@article{fdi:010071942,
  title = {{A}utomated regolith landform mapping using airborne geophysics and remote sensing data, {B}urkina {F}aso, {W}est {A}frica},
  author = {{M}etelka, {V}. and {B}aratoux, {L}enka and {J}essell, {M}ark and {B}arth, {A}. and {J}ezek, {J}. and {N}aba, {S}.},
  editor = {},
  language = {{ENG}},
  abstract = {{W}e have studied the regolith landform distribution in the area of {G}aoua, western {B}urkina {F}aso, using an integration of geophysical and remote sensing data. {C}oncentration maps of {K}, {T}h, {U}, as well as their ratios, were computed from airborne gamma-ray spectrometry data to assess the geochemical composition of the regolith. {T}he mineralogy of the surfaces was mapped via the analysis of multispectral {ASTER} and {L}andsat scenes. {P}auli-decomposition data retrieved from polarimetric {ALOS} {PALSAR} and {R}adarsat-2 images were included to characterize the surface properties of the regolith material. {M}orphometric variables such as slope, curvature, and relative relief were derived from the {SRTM} digital elevation model to quantify the topographic parameters of the different regolith landforms. {A}n artificial neural network implementation, {ADVANGEO}, was then employed to extract four basic regolith landform units from the satellite and airborne data. {R}elic ferruginous duricrusts rich in hematite and goethite belonging to the {H}igh glacis, erosional surfaces represented by rock outcrops and sub outcrops, alluvial sediments, and soft pediment materials of the {M}iddle and {L}ow glacis were mapped successfully in the region. {T}he results were compared with the existing geomorphological maps, an independent visual classification, and field observations. {W}e found that the distribution and shape of the iron-rich duricrusts are more accurate than portrayed in the current maps. {T}he best results, with an overall accuracy of 94.21% and a kappa value of 0.92, were obtained for a dataset consisting of gamma-ray spectrometry data combined with derivatives of the {SRTM} digital elevation model augmented by {L}andsat, and polarimetric radar data. {T}he approach demonstrates for the first time the potential of machine learning in regolith landform mapping. {T}he proposed combined analysis of airborne geophysics and remote sensing data can be adopted easily in other regions with similar long-term lateritic weathering histories worldwide.},
  keywords = {{R}egolith ; {L}andform mapping ; {R}emote sensing ; {A}irborne geophysics ; {N}eural networks ; {G}amma-ray ; {SRTM} ; {R}adar ; {M}ultispectral ; {BURKINA} {FASO} ; {GAOUA}},
  booktitle = {},
  journal = {{R}emote {S}ensing of {E}nvironment},
  volume = {204},
  numero = {},
  pages = {964--978},
  ISSN = {0034-4257},
  year = {2018},
  DOI = {10.1016/j.rse.2017.08.004},
  URL = {https://www.documentation.ird.fr/hor/fdi:010071942},
}