Interpretable Multiscale Machine Learning-Based Parameterizations of Convection for ICON

IF 4.4 2区 地球科学 Q1 METEOROLOGY & ATMOSPHERIC SCIENCES Journal of Advances in Modeling Earth Systems Pub Date : 2024-08-22 DOI:10.1029/2024MS004398
Helge Heuer, Mierk Schwabe, Pierre Gentine, Marco A. Giorgetta, Veronika Eyring
{"title":"Interpretable Multiscale Machine Learning-Based Parameterizations of Convection for ICON","authors":"Helge Heuer,&nbsp;Mierk Schwabe,&nbsp;Pierre Gentine,&nbsp;Marco A. Giorgetta,&nbsp;Veronika Eyring","doi":"10.1029/2024MS004398","DOIUrl":null,"url":null,"abstract":"<p>Machine learning (ML)-based parameterizations have been developed for Earth System Models (ESMs) with the goal to better represent subgrid-scale processes or to accelerate computations. ML-based parameterizations within hybrid ESMs have successfully learned subgrid-scale processes from short high-resolution simulations. However, most studies used a particular ML method to parameterize the subgrid tendencies or fluxes originating from the compound effect of various small-scale processes (e.g., radiation, convection, gravity waves) in mostly idealized settings or from superparameterizations. Here, we use a filtering technique to explicitly separate convection from these processes in simulations with the Icosahedral Non-hydrostatic modeling framework (ICON) in a realistic setting and benchmark various ML algorithms against each other offline. We discover that an unablated U-Net, while showing the best offline performance, learns reverse causal relations between convective precipitation and subgrid fluxes. While we were able to connect the learned relations of the U-Net to physical processes this was not possible for the non-deep learning-based Gradient Boosted Trees. The ML algorithms are then coupled online to the host ICON model. Our best online performing model, an ablated U-Net excluding precipitating tracer species, indicates higher agreement for simulated precipitation extremes and mean with the high-resolution simulation compared to the traditional scheme. However, a smoothing bias is introduced both in water vapor path and mean precipitation. Online, the ablated U-Net significantly improves stability compared to the non-ablated U-Net and runs stable for the full simulation period of 180 days. Our results hint to the potential to significantly reduce systematic errors with hybrid ESMs.</p>","PeriodicalId":14881,"journal":{"name":"Journal of Advances in Modeling Earth Systems","volume":"16 8","pages":""},"PeriodicalIF":4.4000,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2024MS004398","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Advances in Modeling Earth Systems","FirstCategoryId":"89","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1029/2024MS004398","RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"METEOROLOGY & ATMOSPHERIC SCIENCES","Score":null,"Total":0}
引用次数: 0

Abstract

Machine learning (ML)-based parameterizations have been developed for Earth System Models (ESMs) with the goal to better represent subgrid-scale processes or to accelerate computations. ML-based parameterizations within hybrid ESMs have successfully learned subgrid-scale processes from short high-resolution simulations. However, most studies used a particular ML method to parameterize the subgrid tendencies or fluxes originating from the compound effect of various small-scale processes (e.g., radiation, convection, gravity waves) in mostly idealized settings or from superparameterizations. Here, we use a filtering technique to explicitly separate convection from these processes in simulations with the Icosahedral Non-hydrostatic modeling framework (ICON) in a realistic setting and benchmark various ML algorithms against each other offline. We discover that an unablated U-Net, while showing the best offline performance, learns reverse causal relations between convective precipitation and subgrid fluxes. While we were able to connect the learned relations of the U-Net to physical processes this was not possible for the non-deep learning-based Gradient Boosted Trees. The ML algorithms are then coupled online to the host ICON model. Our best online performing model, an ablated U-Net excluding precipitating tracer species, indicates higher agreement for simulated precipitation extremes and mean with the high-resolution simulation compared to the traditional scheme. However, a smoothing bias is introduced both in water vapor path and mean precipitation. Online, the ablated U-Net significantly improves stability compared to the non-ablated U-Net and runs stable for the full simulation period of 180 days. Our results hint to the potential to significantly reduce systematic errors with hybrid ESMs.

Abstract Image

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
基于机器学习的可解释多尺度对流参数化 ICON
为地球系统模型(ESM)开发了基于机器学习(ML)的参数化,目的是更好地表示子网格尺度过程或加速计算。混合 ESM 中基于 ML 的参数化已经成功地从短时高分辨率模拟中学习到了子网格尺度过程。然而,大多数研究使用特定的 ML 方法来参数化源自各种小尺度过程(如辐射、对流、重力波)复合效应的子网格趋势或通量,这些过程大多是理想化设置或超参数化。在这里,我们使用一种过滤技术,在二十面体非流体静力学建模框架(ICON)的模拟中,将对流从这些过程中明确分离出来,并对各种 ML 算法进行离线对比。我们发现,未钝化的 U-Net 虽然显示出最佳离线性能,但却能反向学习对流降水与子网格通量之间的因果关系。虽然我们能够将 U-Net 学习到的关系与物理过程联系起来,但这对于基于非深度学习的梯度提升树来说是不可能的。然后将 ML 算法与主机 ICON 模型进行在线耦合。与传统方案相比,我们的在线性能最佳模型--不包括降水示踪物种的消融 U-Net 表明,模拟的降水极值和平均值与高分辨率模拟的一致性更高。不过,水汽路径和平均降水量都出现了平滑偏差。与未消融的 U-Net 相比,在线消融的 U-Net 显著提高了稳定性,并在 180 天的整个模拟期内稳定运行。我们的研究结果表明,混合 ESM 有可能显著减少系统误差。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Journal of Advances in Modeling Earth Systems
Journal of Advances in Modeling Earth Systems METEOROLOGY & ATMOSPHERIC SCIENCES-
CiteScore
11.40
自引率
11.80%
发文量
241
审稿时长
>12 weeks
期刊介绍: The Journal of Advances in Modeling Earth Systems (JAMES) is committed to advancing the science of Earth systems modeling by offering high-quality scientific research through online availability and open access licensing. JAMES invites authors and readers from the international Earth systems modeling community. Open access. Articles are available free of charge for everyone with Internet access to view and download. Formal peer review. Supplemental material, such as code samples, images, and visualizations, is published at no additional charge. No additional charge for color figures. Modest page charges to cover production costs. Articles published in high-quality full text PDF, HTML, and XML. Internal and external reference linking, DOI registration, and forward linking via CrossRef.
期刊最新文献
Standardized Daily High-Resolution Large-Eddy Simulations of the Arctic Boundary Layer and Clouds During the Complete MOSAiC Drift A Simple Model for the Emergence of Relaxation-Oscillator Convection Online Learning of Entrainment Closures in a Hybrid Machine Learning Parameterization A Refined Zero-Buoyancy Plume Model for Large-Scale Atmospheric Profiles and Anvil Clouds in Radiative-Convective Equilibrium Quantitative Decoupling Analysis for Assessing the Meteorological, Emission, and Chemical Influences on Fine Particle Pollution
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1