A case study was conducted for a severe wintertime dust event that occurred on December 29, 2007 in Korea. Three different dust emission schemes, namely, those of Marticorena and Bergametti (1995), Lu and Shao (1999), and Shao (2004) (hereinafter, referred to MB, LS, and S04 schemes, respectively) were implemented in Weather Research Forecast with Chemistry (WRF/Chem) to assess their performance in simulating wintertime Asian dust events. WRF/Chem simulation reproduces dust emission over Mongolia on December 27 and 28, 2007 and the onset timing of the dust event in Korea well. There is, however, a huge difference among the estimated dust emission amounts for the three schemes; the dust concentration derived by MB scheme is 6 times larger than that from LS scheme. The three schemes overestimate dust concentrations when comparing to observed surface-level $PM_{10}$ mass concentrations in Ganghwa, Seoul, and Yeongwol. This discrepancy is thought to be mainly due to the overestimated model winds and the surface condition such as snow cover fraction, which did not adequately represent the real conditions. Considering frozen soil effect on dust emission, the model results are comparable with observation data: it is important to consider frozen soil in simulating wintertime dust events.