Toward a direct comparison of field and laboratory goniometer measurements

Field and laboratory goniometers are widely used in the remote sensing community to assess spectrodirectional reflection properties of selected targets. Even when the same target and goniometer system are used, field and laboratory results cannot directly be compared due to inherent differences, mainly in the illumination conditions since actual goniometers measure a hemispherical-conical reflectance in the field and a biconical reflectance in the lab. Yet, the ability to compare and combine measurements from different instrumental designs is critical to ensure sensor cross-calibration and for all applications that rely on measurements obtained with both types of instruments. One approach to this problem consists in retrieving the bidirectional reflectance distribution function (BRDF) of the targets of interest for each experimental setup and to compare these, since theoretically they are independent of the particular conditions of illumination and observation. This involves a correction for diffuse incoming radiation in the case of field measurements, and a correction for conicity and inhomogeneity of illumination in the case of laboratory measurements. In this paper, we present a novel BRDF retrieval scheme for typical laboratory goniometers and compare it with the usual correction method assuming Lambertian behavior. We then discuss the first results of measurements and BRDF retrievals using the field and laboratory goniometer systems of the Remote Sensing Laboratories of the University of Zurich, which share the exact observation geometry, on the same inert, highly anisotropic target.