Quantitative precipitation forecast experiments of heavy rainfall over Jeju Island on 14-16 September 2007 using the WRF model

A record-breaking heavy rainfall with a local maximum of 829.5 mm and a maximum intensity of 95 mm h^-1 occurred over Jeju Island, Korea, on 14-16 September 2007. The objective of this study is to examine the capability of the Weather Research and Forecasting (WRF) model in reproducing such a disastrous heavy rainfall. The WRF model is forced by 6-hourly National Centers for Environmental Prediction (NCEP)-Final analysis (FNL) data for 60 hours from 00 UTC 14 to 12 UTC 16 September 2007. A series of quantitative precipitation forecast experiments is carried out by inserting a bogus vortex and changing the domain size, nesting method, physics options, and horizontal grid size. The high-resolution WRF model is capable of reproducing the observed record-breaking rainfall fairly well, even though the model is initialized by a conventional global analysis. The discrepancy in the typhoon movement between the simulation and observation is moderated by inserting a bogus vortex. The quantitative predictability of heavy rainfall over Jeju Island is improved when the outmost domain is smaller than that covering the nominal East-Asian region, indicating that a correct representation of the initial structure of the typhoon is needed to provide the appropriate synoptic-scale environment associated with heavy rainfall. The nesting method of a one-way versus two-way interaction and the choice of physics package do not improve the predictability of such an extreme rainfall event; especially, changes in the cumulus parameterization and planetary boundary layer processes seriously degrade the predictability over the island, indicating realistic designs of these two physics processes are critical in improving the quantitative predictability of heavy rainfall over Korea. It is also concluded that the horizontal grid-spacing of 3-km or less is needed for the success of quantitative precipitation forecast in terms of the rainfall amount and intensity.