How are weather temperature records broken?

30 09 2011

At the Met Office we report weather extremes, such as the maximum monthly temperatures, as well as long-term weather averages. This post is in response to questions posed by our Twitter followers, who have been asking how maximum temperatures are measured and why observations from some thermometers are not valid for official records.

How are official temperatures measured?

Since the 1960’s only observations from stations that meet specific criteria and that are calibrated and regularly checked by the Met Office are used for quoting temperatures.

Stevenson screen

Thermometers must be housed in a Stevenson screen with its door facing north. A Stevenson screen is a white slatted box which houses the thermometer away from direct sunlight and allows air to flow freely through it.

The location of the Stevenson screen is important to ensure consistency across sites. It should be mounted at a height of 1.25m on level grassy ground, away from trees and man made structures. It should also be at least 20m away from any area of concrete or hard-standing, and only half the area within 100m radius of the screen can be covered with buildings or man made surfaces.

The thermometer itself must measure temperatures to the nearest decimal point. This is why observations from some stations – such as METAR stations at airports – aren’t included in official records, as they normally only record the temperature to the nearest whole number.

When are maximum temperatures officially recognised as a new record?

Each day real time data is subject to preliminary quality control before it is released, such as cross checking against nearby stations. A record will not become official until thorough quality control has taken place on the data, which may take several months.

How far do records go back?

Met Office standardised daily temperature records were started in the 1960s, however there are historical paper records for earlier dates held in the National Meteorological Library and Archive. Our National record for the UK goes back to 1910, although there are others that go back further such as the Central England Temperature Record. Individual station statistics go back varying lengths of time dependig on when each station was opened.

Why are some places often the warmest?

Although all thermometers are housed in standard conditions the geographical location of the station does still have an impact on the temperatures recorded. An example is the Gravesend station, which recorded the highest temperature in the UK this Wednesday. The station is surrounded by a built-up industrial area, which could lead to a slightly higher maximum temperature due to factors such as heat being absorbed by buildings.

For full weather extremes listings and historical records go the Met Office Climate section.

Typhoon Nesat brings heavy rain and strong winds to the Philippines.

27 09 2011

Typhoon Nesat made landfall at about 2200 UTC 26 September near Casiguran on the east coast of the island of Luzon in the Philippines. Winds were estimated to be close to 120 mph near the centre of the typhoon and the lowest central pressure was estimated to be near 944mb. Good forecasts of the typhoon’s track in the previous two days allowed authorities to order the evacuation of over 100,000 people in its path.

Satellite image of Typhoon Nesat as it makes landfall

Satellite image of Typhoon Nesat as it makes landfall

Although well south of the typhoon centre the Philippine capital Manila received heavy rain and strong winds causing much disruption including the flooding of the main hospital and US Embassy buildings. Subic Bay on the west coast of Luzon just to the north-west of Manila recorded 350mm (13.8″) rain in 12 hours.

As of 1200 UTC 27 September Nesat was still classified as a typhoon and has started to restrengthen over the South China Sea. Current forecasts predict a second landfall over Hainan island on Thursday and over northern Vietnam on Friday.

You can follow all the latest Tropical Storm News on:

What has brought the warm autumn weather to the UK?

27 09 2011

Over the next few days we are expecting a spell of very warm weather for this time of year across much of England and Wales and even parts of Scotland too. The reason why we are seeing this unseasonable warm spell is due to an area of high pressure which has developed across much of central Europe, centred on Germany and Poland.

This draws up very warm air from a long way south, from parts of France and Spain. That comes across a dry continent removing most of the moisture out of the air. As a result we see very little in the way of cloud with blue skies and plenty of sunshine. As a result the sunshine warms the ground and the ground warms the air so we see high temperatures for this time of the year.

In the video below Paul Gundersen, Met Office Chief Forecaster provides more details about this warm spell, how long it will last and whether this really is an ‘Indian Summer’.

What is an ‘Indian summer’?

26 09 2011

After a cooler than average summer, a spell of settled weather is expected later this week, with temperatures up to 27 °C possible in some areas from mid week.

Many media reports are calling this settled spell an ‘Indian summer’, however according to the Met Office’s Meteorological Glossary, it’s a little too early in the year. An Indian summer is defined as a warm, calm spell of weather occurring in autumn, especially in October and November.

William R Deedler, Weather Historian at the United States National Weather Service, describes it as “any spell of warm, quiet, hazy weather that may occur in October or even early November”.

The origins of the term Indian summer are uncertain, but several writers suggest it may be have been based on the warm, hazy conditions in autumn when native American Indians chose to hunt. The earliest record of the use of the term is in America at the end of the 18th century. Although William R Deedler also refers to a reference by a French man, John de Crevecoeur, in 1778:

“Sometimes the rain is followed by an interval of calm and warmth which is called the Indian Summer; its characteristics are a tranquil atmosphere and general smokiness. Up to this epoch the approaches of winter are doubtful; it arrives about the middle of November, although snows and brief freezes often occur long before that date.”

The term was first used in the British Isles at the beginning of the 19th century, but there is no statistical evidence to show that such a warm spell tends to recur each year. The warmest recorded temperatures in the UK in October and November are 29.4°C on 1 October 1985, in Cambridgeshire, and 21.1°C on 2 November 1938, in Essex and Suffolk.

For the latest weather forecast go to

September to finish on a high for parts of the UK

23 09 2011

Many parts of the UK are expected to see some fine weather during the last few days of September and into the beginning of October.

Early signs suggest there will be several days of warm and dry weather, giving the most settled conditions seen for several weeks.

The best of the weather will be seen in south and east, where temperatures could reach 27 °C. Average maximum temperatures for the UK are 16.1 °C in September and 12.5 °C in October.

Chris Tubbs, Met Office Chief Forecaster, said: “It’s too early to give precise detail, but the dominant signal is for much of the UK to get some fine weather from Tuesday onwards. Temperatures could be unseasonably warm in places, certainly well above the average for this time of year.

“The settled weather is set to last for several days and could carry on into the start of October, giving us a welcome respite from the windy and wet conditions which have dominated for the past few weeks.”

Daytime maximum temperatures are set to be above average even over this weekend, with parts of the south east around 22 °C.

The north and west of the country could see gales overnight on Sunday through to Monday, however, with strong gusts in places. After that, the weather is set to settle and steadily improve as we go through the week.

You can stay up to date with the latest forecast at

Typhoon Roke brings heavy rain and strong winds to Japan

21 09 2011

Typhoon Roke made landfall at about 0600 UTC 21 September near Hamamatsu on the south coast of Japan. Winds were estimated to be close to 100 mph near the typhoon centre and the lowest central pressure recorded was 952mb.

Rainbands extended along way ahead of the typhoon as it approached and 598mm (23.5″) rain were recorded at Tokushima (west of the landfall point) in the 48 hours prior to the centre reaching Japan. Rain and the strongest winds have now cleared from southern and western parts of Japan.

As of 1200 UTC Japan Meteorological Agency still classified Roke as a typhoon centred over eastern parts of Honshu a short distance south-west of Fukushima. Heavy rain is continuing in this region and will do so for a few more hours before clearing. It is likely to be downgraded to a tropical storm soon.

The centre of the storm is currently emerging back out over the ocean east of northern Honshu. It is expected to accelerate north-east and becomes a strong post-tropical storm in the north Pacific Ocean.

Radar imagery of Typhoon Roke

Met Office updates on Twitter: @metofficestorms

Met Office in the Media: 20 September 2011

20 09 2011

There has been widespread coverage in the media today about an impending cold winter and snow in October.  Britain faces an early big freeze in the Express and Britain to be hit by snow in October… in the Daily Mail both report that Exacta Weather forecast snow in parts of the UK as early as next month.  It should be made clear that these forecasts for the coming months are not from the Met Office. When we asked the public what types of forecasts they would like you told us that you would find a monthly forecast more useful. Currently this forecast, which is updated regularly with the latest forecast information, takes us through the first few weeks of October and says:

Temperatures are expected to be around normal for the time of year across much of the UK by day, dropping below normal at night, especially across the Midlands and southeast, leading to an increased incidence of overnight frosts. The cooler conditions at night will be mitigated by day in some parts by sunnier than normal weather, with both the far south and far north of the country favoured to see above normal amounts of sunshine, with nearer normal sunshine hours elsewhere. Rainfall amounts are correspondingly likely to be a little below average in most areas, especially in the west.

To provide some context it certainly would not be a surprise to see overnight frosts and even some snow across higher parts of Scotland in October. Historically, looking at the long term climatology, snow falls on around 3 or 4 days in a typical October across the Highlands of Scotland and on 1 or 2 days in the Southern Uplands and northern Pennines.

Therefore, it’s never too early to be prepared for winter – especially as we know from experience the types of severe weather we can see  in the UK during the winter months. 

As a result it is vital that service providers, ahead of any winter, have plans in place to respond to Met Office 1-5 day forecasts and warnings.  These provide good advice, and can be used with a relatively high degree of confidence. Therefore the Met Office has been working with service providers such as BAA and DH to help them develop robust plans and prepare for any severe weather.

It is out short term forecasts, used for example by government, local councils, train, air and road operators that can help to minimise the impacts of severe weather on you and your community.

Times Atlas of the World and Greenland Icesheet

19 09 2011

The latest edition of the Times Atlas of the World shows a reduced amount of ice and snow over Greenland which has been attributed to climate change by the publishers. This has resulted in debate among climate science experts on if this is really the case. Without seeing the precise mapping methods used in the Times Atlas it would be wrong to comment in detail on the changes between editions.

However, there has certainly been widespread warming in the coastal regions of Greenland over the past 20 years, but whether this has been caused by increasing greenhouse gases or by natural climate variations (e.g. the North Atlantic Oscillation) is not definitely known yet.

The large Greenland ice sheet is of interest because of its potential contribution to global sea level rise. Complete loss of the Greenland ice sheet would result in around 6 metres of sea level rise globally, although the effect would be less around the UK, and sea level might actually fall near Greenland because the gravitational attraction of the ice sheet currently attracts sea water towards it. While the regions shown as lost permanent ice in the Times Atlas are quite large in area, the ice/snow thickness there is small so the effect on sea level would not be expected to be large.

Observations suggest that the Greenland ice sheet has recently been shrinking at a rate that contributes a few tenths of a millimetre per year to global sea level. Some of the outlet glaciers which discharge icebergs from the edge of the ice sheet into the North Atlantic have been accelerating, however it is not yet clear whether this is part of a long term trend or just short term variations.

Over the coming century it is expected that Greenland will contribute up to 20 cm to global sea level rise (from a total of around 20-80 cm), depending on the actual amount of increase in greenhouse gases and other factors. However, current scientific understanding of ice flow and outlet glaciers is relatively limited, so it is hard to make confident predictions of their contribution. Climate scientists are currently working to continue observing the ice sheet for longer and to improve models of the processes controlling ice flow. Over time this will result in more confident predictions of the future of the ice sheet and any changes to sea level.

United States National Snow and Ice Data Centre records second lowest minimum for Arctic sea ice

16 09 2011

Arctic sea ice extent appears to have reached its second lowest minimum since records began, according to the latest figures from the National Snow and Ice Data Centre (NSIDC) in the US.

Sea ice extent was 4.33 million square kilometres at its lowest point on 9 September, just greater than the previous lowest minimum set in 2007.

Arctic Sea Ice Extent

Arctic sea ice data. Grey line indicates 1979 to 2000 average extent for the day shown. Source: NSIDC

Satellite records began in 1979 and have shown a long-term decline in sea ice extent. However, the rate of decline has accelerated in the past 15 years and the last five years make up the lowest five extents in the 32-year record.

Helene Hewitt, an expert in climate modelling at the Met Office, said: “The low ice extent this year appears to be mostly caused by high pressure systems that have persisted for much of the summer over the Arctic. This year’s minimum adds to the continuing pattern of accelerating ice loss over the last 15 years.”

Climate models which simulate future Arctic sea ice extent show wide variations, but Met Office results suggest the area could be nearly ice-free in summer as early as 2040.

Dr Hewitt added that the models do not suggest the current accelerated rate of decline would continue or that there was any ‘tipping point’ from which ice extent could not recover.

She said: “Periods of accelerating ice loss are not unusual in climate models, but there is no reason to expect that to continue. We could see periods of relatively small loss in summer sea ice in the future.

“There is certainly no indication from observations or models that a tipping point in Arctic sea ice has been reached. Indeed, models show that if there is a decrease in global temperature, summer ice extent could recover.”

The NSIDC results broadly agree with those from the other two organisations monitoring sea ice extent, the University of Bremen and the Japan Aerospace Exploration Agency. All use slightly different methods, with Bremen concluding this year is an all time low, while Japanese data shows this year as the second lowest minimum.

Arctic sea ice data . Orange line indicates 1979 to 2000 average extent for the day shown. Source: NSIDC

Next month the Met Office will publish an in-depth article on modelling of Arctic sea ice extent.

Met Office in the Media: 16 September 2011

16 09 2011

The Met Office probability weather game has received positive coverage from around the world after becoming the largest study on the understanding of probabilistic weather forecasts undertaken.  The Washington Post reported on how game players were contributing to the science of communicating uncertainty in weather forecasting, whilst Digital River review the game, reporting: “The efficacy of a well-designed ‘gamification’ strategy has been demonstrated brilliantly by the Met Office in this case.”

The use of probabilities in weather forecasting has been a topic of debate for many years but there is little in the way of research on how to present the extra information contained in these forecasts. So far the game, which sees players helping Brad the ice-cream man by providing probability-based weather advice, has been played nearly 8,000 times. There is only a limited amount of time left to play the game, which is available at The scientists leading the project are hoping that more people can take part to give even more comprehensive results.

Elsewehere several newspapers including The Times have reported on a project to take old weather records and use the to re-analyse past climate. ACRE (Atmospheric Circulation Reconstructions over Earth) aims to recover sparse historical weather observations which are then processed to create reconstructions or ‘reanalyses’ of the world’s climate over the last 200 years. A huge catalogue of old weather data, from the ships’ logs of historic voyages to World War I Royal Navy records, is being used for an international project to recreate the world’s past climate. The reanalysis will show the state of the atmosphere at six hourly intervals to give unprecedented detail about past weather. The end product will have a huge number of potential uses – including understanding future climate. Rob Allan , leading ACRE for the Met Office, said: “This project will help to shed much more light on the patterns, variability and changes in our past climate. This will not only help give us more confidence in our understanding of the past, but also allow us to better assess our predictions for the future.”


Get every new post delivered to your Inbox.

Join 8,486 other followers

%d bloggers like this: