Abstract
The Belgian-ESA PROBA-V(egetation) satellite was successfully launched on May 7, 2013, with the aim of ensuring data continuity for the SPOT-VEGETATION user community. Users are served with daily global land coverage data in four multispectral bands at both 1/3 km and 1 km resolution. Absolutely essential to this continuity is that accurate radiometric calibration is established during commissioning and maintained throughout the mission.
Due to size, weight and power constraints, the satellite does not contain on-board calibration devices and thus the in-flight radiometric calibration relies solely on vicarious calibration methods. A set of distinct methods is being used, including calibration over deep convective clouds, Rayleigh, bright desert surfaces and snow. They complement each other, allow to cross-check results and together they ensure that all calibration requirements are reached. The consistency with other missions, such as ENVISAT-MERIS and SPOT-VGT is, amongst other, verified through cross-calibration over the Libya 4 desert site.
In addition to this, a lunar calibration method has been implemented. Lunar observations are acquired monthly with the instrument using a constant phase angle. The images are compared to the ROLO lunar reflectance model of USGS. The ROLO model, derived from thousands of lunar observations, simulates a disc equivalent reflection. When applied to observations with constant well known phase angle, outstanding temporal stability can be achieved with this model.
A full-fledged implementation of the model has been done within VITO’s OSCAR (Optical Sensor CAlibration with simulated Radiances) facility, automatically generating calibration results when lunar images are acquired. The method allows to obtain multi-temporal calibration of the instrument. The absolute and multi temporal results are compared against those of the desert calibration method, inter-band results are compared against results of the deep convective clouds method.
During the presentation, a brief overview will be given on the complete set of calibration methods used. Next, the results of the lunar calibration and their comparison against the other methods will be discussed in more detail.
Vicarious Calibration of PROBA-V: One Year in Orbit
The Belgian-ESA PROBA-V(egetation) satellite was successfully launched on May 7, 2013, with the aim of ensuring data continuity for the SPOT-VEGETATION user community. Users are served with daily global land coverage data in four multispectral bands at both 1/3 km and 1 km resolution. Absolutely essential to this continuity is that accurate radiometric calibration is established during commissioning and maintained throughout the mission.
Due to size, weight and power constraints, the satellite does not contain on-board calibration devices and thus the in-flight radiometric calibration relies solely on vicarious calibration methods. A set of distinct methods is being used, including calibration over deep convective clouds, Rayleigh, bright desert surfaces and snow. They complement each other, allow to cross-check results and together they ensure that all calibration requirements are reached. The consistency with other missions, such as ENVISAT-MERIS and SPOT-VGT is, amongst other, verified through cross-calibration over the Libya 4 desert site.
In addition to this, a lunar calibration method has been implemented. Lunar observations are acquired monthly with the instrument using a constant phase angle. The images are compared to the ROLO lunar reflectance model of USGS. The ROLO model, derived from thousands of lunar observations, simulates a disc equivalent reflection. When applied to observations with constant well known phase angle, outstanding temporal stability can be achieved with this model.
A full-fledged implementation of the model has been done within VITO’s OSCAR (Optical Sensor CAlibration with simulated Radiances) facility, automatically generating calibration results when lunar images are acquired. The method allows to obtain multi-temporal calibration of the instrument. The absolute and multi temporal results are compared against those of the desert calibration method, inter-band results are compared against results of the deep convective clouds method.
During the presentation, a brief overview will be given on the complete set of calibration methods used. Next, the results of the lunar calibration and their comparison against the other methods will be discussed in more detail.