´óÏó´«Ã½

Last month was the coldest January in Wales since 1987, while December and January combined were the coldest since the winter of 1981/1982.

In a temperature series that goes back to 1914, January was the ninth coldest in Wales with a mean temperature of +1.0 Celsius, the average is 3.9 Celsius.

The coldest January in the temperature series was in 1963, when the mean temperature was a very cold - 2.5 Celsius!

So what about the rest of February? Well, since my last blog when I mentioned there was a hint of more very cold weather after the 6th, the various computer models around the world that predict the weather began to oscillate and change their mind, with big differences from one output to the next.

Because of this, confidence in the forecast dropped and it became more difficult to know what was going to happen 5 to 10 days ahead.

Ensemble forecasts:

A forecast is an estimate of the future state of the atmosphere. It is created by estimating the current state of the atmosphere using weather observations, and then calculating how this state will change in time using a computer model.

As the atmosphere is chaotic, very small errors in its initial state can lead to large errors in the forecast. This means that no forecast will ever be perfect because we can never observe every detail of the atmosphere's initial state.

Tiny errors in the initial state will be amplified, so there is always a limit to how far ahead we can predict any detail.

To test how these small differences in the initial conditions may affect the outcome of a forecast, many forecasts are produced and these are called ensembles.

Instead of producing just one forecast, the computer is run several times, changing the initial starting conditions of the atmosphere very slightly to see what happens.

If the computer forecasts vary a lot then there is a lot of uncertainty about what the weather will actually do in the future, but if the forecasts are all very similar confidence in the outcome will be higher.

Take a look at the ensemble chart for Cardiff from the American model which shows the forecast temperature at 850 hPa or approximately 5000 feet, based on the 6am run of the computer. Temperature in Celsius on the left hand side and date below:

Ìý Ìý Ìý

The different coloured lines represent the computer running with slightly different starting conditions.

What this shows is that up until 8 February, the lines are very close together, indicating high confidence in the forecast up to that point.

After this the lines begin to move apart and confidence in the forecast evolution falls, the lines eventually look like spaghetti which means anything could happen!

Notice the red line which is the mean or average of the ensembles, this does have a downward trend until about the 12 February and this suggests a drop in temperature!

Time will tell but the majority of the ensembles want to develop a block of high pressure to the north of Britain next week.

The thick green line refers to the operational run of the model which uses actual observed data as its starting point, so you would expect this to give you the most reliable outcome.

If most of the ensembles follow the operational run, confidence in the outcome is increased. In this case, the green line has a downward trend until around 11 February and most of the ensembles follow this trend which increases confidence in a drop in temperature!

There are no guarantees, but based on the latest evidence I think we should be prepared for another cold snap next week with some sharp frosts and the possibility of snow for some areas.

Derek