Reporting the weather across the UK

8 01 2016

December 2015 was the wettest calendar month for the UK in a series of monthly weather records stretching back to 1910. But why does the Met Office state 1910 when listing records, especially when some records existed well before that time?

Part of the answer is that the Met Office has a responsibility to collate weather records for the entire UK, the UK countries and historic counties.  The digital archive used to generate our UK analyses includes station observations back to 1853, but only since 1910 has there been a sufficiently dense network of stations to allow an analysis of the whole UK.

One station, the Oxford Radcliffe Observatory, which is managed by Oxford University School of Geography and the Environment – holds rainfall records back to 1767. This allows a greater understanding of the rainfall in Oxfordshire, but doesn’t allow greater comparison with England or the UK: vital when you are trying to provide a complete picture.

The England and Wales Precipitation (EWP) series stretches back to 1766. In recent times the EWP – a highly significant climate series – is based on records from around 100 stations, but the further you go back the fewer recording stations there were. This provides a good analysis of records for England and Wales, but doesn’t capture the remainder of the UK: Scotland and Northern Ireland. Additionally, it doesn’t take account of the thousands of recording stations which provide more detailed picture for the UK in more modern times.

Dr Mark McCarthy is the head of the Met Office’s National Climate Information Centre. Commenting on the results he said: “Although our UK dataset currently only stretches back to 1910 we are adding to it by digitising more of our extensive paper archives in order to extend these records further back in time. When we have done that it is possible that months like October 1903 may rival or even surpass some of the UK records set in December 2015.

“However for December 2015 we have a good picture  of the rainfall patterns across the UK such as the record breaking rainfall in: Cumbria, North Wales; eastern Dumfries and Galloway; and parts of the Cairngorms.”

December 2015 rainfall anomaly map

December 2015 rainfall anomaly map

“In fact, as our very high-resolution rainfall map in December 2015 shows, parts of England were close to average and some places actually recorded lower than average precipitation. Just like a digital photograph, greater resolution allows you to observe finer detail.  Therefore picking any one place or region may not be representative of the UK as a whole.”

Met Office national records are created using a method to interpolate observations from our network of stations onto a 5km by 5km grid covering the UK. The gridding method is a more sophisticated approach for analysing rainfall than simply taking an average of station data. However, because it also requires a denser network of stations it is not as long running a series as the EWP and some long running observing sites. The different datasets are therefore complementary and we use both to monitor our changing climate.

So, the UK’s national climate series – the records you will see quoted when the Met Office routinely releases statistics – is a comprehensive rainfall analysis covering the whole of the UK back to 1910 using all available observations. Other series including the EWP are also a vital part of our national climate monitoring and provide us with an even longer historical context for some parts of the UK.

Professor Adam Scaife is a climate scientist with the Met Office’s Hadley Centre. He said: “It’s clear that December 2015 was a very significant month for rainfall and was the highest since our records began in 1910.  We have been asked about the link between climate change and the rainfall in December 2015.

“With or without climate change there have always been exceptional spells of weather and there always will be. But climate change can add to the natural variations in our climate and it is this that increases the chance of record breaking weather and unprecedented extremes.  It is therefore vital that we monitor our weather and climate in as much detail as possible to assess and predict future weather extremes.”

Met Office rainfall records – how far do they go back and what can they tell us?

14 02 2014

As we have seen over recent weeks and months, observations for the UK are essential to put recent weather into context and to detect variations and possible long-term trends in UK climate. So, when the Met Office quotes “the wettest on record” what does that mean?

Station records
All our time-series of rainfall come from observations made by rain gauges and their length is determined by how long the recording stations have been open.

Stations with long records are a very important part of the UK’s weather station network. These time series provide an accurate picture of rainfall for that particular location, provided there are no significant changes in instrument type or station exposure. One of the longest in the UK is the weather station at Oxford Radcliffe Observatory, which holds nearly 250 years of rainfall observations from 1767 to the present day and is maintained by Oxford University School of Geography and the Environment.

This will give an accurate picture of rainfall for the city and will be broadly representative of the year-to-year variations of rainfall across Oxfordshire or even central England, but it won’t be useful when looking more widely across England and Wales or the UK.

Regional records
So, to accurately make comparisons across regions and countries we need to create a different time-series.

One way of doing this is to calculate the average of different weather stations across an area – for example England and Wales. The England and Wales Precipitation series (EWP) is such a series of monthly rainfall totals from 1766 to the present day. For the most recent decades, the EWP series is based on over 100 stations, although this number decreases as you go back through the 20th, 19th and 18th Centuries.

The EWP series is much more reliable than a single station at representing rainfall for England and Wales overall and, because it extends back to 1766, it is very important climate series.

However, it does not cover the whole of the UK or take full advantage of the complete network of several thousand stations currently recording rainfall across the country.

National records
Met Office national records are created using gridded datasets which interpolate observations from the full network of stations onto a 5km by 5km grid covering the UK. The gridding method is a more sophisticated approach for analysing rainfall than simply taking an average of station data. However, because it is a digital series it is shorter than the EWP – the number of rain gauges with data in our electronic archive decreases rapidly by the early 20th Century.

So, the UK’s national climate series – the records you will see quoted when the Met Office releases statistics – is a comprehensive gridded rainfall analyses back to 1910. This series provides the best estimate of overall rainfall and its distribution across the UK.

The gridded rainfall analysis also enables us to produce maps showing UK rainfall patterns, for example January 2014:

Rainfall anomaly January 2014

Rainfall anomaly January 2014

How do the series compare?
Here are the headlines for January 2014:

  • For the England & Wales areal series, January 2014 (173.5mm) was the wettest January since 1948 (176.8mm) and the second wettest January in the series from 1910
  • For the EWP series, January 2014 (185.0mm) was the wettest January in the series from 1766, marginally wetter than 1948 (176.8mm)
  • For Oxford, January 2014 (146.9mm) was the wettest January in the series from 1767, wetter than 1852 (138.7mm)

You can see from the graph below that the EWP and the national England & Wales series both represent the same area and are very similar. Rainfall totals for the Oxford Radcliffe Observatory series are generally lower.

For any individual year there can be significant differences between series. For example we would probably conclude that January 1988 was climatologically more extreme for England and Wales than for Oxford, similarly 1997 is the driest January for England and Wales by a reasonable margin, but there are a number of similarly dry or drier Januarys than 1997 for Oxford.

January rainfall comparison

January rainfall comparison

So which is the best series to use?
Well, the answer is that we need to use them all. The Met Office routinely quotes rainfall statistics based on the gridded data, because these are considered the most reliable estimates, are based on the full network of observations, and can provide the regional pattern of rainfall.

The EWP series is an invaluable climate series because it provides a much longer near 250-year perspective but has less regional detail.

However, our climate analyses would not be possible without the long running high quality individual station series such as Oxford. These are the foundations of historical climate analysis.


Get every new post delivered to your Inbox.

Join 8,484 other followers

%d bloggers like this: