Jump to content

1916-17; The severe winter of the Great War

Recommended Posts

looks like 1917 was also the 3rd coldest year of the 20th C only being pipped by 1963 & 1919

1963 = 8.52c

1919 = 8.53c

1917 = 8.55c 

Edited by cheeky_monkey
  • Like 2
  • Thanks 1
Link to post
Share on other sites

Very good analysis -- interesting that this winter came right after the record mild January of 1916. The same sequence occurred in eastern North America. From my Toronto study (in the climate change area although it probably should be in this section), January 1916 after UHI adjustment ranked 12th warmest out of 181 (1840-2020) while February 1917 ranked tied 13th coldest. 

As you said, 1917 and 1929 were the only two winters with much "form" in that long stretch between 1895 and 1940. In the case of eastern North America, 1929 was not particularly cold, more like an average to cold winter, and there were a few others in that stretch which had similar definition (1895 had extreme cold at exactly the same time, early Feb; Jan 1940 was quite a cold month). Winters of 1903-04, 1911-12, 1919-20 and 1933-34 were all quite cold. Winter 1935-36 is regarded as one of the most severe further west (midwest, upper plains states) but marginal for inclusion among cold winters at Toronto or NYC.  

That interval 1896-1939 still marks a bit of a pause with more frequent cases both before 1895 and after 1940. From this we can learn that an interval of natural variation caused warming peaked between 1895 and 1940, the most obvious peak was 1919 to 1922. Some concerns were raised then about melting in Greenland, but these abated when things turned a lot colder in the latter half of the 1930s and 1940s. 

I found in the Toronto and NYC analyses that there is something like a 2:1 ratio of correlated trends to anti-correlated trends on the two sides of the Atlantic (meaning, it's about twice as likely for the two sides to be in sync as running opposite, when we look at the more extreme thirds of the data -- and also the middle thirds tend to occur together as well). This is probably because with 70-80 deg of longitude separating the two climate zones, a five-wave structure would place both in the same ridge or trough positions (360/5=72) and five waves are common with the mild variant having ridges around 0, 72E, 144E, 144W and 72W. If the hemispheric wave pattern reduces to four, the 90 degree separations can still be very similar in outcomes (let's say you had ridges near 80W and 10E). A three wave pattern would likely place the two zones out of sync. This may be what occurred in 1933-34 which was very cold in eastern North America and very mild in western Europe in a period of zonal flow indices peaking near maximum levels. It appeared to repeat in summer 1936 which was blistering hot in east/central North America and cool/wet in western Europe. 

I think we could make significant gains in LRF if we understood cause and effect of these wave length patterns (and plenty of research has been done and continues to be done on that subject). Does the pattern ever compress to six or seven? I think that would begin to look like a blocking pattern. I think it has been established that three is the minimum (two would be a very nasty zonal mildfest just about everywhere, maybe it tried to form in 2015-16). 

Another thing I find interesting is that almost every solar maximum since the end of the colder climate period (arbitrarily, let's say 1895) has seen a cold winter. The cases in the CET would be 1917, 1929, 1940, 1947, 1956, 1969 (mostly Feb there), 1979, 1991 ... then the 1999-2001 peak didn't do as well, but the 2012-13 peak came at the end of a run of cold winters of which 2009-10 and 2010-11 were obviously most significant. In fact, with the exception of 1963, almost all the modern cold winters have come fairly close to a sunspot maximum except those two. Before 1895, there were more frequent cold winters anyway so finding some near sunspot peaks is not that extraordinary, they probably clustered slightly more into the quiet sun years (the Dalton was mostly a quiet sun interval). 

The orthodox theory of climate change says we are heading in a steady upward climb but the experience of 1895 to 1940 says hold on, maybe there is some hope for the climate to shift somewhat back to colder (although the human AGW signal is obviously going to be part of the total picture). 

In fact we have had more significant cold winters (CET) in the 44 winter span 1977 to 2020 than they saw from 1896 to 1939 (2). I didn't specify a total for 1977 to 2020 because there were a few debatable ones or brief duration severe winter conditions, however 1896 to 1939 didn't even produce debatable ones beyond 1917 and 1929. If one took 1895 = 1963 and 1940 = 2010 (68, 70 yrs) then the more frequent cold from 2010 to 2013 resembles the early to mid 1940s, and the pulse of milder and a few colder winters to come would be expected to resemble the 1950s and 1960s. Instead of the brutal 1947 we had the brief 2018 (similar relative timing and a similar although shortened event with the very heavy sea effect snowfalls). This would imply we are now around 1949-1950, not that thrilling a prospect for coldies as only 1956 then 1963 would be cold winters anywhere near the event horizon. Just speculation on my part though. With this longer quiet sun interval, I think the odds are mounting in our favour to have a more rapid appearance of the next "big one" and who knows, it could be this winter, or next ... 

The great thing about 1895 to 1940 is, we've been here before and it wasn't the end of winter but it did mark a general downturn in frequency of cold winters from which we never fully recovered at any point (closest being mid 1980s and 2009-13). 

Edited by Roger J Smith
  • Like 1
  • Thanks 2
Link to post
Share on other sites

Unusual to have such a cold winter considering what was happening in the ENSO region

Nov 1916                                                   Dec 1916


Jan 1917                                                     Feb 1917


That is one strong La Nina event taking place there. Don't know whether it achieved Super Nina status or not but it looks close to it

There was at least some colder than average anomalies in the NE Pacific


A nice tripole in the N Atlantic too just heading into the winter


A very chilly Indian Ocean too


Don't know if the cold Indian Ocean played its part in the winter but with that tiny warm blob near Indonesia it could be the tell tale sign of a -IOD too.

  • Thanks 1
Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Create New...