What’s the sunniest month of the year?

I thought that I would investigate and find out exactly what climatologically is the sunniest month of the year using monthly sunshine data from the Met Office. The gridded sunshine record only started in 1929, but I’ve been reliably informed by someone in the know at the Met Office, that plans are well advanced to extend this (along with the rainfall and temperature series), back to at least the start of the 20th century, which I applaud, although in my opinion it’s a long way overdue. The add-on to my application looks not only at sunshine, but rainfall and temperature too, and for any region within the United Kingdom. The pie chart I’ve include is for England & Wales (fig 1), and reveals, surprisingly to many I bet, that May is climatologically the sunniest month of the year in 35.2% of years since 1929, followed by June in 31.8% of years, and July in 19.3% of years.

Figure 1 – Data courtesy of the Met office

Also rather surprisingly, it can happen that in some years April is the sunniest month. According to my application that’s occurred five times, in 1945, 1954, 2002, 2007 and 2011, so three times in the last 15 years. There’s always a chance that I’ve made an error let me know if I have.

The sunniest month in the entire England and Wales record was the July of 2006 with 287.6 hours of sunshine (fig 2). The sunniest month in the famous summer of 1976 was also a July with 251.5 hours of sunshine, but ranked only 13th in the list of sunniest months in a year.

Figure 2 – Data courtesy of the Met Office

The sunniest May’s in the UK since 1929

What was the sunniest May in the UK? That of course depends on where you live, so using the regional gridded sunshine data series from the Met Office that started in 1929 here is a map of the sunniest May for each of the nine individual regions, and for the whole series in a horizontal bar chart (fig 1). May is climatologically the sunniest month of the year in the UK, but more of that in an upcoming post, and by the look of it, the May of 1989 was the sunniest in nine of the seventeen regions, including the UK and England and Wales, if the gridded sunshine values are anything to by.

Figure 1 – Data and map courtesy of the Met Office.

The 1989 record certainly is in no danger of being broken in the southeast this year, but it’s certainly could in Northern Ireland and the west of Scotland, where they’ve got off to a flying start (fig 2).

Figure 2

April sunshine totals 1981-2010

Figure 1

The sunniest place on average in the UK in the period between 1981 and 2010 is Shanklin on the Isle of Wight with 201.4 hours. In fact most of the climate stations in the top 17 can be found on the English Channel coast somewhere. On the southwest peninsula the fall off in sunshine the further away from the coast you are looks to be around 10%. The Met Office provide these 1981-2010 average in their DataPoint web service and are not neither straightforward to download or to parse, because they’re all in XML format and come as individual files for almost 300 locations. The things I have to go through for a climate story.

Figure 2

I contacted the Met Office at Jersey yesterday and asked them what their record highest April sunshine total and got this tweet back.

Figure 3

I make their total for the same period 249.5 hours, yesterday they had another 2.5 hours taking them up to 252.0 hours. I’ve tweeted the Met Office at Jersey and asked them to check their total, but so far have had no response. As far as I can see I’ve done the maths correctly, and my old maths teacher Mr Brightmoor I’m sure would have been proud of me. The 1981-2010 average for Jersey is 196.5 hours in April, so that makes the latest anomaly just over 28% above the average.

Figure 4

A little white lie from the Met Office

Quite a lot of things annoy me about the Met Office as regulars readers of this blog will have no doubt picked up over the years. One of the worst traits that they have, is when talking about climate statistics for the UK, they use the following phrase:

“When records began in 1910…”

When you know full well that detailed records existed much earlier than that. What they ought to say, which is much close to the truth, is:

“We’ve only managed to digitize records back as far back as 1910, and even though we hold detailed climate records that go as far back as the early 1850’s, we can’t be bothered to do anything with because we’ve already spent the money on a big new shiny supercomputer.”

There is a classic example of that on their website today concerning the occurrence of frost in the UK during April.

Figure 1 – Courtesy of the Met Office

“Detailed frost recording in the UK began in 1961”. What a shocking admission for the national weather service of a country that was established in 1854. That must mean all of the sterling work done by meteorologists before 1961 was all in vain and a complete waste of time. What happened to all the temperature records between 1854 and 1960? At random, I picked out a copy of the DWR for January 1917 just to see if there were any extreme temperatures reported back then, and unsurprisingly there they were, for thirty-three stations across the British Isles (fig 2). Of course that’s just the tip of the iceberg, because like today, many more climatological returns from stations remain unpublished and go straight into the archives.

Figure 2 – Courtesy of the Met Office ©Crown Copyright

Absolutely no excuse

The Met Office were one of the first big users of computers and still are, here is a list of the computers that they’ve used down through the years (fig 3). I realise that looking forward with NWP model is vital, but so is looking back, and treasuring the climate data that previous generations have made since 1854. But why is it that climate records have always seemed to have been neglected, and climate data limited and inaccessible when there seems to have been no lack of one of the world’s fastest supercomputers at their disposal?

Figure 3 – Courtesy of Wikipedia

The Met Office have the resources, both in terms of staff and computing power to digitize the rainfall, temperature and sunshine records of the past and extend the gridded data series that they already have, back at least 50 years before 1910, but they have chosen to leave that wealth of old climate data untapped, and for the life of me, I just can’t understand why?

April 2011 – warmest on record*

Figure 1 – Image and data courtesy of the Met Office, CRU & Wetterzentrale

April 2011 – warmest on record*

*Or what you can do with a load of climate and weather data

Figure 2 – Data courtesy of the Met Office

It seems that we have incredibly short memories when it comes to weather across the country, well at least I know that I have, and that’s why I never noticed that April 2011 was, and still is the warmest April on record since at least 1659 in Central England (figs 2, 3 & 4). The mean CET temperature for the month was +3.91°C above the 1961-1990 long-term average, and 0.6°C warmer than the second warmest April in the list 2007. I’ve just put together a blog that consists of a number of tables, charts and graphs of climate data for the month that I’ve constructed from data or images that I’ve downloaded from the Internet.

Figure 3  – Data courtesy of the Met Office

Here’s a more detailed of daily temperatures, in what was a very mild Spring in Central England, particularly in March and April (fig 4).

Figure 4 – Data courtesy of the Met Office

The monthly mean 12 UTC temperature anomaly of +5°C sits across northeast France and dominates most of western Europe by its influence (fig 5).

Figure 5  – Data courtesy of NCEP/NOAA

The mean pressure field for the month of April was ideal for warm and dry weather with plenty of sunshine. An anomaly of +6 hPa over the North Sea enabled a cut-off anticyclonic cell over England and Wales, although the weather was more southwesterly in the northwest (figs 6 & 7).

Figure 6 & 7 – Data courtesy of NCEP/NOAA

Here’s a chart of daily rainfall from the EWP series (fig 8), apart from a wet start, the month was predominantly dry in all areas of the UK except the northwest of Scotland.

Figure  8 – Data courtesy of the Met Office

And here are the 12 UTC plotted SYNOP observations for Northolt  in London for each day of that April (fig 9).

Figure 9 – Images courtesy of OGIMET

And finally here are anomaly charts of rainfall, temperature and sunshine for the month courtesy of the Met Office (fig 10). The thing is I can produce maps, charts and graphs easier than I can write the text that glues them all altogether.

Figure 10 – Images courtesy of the Met Office

25 March 2017 – sunshine total too low for Exeter?

I was a bit puzzled to see that there had been just 9.8 hours of ‘bright’ sunshine at Exeter airport yesterday (25th March 2017). As far as I can see the sensor there is either faulty, or is perhaps shaded by building or something. It was a sunny day in our part of Devon from dawn to dusk, as it was in most parts of the British Isles yesterday, and the only cloud there was was some thin high cirrus late in the afternoon, but in my experience with the Campbell-Stokes sunshine recorder of old is the sunshine trace for yesterday would have been uninterrupted from dawn, to maybe half an hour from sunset. I know that the electronic sensors the Met Office use are calibrated to map as closely as possible to the old sunshine recorders, but the bright sunshine threshold for yesterday must be set far too high for Exeter and Liscombe. Here are yesterday’s automated hourly SYNOPs from the airport (fig 1).

Figure 1

Admittedly Liscombe in North Devon also reported a low total of 10.0 hours, but Yeovilton to the east in Dorst reported 1.3 hours more, and Camborne in Cornwall reported 2 hours more (fig 4). I can understand that Camborne maybe has a clear ‘sea’ horizon to the west, and would maybe report up to 0.5 hours more than Exeter, because Exeter has hills to both its west (Dartmoor) and to its east (Blackdown), which will would make sunrise a little later, and sunset a little earlier. Here are the raw SYNOPs for Exeter which include an hourly total (fig 2) which give us a better understanding of the missing sunshine.

Figure 2

As you can see from the observations, the only hours that weren’t completely sunny were 06-07 (0.2), 16-17 (0.5), 17-18 (0.1) and 18-19 (0.0). Sunrise was at 0605 UTC, and sunset 1835 UTC (fig 3), so there were was 12.5 hours of daylight, and theoretically ~12 hours of bright sunshine in this part of Devon.

The night of the 24-25th was dry and the strength of the wind prevented any dew or frost from affecting the record from 06-07 UTC. The sunrise was at 0605 UTC, but only 0.2 hours of bright sunshine was recorded that first hour, I would have thought that at least 0.5 hours could have been recorded, maybe the Blackdown hills restrict the maximum total by more than I imagine.

Figure 4

Finally, here is a tabulated ranked list of yesterday’s sunshine totals from around the country, with a percentage of the possible maximum sunshine for each location (fig 5). The algorithm that I employ for the maximum amount of daylight is simply, the time difference between sunrise and sunset for each location, which I know is not a realistic value that any sensor could ever achieve, so the percentages will be slightly higher (~3%) than those in the table.

Figure 5

The acid test of course would be to examine a visible satellite image for yesterday evening, and by a stroke of luck, I just happen to have one that I archived. Why I didn’t think to look at this in the first place I don’t know, but it does show a thick wave-like cloud that sprang up late yesterday afternoon over west Devon, which no doubt is what limited sunshine at Exeter airport in the 90 minutes prior to sunset. I’ve learnt a salutary lesson from this particular blog, and that is don’t jump to conclusions and spout off about sunshine totals from remote observing sites, even if they are just 11 kilometers away!

Figure 6 – 1730 UTC 25 March 2017 (Courtesy of the Met Office)

I discovered quite by chance that a friend of mine had taken pictures of the sunset on Saturday evening from close by in Silverton, which do show the wave-like aspect of the cloud that sprang up over Dartmoor late that afternoon (fig 7).

Figure 7 – Courtesy of Warren Radmore

21st March 2017

These are some plotted climate charts for yesterday across the British Isles and near continent. They include sunshine for yesterday, precipitation [06-06 UTC] and maximum [06-18] and minimum temperatures [18-06]. It’s not particularly exciting I know, but I was just looking at another way of displaying the charts that I generate from how I have done it in the past.

Is the UK getting wetter?

I was just looking through the climate records that the Met Office provide for downloading across the country. I had written an application, as is often the case it was another one of my climate and weather applications that I had sadly neglected to finish off. The Met Office are very frugal when it comes to allowing people to access climate data, and it’s no different with this historic station data series. Take a look at the scanty coverage of monthly climate data they make freely available (fig 1). The monthly climate data is limited to just 37 stations across the entire UK, three of those stations have closed. Here in the entire southwest of England, there are just three long time series available to download, Camborne, Chivenor and Yeovilton. No climate data available for St Mary’s, Plymouth, Exeter or Dunkeswell, even with the headquarters in Exeter.

Not that you get a great deal of climate information in the monthly data that they do release (fig 2). There is no information regarding the frequency of wind speed, direction or highest gust during the month, no MSLP information, no frequency of visibilities, no days of fog, thunder, gale, ground frost, hail, sleet or snow, just the barest of bones – why give it away when someone will pay you for it?

Figure 2

Following on from the article that I wrote about Climate changes in the UK since 1910, I thought that I would use this monthly data to see if I could find any long-term trends in annual precipitation from some of the 28 available climate records from stations across the country. I start with Sheffield, which has a long climate record that started in 1883 (fig 3). In my application I decided to display three charts for precipitation, the top one displays a 12 month rolling accumulation for the entire record. With the top graph you can select a period (yellow overlay), and display that period as a bar chart of monthly totals and monthly anomalies in the lower two charts.

As far as I can see using a simple linear trend in the accumulation series, there has been a +11.5% increase in annual precipitation in the last 133 years in the City.

Figure 3

Oxford has an even longer record that stretches back to 1853, but the increase there is only +2.6% in the 163 year record (fig 4).

Figure 4

Armagh has a rainfall record that also stretches back to 1853, but there there’s been a -1.4% decrease in annual rainfall (fig 5).

Figure 5

The Wick record which started in 1914 shows an increase in annual precipitation of +7.3% (fig 6).

Figure 6

The Stornoway record that started in 1874 shows a decrease in annual precipitation of -1.7% (fig 7).

Figure 7

It’s a bit difficult to find a climate station with a long climate record in the south so Heathrow which started in 1948 will have to do. It shows an increase in annual precipitation of +3.7% (fig 8).

Figure 8

These results obviously need plotting on a map if you are to try to make any sense from them, which I will get round to, but for now these results will have to do. I am surprised to find that a couple of stations did come up with drier trends, it could of course be down to the fact that I haven’t used a fixed time period for the linear trends.

To finish, I’ll just display the monthly England and Wales precipitation series that’s also free to download from the Met Office (fig 9). That series started in 1766, and shows a +5.3% increase in annual precipitation in the last 250 years across England and Wales. So I would say that it has been slowly getting wetter, certainly in England and Wales, but I need to do a much fuller analysis than this very limited quick look. There is one thing that I have found out conclusively though with this bit of programming, and that is rainfall records extend much further back than 1910!

Figure 9

Let me know if you spot any statistical howlers in this piece.

Exeter Airport – January 2017

I’ve just been putting some finishing touches to an application that generates a meteogram of climate data for any SYNOP station in the world. I call it a meteogram, because it attempts to generate a cross-section using graphs of the weather for a location. Because OGIMET allows you to access hourly observational data for any station, this provides a very good way of looking back at the climate of a place in the absence of NCM data being available. Here’s one hot of the press for last month at Exeter airport. Although reception was 100% for all observations, some climate sections could still be missing resulting in incomplete totals for rainfall and sunshine.

Figure 1

Figure 2

Climate changes in the UK since 1910

Just how has the climate of the UK being done since 1910?

The short answer

It’s sunnier, warmer and wetter in the UK than it was in 1910.

The slightly longer answer

Well, with the help of the Met Office, and downloading the monthly climate data series that they maintain, in a bespoke application that I have written, I am now able to generate a chart for any period since 1910 for the following observational elements:

  • Maximum Temperature
  • Minimum Temperature
  • Mean Temperature
  • Precipitation
  • Sunshine

Because the data the Met Office generate is from gridded data, they also produce subsets of the data for all the following regional and National areas:

  • UK
  • England
  • Wales
  • Scotland
  • Northern Ireland
  • England & Wales
  • Northern England
  • Southern England
  • Northern Scotland
  • Eastern Scotland
  • Western Scotland
  • North East England
  • Northwest England & North Wales
  • The Midlands
  • East Anglia
  • South West England & Wales
  • Southeast England

So with this application I can easily delve into any area for any observational element to see just what’s going on under the hood so to speak. I have constructed each chart in the same way, that is I’ve plotted a running 12 month total or average for each month (grey area series), for instance the first value I’ll plot for rainfall will be in December 1911 and it will be the total rainfall for that year. In January 1911 I’ll plot the total for the period between February 1910 and January 1911, and so on. I like this way of looking at data because it gives you a complete chart with no cold or wet spells that fall between the cracks and get hidden in a simple annual total or average chart.

On top of that I’ve overlaid a moving average (dashed line with yellow outline series), which can be for a period of 1 to 30 years, although the three charts below (fig 1-3) all have a three-year moving average. This is a bit tricky to explain because it’s a moving average of another average if you like.

Finally I’ve added a simple linear trend (red dashed series) so that I can display the trend in the annotation box in the top right.

The first chart (fig 1) shows that annually since 1930 the UK has become 10 hours per decade sunnier in 2016 than it was in 1930. That’s around 80 hours a year sunnier. My devilishly fiendish mind can think of a couple of reasons why that might be:

  • The Clean Air Act of 1956 and because the air is now less polluted by coal fires, places, particularly in Cities and particularly in Autumn and Winter, will naturally be much brighter places to live, with none of the smogs of the 1950’s.
  • The Change in the way we now measure sunshine, perhaps the new sensors are now much more accurate that the old Campbell–Stokes recorder were.

Figure 1 – UK Sunshine

Temperature wise the mean annual temperature across the UK has increased by +0.09°C per decade since 1910 (fig 2), so the mean temperature is now close to being +0.9°C warmer than it was in 1910. This agrees very well with the CET series in this regard, but I’ll leave a comparison of the two datasets for another article.

Figure 2 – UK Mean Temperature

Annual Precipitation totals are also showing an increase of around +7 mm per decade (fig 3). This doesn’t sound a lot, but over 106 years that’s almost 80 mm of rainfall. I could delve into what particular months are wetter, but the purpose of this particular blog and the new coding that I’ve done, was to look at the annual picture rather than break it down into a seasonal or monthly one.

Figure 3 – UK Precipitation

As far as I can see the three charts look comparable to the ones on the Met Office site, not only that, my charts look much better, as if someone has put a lot of care into their construction, which of course they have. If you do see a problem with any of them please let me know.