The aim of this work was (i) to examine whether the Met Office Unified Model (MetUM) is able to correctly predict orographic rain enhancement when run at its current highest operational resolution of 1.5 km, and (ii) to investigate how much of the observed orographic rain enhancement is reproduced by the MetUM at lower horizontal resolutions.
The performance of the Met Office operational weather forecasts was assessed against rain gauge observations for a large number of cases: 12 in the hills of the Lake District in Cumbria (North-West England), 12 over the Welsh hills and 7 over the Scottish mountains. Observed patterns of orographic rainfall averaged over all cases exhibit large amounts of orographic enhancement over the first hills encountered by the south-westerly low-level flow in each region.
The operational 1.5 km resolution forecasts produced very realistic looking rainfall patterns with a correlation coefficient of 0.9 and an area-averaged rain accumulation error of less than 2%. Increasing the horizontal model grid spacing decreases the amount of rain produced over the hills, thereby decreasing forecast accuracy. The mean area-averaged rain accumulations are under predicted by 11 to 24% at 12 km resolution and by 33 to 48% at 40 km resolution, with the largest under predictions occurring over the Lake District. Changes due to using smoother orography while everything else remained unchanged (including the model grid spacing) were found to be 10% for the 12 km resolution forecasts and 24% for the 40 km resolution forecasts, based on simulations from a single case study over the Lake District. It was also found that ignoring the effect of raindrop advection results in too much rain falling on the windward slopes and not enough falling on the leeslopes.
The Kinematic Driver Model (KiD) has recently been developed at the Met Office. The flow is prescribed using a linear model of two-dimensional flow over a Witch of Agnesi hill. The KiD model then allows advective transport and particle sedimentation while avoiding the complexity caused by feedbacks between dynamics and microphysics. The relative contribution of various microphysical processes to the orographic surface rain enhancement has been investigated for a range of hill heights and widths.