@article{fdi:010077974,
  title = {{E}valuation of backscattering models and support vector machine for the retrieval of bare soil moisture from {S}entinel-1 {D}ata},
  author = {{E}zzahar, {J}. and {O}uaadi, {N}. and {Z}ribi, {M}. and {E}lfarkh, {J}. and {A}ouade, {G}. and {K}habba, {S}. and {E}r-{R}aki, {S}. and {C}hehbouni, {A}bdelghani and {J}arlan, {L}ionel},
  editor = {},
  language = {{ENG}},
  abstract = {{T}he main objective of this work was to retrieve surface soil moisture ({SSM}) by using scattering models and a support vector machine ({SVM}) technique driven by backscattering coefficients obtained from {S}entinel-1 satellite images acquired over bare agricultural soil in the {T}ensfit basin of {M}orocco. {T}wo backscattering models were selected in this study due to their wide use in inversion procedures: the theoretical integral equation model ({IEM}) and the semi-empirical model ({O}h). {T}o this end, the sensitivity of the {SAR} backscattering coefficients at {VV} (sigma vv degrees) and {VH} (sigma vh degrees) polarizations to in situ soil moisture data were analyzed first. {A}s expected, the results showed that over bare soil the sigma vv degrees was well correlated with {SSM} compared to the sigma vh degrees, which showed more dispersion with correlation coefficients values (r) of about 0.84 and 0.61 for the {VV} and {VH} polarizations, respectively. {A}fterwards, these values of sigma vv degrees were compared to those simulated by the backscatter models. {I}t was found that {IEM} driven by the measured length correlation {L} slightly underestimated {SAR} backscatter coefficients compared to the {O}h model with a bias of about -0.7 d{B} and -1.2 d{B} and a root mean square ({RMSE}) of about 1.1 d{B} and 1.5 d{B} for {O}h and {IEM} models, respectively. {H}owever, the use of an optimal value of {L} significantly improved the bias of {IEM}, which became near to zero, and the {RMSE} decreased to 0.9 d{B}. {T}hen, a classical inversion approach of sigma vv degrees observations based on backscattering model is compared to a data driven retrieval technic ({SVM}). {B}y comparing the retrieved soil moisture against ground truth measurements, it was found that results of {SVM} were very encouraging and were close to those obtained by {IEM} model. {T}he bias and {RMSE} were about 0.28 vol.% and 2.77 vol.% and -0.13 vol.% and 2.71 vol.% for {SVM} and {IEM}, respectively. {H}owever, by taking into account the difficultly of obtaining roughness parameter at large scale, it was concluded that {SVM} is still a useful tool to retrieve soil moisture, and therefore, can be fairly used to generate maps at such scales.},
  keywords = {soil moisture ; synthetic aperture radar ({SAR}) ; {S}entinel-1 ; semi-empirical and theoretical backscatter models ; support vector ; machine ; bare soil ; {MAROC}},
  booktitle = {},
  journal = {{R}emote {S}ensing},
  volume = {12},
  numero = {1},
  pages = {art. 72 [20 p.]},
  year = {2020},
  DOI = {10.3390/rs12010072},
  URL = {https://www.documentation.ird.fr/hor/fdi:010077974},
}