5:15 PM - 6:45 PM
[AAS01-P01] Analysis of vertical winds using radar observations and numerical models in a hailstorm on the Sendai Plain in June 2022
Keywords:Vertical vecity, Doppler radar, Numerical weather prediction model, Hailstorm
We conducted an analysis of a hailstorm that occurred on June 2, 2022, in the vicinity of Sendai, Japan. Our study estimated the three-dimensional wind speed over the Sendai Plain and used it to evaluate the numerical outcomes of the Japan Meteorological Agency’s non-hydrostatic model (JMA-NHM) at various horizontal resolutions.
In our radar analysis, we utilized XRAIN, a dual-polarization meteorological radar data provided by the Ministry of Land, Infrastructure, Transport, and Tourism. The computations were executed at a horizontal resolution of 500 m and a vertical resolution of 300 m. We employed three radars located in Miyagi and Fukushima prefectures and performed calculations using the MUSCAT method (Yamada 2021). A quality control process was also performed for Doppler velocity data.
For numerical simulations using JMA-NHM, we performed them at three different horizontal resolutions: 2km, 1km, and 500m to investigate how simulated convection changes with different resolutions.
Both the radar analysis and all model calculations at different horizontal resolutions were successful in representing strong precipitation, exceeding a reflectivity of 40 dBZ over the Sendai Plain. In our statistical analysis, we observed significant differences depending on the horizontal resolution, especially in developed updrafts. All results depicted the development and decay of convective clouds within the analysis period.
The model results reproduce stronger vertical flows as horizontal resolution increased; particularly at a resolution of 500m where it could represent vertical flows exceeding 10 m/s. All horizontal resolutions could represent updraft development but most accurately at a resolution of 500m which also best represented statistical properties related to characteristics or variations in convective clouds. Thus, the present study suggests that a numerical weather prediction model with finer resolution can more faithfully represent convective clouds.
In our radar analysis, we utilized XRAIN, a dual-polarization meteorological radar data provided by the Ministry of Land, Infrastructure, Transport, and Tourism. The computations were executed at a horizontal resolution of 500 m and a vertical resolution of 300 m. We employed three radars located in Miyagi and Fukushima prefectures and performed calculations using the MUSCAT method (Yamada 2021). A quality control process was also performed for Doppler velocity data.
For numerical simulations using JMA-NHM, we performed them at three different horizontal resolutions: 2km, 1km, and 500m to investigate how simulated convection changes with different resolutions.
Both the radar analysis and all model calculations at different horizontal resolutions were successful in representing strong precipitation, exceeding a reflectivity of 40 dBZ over the Sendai Plain. In our statistical analysis, we observed significant differences depending on the horizontal resolution, especially in developed updrafts. All results depicted the development and decay of convective clouds within the analysis period.
The model results reproduce stronger vertical flows as horizontal resolution increased; particularly at a resolution of 500m where it could represent vertical flows exceeding 10 m/s. All horizontal resolutions could represent updraft development but most accurately at a resolution of 500m which also best represented statistical properties related to characteristics or variations in convective clouds. Thus, the present study suggests that a numerical weather prediction model with finer resolution can more faithfully represent convective clouds.