Great piece of research Brian!
I have compared the two 30 year periods for each temperature category with t-tests:
-10 C or lower. -9 to -5 C. -4 to 0 C
1961-90. 667444. 287660. 309076
1991-2020. 625163. 288932. 316544
-10 or lower was a highly significant decline at the 0.1% level
-9 to -5 was a non significant increase.
-4 to 0C was a highly significant increase at the 0.1% level
It is surprising that the middle category was non-significant.
My guess at an explanation would be it is something to do with the reduction in arctic sea ice extent. Now at the beginning and end of the Arctic sea ice season, there are big areas which are ice-free compared to 1961-90 which would reduce cooling of t850s. Redoing the analyses on a monthly basis might show this but that would be a big undertaking.
As to the impact on UK cold spells in winter, yes, this reduction in the -10 C t850 cold pool will lessen to some extent the chances of cold spells. BUT it is clear from historical data that cold UK winters come in groups (1980s, 2008/9 to 2010/11) and this is related to the strength of the westerlies and the increased likelihood of retrogressive synoptics bringing easterlies when westerlies are weaker and the jet stream slows and buckles. This is what brings very cold air to the UK in winter. And happily, we seem to be having an example of this phenomenon currently.
Originally Posted by: Iceman