hours of the last post here, I will begin my AWB run through the US articles within two hours. Cleaning up the other nations may be too daunting of a task (even for AWB), however. GotR ^Talk 01:01, 8 March 2013 (UTC)

Lua migration

I have just migrated this template to use Lua via Module:WeatherBox and Module:WeatherBoxColors, the template is now less than 1/3 the size and generates a full box in about 0.6 seconds compared with 7.9 seconds for the old version. There are slight differences in formatting, which you can see at Template:Weather box/testcases. Long live the new Weather Box, death to slow templates. Dragons flight (talk) 20:44, 9 March 2013 (UTC)

Gradient playground

I've roughed out a tool for testing out gradients. The one shown below is the current temperature map. People could use this to discuss new color schemes, if that is what you want to do. Dragons flight (talk) 02:21, 11 March 2013 (UTC)

Current temperature map

{{#invoke:Gradient| gradient_test | gradient=
  {{#invoke:Gradient|export_gradient
  | -90 | 000000 
  | -42.75 | 0000FF 
  | 4.5 | FFFFFF 
  | 23 | FF7F00
  | 41.5 | FF0000
  | 100 | 000000 
  }} 
| step = 1
| cells_per_row = 25
}}

-90 -90	-89 -89	-88 -88	-87 -87	-86 -86	-85 -85	-84 -84	-83 -83	-82 -82	-81 -81	-80 -80	-79 -79	-78 -78	-77 -77	-76 -76	-75 -75	-74 -74	-73 -73	-72 -72	-71 -71	-70 -70	-69 -69	-68 -68	-67 -67	-66 -66
-65 -65	-64 -64	-63 -63	-62 -62	-61 -61	-60 -60	-59 -59	-58 -58	-57 -57	-56 -56	-55 -55	-54 -54	-53 -53	-52 -52	-51 -51	-50 -50	-49 -49	-48 -48	-47 -47	-46 -46	-45 -45	-44 -44	-43 -43	-42 -42	-41 -41
-40 -40	-39 -39	-38 -38	-37 -37	-36 -36	-35 -35	-34 -34	-33 -33	-32 -32	-31 -31	-30 -30	-29 -29	-28 -28	-27 -27	-26 -26	-25 -25	-24 -24	-23 -23	-22 -22	-21 -21	-20 -20	-19 -19	-18 -18	-17 -17	-16 -16
-15 -15	-14 -14	-13 -13	-12 -12	-11 -11	-10 -10	-9 -9	-8 -8	-7 -7	-6 -6	-5 -5	-4 -4	-3 -3	-2 -2	-1 -1	0 0	1 1	2 2	3 3	4 4	5 5	6 6	7 7	8 8	9 9
10 10	11 11	12 12	13 13	14 14	15 15	16 16	17 17	18 18	19 19	20 20	21 21	22 22	23 23	24 24	25 25	26 26	27 27	28 28	29 29	30 30	31 31	32 32	33 33	34 34
35 35	36 36	37 37	38 38	39 39	40 40	41 41	42 42	43 43	44 44	45 45	46 46	47 47	48 48	49 49	50 50	51 51	52 52	53 53	54 54	55 55	56 56	57 57	58 58	59 59
60 60	61 61	62 62	63 63	64 64	65 65	66 66	67 67	68 68	69 69	70 70	71 71	72 72	73 73	74 74	75 75	76 76	77 77	78 78	79 79	80 80	81 81	82 82	83 83	84 84
85 85	86 86	87 87	88 88	89 89	90 90	91 91	92 92	93 93	94 94	95 95	96 96	97 97	98 98	99 99	100 100

More rows?

Like the record highs/lows for temperatures, should we include a record high/low for rainfall and snowfall? Also, for the temperatures, should a "record high for low"/"record low for high" (highest recorded monthly min. temp/lowest recorded monthly max. temp) be added to the box? Koopatrev (talk) 08:49, 18 March 2013 (UTC)

unnecessary precision

Currently if I supply rainfall in millimetres, it converts to 3 decimal places of inches, which seems quite absurd. For instance 2293 mm comes out as 90.276 inches. Can the precision be reduced to 1D, either for all or as a parameter?

John of Cromer in Philippines (talk) mytime= Wed 20:42, wikitime= 12:42, 20 March 2013 (UTC)

John, if you are specifying rain to the nearest 0.1 mm (as appears in this article you recently edited), then the corresponding uncertainty on the inch value is 0.004 inches and three decimal places is a reasonable way to represent that. In such a case, rounding inches to the nearest 0.1 inch would significantly misrepresent the implied precision. For the record, in your example 2293 mm would be shown as 90.28 inches, since in this case your implied rounding is to the nearest mm (equivalent to 0.04 inches). Dragons flight (talk) 15:51, 20 March 2013 (UTC)

The thing is that the only country that uses inches for rain/snow/precipitation is the US and they measure to two decimal places. There has to be a cut off in the precision somewhere and given the US standard, two decimal places seems logical. CambridgeBayWeather (talk) 16:22, 20 March 2013 (UTC)

"then the corresponding uncertainty on the inch value is 0.004 inches and three decimal places is a reasonable way to represent that."

I think 3 decimal places of an inch is both unreasonable and infeasible – I doubt it is possible to measure rainfall to a thousandth of an inch, and it just looks absurd. In fact I doubt it is possible to measure to 2D with normal equipment. We are trying to present useful information here, not an exercise in heuristics. The data in Tacloban was obtained from weatherbase. I guess they already converted integer inches to obtain mm in the first place. I have now rounded data to integer mm but it still appears as 2D inches. Incidentally trailing zero is not supplied so one figure appears as 99.9 not 99.90 as it should by your reckoning.

I also don't like the shading - why can't I specify a different colour other than the two you offer? My choice is lightgrey (or lightgray), or white.

I also note there is a line break in the first column in a strange place - so it appears as

high (

°F)

I'd be quite happy for no imperial (US) units to appear at all.

I also don't like the tgng on the dates and the word 'avg'. What's the point of it?

John of Cromer in Philippines (talk) mytime= Thu 09:10, wikitime= 01:11, 21 March 2013 (UTC)

seems like simply ditching the template for a hard-coded table is a step backwards. Frietjes (talk) 15:11, 21 March 2013 (UTC)

It's actually not that hard, modern rain gauges will concentrate water over a large collecting area into a narrow measuring vessel. If the cross-section of the measuring column is 1/100 of the collection area cross-section, then you'll get a column that amplifies the height of the collected water by 100-fold. Each 0.1 mm of rainfall can be represented by 10 mm of column height, which is then easy to measure. Personally, I think it is rather silly to be that precise since microscale effects will contribute significant noise on that scale, but it is not an intractable problem from the point of view of reading a gauge.

Back to the other point if one reports the rainfall as 20.2 mm, then presumably the person reading the range gauge intends to say that the measured value was in the range 20.15 to 20.25 mm. If you convert that to inches you get 0.7933 to 0.7972 inches, with a central average of 0.7952 inches. If you round that to one decimal place, you get 0.8 inches, which is outside the implied range. Similarly, rounding the central value to 2 decimal places is 0.80 in, which is against outside the implied precision range for a number like 20.2 mm. The only way to consistently represent numbers like 20.2 mm in inches without sometimes falling outside the range implied by 0.1 mm precision reporting is to use three decimal places, i.e. 0.795 inches. Personally, as a scientist, if one is going to convert numbers at all then I believe one should include enough digits to maintain a comparable level of implied precision, which in this case means adding two digits when converting mm to inches.

Also, I agree with Frietjes that substituting the chart with a fixed table is generally the wrong thing to do. Dragons flight (talk) 17:08, 21 March 2013 (UTC)

To answer your questions;

"In fact I doubt it is possible to measure to 2D with normal equipment." That depends on what you mean by "normal equipment". To me that means the equipment normally used by the meteorological equipment used by the agency in charge of collecting the data. In the US that would be the NOAA National Weather Service. They would certainly have the rain gauges capable of measuring to .01 in. I found an old tube in our storage room that is in inches and it is marked in .01 graduations. Not sure how old the tube is but it must be from before 1978 as that's when I started and we have always used mm. If on the other hand you mean equipment that any body can buy then the answer is yes, Rain Gauge, Gauges - Digital Self Emptying and Sight Glass and the inexpensive Stratus RG202 Long Term Professional Rain Gauge and Snow Gauge.
"The data in Tacloban was obtained from weatherbase. I guess they already converted integer inches to obtain mm in the first place." I doubt it. Weatherbase appears to be a US company, contact section and the Philippines seems to be a metric country. Also the information at Philippine Atmospheric, Geophysical and Astronomical Services Administration is presented in metric. I would think that it is more likely that Weatherbase converted from mm/C to in/F rather than the other way round.
"Incidentally trailing zero is not supplied so one figure appears as 99.9 not 99.90 as it should by your reckoning." Yes I agree the figure should include the zero and all figures (inches) should be to two decimal places. Right now the figures are one, two and three decimal places.
"I also don't like the shading - why can't I specify a different colour other than the two you offer? My choice is lightgrey (or lightgray), or white." I'm not sure who the "you" is supposed to be. If you wanted a different colour why did you not take part in the discussions that can be found further up this page? Or start a new section.
You noted that there is a line break in the first column and I assume you mean here I'm not seeing that so it is probably due to your screen size and/or the fact that the weather box was set to 70%. Because of the screen I'm using combined with the clear template and the "Statue of MacArthur and Osmeña" image I'm seeing a large amount of white space. Looking at the page without the clear template seems fine to me but may not be to others.
"I'd be quite happy for no imperial (US) units to appear at all." Who wouldn't but I suspect that a lot of US based readers and editors along with a certain segment of other from countries that changed to the metric system in the last 30 - 40 years would be upset.
"I also don't like the tgng on the dates and the word 'avg'. What's the point of it?" Take a look at Template talk:Weather box#Abbreviation of "Average" (Avg.), Template talk:Weather box/Archive 4#Abbreviations and most importantly Category talk:Wikipedia formatting and function templates#New template requested — Discreet abbreviation. Basically the tagging is to make the page more accessible to people using screen readers.

Hope this helps. CambridgeBayWeather (talk) 19:29, 21 March 2013 (UTC)

Why did I change? Well

(a) we're talking historic weather data for an equatorial climate. I doubt the figures will move at all, and even if they did, I don't see how it's more of a hardship to change a fixed table than a dynamic one. (I did try SUBSTing, but it didn't work, probably because of the {{stack}}). I don't actually see how {{table}} is more hard-coded than {{weather box}} (it was Frietjes who changed it to really hard-coded). In either case any change has to be entered by hand.

(b) I already noted my objections to the shading, which apart from being hideous, didn't actually offer me any explanation of what it meant.

(c) The generated table didn't align along the right edge of the page

(d) For the interested passer-by, I'm sure that integer millimetres and inches to 1D are entirely adequate. This is a general encyclopedic article (or meant to be), not a scientific treatise.

(e) Even if the data were presented in a completely foreign language, for instance Cherokee, I could deduce that 12 columns represented months, and one for the year.

The only rain gauge I recall seeing was basically a funnel and a graduated beaker.

I have a notebook with a 10" screen.

John of Cromer in Philippines (talk) mytime= Fri 20:48, wikitime= 12:48, 22 March 2013 (UTC)

(b) - What is or is not hideous is a matter of opinion. I find the the white box ugly and I don't see that it gives a better explanation than the standard.

(c) - That's because it was set to a 70% width. Remove that and the box is centred.

(d) - It should reflect the original and the conversions should be standard throughout.

(e) - It's not what we could deduce after looking at the table. It's what the screen reader can speak to the person who is unable to see the table.

That's all a standard rain gauge is, File:Standard rain Guage.JPG and http://commons.wikimedia.org/wiki/Rain_gauge. It's inexpensive, long lasting and hard to screw up. CambridgeBayWeather (talk) 16:59, 25 March 2013 (UTC)

The colour used in Weatherboxes

From the previous post

Is there any recommendation about the usage of temperature colour= "pastel" and rain colour= "green"? Why are there several colour schemes in the first place? Isn't it better if the colours are consistent across articles? Calimo (talk) 09:18, 12 February 2012 (UTC)

The green precipitation/rain colour is useful to avoid blending of snowfall with record lows, wind chill values, # of precipitation, snowy, and rainy days and humidity values an example with blending of blue colours and green colouring is used to avoid blending. In addition, the green precipitation/rain colour is more useful in warm and places that are not really cold (such as humid subtropical, humid continental climate, semi-arid, and arid climates) as the blue colouring can leave a false impression of a place that is very wet, and cold, even though it is not cold from a thermal standpoint sample. No it is not better to make the colours consistent in every article due to the different climates that each city has, although it would be better for all the cities in one country to have the same colouring since the cities in that country may have a similar climate pattern. Pastel colouring is better to use in tropical areas since the temperature range is not large (as pastel goes only down to −30 °C (−22.0 °F)) before all the temperatures below that look the same and since using in places with cold winters and warm summers make it confusing to read. The green colour is not good for places that are really cold like Yellowknife and Iqaluit or subpolar oceanic climates like Ushuaia (so standard colouring is better). Ssbbplayer (talk) 20:28, 17 September 2012 (UTC)

Removal of Green Colours

User:Subtropical-man has objected to the addition of green colours used in the weatherboxes stating that it is not the standard on wikipedia. If he/she stated that it is not appropriate then please explain why before there is an outright war over whether to include green or use blue or that why the green colour was introduced in the first place. Ssbbplayer (talk) 13:06, 8 October 2012 (UTC)

Well, green is not standard because it isn't used much. You give good reasons to change that. However you need a consensus before you make this change. Until then it is not standard, and likely to be reverted.

IMO we could remove the blue altogether, it is not so bad for places like Iqaluit to be in green. What is important is to stay consistent.

Also I'm not really convinced by the pastels. If temperature doesn't change (as in tropical climates), then the color shouldn't change much either. Potentially we could use pastels of green for consistency.

Calimo (talk) 14:50, 9 October 2012 (UTC)

Temperature colours change in subtropical/tropical climates, Ssbbplayer is wrong. Please see: [2]; 28, 33, 37, 39, 45 or 47°C - the next higher temperatures = more darker red. Subtropical-man (talk) 23:06, 16 October 2012 (UTC)

Very strong oppose for pastels. Pastels colours is totally unacceptable, weatherbox looks like a colorful Christmas tree or colorful puzzle or coloring book. These colors are good for reflective vests, not for encyclopedia. Sore eyes while watching "pastels" weatherbox. Second: green has one advantage - it does not blend with very low temperatures but this green tint (from option of "|precipitation colour= green") is too bright and unnatural = I am against. Current standard colours is good, I do not see the need for changes. Subtropical-man (talk) 16:45, 16 October 2012 (UTC)

How is the green colouring unnatural? Rain is colourless. Ssbbplayer (talk) 21:45, 16 October 2012 (UTC)

Green tint from option of "|precipitation colour= green" is too bright and unnatural. Rain is colourless? It does not matter. Green tint from option of "|precipitation colour= green" is good for reflective vests. Subtropical-man (talk) 16:45, 16 October 2012 (UTC)

I like using the green colour. It makes the precipitation and rainfall values stand out from the other values like record lows and snowfall. It is better, because I think that too much blue colouring makes it hard to read and gives a place that is relatively dry a false impression of a wet and cold climate. Look at these examples then.Ssbbplayer (talk) 21:41, 17 October 2012 (UTC)

This is your personal opinion. Though you may think it is an eyesore, the only thing that is more of an eyesore is if a vandal messes up with the climate data and puts in a bunch of ridiculous values. Similarly, I think green colours is better. None of us is right or wrong and it does not mean I and you have the right to just randomly go to city articles and just change colours to fit our preferences. Please explain why these options were available in the first place?. Remember that this isn't the first time about the usage of green colours.Ssbbplayer (talk) 22:05, 16 October 2012 (UTC)

This discussion should involve more users, not just three editors.Ssbbplayer (talk) 22:20, 16 October 2012 (UTC)

I am strongly opposed to pastels. I am not strongly opposed to using the green colour. You're right, it makes the precipitation and rainfall values stand out from the other values like record lows but this colour (specifically this tone of colour) has a major flaw: looks bad (as reflective vest). So, if blue colour has a disadvantage and advantage and also green colour has a disadvantage and advantage - there is no need to change. Also, if green option is to be effective, must create "green" option for avg. precipitation/rain days because it can not be that Precipitation/Rain mm (inches) is green and avg. precipitation/rain days is blue. Subtropical-man (talk) 22:01, 17 October 2012 (UTC)

The green was put in because an editor thought it a good idea. That's what Wikipedia is about. I prefer the blue for precipitation but won't edit war to restore it. I will point out that having the precipitation and rain line in green and the snow in blue is very wrong. If the first two lines are green then the snow must be green as well. Snow is a form of precipitation and having it a separate colour seems to indicate that it has nothing to with the total precipitation in the template. See Rainfall, Snowfall, and Precipitation. CambridgeBayWeather (talk) 07:12, 18 October 2012 (UTC)

Yes, I noticed the same thing. Precipitation/rain/snow and avg. precipitation/rain/snow days should be one colour. Subtropical-man (talk) 13:16, 18 October 2012 (UTC)

New proposal

I also have a new proposal - change the colour of cold temperatures from blue to violet. Violet colour is more suitable for cold temperatures than blue and also solves the problem of mixing colours of low temperatures with precipitations. For example (this example is merely indicative, colors used from pastels option, I'd prefer a more gentle tones):

Climate data for some places
Month	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec	Year
Mean daily minimum °C (°F)	0 (32)	−2 (28)	−4 (25)	−6 (21)	−8 (18)	−10 (14)	−12 (10)	−14 (7)	−16 (3)	−18 (0)	−20 (−4)	−22 (−8)	−11 (12)
Average rainfall mm (inches)	7 (0.3)	3 (0.1)	2 (0.1)	8 (0.3)	48 (1.9)	105 (4.1)	119 (4.7)	121 (4.8)	84 (3.3)	59 (2.3)	44 (1.7)	24 (0.9)	624 (24.5)
Average rainy days (≥ 1.0 mm)	1.3	0.5	0.5	1.2	4.7	10.8	12.7	12.7	9.4	7.4	5.8	3.6	70.6
Source: some source

Generally, this idea is based only on change one colour, from blue to violet in cold temperatures. Changes will only from 0°C (32°F) to negative temperature, rest of colours remained to be the same. Subtropical-man (talk) 13:20, 18 October 2012 (UTC)

I supported it, also user Ssbbplayer initially supported it (voice paused[6]). What is the opinion of other users? Subtropical-man (talk) 17:20, 26 January 2013 (UTC)

If I understand it correctly, we would have rain in blue, and temperature in purple (cold) / red (warm). I think that would be good. Water is intrinsically slightly blue, not green. I still disagree with the usage of pastels though. Equatorial places with constant or nearly constant temperature should have no change in color as well.

I guess the discussion should now move to the Village Pump to get a broader consensus? This change will affect lots of articles. Calimo (talk) 16:05, 28 January 2013 (UTC)

there is no need a move discussion and get a broader consensus, discussion and consensus on this talk page is sufficient. Subtropical-man (talk) 17:35, 28 January 2013 (UTC)

I think that Calimo is correct. Given the number of pages using this more voices would be better. CambridgeBayWeather (talk) 17:56, 28 January 2013 (UTC)

Climate data for some places
Month	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec	Year
Mean daily maximum °C (°F)	−20 (−4)	−25 (−13)	−30 (−22)	−35 (−31)	−40 (−40)	−45 (−49)	−21 (−6)	−22 (−8)	−23 (−9)	−24 (−11)	−25 (−13)	—	—
Mean daily minimum °C (°F)	40 (104)	35 (95)	30 (86)	25 (77)	20 (68)	15 (59)	10 (50)	5 (41)	0 (32)	−5 (23)	−10 (14)	−15 (5)	13 (55)
Source: some source

Currently I'm not sure if the colour change is good or not. Looking at the above I see that -22 to -23 are the same colour and -25 and colder are all one colour. There needs to be a greater range of colours after -25. CambridgeBayWeather (talk) 18:34, 28 January 2013 (UTC)

Note, that the entire range of colors will be moved, +8 in currently pastel weatherbox = -1 in new version weatherbox. Your example concerns the current weatherbox with the pastels. Current weatherbox with the pastels uses the colour violet from +8 (see below), therefore to about -35 there are changes in colors. Temperature of -35 and below is extreme temperatures, only a small part of the articles of Wikipedia will be show such low temperatures (for example: Murmansk located in the extreme northwest part of Russia (Arctic Circle) has this temperatures only as Record low in winter, McMurdo Station in Antarctica has −31.8 as low °C in one month). Therefore it is not necessary to greater range of colours. Notwithstanding, I think, though that we can do it.

Climate data for some places
Month	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec	Year
Mean daily minimum °C (°F)	10 (50)	8 (46)	6 (43)	4 (39)	2 (36)	0 (32)	−2 (28)	−4 (25)	−6 (21)	−8 (18)	−10 (14)	−12 (10)	−1 (30)
Source: some source

Subtropical-man (talk) 19:04, 28 January 2013 (UTC)

There are many varieties of violet, range of colour of violet is similarly large as range of colour of blue. Examples of shades:

{{#invoke:Gradient| gradient_test | gradient=
  {{#invoke:Gradient|export_gradient
  | 0 | #E6E6FA
  | -45 | #B57EDC
  | -90 | 000000
  }} 
| step = 1
| cells_per_row = 25
}}

-90 -90	-89 -89	-88 -88	-87 -87	-86 -86	-85 -85	-84 -84	-83 -83	-82 -82	-81 -81	-80 -80	-79 -79	-78 -78	-77 -77	-76 -76	-75 -75	-74 -74	-73 -73	-72 -72	-71 -71	-70 -70	-69 -69	-68 -68	-67 -67	-66 -66
-65 -65	-64 -64	-63 -63	-62 -62	-61 -61	-60 -60	-59 -59	-58 -58	-57 -57	-56 -56	-55 -55	-54 -54	-53 -53	-52 -52	-51 -51	-50 -50	-49 -49	-48 -48	-47 -47	-46 -46	-45 -45	-44 -44	-43 -43	-42 -42	-41 -41
-40 -40	-39 -39	-38 -38	-37 -37	-36 -36	-35 -35	-34 -34	-33 -33	-32 -32	-31 -31	-30 -30	-29 -29	-28 -28	-27 -27	-26 -26	-25 -25	-24 -24	-23 -23	-22 -22	-21 -21	-20 -20	-19 -19	-18 -18	-17 -17	-16 -16
-15 -15	-14 -14	-13 -13	-12 -12	-11 -11	-10 -10	-9 -9	-8 -8	-7 -7	-6 -6	-5 -5	-4 -4	-3 -3	-2 -2	-1 -1	0 0

OR

{{#invoke:Gradient| gradient_test | gradient=
  {{#invoke:Gradient|export_gradient
  | 0 | #8F00FF
  | -90 | 000000
  }} 
| step = 1
| cells_per_row = 25
}}}

-90 -90	-89 -89	-88 -88	-87 -87	-86 -86	-85 -85	-84 -84	-83 -83	-82 -82	-81 -81	-80 -80	-79 -79	-78 -78	-77 -77	-76 -76	-75 -75	-74 -74	-73 -73	-72 -72	-71 -71	-70 -70	-69 -69	-68 -68	-67 -67	-66 -66
-65 -65	-64 -64	-63 -63	-62 -62	-61 -61	-60 -60	-59 -59	-58 -58	-57 -57	-56 -56	-55 -55	-54 -54	-53 -53	-52 -52	-51 -51	-50 -50	-49 -49	-48 -48	-47 -47	-46 -46	-45 -45	-44 -44	-43 -43	-42 -42	-41 -41
-40 -40	-39 -39	-38 -38	-37 -37	-36 -36	-35 -35	-34 -34	-33 -33	-32 -32	-31 -31	-30 -30	-29 -29	-28 -28	-27 -27	-26 -26	-25 -25	-24 -24	-23 -23	-22 -22	-21 -21	-20 -20	-19 -19	-18 -18	-17 -17	-16 -16
-15 -15	-14 -14	-13 -13	-12 -12	-11 -11	-10 -10	-9 -9	-8 -8	-7 -7	-6 -6	-5 -5	-4 -4	-3 -3	-2 -2	-1 -1	0 0

OR

{{#invoke:Gradient| gradient_test | gradient=
  {{#invoke:Gradient|export_gradient
  | 0 | #E6E6FA
  | -15 | #B57EDC
  | -50 | #8F00FF
  | -90 | 000000
  }} 
| step = 2
| cells_per_row = 25
}}

-90 -90	-88 -88	-86 -86	-84 -84	-82 -82	-80 -80	-78 -78	-76 -76	-74 -74	-72 -72	-70 -70	-68 -68	-66 -66	-64 -64	-62 -62	-60 -60	-58 -58	-56 -56	-54 -54	-52 -52	-50 -50	-48 -48	-46 -46	-44 -44	-42 -42
-40 -40	-38 -38	-36 -36	-34 -34	-32 -32	-30 -30	-28 -28	-26 -26	-24 -24	-22 -22	-20 -20	-18 -18	-16 -16	-14 -14	-12 -12	-10 -10	-8 -8	-6 -6	-4 -4	-2 -2	0 0

OR other.

You can choose different shades, and experiment further. Subtropical-man (talk) 16:11, 20 March 2013 (UTC)

See also: Template_talk:Weather_box#RfC_-_Colours_to_be_used_in_the_weather_box. Subtropical-man (talk) 20:35, 17 May 2013 (UTC)

Abbreviation of "Average" (Avg.)

Are we so short of space that it is necessary to abbreviate "Average" to "Avg."?

On pages that don't have all the lines shown, such as Timbuktu#Climate:

Climate data for Timbuktu
Month	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec	Year
Record high °C (°F)	36 (97)	39 (102)	42 (108)	45 (113)	48 (118)	47 (117)	47 (117)	43 (109)	47 (117)	47 (117)	40 (104)	42 (108)	48 (118)
Mean daily maximum °C (°F)	30.0 (86.0)	33.2 (91.8)	36.6 (97.9)	40.0 (104.0)	42.2 (108.0)	41.6 (106.9)	38.5 (101.3)	36.5 (97.7)	38.3 (100.9)	39.1 (102.4)	35.2 (95.4)	30.4 (86.7)	36.8 (98.3)
Mean daily minimum °C (°F)	13.0 (55.4)	15.2 (59.4)	18.5 (65.3)	22.5 (72.5)	26.0 (78.8)	27.3 (81.1)	25.8 (78.4)	24.8 (76.6)	24.8 (76.6)	22.7 (72.9)	17.7 (63.9)	13.5 (56.3)	21.0 (69.8)
Record low °C (°F)	5 (41)	8 (46)	7 (45)	10 (50)	15 (59)	21 (70)	17 (63)	21 (70)	18 (64)	12 (54)	7 (45)	1 (34)	1 (34)
Average precipitation mm (inches)	0.6 (0.02)	0.1 (0.00)	0.1 (0.00)	1.0 (0.04)	4.0 (0.16)	16.4 (0.65)	53.5 (2.11)	73.6 (2.90)	29.4 (1.16)	3.8 (0.15)	0.1 (0.00)	0.2 (0.01)	182.8 (7.2)
Average rainy days	1	0	0	0	1	1	3	4	3	1	0	0	14
Source 1: World Meteorological Organization
Source 2: Weatherbase

it looks strange to abbreviate "Average" when

There is clearly enough space (lines above it are longer), and
Other instances of "Average" are not abbreviated.

I can understand abbreviating month names, but "Average" seems unnecessary. Thanks, cmɢʟee୯ ͡° ̮د ͡° ੭ 18:45, 28 January 2013 (UTC)

Another example, please see:

Climate data for London
Month	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec	Year
Record high °C (°F)	18.5 (65.3)	19.7 (67.5)	25.3 (77.5)	29.8 (85.6)	32.8 (91.0)	35.6 (96.1)	36.5 (97.7)	38.5 (101.3)	35.4 (95.7)	29.9 (85.8)	21.1 (70.0)	17.7 (63.9)	38.5 (101.3)
Mean daily maximum °C (°F)	8.3 (46.9)	8.5 (47.3)	11.4 (52.5)	14.2 (57.6)	17.7 (63.9)	20.7 (69.3)	23.2 (73.8)	22.9 (73.2)	20.1 (68.2)	15.6 (60.1)	11.4 (52.5)	8.6 (47.5)	15.2 (59.4)
Mean daily minimum °C (°F)	2.6 (36.7)	2.4 (36.3)	4.1 (39.4)	5.4 (41.7)	8.4 (47.1)	13.2 (55.8)	14.2 (57.6)	14.0 (57.2)	11.2 (52.2)	8.3 (46.9)	5.1 (41.2)	2.8 (37.0)	7.5 (45.5)
Record low °C (°F)	−10.0 (14.0)	−9.0 (15.8)	−8.0 (17.6)	−2.0 (28.4)	−1.0 (30.2)	5.0 (41.0)	7.0 (44.6)	6.0 (42.8)	3.0 (37.4)	−4.0 (24.8)	−5.0 (23.0)	−7.0 (19.4)	−10.0 (14.0)
Average rainfall mm (inches)	51.6 (2.03)	38.2 (1.50)	40.5 (1.59)	45.0 (1.77)	46.5 (1.83)	47.3 (1.86)	41.1 (1.62)	51.6 (2.03)	50.4 (1.98)	68.8 (2.71)	58.0 (2.28)	53.0 (2.09)	591.8 (23.30)
Average rainy days (≥ 1.0 mm)	10.8	8.5	9.6	9.4	9.0	8.3	8.0	7.6	8.5	10.7	10.1	9.9	110.4
Source: Met Office

In this case, the whole word of "Average" may extend field. Subtropical-man (talk) 19:44, 28 January 2013 (UTC)

If the weather box is too large and creates too much excess white space, I suggest using the parameter shown on the template page (you can choose between 0%-100%, not just 50%. I prefer using 80%). I prefer keeping the whole term than the abbreviated form. Either you can use that way (I see it as a more practical way) or maybe the default size of the weatherbox can be changed (but it is more controversial as it affects all weatherboxes). On other wikis, the weatherboxes tend to be smaller than the one in English wiki. Ssbbplayer (talk) 15:18, 7 May 2013 (UTC)

Precipitation colours

Currently the precipitation/rain/snow lines can be set to green as opposed to the default blue. What I would like to see is that the green is extended to the three average days lines for consistency. At the same time rather than have to put in "|precipitation colour =", "|rain colour =" and "|snow colour =" it would only be necessary to put in "|precipitation colour=" and all the lines would be green. CambridgeBayWeather (talk) 18:05, 10 March 2013 (UTC)

By "all the lines would be green", do you mean that the average snow accumulation line would be green as well? For locations with both heavy precipitation and snow (such as Yakutat, AK), the colours would blend far too much, and the idea of "green snow" seems absurd. GotR ^Talk 18:20, 10 March 2013 (UTC)

Precipitation/rain/snow lines should be blue, and needs to be improved colour of frost to violet colour, see: Template_talk:Weather_box#New_proposal. Subtropical-man (talk) 18:37, 10 March 2013 (UTC)

Of course the precip/rain/snow should be blue but we currently have the option to make them green. Snow is a type of Precipitation (meteorology) so if one type is set to green all should be green.

Climate data for Yakutat, Alaska (Yakutat State Airport), 1981−2010 normals
Month	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec	Year
Average precipitation inches (mm)	13.66 (347)	10.86 (276)	10.98 (279)	9.19 (233)	8.21 (209)	6.39 (162)	7.88 (200)	14.15 (359)	21.11 (536)	22.01 (559)	14.46 (367)	16.28 (414)	155.18 (3,941)
Average snowfall inches (cm)	31.3 (80)	28.2 (72)	28.5 (72)	11.0 (28)	0.5 (1.3)	0 (0)	0 (0)	0 (0)	0 (0)	3.2 (8.1)	20.4 (52)	27.4 (70)	150.4 (382)
^{[citation needed]}

The reason for the green was to stand out from the blue colour. The green blends no worse than the blue does;

Climate data for Yakutat, Alaska (Yakutat State Airport), 1981−2010 normals
Month	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec	Year
Average precipitation inches (mm)	13.66 (347)	10.86 (276)	10.98 (279)	9.19 (233)	8.21 (209)	6.39 (162)	7.88 (200)	14.15 (359)	21.11 (536)	22.01 (559)	14.46 (367)	16.28 (414)	155.18 (3,941)
Average snowfall inches (cm)	31.3 (80)	28.2 (72)	28.5 (72)	11.0 (28)	0.5 (1.3)	0 (0)	0 (0)	0 (0)	0 (0)	3.2 (8.1)	20.4 (52)	27.4 (70)	150.4 (382)
^{[citation needed]}

I meant all six lines that refer to prcip, precip/snow/rain/prcip days/rain days/snow days. I was happy enough with the violet colour as long as the range of colours at low temperatures was greater. CambridgeBayWeather (talk) 20:31, 10 March 2013 (UTC)

The range of violet colours at low temperatures is large (not less than blue), only in pastels option is smaller range. Subtropical-man (talk) 20:44, 10 March 2013 (UTC)

CBW, I meant to state that setting precip to green and snow to blue amends the blending. Apologies for any confusion

Climate data for Yakutat, Alaska (Yakutat State Airport), 1981−2010 normals
Month	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec	Year
Average precipitation inches (mm)	13.66 (347)	10.86 (276)	10.98 (279)	9.19 (233)	8.21 (209)	6.39 (162)	7.88 (200)	14.15 (359)	21.11 (536)	22.01 (559)	14.46 (367)	16.28 (414)	155.18 (3,941)
Average snowfall inches (cm)	31.3 (80)	28.2 (72)	28.5 (72)	11.0 (28)	0.5 (1.3)	0 (0)	0 (0)	0 (0)	0 (0)	3.2 (8.1)	20.4 (52)	27.4 (70)	150.4 (382)
^{[citation needed]}

GotR ^Talk 21:46, 10 March 2013 (UTC)

Comment: After discussing it with other editors (see Talk:Toronto#colours and Colour of weatherboxes in Canada) it is appropriate that the precipitation colours for all fields should be one colour. It makes sense too because snow is a type of precipitation as well and would reduce the confusion from other editors/readers when they look at the climate data. Given that blending can be of an issue with green colours, it is rare to occur because snowfall is pretty much measured only in Canada and USA and parts of Europe. It is rare for a climate in the above example like that to occur in which blending occurs for both colours and only green for precip and blue for snow will resolve blending. 15:50, 9 May 2013 (UTC)Ssbbplayer (talk)

See also: Template_talk:Weather_box#RfC_-_Colours_to_be_used_in_the_weather_box for more information. Ssbbplayer (talk) 19:49, 18 May 2013 (UTC)

^ Cite error: The named reference NCDC was invoked but never defined (see the help page).
^ Cite error: The named reference SE MI extremes was invoked but never defined (see the help page).
^ "Climatological Normals of Detroit". Hong Kong Observatory. Retrieved May 13, 2010.

[NCDC-1] Cite error: The named reference NCDC was invoked but never defined (see the help page).

[SE_MI_extremes-2] Cite error: The named reference SE MI extremes was invoked but never defined (see the help page).

[HKO-3] "Climatological Normals of Detroit". Hong Kong Observatory. Retrieved May 13, 2010.

[1]

[2]

[3]