Data mining-based machine learning methods for improving hydrological data - a case study of salinity field in the Western Arctic Ocean

Aug 4, 2026

Upload by Ezra

The Beaufort Gyre functions as the largest freshwater reservoir in the Arctic Ocean, so accurate freshwater and salinity characterization is essential for understanding Arctic stratification, circulation, and biogeochemical cycles. Intensive observations and ship surveys provide abundant CTD measurements, which are leveraged to reconstruct annual salinity products from 2003–2022. Multi-machine learning models are data-mining based and use physically driven predictors such as sea level pressure, bathymetry, sea ice concentration, and sea ice drift, achieving controlled salinity error metrics. Results show reliable freshwater content characterization across halocline depth and offer an approach to generate robust hydrological data products where observations are scarce.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Study context: Beaufort Gyre and Arctic freshwater evolution

Data availability and CTD observations

Rationale for data mining-based multi-machine learning

TYPE Original Research PUBLISHED 19 November 2024 DOI 10.3389/fmars.2024.1490548 OPEN ACCESS EDITED BY Francisco Machn, University of Las Palmas de Gran Canaria, Spain REVIEWED BY Tien Anh Tran, Seoul National University, Republic of Korea Longjiang Mu, Laoshan Laboratory, China *CORRESPONDENCE Ling Du [duling@ouc.edu.cn](duling@ouc.edu.cn) RECEIVED 03 September 2024 ACCEPTED 04 November 2024 PUBLISHED 19 November 2024 CITATION Tao S, Du L and Li J (2024) Data miningbased machine learning methods for improving hydrological data: a case study of salinity ﬁeld in the Western Arctic Ocean. Front. Mar. Sci. 11:1490548 . doi: 10.3389/fmars.2024.1490548 COPYRIGHT © 2024 Tao, Du and Li. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Data mining-based machine learning methods for improving hydrological data: a case study of salinity ﬁeld in the Western Arctic Ocean Shuhao Tao 1,2, Ling Du 1,2* and Jiahao Li 1,2 1 Frontier Science Center for Deep Ocean Multispheres and Earth System (FDOMES) and Physical Oceanography Laboratory, Ocean University of China, Qingdao, China, 2College of Oceanic and Atmospheric Sciences, Ocean University of China, Qingdao, China The Beaufort Gyre is the largest freshwater reservoir in the Arctic Ocean. Longterm changes in freshwater reservoirs are critical for understanding the Arctic Ocean, and data from various sources, particularly observation or reanalysis data, must be used to the greatest extent possible. Over the past two decades, a large number of intensive ﬁeld observations and ship surveys have been conducted in the western Arctic Ocean to obtain a large amount of CTD (Conductivity, Temperature, and Depth) data. Multi-machine learning methods were assessed and merged to reconstruct the annual salinity product in the Western Arctic Ocean over the period 2003-2022 . Data mining-based machine learning methods reconstructed salinity product based on input variables determined by physical processes, such as sea level pressure, bathymetry, sea ice concentration, and sea ice drift. The root-mean-square error of sea surface salinity, in comparison to deep water, was effectively managed during machine learning, which exhibits higher sensitivity to variations in the atmosphere, sea ice, and ocean. The mean absolute errors in freshwater content and halocline depth within the Beaufort Gyre region for the salinity product from 2003 to 2022 are 0. 98 m and 1 .31 m, respectively, when compared to observational data. The salinity product provides reliable characterizations of freshwater content in the Beaufort Gyre and its variations at halocline depth. In polar regions where lacking observed data, we can build data mining-based machine learning methods to generate reliable data products to compensate for the inconvenience. Furthermore, the application potential of this multi-machine learning results approach for evaluating and integrating extends beyond the salinity ﬁeld, encompassing hydrometeorology, sea ice thickness, polar biogeochemistry, and other related ﬁelds. KEYWORDS salinity product, multi-machine learning, data merging, post calibrating, Western Arctic Ocean Frontiers in Marine Science 01 [frontiersin.org](frontiersin.org) 1 Introduction In contrast to the low- and mid-latitude oceans, the Arctic Ocean is characterized by its extensive sea ice coverage and nearfreezing sea surface water. Variations in salinity in the Western Arctic Ocean have profound implications for stratiﬁcation strength, ocean circulation patterns, and biogeochemical cycles (

FAQs

What does the study aim to improve and why is it important in the Arctic Ocean?

The study improves hydrological salinity data products to better characterize freshwater content and its evolution in the Beaufort Gyre. Accurate salinity is crucial for understanding stratification, circulation patterns, and biogeochemical cycles in the Arctic.

Which period and data sources are used to reconstruct the salinity product?

The annual salinity product is reconstructed for 2003–2022 using CTD observations collected through intensive field observations and ship surveys in the western Arctic Ocean.

What predictors and machine-learning strategy are used in the reconstruction?

The method uses input variables determined by physical processes, including sea level pressure, bathymetry, sea ice concentration, and sea ice drift. Multi-machine learning models are assessed and merged, with post-calibrating to manage reconstruction errors.

Aug 4, 2026

Upload by Ezra