Seasonal climate forecast for Europe

A thermometer in a window, showing a temperature of -25°C

At this time of year, in the midst of contract renewals, property insurers are keeping an eye on the long-range forecasts for the rest of the winter in Europe. We’ve all seen the headlines in the tabloids about the next “big freeze,” or “stormy winter” doom and gloom forecasts, but what can we really say about the season ahead, and how that might translate into the weather we actually get?

Conventional everyday weather forecasts give us accurate predictions of weather conditions over the next few days, but accuracy starts to drop significantly seven to ten days into the future. To look beyond this time horizon, weather forecasting models focus on larger-scale climate influences such as upper air flows, sea surface temperatures and large scale pressure patterns, among others.

Seasonal forecasts are less computationally intensive, since they don’t explicitly model as much detail as the weather forecasts we’re used to seeing in the news. Therefore, they can be used to look further into the future and give us clues as to what type of weather we might expect over a longer period. Seasonal forecasts are expressed probabilistically, as weighted dice, or a shift of the probability distributions and use language such as “colder-than-normal” or “above average” to describe the most likely seasonal conditions.

So what can we expect for this season?

Natural climate variability is the dominant influence on the year-to-year differences between winters. Whether they are cold or mild, wet or dry, climate phenomena such as the El Niño-Southern Oscillation (ENSO), the North Atlantic Oscillation, and the Quasi-biennial Oscillation, (QBO) influence European winters on a seasonal scale. Other influences such as the amount of snow and ice cover across Eurasia, even fluctuations in solar activity, can drive climate conditions in winter.have been identified to influence our European winters on a seasonal scale. Other influences such as the amount of snow and ice cover across Eurasia, even fluctuations in solar activity can all drive climate conditions over a winter.

The biggest influence on global climate is ENSO. Currently, the latest forecasts indicate a swing to the cool phases known as La Niña, which is determined by colder-than-average sea-surface temperature in the tropical Pacific Ocean. This has far-reaching impacts around the world and could encourage more tropical cyclones in the North Atlantic (if it persists into next year’s hurricane season), while at the same time encouraging cool conditions in the north-western part of North America, and extra rainfall on the U.S. Pacific coast.

Graph ECMWF forecast from December 1 2017 to July 2018

In Europe, winter conditions are influenced less by La Niña than elsewhere, but there are still some effects that may be relevant for the next few months as La Niña persists.

For Western Europe, La Niña can increase the likelihood of blocking weather patterns, which means that milder air is less likely to move in from the Atlantic. In other words, the normal west to east flow of air from the Atlantic into Europe is “blocked.” In early winter, this has the effect of keeping conditions colder than normal, but in late winter, it could actually create more clouds and stop temperatures from dropping too far. On top of this, the NAO is predicted to be, on average, more positive this winter, meaning there could be a higher propensity for storms to move in across Western Europe, which could result in more rainfall and windier conditions, along with milder-than-average temperatures.

The outlook becomes more complex when combined with the effects of the QBO. The QBO represents stratospheric winds over the equator, which can influence the phase of the NAO. It’s currently easterly, which encourages negative NAO (associated with decreased storminess for Western Europe). In short, the combination of many different climate influences requires the use of general circulation models to resolve all of the interactions. When run for longer forecasts, those models can capture the potential uncertainty derived from the current situation and the complex non-linear interactions in the atmosphere.

After a run of relatively mild winters, current seasonal forecasts are predicting a colder-than-normal start to the winter for Western Europe, but as the season goes on, milder-than-normal conditions will become more likely, and are associated with a positive NAO. The means that extreme wintery conditions involving snow, frost and fog should be slightly less likely than normal during January and February across north-western Europe. In terms of rainfall, to accompany the mild conditions on average over the winter, we’re also likely to see wetter-than-normal conditions for Western Europe, but around average conditions elsewhere.

Nevertheless, it’s worth reiterating that long-range seasonal forecasts are probabilistic. This means that weather extremes cannot be ruled out. Even a mild winter will have periods of cold Arctic temperatures as wind shifts to the north, and a dry winter will still see the odd storm swing through, potentially bringing heavy rain and strong winds. Furthermore, a seasonal forecast highlights what could happen, but doesn’t preclude any eventuality.

Research into climate dynamics

There’s a great deal of academic research into climate variability and the dynamics and interactions of different climate phenomena. Willis Research Network ( WRN) has helped create a deeper understanding of relationship between these climate influences and losses. Recent research suggests that assessing the relationship between climate indices, such as the East Atlantic Pattern and Scandinavian Pattern, may help explain losses more accurately than the NAO. These subtleties underpin the utility of seasonal forecasts for the insurance sector.

The WRN and Willis Re Catastrophe Analytics teams have also been working on ways to use this probabilistic information on the predicted climate conditions for the season ahead, both for regional concerns and globally. The latest insights on winter storm season forecasting for Europe or hurricane forecasting for the U.S., Caribbean and Central America is routinely summarized and delivered as reports or blogs as conditions develop and strong climate influences appear. Sometimes, when there are competing climate influences, the forecast advice is to stick with the climatological norm, but when strong drivers occur, such as the conditions expected during either phase of ENSO, there’s potential to use this information for improved risk management and pricing of premiums.

About Geoffrey Saville

Geoffrey Saville is a member of Willis Towers Watson's Analytics Technology Team, having joined the company in 2013…
Categories: Environmental Liability, Insurance and Risk Management, Property, Reinsurance | Tags: , , ,

Leave a Reply

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