Application of the multigrid 3D variation method to a combination of aircraft observations and bogus data for Typhoon Nida (2016)

Abstract

Extremely inhomogeneous aircraft observations were obtained by the Hong Kong Observatory when Typhoon Nida (2016) approached South China. However, only part of Typhoon Nida was observed due to limitations of the flight range. To investigate an effective scheme for assimilating inhomogeneous aircraft observations, a multigrid 3D variation (m3DVAR) system, with the assistance of a bogus vortex, was employed. The arithmetic mean was used to reduce the representative error of the turbulence scale before assimilation. Track and intensity forecasts were improved by assimilating aircraft observations and bogus data. The assimilation of aircraft observations could play an important role in the improvement of intensity forecasts for Nida because adding bogus data is of help to propagate the sparse and inhomogeneous information in the m3DVAR system. The assimilation of pressure (horizontal wind) was also found mainly to contribute to the large magnitude (sophisticated distribution) of increments.