Tropical rainforest investigation in Brazil using multitemporal ERS-1 SAR data
Methods were developed to use earth resource satellite (ERS)-1 data for forest/nonforest mapping and rainforest monitoring and, if possible, to differentiate landuse intensity. Several classification approaches of ERS-1 synthetic aperture radar (SAR) data have been tested. The evidence based classifier EBIS (evidence based interpretation of satellite imagery) proved to be a suitable tool for forest/nonforest classification and diversification of landuse, especially when using a new version including textural classification. The separation of forest/nonforest could be improved by multitemporal ERS-1 SAR data of the dry and wet seasons. The approach developed in an area of about 50 km by 60 km around the town Sena Madureira was successfully transfered to a region North of Rio Branco, the capital of Acre.
Bibliographic Reference: Paper presented: Second ERS-1 Application Workshop, London (GB), December 6-8, 1995
Availability: Available from (1) as Paper EN 41055 ORA
Record Number: 199810046 / Last updated on: 1998-02-03
Original language: en
Available languages: en