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. The monthly mean surface air temperatures, for the Midlands region of England, are given in degrees Celsius from the year 1659 to the present.

This record represents the longest series of monthly temperature observations in existence. It is a valuable dataset for meteorologists and climate scientists. It is monthly from 1659, and a daily version has been produced from 1772. The monthly means from November 1722 onwards are given to a precision of 0.1 °C. The earliest years of the series, from 1659 to October 1722 inclusive, for the most part only have monthly means given to the nearest degree or half a degree, though there is a small ‘window’ of 0.1 degree precision from 1699 to 1706 inclusive. This reflects the number, accuracy, reliability and geographical spread of the temperature records that were available for the years in question (intro courtesy of Wikipedia).

It’s best to navigate all the articles by moving your mouse cursor over the ‘Look Back’ tabbed page and a drop down menu of items will appear for you to select from.

Late April cold spells and the Easter Snowstorm of 1908

Figure 1 – Lymington High Street – April 25th 1908 (courtesy of lymington.org)

Easter in 1908 fell late, so the snow that fell over much of southern England must have come as a big surprise on the Easter Sunday on the 19th of April (fig 1). The following week was intensely cold for late April, and there were periods of heavy snow across much of southern England. In an article in the Met Mag of May 1908, Fred J Brodie said this about the snow at Oxford:

The conditions at Oxford are interesting in a special degree on account of the length of the meteorological records at the Radcliffe Observatory which run from 1853. The depth of snow there was 17 inches, and the only instance of a greater amount being recorded at any time of year was on February 13th and 14th, 1888, when 24 inches of undrifted snow was measured.

I love the comment that Fred went onto make a few lines further on…

The practice of comparing, for the purpose of record making, observations made in two different localities is not to
be commended…

He of course is completely right in what he says, but he must be spinning in his grave these days, on the goings on in the early 21st century with extreme temperature records I would have thought, because no one, and that includes myself seems to give a hoot these days about comparing extremes from weather stations without knowing thinking much about their actual location. You can find an article about the events of April 1908 on the Weather Outlook forum, which includes details of snow depths recorded at the time, plus a lot of other information and photographs about the blizzard. The Weather Magazine of December 1981 also had an article about April 1908 in which it linked it to the April of 1981 and said:

The marked similarity of the graphs for 1908 and 1981, especially in the second half of each, is confirmed by a correlation coefficient of 0.93 for the last 15 days of the month. For the full month the correlation coefficient is 0.65. The weather of late April was remarkably similar in these years.

Since 1981, the daily CET series may well have undergone some slight modifications, but there is most definitely a cold spell that occurred during at the second half of each month, the minimum CET in 1908 was a couple of degrees colder than it was in 1981 though, and those on the 24th and 25th still hold the record for lowest minimums on those two days (blue stars). Personally I only see a broad similarity between the two, I’ll have to spend some time and write some code to generate a correlation coefficients between these two months and see what I come up with. If you look closely at the graph of CET (fig 2), you’ll notice that in just over a week, maximum anomalies rose from around -8°C to +8°C. The resultant rapid thawing of lying snow from the week-long cold spell lead to great flooding in places along rivers in the southeast especially the Thames, and the Great Ouse at Buckingham.

Figure 2

Synoptically, the 25th of April in both 1908 and 1981 were slightly similar in that they were both cyclonic in nature.

Figure 3 – Courtesy of NCEP/NOAA reanalysis

But up aloft in the atmosphere the cold air of 1908 was much deeper than it was in 1981 (figs 4 & 5).

It seems cold outbreaks towards the end of April are not at all uncommon, I’ve just picked on probably two of the more extreme events. Next week promises its own cold outbreak (fig 6), but synoptically, if the GFS model is correct, it will be more of a cold northerly rather than cyclonic as it was either in 1908 or 1981.

Figure 6 – Courtesy of OGIMET


Extreme Easters since 1772 in Central England

Figure 1 – Data courtesy of NCEP/NOAA 20th Century Reanalysis

The Met Office beat me to a story about extreme Easters of the past, but undaunted, and without the masses of climate data they have at their disposal, I pressed on with a bit of research of my own.

Because Easter is not at a fixed time each year it’s difficult to compare one with another. Easter Sunday can fall as early as the 22nd of March or as late as the 25th of April. I’ve used the daily CET series from 1772 (now there’s a surprise), and calculated a five-day mean, from Maundy Thursday to Bank Holiday Monday to do my comparison with. Because of the time range that easter can fall, I have base it on mean temperature anomalies rather than the mean temperature. So from what I’ve found the coldest Easter period since 1772 occurred in 1892 (fig 3). The Easter Sunday that year fell on the 17th of April so it was by no means early.

Figure 2 – Data courtesy of the Met Office

The mean temperature for the five days between Maundy Thursday and the Bank Holiday Monday in 1892 was 2.1°C, which was -6.4°C below the long-term average for that period (fig 3). The weather chart for the Sunday (fig 1) shows just what a bleak and cold Easter that must have been in eastern parts.

Figure 3 – Data courtesy of the Met Office

I couldn’t resist including the Monthly Weather Report for April 1892 after using it to check out my story because I was taken with some of the phrases that were used by whoever wrote the report. I have highlighted some of them from the PDF that I accessed courtesy of the Met Office (fig 4). The remark about Vapour Tension exceeding 0.25 on the south coast of Ireland and England was a real charmer, I bet sixpence was a lot of money in 1892, and what happened to the Weekly Weather Report?

Figure 4 – Courtesy of the Met Office and ©Crown Copyright

Conversely, the warmest Easter using the same method, fell in 1926 in Central England at any rate.  Easter Sunday that year fell on April 4th, and the five-day mean was +6.5°C above today’s long-term average, and if you look at the synoptic situation (fig 5) you can understand why. I did look for any weather related news for Easter 1926, and mistakenly thought that this was the year of the Easter Rebellion in Ireland, but I was 10 years too late, that occurred in 1916.

Figure 5 – Good Friday 1926, data courtesy of NCEP/NOAA 20th Century Reanalysis

Last 30 days second warmest since 1772

A fascinating battle is going on at the moment in the climate statistics of Central England temperatures, well at least it is for people who have the time to notice these things. In the last 30 day period (the 10th of March to the 9th of April), 2017 currently has the second highest mean temperature in the CET series since at least 1772, sandwiched as it is between two other very memorable periods in 1957 and 1938. If you are interested, 2017 is currently joint 9th warmest year since 1772, which is a reflection of just how mild it’s been so far this year.

Figure 1

3rd mildest March since 1659

Figure 1 – Data courtesy of the Met Office

The Met Office finally got there act together today, and fixed their web service and updated their CET web page with the latest data. As expected, March 2017 was a very mild month, and when the temperatures were finally confirmed today, it turned out that it had been the 3rd mildest since the monthly series started in 1659. I make the mean temperature for the month 8.68°C, which was exactly 3°C above the 1961-1990 long-term average. It couldn’t quite beat the CET of either March 1938 or 1957, so it ended up being the warmest March since 2012.

Figure 2 – Data courtesy of the Met Office

The month saw four new high minimum temperature records set and one highest maximum record on the 30th.

Figure 3 – Data courtesy of the Met Office

7th mildest March on record to date

With 28 days of the month now gone, March 2017 provisionally stands as the 7th warmest March in Central England back to 1772, with a mean temperature of 8.11°C which is +2.55°C above the 1961-1990 long-term average. Maximum anomalies are a little higher than minimum anomalies this month, indicating that the days have been a little milder than the nights. Mild has it’s been, there was little chance that it would catch the exceptionally mild March of 1957 though.

Figure 1

The relationship between volcanoes and Central England Temperature in recent years


Mount Pinatubo June 1991 Courtesy of Wikipedia

Now that I’ve discovered the VEI database from the NCEI, I can now overlay volcanic eruption events on top of the monthly CET anomalies and chart the results. In the above chart (fig 2) I’ve overlaid all the VEI 4 events or greater from 1980, and was surprised to find that there seemed little in the way of correlation between them. The Pinatubo eruption of the 15th of June 1991 for example was the first VEI 6 event since Tambora in 1815 (fig 1), and lifted more than 5 cubic kilometres of material 25 miles straight up into the stratosphere, coincidentally a typhoon that was passing close by to the Philippines at the same time scattered the ash from the volcano to the four winds. I thought the effects of this would have had a dramatic cooling effect on CET in 1992, but not that you would notice. Of course any cooling in a local temperature series may well be masked by other regional and global factors that influence CET that are going on at the same time, global climate is complicated.

Figure 2

The last 30 years CET

Figure 1

This chart is of 365 day centred rolling daily mean CET temperature anomalies for the last 30 years. Just to complicate things, I’ve overlaid a 14 day moving average on top of this just to smooth things out a bit. The LTA that I have used to calculate the anomalies is the latest possible 30 year period that I could use i.e. 1987-2016.  The decadal linear trend for the last 30 years is just +0.08°C per decade. Of course, things would look different if I had used a longer period or an earlier LTA, but such is the way with statistics.

Joint 7th mildest start to March since 1772

Central England has been super mild this March, with mean CET anomalies almost +3°C above average, but today is maybe the last mild day for a while as colder weather sets in for the rest of the week.

Top ten mild start to March

The first 15 days of March 2017 have provisionally made it the 10th mildest start to March in Central England since 1772. Currently the mean to the 15th is 7.97°C or +2.77°C above the 1961-1990 long-term average. Whether it will be able to keep to through the rest of the month remains to be seen. I did make a rash statement that it wouldn’t beat March 1957 in one of my earlier missives, which may now be looking a bit rash on my part! The blue line in the top chart is the 2017 mean, and the red one 1957 (fig 1).

Figure 1

March 1957 – warmest in over 357 years

Figure 1

March 1957 stands out as being the mildest in the CET series which started in 1659, and as mild as this current March is, the mean anomaly of which stands at +2.48°C on the 12th, it will have to do a lot more to surpass the mean temperature of 9.2°C of March 1957, which was +3.53°C above the 1960-1991 long-term average (fig 1). Here are the daily surface charts for the month courtesy of Wetterzentrale (fig 2). In fact even 60 years later, the month still holds claim to three of the warmest March days in the CET series, the 3rd, 11th and 12th. The MWR explains how a pulse of warm dry air from Spain on the 10th, produced a period of Summer like weather that lasted to the 13th. Temperatures at several places in England and Scotland reached 70°F on the 12th, at Haydon Bridge (Northumberland) they reached 74°F (23.3°C), and unusually 69°F was recorded as far north as Cape Wrath.

Figure 2 – Courtesy of Wetterzentrale

As you can probably see from the daily charts (fig 2) the country was flooded by mild south or southwesterly for much of the month, the mean pressure chart for the month shows a negative pressure anomaly of -17 hPa at around 30° west and 50° north (fig 3 & fig 4), with higher than average pressure over Greenland (+11 hPa) and Scandinavia (+6 hPa).

Figure 3

Figure 4

This of course resulted in unseasonably warm temperatures, not just affecting the British Isles, but also across much of western Europe as well, with a positive warm anomaly (+4°C) over eastern France and Switzerland (fig 5).

Figure 5

So March 1957 ended up with very few stations reporting an air frost, and ended up warmer than the following April, quite an exceptional month.