期刊目录 | 《中国科学：地球科学》英文版2024年第12期文章速览

1

Improving global weather and ocean wave forecast with large artificial intelligence models 

A hybrid deep learning and data assimilation method for model error estimation

A deep learning-based global tropical cyclogenesis prediction model and its interpretability analysis

FuXi-Extreme: Improving extreme rainfall and wind forecasts with diffusion model

Real-time predictions of the 2023–2024 climate conditions in the tropical Pacific using a purely data-driven Transformer model 

2

Short-term solar eruptive activity prediction models based on machine learning approaches: A review

From fundamental theory to realistic modeling of the birth of solar eruptions

Stable isotopes in atmospheric water vapour: Patterns, mechanisms and perspectives

Advances in understanding the mechanisms of Arctic amplification

3

Records of Fukushima accident-derived cesium-137 in the Chukchi Sea sediment: Implication for a new time marker?

ESDC: An open Earth science data corpus to support geoscientific literature information extraction

Response to climate warming of winter wheat varieties bred across different eras in the North China Plain

Controls on sediment transport from rivers to trenches in passive and active continental margins

Effects of upper mantle wind on mantle plume morphology and hotspot track: Numerical modeling

Differences in both the structure and interaction of the crust and mantle on the eastern and western sides of the Ordos Block

Reflection and transmission coefficient approximation at weak-contrast interfaces for strong VTI media

Simultaneous inversion of seismic scattering and absorption attenuation using coda energies

4

5

New insights on driving factors of East Asian droughts and floods

Promises and challenges in inferring the evolution of ancient organisms using optimum growth temperature

Prediction of severe weather events and projection of climate extremes are essential research areas in atmospheric sciences.Traditionally, prediction and projection rely on observational technology,theoretical studies,and development of statistical and dynamic models.

Recent advancements in artificial intelligence(AI)have greatly accelerated progress in atmospheric sciences.AI has deepened our understanding of atmospheric dynamics and physics,enhanced numerical simulation capabilities,and refined prediction and projection techniques,while advancing the traditional“observationmechanism-numerical simulation”research paradigm.

This special topic highlights the integration of AI in forecasting severe weather and climate events. Featured studies include a hybrid deep learning and data assimilation method for model error estimation,a Swin Transformer-based tropical cyclone genesis prediction model(TCGP-Net),the Fuxi-Extreme model for extreme rainfall and wind speed,an improved 3D-Geoformer model for predicting the El Niño-Southern Oscillation, and a comprehensive review of the evolution of AI-driven forecasting models.

The review suggests that hybrid models integrating AI and numerical simulations may lead the future of prediction.These studies offer novel methodologies for severe weather and climate forecasting and provide valuable insights and scientific ref-erences for interdisciplinary research at the intersection of AI and atmospheric sciences.For details, see the papers on pages 3641–3726.