The Impact of the North Atlantic Oscillation on European Windstorms

Windstorms are the largest driver of annual insured losses in Europe. One way to predict the likelihood of windstorms occurring at time-scales longer than typical weather forecasts is to investigate teleconnection patterns. Teleconnection patterns are recurring and persistent large-scale patterns of pressure and circulation anomalies that span vast geographical areas.

The North Atlantic Oscillation (NAO) explains a significant amount of the atmospheric variability over Europe, and turns out to be the teleconnection pattern with the highest correlation to European windstorms.

Here we explain the NAO, illustrate its impact on windstorms and discuss current NAO forecasts.

What is the NAO?

The NAO describes the relative position and strength of two semi-permanent pressure systems: the Icelandic Low and the Azores High.

Positive NAO

A positive phase of the NAO is associated with below-normal pressure near the Icelandic low and above-normal high pressure near the Azores High—resulting in:

  • an increase in the strength of westerly winds from the Atlantic
  • additional moisture transported to Europe
  • a shifting of the storm track northward.

The overarching effect of a positive NAO is warm, wet and stormy winters over northern Europe.

Negative NAO

Conversely, a negative NAO is associated with cold and dry winters in northern Europe, and a storm track positioned over southern Europe and North Africa. A schematic showing the typical storm tracks associated with positive and negative phases of the NAO is shown above.

NAO evidence

One simple way to highlight the impact of the NAO on the storm track is to compute and map the correlation between the NAO and a proxy for storminess: surface wind speed. An example of such a plot is shown below for January. A clear positive correlation across northern Europe can be seen (red-coloured area), indicating a positive NAO leads to increased winds.

The correlation between the NAO index and surface wind speed for January, based on data from 1979-2016.

Do positive NAOs increase the likelihood of severe storms?

From an insurance industry perspective it is the severe storms—with very high wind speeds and/or large footprints—that are by far the most costly. Therefore, it is important to establish whether a positive phase of the NAO leads to not just generally higher wind speeds, but an increased chance of severe storms.

To investigate this we have utilised the eXtreme Wind Storms (XWS) Catalogue, a Willis Research Network project categorising severe storms with substantial insurance loss from 1979-2014. When we consider the XWS catalogue in tandem with the daily NAO index, we find the vast majority of storms (41 out of 52) coincided with a positive NAO value (shown below – red dots).

The daily NAO index (blue) with XWS Catalogue storms (red dots) for 1979-2014.

To further illustrate how a persistently positive NAO can be associated with a series of damaging storms we show the notorious 1989-1990 storm season below. The storm series had aggregate insured losses exceeding 17 billion US dollars (indexed to 2016), and all occurred during the positive phase of the NAO.

The daily NAO index (blue) with XWS Catalogue storms (red) for the 1989-1990 storm season.

What the future holds

The National Oceanic and Atmospheric Administration (NOAA) release a 14-day forecast for the NAO. The forecast consists of an ensemble of global climate model runs, shown below (red lines), with the last 120 days of the observed NAO (black line).

The ensemble mean of all of the forecasts predict the NAO will remain slightly negative from the 16th December to 23rd December, with a wide range of possibilities thereafter.

On a shorter forecast time-scale (e.g. 48 hours), Willis Re has recently launched an Event Response Service that allows users to monitor the current risk to European windstorms and pursue real-time actions based on the relevant information.


NAO index for the last 120 days (black lines) with 14-day ensemble forecasts (red lines).

Longer-term predictions of the NAO are not publically available. Nevertheless, the Met Office has been able to predict the NAO with some skill several months ahead over the last couple of years (Scaife et al., 2014), and more recently they have claimed the ability to make skilful predictions of the NAO one year ahead (Dunstone et al., 2016).

The Met Office does publically release a three-month outlook for temperature and precipitation in the UK; this does not directly predict the NAO, but does indicate its likely state. Their latest prediction for December-January-February was issued on the 24th November and it projects colder and drier conditions than usual, indicative of a negative NAO.

Overall this indicates a slightly decreased likelihood of a severe windstorm occurring in the near future. Of course severe storms can and do occur when the NAO is negative, as was the case with windstorm Xynthia (2010), for example.

What the data shows

The NAO is one of the dominant patterns of atmospheric variability in the Northern Hemisphere, and it has important implications in the development and path of windstorms over Europe. We have shown evidence that severe storms tend to occur more frequently during a positive phase of the NAO.

In terms of forecasts, the short and seasonal predictions of the NAO suggest the oscillation will remain in a negative phase in the immediate future, slightly reducing the likelihood of severe storms over northern Europe this winter. Nonetheless, the forecasts showing an increasing chance of a positive NAO after Christmas.


Scaife, A. A., et al. “Skillful long‐range prediction of European and North American winters.” Geophysical Research Letters (2014)

Dunstone, Nick, et al. “Skilful predictions of the winter North Atlantic Oscillation one year ahead.” Nature Geoscience (2016)


Dr. Samuel Phibbs joined Willis Re International in 2016 as an Atmospheric Scientist on the Model Research and Evaluation Team. He holds a PhD in atmospheric physics from Imperial College London, and an MSc in theoretical physics from University College London. During his PhD, Sam investigated the sensitivity of different weather systems to changing temperature, and published in peer-reviewed journals on both tropical and extra-tropical cyclones.

Dr. Ioana Dima-West joined Willis Re International in 2014 as Lead Atmospheric Scientist, part of the Model Research and Evaluation Team. She has extensive knowledge in building and validating cat-models and experience estimating climate change-related loss impacts for wind perils around the world. Ioana has published in both the academic and the insurance/reinsurance worlds and she works closely with scientists of the Willis Research Network.

Categories: Europe, Natural Catastrophe, Reinsurance, Weather risk | Tags: ,

Leave a Reply

Your email address will not be published. Required fields are marked *