Multimodal Semisupervised Deep Graph Learning for Automatic Precipitation Nowcasting

Figure 1

Motivation and the problem definition of precipitation nowcasting in this paper. In this paper, we assume that the future precipitation can be predicted via modeling the following factors: radar echo maps, air humidity images, satellite images, temperature images, and available precipitation data over the past several hours with the corresponding topographic map. In addition, we assume that if we can divide a certain region to many cells, then precipitation at different locations in the same cell can be deemed as almost the same.