@article{fdi:010066132,
  title = {{S}atellite-derived {CO}2 fugacity in surface seawater of the tropical {A}tlantic {O}cean using a feedforward neural network},
  author = {{M}oussa, {H}. and {B}enallal, {M}. {A}. and {G}oyet, {C}. and {L}ef{\`e}vre, {N}athalie},
  editor = {},
  language = {{ENG}},
  abstract = {{A} feedforward neural network is used to quantify the fugacity of {CO}2 in surface seawater (f ({CO}2sw)) of the tropical {A}tlantic {O}cean, exclusively from satellite data: sea-surface temperature, sea-surface salinity, and chlorophyll-a (chl-a), at a 4 km x 4 km spatial resolution, for the period of spring ({M}arch and {A}pril). {T}he model was constructed using 7188 in situ data provided by the '{S}urface {O}cean {CO}2 {AT}las' ({SOCAT}) products, and the '{EC}-funded project {CARBOOCEAN} {IP} program' products, available for the years 2001, 2002, 2004, 2006, 2007, and 2009. {T}he model was tested using remote sensing data of the {M}oderate {R}esolution {I}maging {S}pectroradiometer {A}qua. {T}his approach was validated over the area extending from 8 degrees {N}-61 degrees {W} to 23 degrees {N}-20 degrees {W}. {A} comparison with multiple linear regression model was established. {T}he neural network has provided better results (root mean square error ({RMSE}) of 8.7 atm (0.881 {P}a)) than linear regression ({RMSE} of 9.6 atm (0.973 {P}a)) for f ({CO}2sw) interpolation using remote sensing data. {S}ince the required input data are available, this approach could be applied to the whole tropical {A}tlantic {O}cean and for the remaining seasons (summer, fall, and winter).},
  keywords = {{ATLANTIQUE} ; {ZONE} {TROPICALE}},
  booktitle = {},
  journal = {{I}nternational {J}ournal of {R}emote {S}ensing},
  volume = {37},
  numero = {3},
  pages = {580--598},
  ISSN = {0143-1161},
  year = {2016},
  DOI = {10.1080/01431161.2015.1131872},
  URL = {https://www.documentation.ird.fr/hor/fdi:010066132},
}