3D Selection of ENVISAT Data for Improved Water Stage Times Series on the Rio Negro and Adjacent Wetlands (Amazon Basin)
Résumé
Time series of water stages derived from satellite radar altimetry are of very variable quality, depending on the mission, the tracking scheme and the geomorphological context of the crossing between the satellite track and the river. Also, it has long been observed that this quality is dependent on the hydrological cycle itself: stage values are often much worse at low water compared to those determined at high waters. This problem is mostly due to the data selection. Indeed, common window for data selection is mostly geographical. At low waters, measurements contaminated/distorted by non-water reflectors are more likely present within the selection window because of emerged river banks and/or islets. We have developped a processing scheme based on the GRASS freeware GIS to improve the selection of the satellite measurements and reduce the incorporation of non-water measurement within the dataset retained for estimating the water stage. This scheme has been successfully applied to the ENVISAT mission over the Rio Negro where reliable water stages could be obtained far upstream at both high and low waters, and also over adjacent wetlands.
Origine : Fichiers produits par l'(les) auteur(s)
Loading...