Standard weather forecasts provide information about the weather expected over the next few days. While it is generally not possible to predict these day-to-day changes in detail beyond about a week ahead, it is possible to say something about likely conditions averaged over the next few weeks and months ahead. Specifically, the Monthly Outlook provides information on conditions averaged over one- and two-week periods for the coming month, while seasonal forecasts provide information for three-month averages over the coming six months.

As for the standard weather forecasts, computer models of the atmosphere lie at the heart of the Monthly Outlook (and seasonal) forecasts. Three additional features of the forecast method are:

• the models are run forward in time to several months ahead rather than a few days
  
• the models include forecasts of the world's ocean circulation and temperatures as well as the atmosphere - the ocean has an important role in influencing long-term weather conditions
  
• the forecast is run not once but many times, with slight variations in the starting conditions. The result is an 'ensemble' of forecasts that provide a picture of the range of likely long-term weather conditions
  

The 'ensemble' of forecasts allows us to assess the probability of weather events. For example, for a particular event (say, temperature being well above normal), at a particular location (say, the north-east of England), in a particular period (say, the second half of the month ahead), the proportion of forecasts in the ensemble that predict this event gives the forecast probability that the event will occur. To illustrate, for an ensemble containing 50 forecasts (known as a '50-member' ensemble), if 30 members predict the event will occur and 20 members predict it will not occur we would say there is a 60% (30/50) chance of the event.

This probabilistic forecasting approach is the most appropriate method to use for monthly and seasonal forecasts - because of the greater uncertainties in weather forecasting at long range. Some users, however, find it difficult to incorporate probability (risk) information into their planning process. For these users it is possible to provide a month-ahead outlook as a deterministic (traditional) forecast. In many cases this recommended forecast is simply the most likely outcome (the one most ensemble members predict).

The Met Office Monthly Outlook is a month-ahead forecast for the whole of the UK, produced once a fortnight. The forecast is split into three Monday to Sunday periods - two seven-day periods and a subsequent fourteen-day period. For each period, temperature, rainfall and sunshine forecasts for 10 climatological regions of the UK are derived from the long-range forecast ensemble. The forecast is presented together with climatological data that help in using the forecast. For example, the forecast values may be compared to the long-term average for the period and to the conditions that were observed over the same period last year.