English
 
Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
 
 
DownloadE-Mail
  Physically constrained generative adversarial networks for improving precipitation fields from Earth system models

Hess, P., Drüke, M., Petri, S., Strnad, F., Boers, N. (2022): Physically constrained generative adversarial networks for improving precipitation fields from Earth system models. - Nature Machine Intelligence, 4, 10, 828-839.
https://doi.org/10.1038/s42256-022-00540-1

Item is

Files

show Files
hide Files
:
Hess etal_s42256-022-00540-1.pdf (Any fulltext), 4MB
 
File Permalink:
-
Name:
Hess etal_s42256-022-00540-1.pdf
Description:
-
Visibility:
Private
MIME-Type / Checksum:
application/pdf
Technical Metadata:
Copyright Date:
-
Copyright Info:
-
License:
-

Locators

show
hide
Locator:
https://doi.org/10.5281/zenodo.4700270 (Supplementary material)
Description:
-

Creators

show
hide
 Creators:
Hess, Philipp1, Author              
Drüke, Markus1, Author              
Petri, Stefan1, Author              
Strnad, Felix1, Author              
Boers, Niklas1, Author              
Affiliations:
1Potsdam Institute for Climate Impact Research, Potsdam, ou_persistent13              

Content

show
hide
Free keywords: -
 Abstract: Precipitation results from complex processes across many scales, making its accurate simulation in Earth system models (ESMs) challenging. Existing post-processing methods can improve ESM simulations locally but cannot correct errors in modelled spatial patterns. Here we propose a framework based on physically constrained generative adversarial networks to improve local distributions and spatial structure simultaneously. We apply our approach to the computationally efficient CM2Mc–LPJmL ESM. Our method outperforms existing ones in correcting local distributions and leads to strongly improved spatial patterns, especially regarding the intermittency of daily precipitation. Notably, a double-peaked Intertropical Convergence Zone, a common problem in ESMs, is removed. Enforcing a physical constraint to preserve global precipitation sums, the generative adversarial network can generalize to future climate scenarios unseen during training. Feature attribution shows that the generative adversarial network identifies regions where the ESM exhibits strong biases. Our method constitutes a general framework for correcting ESM variables and enables realistic simulations at a fraction of the computational cost.

Details

show
hide
Language(s): eng - English
 Dates: 2022-09-012022-10-032022-10
 Publication Status: Finally published
 Pages: 15
 Publishing info: -
 Table of Contents: -
 Rev. Type: Peer
 Identifiers: DOI: 10.1038/s42256-022-00540-1
PIKDOMAIN: RD4 - Complexity Science
PIKDOMAIN: RD1 - Earth System Analysis
Research topic keyword: Atmosphere
Regional keyword: Global
Model / method: Machine Learning
Model / method: POEM
Organisational keyword: FutureLab - Artificial Intelligence in the Anthropocene
MDB-ID: No MDB - stored outside PIK (see DOI)
 Degree: -

Event

show

Legal Case

show

Project information

show

Source 1

show
hide
Title: Nature Machine Intelligence
Source Genre: Journal, SCI, Scopus
 Creator(s):
Affiliations:
Publ. Info: -
Pages: - Volume / Issue: 4 (10) Sequence Number: - Start / End Page: 828 - 839 Identifier: CoNE: https://publications.pik-potsdam.de/cone/journals/resource/nature-machine-intelligence
Publisher: Nature