The impact of the presence of water ice on the analysis of debris disk observations

Context. The analysis of debris disk observations is often based on the assumption of a dust phase composed of compact spherical grains consisting of astronomical silicate. Instead, observations indicate the existence of water ice in debris disks.
Aims: We quantify the impact of water ice as a potential grain constituent in debris disks on the disk parameter values estimated from photometric and spatially resolved observations in the mid- and far-infrared.
Methods: We simulated photometric measurements and radial profiles of debris disks containing water ice and analyzed them by applying a disk model purely consisting of astronomical silicate. Subsequently, we quantified the deviations between the derived and the true parameter values. As stars in central positions we discuss a β Pic sibling and main-sequence stars with spectral types ranging from A0 to K5. To simulate observable quantities we employed selected observational scenarios regarding the choice of wavelengths and instrument characteristics.
Results: For the β Pic stellar model and ice fractions ≥50% the derived inner disk radius is biased by ice sublimation toward higher values. However, the derived slope of the radial density profile is mostly unaffected. Along with an increasing ice fraction, the slope of the grain size distribution is overestimated by up to a median factor of ~1.2 for an ice fraction of 90%. At the same time, the total disk mass is underestimated by a factor of ~0.4. The reliability of the derived minimum grain size strongly depends on the spectral type of the central star. For an A0-type star the minimum grain size can be underestimated by a factor of ~0.2, while for solar-like stars it is overestimated by up to a factor of ~4-5. Neglecting radial profile measurements and using solely photometric measurements, the factor of overestimation increases for solar-like stars up to ~7-14.