FN Archimer Export Format
PT J
TI Reconstruction of global surface ocean pCO(2) using region-specific predictors based on a stepwise FFNN regression algorithm
BT 
AF ZHONG, Guorong
   LI, Xuegang
   SONG, Jinming
   QU, Baoxiao
   WANG, Fan
   WANG, Yanjun
   ZHANG, Bin
   SUN, Xiaoxia
   ZHANG, Wuchang
   WANG, Zhenyan
   MA, Jun
   YUAN, Huamao
   DUAN, Liqin
AS 1:1,2,3,4;2:1,2,3,4;3:1,2,3,4;4:1,3,4;5:1,2,3,4;6:1,4;7:1,4;8:1,2,3,4;9:1,3,4;10:1,3,4;11:1,3,4;12:1,2,3,4;13:1,2,3,4;
FF 1:;2:;3:;4:;5:;6:;7:;8:;9:;10:;11:;12:;13:;
C1 Chinese Acad Sci, Inst Oceanol, Key Lab Marine Ecol & Environm Sci, Qingdao 266071, Peoples R China.
   Univ Chinese Acad Sci, Beijing 101407, Peoples R China.
   Pilot Natl Lab Marine Sci & Technol, Marine Ecol & Environm Sci Lab, Qingdao 266237, Peoples R China.
   Chinese Acad Sci, Ctr Ocean Megasci, Qingdao 266071, Peoples R China.
C2 CHINESE ACAD SCI, CHINA
   UNIV CHINESE ACAD SCI, CHINA
   PILOT NATL LAB MARINE SCI & TECHNOL, CHINA
   CHINESE ACAD SCI, CHINA
IN DOAJ
IF 4.9
TC 6
UR https://archimer.ifremer.fr/doc/00755/86712/92183.pdf
   https://archimer.ifremer.fr/doc/00755/86712/92184.pdf
   https://archimer.ifremer.fr/doc/00755/86712/92185.pdf
LA English
DT Article
CR OISO - OCÉAN INDIEN SERVICE D'OBSERVATION
AB Various machine learning methods were attempted in the global mapping of surface ocean partial pressure of CO2 (pCO(2)) to reduce the uncertainty of the global ocean CO2 sink estimate due to undersampling of pCO(2). In previous research, the predictors of pCO(2) were usually selected empirically based on theoretic drivers of surface ocean pCO(2), and the same combination of predictors was applied in all areas except where there was a lack of coverage. However, the differences between the drivers of surface ocean pCO(2) in different regions were not considered. In this work, we combined the stepwise regression algorithm and a feed-forward neural network (FFNN) to select predictors of pCO(2) based on the mean absolute error in each of the 11 biogeochemical provinces defined by the self-organizing map (SOM) method. Based on the predictors selected, a monthly global 1 circle x 1 circle surface ocean pCO(2) product from January 1992 to August 2019 was constructed. Validation of different combinations of predictors based on the Surface Ocean CO2 Atlas (SOCAT) dataset version 2020 and independent observations from time series stations was carried out. The prediction of pCO(2) based on region-specific predictors selected by the stepwise FFNN algorithm was more precise than that based on predictors from previous research. Applying the FFNN size-improving algorithm in each province decreased the mean absolute error (MAE) of the global estimate to 11.32 mu atm and the root mean square error (RMSE) to 17.99 mu atm. The script file of the stepwise FFNN algorithm and pCO(2) product are distributed through the Institute of Oceanology of the Chinese Academy of Sciences Marine Science Data Center (IOCAS, , Zhong, 2021.
PY 2022
PD FEB
SO Biogeosciences
SN 1726-4170
PU Copernicus Gesellschaft Mbh
VL 19
IS 3
UT 000758150300001
BP 845
EP 859
DI 10.5194/bg-19-845-2022
ID 86712
ER

EF