At short range, six- to 48-hours, it is possible to predict many fine scale details about the weather conditions.
Short-range forecast are mainly provided by the North Atlantic and European (NAE) numerical forecast model. By running the model over only the North Atlantic region, rather than globally, we are able to increase the resolution. This leads to an improved representation of atmospheric processes and an increase in the level of detail in the forecasts. Adding increased detail inevitably increases the risk of errors.
The North Atlantic and European model is also run in an ensemble forecast to help understand the uncertainty in each individual forecast.
A global model is also needed to provide forecasts for cities around the world and worldwide aviation forecasts of upper-level winds, icing and turbulence.
When forecasting more than two days ahead, the global model is needed because the weather we experience several days ahead can be influenced by the weather happening right now at the other side of the world.
Forecasts out to six days ahead are provided by the global forecast model. The global model also feeds weather into the North Atlantic and European model from outside its region.
As we extend the forecast time range even further, out as far as two weeks ahead, the use of ensemble forecasts is essential.
One to two days ahead the uncertainty is usually in the local details of the weather, but several days ahead the whole weather pattern can often be uncertain.
In ensemble forecasting, many separate forecasts are run for the same forecast period, each with slightly different starting conditions. The resulting forecasts become increasingly different from each other as the forecast range increases.
The spread of different forecasts, for a given time, gives the forecaster an indication of the uncertainty of the forecast.
Forecasters are also able to compare the results from our model with those from other centres. If all models are producing approximately the same solution, confidence in the forecast would be high.
Confidence can also be estimated by the consistency between the latest model forecast, earlier forecasts and ensemble forecasts. If the model is consistent then confidence may be high but if it suddenly changes then confidence falls rapidly. In these situations the solutions of other models may be crucial.
Research is ongoing at the Met Office on the use of ensemble forecasts to estimate the inevitable uncertainty associated with weather forecasting and in how to best communicate that uncertainty to forecast users.
Early warnings and advisories of severe weather are issued by our forecasters as part of the National Severe Weather Warning Service. The decision to issue a warning is made after the forecaster has examined forecast data from all of these sources, combined with their experience and judgement about the likelihood of severe weather occurring. This human-machine partnership is very important in producing accurate weather forecasts.
Last updated: 5 August 2013