Synoptic systems influence the effectiveness of spectral nudging in high-resolution simulations of extreme precipitation

Abstract

Spectral nudging (SN) is an effective numerical technique that prevents the background field from deviating excessively from the boundary conditions in regional climate modeling. This study investigates the effectiveness of SN in convection-permitting model simulations of three typical events of warm-season extreme precipitation in South Korea. The case studies show that SN improves extreme precipitation simulations by keeping synoptic circulations more consistent with observations. Trajectory analysis also shows that SN influences the moisture transport and local ascent that trigger extreme precipitation. However, the extent of the improvement depends on the synoptic condition and the domain size. The influence of SN generally increases with increasing domain size, especially for events driven by the developing upper-level trough with strong baroclinic instability compared to those influenced by surface processes. SN is also found to be more effective in events whose dominant wave scales fall between the domain size and the cut-off wavelength used for SN. Based on a series of sensitivity experiments, it is proposed that the most effective configuration for simulating extreme precipitation events is to apply SN to a domain approximately 3000 km wide.