Methods for the surface reflectance retrieval from CHRIS/PROBA data over land and inland waters

The Compact High Resolution Imaging Spectrometer (CHRIS) on board the Project for On–Board Autonomy (PROBA) platform system provides the first high spatial resolution hyperspectral/multiangular remote sensing data from a satellite system, what represents a new source of information for Earth Observation purposes. When dealing with the retrieval of surface reflectance from such kind of hyperspectral data, a radiative transfer approach is commonly preferred. However, since CHRIS 2003 and 2004 data present reported calibration problems in several bands, especially in the near–infrared region, a standard atmospheric correction based on radiative transfer models should not be performed. A dedicated atmospheric correction algorithm for CHRIS/PROBA data over land is presented in this work. It consists in the combination of radiative transfer and empirical line approaches to atmospheric correction, in order to retrieve surface reflectance images free from both the atmospheric distortion and artifacts due to mis–calibration. The atmospheric optical parameters and the updated set of calibration coefficients are obtained jointly in an autonomous process, without the need for any ancillary data. Results from the application of the algorithm to CHRIS/PROBA data from the two ESA SPectra bARrax Campaign (SPARC) held at the Barrax study site (La Mancha, Spain) in 2003 and 2004 are presented in this work, focusing on the validation of the final surface reflectance using in–situ measurements acquired simultaneously to PROBA overpasses. Besides, the first version of an atmospheric correction module for inland waters, which is currently under development, is also presented, as well as the first results obtained from its application to data from the Rosarito reservoir. The potential of CHRIS/PROBA data for Earth observation purposes is shown.