Abstract
The on-orbit Landsat 9 Operational Land Imager (OLI) absolute radiometric calibration was adjusted during commissioning to improve the agreement with Landsat 8 OLI based on five days of underfly collections. These adjustments were cross-checked with data from a near simultaneous solar diffuser collects made with both platforms. Post commissioning the two platforms follow orbits that set their ground tracks to be eight days apart, limiting direct earth scene comparisons between the two OLIs. The current work describes an approach relying on the solar diffuser data to evaluate the radiometric consistency between the two OLI radiance scales for the period of the two years following the underfly adjustments. Background information about the solar diffuser panels pre-launch calibrations, and the Landsat 8 on orbit working diffuser aging degradation will be reviewed. During routine operation intercomparisons between Landsat 8 and 9 OLI make use of the once per week calibrations based on the working solar diffuser panel collects as well as the twice per year pristine diffuser collects. Data from the solar diffuser collects, lead to generating a continuum of relative radiometric scales comparisons. Typical diffuser measurements from a given OLI occurs within one day of the other minimizing effects of sun-sensor geometry such as OLI-Sun distance and beta-angle differences between the sun and the Landsat 8/9 orbit. Newly added near simultaneously planned diffuser collects, eliminate the need to correct for OLI-Sun distance and the artifacts induced by residual error related to such distance normalization step. The sensitivity of the retrievals to these effects as well as impacts from other effects such as spacecraft orbit-maintenance activities is discussed. The results of the diffuser-based intercomparisons show that the radiometric calibration of Landsat 9 OLI after more than 2 years, still agrees with that of Landsat 8 OLI at a level <0.3% in all spectral bands. Such level of agreement is within the uncertainty for the values obtained from the underfly early in the mission. The results presented indicate that this method can be applied to larger constellations of sensors that make use of solar diffusers as on-orbit calibrations leading to a single calibration scale per constellation that can be determined early in the mission and tracked overtime.
On-Orbit Temporal Trending of the OLI Radiance Scale Cross-Cal for the Landsat Mini-Constellation Using Solar Diffuser Panels
The on-orbit Landsat 9 Operational Land Imager (OLI) absolute radiometric calibration was adjusted during commissioning to improve the agreement with Landsat 8 OLI based on five days of underfly collections. These adjustments were cross-checked with data from a near simultaneous solar diffuser collects made with both platforms. Post commissioning the two platforms follow orbits that set their ground tracks to be eight days apart, limiting direct earth scene comparisons between the two OLIs. The current work describes an approach relying on the solar diffuser data to evaluate the radiometric consistency between the two OLI radiance scales for the period of the two years following the underfly adjustments. Background information about the solar diffuser panels pre-launch calibrations, and the Landsat 8 on orbit working diffuser aging degradation will be reviewed. During routine operation intercomparisons between Landsat 8 and 9 OLI make use of the once per week calibrations based on the working solar diffuser panel collects as well as the twice per year pristine diffuser collects. Data from the solar diffuser collects, lead to generating a continuum of relative radiometric scales comparisons. Typical diffuser measurements from a given OLI occurs within one day of the other minimizing effects of sun-sensor geometry such as OLI-Sun distance and beta-angle differences between the sun and the Landsat 8/9 orbit. Newly added near simultaneously planned diffuser collects, eliminate the need to correct for OLI-Sun distance and the artifacts induced by residual error related to such distance normalization step. The sensitivity of the retrievals to these effects as well as impacts from other effects such as spacecraft orbit-maintenance activities is discussed. The results of the diffuser-based intercomparisons show that the radiometric calibration of Landsat 9 OLI after more than 2 years, still agrees with that of Landsat 8 OLI at a level <0.3% in all spectral bands. Such level of agreement is within the uncertainty for the values obtained from the underfly early in the mission. The results presented indicate that this method can be applied to larger constellations of sensors that make use of solar diffusers as on-orbit calibrations leading to a single calibration scale per constellation that can be determined early in the mission and tracked overtime.