Standard time ranges for max and min temps

[h2005__uk__]

Hi

Are there standard time ranges for maximum and minimum temperatures?

I believe the data through Ogimet takes the max as being the max between 6am and 6pm, with the min being between 6pm and 6am (the following day). However I'd have thought that could miss some minimums in the winter when often the minimum is just before sunrise at nearer 8am?

Would 9am - 9pm for max and 9pm - 9am for min be better, or even 6am - 9pm for max and 6pm - 9am for min?

Edited December 31, 2017 by h2005__uk__

[NorthYorksWeather]

I use 9am to 9am as Im fairly sure thats what the Met Office use for their data.

[johnholmes]

4 hours ago, vizzy2004 said:

I use 9am to 9am as Im fairly sure thats what the Met Office use for their data.

correct

[reef]

The official method is 09:00 to 09:00 GMT (or 10:00 to 10:00 BST), with the maximum temperature in that time period being for the previous day (so a max temp in that 24 hours on Monday at 7am would go down as Sunday's maximum) and then the minimum as the following day (a min temp in that 24 hours at 11am on Monday would be Tuesday's value). For rainfall, all that falls between 09:00 and 09:00 GMT is for the previous day (so if 5mm of rain falls on Monday at 6am that will be included in Sunday's figure).

Personally I prefer 00:00 to 00:00 (especially with an AWS like a Davis VP2) but keep up two records using both methods these days.

[johnholmes]

I think you will find the hours are 0900-0900 clock time, not as reef has suggested?

[h2005__uk__]

2 hours ago, reef said:

The official method is 09:00 to 09:00 GMT (or 10:00 to 10:00 BST), with the maximum temperature in that time period being for the previous day (so a max temp in that 24 hours on Monday at 7am would go down as Sunday's maximum) and then the minimum as the following day (a min temp in that 24 hours at 11am on Monday would be Tuesday's value). For rainfall, all that falls between 09:00 and 09:00 GMT is for the previous day (so if 5mm of rain falls on Monday at 6am that will be included in Sunday's figure).

Personally I prefer 00:00 to 00:00 (especially with an AWS like a Davis VP2) but keep up two records using both methods these days.

The Met Office don't seem to use the 'official' method for the 'day's extremes' pages on their website, as on there, they define the time ranges as:

Highest maximum temperature - (0900 to 2100 on the date shown); Lowest minimum temperature - (2100 on the previous day to 0900 on the date shown)

I've been playing around with some of my weather data recently and have been using these ranges, but the problem is that on the rare occasion a maximum occurs during the night, it gets missed. On the other hand, using 0000-0000 or 0900-0900 seems to have its drawbacks too.

[h2005__uk__]

PS. I suppose it depends whether you want the maximum of the day and the minimum of the night, or the max/mins of a day.

[reef]

On ‎31‎/‎12‎/‎2017 at 17:25, johnholmes said:

I think you will find the hours are 0900-0900 clock time, not as reef has suggested?

Are you sure John? I was always under the impression it was 0900-0900 UTC, which in British summer time is an hour later?

If it transferred when the clocks do then you'd essentially lose an hour in spring and gain one in Autumn?

[johnholmes]

1 hour ago, reef said:

Are you sure John? I was always under the impression it was 0900-0900 UTC, which in British summer time is an hour later?

If it transferred when the clocks do then you'd essentially lose an hour in spring and gain one in Autumn?

You might be right, my memory is not what it was, best you check with UK Met perhaps?

[johnholmes]

This link seems to be what I could not find before I posted suggesting Z time. It does seem I am correct, which is a relief.

para 2.4 is the relevant para

http://artefacts.ceda.ac.uk/badc_datadocs/ukmo-midas/ukmo_guide.html#2.4

[iapennell]

The Met Office Standard is indeed 09.00 to 09.00 (GMT) for recording temperatures, with the maximum being attributed to the first day and the minimum temperature attributed to the next day. This is how I keep temperature records and for the most part it accurately reflects the minimum and maximum temperatures as you would expect. 

I used to help take weather records at Lancaster University Weather Station (where I did my degree many years ago) and rainfall, run of wind, etc was always logged at 09.00 GMT and the accumulated totals attributed to the previous day. There was also an automatic rain-guage and this was set to record midnight to midnight (GMT), now that I have an Automatic Weather Station I record rainfall this way. 

I record the wind (average speed and direction) overnight and mid-morning to mid-evening so that I can gauge what sort of conditions we have overnight and during the day. I have the cut off at 21.00 (GMT) so that the average winds for the day correspond closer to the midnight to midnight values. However, if (say) strong winds occur in the late evening I would log these for the day in question and apply a small adjustment for the mean winds for that day. I am certain that this would probably not meet with official approval, but it enables one to see the differences of wind speed and directions between night and day, which has an all-important bearing on the occurrence of dew or frost and (in summer) daytime heat. I could then calculate midnight to midnight mean wind speeds and directions but, hey, that's quite a bit of a faff when you could just average the night and day-evening means which, most of the time, are not going to be far wrong!

I do believe, however that many AWSs that provide weather data for the Met Office log winds as midnight to midnight and this is quite acceptable to the Met. 

Ian Pennell