In message 20150106115132.1f7eb84f@linux-pkou, Graham P Davis
writes
On Tue, 6 Jan 2015 11:08:11 +0000
John Hall wrote:

Though some notable winters (notably 1947) haven't begun till
considerably later than this. And looking at that winter's charts for
the first couple of weeks of January seemed to give no hint of what
was to come.

But December and January were both cold, about 1-1.5C below average.
December CET was 3.0C below 2014.

Oh yes, I'm certainly not trying to claim that we could see a repeat of
1947.

As for minimum temperatures, December
1946 was 19= coldest in the full CET record, whilst Jan '47 was 25=.

For January, that's largely because the real cold set in around the
23rd.

So 1946-7 was already a cold winter well before the February snowfall
occurred. The frosty December and January would have gotten into the
ground and, like the build-up to the 1962-3 winter, would have aided
the persistence of the lying snow.


I don't think much frost would still have been in the ground when the
main cold spell started, as there had been a mild spell from around the
8th to the 19th, with daily mean CETs of 3.3, 3.9, 2.9, 5.1, 5.9, 5.1,
6.7, 9.1, 10.0, 6.9, 6.3 and 2.8. Given that those are daily means
rather than daily maxima, that is pretty mild.
--
I'm not paid to implement the recognition of irony.
(Taken, with the author's permission, from a LiveJournal post)