Gravity Waves in Titan's Atmosphere

Abstract

Scintillations (high frequency variations) observed in the radio signal during the occultation of Voyager 1 by Titan (Hinson and Tyler, 1983, Icarus 54, 337-352) provide information concerning neutral atmospheric density fluctuations on scales of hundreds of meters to a few kilometers. Those seen at altitudes higher than 25 km above the surface were interpreted by Hinson and Tyler as being caused by linear, freely propagating (energy-conserving) gravity waves, but this interpretation was found to be inconsistent with the scintillation data below the 25-km altitude level. Here an attempt is made to interpret the entire scintillation profile between the surface and the 90-km altitude level in terms of gravity waves generated at the surface. Numerical calculations of the density fluctuations caused by two-dimensional, nonhydrostatic, finite-amplitude gravity waves propagating vertically through Titan's atmosphere are performed to produce synthetic scintillation profiles for comparison with the observations. The numerical model accurately treats the effects of wave transience, nonlinearity, and breakdown due to convective instability in the overturned part of the wave. The results indicate that wave phase speeds could not have exceeded 2 m sec-1 and must have been oriented in the meridional (north-south) direction if there were strong zonal winds on Titan. The high-altitude scintillation data were accurately recovered with a freely propagating wave solution, confirming the analytic model of Hinson and Tyler. The amplitude, phase speed, and horizontal and vertical wavelengths of the freely propagating waves are consistent with their having been generated in the convective boundary layer at Titan's surface. It is found that the low-altitude scintillation data can be fit by a model where a component of the gravity waves becomes convectively unstable and breaks near the 15 km level. A definitive value for the amplitude of the breaking wave cannot be obtained without better knowledge of its horizontal wave-length and phase speed. If the breaking wave had the same horizontal wavelength, approximately 4 km, as the freely propagating waves, then its vertical perturbation velocity near the surface would have been ∼1.4 cm sec-1, and its phase speed would have been ∼20 cm sec-1. This component could also have been generated by convection near the surface. Alternatively, it is estimated that the breaking wave could have been forced by topographic relief of 60-300 meters. The large-scale structure of the observed scintillation profile in the entire altitude range between 5 and 85 km can be simulated by a model where the freely propagating and breaking waves are forced at the surface simultaneously. Further analysis of the Voyager 1 Titan low-altitude scintillation data, using inversion theory appropriate for strong scattering, could potentially remove some of the ambiguities remaining in this analysis and allow a better determination of the strength and source of the waves.