Author Topic: Daily mean temperatures using hourly (or more frequent) readings  (Read 1410 times)

0 Members and 1 Guest are viewing this topic.

Offline rissole

  • Member
  • *
  • Posts: 4
Is there any Web site where you can find the daily mean temperatures for major weather stations that use the 24 hourly readings—or maybe even more than 24 readings for a day?  In this day and age, with all our computers, averaging the daily high and low just seems ludicrously primitive.  :eek:  For instance, where I live, if you just averaged the high and low yesterday, the average temperature would have been 58.  But if you used the hourly readings, the average was 55.7—a pretty big difference.

Anyway, surely there's some place online where we can find more accurate data for daily mean temperatures. :-) If anyone can offer any further info, I would greatly appreciate it.

Offline WeatherHost

  • Forecaster
  • *****
  • Posts: 3649
Re: Daily mean temperatures using hourly (or more frequent) readings
« Reply #1 on: January 01, 2019, 01:50:31 PM »
It's been discussed a few times and various examples have been posted, but overall, the NOAA/NWS method seems to work out over time.

2.3 degrees really isn't that much in the grand scheme of things.


Offline rissole

  • Member
  • *
  • Posts: 4
Re: Daily mean temperatures using hourly (or more frequent) readings
« Reply #2 on: January 01, 2019, 03:43:09 PM »
It's been discussed a few times and various examples have been posted, but overall, the NOAA/NWS method seems to work out over time.

2.3 degrees really isn't that much in the grand scheme of things.

But with modern computers, why stick with the old scheme all?  Why settle for less than outstanding accuracy?  I mean, how difficult could it really be to implement such a change?  It just doesn't make much sense to me. 

Anyway, thanks for your response, WeatherHost.  And since you didn't mention it, I assume there currently aren't any Web sites where you can easily find this information.  ;)

Offline mcrossley

  • Forecaster
  • *****
  • Posts: 1132
    • Wilmslow Astro
Re: Daily mean temperatures using hourly (or more frequent) readings
« Reply #3 on: January 01, 2019, 05:15:53 PM »
The other thing to bear in mind is if you change the method then you invalidate any comparison with historic records.
Mark

Offline WeatherHost

  • Forecaster
  • *****
  • Posts: 3649
Re: Daily mean temperatures using hourly (or more frequent) readings
« Reply #4 on: January 01, 2019, 05:41:16 PM »
  I mean, how difficult could it really be to implement such a change?

Reprogramming government computers?


Offline rissole

  • Member
  • *
  • Posts: 4
Re: Daily mean temperatures using hourly (or more frequent) readings
« Reply #5 on: January 02, 2019, 09:15:49 PM »
The other thing to bear in mind is if you change the method then you invalidate any comparison with historic records.

Well, the solution would be to offer both types of calculations.  That way, you could still use the older style to compare with the past if you wanted to. 

Offline rissole

  • Member
  • *
  • Posts: 4
Re: Daily mean temperatures using hourly (or more frequent) readings
« Reply #6 on: January 02, 2019, 09:24:29 PM »
Reprogramming government computers?

Well, compared with all the money that gets thrown into creating super-complex computer weather models these days, I hardly think that adding one extra field to compute the average of 24 temperature readings a day instead of just 2 would push the government into bankruptcy.  ;)

Actually, I don't really care whether the government does this or not.  Is there not some privately owned weather company that can offer this data for weather enthusiasts?  For instance, this Web site seems to do that very thing in order to calculate "degree days" for energy professionals:

https://www.degreedays.net/calculation