All this interesting stuff of course arises from the fact that we are trying to
discretize a continuous variable.

Will
--

wrote in message
...
On 18 Jan, 20:36, Mike Tullett
wrote:
On Fri, 18 Jan 2008 19:39:51 +0000, Trevor Harley wrote in
news:2008011819395116807-taharley@dundeeacuk

I really hate it when the temperature soars between 8 and 9 am. Quite
often in winter the maximum is 8.59 a.m. December 21st here was very
cold - a maximum of -2.4C during the day, dropping to -6.8C around 2
am, then climbing rapidly to -1.4C at 8.59, thus spoiling my record
lowest maximum. And there have been several occasions where an
otherwise "frost day" has been ruined by rising temperatures around
dawn (which is 8.45 a.m. here).

The most notable occasion that happened here was in the early 80s. The
daytime max was about -11C but by 0900 next day the temp was up to +3C.

--
Mike Tullett - Coleraine 55.13°N 6.69°W posted 18/01/2008 20:36:23 GMT

I can remember in the cold spell at the end of January 1972, the
morning of 31 Jan was bitterly cold (my min that morning, in
Warwickshire, was -12.2°C, around 8.30 am). At around a third of
climatological sites in the Midlands, the lowest temperature occurred
a little after the (nominal) 0900 UTC observation, so the minimum
thermometer reading was, following standard practice, credited to the
following morning - i.e. 1 February. However, milder air spread in
quickly during the day on 31 January, and February was a mostly mild
month, resulting in most places seeing their official lowest February
(and annual) minimum occurring ... erm, in January.

I'd say it's about once every 2-3 years here when the nominal 24 hour
09-09h max and min actually occur exactly 48 hours apart. How can this
happen?

Day 1 - temp at 0900 0°C
Day 2 - temp rising throughout, temp 5°C at 0900
Day 3 - temp rising throughout, now 10°C at 0900

For Day 2, the 09-09h min will be 0°C (temp at 0900 Day 1) while the
max will be 10°C (temp at 0900 Day 3). The 24 hour extremes are
actually exactly 48 hours apart.

Of course, if the temp is falling throughout then min = max = 0.0
degC. But that's another quirk of the system.


Stephen Burt
Stratfield Mortimer, Berkshire