NA returned for night with some coordinates via getSunlightTimes

[mfidino]

Using suncalc 0.5.0

Some code that recreates the issue:

suncalc::getSunlightTimes(
  date = as.Date("2018-05-24"),
  lat = 53.53043,
  lon = -113.51,
  keep = c("dusk", "night", "dawn","nightEnd"),
  tz = "US/Mountain"
)

which returns

        date      lat     lon                dusk night                dawn nightEnd
1 2018-05-24 53.53043 -113.51 2018-05-24 22:30:49  <NA> 2018-05-24 04:33:53     <NA>

Working through the time calculations it looks as if it is tied to the .hourAngle() function. Essentially, the value that gets evaluated within acos() is < -1 and therefore returns NA. Warnings are suppressed here and does not return any type of recognition to the user that this is something that there is trouble to calculate.

[VirginiaMorera]

I have found the same issue, only for calculating the night and endNight variables. Attached 100 observations sampled from the data that were causing problems. In the same dataset, with observations in similar dates and locations, the night and endNight times can be calculated, and no error or warning is produced, so I can't see what the problem is!

example_suncalc.csv

[natpac]

Hey,
Any news on this issue?

[Sofya2024]

Yeah I have encountered the same issue but only for night and nightEnd values at specific dates irrespective of the location points I use. the dates range from 25/05/2023 to 22/07/2023. I don't have this issue for later dates or for other time calculations like nauticalDawn. I would have hoped someone has found a solution by now

[natpac]

Up

[etgriffiths]

Hey, As far as I can tell this is still an issue. Between 2023-05-04 and 2023-08-08 for night and nightEnd and between 06-02 and 07-12 for nauticalDawn and nauticalDusk are all NA. Coordinates are lat = 56, lon = 7.5. ;(