Rampal et al. (2026) Downscaling with AI reveals the large role of internal variability in fine-scale projections of climate extremes

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Climate Dynamics
• Year: 2026
• Date: 2026-07-23
• Authors: Neelesh Rampal, Peter B. Gibson, Steven C. Sherwood, Laura E. Queen, Hamish Lewis, Gab Abramowitz
• DOI: 10.1007/s00382-026-08269-y

Research Groups

• Climate Change Research Institute (CCRI)
• Department of Atmospheric Science (DAS)

Short Summary

This study introduces a generative-AI approach to enhance the scope of regional downscaling, enabling characterization of fine-scale future climate changes and rare extremes. The research reveals that internal variability dominates uncertainty in projected precipitation extremes.

Objective

• Investigate the feasibility of using generative-AI for dynamical downscaling to expand ensemble sizes and improve predictions of fine-scale future climate changes.

Study Configuration

• Spatial Scale: Regional (New Zealand)
• Temporal Scale: End-of-century projections

Methodology and Data

• Models used: Generative-AI approach combined with traditional dynamical downscaling methods
• Data sources: Reanalysis datasets, satellite observations

Main Results

• Projected daily precipitation extremes increase with climate change at rates of $$5.6\%^{\circ }\textrm{C}^{-1}$$ for annual extremes and $$9\%^{\circ }\textrm{C}^{-1}$$ for 100-year extremes.
• Internal variability dominates uncertainty in projected precipitation extremes, increasing with warming much faster than the mean.

Contributions

• The study provides new insights into the role of internal variability in shaping fine-scale precipitation changes under climate change.
• The generative-AI approach enables characterization of rare and extreme events that were previously inaccessible to traditional downscaling methods.

Funding

• This research was funded by the New Zealand Ministry for Business, Innovation and Employment (MBIE) through grant number UOAX1704.

Citation

@article{Rampal2026Downscaling,
  author = {Rampal, Neelesh and Gibson, Peter B. and Sherwood, Steven C. and Queen, Laura E. and Lewis, Hamish and Abramowitz, Gab},
  title = {Downscaling with AI reveals the large role of internal variability in fine-scale projections of climate extremes},
  journal = {Climate Dynamics},
  year = {2026},
  doi = {10.1007/s00382-026-08269-y},
  url = {https://doi.org/10.1007/s00382-026-08269-y}
}