• Hourly
  • 7 day
  • 16 day
  •  Hourly
  •  16 day
  • 10 day
  • More

News and information

UK winter 2021-22

Update 2



For context and background information please read:

Winter 2021-22 weather, an initial look issued on September 20th 2021.

For the third update please read Winter 2021-22 final pointers which was issued on 19/11/2021


Berkhamsted snow, December 2009

Heavy snow in Berkhamsted, December 2009

September and October

The Central England Temperature (CET) for September was well above the 30 year average. On balance TWO considers a cooler month to be more likely to be followed by a colder winter.

The October CET is also likely to finish above the 30 year average. A correlation between warm and dry Octobers and colder than average winters has been discussed. 

Seasonal models

Publicly available data from most of the seasonal models covers the entire period of the meteorological winter.

Model Temperature Precipitation
UK Met Office GloSea (UK) D/J/F Above average Above normal
Meteo France (France) D/J/F No bias Mostly no bias
ECMWF D/J/F Above in the south west. Mostly no bias elsewhere Mostly no bias
DWD (Germany) D/J/F Above average Above average in the north west
ECCC (Canada) D/J/F Above average Above average
International Research Institute (IRI) D/J/F Below average in Scotland. Above average or no bias elsewhere  Mostly above average
CMCC D/J/F Above average Mostly above average
JMA D/J/F No bias in the north. Above average in the south Mostly no bias
C3S multi system (European combi) D/J/F Above average Mostly no bias. Above average in central regions
CFS v2 (USA) D/J/F Above average  Above average in the west

D/J/F = December, January, February

The Climate Forecast System v2 is available on TWO. View the latest CFS v2 charts.

Since the first review, issued on September 20th 2021, the seasonal models have firmed up in suggesting:

1) Above average temperatures

2) Precipitation more likely above than below the average

It is important to appreciate that the skill level of seasonal models for the UK and north western Europe is low, in other words they are not very accurate.

North Atlantic Oscillation (NAO)

The NAO is essentially a measure of pressure patterns across the North Atlantic. During the winters when a negative NAO develops, blocking areas of high pressure form at high latitudes and displace the cold Arctic air down to mid latitude locations such as the UK.

This year the likelihood of a neutral or negative NAO is considered to be slightly higher than the norm, but a predominantly positive phase is still favoured.

Quasi-biennial oscillation (QBO)

The QBO index is determined by the strength and direction of equatorial zonal winds in the tropical stratosphere. When the winds are in a westerly phase the index is positive and when in an easterly phase it is negative. A correlation between the strength of the jet stream across the North Atlantic and the QBO has been identified. A negative (easterly) QBO favours a weaker jet stream which in turn means a greater chance of cold spells during the winter months.

An easterly QBO is expected to continue through the winter. This is believed to increase the likelihood of a colder than average winter.

Grand Union canal in Berkhamsted, February 2018
Grand Union Canal, Berkhamsted. February 2018.

El Niño Southern Oscillation (ENSO)

El Nino takes place when SSTs in the central-east Pacific are anomalously warm and La Nina when they are colder than average. ENSO has an impact on global weather patterns, although the link to the UK is quite weak. A weak El Nino event is believed to increase the chance of colder weather during the second half of the winter.

Forecasting the ENSO conditions several months ahead is prone to error, but at the moment there is a 87% chance of a La Nina event persisting through the Northern Hemisphere winter. The percentage chance has further increased since the previous update.

The TWO view is a weak La Nina increases the chance of a cold winter in western Europe. A strong event probably diminishes it.

North Atlantic Sea Surface Temperatures (SST)

SSTs to the north and west of the UK are mostly above the norm. This will help to make northerly air streams less cold than would otherwise be the case.

Solar Activity

The link between solar activity and the weather remains controversial. There is a suggestion that colder winters are more likely to occur in the UK close to or shortly after a solar minimum is reached.

The coming winter is expected to coincide with the early phase of Solar Cycle 25. The position of the solar cycle may increase the chance of a cold winter in western Europe.

Analogue years

The Winter Analogue Index (WAI) 30 day tracker at the time of publication has the 10 best Northern Hemisphere matches as follows: 

1) 1955 (Cold)

2) 2005 (Close to average)

3) 2013 (Very mild)

4) 1973 (Mild)

5) 1968 (Cold)

6) 1995 (Cold)

7) 2007 (Very mild)

8) 1982 (Close to average)

9) 2017 (Close to average) 

10) 1998 (Very mild)

At this stage the WAI has an even split between cold, mild and average winters. Later in the autumn the 60 day tracker will become available and it will be used as an input in the winter forecast. 

Taken as is, the current 30 day tracker possibly suggests a greater chance of a cold winter than has been the case in recent years.  


Since the first update the seasonal models have firmed up on pointing towards a mild winter in the UK. The precipitation forecast has also leaned further towards a wetter than normal season.

Other factors such as NAO, QBO and La Nina are consistent with the previous update.

The WAI was not available when the first update was issued. However, it may offer weak support for an increased chance of an average or colder than average winter compared to recent years.

Nonetheless, recent climatology favours milder conditions. Most months this year have produced positive temperature anomalies, although spring (March, April and May) was colder than the 30 year average and winter 2020-21 was close to the norm in the south and colder in the north.  

Taking the above into account the current TWO view is that a milder season is favoured, but there are a number of pointers which support a cold scenario.

The TWO winter forecast is released at the end of November and it will also consider how weather patterns developed during the last weeks of the meteorological autumn.

Press and media enquiries

We welcome all press enquiries, whether relating to this news feature, or seeking comment on other weather news.

Contact us



14 day outlook

Monthly outlook

Seasonal outlook


A huge range of charts and data is freely available.

Models include UK Met Office UKV and MOGREPS-G, ECMWF, NCEP GFS, Meteo France Arpege and Arome.