Narrow Results

This study explores the temporal precipitation organization of 126 landfalling typhoons around Japan during 2006–2019. The internal structure particularly the cell sizes and spell durations of precipitation induced by these 126 typhoons are investigated from the Radar Automated Meteorological Data Acquisition System (Radar-AMeDAS) observed hourly precipitation dataset. The best track data from the Regional Specialized Meteorological Center (RSMC) Tokyo are utilized to identify the typhoon locations. We first stratified the independent precipitation cell sizes and precipitation spell durations with intensity exceeding various thresholds into different bins. Then the frequency distribution of the precipitation sizes and durations in each bin are computed. Our results indicate that the occurrence of typhoon induced heavier precipitation is higher compared to the typhoon induced lighter precipitation. The typhoon induced heavier precipitation over Japan last up to a day, while the lighter precipitation last about 12 hours. The long-lived precipitation cases are also noticed that last up to 2 days, but they don’t occur so frequently. We analyzed the spell durations over different regions of Japan with various radii from the typhoon center starting from 10 km to 300 km. The results also indicate that the pattern of spell durations are mostly same over all the regions within 300 km radius.

Atmospheric motion vectors (AMVs) have produced positive impacts on global weather forecasts, but few studies have evaluated the impacts of AMVs data from Fengyun (FY) geostationary satellite series, especially from FY-2G and FY-4A, on typhoon forecasts in a regional model. In this study, the AMVs data of FY-2G and FY-4A were compared and evaluated by pre-processing methods such as height assignment, quality control, channel combination and thinning. Typhoon Haishen (No.10 super typhoon in 2020) was taken as an example. The AMVs data of the two satellites were assimilated by using 3DVAR provided by WRFDA and simulated by the WRF model to evaluate the forecast results of the two satellites, respectively. The results show that the AMV data from FY-4A are better overall than those from FY-2G, with smaller RMSEs and biases for full wind speeds. On the other hand, assimilation of AMVs data improves the forecasts of environmental fields, resulting in the simulated track closer to the best track. Another experiment shows that the assimilation of AMVs data has a good impact on precipitation prediction In general, the assimilation of FY-2G and FY-4A AMV data has a relatively positive impact on typhoon prediction, and the AMVs data combined with multiple channels can provide better prediction.