The python scripts presented here are to stack multiple spectra, make a composite spectrum, and fit optical emission lines with Gaussian profiles to calculate the line luminosities.

The details of the stacking method is discussed in Shivaei et al. (2018, ApJ 855, 42).

Contributions are very welcomed!

This is the driver that calls the two main codes stackspec_script.py and linefit.py.

This code stacks the spectra. The output is a fits file with eight extensions:

• Ext 1: weighted average
• Ext 2: error
• Ext 3: unweighted average
• Ext 4: median
• Ext 5: 3 sigma-clipped average
• Ext 6: standard deviation in each wavelength bin
• Ext 7: average redshift of objs contributing to each wavelength bin
• Ext 8: number of objs in each wavelength bin

This code fits the H-alpha, H-beta, and [OIII]5007 lines with single Gaussian profiles, and the [OII]3727 line with a double Gaussian profile to calculate line luminosities and their errors.

• The code uses python 3.6.1 (also compatible with 2.7.9)

• Required libraries: numpy 1.11.3 and astropy 1.3.2. Other standard libraries include glob, time, argparse, multiprocessing, subprocess, sys, os.

The driver code is stack_driver.py. Before running this code you need to modify a few lines:

1. lines 21-24: First you need to specify paths for output and source files

   • specpath: path to the directory that contains the spectra

   • tblpath: path of the catalog of your objects (should be a fits file). For this specific code, the catalog needs to have these entries:

     • 'maskname' (name of the observed mask),
     • 'id' (object ID), 'z_mosfire' (redshift),
     • 'ha6565_lum' (H-alpha luminosity),
     • 'hb4863_lum' (H-beta luminosity),
     • 'ha6565_lum_err' (H-alpha luminosity error),
     • 'hb4863_lum_err' (H-beta luminosity error),
     • 'luv' (dust-corrected UV luminosity),
     • 'ha6565_abs_flux' (H-alpha absorption flux),
     • 'hb4863_abs_flux' (H-beta absorption flux),

     In case you want to modify/remove any of these entries, edit lines 90-99 in stackspec_script.py

   • outfile: path for the output composite spectrum (should be a fits file: ~/Example.fits)

   • lineoutfile: path for the output text file that contains the line luminosity measurements (a text file, e.g. ~/Example.txt)

2. line 28:

   • Here you set the criteria for the objects you want to draw from the main catalog (tblpath above) and stack. For example:

   obj_ind=((tbl['z_mosfire'] < 2.) & (tbl['ha6565_lum'] != -999.))
   

3. lines 40 and 47:

   • Edit the paths to point to where you saved the stackspec_script.py (line 40) and linefit.py (line 47) scripts.

Once you modify the driver code, you are ready to run it through your terminal. Type in the terminal:

python stack_driver.py [arg1] [arg2] [arg3-optional]

The first argument [arg1] indicates the stack spectra normalization. You have four choices:

• 'Ha' : each individual spectra will be first normalized to H-alpha luminosity, taken from the objects catalog (tblpath, see above).
• 'Hb' : same as above, but normalized to H-beta luminosity.
• 'UV' : same as above, but normalized to UV luminosity.
• 'none' : no normalization will be used.

You can define other normalization terms in line 134 of stackspec_script.py.

The second argument [arg2] indicates which line you want to measure its luminosity at the end. You have these options:

• 'Ha' : H-alpha 6565 ang
• 'Hb' : 'H-beta' 4863 ang
• 'oiii' : [OIII] 5008 ang
• 'oii' : [OII] 3727 ang doublet

If you want to fit two lines, you can add the third optional argument [arg3-optional].

You can only call these combinations (the order is important!):

• 'Ha' 'Hb'
• 'Ha' 'oiii'
• 'Ha' 'oii'
• 'Hb' 'oiii'
• 'Hb' 'oii'
• 'oiii' 'oii'

Example:

python stack_driver.py 'Ha' 'Ha' 'Hb' 

This will normilize the individual spectra to H-alpha luminosity, stack them, and measure the Ha and Hb luminosities from the composite spectrum.

The details of the stacking method is discussed at Shivaei et al. (2018, ApJ 855, 42).