Preprints
https://doi.org/10.5194/gmd-2022-213
https://doi.org/10.5194/gmd-2022-213
Submitted as: development and technical paper
01 Sep 2022
Submitted as: development and technical paper | 01 Sep 2022

Customized Deep Learning for Precipitation Bias Correction and Downscaling

Fang Wang1, Di Tian1, and Mark Carroll2 Fang Wang et al.
  • 1Department of Crop, Soil, and Environmental Sciences, Auburn University, Auburn, AL 36849, USA
  • 2Computational and Information Science Technology Office, NASA Goddard Space Flight Center Greenbelt, MD 20771, USA

Abstract. Systematic biases and coarse resolutions are major limitations of current precipitation datasets. Many deep learning (DL) based studies have been conducted for precipitation bias correction and downscaling. However, it is still challenging for the current approaches to handle complex features of hourly precipitation, resulting in incapability of reproducing small scale features, such as extreme events. This study developed a customized DL model by incorporating customized loss functions, multitask learning, and physically relevant covariates to bias correct and downscale hourly precipitation data. We designed six scenarios to systematically evaluate the added values of weighted loss functions, multi-task learning, and atmospheric covariates compared to the regular DL and statistical approaches. The model was trained and tested using the Modern-Era Retrospective analysis for Research and Applications version 2 (MERRA2) reanalysis and the Stage IV radar observations over northern coastal region of Gulf of Mexico. We found that all the scenarios with weighted loss functions performed notably better than the other scenarios with conventional loss functions and a quantile mapping-based approach at hourly, daily, and monthly time scales as well as extremes. Multitask learning showed improved performance on capturing hourly precipitation climatology, aggregated precipitation at daily and monthly scales, and detailed features of extreme events, while the improvement is not as large as from weighted loss functions. Accounting for atmospheric covariates further improved the model performance for capturing extreme events. We show that the customized DL model can better downscale and bias correct precipitation datasets and provide improved precipitation estimates at fine spatial and temporal resolutions where regular DL and statistical methods experiencing challenges.

Journal article(s) based on this preprint

Fang Wang et al.

Interactive discussion

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on gmd-2022-213', Anonymous Referee #1, 28 Sep 2022
    • AC1: 'Reply on RC1', Di Tian, 22 Nov 2022
  • RC2: 'Comment on gmd-2022-213', Anonymous Referee #2, 05 Oct 2022
    • AC2: 'Reply on RC2', Di Tian, 22 Nov 2022
  • RC3: 'Comment on gmd-2022-213', Anonymous Referee #3, 06 Oct 2022
    • AC3: 'Reply on RC3', Di Tian, 22 Nov 2022
  • RC4: 'Comment on gmd-2022-213', Anonymous Referee #4, 06 Oct 2022
    • AC4: 'Reply on RC4', Di Tian, 22 Nov 2022
  • RC5: 'Comment on gmd-2022-213', Anonymous Referee #5, 08 Oct 2022
    • AC5: 'Reply on RC5', Di Tian, 22 Nov 2022
  • RC6: 'Comment on gmd-2022-213', Anonymous Referee #6, 18 Oct 2022
    • AC6: 'Reply on RC6', Di Tian, 22 Nov 2022
  • CEC1: 'Comment on gmd-2022-213', Juan Antonio Añel, 25 Oct 2022
    • AC7: 'Reply on CEC1', Di Tian, 23 Nov 2022

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision
AR by Di Tian on behalf of the Authors (22 Nov 2022)  Author's response    Author's tracked changes    Manuscript
ED: Referee Nomination & Report Request started (25 Nov 2022) by Charles Onyutha
RR by Anonymous Referee #3 (03 Dec 2022)
RR by Anonymous Referee #4 (06 Dec 2022)
RR by Anonymous Referee #2 (09 Dec 2022)
RR by Anonymous Referee #1 (12 Dec 2022)
ED: Publish as is (13 Dec 2022) by Charles Onyutha

Interactive discussion

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on gmd-2022-213', Anonymous Referee #1, 28 Sep 2022
    • AC1: 'Reply on RC1', Di Tian, 22 Nov 2022
  • RC2: 'Comment on gmd-2022-213', Anonymous Referee #2, 05 Oct 2022
    • AC2: 'Reply on RC2', Di Tian, 22 Nov 2022
  • RC3: 'Comment on gmd-2022-213', Anonymous Referee #3, 06 Oct 2022
    • AC3: 'Reply on RC3', Di Tian, 22 Nov 2022
  • RC4: 'Comment on gmd-2022-213', Anonymous Referee #4, 06 Oct 2022
    • AC4: 'Reply on RC4', Di Tian, 22 Nov 2022
  • RC5: 'Comment on gmd-2022-213', Anonymous Referee #5, 08 Oct 2022
    • AC5: 'Reply on RC5', Di Tian, 22 Nov 2022
  • RC6: 'Comment on gmd-2022-213', Anonymous Referee #6, 18 Oct 2022
    • AC6: 'Reply on RC6', Di Tian, 22 Nov 2022
  • CEC1: 'Comment on gmd-2022-213', Juan Antonio Añel, 25 Oct 2022
    • AC7: 'Reply on CEC1', Di Tian, 23 Nov 2022

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision
AR by Di Tian on behalf of the Authors (22 Nov 2022)  Author's response    Author's tracked changes    Manuscript
ED: Referee Nomination & Report Request started (25 Nov 2022) by Charles Onyutha
RR by Anonymous Referee #3 (03 Dec 2022)
RR by Anonymous Referee #4 (06 Dec 2022)
RR by Anonymous Referee #2 (09 Dec 2022)
RR by Anonymous Referee #1 (12 Dec 2022)
ED: Publish as is (13 Dec 2022) by Charles Onyutha

Journal article(s) based on this preprint

Fang Wang et al.

Fang Wang et al.

Viewed

Total article views: 1,197 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
732 429 36 1,197 8 10
  • HTML: 732
  • PDF: 429
  • XML: 36
  • Total: 1,197
  • BibTeX: 8
  • EndNote: 10
Views and downloads (calculated since 01 Sep 2022)
Cumulative views and downloads (calculated since 01 Sep 2022)

Viewed (geographical distribution)

Total article views: 1,082 (including HTML, PDF, and XML) Thereof 1,082 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 25 Jan 2023
Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Short summary
Precipitation datasets suffer from biases and limited resolutions. We developed a customized deep learning (DL) model to bias correct and downscale precipitation data using radar observations. The results showed that the customized DL model can provide improved precipitation at fine resolutions where regular DL and statistical methods experiencing challenges. The new model can be used to enhance precipitation estimates, especially for capturing extremes at the smaller scale.