@article{fdi:010090042,
  title = {{A}bove-ground biomass estimation based on multi-angular l-band measurements of brightness temperatures},
  author = {{S}alazar-{N}eira, {J}. {C}. and {M}ialon, {A}. and {R}ichaume, {P}. and {M}ermoz, {S}. and {K}err, {Y}. {H}. and {B}ouvet, {A}lexandre and {L}e {T}oan, {T}. and {B}oitard, {S}. and {R}odriguez-{F}ernandez, {N}. {J}.},
  editor = {},
  language = {{ENG}},
  abstract = {{T}here is growing interest in using passive microwave observations and vegetation optical depth ({VOD}) to study the above-ground biomass ({AGB}) and carbon stocks evolution. {L}-band observations, in particular, have been shown to be very sensitive to {AGB}. {H}ere, thanks to the multiangle capabilities of the soil moisture and ocean salinity mission, a new approach to estimate {AGB} directly from multiangular {L}-band brightness temperatures ({TB}s) is proposed, thus surpassing the use of intermediate variables such as {VOD}. {T}he {E}uropean {S}pace {A}gency ({ESA}) {C}limate {C}hange {I}nitiative ({CCI}) {B}iomass maps for the years 2010, 2017, and 2018 are used as the {AGB} reference. {AGB} estimates from artificial neural networks ({ANN}) using a purely data-driven approach explained up to 88% of {AGB} variability globally; even so, a decrease in retrieval performance was observed when models are applied to data from years different than the year used for their training. {A} new training methodology based on multiyear training sets is presented, leading to results showing more stability for temporal analyses. {T}he best set of predictors and an optimal learning dataset configuration are proposed based on an assessment of the accuracy of the estimates. {T}he {ANN} methodology using {TB}s is a promising alternative with respect to the common method of using a parametric function to estimate {AGB} from {VOD}. {ANN}s {AGB} estimates showed a higher correlation with {CCI} {AGB} maps ({R}-2 similar to 0.87 instead of similar to 0.84) and presented a stronger agreement with their spatial structure and less differences in residual maps.},
  keywords = {{A}bove-ground biomass ({AGB}) ; forest biomass ; {L}-band ; machine learning ; neural networks ; passive microwaves ({PMW}s) ; soil moisture and ocean salinity ({SMOS})},
  booktitle = {},
  journal = {{IEEE} {J}ournal of {S}elected {T}opics in {A}pplied {E}arth {O}bservations and {R}emote {S}ensing},
  volume = {16},
  numero = {},
  pages = {5813--5827},
  ISSN = {1939-1404},
  year = {2023},
  DOI = {10.1109/jstars.2023.3285288},
  URL = {https://www.documentation.ird.fr/hor/fdi:010090042},
}