2019 Volume 15 Pages 107-112
High-frequency variations are excluded in the merged satellite and in-situ data global daily sea surface temperature (MGDSST) used in weather forecasting in Japan Meteorological Agency. We investigated the importance of temporal resolution on sea surface temperature (SST) when predicting winter precipitation using the Non-Hydrostatic Regional Climate Model. We used seven-day temporal smoothing to investigate the influence of temporal resolution on prediction. The Gaussian filter was used as spatial smoothing for comparison with the influence of spatial resolution. The influence of the temporal resolution of SST on monthly precipitation is smaller than that of spatial resolution. However, the influence of the temporal resolution on daily precipitation is comparable to that of spatial resolution. The temporal resolution of SST greatly affects precipitation, particularly in December, as the variations in SST are largest compared to the rest of the year. Furthermore, the winter monsoon promotes the effect of SST on winter precipitation. Our experiments using seven-day moving average smoothing indicates that the temporal resolution of the SST on precipitation become about 15 %/K under the winter monsoon.