Submitted as: development and technical paper
01 Sep 2022
Submitted as: development and technical paper | 01 Sep 2022
Status: this preprint is currently under review for the journal GMD.

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.

Fang Wang et al.

Status: open (until 27 Oct 2022)

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 reply

Fang Wang et al.

Fang Wang et al.


Total article views: 358 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
199 155 4 358 2 3
  • HTML: 199
  • PDF: 155
  • XML: 4
  • Total: 358
  • BibTeX: 2
  • EndNote: 3
Views and downloads (calculated since 01 Sep 2022)
Cumulative views and downloads (calculated since 01 Sep 2022)

Viewed (geographical distribution)

Total article views: 269 (including HTML, PDF, and XML) Thereof 269 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 28 Sep 2022
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.