SENSITIVITY OF TOTAL MOUNTAIN DRAG TO MODEL RESOLUTION FOR MAP Case Siudies

A fundameatal requirement of any parametrization of sub-gridscale orography (SSO) is that the total (resolved plus pa,rametrized) drag on a mountain rarge should be independent of the model horizontal resolution. However, parametrizations are typically developed and tested (ar:d inevitably tuned) for a single model resolution and so this issue tends not to be addressed. In ihis study, this issue is addressed by irvesl,igating the behaviour of the Met Ofrce Unified Model {UM) flow-blocking and gravity-wave drag scheme (herea,fter the SSO drag scheme) and boundary layer scheme (incorporating a,n orographic rolghness scheme, herea.fter the OR scheme) in UM simulations at horizontal resolutions ranging between 60 km and 2 km. At 60 km resolution most of the toial orographic drag must be parametrized, whilst at 2 km resolution most of the total drag should be resolved. This paper builds on the Smith et al. (paper A.9, these proceedings) resolved drag convergence study, with the primary focus here being to aesess the degree to which the tctal mountain drag is independent of the model resolution. Thus, as in the Smith et aJ. sludy, the UM has been used io simulate seven different MAP case, although here we concentrate on ihe 8l11l99IOP15 case. The two main questions we aim to answer are:-