This paper investigated several approaches to remapping and combining multiple-radar reflectivity fields onto a unified 3D Cartesian grid with high spatial and temporal resolutions, and analyzed systematic observation...This paper investigated several approaches to remapping and combining multiple-radar reflectivity fields onto a unified 3D Cartesian grid with high spatial and temporal resolutions, and analyzed systematic observation differences among multiple radars. The remapping approaches were evaluated by inspecting the spatial consistency of the reflectivity fields on vertical and horizontal cross sections on the equidistant line of radars, and the intensity change of 1-h accumulated precipitation before and after interpolation. The combining approaches were evaluated by continuity examination. The results show that for remapping schemes, the vertical interpolation with nearest neighbor on the range-azimuth plane is the most reasonable scheme that provides consecutive reflectivity fields and retains the high-resolution structure comparable to that of the raw data; for multiple-radar data mosaics, the distance-exponential-weighted mean scheme provides spatially consistent reflectivity mosaics. The mosaics can mitigate various problems caused by the radar beam geometry such as the cone of silence.展开更多
基金the National Major Basic Researched Programe of China(2004CB418305)the National Natural Foundation of China(40375008)
文摘This paper investigated several approaches to remapping and combining multiple-radar reflectivity fields onto a unified 3D Cartesian grid with high spatial and temporal resolutions, and analyzed systematic observation differences among multiple radars. The remapping approaches were evaluated by inspecting the spatial consistency of the reflectivity fields on vertical and horizontal cross sections on the equidistant line of radars, and the intensity change of 1-h accumulated precipitation before and after interpolation. The combining approaches were evaluated by continuity examination. The results show that for remapping schemes, the vertical interpolation with nearest neighbor on the range-azimuth plane is the most reasonable scheme that provides consecutive reflectivity fields and retains the high-resolution structure comparable to that of the raw data; for multiple-radar data mosaics, the distance-exponential-weighted mean scheme provides spatially consistent reflectivity mosaics. The mosaics can mitigate various problems caused by the radar beam geometry such as the cone of silence.