Ground-reference techniques for the absolute radiometric calibration of MODIS

The Remote Sensing croup at the University of Arizona has used ground-based test sites for the vicarious calibration of airborne and satellite-based sensors. Past work has focused on high-spatial-resolution sensors that are well-suited to the reflectance-, irradiance-, and radiance-based methods. Application of these methods to the recently launched Moderate Resolution Imaging Spectroradiometer (MODIS) with its lower spatial resolution poses a challenge for vicarious calibration. This work presents the modifications that must be made to reflectance-, irradiance-, and radiance-based approaches in order to use them for MODIS. The reflectance-based method described here relies on ground-based measurements of the reflectance of both large- and small-scale areas of the test site as well as low-level aircraft data to scale the ground-based measurements to the spatial scale of a MODIS pixel. The radiance-based approach relies on a recently-developed airborne, non-imaging radiometer with an 80-m footprint sampling the test site with a predetermined strategy to account for the differing spatial resolutions. Because this sampling strategy depends upon the test site being use, this work describes the two primary test sites of Railroad Valley Playa in Nevada and White Sands Missile Range in New Mexico and the spectral and spatial effects these sites will have on the calibration of MODIS. Early results from application of the reflectance-based method to MODIS using data from April 2000 indicate that the radiometric response of MODIS has not changed significantly.