The High-Altitude Southern-Scandinavian Mountain Wave of 14 January 2003

Based on an unusually large complement of observations and modeling, we review the middle-atmospheric properties of a long-wavelength mountain wave that formed over the southern Scandinavian Mountains on 14 January 2003. This wave was predicted several days in advance by forecast models run to support aircraft flight planning for the second SAGE III Ozone Loss and Validation Experiment (SOLVE II). These forecasts predicted that the wave would propagate into the lower stratosphere and form polar stratospheric clouds (PSCs), motivating DC-8 underflights to probe these predicted wave-induced PSCs using onboard aerosol lidars. This PSC-producing mountain wave was subsequently identified in radiosonde data, and in swath-scanned satellite radiances acquired by the Advanced Microwave Sounding Unit (AMSU-A) instruments (deployed at the time on 4 different satellites) and by the Atmospheric Infrared Sounder (AIRS) on Aqua. The satellite observations show the wave propagating right though the stratosphere to an altitude of at least 1 hPa (the highest AMSU-A/AIRS channel). This wave imagery reveals a remarkably abrupt and major change in the wave's horizontal phase structure above 10 hPa altitude. This peculiar feature is reproduced using a linear Fourier-ray model solution that is forward modeled through the vertical weighting functions of relevant AMSU-A channels. It originates due to an intrinsically three-dimensional ship wave-like mountain wave pattern, emanating from the elliptical orography of southern Scandinavia, being distorted (refracted) by a stratospheric horizontal wind vector that abruptly rotates anticlockwise with height (backs) at 10-20 hPa, a dynamical process analogous to the “asymptotic wake” phenomenon originally predicted by Shutts [1998]. The asymptotic wake waves in the northern wing progressively dissipate throughout the stratosphere in response to this backing stratospheric flow, whereas waves in the southern wing propagate rapidly into the mesosphere. The detailed three-dimensional observations and modeling allow the vertical flux of horizontal momentum density of waves in the southern wing to be accurately estimated at ~300mPa, a large flux likely to induce huge local mean-flow accelerations when deposited in the mesosphere. We show that high-resolution high-altitude global forecast model runs resolve these waves (though underestimate their amplitude) and reproduce vigorous wave breaking in the mesosphere. We also discuss how various subgrid-scale orographic gravity wave drag parameterizations perform in modeling the middle atmospheric wave drag from this particular event.