Precipitaion observation and forecast in North China in 2022 by numerical model and deep learning model

This dataset contains multi-source grid precipitation data of 2022 for North China (35°N ~ 44.55°N, 112°E ~ 121.55°E) with a spatial resolution of 0.05°×0.05° (5km).CMPAS. CMPAS is a three-source fusion precipitation product developed by the National Meteorological Information Center of China Meteorological Administration based on ground automatic station, satellite and radar in China, with a time resolution of 1h and a spatial resolution of 0.05°×0.05° (5km). Using CMPAS as the true value of precipitation field.Topography. Topography contains topography height, slope, longitude and latitude information about North China.EMCWF. European Centre for Medium-Range Weather Forecasts High resolution global model forecast.CMA-SH9. The Mesoscale forecast of East China Regional Numerical Center.CMA-3KM. The high-resolution regional numerical forecast independently developed by the Numerical Prediction Center of the China Meteorological Administration.FRNet. Precipitation predicted of deep learning model without using generate adversarial strategies.GFRNet. The precipitation inferred by GFRNet model which is trained with generate adversarial strategies.In time, the initial prediction of 00UTC and 12UTC was selected, and the prediction time of 24h was retained. In space, bilinear interpolation algorithm was used to interpolate the prediction of the numerical model onto a uniform grid of 0.05°×0.05°, corresponding to the size of the target region of 192 ×192. The regional scope, data structure of FRNet and GFRNet were consistent with NWPs, and the 3h cumulative precipitation forecast at the 8 forecast periods of the 3rd, 6th, 9th, 12th, 15th, 18th, 21st and 24h of each cycle was retained.