JAMES: 地球系统模式机器学习应用专刊

近年来，测量和模拟得到的地球系统数据急剧增加，已经超出了当前我们处理、理解和使用这些数据的能力。机器学习方法的兴起为我们提供了机会促进我们处理、分析以及从大量的地球系统数据中学习，并应用到模式参数化和预测。此专刊将征集利用机器学习促进地球系统建模的稿件，包括为推动地球系统建模开发的新机器学习方法（比如可解释性机器学习算法、物理指导算法、因果推断和混合模型）以及地球系统建模的机器学习应用（比如天气和气候的可预测性、机器学习参数化、不确定性量化）。

In recent years, measured and simulated Earth system data has grown dramatically, and our current ability to collect and generate high-quality data surpasses our ability to process, understand, and use them. The rise of novel machine learning (ML) methods provides opportunities to boost how we process, analyze, learn from large volumes of Earth systems data, and apply them to model parameterizations and predictions. In this special collection we invite manuscripts that use ML to advance Earth system modeling. The scope of this special collection includes both new ML methodologies developed for advancing Earth system science (e.g., interpretability of ML algorithms, physics-guided algorithms, causal inference, hybrid modeling) and ML applications to Earth system modeling (e.g., predictability of weather and climate, ML parameterizations, uncertainty quantification).

开放提交: 2021.6.1

提交截止: 2023.12.31

组织者:

Janni Yuval, Massachusetts Institute of Technology

Mike Pritchard, University of California Irvine

Pierre Gentine, Columbia University

Laure Zanna, New York University

Jiwen Fan, Pacific Northwest National Laboratory

点击阅读原文了解关于投稿的详细信息。

—END—