Compiled from temperature observations obtained over land and sea, HadCRUT is used as a basis for a global temperature record going back to 1850.

The latest version of the dataset, called HadCRUT4, includes newly available data – notably adding much more information from the sparsely observed northern higher latitude regions.

Differences in the way sea surface temperature observations have been collected have been taken account of and the new version also provides much more detail on uncertainty.

Colin Morice, climate monitoring research scientist at the Met Office said: “The new study brings together our latest and most comprehensive databases of land and marine temperature observations, along with recent advances in our understanding of how measurements were made at sea. These have been combined to give us a clearer picture of what the historical data can tell us about global climate change over the past 161 years.

“Updates have resulted in some changes to individual years in the nominal global mean temperature record, but have not changed the overall warming signal of about 0.75 °C since 1900.”

One of the key reasons for slight changes to mean temperature for later years in HadCRUT4 is the inclusion of much more data from the Arctic, an area which is warming faster than other parts of the world.

Phil Jones, research director of the Climatic Research Unit, said: “HadCRUT is underpinned by observations and we’ve previously been clear it may not be fully capturing changes in the Arctic because we have had so little data from the area.

“For the latest version we have included observations from more than 400 stations across the Arctic, Russia and Canada. This has led to better representation of what’s going on in the large geographical region.”

Another change relates to dealing with the different ways sea-surface temperatures have been measured. This has had an effect on some years further back in the record, particularly in the mid 20th century.

Peter Stott, head of climate monitoring and attribution at the Met Office, said: “An example of this is the rapid changes in the kinds of measurements we see in the digital archives around the Second World War. Some sea surface temperature observations were taken from buckets hauled on board ships and others were made in the engine rooms.

“Research has shown readings from buckets were generally cooler so when the database changes from one source to another you see artificial jumps in the raw data. We’ve quantified these effects and corrected for them providing a clearer view of the evolution of global temperatures.”