Preprints
https://doi.org/10.5194/egusphere-2023-1775
https://doi.org/10.5194/egusphere-2023-1775
18 Sep 2023
 | 18 Sep 2023
Status: this preprint is open for discussion.

TemDeep: A Self-Supervised Framework for Temporal Downscaling of Atmospheric Fields at Arbitrary Time Resolutions

Liwen Wang, Qian Li, Qi Lv, Xuan Peng, and Wei You

Abstract. Numerical forecast products with high temporal resolution are crucial tools in atmospheric studies, allowing for accurate identification of rapid transitions and subtle changes that may be missed by lower-resolution data. However, the acquisition of high-resolution data is limited due to excessive computational demands and substantial storage needs in numerical models. Current deep learning methods for statistical downscaling still require massive ground truth with high temporal resolution for model training. In this paper, we present a self-supervised framework for downscaling atmospheric variables at arbitrary time resolutions by imposing a temporal coherence constraint. Firstly, we construct an encoder-decoder structured temporal downscaling network, and then pretrain this downscaling network on a subset of data that exhibits rapid transitions and is filtered out based on a composite index. Subsequently, this pretrained network is utilized to downscale the fields from adjacent time periods and generate the field at the middle time point. By leveraging the temporal coherence inherent in meteorological variables, the network is further trained based on the difference between the generated field and the actual middle field. To track the evolving trends in meteorological system movements, a flow estimation module is designed to assist with generating interpolated fields. Results show that our method can accurately recover evolution details superior to other methods, reaching 53.7 % in the restoration rate on the test set. In addition, to avoid generating abnormal values and guide the model out of local optima, two regularization terms are integrated into the loss function to enforce spatial and temporal continuity, which further improves the performance by 7.6 %.

Liwen Wang, Qian Li, Qi Lv, Xuan Peng, and Wei You

Status: open (extended)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2023-1775', Anonymous Referee #1, 16 Oct 2023 reply
  • CEC1: 'Executive Editor comment on egusphere-2023-1775', Astrid Kerkweg, 23 Oct 2023 reply
Liwen Wang, Qian Li, Qi Lv, Xuan Peng, and Wei You
Liwen Wang, Qian Li, Qi Lv, Xuan Peng, and Wei You

Viewed

Total article views: 329 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
237 79 13 329 7 9
  • HTML: 237
  • PDF: 79
  • XML: 13
  • Total: 329
  • BibTeX: 7
  • EndNote: 9
Views and downloads (calculated since 18 Sep 2023)
Cumulative views and downloads (calculated since 18 Sep 2023)

Viewed (geographical distribution)

Total article views: 319 (including HTML, PDF, and XML) Thereof 319 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 24 Feb 2024
Download
Short summary
Our research presents a novel deep learning approach called "TemDeep" for downscaling atmospheric variables at arbitrary time resolutions based on temporal coherence. Results show that our method can accurately recover evolution details superior to other methods, reaching 53.7 % in the restoration rate. Our findings are important for advancing weather forecasting models, enabling more precise and reliable predictions to support disaster preparedness, agriculture, and sustainable development.