Regional climate projections using a deep learning--based model-ranking and downscaling framework: Application to European climate zones

2025-02-27

AI Summary

This study addresses the insufficient regional downscaling accuracy of Global Climate Models (GCMs) by proposing a deep learning–driven framework for multi-model evaluation and high-resolution downscaling. First, we develop DL-TOPSIS—a deep learning–enhanced Technique for Order Preference by Similarity to Ideal Solution—to perform weighted ranking of 32 CMIP6 models. Second, we conduct 0.1°-resolution temperature downscaling across five Köppen climate zones. Our key innovation is GeoSTANet, a Geographic Spatio-Temporal Attention Network that uniquely integrates imbalance-aware learning with explicit spatio-temporal modeling. Experiments demonstrate that GeoSTANet achieves RMSE = 1.57°C and Kling–Gupta Efficiency (KGE) = 0.89—outperforming ConvLSTM by a 20% RMSE reduction—and significantly improves reliability in simulating extreme temperature indices (TXx/TNn). The framework thus delivers high-fidelity, actionable climate projections to support regional climate adaptation planning.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Computer Vision: Remote Sensing / Geospatial AI

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web search models and ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Accurate regional climate forecast calls for high-resolution downscaling of Global Climate Models (GCMs). This work presents a deep-learning-based multi-model evaluation and downscaling framework ranking 32 Coupled Model Intercomparison Project Phase 6 (CMIP6) models using a Deep Learning-TOPSIS (DL-TOPSIS) mechanism and so refines outputs using advanced deep-learning models. Using nine performance criteria, five K""oppen-Geiger climate zones -- Tropical, Arid, Temperate, Continental, and Polar -- are investigated over four seasons. While TaiESM1 and CMCC-CM2-SR5 show notable biases, ranking results show that NorESM2-LM, GISS-E2-1-G, and HadGEM3-GC31-LL outperform other models. Four models contribute to downscaling the top-ranked GCMs to 0.1$^{circ}$ resolution: Vision Transformer (ViT), Geospatial Spatiotemporal Transformer with Attention and Imbalance-Aware Network (GeoSTANet), CNN-LSTM, and CNN-Long Short-Term Memory (ConvLSTM). Effectively capturing temperature extremes (TXx, TNn), GeoSTANet achieves the highest accuracy (Root Mean Square Error (RMSE) = 1.57$^{circ}$C, Kling-Gupta Efficiency (KGE) = 0.89, Nash-Sutcliffe Efficiency (NSE) = 0.85, Correlation ($r$) = 0.92), so reducing RMSE by 20% over ConvLSTM. CNN-LSTM and ConvLSTM do well in Continental and Temperate zones; ViT finds fine-scale temperature fluctuations difficult. These results confirm that multi-criteria ranking improves GCM selection for regional climate studies and transformer-based downscaling exceeds conventional deep-learning methods. This framework offers a scalable method to enhance high-resolution climate projections, benefiting impact assessments and adaptation plans.

Problem

• Develops a deep learning-based model-ranking framework
• Downscales GCMs to 0.1° resolution for Europe
• Evaluates 32 CMIP6 models across climate zones

Innovation

• Deep-learning-based multi-model evaluation framework
• Transformer-based models for downscaling GCMs
• Multi-criteria ranking improves GCM selection