The Central England Temperature (CET) record is a meteorological dataset originally published by Professor Gordon Manley in 1953 and subsequently extended and updated in 1974, following many decades of painstaking work. It is now maintained by the Met Office and extends back to 1659 in its monthly and to 1772 in its daily form.
Despite the last record couple of cold days in Central England, I reckon that the average number of degree days between the winter solstice and the vernal equinox was exceeded on the 14th of March this year (fig 1).
This normally would make the first day of spring around a week early, but because springs have been occurring earlier on average (20 days earlier than in 1772 and now around the 11th of March each year), 2018 was around three days late this year, it was also the latest spring to have occurred since 2013.
I personally think that using degree days like this is a crude but effective way of coming up with a date for the start of spring, let me know if you have a better idea.
Yesterday, the 18th of March 2018 was the coldest for that particular day in March since 1853. That’s according to the latest provisional Central England mean temperatures released from the Met Office (fig 1). The maximum for Sunday was 0.6°C and the minimum -2.2°C, and the mean of -0.8°C, was almost 6.6°C lower than the 1961-1990 long-term average of 5.8°C.
Only four other years have reported negative mean temperatures for the 18th of March (fig 2 & 3) in the daily mean CET series that started in 1772, they were 1900, 1853, 1814 and 1812.
Don’t forget that you can keep an eye on the latest daily CET values here
Looking at the observations for yesterday and the reported 06 to 18 UTC maximum temperature (fig 4), many stations across central England and Wales had sub-zero maximums, in fact at a number of stations, the temperature has remained below zero the whole weekend, which just two days from the vernal equinox is highly unusual.
I think the thermograph for the last few days from Rothamsted in Hertfordshire gives a good account of how quickly the cold snap set in last Friday, and just how cold the last 48 hours have been across the country (fig 5).
I was just doing some research into severe winters of the past in Central England and produced this intercomparison between this last winter (2017-18) and the snowy winter of 1946-47 (fig 1). The 1946-47 winter didn’t really get going till mid-January of 1947, and the cold of this last winter was only really a match for it either before mid-January or in the last couple of days in February.
An even more interesting intercomparison is between the severe winter of 1962-63 and the snowy winter of 1946-47 (fig 2). The winter of 1962-63 like 1946-47 was also a bit fitful in starting, and only really got into its stride just before Christmas 1962. After that there was no stopping it, and it more or less reigned supreme over 1946-47 until the end of January. The year 1947 produced the coldest February in the entire 359 years of the CET series and 1962-63 didn’t stand much of a chance against in the intercomparison, and 1947 edged out 1963 right through to the end of the meteorological winter.
Below are the two winters in a lot more detail (fig 3 & 4) and you can see that although 1962-63 was very severe from around the 23rd of December 1962 to the 25th of January 1963, 1947 went hyper from around the 22nd of January 1947 right through to the end of the winter. The blue diamonds are daily extreme records that these winters still hold.
Collectively the last three months of December, January and February which constitute the so-called ‘meteorological’ winter were +0.24°C warmer than the 1961-1990 long-term average, but if you had use the 1981-2010 average instead, they would have been -0.24°C colder, I suppose AGW deniers might prefer the latter. Winter 2017-18 statistically was the coldest winter in five years (fig 2).
There were two main cold spells in the winter, one in the second week of December, which looked like it might deliver but the Atlantic ocean had other ideas, and of course the last week in February, which was probably helped along by the SSW earlier in the month. There were a number of mild spells, one either side of Christmas, one in mid January and another in mid-February. Only one daily record extreme was broken during the winter of 2017-18 in central England, that of coldest 28th of February since 1785 (fig 1).
Don’t forget you can keep up to date with both the latest daily and monthly temperatures in Central England by using the interactive graph on the Meteograph website.
After reporting yesterday that we’ve just experienced the coldest 28th of February since 1785 in Central England, it now transpires that the first of March provisionally has the lowest mean temperature for that day since 1785 (fig 1), and additionally was the coldest day (lowest maximum of -0.8°C) for any March day since 1878.
The last seven days have been cold, but they still only rank 6th in the all time coldest weeks (23rd February – 1st March) since 1772 (fig 2). So if you think the mean of -0.5°C for the last seven days has been cold, the -2.56°C of 1947 was a magnitude colder. It’s very rare these days to see any new extreme minimum records broken, but setting three records in two days isn’t bad going!
Don’t forget you can keep up to date with the latest daily temperatures in Central England by using the interactive graph on my Meteograph website.
I did think that yesterday’s cold was exceptional for late February, but it wasn’t until I download the finalised daily CET data from the Met Office this morning that I realised it had in fact been the coldest 28th of February since 1785 in Central England – that’s 233 years ago.
The mean temperature for yesterday was just -3.6°C, which was just a smidgen short of the -3.8°C lowest for that day in 1785, and -8.1°C lower than the 1961-1990 long-term average.
The mean daily temperature series started in 1772, but if you compare the day maximum for yesterday of -1.3°C that was the coldest since at least 1878 when the more detailed max/min series began.
I’ve just noticed that there seems to be a disproportionate number of cold last weeks of February and first week in March in the daily CET records (fig 1). Only someone as crazy as me would be writing code on a lovely sunny Sunday with Mediterranean blue skies outside to discover this trivial and useless nugget of information regarding temperatures in central England since the millennium. Statistically six out of the last 245 years in the top 30 is probably no more or less than what you would expect, but in recent years there have been so few examples of any month that was colder than average (apart from 2010) that it came as a bit of a surprise to me. These colder inter month periods have probably gone unnoticed because you often view climate periods on a calendar month basis, or at least I do.
Anyway the original reason why I started doing this little spot of research was to find out just what were the coldest end of February start of March periods since 1772. I could then gauge statistically just how cold the next couple of weeks of cold had been in the series.
The upshot of that work confirmed what I already thought might be the result, and that is that this coming cold spell of very cold weather is going to have been pretty spectacular if it’s to equal or surpass the exceptional same period in 1947 (fig 2).
According to results that I’ve generated, the mean temperature for the period 21st February to the 7th of March was -2.2°C, that ‘s -6.6°C below the long-term average (1961-1990). That’s a clear 1.4°C colder than 1786 the next coldest such like period. Even if we do have a couple of ice days in this coming week in central England I can’t see how the mean temperature can get anywhere near that of 1947, although it might push the means for 2005 and 2006.
This period has shown only a very slight warming over the years of just 0.006°/decade, which equates to a small warming of 0.15°C in the last 245 years (fig 3). Perhaps this is the last vestige of how the temperatures were during the late 18th and 19th centuries in central England, now that’s another crazy thought.
I plan to eventually get round to display the following data types graphically in some form or other:
Daily CET (I’ve included data back to 1954 for starters)
As well as writing the code to visualise the graphs in a web page, I’m also in the process of adding extra functionality to export processed climate and observational data to my ISP to the many desktop Windows apps that I use. If I had any web server-side skills this could be done using something like PHP, but since I don’t, a humble CSV file will have to do!
Here’s a look at what the Highmaps component looks like (fig 2). I can never understand why the Met Office can’t use something similar as the basis for their warnings system using ESRI GIS.
I’ll warn you now that a lot of the pages are work-in-progress and many are still broken and require reworking. I’m getting up to speed so hopefully I’ll get round to fixing most of them in the next few months. I’ll also have to try to remember to upload the latest climate data on a regular basis, but if I do happen to forget, just drop me an email to remind me.
The 13/14th of February is statistically the coldest night in Central England, with a 40% chance of an air frost occurring (fig 2), and last night it lived up to its statistical reputation (fig 1), with a widespread sharp frost across a large part of the country. That should push the frequency up for that day by maybe 1/140 of a percent in the last 140 year, or does it?
I also noticed a jump of +6.3° in the air temperature at Exeter between 03 and 04 UTC this morning as the cloud rolled in (fig 3), temperatures are on a bit of a roller coaster ride at the moment.
St Valentines day, the 13/14th of February is the coldest night of the year in Central England, this is based on the lowest mean daily minimum temperature record that started in 1878, the mean for that day is just 0.67°C, with a 40% chance of an air frost, and a 4.3% of an ice day. The lowest mean daily maximum on the other hand occurs around four weeks earlier on the 17th of January with a temperature of 6.1°C (fig 1).