Downscaling Taiwan precipitation with a residual deep learning approach

被引:0
|
作者
Hsu, Li-Huan [1 ]
Chiang, Chou-Chun [1 ]
Lin, Kuan-Ling [1 ]
Lin, Hsin-Hung [1 ]
Chu, Jung-Lien [1 ]
Yu, Yi-Chiang [1 ]
Fahn, Chin-Shyurng [2 ]
机构
[1] Natl Sci & Technol Ctr Disaster Reduct, New Taipei City 231007, Taiwan
[2] Natl Taiwan Univ Sci & Technol, Dept Comp Sci & Informat Engn, Taipei 106335, Taiwan
关键词
Downscaling; Super-resolution; Deep learning; Residual network; Precipitation forecasting; RAINFALL;
D O I
10.1186/s40562-024-00340-y
中图分类号
P [天文学、地球科学];
学科分类号
07 ;
摘要
In response to the growing demand for high-resolution rainfall data to support disaster prevention in Taiwan, this study presents an innovative approach for downscaling precipitation data. We employed a hierarchical architecture of Multi-Scale Residual Networks (MSRN) to downscale rainfall from a coarse 0.25-degree resolution to a fine 0.0125-degree resolution, representing a substantial challenge due to a resolution increase of over 20 times. Our results demonstrate that the hierarchical MSRN outperforms both the one-step MSRN and linear interpolation methods when reconstructing high-resolution daily rainfall. It surpasses the linear interpolation method by 15.1 and 9.1% in terms of mean absolute error and root mean square error, respectively. Furthermore, the hierarchical MSRN excels in accurately reproducing high-resolution rainfall for various rainfall thresholds, displaying minimal biases. The threat score (TS) highlights the hierarchical MSRN's capability to replicate extreme rainfall events, achieving TS scores exceeding 0.54 and 0.46 at rainfall thresholds of 350 and 500 mm per day, outperforming alternative methods. This method is also applied to an operational global model, the ECMWF's daily rainfall forecasts over Taiwan. The evaluation results indicate that our approach is effective at improving rainfall forecasts for thresholds greater than 100 mm per day, with more significant improvement for the 1- to 3-day lead forecast. This approach also offers a realistic visual representation of fine-grained rainfall distribution, showing promise for making significant contributions to disaster preparedness and weather forecasting in Taiwan.
引用
收藏
页数:13
相关论文
共 50 条
  • [1] A Generative Deep Learning Approach to Stochastic Downscaling of Precipitation Forecasts
    Harris, Lucy
    McRae, Andrew T. T.
    Chantry, Matthew
    Dueben, Peter D.
    Palmer, Tim N.
    [J]. JOURNAL OF ADVANCES IN MODELING EARTH SYSTEMS, 2022, 14 (10)
  • [2] On the modern deep learning approaches for precipitation downscaling
    Kumar, Bipin
    Atey, Kaustubh
    Singh, Bhupendra Bahadur
    Chattopadhyay, Rajib
    Acharya, Nachiketa
    Singh, Manmeet
    Nanjundiah, Ravi S.
    Rao, Suryachandra A.
    [J]. EARTH SCIENCE INFORMATICS, 2023, 16 (2) : 1459 - 1472
  • [3] Deep Learning for Daily Precipitation and Temperature Downscaling
    Wang, Fang
    Tian, Di
    Lowe, Lisa
    Kalin, Latif
    Lehrter, John
    [J]. WATER RESOURCES RESEARCH, 2021, 57 (04)
  • [4] On the modern deep learning approaches for precipitation downscaling
    Bipin Kumar
    Kaustubh Atey
    Bhupendra Bahadur Singh
    Rajib Chattopadhyay
    Nachiketa Acharya
    Manmeet Singh
    Ravi S. Nanjundiah
    Suryachandra A. Rao
    [J]. Earth Science Informatics, 2023, 16 : 1459 - 1472
  • [5] A Deep Learning-Based Downscaling Method Considering the Impact on Typhoons to Future Precipitation in Taiwan
    Lin, Shiu-Shin
    Zhu, Kai-Yang
    Wang, Chen-Yu
    [J]. ATMOSPHERE, 2024, 15 (03)
  • [6] PreciDBPN: A customized deep learning approach for hourly precipitation downscaling in eastern China
    Xia, Hanmeng
    Wang, Kaicun
    [J]. Atmospheric Research, 2024, 311
  • [7] Customized deep learning for precipitation bias correction and downscaling
    Wang, Fang
    Tian, Di
    Carroll, Mark
    [J]. GEOSCIENTIFIC MODEL DEVELOPMENT, 2023, 16 (02) : 535 - 556
  • [8] Deep Precipitation Downscaling
    Yu, Tingzhao
    Kuang, Qiuming
    Zheng, Jiangping
    Hu, Junnan
    [J]. IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2022, 19
  • [9] Convolutional Residual-Attention: A Deep Learning Approach for Precipitation Nowcasting
    Yan, Qing
    Ji, Fuxin
    Miao, Kaichao
    Wu, Qi
    Xia, Yi
    Li, Teng
    [J]. ADVANCES IN METEOROLOGY, 2020, 2020
  • [10] Deep-learning-based downscaling of precipitation in the middle reaches of the Yellow River using residual-based CNNs
    Fu, He
    Guo, Jianing
    Deng, Chenguang
    Liu, Heng
    Wu, Jie
    Shi, Zhengguo
    Wang, Cailing
    Xie, Xiaoning
    [J]. QUARTERLY JOURNAL OF THE ROYAL METEOROLOGICAL SOCIETY, 2024, : 3290 - 3304