A Bayesian MCMC approach to fitting Spectral Energy Distributions of AGN and galaxies

Welcome to AGNfitter!

AGNfitter is a Python algorithm implementing a fully Bayesian MCMC method to fit the spectral energy distributions (SEDs) of active galactic nuclei (AGN) and galaxies from the sub-mm to the UV. Through this method, you will be able to robustly disentangle the physical processes responsible for the emission of your sources.

You only need a catalog of photometric data (wavelengths, fluxes and errors), take a few decisions (if you wish), and you are ready to go (see Example).

AGNfitter makes use of a large library of theoretical, empirical, and semi-empirical models to characterize both the nuclear and host galaxy emission simultaneously. The model consists of four physical emission components: an accretion disk, a torus of AGN heated dust, stellar populations, and cold dust in star forming regions. A detailed presentation, test and discussion of AGNfitter can be found in https://arxiv.org/abs/1606.05648#

• Numpy version 1.6 or later
• Matplotlib version 1.4.0 or later
• Scipy
• Astropy version 1.2 or later (pip install --no-deps astropy)
• acor (pip install acor)

Installation can be done by cloning this Github repository.

After installation, let's do a quick test:

1) In example/SETTINGS_AGNfitter.py, go to def CATALOG_settings() and change

cat['path'] ='/Users/USER/AGNfitter/'

to your AGNfitter path. These test settings point to the example catalog contained in data/catalog_example.txt.

2) In the terminal, go to your AGNfitter path and start

./RUN_AGNfitter_multi.py  example/SETTINGS_AGNfitter.py

You should have a nice example in your cat['path']/OUTPUT folder.

Either make sure that the root AGNfitter directory is on your PATH or specify the full path to RUN_AGNfitter_multi.py.

TASK 0 (optional): If you wish to have a working path other than the AGNfitter code path, please change

cat['workingpath'] = cat['path']

to your costumized working path.

TASK 1: In your working path, configure your settings creating a file my_SETTINGS_AGNfitter.py. This file should be created based on the example in example/SETTINGS_AGNfitter.py (copy+paste). To get AGNfitter running this is the ONLY file you need to modify.

TASK 1a: Specify your catalog's format in:

def CATALOG_settings()
    cat['path'] ='/Users/USER/AGNfitter/'
    cat['filename'] = 'data/catalog_example.txt
    cat['filetype'] = 'ASCII' ## catalog file type: 'ASCII' or 'FITS'. 
    cat['name'] = 0#'ID'            ## If ASCII: Column index (int) of source IDs
    cat...

TASK 1b: To construct the dictionary please go to

def FILTERS_settings():
    filters['dict_zarray'] = np.arange(zmin, zmax, zinterval)

Here you can specify the redshift ranges or a redshift array you need for you catalog. The DICT_default only includes z=[0.283, 1.58] for the test. Please, consider this process takes around 0.1 minute per redshift element. This process might be lengthy but you only have to do it once.

You can use the default combination of photometric bands by leaving

    filters['Bandset'] = 'BANDSET_default'.

Otherwise, if you like, you can specify the photometric bands included in your catalog by setting

def FILTERS_settings():
    ...
    filters['Bandset'] = 'BANDSET_settings' 
    
    filters['SPIRE500']= False
    filters['SPIRE350']= True        

and assigning 'True' to the keys corresponding to the photometric bands in your catalog.

TASK 2: Run AGNfitter with

./RUN_AGNfitter_multi.py my_SETTINGS_AGNfitter.py

This will run AGNfitter in series. In general there are a few more runtime options (see below).

Done!

A careful documentation will be soon available. In the meantime, some notes are available in the Wiki as response to questions asked by users.

Please cite Calistro Rivera et al. (2016)_ if this code has achieved its purpose and contributed to your research. The BibTeX entry for the paper is:

@ARTICLE{2016arXiv160605648C,
author = {{Calistro Rivera}, G. and {Lusso}, E. and {Hennawi}, J.~F. and {Hogg}, D.~W.},
title = "{AGNfitter: A Bayesian MCMC approach to fitting spectral energy distributions of AGN}",
journal = {ArXiv e-prints},
archivePrefix = "arXiv",
eprint = {1606.05648},
keywords = {Astrophysics - Astrophysics of Galaxies, Astrophysics - Instrumentation and Methods for Astrophysics},
year = 2016,
month = jun,
adsurl = {http://adsabs.harvard.edu/abs/2016arXiv160605648C},
adsnote = {Provided by the SAO/NASA Astrophysics Data System}
}

AGNfitter is an open-source software made available under the MIT License. See the LICENSE file for details.