@article{fdi:010082761,
  title = {{R}etrieving crop albedo based on radar {S}entinel-1 and random forest approach},
  author = {{A}mazirh, {A}. and {B}ouras, {E}. and {O}livera-{G}uerra, {L}. {E}. and {E}r-{R}aki, {S}. and {C}hehbouni, {A}bdelghani},
  editor = {},
  language = {{ENG}},
  abstract = {{M}onitoring agricultural crops is of paramount importance for preserving water resources and increasing water efficiency over semi-arid areas. {T}his can be achieved by modelling the water resources all along the growing season through the coupled water-surface energy balance. {S}urface albedo is a key land surface variable to constrain the surface radiation budget and hence the coupled water-surface energy balance. {I}n order to capture the hydric status changes over the growing season, optical remote sensing becomes impractical due to cloud cover in some periods, especially over irrigated winter crops in semi-arid regions. {T}o fill the gap, this paper aims to generate cloudless surface albedo product from {S}entinel-1 data that offers a source of high spatio-temporal resolution images. {T}his can help to better capture the vegetation development along the growth season through the surface radiation budget. {R}andom {F}orest ({RF}) algorithm was implemented using {S}entinel-1 backscatters as input. {T}he approach was tested over an irrigated semi-arid zone in {M}orocco, which is known by its heterogeneity in term of soil conditions and crop types. {T}he obtained results are evaluated against {L}andsat-derived albedo with quasi-concurrent {L}andsat/{S}entinel-1 overpasses (up to one day offset), while a further validation was investigated using in situ field scale albedo data. {T}he best model-hyperparameters selection was dependent on two validation approaches ({K}-fold cross-validation 'k = 10', and holdout). {T}he more robust and accurate model parameters are those that represent the best statistical metrics (root mean square error '{RMSE}', bias and correlation coefficient '{R}'). {C}oefficient values ranging from 0.70 to 0.79 and a {RMSE} value between 0.0002 and 0.00048 were obtained comparing {L}andsat and predicted albedo by {RF} method. {T}he relative error ratio equals 4.5, which is acceptable to predict surface albedo.},
  keywords = {surface albedo ; random forest ; {S}entinel-1 ; crop vegetation ; {L}andsat ; {MAROC} ; {ZONE} {SEMIARIDE}},
  booktitle = {},
  journal = {{R}emote {S}ensing},
  volume = {13},
  numero = {16},
  pages = {3181 [17 p.]},
  year = {2021},
  DOI = {10.3390/rs13163181},
  URL = {https://www.documentation.ird.fr/hor/fdi:010082761},
}