Submitted as: model experiment description paper
28 Jun 2023
Submitted as: model experiment description paper |  | 28 Jun 2023
Status: this preprint is currently under review for the journal GMD.

High resolution downscaling of CMIP6 Earth System and Global Climate Models using deep learning for Iberia

Pedro M. M. Soares, Frederico Johannsen, Daniela C. A. Lima, Gil Lemos, Virgílio Bento, and Angelina Bushenkova

Abstract. Deep learning (DL) methods have recently garnered attention from the climate change community, as an innovative approach for downscaling climate variables from Earth System and Global Climate Models (ESGCMs) with horizontal resolutions still too coarse to represent regional-to-local-scale phenomena. In the context of the Coupled Model Intercomparison Project phase 6 (CMIP6), ESGCMs simulations were conducted for the Sixth Assessment Report (AR6) of the Intergovernmental Panel on Climate Change (IPCC), at resolutions ranging from 0.70º to 3.75º. Here, four Convolutional Neural Network (CNN) architectures were evaluated for their ability to downscale, to a resolution of 0.1º, seven CMIP6 ESGCMs over the Iberian Peninsula - a known climate change hotspot, due to its increased vulnerability to projected future warming and drying conditions. The study is divided into three stages: (1) evaluating the performance of the four CNN architectures in predicting mean, minimum, and maximum temperatures, as well as daily precipitation, trained using ERA5 data, and compared with the Iberia01 observational dataset; (2) downscaling the CMIP6 ESGCMs using the trained CNN architectures and further evaluating the ensemble against Iberia01; and (3) constructing a multi-model ensemble of CNN-based downscaled projections for temperature and precipitation over the Iberian Peninsula at 0.1º resolution throughout the 21st century, under four Shared Socioeconomic Pathway (SSP) scenarios. Upon validation and satisfactory performance evaluation, the DL downscaled projections demonstrate overall agreement with the CMIP6 ESGCM ensemble in terms of temperature and precipitation projections. Moreover, the advantages of using a high-resolution DL downscaled ensemble of ESGCM climate projections are evident, offering substantial added value in representing regional climate change over Iberia. Notably, a clear warming trend is observed, consistent with previous studies in this area, with projected temperature increases ranging from 2 ºC to 6 ºC depending on the climate scenario. Regarding precipitation, robust projected decreases are observed in western and southwestern Iberia, particularly after 2040. These results may offer a new tool for providing regional climate change information for adaptation strategies based on CMIP6 ESGCMs prior to the next phase of the European branch from the Coordinated Regional Climate Downscaling Experiment (EURO-CORDEX) experiments.

Pedro M. M. Soares et al.

Status: open (extended)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • CEC1: 'Comment on gmd-2023-136', Juan Antonio Añel, 31 Jul 2023 reply
    • AC1: 'Reply on CEC1', Frederico Johannsen, 07 Aug 2023 reply
    • AC2: 'Reply on CEC1', Frederico Johannsen, 14 Sep 2023 reply
  • RC1: 'Comment on gmd-2023-136', Anonymous Referee #1, 08 Sep 2023 reply
    • AC3: 'Reply on RC1', Frederico Johannsen, 15 Sep 2023 reply

Pedro M. M. Soares et al.

Pedro M. M. Soares et al.


Total article views: 554 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
376 161 17 554 6 4
  • HTML: 376
  • PDF: 161
  • XML: 17
  • Total: 554
  • BibTeX: 6
  • EndNote: 4
Views and downloads (calculated since 28 Jun 2023)
Cumulative views and downloads (calculated since 28 Jun 2023)

Viewed (geographical distribution)

Total article views: 522 (including HTML, PDF, and XML) Thereof 522 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 03 Oct 2023
Short summary
This study uses deep learning (DL) to downscale global climate models for the Iberian Peninsula. Four DL architectures were evaluated and trained using historical climate data, and then used to downscale future projections from the global models. These show agreement with the original models and reveal a warming of 2 ºC to 6 ºC, along with decreasing precipitation in western Iberia after 2040. This approach offers key regional climate change information for adaptation strategies in the region.