On Thursday, February 27, 2020 at 8:20:16 PM UTC, John Hall wrote:
In message , Len
writes
If Storm Jorge gives 25 mm of rain over the next two days then my Feb
record will be beaten.
However my record is 205.1 mm in Feb 1990.
This was not a leap year.

It was like 1st March 2018 during the Beast from the East being a new
record cold day for March.
It would n't have been if it was a leap year.

How can one reconcile such things?

Len
Wembury

I don't think you can. Something similar is when an exceptional spell
doesn't show up in the records as prominently as it merits because it
was partly in one calendar month and partly in the next. A classic
example is the hottest spell of the summer of 1976, which lasted from
around mid-June to mid-July. If that had happened two weeks later, I
suspect it would have been easily the hottest calendar month in the
British instrumental record.
--
John Hall
"If you haven't got anything nice to say about anybody, come
sit next to me."
Alice Roosevelt Longworth (1884-1980)

What you can do is say something like 'The latest date for an ice day is 1st March or 29th Feb (just an example). The day of the year is the same in both cases even though the month is different..

In fact in my weather database I often use a macro, which calls the DOY function, to return the latest dates for things like that. Likewise I can easily identify the hottest '4 week period' Helps overcome the fact that most weather reports are based on months and can be mentioned in the monthly report.