Stratospheric gravity waves over the mountainous island of South Georgia: testing a high-resolution dynamical model with 3-D satellite observations and radiosondes

Abstract. Atmospheric gravity waves (GWs) play an important role in atmospheric dynamics but accurately representing them general circulation models (GCMs) is challenging. This especially true for orographic GWs generated by wind flow over small mountainous islands the Southern Ocean. Currently, these lie “grey zone” of global model resolution, where they are neither fully resolved nor parameterised. It expected that as GCMs approach spatial resolution current high-resolution local-area models, small-island GW sources may be without need parameterisations. But how realistic simulations compared to observations? Here, we test a (1.5 km horizontal grid, 118 vertical levels) configuration Met Office Unified Model island South Georgia (54∘ S, 36∘ W), running The island's orography well model, and real-time boundary conditions used two time periods during July 2013 June–July 2015. We compare simulated coincident 3-D satellite observations from Infrared Sounder (AIRS) on board Aqua. By carefully sampling using AIRS measurement footprints (denoted sampled hereafter), present first like-for-like comparison observed amplitudes, wavelengths directional momentum flux (GWMF) S-transform method. find timing, magnitude direction GWMF good agreement with observations, once applied model. Area-averaged zonal 2 months westward at around 5.3 5.6 mPa datasets respectively, values directly can exceed 50 mPa. However, up 35 % total actually found upwind only 17 AIRS, suggesting non-orographic underestimated our configuration. Meridional results show northward bias (∼20 %) correspond southward radiosonde measurements. Finally, one example large-amplitude (T′≈15–20 K 45 altitude) short (λH≈30–40 km) measurements excellent AIRS. suggests full-resolution T′≈45 λH≈30–40 occur reality. Our study demonstrates not simulate stratospheric application also highly effective method their validation.