TianQuan-Climate: A Subseasonal-to-Seasonal Global Weather Model via Incorporate Climatology State

2025-04-14

AI Summary

Subseasonal-to-seasonal (S2S; 15–45 days) global weather forecasting faces fundamental challenges—including initial condition error accumulation, delayed responses to external forcings, and prediction smoothing—limiting its utility in climate resilience, agriculture, and renewable energy planning. To address these, we propose the first end-to-end deep learning meteorological model specifically designed for the S2S timescale. Our method introduces a multi-model collaborative forecasting framework to suppress systematic error propagation, a climate-state dynamic embedding module for effective content fusion, and a novel uncertainty-aware Vision Transformer (UD-ViT) to enhance generalization and robustness. Experiments demonstrate that our model consistently outperforms state-of-the-art numerical weather prediction systems (e.g., ECMWF-SEAS5) and existing AI-based approaches across five upper-air variables (at 13 pressure levels) and two surface variables, evaluated on daily mean forecasts. It achieves significant improvements in both forecast accuracy and reliability at extended lead times.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Planning, Routing, and Scheduling: Planning with Language Models
• Computer Vision: Diffusion Models for Vision

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Subseasonal forecasting serves as an important support for Sustainable Development Goals (SDGs), such as climate challenges, agricultural yield and sustainable energy production. However, subseasonal forecasting is a complex task in meteorology due to dissipating initial conditions and delayed external forces. Although AI models are increasingly pushing the boundaries of this forecasting limit, they face two major challenges: error accumulation and Smoothness. To address these two challenges, we propose Climate Furnace Subseasonal-to-Seasonal (TianQuan-Climate), a novel machine learning model designed to provide global daily mean forecasts up to 45 days, covering five upper-air atmospheric variables at 13 pressure levels and two surface variables. Our proposed TianQuan-Climate has two advantages: 1) it utilizes a multi-model prediction strategy to reduce system error impacts in long-term subseasonal forecasts; 2) it incorporates a Content Fusion Module for climatological integration and extends ViT with uncertainty blocks (UD-ViT) to improve generalization by learning from uncertainty. We demonstrate the effectiveness of TianQuan-Climate on benchmarks for weather forecasting and climate projections within the 15 to 45-day range, where TianQuan-Climate outperforms existing numerical and AI methods.

Problem

• Addresses error accumulation in subseasonal weather forecasting
• Reduces system errors in long-term climate predictions
• Improves generalization using climatological integration and uncertainty learning

Innovation

• Multi-model strategy reduces long-term forecast errors
• Content Fusion Module integrates climatological data
• UD-ViT improves generalization via uncertainty learning