-3

What percent of the world's population is in a timezone whose UTC offset (in hours) is not a multiple of 1?

Notes:

  • To answer this, we need to know what % of the world's population live in each time zone
  • A precise answer may be difficult because of changes to offsets throughout the year (e.g. with daylight savings and other adjustments)

Example

Suppose we compare Melbourne's UTC offset to Adelaide's. Melbourne is UTC + 10 hours, whereas Adelaide is UTC + 9.5 hours.

In pseudocode / ruby:

time = Time.current.beginning_of_hour

time.in_time_zone("Melbourne")
# => Sun, 20 Sep 2020 15:00:00 AEST +10:00

time.in_time_zone("Adelaide")
# => Sun, 20 Sep 2020 14:30:00 ACST +09:30

So we can state that (at this time of year), Melbourne's UTC offset (10) is a multiple of 1, whereas Adelaide's (9.5) is not.

Known so far

  • A cursory look at the wikipedia page on UTC offsets suggests ~11 out of 38 timezones appear to end in 30 or 45 (i.e. they don't end in 00)
  • This resource is very interesting, but doesn't appear to offer the raw data
    • It suggests there are 39 time zones, however, it's from 2015, so perhaps something changed since
  • This quora question offers some useful contextual graphics
  • 1
    While GIS could certainly be used to solve this trivia question, there doesn't seem to be a GIS component to the question, as asked. It might be more appropriate in Puzzling (though it doesn't seem a good fit for their style). If you want to solve this yourself, you'll need to choose a GIS platform, locate current time zone and population data, then perform analysis. – Vince Sep 20 at 14:17
  • @Vince this information is being used to determine the priority of a bug fix in some open source software. Right now users whose time zones aren’t multiples of 1 have degraded performance and even crashes in some cases. We need to know what % of the user base this is affecting – stevec Sep 20 at 15:15
  • 1
    You can get a shapefile of timezones (github.com/evansiroky/timezone-boundary-builder/releases) and a raster of population (various sources, eg sedac.ciesin.columbia.edu/data/set/gpw-v3-population-density/… for 1 degree grid, higher resolutions available elsewhere). Aggregate raster by polygon and identify the half-hour TZs and sum. – Spacedman Sep 20 at 16:09
  • 2
    It's still off-topic without an attempt, but making an Edit so that it's not worded like a Puzzling question is probably a first step in making this on-topic. – Vince Sep 20 at 16:10
  • @Spacedman thank you. I managed to solve it. The answer is 21%. Workings here – stevec Sep 20 at 16:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.