the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Brief Communication: Training of AI-based nowcasting models for rainfall early warning should take into account user requirements
Abstract. In the field of precipitation nowcasting, deep learning (DL) has emerged as an alternative to conventional tracking and extrapolation techniques. However, DL struggles to adequately predict heavy precipitation, which is essential in early warning. By taking into account specific user requirements, though, we can simplify the training task and boost predictive skill. As an example, we predict the cumulative precipitation of the next hour (instead of five minute increments), and the exceedance of thresholds (instead of numerical values). A dialogue between developers and users should identify the requirements to a nowcast, and how to consider these in model training.
- Preprint
(849 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 09 Sep 2024)
Model code and software
The RainNet2024 family of deep neural networks for precipitation nowcasting Georgy Ayzel and Maik Heistermann https://doi.org/10.5281/zenodo.12547127
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
77 | 25 | 8 | 110 | 3 | 3 |
- HTML: 77
- PDF: 25
- XML: 8
- Total: 110
- BibTeX: 3
- EndNote: 3
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1