In this article we look at our developments in high resolution forecasting which enables us to forecast potentially high-impact but small-scale weather features - such as thunderstorms - with improved accuracy.
Weather forecasting is complex and challenging at any time - a challenge the Met Office rises to every day, making us one of the top two national weather services in the world - but certain types of weather are much harder to predict with precision than others.
A prime example is the heavy, often thundery but highly localised showers we can get here in the UK - particularly in summer months.
These storms can be just a few hundred metres across and can bring exceptionally heavy rain in short periods of time, potentially causing flash flooding as we saw in parts of the UK in June 2012.
The Met Office introduced cutting edge technology into its operational forecasting to help improve the accuracy of forecasting for weather like this and other 'small-scale' weather features.
This involves using a much higher resolution version of the Met Office forecast model which is used to simulate what the atmosphere - and weather - will do next.
Instead of using grid boxes of 4 x 4 km, the new UKV model uses a much finer scale at 1.5 x 1.5 km over the whole UK. This can be likened to vastly improving the resolution of a digital picture, revealing much more detail and providing much more clarity than before.
Brian Golding, Deputy Director of Weather Science at the Met Office, explains: "The Met Office UKV system enables us to forecast certain types of weather with much more detail and helps us get a better picture of the type of weather we may see, for example - instead of just predicting showers, we can see how intense or prolonged those showers or thunderstorms may be.
"It's effectively our best forecast of what will happen, but the drawback is there is only one version run at any one time, so it's difficult to know how likely that forecast is."
For the Olympics, the Met Office took high-resolution forecasting a step further by running multiple forecasts at the same time, a technique called ensemble forecasting.
The Met Office has used ensemble forecasting for many years and it is a vital tool for assessing probabilities attached to aspects of the forecast - for example, the chance that rain will fall in a given place over a given time period (known as 'probability of precipitation').
Previous ensembles have worked at a lower resolution of 18 x 18 km - mainly because running so many forecasts requires a great deal of computing power. However, a recent scheduled upgrade in the Met Office's supercomputer capacity has enabled trialling of a much finer resolution ensemble system - using a resolution of 2 km x 2 km.
Brian Golding, Deputy Director of Weather Science at the Met Office, continued: "Some types of weather, and even individual weather systems, are very sensitive to how well the atmosphere is observed and it's very difficult to be sure you have that exactly right, where only small differences can lead to big changes in the forecast.
"By running multiple forecasts with slightly different starting conditions we can tackle that by estimating how likely a forecast is. If the weather outcome is very sensitive to starting conditions, we know there is a lot of uncertainty, if the outcome is more robust to slight variations, then we can be more confident.
"This means we can assess the chances of weather impacts in a certain area at a certain time, so we can give much more useful guidance on the chances of, for example, flooding from a heavy thunderstorm, or of trees being blown down by strong gusts from a storm."
Even with all this technology, it is not be possible to predict the exact location and timing of each weather feature over the UK, or the impacts of that weather. However, the increased detail from these high resolution forecasting techniques will enable the Met Office to give ever improving guidance to its customers to help protect lives and property.
The high-resolution ensembles was tested throughout the Olympics before being subject to further research with a view that the facility could be introduced operationally in the future, potentially leaving a legacy that will benefit the UK well after the Olympic and Paralympic Games.
Ensemble forecasts quantify uncertainties in weather prediction and estimate risks of particular weather events.
What is an ensemble forecast?
Instead of running just a single forecast, the computer model is run a number of times from slightly different starting conditions.
Ensemble probability and decision-making
Ensemble prediction allows the uncertainty in forecasts to be assessed quantitatively.
Last updated: 17 September 2015
Daily air quality forecasts
Air quality forecasting for the Olympic Games.
Wind and wave modelling
When it comes to sailing, knowing what the wind and sea are doing is everything.
Doppler radar and weather observations
Improving our weather observing capability ahead of London 2012
Olympic and Paralympic weather statistics
Weather statistics for Olympic venues.
Weather and sport
Find out more about how weather affects sport from a range of competitors.
London 2012 Olympic Games
25 July - 12 August - Check the five day weather forecast for the London 2012 Olympic Games.