Science desk
< July 11	<< Jun \| July \| Aug >>	July 13 >

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

July 12

Variation in time of dusk

In Frederick Forsyth's Day Of The Jackal, one attempt on De Gaulle's life failed because it was darker than expected, so his car was on the gunmen before they realised. The organiser had used a calendar of the previous year to give the time of dusk (20.35 on 22 August 1961), but on the same date in 1962 dusk was at 20.10, fifteen minutes earlier. Supposedly this was an astronomical fact, nothing to do with local weather conditions. Is such an annual variation possible? If so, what is the pattern, as continuing movement in one direction is not possible? →2A00:23C6:AA08:E500:41E5:1052:C726:931C (talk) 11:52, 12 July 2020 (UTC)[reply]

Good catch! Same-date year-to-year changes are primarily due to the difference between the tropical year the and calendar year and are never larger than the day-to-day change, for which fifteen minutes is way too big for Paris's moderate latitude of 49°. (Expect a same-date year-to-year change to be roughly 3/4 of the day-to-day change if there is an intervening leap day and 1/4 of the day-to-day change in the opposite direction if there isn't.)

The NOAA Improved Sunrise/Sunset Calculator doesn't give dusk or twilight times, but does gives the following apparent sunset times for Paris:

1961-08-22: Sunrise=05:51 Sunset=19:52
1962-08-22: Sunrise=05:51 Sunset=19:53

Timeanddate.com does calculate twilight times, and gives:

1961-08-22: Sunrise=05:52 Sunset=19:53 End of Civil Twilight=20:27
1962-08-22: Sunrise=05:52 Sunset=19:53 End of Civil Twilight=20:27

I don't know if the 1 minute discrepancy is due to differences in the calculators or in their precise chosen location for Paris, but they are rounding errors compared to Forsyth's claimed 15 minutes. -- ToE 13:41, 12 July 2020 (UTC)[reply]

We have a description of the actual assassination attempt at Jean Bastien-Thiry#Assassination attempt, and to a lesser extent at Charles de Gaulle#Assassination attempts. From the former:

On 22 August 1962, while Bastien-Thiry functioned as a lookout, de Gaulle's car (a Citroën DS) and some nearby shops were raked with machine-gun fire. De Gaulle and his wife and entourage escaped, uninjured. After the attempt, holes from fourteen bullets were found in the president's vehicle, one of which barely missed the president's head; another twenty were found to have struck the nearby Café Trianon; and an additional 187 spent shell casings were found on the pavement. De Gaulle was said to have credited the unusual resilience of the Citroën DS with saving his life: even though the shots had punctured two of the armoured tires, the car escaped at full speed.

-- ToE 13:52, 12 July 2020 (UTC)[reply]

Ummmmm.... the difference between those two times is 25 minutes, not 15.--Khajidha (talk) 13:56, 12 July 2020 (UTC)[reply]

Yes, I've just noticed that, so even more unreasonable. Thanks for clarification.→2A00:23C6:AA08:E500:75AB:AB90:72E8:F3BF (talk) 14:02, 12 July 2020 (UTC)[reply]

And this is why we don't replace Gregorian with leap week calenders, having the same calendar every year is not worth fake dates. Sagittarian Milky Way (talk) 14:04, 12 July 2020 (UTC)[reply]

Anyway, having the same calendar every year would be less convenient than the present system for multiple reasons. --174.89.49.204 (talk) 16:27, 13 July 2020 (UTC)[reply]

It's not even the same every year cause 52/293rds of years are 371 days, if and only if the remainder of ( 52 * 4 digit year + 146 ) / 293 is less than 52. Simple. /s Sagittarian Milky Way (talk) 03:41, 14 July 2020 (UTC)[reply]

fr-Wikipedia has an article about that assassination attempt. It says the shooting starts at 20:20 but is otherwise silent on the luminosity conditions. It also says Bastien-Thiry was posted as a lookout and gave the signal by waving a newspaper, from which one could infer it was not too dark yet. The relevant sources are not online (refs 18 to 22 in the fr article) so I did not check them. Tigraan^{Click here to contact me} 12:49, 13 July 2020 (UTC)[reply]

I presume that B-T waved the paper when he first saw the car, which in view of the conditions may well have been when it was nearly at him. The gunmen would have assumed that he did so at the agreed distance and that they had time to aim, and would have been totally disoriented by the sudden appearance of the black Citroen travelling nearly head-on at 70 mph. A white newspaper being waved transversely would have been visible in relative gloom. →2A00:23C6:AA08:E500:789B:13C:EAB8:733B (talk) 21:43, 15 July 2020 (UTC)[reply]

Analemma would be a good read for this topic. Specifically this is roughly the extent of the 12-month cycle of the deviations of clock noon vs. astronomical noon (Sun in zenith). This will probably dominate the day-to-day changes in sunset times. (For example I live on the 46th parallel and earliest sunset comes more than a month before the latest sunrise).

However, in more northern latitudes near the polar circle, sunrise/sunset times don't follow a sinusoidal-like pattern and day-to-day sunset time variations near the polar day/night onset are much bigger. Being a day off due to the tropical year wandering could definitely mean screwing up by 25 minutes. For example using the NOAA calculator above: in Thule (Greenland), on 23 August 2016 the sun rose at 2:40, but on 23 August 2015 sunrise was already at 2:17. 93.136.16.89 (talk) 18:13, 16 July 2020 (UTC)[reply]