| During the Descent of the Huygens probe towards Titan’s surface, the Descent
Imager/Spectral Radiometer (DISR) aboard the probe took about 400Â images. About half of
these were taken with low enough altitude and emission angle to show surface
features through the optically thick atmosphere. Soderblom et al. (2007) used two
stereo pairs of images to photometrically reconstruct the surface topography of two
areas.
Here, we investigate the possibilities to retrieve the surface topography of the complete
landing site region by means of shape from shading. We base the reconstruction on the
stitched image mosaic which has been put together by Karkoschka et al. (2007). In this
mosaic, the atmospheric component in the observed surface brightnesses has already been
removed. In order to take into account the complete illumination from Titan’s sky for the
shape from shading reconstruction, we use the results of the radiative transfer modelling by
Tomasko et al. 2008.
For a shape from shading approach, it is usually assumed that the intrinsic surface albedo
is constant, and that all observed brightness variations are due to topographic shading. For the
Huygens landing site region, this is clearly not the case. The observed brightness distribution
is bimodal, and a dichotomy of bright and dark areas is evident. We investigate different a
priori assumptions to constrain the underdetermined problem of simultaneously retrieving
topography and surface albedo variation. E. g., assuming that there are only two types of
surface material (a bright one and a dark one), we allow for only two different values for
the surface albedo and attribute all additional brightness variations to topographic
shading.
It has been argued that the darker “river beds” withing the brighter areas contain deposits
of the same material that makes up the darker areas, however, Keller et al. (2008) have shown
that the magnitude of darkening observed in the river beds could also be caused just by
topographic shading due to moderate terrain. We assess different assumptions about the
nature of the river beds by comparing their consequences for the retrieved topographic
models. |