NIES-ML3 ensemble product of surface ocean CO2 concentrations and air-sea CO2 fluxes reconstructed by using three machine learning models with new CO2 trends
Click Here to Download 
Global oceans have absorbed a substantial portion of the anthropogenic carbon dioxide (CO2) emitted into the atmosphere. Data-based machine learning estimates for the oceanic CO2 sink have become an import part of the Global Carbon Budget in recent years. This product is the result of our new study on ocean CO2 trends using Random Forest, Gradient Boost Machine, and Feedforward Neural Network. Using the time-dependent trends for ocean CO2 reconstruction substantially reduced the bias of using a constant trend and therefore improved the oceanic sink estimate.
Description
|
Creator
|
|
|
Release date
|
2022/03/11
|
|
Temporal coverage
|
1982 - 2023
|
|
Data provider
|
NIES
Email: cgerdb_admin(at)nies.go.jp |
|
DOI
|
|
|
File format
|
|
|
Data volume
|
1.38 GB
|
|
Version
|
ver.2024.0 (Last updated: 2024/08/30)
|
|
Language
|
English
|
Data Set
|
Parameters
|
Surface ocean CO2 concentration and air-sea CO2 flux
|
| Domain |
Global
|
| Time resolution |
Monthly
|
| Spatial resolution |
1 x 1 degree
|
| Calculation method |
Random Forest, Gradient Boost Machine, Feedforward Neural Network
|
|
Keywords
|
[GCMD_Platform]
Models/Analyses > Models
[GCMD_Science]Oceans > Ocean Chemistry > Carbon Dioxide
[Free keywords]Carbon Dioxide, CO2, Ocean, Flux, Budget, Global, Machine Learning, Random Forest, Gradient Boost Machine, Neural Network
|
|
Update history
|
[2024/08/30]
Data from 1982 to 2023 are released. ver.2024.0. Data sources used for the update:
1.Ocean CO2: OSCATv2024
[2023/07/21]2.Atmospheric CO2: NOAA Greenhouse Gas Marine Boundary Layer Reference 3.Sea Surface Temperature and Ice Cover: NOAA Optimum Interpolation (OI) SST V2 4.Mixed Layer Depth: world-ocean-atlas-2018 5.Salinity: world-ocean-atlas-2023 6.Chlorophyll-a: VIIRSJ1 Level-3 2018-2024 and VIIRSN Level-3 2012-2024 7.Wind Speed and Surface Pressure: ECMWF reanalysis-era5-single-levels-monthly-means Data from 1980 to 2022 are released. ver.2023.0.
[2022/06/30]New ocean CO2 trends were used for the reconstruction of CO2 in ver.2022.2.
[2022/05/19]A new coefficient was used to calculate air-sea flux. ver.2022.1.
[2022/03/11]Data from 1980 to 2020 are released. ver.2022.0.
|
Reference Information
| References |
J. Zeng, Y. Iida, T. Matsunaga, T. Shirai (2022), Surface ocean CO2 concentration and air-sea flux estimate by machine learning with modelled variable trends, Front. Mar. Sci., 9, 989233, doi:10.3389/fmars.2022.989233.
|
Terms and Conditions of Use*
*By accessing or using the Service you agree to follow these Terms. If you disagree with any part of the Terms, you may not access the Service.
| License | |
| Citation format |
When this data set is referred to in publications, it should be cited in the following format.
Zeng, J (2022), NIES-ML3 ensemble product of surface ocean CO2 concentrations and air-sea CO2 fluxes reconstructed by using three machine learning models with new CO2 trends, ver.xxxx.x *1, NIES, DOI:10.17595/20220311.001, (Reference date*2: YYYY/MM/DD)
*1 The version number is indicated in the name of each data file. *2 As the reference date, please indicate the date you downloaded the files. |
Advisory Service
|
Advisory service
|
If you need scientific advice or expert opinion regarding the contents or scientific validity of this data set or Products derived from it, we can provide an advisory service based on an individual contract, different from the above one.
Depending on the extent of the help required, we can offer collaboration and/or supervision of the work based on the present data set. If you want to use our advisory service, contact the Data Provider. |