@article{fdi:010061761,
  title = {{S}tatistical relationship between surface {PM}10 concentration and aerosol optical depth over the {S}ahel as a function of weather type, using neural network methodology},
  author = {{Y}ahi, {H}. and {M}articorena, {B}. and {T}hiria, {S}. and {C}hatenet, {B}ernadette and {S}chmechtig, {C}. and {R}ajot, {J}ean-{L}ouis and {C}repon, {M}.},
  editor = {},
  language = {{ENG}},
  abstract = {{T}his work aims at assessing the capability of passive remote-sensed measurements such as aerosol optical depth ({AOD}) to monitor the surface dust concentration during the dry season in the {S}ahel region ({W}est {A}frica). {W}e processed continuous measurements of {AOD}s and surface concentrations for the period (2006-2010) in {B}anizoumbou ({N}iger) and {C}inzana ({M}ali). {I}n order to account for the influence of meteorological condition on the relationship between {PM}10 surface concentration and {AOD}, we decomposed the mesoscale meteorological fields surrounding the stations into five weather types having similar 3-dimensional atmospheric characteristics. {T}his classification was obtained by a clustering method based on nonlinear artificial neural networks, the so-called self-organizing map. {T}he weather types were identified by processing tridimensional fields of meridional and zonal winds and air temperature obtained from {E}uropean {C}entre for {M}edium-{R}ange {W}eather {F}orecasts ({ECMWF}) model output centered on each measurement station. {F}ive similar weather types have been identified at the two stations. {T}hree of them are associated with the {H}armattan flux; the other two correspond to northward inflow of the monsoon flow at the beginning or the end of the dry season. {A}n improved relationship has been found between the surface {PM}10 concentrations and the {AOD} by using a dedicated statistical relationship for each weather type. {T}he performances of the statistical inversion computed on the test data sets show satisfactory skills for most of the classes, much better than a linear regression. {T}his should permit the inversion of the mineral dust concentration from {AOD}s derived from satellite observations over the {S}ahel.},
  keywords = {mineral dust concentration ; aerosol optical depth ; weather types ; nonlinear artificial neural network ; {ZONE} {SAHELIENNE}},
  booktitle = {},
  journal = {{J}ournal of {G}eophysical {R}esearch.{A}tmospheres},
  volume = {118},
  numero = {23},
  pages = {13265--13281},
  ISSN = {2169-897{X}},
  year = {2013},
  DOI = {10.1002/2013jd019465},
  URL = {https://www.documentation.ird.fr/hor/fdi:010061761},
}