An approach to cartographic exploitation of ERS imagery in tropical area (Senegal - West Africa): principles and first results.

   William Stroobants (1), Marilyne B? (2), Jean-Paul Rudant (3), Edtih
   Th?odorakopoulos (1).

  (1) Cellule d'Etudes en G?ographie Num?rique, 16 Bis Av. Prieur de la
      C?te d'Or, 94114 Arcueil Cedex, France. T?l: 01 42 31 88 26, Fax: 01
      42 31 99 77, E-mail: wstrob@ctme.etca.fr, edith@ctme.etca.fr.
  (2) Universit? de Dakar - Unit? de traitement d'images spatiales.
      T?l: 221 27 34 13, Fax: 221 27 34 13, E-mail: diarra@chris.isra.sn.
  (3)  Universit? de Marne-La-Vall?e, Laboratoire de physique et chimie des
      g?omat?riaux. 2 rue de la Butte Verte 93166 Noisy-Le-Grand, France.
      T?l: 01 44 27 38 70, Fax: 01 44 27 50 85, E-mail: jpr@lgs.jussieu.fr.

  ABSTRACT

  This paper first addresses the main steps of cartographic production in
  tropical areas using ERS SAR imagery. We analyse the needs for cartographic
  products of these countries with regard to their landscape. A study has
  demonstrated the lack of up-to-date maps at the small and medium scales in
  such tropical countries. Sattelite radar imagery, thanks to its day/night
  capture ability and relative insensivity to weather conditions, is an
  attractive source for updating these maps. We show that the needs of these
  countries can be satisfied by the fusion of external data with radar imagery.
  Then, we focus on Senegal and to be more precise on the Saloum delta where the   great diversity of landscapes and the strong temporal dynamics of the surface   phenomena allow us to test our multitemporal approach to ERS imagery
  processing. The interpretation of these images is dictated by the reflectivity   of electromagnetics waves on complex surfaces. The roughness of these
  surfaces are determined by oceanic and climatic conditions which are most
  often imprecisely known. We demonstrate with examples that larger landscape
  units such as forests, cultivated parcels and mangroves can be correctly
  interpretable. But multitemporal fusion of SAR images may limit the final
  cartographic product quality if the surface temporal phenoma frequency is
  larger than the number of images we are using. However the complementarity of
  SAR data such as ERS and J-ERS with optical imagery (Landsat, Spot) allows us
  to remove some uncertainties of the interpretation.

  keywords: Cartography, SAR imagery, filtering, photo-interpretation, Senegal.

Keywords: ESA European Space Agency - Agence spatiale europeenne, observation de la terre, earth observation, satellite remote sensing, teledetection, geophysique, altimetrie, radar, chimique atmospherique, geophysics, altimetry, radar, atmospheric chemistry