Are long range weather forecasts any good?

There is no doubt that accurate weather alerts, months in advance could help industry manage energy costs and prepare for potentially damaging cold-snaps and flood conditions. However, Paul Knightley of MeteoGroup UK warns that such meteorological techniques are actually still very much in their infancy...

ONCE AGAIN the news is full of predictions of the coming winter's weather, with some anticipating plenty of cold weather, and others expecting a rather milder season. However, all forecasts of this nature should be treated with caution.

Accurate long range forecasts are really the "Holy Grail" of meteorology. The economic benefits of accurate long range predictions are enormous.

In winter, energy companies would know the likely demand for gas and electricity. Local authorities and the Highways Agency could order in the correct amount of salt to keep the roads free of ice and snow. In the warmer parts of the year, retailers would know when to bring out new season fashions.

Farmers would know when to sow, when to harvest and when to protect crops, and water authorities could start to reserve water before a drought occurs rather than after. All in all, the world would be a very different place.

Unfortunately, the science of long range forecasting has not advanced a great deal in recent years. Forecasts beyond 10 days cannot be trusted and are likely to go wrong nearly as often as they go right.

Unlocking the door

The techniques involved in long range prediction are quite different from those used in short term forecasting.

The weather in the next few days is predicted by super computers which use very complex maths and physics to model how the atmosphere will change through time from one hour ahead to a few days ahead. As everybody knows, even these forecasts are not always reliable.

Chaos theory means movements in the atmosphere are not always predictable. If the super computers get tomorrow's weather just slightly wrong, that small error may have multiplied many times over by day seven and so the forecast for a week ahead may be hopelessly out.

For such reasons, using these super computers to accurately predict six months in advance with conventional methods is simply not possible.

Despite this, all is not lost for meteorologists in pursuit of the "Holy Grail". There are other techniques which can allow forecasters to attempt predictions for a few months ahead.

Matching analogue weather patterns, studying ocean currents and how they correlate with future atmospheric patterns, looking at sea surface temperature anomalies and analysing solar variations are just a few of the techniques that just might unlock the door to long range accuracy in the future.

One way or the other!

Some winter forecasts are based on predictions of the North Atlantic Oscillation (NAO) Index. The NAO is not as complex as it sounds. It is simply an index to represent the average variation in air pressure between the Azores and Iceland.

The "normal" is for low pressure to dominate the Iceland area with high pressure over the Azores. This pressure distribution is the reason for our prevailing south-westerly winds and relatively mild weather in the UK.

However, when pressure is higher than normal across Iceland and lower than usual around the Azores, the NAO is said to be in a negative phase. Our balmy south-westerly winds will then often give way to chilly winds from the east.

If, for example, the NAO index turns out to be negative this winter, then we have a higher than normal chance of experiencing a cold winter with an increased threat of snow.

However, the science of long range forecasting is very much in its infancy and there is still every chance that winter will follow the trend of the last decade or so and be predominantly mild with just occasional wintry interludes.