DST start/stop in WeeWX
42 views
Skip to first unread message
Joel Bion
Mar 11, 2024, 9:37:48 AMMar 11
to weewx-development
In my country, we changed to DST at 2AM this past Sunday early morning.
I’m assuming that in the database, the timestamps are all UTC.
It’s the display of the information that I am wondering about, because the graphs show local times. I live in the Pacific Time Zone in North America. Right now, my server displays on my WeeWX home page that I am in UTC -7, which is correct. Until 2am my time Sunday, I was in UTC-8.
So, at 4AM Sunday (yes I was awake), my graph for the day showed 4 hours of data in my day. Yet, because of the time change, there were only 3 realtime hours in the day. Again, 4 hours of data was displayed in that day. Yet, that day only had 3 hours.
My guess is, when I am in “standard time” and ask weewx to graph me historic data it assumes my decade+ of data is all in standard time, ignoring the date and basing off a chosen date-independent “start of epoch?”, or some such? And then when I am in daylight savings time, do something similar during Summer time? so, if I was in a cloudburst in the last hour of the day before time-change day, during daylight savings time, that data is shown in Sunday’s graph, during standard time it’s shown in Saturday’s?
And what about “tallies” in the tables, like “rain each day of the month”? How do they choose what day events the last hour of the day before a time change are bucketed to?
I’m assuming that in the database, the timestamps are all UTC.
It’s the display of the information that I am wondering about, because the graphs show local times. I live in the Pacific Time Zone in North America. Right now, my server displays on my WeeWX home page that I am in UTC -7, which is correct. Until 2am my time Sunday, I was in UTC-8.
So, at 4AM Sunday (yes I was awake), my graph for the day showed 4 hours of data in my day. Yet, because of the time change, there were only 3 realtime hours in the day. Again, 4 hours of data was displayed in that day. Yet, that day only had 3 hours.
My guess is, when I am in “standard time” and ask weewx to graph me historic data it assumes my decade+ of data is all in standard time, ignoring the date and basing off a chosen date-independent “start of epoch?”, or some such? And then when I am in daylight savings time, do something similar during Summer time? so, if I was in a cloudburst in the last hour of the day before time-change day, during daylight savings time, that data is shown in Sunday’s graph, during standard time it’s shown in Saturday’s?
And what about “tallies” in the tables, like “rain each day of the month”? How do they choose what day events the last hour of the day before a time change are bucketed to?
Tom Keffer
Mar 11, 2024, 10:50:56 AMMar 11
to Joel Bion, weewx-development
The plot is always time_length long. For example, for a day plot, this defaults to 27 hours. The DST transition only affects the labeling.
Take this example from yesterday:
The total width is 27 realtime hours --- that doesn't change with DST. However, the advent of DST causes the major increments to the right of the DST transition to shift a little bit to the left --- if you will, we reached them quicker thanks to the "spring forward."
The density of data points across the graph remains the same --- one every 5 minutes --- but the labeling shifts around a bit.
As for the tallies: the March NOAA report starts with 1 March 2024 00:00, exclusive, and ends with 1 April 2024 00:00, inclusive. Those are local times. Because of DST, the table actually contains 1 hour less data than if DST had not happened.
Hope that helps.
--
You received this message because you are subscribed to the Google Groups "weewx-development" group.
To unsubscribe from this group and stop receiving emails from it, send an email to weewx-developm...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-development/35200E9A-9AB7-4B33-985F-AB80CB35D890%40gmail.com.
Tom Keffer
Mar 11, 2024, 1:24:03 PMMar 11
to Joel Bion, weewx-development
Reply all
Reply to author
Forward
0 new messages