Potential and Limitations of Machine Learning for Modeling Warm‐Rain Cloud Microphysical Processes

Potential and Limitations of Machine Learning for Modeling Warm‐Rain Cloud Microphysical Processes
复制标题

机器学习在暖雨云微物理过程建模中的潜力和局限性

DOI:
10.1029/2020ms002301
复制
发表时间:
2020
影响因子:
6.8
通讯作者:
Seifert
Seifert
中科院分区:
地球科学2区
文献类型:
--
作者:
Seifert

文献摘要

参考文献

被引文献

相似文献

研究了基于神经网络的机器学习在云微物理参数化中的应用。作为一个例子,我们使用了由碰撞聚并形成的暖雨,即在两矩框架中水滴的自转换、吸积和自收集的参数化。采用蒙特卡罗超液滴算法求解了动力学收集方程的基准解。超液滴法提供了可靠但有噪声的暖雨过程速率估计。对于每个处理速率,使用标准的机器学习技术训练神经网络。当与测试数据进行比较时,所得到的模型对过程速率做出了熟练的预测。然而,当求解常微分方程时,解不如建立的暖雨参数化的解好。这一缺陷可以看作是所应用的机器学习方法的局限性,但与此同时,它指出了常用的两矩暖雨方案的根本缺陷。因此,包含时间导数概念的更先进的机器学习方法有可能克服这些问题。
The use of machine learning based on neural networks for cloud microphysical parameterizations is investigated. As an example, we use the warm‐rain formation by collision‐coalescence, that is, the parameterization of autoconversion, accretion, and self‐collection of droplets in a two‐moment framework. Benchmark solutions of the kinetic collection equations are performed using a Monte Carlo superdroplet algorithm. The superdroplet method provides reliable but noisy estimates of the warm‐rain process rates. For each process rate, a neural network is trained using standard machine learning techniques. The resulting models make skillful predictions for the process rates when compared to the testing data. However, when solving the ordinary differential equations, the solutions are not as good as those of an established warm‐rain parameterization. This deficiency can be seen as a limitation of the machine learning methods that are applied, but at the same time, it points toward a fundamental ill‐posedness of the commonly used two‐moment warm‐rain schemes. More advanced machine learning methods that include a notion of time derivatives, therefore, have the potential to overcome these problems.
DOI: 10.1029/2018ms001351
发表时间: 2018-10-01
影响因子: 6.8
作者:
O'Gorman, Paul A.;Dwyer, John G.
通讯作者: Dwyer, John G.
DOI: 10.5194/gmd-10-1521-2017
发表时间: 2016-11
影响因子: 5.1
作者:
S. Unterstrasser;F. Hoffmann;Marion Lerch
通讯作者: S. Unterstrasser;F. Hoffmann;Marion Lerch
拉格朗日云微物理中的随机合并
DOI: 10.5194/acp-17-13509-2017
发表时间: 2017
影响因子: 6.3
作者:
P. Dziekan;H. Pawłowska
通讯作者: H. Pawłowska
DOI: 10.1175/jas-d-20-0082.1
发表时间: 2020-12-01
影响因子: 3.1
作者:
Brenowitz, Noah D.;Beucler, Tom;Bretherton, Christopher S.
通讯作者: Bretherton, Christopher S.
DOI: 10.21203/rs.3.rs-55125/v1
发表时间: 2020-01
期刊: ArXiv
影响因子: --
作者:
Christopher Rackauckas;Yingbo Ma;Julius Martensen;Collin Warner;K. Zubov;R. Supekar;Dominic J. Skinner;A. Ramadhan
通讯作者: Christopher Rackauckas;Yingbo Ma;Julius Martensen;Collin Warner;K. Zubov;R. Supekar;Dominic J. Skinner;A. Ramadhan