2008ApJ...687..608K

Debris disks around main-sequence stars are believed to derive from planetesimal populations that have accreted at early epochs and survived possible planet formation processes. While debris disks must contain solids in a broad range of sizes–from big planetesimals down to tiny dust grains–debris disk observations are only sensitive to the dust end of the size distribution. Collisional models of debris disks are needed to ``climb up'' the ladder of the collisional cascade, from dust toward parent bodies, representing the main mass reservoir of the disks. We have used our collisional code to generate five disks around a Sun-like star, assuming planetesimal belts at 3, 10, 30, 100, and 200 AU with 10 times the Edgeworth-Kuiper Belt mass density, and to evolve them for 10 Gyr. Along with an appropriate scaling rule, this effectively yields a three-parametric set of reference disks (initial mass, location of planetesimal belt, and age). For all the disks, we have generated spectral energy distributions (SEDs), assuming homogeneous spherical astrosilicate dust grains. A comparison between generated and actually observed SEDs yields estimates of planetesimal properties (location, total mass, etc.). As a test and a first application of this approach, we have selected five disks around Sun-like stars with well-known SEDs. In four cases, we have reproduced the data with a linear combination of two disks from the grid (an ``asteroid belt'' at 3 AU and an outer ``Kuiper Belt''); in one case a single, outer component was sufficient. The outer components are compatible with ``large Kuiper Belts'' of 0.2-50 Earth masses (in bodies up to 100 km in size) with radii of 100-200 AU.