2014A&A...565A..24F

The probability distribution function (PDF) of the mass surface density is an essential characteristic of the structure of molecular clouds or the interstellar medium in general. Observations of the PDF of molecular clouds indicate a composition of a broad distribution around the maximum and a decreasing tail at high mass surface densities. The first component is attributed to the random distribution of gas which is modeled using a log-normal function while the second component is attributed to condensed structures modeled using a simple power-law. The aim of this paper is to provide an analytical model of the PDF of condensed structures which can be used by observers to extract information about the condensations. The condensed structures are considered to be either spheres or cylinders with a truncated radial density profile at cloud radius r_cl. The assumed profile is of the form ρ(r)=ρ_c/(1+(r/r₀)²)^n/2 for arbitrary power n where ρ_c and r₀ are the central density and the inner radius, respectively. An implicit function is obtained which either truncates (sphere) or has a pole (cylinder) at maximal mass surface density. The PDF of spherical condensations and the asymptotic PDF of cylinders in the limit of infinite overdensity ρ_c/ρ(r_cl) flattens for steeper density profiles and has a power law asymptote at low and high mass surface densities and a well defined maximum. The power index of the asymptote Σ^–γ of the logarithmic PDF (ΣP(Σ)) in the limit of high mass surface densities is given by γ=(n+1)/(n-1)-1 (spheres) or by γ=n/(n-1)-1 (cylinders in the limit of infinite overdensity).