The most objective way to overcome the limitation of numerical weather prediction model is to represent the uncertainty of prediction by introducing probabilistic forecast. The uncertainty of numerical weather prediction system is occurred by parameterization of unresolved scale and energy losses from the sub-scale physical processes etc. In this study, we focused on model error, and performed ensemble forecast to represent model uncertainty. By applying multi-physics scheme (PHYS) and stochastic kinetic energy backscatter scheme (SKEBS) to typhoon Rusa (2002), we verified the performance of two schemes. Both ensemble mean forecast of typhoon track were improved compared to control run. The results using PHYS was improved by 28% and the results using SKEBS was improved by7%. Both of ensemble mean errors wereincreased rapidly when forecast time 84 hours, and both of ensemble spread wereincreased gradually during time integration. The results using SKEBS represented model error well duringforecast time 96 hours, and after that it tended to simulate as an under-dispersive way. The model simulationusing PHYS overestimated the ensemble mean error during time integration, andrepresented well at the forecast time 120 hours. The moving speedusing PHYS was closest to the best track, especially after landfall. In the sensitivity tests onthe model uncertainty ofSKEBS, ensemble mean forecast was sensitive to physics parameterization. By changing the value of SKEBS forcing parameter, the results of ensemble spread, ensemble mean errors, and moving speed were improved by default experiment.
PDF
download
878987797
028-85220240
