Articles | Volume 17, issue 9
https://doi.org/10.5194/gmd-17-3839-2024
https://doi.org/10.5194/gmd-17-3839-2024
Model description paper
 | 
14 May 2024
Model description paper |  | 14 May 2024

DEUCE v1.0: a neural network for probabilistic precipitation nowcasting with aleatoric and epistemic uncertainties

Bent Harnist, Seppo Pulkkinen, and Terhi Mäkinen

Download

Interactive discussion

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2023-1100', Anonymous Referee #1, 25 Sep 2023
    • AC1: 'Reply on RC1', Bent Harnist, 06 Nov 2023
  • RC2: 'Comment on egusphere-2023-1100', Anonymous Referee #2, 09 Oct 2023
    • AC2: 'Reply on RC2', Bent Harnist, 06 Nov 2023
    • AC3: 'Supplement for the reply on RC2', Bent Harnist, 07 Nov 2023

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload
AR by Bent Harnist on behalf of the Authors (13 Nov 2023)  Author's response   Author's tracked changes   Manuscript 
ED: Referee Nomination & Report Request started (09 Jan 2024) by Shu-Chih Yang
RR by Jatan Buch (16 Jan 2024)
ED: Publish subject to minor revisions (review by editor) (27 Feb 2024) by Shu-Chih Yang
AR by Bent Harnist on behalf of the Authors (08 Mar 2024)  Author's response   Author's tracked changes   Manuscript 
ED: Publish as is (24 Mar 2024) by Shu-Chih Yang
AR by Bent Harnist on behalf of the Authors (26 Mar 2024)

Post-review adjustments

AA: Author's adjustment | EA: Editor approval
AA by Bent Harnist on behalf of the Authors (03 May 2024)   Author's adjustment   Manuscript
EA: Adjustments approved (09 May 2024) by Shu-Chih Yang
Download
Short summary
Probabilistic precipitation nowcasting (local forecasting for 0–6 h) is crucial for reducing damage from events like flash floods. For this goal, we propose the DEUCE neural-network-based model which uses data and model uncertainties to generate an ensemble of potential precipitation development scenarios for the next hour. Trained and evaluated with Finnish precipitation composites, DEUCE was found to produce more skillful and reliable nowcasts than established models.