We present basic properties of protostellar disks in the embedded phase of star formation ( EPSF ) , which is difficult to probe observationally using available observational facilities .
We use numerical hydrodynamics simulations of cloud core collapse and focus on disks formed around stars in the 0.03 - 1.0 ~ { } M _ { \sun } mass range .
Our obtained disk masses scale near-linearly with the stellar mass .
The mean and median disk masses in the Class 0 and I phases ( M _ { d,C 0 } ^ { mean } = 0.12 ~ { } M _ { \odot } , M _ { d,C 0 } ^ { mdn } = 0.09 ~ { } M _ { \odot } and M _ { d,CI } ^ { mean } = 0.18 ~ { } M _ { \odot } , M _ { d,CI } ^ { mdn } = 0.15 ~ { } M _ { \odot } , respectively ) are greater than those inferred from observations by ( at least ) a factor of 2–3 .
We demonstrate that this disagreement may ( in part ) be caused by the optically thick inner regions of protostellar disks , which do not contribute to millimeter dust flux .
We find that disk masses and surface densities start to systematically exceed that of the minimum mass solar nebular for objects with stellar mass as low as M _ { \ast } = 0.05 - 0.1 ~ { } M _ { \odot } .
Concurrently , disk radii start to grow beyond 100 AU , making gravitational fragmentation in the disk outer regions possible .
Large disk masses , surface densities , and sizes suggest that giant planets may start forming as early as in the EPSF , either by means of core accretion ( inner disk regions ) or direct gravitational instability ( outer disk regions ) , thus breaking a longstanding stereotype that the planet formation process begins in the Class II phase .