The Snowiest Winters

[John Hall]

Inspired by a post from Scott, but I felt this veered off-topic from the original thread so I've started a new one.

The Bonacina and O'Hare snowfall index rates the following eight winters as "very snowy" in some 140 years of the index stretching back to 1875-6: 1875-76, 1878-79, 1885-86, 1916-17, 1946-47, 1962-63, 1978-79 and 2009-10. With three cases within the first ten years, the coldness and snowiness of Victorian winters shows up very clearly. After that, if you treat 1962-3 as a freak interloper, there seems to be a remarkably regular periodicity of about 31 years, with no obvious sign of our warming climate as yet causing a tailing off. Based on this, we can expect our next very snowy winter around 2040. :) Sadly, if I survive I will be in my nineties by then, so I doubt whether I'll get much enjoyment from it!

Though there's little sign of a reduction in very snowy winters, snowy and even average winters have declined in recent decades, with winters with little snow becoming far more common. What that means is that a so-called "average" winter is actually a good deal snowier than the modern-day average.

http://www.neforum2.co.uk/ferryhillweather/bonacina.html

[Richard Dixon]

John - I took a look at that data just to understand how our "coldest months" have changed.

Data below plots the lowest monthly CET by winter and the orange line is the decadal average of this value. You can see how it's risen since 1990 more than anything - but for the climate change sceptics - you can also see how 1910s-1920s we're that dissimilar to 1990 onwards.

One thing I noticed that rung so true with my childhood is the number of years which had a monthly CET <2c compared to, say, kids growing up today, and also how the frequency of >4.5c has risen since 1990.

Richard

[xmetman]

Out of interest to find the snowiest winter a number of years ago I developed a Central England Snowfall application, that combined the daily CET with the daily UKP and estimated daily snow depths. 

This is a table of the results from 1931 (depths in cm):

You obviously need to use a "guessing algorithm" that gives you the chances of snow on any particular day, depending on the minimum and maximum CET.

I think I assigned a maximum of less than 1°C a 100% chance that any precipitation would be of snow.

I'm not saying it's particularly accurate, but it does consistently apply the same algorithm to each day of each year.

Interestingly it did report 3 cms of snow last month on the 30th of November.

One thing that I did discover was an almost 60% reduction in the depth of annual accumulated snow since 1931.

This reduction compares quite well with the reduction in snowfall accumulations at Weston Park Museum in Sheffield since 1950.

There are two years 1988-89 and 2013-24 when it didn't generate any accumulations.

You can read my full report in more detail here.

Since I first published the article in May, a grand total of 25 people have viewed it, with your help maybe it'll go viral!

[xmetman]

And just for completeness here is my best effort at the snowiest winter of them all...

[John Hall]

Very interesting. There's such a huge amount of scatter, that too much reliance should probably not be placed on the precise value of that trend line, but for recent decades it seems to bear out one's subjective experience.

[John Hall]

"Out of interest to find the snowiest winter a number of years ago I developed a Central England Snowfall application, that combined the daily CET with the daily UKP and estimated daily snow depths."

That's very interesting. I'm surprised by a few things, though. One is how 1980-1 could rank so high and 1981-2 rank so low. The other is how 1962-3 could rank only 7th (it looks at first glance like it ranks 8th, but that's because the heading row has been assigned number 1 spot). Here in Surrey, I can remember far enough back to know that it was well ahead of 1978-9 here. In fact I suspect that it would be second only to 1946-7, and possibly even ahead of it. I realise that in the Midlands the comparison would be different, with more snow in 1946-7 and probably less in 1962-3.

One factor to bear in mind is that 10 mm of precipitation might give a snow depth of say 10 cm if the temperature was -1C but only half that if the temperature was 1C, being the difference between dry, powdery snow and wet snow.

[Richard Dixon]

On Thursday, 21 December 2017 16:49:48 UTC, John Hall wrote:

Very interesting. There's such a huge amount of scatter, that too much reliance should probably not be placed on the precise value of that trend line, but for recent decades it seems to bear out one's subjective experience.

Actually this isn't a trend line in the traditional sense: I'm not typically a fan of them. It's an average of each decade centred on middle of the decade over which the average is calculated just to try and make some sort of sense of the data.

Richard

[George in Edinburgh]

Epping

http://www.swanstonweather.co.uk/EppingWeatherPages/snow.htm

http://www.swanstonweather.co.uk/EppingWeatherPages/TableofSnowdata.htm

George in Swanston, Edinburgh but once of Epping (and Sunderland and Newport, Mon for that matter)

www.swanstonweather.co.uk

[Dave C]

I think I might not be too excited about waiting till 2040 apart from the same reason as you John. Being in Essex pretty much all of the last 65 years or so 2009-2010 couldn't hold a candle to the other "snowy" winters although it certainly is the stand out in the last 25 years or so. There were hardly any deep drifts and that is certainly something we don't get these days. Transient blizzards or here in particular, frequent heavy snow showers off the North Sea causing blowing snow piling up in the kerbs on a bone dry road. I always go on about this and people think I imagine it (Ron??) but the times I would arrive late at work in London with the quizzical look "What snow?". Not many East Anglian posters on these forums but I know Darren P has commented on this re: N.Kent coast but he doesn't go back as far as me.