Radiative Transfer in Dusty Circumstellar Disks and Envelopes

C.P. Dullemond

One of the most important tools for my research is a continuum radiative transfer package called RADMC. This is a Monte Carlo code that can compute the transport of continuum radiation through an axisymmetric dusty medium surrounding a star. I also developed other radiative transfer tools, such as a program called RADICAL that can do various types of radiative transfer including non-LTE molecular line transfer. However, this program is old and will be phased out. Instead, to replace both RADMC and RADICAL, I am currently developing RADMC-3D. This includes dust and line transfer in one code, has 1-D, 2-D and 3-D capabilities with adaptive mesh refinement (AMR) etc. That will be our workhorse radiative transfer tool for the future.


Computer programs for radiative transfer

  • RADMC: An easy-to-use Monte Carlo radiative transfer code for axisymmetric dusty media surrounding a star. RADMC is very easy to use. It is basically plug-and-play, and it is not too sensitive to issues of gridding (though still care has to be taken). For most problems of dust continuum radiative transfer in axisymmetric circumstellar disks and envelopes, this is the method of choice. See the RADMC web page. A public domain version of this program will be released some time in 2008. I apologise to those who have been waiting for the public code to appear: I was not satisfied with the package yet, and only recently managed to find time to write a manual for it.
  • RADMC-3D: This is the 3-D version of RADMC. It is a completely new code, written from scratch though with many of the subroutines imported from RADMC and from RADICAL. It is much more flexible, has 1-D, 2-D and 3-D capabilities, has adaptive mesh refinement (AMR), has both dust and line transfer. The first version of the code is now ready, though lacking a number of important options. But for a sneak-preview, you can visit the RADMC-3D web page. If you are interested in becoming a beta-tester, please send me an email.
  • RADICAL: A multi-purpose 2-D radiative transfer program for axi-symmetric circumstellar clouds/envelopes/disks. It can do dust-continuum transfer as well as gas line transfer. This is an old code which is now rarely used. We will use RADMC-3D instead in the future.
  • TRANSPHERE-1D (V3): A dust continuum radiative transfer code for 1-D spherically symmetric circumstellar envelopes. The code is special in that it can cope with arbitrary optical depths without difficulty, using the method of Variable Eddington Factors. It converges within about 10 to 15 iteration (independent of tau) and has flux conservation usually within the percent level. It is specially well suited for high optical depth circumstellar envelopes / winds around e.g. OHIR stars and Class 0 protostars, but can be used for any application in which 1-D continuum radiative transfer is necessary, be it at low, intermediate or high optical depth. A scaled-down version of this program can be downloaded here. Note that this is Version-3 of the code. It now can also handle starless cores with an external interstellar radiation field.
    Author: C.P. Dullemond
  • DISKSTRUCT: The diskstruct code also has a 1-D variable eddington factor radiative transfer code inside. See diskstruct below.


Computer programs for models of passive circumstellar disks

  • CGPLUS: A simple Chiang & Goldreich model program for computing the SED of a flared irradiated passive circumstellar dust+gas disk. It includes all the additional elements of the model as discussed in the paper by Dullemond, Dominik & Natta (2001), for instance: the inner rim, the shadowing and a more consistent energy conserving treatment of the flared disk. The program is fast (it computes a model in a few seconds in the IDL version of the program). A sophisticated IDL widget interface makes it a very easy-to-use program. This interface displays the observed SED with the current model fit overplotted. Parameters can be adjusted on the fly using the input widgets, and a new model is computed. One can zoom in, plot in different forms (e.g.lin/log), and save/load the latest best-fit model. Download it here. NOTE: This model is evidently highly limited in its realism. Multi-dimensional radiative transfer codes are better!!
  • DISKSTRUCT: The diskstruct code package is the set of programs used for the papers Dullemond, van Zadelhoff and Natta (2002) A&A 389, 464, Dullemond & Natta (2003) A&A 405, 597, and Dullemond & Natta (2003) A&A 408, 161. These programs compute 1+1D disk structure models. The vertical radiative transfer is a detailed frequency-dependent radiative transfer calculation using variable eddington factor method (very similar to that of the TRANSPHERE code above, but now for 1-D vertical plane parallel problems instead of spherically symmetric problems). The codes are not very well documented, but there is a README_MANUAL file with the very basics of the code package. The rest will be up to the user to figure out. There IS, however, an example model which should work, and which could give the user a reasonable starting point. I refer to the above papers for details of the methods etc. You can download a beta (!!) release, but please keep in mind that it really IS a beta release, so beware of strange results. Everything is the responsibility of the user... You can download the zip package of the April 2009 release here or you can download the zip package of the August 2010 release here. The newest version (August 2010) has only a few minor improvements over the older one.
  • RADMC: The above listed RADMC code is also specialized in making ready-to-use disk structure models.


Some model results for downloading

For a long time these pages were simply lost (unlinked). This was just a mistake. So here are some results:
  • The flaring and self-shadowing disks of the Dullemond and Dominik 2004a paper can be found here.
  • The disk models of Dullemond, van Zadelhoff and Natta 2003 can be found here.
  • A little benchmark test for 1-D vertical disk models can be accessed here.



Last update: August, 2010
dullemon@mpia.de