Spatiotemporal convolutional approach for the short-term forecast of hourly heavy rainfall probability integrating numerical weather predictions and surface observations

The hourly rainfall observations used in this project are recorded from 5718 automatic weather stations (AWS) located in the Jiangsu region in eastern China (30° -35.95°N, 116.05° -122°N. The NWP results are derived from the Precision Weather Analysis and Forecasting System (PWAFS), which is the operational NWP model run by the Jiangsu Meteorological Bureau, China. It consists of the WRF-ARW model version 3.9.1 and a three-dimensional variational data assimilation system. example: import numpy as np lat_min = 30 lat_max = 35.95 lon_min = 116.05 lon_max = 122 lat1d = np.linspace(lat_min, lat_max, 120) lon1d = np.linspace(lon_min, lon_max, 120) grid_lat, grid_lon = np.meshgrid(lat1d, lon1d, indexing=\"ij\") np.load(\"/AWS/xxx/xxx/xxx.npy\") ## UTC time