|Title||Evaluation of revised parameterizations of sub-grid orographic drag.|
|Publication Type||Technical memorandum|
|Secondary Title||ECMWF Technical Memoranda|
CY31R1 of the IFS became operational on 12 September 2006, and included a Turbulent Orographic Form Drag (TOFD) scheme to compute form drag from sub-grid scale orography (SSO), a 'cutoff' or 'effective' mountain height in the computation of gravity wave drag, and a joint implicit calculation of momentum coefficients from the TOFD and SSO scheme. This technical memorandum evaluates these revised parameterizations by comparing their impact over the Himalayas and Rockies with control CY29R1 T511 and T95 winter and spring forecasts. The more physically realistic cutoff mountain height reduced the excessive deceleration of flow over the Himalayas and Rockies, and generally over areas of significant orography. Through thermal wind balance this is associated with a reduction in temperature bias over the Himalayas. The TOFD scheme, which directly parameterises the drag and redistributes it vertically, typically resulted in an increase in the near-surface wind and was shown to reduce the predominately negative 10 m wind bias over orography. The joint implicit calculation of momentum coefficients introduced some degree of dependency into the coupled TOFD and SSO processes, and in doing so reduced the time step sensitivity that existed when the schemes were evaluated independently of each other. However, when compared to the previous cycle, CY31R1 climate runs showed an increase in the winter positive zonal wind bias over winter northern hemisphere mid-latitudes, and a corresponding increase in the cold bias over the pole. This suggested that the reduction in gravity wave drag had been excessive (which had been apparent in the T95 results over the Rockies), and to remedy this the cutoff mountain height was doubled. This retuning was implemented operationally in CY32R2 on 5 June 2007.