| Title: |
Self-supervised Spatial-Temporal Learner for Precipitation Nowcasting |
| Authors: |
Li, Haotian; Siebes, Arno; Mehrkanoon, Siamak; Sub Algorithmic Data Analysis |
| Publication Year: |
2025 |
| Subject Terms: |
Precipitation nowcasting; Self-supervised learning; Spatial-temporal learning; Taverne; Artificial Intelligence; Computer Science Applications; Mathematical Physics |
| Description: |
Nowcasting, the short-term prediction of weather, is essential for making timely and weather-dependent decisions. Specifically, precipitation nowcasting aims to predict precipitation at a local level within a 6-hour time frame. This task can be framed as a spatial-temporal sequence forecasting problem, where deep learning methods have been particularly effective. However, despite advancements in self-supervised learning, most successful methods for nowcasting remain fully supervised. Self-supervised learning is advantageous for pretraining models to learn representations without requiring extensive labeled data. In this work, we leverage the benefits of self-supervised learning and integrate it with spatial-temporal learning to develop a novel model, SpaT-SparK. SpaT-SparK comprises a CNN-based encoder-decoder structure pretrained with a masked image modeling (MIM) task and a translation network that captures temporal relationships among past and future precipitation maps in downstream tasks. We conducted experiments on the NL-50 dataset to evaluate the performance of SpaT-SparK. The results demonstrate that SpaT-SparK outperforms existing baseline supervised models, such as SmaAt-UNet, providing more accurate nowcasting predictions. |
| Document Type: |
book part |
| File Description: |
application/pdf |
| Language: |
English |
| Relation: |
https://dspace.library.uu.nl/handle/1874/483033 |
| Availability: |
https://dspace.library.uu.nl/handle/1874/483033 |
| Rights: |
info:eu-repo/semantics/OpenAccess |
| Accession Number: |
edsbas.A9A88A34 |
| Database: |
BASE |