Cummins, Donald P., Guemas, Virginie, Blein, Sébastien, Brooks, Ian M., Renfrew, Ian A. ORCID: https://orcid.org/0000-0001-9379-8215, Elvidge, Andrew D.
ORCID: https://orcid.org/0000-0002-7099-902X and Prytherch, John
(2024)
Reducing parametrization errors for polar surface turbulent fluxes using machine learning.
Boundary-Layer Meteorology, 190 (3).
ISSN 0006-8314
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.
Item Type: | Article |
---|---|
Additional Information: | Data availability: ACCACIA flight data are available from the CEDA archive: https://doi.org/10.5285/0844186db1ba9e20319a2560f8d61651 (MASIN); https://catalogue.ceda.ac.uk/uuid/c064b0c150274a1cbd18c563573f392e (FAAM). ACSE cruise data are available from the CEDA archive (https://doi.org/10.5285/c6f1b1ff16f8407386e2d643bc5b916a, Brooks et al. 2022a). AO16 cruise data are available from the CEDA archive (https://doi.org/10.5285/614752d35dc147a598d5421443fb50e8, Brooks et al. 2022b). SHEBA data are available from the NCAR Earth Observing Laboratory: (https://doi.org/10.5065/D65H7DNS, Andreas et al. 2007) (ASFG tower); (https://doi.org/10.5065/D6ZC8170, Andreas et al. 2012) (PAM stations). NSIDC sea ice concentration data are available from the NSIDC archive (https://doi.org/10.7265/efmz-2t65, Meier et al. 2021). Code availability: The polar-specific bulk algorithm used in this study and described in Sect. 3.1 is available to download as a Python library from GitHub (https://github.com/virginieguemas/CDlib). Funding Information: This work was supported by a national funding by the Agence Nationale de la Recherche within the framework of the Investissement d’Avenir programme under the ANR-17-MPGA-0003 ASET reference. This article has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 101003826 via project CRiceS (Climate Relevant interactions and feedbacks: the key role of sea ice and Snow in the polar and global climate system). The ACCACIA field campaign was supported by the UK Natural Environment Research Council (NERC) grant numbers NE/I028653/1, NE/I028858/1, and NE/I028297/1. The participation of Ian Brooks and John Prytherch in the ACSE field campaign was supported by NERC grant number NE/K011820/1. The AO16 measurements were supported by the the Swedish Polar Research Secretariat. John Prytherch was also supported by the Knut and Alice Wallenberg Foundation (grant number 2016-0024). |
Uncontrolled Keywords: | arctic,artificial neural networks,machine learning,monin-obukhov similarity theory,sea ice,surface layer,atmospheric science,sdg 13 - climate action,2* ,/dk/atira/pure/subjectarea/asjc/1900/1902 |
Faculty \ School: | Faculty of Science > School of Environmental Sciences University of East Anglia Research Groups/Centres > Theme - ClimateUEA |
UEA Research Groups: | Faculty of Science > Research Groups > Centre for Ocean and Atmospheric Sciences |
Related URLs: | |
Depositing User: | LivePure Connector |
Date Deposited: | 26 Mar 2024 13:30 |
Last Modified: | 03 Apr 2024 09:31 |
URI: | https://ueaeprints.uea.ac.uk/id/eprint/94767 |
DOI: | 10.1007/s10546-023-00852-8 |
Actions (login required)
![]() |
View Item |