Reducing Parametrization Errors for Polar Surface Turbulent Fluxes Using Machine Learning

IF 2.3 3区 地球科学 Q3 METEOROLOGY & ATMOSPHERIC SCIENCES
{"title":"Reducing Parametrization Errors for Polar Surface Turbulent Fluxes Using Machine Learning","authors":"","doi":"10.1007/s10546-023-00852-8","DOIUrl":null,"url":null,"abstract":"<h3>Abstract</h3> <p>Turbulent exchanges between sea ice and the atmosphere are known to influence the melting rate of sea ice, the development of atmospheric circulation anomalies and, potentially, teleconnections between polar and non-polar regions. Large model errors remain in the parametrization of turbulent heat fluxes over sea ice in climate models, resulting in significant uncertainties in projections of future climate. Fluxes are typically calculated using bulk formulae, based on Monin-Obukhov similarity theory, which have shown particular limitations in polar regions. Parametrizations developed specifically for polar conditions (e.g. representing form drag from ridges or melt ponds on sea ice) rely on sparse observations and thus may not be universally applicable. In this study, new data-driven parametrizations have been developed for surface turbulent fluxes of momentum, sensible heat and latent heat in the Arctic. Machine learning has already been used outside the polar regions to provide accurate and computationally inexpensive estimates of surface turbulent fluxes. To investigate the feasibility of this approach in the Arctic, we have fitted neural-network models to a reference dataset (SHEBA). Predictive performance has been tested using data from other observational campaigns. For momentum and sensible heat, performance of the neural networks is found to be comparable to, and in some cases substantially better than, that of a state-of-the-art bulk formulation. These results offer an efficient alternative to the traditional bulk approach in cases where the latter fails, and can serve to inform further physically based developments. </p>","PeriodicalId":9153,"journal":{"name":"Boundary-Layer Meteorology","volume":"9 1","pages":""},"PeriodicalIF":2.3000,"publicationDate":"2024-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Boundary-Layer Meteorology","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.1007/s10546-023-00852-8","RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"METEOROLOGY & ATMOSPHERIC SCIENCES","Score":null,"Total":0}
引用次数: 0

Abstract

Turbulent exchanges between sea ice and the atmosphere are known to influence the melting rate of sea ice, the development of atmospheric circulation anomalies and, potentially, teleconnections between polar and non-polar regions. Large model errors remain in the parametrization of turbulent heat fluxes over sea ice in climate models, resulting in significant uncertainties in projections of future climate. Fluxes are typically calculated using bulk formulae, based on Monin-Obukhov similarity theory, which have shown particular limitations in polar regions. Parametrizations developed specifically for polar conditions (e.g. representing form drag from ridges or melt ponds on sea ice) rely on sparse observations and thus may not be universally applicable. In this study, new data-driven parametrizations have been developed for surface turbulent fluxes of momentum, sensible heat and latent heat in the Arctic. Machine learning has already been used outside the polar regions to provide accurate and computationally inexpensive estimates of surface turbulent fluxes. To investigate the feasibility of this approach in the Arctic, we have fitted neural-network models to a reference dataset (SHEBA). Predictive performance has been tested using data from other observational campaigns. For momentum and sensible heat, performance of the neural networks is found to be comparable to, and in some cases substantially better than, that of a state-of-the-art bulk formulation. These results offer an efficient alternative to the traditional bulk approach in cases where the latter fails, and can serve to inform further physically based developments.

利用机器学习减少极地表面湍流通量的参数化误差
摘要 众所周知,海冰与大气之间的湍流交换会影响海冰的融化速度、大气环流异常的发展,并有可能影响极地与非极地区域之间的远程联系。气候模式中海冰上湍流热通量的参数化仍存在很大的模式误差,导致对未来气候的预测存在很大的不确定性。通量通常使用基于莫宁-奥布霍夫相似性理论的大体积公式计算,这在极地地区显示出特别的局限性。专为极地条件开发的参数化(如代表海脊或海冰熔池的形式阻力)依赖于稀少的观测数据,因此可能并不普遍适用。在这项研究中,针对北极地区的表面动量、显热和潜热的湍流通量,开发了新的数据驱动参数。机器学习已被用于极地以外地区,以提供精确且计算成本低廉的地表湍流通量估计值。为了研究这种方法在北极地区的可行性,我们将神经网络模型与参考数据集(SHEBA)进行了匹配。使用其他观测活动的数据对预测性能进行了测试。在动量和显热方面,我们发现神经网络的性能与最先进的批量公式相当,在某些情况下甚至大大优于后者。这些结果为传统的大容量方法失效时提供了有效的替代方法,并可为进一步基于物理的开发提供参考。
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 求助全文
来源期刊
Boundary-Layer Meteorology
Boundary-Layer Meteorology 地学-气象与大气科学
CiteScore
7.50
自引率
14.00%
发文量
72
审稿时长
12 months
期刊介绍: Boundary-Layer Meteorology offers several publishing options: Research Letters, Research Articles, and Notes and Comments. The Research Letters section is designed to allow quick dissemination of new scientific findings, with an initial review period of no longer than one month. The Research Articles section offers traditional scientific papers that present results and interpretations based on substantial research studies or critical reviews of ongoing research. The Notes and Comments section comprises occasional notes and comments on specific topics with no requirement for rapid publication. Research Letters are limited in size to five journal pages, including no more than three figures, and cannot contain supplementary online material; Research Articles are generally fifteen to twenty pages in length with no more than fifteen figures; Notes and Comments are limited to ten journal pages and five figures. Authors submitting Research Letters should include within their cover letter an explanation of the need for rapid publication. More information regarding all publication formats can be found in the recent Editorial ‘Introducing Research Letters to Boundary-Layer Meteorology’.
×
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
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
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 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信