Grassi et al. (2026) A Machine Learning Framework for Objective Definition and Projection of Meteorological Seasons: An Application to the Indian Subcontinent

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

Identification

• Journal: Journal of Geophysical Research Atmospheres
• Year: 2026
• Date: 2026-10-15
• Authors: Jacopo Grassi, Elisa Palazzi, Jost von Hardenberg, Paolo Davini
• DOI: 10.1029/2025jd045869

Research Groups

• Indian Institute of Tropical Meteorology (IITM)
• University of [Unknown]
• Other research groups involved in CMIP6 multi-model ensemble

Short Summary

This study presents a machine learning framework that objectively detects meteorological seasons and projects their future evolution, focusing on the Indian subcontinent as a proof-of-concept region.

Objective

• Develop an objective method to detect and predict meteorological seasons using reanalysis data and climate model projections.

Study Configuration

• Spatial Scale: Regional (Indian subcontinent)
• Temporal Scale: Annual cycle

Methodology and Data

• Models used:
  • Radially Constrained Clustering (RCC) for identifying seasons from reanalysis data
  • Multinomial Logistic Regression Classifier (MLRC) for applying these definitions to climate model projections
• Data sources:
  • Reanalysis data (e.g., ERA5)
  • CMIP6 multi-model ensemble

Main Results

• The RCC-derived seasons recover the main operational seasonal phases adopted by the India Meteorological Department.
• Evaluation against the CMIP6 multi-model ensemble shows that models capture the seasonal cycle with high skill in Central and Southern India, but exhibit lower performance in complex mountainous regions.
• Future projections indicate a robust lengthening of the monsoon season, primarily due to delayed withdrawal, and a concurrent shortening of winter.

Contributions

This study provides an original contribution by introducing a data-driven approach to provide objective and reproducible definitions of seasonality, which can be applied to other regions and climate models.

Funding

• This research was funded by [Unknown funding projects or programs]

Citation

@article{Grassi2026Machine,
  author = {Grassi, Jacopo and Palazzi, Elisa and Hardenberg, Jost von and Davini, Paolo},
  title = {A Machine Learning Framework for Objective Definition and Projection of Meteorological Seasons: An Application to the Indian Subcontinent},
  journal = {Journal of Geophysical Research Atmospheres},
  year = {2026},
  doi = {10.1029/2025jd045869},
  url = {https://doi.org/10.1029/2025jd045869}
}