Weather Radar Long-sequence Product Dataset in China

Weather radar plays a crucial role in the monitoring, forecasting, and early warning of severe convective weather. With the widespread application of artificial intelligence technology in the meteorological field, the demand for high-quality, long-sequence weather radar product is becoming increasingly urgent. To support the development of domestic independent severe convective weather models, enhance the identification and monitoring technology for severe convective weather, and promote in-depth research in numerical forecasting and mechanisms, the Meteorological Observation Center of China Meteorological Administration has applied techniques such as feature analysis and recognition, multi-source data collaborative quality control, inspection evaluation, and diagnostic error correction. Through a hierarchical processing approach involving single-station quality control, network quality control, and inspection evaluation, coarse data errors are eliminated. This process has resulted in the formation of a long-sequence, high-quality, and high spatiotemporal resolution weather radar basic product dataset (V1.0) covering the years from 2011 to 2023. This dataset incorporates post-quality control base data from nationwide weather radars, along with four types of two-dimensional mosaic products: composite reflectivity, hybrid scan reflectivity, echo top height, and vertically integrated liquid water content, as well as three-dimensional mosaic products of constant altitude plan position indicator (CAPPI). The post-quality control base data amounts to roughly 200TB, encompassing data from 247 stations with a temporal resolution of approximately 6 minutes. The two-dimensional mosaic product data totals approximately 1.2TB, featuring a temporal resolution of 6 minutes. These two-dimensional mosaic products cover a horizontal spatial range from 73.0° to 135.0°E and from 12.2° to 53.2°N, with a spatial resolution of 0.01° × 0.01°. The three-dimensional mosaic product data totals approximately 27.5TB, sharing the same temporal and horizontal resolution as the two-dimensional mosaic product data. In terms of vertical spatial coverage, it spans from 0.5 km to 16 km, consisting of 24 layers. The vertical resolution is 0.5 km for altitudes up to 8 km and 1 km for altitudes above 8 km. Currently, this dataset has played a crucial role in severe convective weather model training and the development of reanalysis data in the Chinese region.