[A-AS10] General Meteorology

[AAS10-16] Evaluation of JMA-GSM typhoon forecast using a new spectral cumulus parameterization in Parapiroon (2018) and Hagibis (2019)

*Yuya Baba¹, Masashi Ujiie² (1.Japan Agency for Marine-Earth Science and Technology, 2.Japan Meteorological Agency)

Keywords:cumulus parameterization, typhoon, numerical weather prediction

Typhoon forecast of Global Spectral Model of Japan Meteorological Agency (JMA-GSM, GSM hereafter) using a new spectral cumulus parameterization (spectral scheme) was evaluated using forecast experiments and the original convection scheme of GSM. The experiments were conducted for two typhoon cases, i.e., Prapiroon (2018) and Hagibis (2019). In the Prapiroon case, the spectral scheme improved both typhoon track and minimum pressure errors. In the Hagibis case, the spectral scheme much improved the minimum pressure error, but it increased the track error. The cause of error difference was analysed using an ensemble singular vector sensitivity analysis. In both cases, the spectral scheme generally improved environmental condition around the typhoon after the initialization. However, in the Hagibis case, the spectral scheme predicted a rapid minimum pressure decrease at the center of the typhoon, and this induced a pressure perturbation behind the track enhancing environmental easterly wind. This easterly wind eventually moved the typhoon westward, resulting in the track error. It is concluded that the spectral scheme can potentially improve the typhoon forecast of GSM by improving the environmental condition, but updates in creating the initial condition should be considered for further better forecast.