Improved sea surface salinity data for the Arctic Ocean derived from SMAP satellite data using machine learning approaches
Salinity is among the key climate characteristics of the World Ocean. During the last 15 years, sea surface salinity (SSS) is measured using satellite passive microwave sensors. Standard retrieving SSS algorithms from remote sensing data were developed and verified for the most typical temperature a...
Published in: | Frontiers in Marine Science |
---|---|
Main Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
Frontiers Media SA
2024
|
Subjects: | |
Online Access: | http://dx.doi.org/10.3389/fmars.2024.1358882 https://www.frontiersin.org/articles/10.3389/fmars.2024.1358882/full |
id |
crfrontiers:10.3389/fmars.2024.1358882 |
---|---|
record_format |
openpolar |
spelling |
crfrontiers:10.3389/fmars.2024.1358882 2024-04-28T08:06:41+00:00 Improved sea surface salinity data for the Arctic Ocean derived from SMAP satellite data using machine learning approaches Savin, Alexander Krinitskiy, Mikhail Osadchiev, Alexander 2024 http://dx.doi.org/10.3389/fmars.2024.1358882 https://www.frontiersin.org/articles/10.3389/fmars.2024.1358882/full unknown Frontiers Media SA https://creativecommons.org/licenses/by/4.0/ Frontiers in Marine Science volume 11 ISSN 2296-7745 Ocean Engineering Water Science and Technology Aquatic Science Global and Planetary Change Oceanography journal-article 2024 crfrontiers https://doi.org/10.3389/fmars.2024.1358882 2024-04-02T07:44:13Z Salinity is among the key climate characteristics of the World Ocean. During the last 15 years, sea surface salinity (SSS) is measured using satellite passive microwave sensors. Standard retrieving SSS algorithms from remote sensing data were developed and verified for the most typical temperature and salinity values of the World Ocean. However, they have far lower accuracy for the Arctic Ocean, especially its shelf areas, which are influenced by large river runoff and have low typical temperature and salinity values. In this study, an improved algorithm has been developed to retrieve SSS in the Arctic Ocean during ice-free season, based on Soil Moisture Active Passive (SMAP) mission data, and using machine learning approaches. Extensive database of in situ salinity measurements in the Russian Arctic seas collected during multiple field surveys is applied to train and validate the machine learning models. The error in SSS retrieval of the developed algorithm compared to the standard algorithm reduced from 3.15 to 2.15 psu, and the correlation with in situ data increased from 0.82 to 0.90. The obtained daily SSS fields are important to improve accurate assessment of spatial and temporal variability of large river plumes in the Arctic Ocean. Article in Journal/Newspaper Arctic Arctic Ocean Frontiers (Publisher) Frontiers in Marine Science 11 |
institution |
Open Polar |
collection |
Frontiers (Publisher) |
op_collection_id |
crfrontiers |
language |
unknown |
topic |
Ocean Engineering Water Science and Technology Aquatic Science Global and Planetary Change Oceanography |
spellingShingle |
Ocean Engineering Water Science and Technology Aquatic Science Global and Planetary Change Oceanography Savin, Alexander Krinitskiy, Mikhail Osadchiev, Alexander Improved sea surface salinity data for the Arctic Ocean derived from SMAP satellite data using machine learning approaches |
topic_facet |
Ocean Engineering Water Science and Technology Aquatic Science Global and Planetary Change Oceanography |
description |
Salinity is among the key climate characteristics of the World Ocean. During the last 15 years, sea surface salinity (SSS) is measured using satellite passive microwave sensors. Standard retrieving SSS algorithms from remote sensing data were developed and verified for the most typical temperature and salinity values of the World Ocean. However, they have far lower accuracy for the Arctic Ocean, especially its shelf areas, which are influenced by large river runoff and have low typical temperature and salinity values. In this study, an improved algorithm has been developed to retrieve SSS in the Arctic Ocean during ice-free season, based on Soil Moisture Active Passive (SMAP) mission data, and using machine learning approaches. Extensive database of in situ salinity measurements in the Russian Arctic seas collected during multiple field surveys is applied to train and validate the machine learning models. The error in SSS retrieval of the developed algorithm compared to the standard algorithm reduced from 3.15 to 2.15 psu, and the correlation with in situ data increased from 0.82 to 0.90. The obtained daily SSS fields are important to improve accurate assessment of spatial and temporal variability of large river plumes in the Arctic Ocean. |
format |
Article in Journal/Newspaper |
author |
Savin, Alexander Krinitskiy, Mikhail Osadchiev, Alexander |
author_facet |
Savin, Alexander Krinitskiy, Mikhail Osadchiev, Alexander |
author_sort |
Savin, Alexander |
title |
Improved sea surface salinity data for the Arctic Ocean derived from SMAP satellite data using machine learning approaches |
title_short |
Improved sea surface salinity data for the Arctic Ocean derived from SMAP satellite data using machine learning approaches |
title_full |
Improved sea surface salinity data for the Arctic Ocean derived from SMAP satellite data using machine learning approaches |
title_fullStr |
Improved sea surface salinity data for the Arctic Ocean derived from SMAP satellite data using machine learning approaches |
title_full_unstemmed |
Improved sea surface salinity data for the Arctic Ocean derived from SMAP satellite data using machine learning approaches |
title_sort |
improved sea surface salinity data for the arctic ocean derived from smap satellite data using machine learning approaches |
publisher |
Frontiers Media SA |
publishDate |
2024 |
url |
http://dx.doi.org/10.3389/fmars.2024.1358882 https://www.frontiersin.org/articles/10.3389/fmars.2024.1358882/full |
genre |
Arctic Arctic Ocean |
genre_facet |
Arctic Arctic Ocean |
op_source |
Frontiers in Marine Science volume 11 ISSN 2296-7745 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.3389/fmars.2024.1358882 |
container_title |
Frontiers in Marine Science |
container_volume |
11 |
_version_ |
1797576082085380096 |