Improving the skill of visibility and fog forecasting from the Met Office high-resolution models

The presence of fog can have serious impacts on many surface-based operations, particularly in those of busy airports but it has always been difficult to predict because of its sensitivity to tiny errors in relative humidity. This paper explores how the existing Met Office Unified Model algorithm for visibility and fog probability, which is also used as a post processing technique to take advantage of terrain-adjusted temperature fields, can be extended to be more skilful through the use a range of commonly observed aerosols rather than a single representative aerosol.