On Fri, 1 Jan 2016 10:27:25 -0800 (PST), Scott W wrote:

On Friday, 1 January 2016 17:30:19 UTC, Norman Lynagh wrote:
On Fri, 1 Jan 2016 08:17:01 -0800 (PST), Scott W wrote:




Because I'm 09-09 here in Wanstead December was technically not frost free. Temp fell to -0.2C at 0639. Obviously lots of 'what ifs' but I'm just keeping it consistent with the old stats at Greenwich which were 09-09.
The best argument I've seen for 09-09 is that most observers are not around at midnight to read the manual instruments.

Scott, If you are using the standard 09-09 reporting the minimum read at 0900 is
credited to the day on which it is measured. Therefore, the 24-hour minimum
measured at 0900 this morning (1st Jan) is the minimum for 1st Jan. The min of
-0.2 at 0639 this morning therefore 'belongs' to 1st Jan.

In contrast, the 24-hour max temp and the 24-hour rainfall measured at 0900 are
'thrown back' to the previous day. Therefore the 24-hour max temp mearured at
0900 today 'belongs' to 31st Dec, as does the 24-hour rainfall measured at 0900
today.

I'm not saying that this is necessarily the best system but it is the standard
system used at 'official' observing stations and it does enable comparison with
those stations.

Thanks, Norman. Am I correct in thinking that the minimum (in this case) must occur before 2359z on 31st? I'd always counted the min (as the max and rainfall) in each 24hr period. I suppose over the course of a month it makes little difference to the overall result whether you use 00-00 or 09-09 - it is merely a technicality?

Scott, In standard 09-09z reporting midnight (or 2359) has no relevance. There
are only 3 numbers:

1) The min temp in the 09-09z period
2) The max temp in the 09-09z period
3 The rainfall total in the 09-09z period.

Item 1) is assigned to the date in which the end of the 09-09z period falls.
Items 2) and 3) are assigned to the date in which the start of the 09-09z period
falls. They are said to be 'thrown back' to the previous day.

Therefore, the minimum temperature for the 24-hour period ending at 09z on 1st
January is assigned to 1st January. The maximum temperature for the 24-hour
period ending at 09z on 1st Jan is assigned to 31st Dec as is the 24-hour
rainfall measured at 09z on 1st Jan.

These rules apply irrespective of where in the 24-hour period the minimum and
maximum temperatures apply. This can thow up occasional anomalies. For example,
it is perfectly possible for the minimum temperature assigned to 1st Januray to
actually occur during 31st December. Similarly, it is perfectly possible for the
maximum temperature assigned to 31st December to actually occur in January.

Hope this makes sense :-)

--
Norman Lynagh
Tideswell, Derbyshire
303m a.s.l.
http://peakdistrictweather.org