Optical depth of the Martian atmosphere and surface albedo from high-resolution orbiter images

In this paper we describe and evaluate the so-called shadow method. This method can be used to estimate the optical depth of the Martian atmosphere from the differences in brightness between shadowed and sunlit regions observed from an orbiter. We present elaborate and simplified versions of the method and analyze the capabilities and the sources of errors. It proves essential to choose shadowed and sunlit comparison regions with similar surface properties. Accurate knowledge of the observing geometry, including the slopes of the observed region, is important as well, since the procedure should be corrected for the non-horizontal surface. Moreover, the elaborate version of the shadow method can be sensitive to (i) the optical model of aerosols and (ii) the assumed bi-directional reflectance function of the surface. To obtain reliable estimates, the analyzed images must have a high spatial resolution, which the HiRISE camera onboard the MRO provides. We tested the shadow method on two HiRISE images of Victoria crater (TRA_0873_1780 and PSP_001414_1780) that were taken while this crater was the exploration site of the Opportunity rover. While the rover measured optical depth τ approximately in the ranges from 0.43 to 0.53 and from 0.53 to 0.59 by imaging the sun, our shadow procedure yielded τ about 0.50 and 0.575, respectively (from the HiRISE's red images). Thus, the agreement is quite good. The obtained estimates of the surface albedo are about 0.20 and 0.17, respectively.