Anyone looking at our current 16-30 day UK forecast (for the first half of February) will have noticed that uncertainty is the theme – much more so than usual, even when looking this far ahead.
The outlook for this period is finely balanced, with two scenarios looking equally possible: the first being a continuation of the changeable and relatively mild conditions we’ve seen so far this winter; the second being a shift to a period of much colder weather.
Whilst two different outcomes are possible, it does look as though the general pattern that establishes itself by the start of February could last for quite a while, perhaps even until the middle of the month.
However, it is not possible to say which of the two scenarios is more likely due to an unusual amount of uncertainty in the forecast.
For these longer range outlooks, we use sophisticated techniques to understand the uncertainties – called ensemble forecasts.
In an ensemble forecast we run a forecast model many times from very slightly different starting conditions. The range of different outcomes gives us a measure of how confident or uncertain we should be in the overall forecast.
Currently the ensembles for the start of February onwards show a fairly even split that does not clearly favour either the mild or the cold option – and there are relatively few solutions in between.
There is a possible explanation for this relatively rare situation. In recent days there has been an increase in temperatures in the high atmosphere (the stratosphere) over the northern hemisphere. This change seems to have been driven in part by marked changes in the weather patterns lower down in the atmosphere – over the north Pacific for example.
Research, developed in part by the Met Office, suggests that changes such as this high up in our atmosphere can – in time – go on to affect weather patterns at surface-level. However, this is a new and very complex area of meteorology where we are still developing our understanding of all the mechanisms involved.
It is possible this is why we are seeing the volatility in the current forecast, as different model runs handle the interactions differently.