ISSN 1000-0526
CN 11-2282/P
CN 11-2282/P
Archive > Volume 42 Issue 10 > 2016,42(10):1161-1175. DOI:10.7519/j.issn.1000-0526.2016.10.001 Prev Next
Stochastic Parameterization Toward Model Uncertainty for the GRAPES Mesoscale Ensemble Prediction System
- Article
- Figures
- Metrics
- Preview PDF
- Reference
- Related
- Cited by
- Materials
Abstract:
In order to describe the model uncertainly of the GRAPES_MEPS (Global/Regional Assimilation and Prediction System, Mesoscale Ensemble Prediction System), we used Stochastically Perturbed Parameterization Tendencies (SPPT) scheme in this system. The random field which is described with first order Markov chain has a timerelated characteristics and Gaussian distribution, and also has a continuous and smooth horizontal structure since it is a combination through the spectral transform. This paper presents experiments on GRAPES_MEPS ensemble forecasts based on SPPT scheme, with a series of sensitivity tests on random perturbation amplitude and timescale correlation coefficient carried out. Verification on ensemble forecasts is also implemented, and the impact of SPPT scheme on precipitation prediction is analyzed. The experimental results indicate that SPPT scheme can improve forecasting skills of GRAPES_MEPS system and reduce the false negative rate to a certain extent, and improve the prediction of heavy rain forecast skill significantly. Through the sensitivity tests we found that the effect of SPPT scheme for GRAPES_MEPS system is related to the amplitude of the random perturbation field and time correlation scale, more suitable parameters should be determined through sensitivity experiments.
Keywords:
Project Supported:
Clc Number:
Mobile website
