Preprints
https://doi.org/10.5194/egusphere-2024-1022
https://doi.org/10.5194/egusphere-2024-1022
29 Apr 2024
 | 29 Apr 2024
Status: this preprint is open for discussion.

Using Multi-Head Attention Deep Neural Network for Bias Correction and Downscaling for Daily Rainfall Pattern of a Subtropical Island

Yi-Chi Wang, Chia-Hao Chiang, Chiung-Jui Su, Ko-Chih Wang, Wan-Ling Tseng, Cheng-Ta Chen, and Hsin-Chien Liang

Abstract. This study investigates the capability of a deep learning approach, employing a multi-head attention mechanism within a deep neural network (DNN) framework, aimed at refining the bias correction and downscaling process for the fifth generation European Centre for Medium-Range Weather Forecasts reanalysis rainfall datasets to provide local-scale daily rainfall data across Taiwan, a mountainous subtropical island. Leveraging gridded 5-km daily rainfall observations across Taiwan, the proposed DNN model, the Encoder-Decoder with multi-head Attention for auxiliary channels (EDA) model, can adeptly correct biases and downscale rainfall statistics from coarse-resolution reanalysis data by incorporating auxiliary inputs, such as surface wind information, and invariant data, such as high-resolution topography data. Our evaluation, centred on the distinct seasonal rainfall characteristics of Taiwan, uses mean rainfall patterns, rainfall statistics, extreme climate indices, and their interannual variation for the rainy seasons. The findings show the EDA model's ability to correct for overestimated low-intensity rainfall and inaccurately positioned orographic rainfall in reanalysis datasets, achieving better accuracy than conventional quantile-mapping methods. Further analysis reveals the critical role of auxiliary information of surface winds used by the EDA to enhance the downscaling accuracy across various performance metrics. This study underscores the significant potential of DNN architectures for statistical bias correction and downscaling in regions with complex terrains, by effectively integrating auxiliary data to capture the interplay between synoptic and local circulations influenced by topography.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.
Yi-Chi Wang, Chia-Hao Chiang, Chiung-Jui Su, Ko-Chih Wang, Wan-Ling Tseng, Cheng-Ta Chen, and Hsin-Chien Liang

Status: open (until 24 Jun 2024)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
Yi-Chi Wang, Chia-Hao Chiang, Chiung-Jui Su, Ko-Chih Wang, Wan-Ling Tseng, Cheng-Ta Chen, and Hsin-Chien Liang
Yi-Chi Wang, Chia-Hao Chiang, Chiung-Jui Su, Ko-Chih Wang, Wan-Ling Tseng, Cheng-Ta Chen, and Hsin-Chien Liang

Viewed

Total article views: 119 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
89 26 4 119 9 2 2
  • HTML: 89
  • PDF: 26
  • XML: 4
  • Total: 119
  • Supplement: 9
  • BibTeX: 2
  • EndNote: 2
Views and downloads (calculated since 29 Apr 2024)
Cumulative views and downloads (calculated since 29 Apr 2024)

Viewed (geographical distribution)

Total article views: 131 (including HTML, PDF, and XML) Thereof 131 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 20 May 2024
Download
Short summary
Our study introduces a deep learning model, the EDA, to refine rainfall data in Taiwan. This model significantly improves bias correction by integrating surface wind and topography data, crucial in areas like Taiwan where traditional methods fall short. The EDA excels in adjusting low-intensity and misplaced rainfall, enhancing water management, agriculture, and disaster prevention. This work showcases deep learning's potential to improve climate downscaling in complex terrains.