Abstract/Summary

Rainbands that form downwind of mountainous terrain are investigated using both idealised simulations and a case study. While relatively small in scale (a few tens of km across by up to 100 km in length), these bands can cause localised flooding if associated with intense precipitation over several hours due to the anchoring effect of orography. The aims are to evaluate the ability of weather forecast models to robustly simulate rainband formation at convection-permitting and higher resolution and to identify the conditions, including mountain height, wind speed and flow stability, leading to the formation of these rainbands through different mechanisms. The operational UK variable resolution (UKV) configuration of the Met Office Unified Model (MetUM) failed to represent the case study small-scale orographic rainbands over Scotland, UK, with its current resolution (l.5-km horizontal grid spacing), although increased CAPE existed in the region of rainband formation. In contrast, the rainbands were simulated when the grid spacing was decreased to 500 m and they occurred robustly in ensemble forecasts generated using boundary layer temperature perturbations. A novel convection-permitting (]-km grid spacing) moist idealised flow over isolated orography MetUM configuration was set up and compared with an equivalent dry experiment, which was validated against previous work. Rainbands could form, triggered by the mountain wave in seven of the eight cases, when the approaching flow had low non-dimensional mountain height (:S2), indicative of the flow-over or split-flow regime, provided the windspeed exceeded a threshold value (of about 8 m s-1 ). A long convective timescale (exceeding 4.6 minutes) favours rainband formation. The distance the precipitation structure propagates in the lee of the mountain decreases, leading to shorter rainbands, as mountain height increases. The need for (at least) convection-permitting resolution, together with accurate orographic and environmental representation, for accurate forecasts of small-scale, orographically-driven rainbands is highlighted by this study.