AtomNeb – Library for Atomic Data of Ionized Nebulae

AtomNeb is an atomic database useful for nebular spectral analysis, plasma diagnostics, and abundance analysis.


AtomNeb-py is a library written in Python for reading atomic data from AtomNeb, which is a database containing atomic data stored in the Flexible Image Transport System (FITS) file format for collisionally excited lines and recombination lines typically observed in spectra of ionized gaseous nebulae. The AtomNeb database were generated for use in pyEQUIB, proEQUIB, and other nebular spectral analysis tools.

Collisionally Excited Lines

AtomNeb for collisionally excited lines contains sets of atomic datasets, which include energy levels (Ej), collision strengths (Ωij), and transition probabilities (Aij) of the most ions commonly observed in ionized nebulae.

The atomic datasets for collisionally excited lines are as follows:

Each dataset contains the following atomic data FITS files: AtomElj.fits for Energy Levels (Ej), AtomOmij.fits for Collision Strengths (Ωij), and AtomAij.fits for Transition Probabilities (Aij).

Recombination Lines

AtomNeb for recombination lines contains sets of effective recombination coefficients (αeff) of recombination lines of H I, He I, He II, C I, C II, C III, C VI, N II, N III, N IV, N V, N VI, N VII, O II, O III, O IV, O V, O VI, O VIII, and Ne II ions typically observed in ionized nebulae, as well as Branching ratios (Br) of O II and N II lines.

The atomic datasets for recombination lines are as follows:


Dependent Python Packages

This package requires the following packages:

The previous version relied on pandas, but all the data structures were changed from pandas.DataFrame to those defined by NumPy that speed up the computations and reduce the memory usage.

  • To get this package with all the FITS file, you can simply use git command as follows:
git clone
cd AtomNeb-py/atomic-data-rc/
tar -xvf *.fits.tar.gz

To install the last version, all you should need to do is

$ python install

To install the stable version, you can use the preferred installer program (pip):

$ pip install atomneb

or you can install it from the cross-platform package manager conda:

$ conda install -c conda-forge atomneb

How to Use

The Documentation of the functions provides in detail in the API Documentation ( There are two main categories: collisionally excited lines (CEL) and recombination lines (RC).

See Jupyter Notebooks: Notebooks.ipynb

Run Jupyter Notebooks on Binder

  • The atomic data for collisionally excited lines (CEL) contain Energy Levels (Ej), Collision Strengths (Ωij), and Transition Probabilities (Aij). We have four atomic datasets for them: collection, chianti52, chianti60, and chianti70.

    You need to load the atomneb library as follows:

    import atomneb
    import numpy as np
    import os
    base_dir = '.'
    data_dir = os.path.join('atomic-data', 'chianti70')
    atom_elj_file = os.path.join(base_dir,data_dir, 'AtomElj.fits')
    atom_omij_file = os.path.join(base_dir,data_dir, 'AtomOmij.fits')
    atom_aij_file = os.path.join(base_dir,data_dir, 'AtomAij.fits')
    elj_data_list = atomneb.read_elj_list(atom_elj_file)
    omij_data_list = atomneb.read_omij_list(atom_omij_file)
    aij_data_list = atomneb.read_aij_list(atom_aij_file)

    Now you have access to:

    • Energy Levels (Ej):

      oiii_elj_data = atomneb.read_elj(atom_elj_file, atom, ion, level_num=6)

      which gives:

      0.00000      1.00000      2.00000      2.00000      0.00000      2.00000
      0.00000      113.200      306.200      20273.30     43185.69     60324.80
    • Collision Strengths (Ωij):

      oiii_omij_data = atomneb.read_omij(atom_omij_file, atom, ion)

      which gives:

      0       1       1       1       1       ...
      0       2       3       4       5       ...
      100.0      158.50       251.20       398.10       631.0       ...
    • Transition Probabilities (Aij):

      oiii_aij_data = atomneb.read_aij(atom_aij_file, atom, ion)

      which gives:

      0.0000   2.5969e-05       0.0000   2.3220e-06      ...
  • The atomic data for recombination lines (RC) contain effective recombination coefficients (αeff) of emission lines from different collections: RC Collection, SH95 Collection, PPB91 Collection, PFSD12 He I data, FSL13 N II data, and SSB17 O II data.

    You need to load the atomneb libary:

    import atomneb
    import numpy as np
    import os
    base_dir = '.'
    data_rc_dir = os.path.join('atomic-data-rc')

    Now you have access to effective recombination coefficients (αeff) of the following collections:

    • RC Collection:

      atom_rc_file = os.path.join(base_dir,data_rc_dir, 'rc_collection.fits')
      cii_rc_data = atomneb.read_aeff_collection(atom_rc_file, atom, ion)
      n_line = len(cii_rc_data['wavelength'])
      for i in range(0, n_line):
           print(cii_rc_data['wavelength'][i], cii_rc_data['a'][i],
                 cii_rc_data['b'][i], cii_rc_data['c'][i],
                 cii_rc_data['d'][i], cii_rc_data['f'][i])

      which gives:

      914.00000      0.69280000     0.021400000    -0.016300000     -0.24310000     -0.88000000
      962.00000       1.0998000   -0.0042000000    -0.027900000     -0.22940000     -0.96560000
      997.00000      0.78210000     -0.36840000   0.00030000000     -0.12170000     -0.78740000
    • SH95 Collection:

      atom_rc_file = os.path.join(base_dir,data_rc_dir, 'rc_SH95.fits')
      hi_rc_data = atomneb.read_aeff_sh95(atom_rc_file, atom, ion)

      which gives:

      100.00000       500.00000       0.0000000   4.2140000e-27   1.7560000e-27   1.0350000e-27
    • PPB91 Collection:

      atom_rc_file = os.path.join(base_dir,data_rc_dir, 'rc_PPB91.fits')
      cii_rc_data = atomneb.read_aeff_ppb91(atom_rc_file, atom, ion)
      n_line = len(cii_rc_data['wavelength'])
      for i in range(0, n_line):
         print(cii_rc_data['ion'][i], cii_rc_data['case1'][i], cii_rc_data['wavelength'][i],
               cii_rc_data['a'][i], cii_rc_data['b'][i], cii_rc_data['c'][i],
               cii_rc_data['d'][i], cii_rc_data['br'][i], cii_rc_data['q'][i], cii_rc_data['y'][i])

      which gives:

      C2+A       9903.4600      0.69700000     -0.78400000       4.2050000      0.72000000       1.0000000       1.6210000
      C2+A       4267.1500       1.0110000     -0.75400000       2.5870000      0.71900000      0.95000000       2.7950000
    • PFSD12 He I data:

      atom_rc_file = os.path.join(base_dir,data_rc_dir, 'rc_he_ii_PFSD12.fits')
      hei_rc_data = atomneb.read_aeff_he_i_pfsd12(atom_rc_file, atom, ion)
      hei_rc_data_wave = atomneb.read_aeff_he_i_pfsd12(atom_rc_file, atom, ion, wavelength=True)

      which gives:

      5000.0000       10.000000      -25.379540      -25.058970      -25.948440      -24.651820      -25.637660
    • FSL13 N II data:

      atom_rc_file = os.path.join(base_dir,data_rc_dir, 'rc_n_iii_FSL13.fits')
      wavelength_range=[4400.0, 7100.0]
      nii_rc_data = atomneb.read_aeff_n_ii_fsl13(atom_rc_file, atom, ion, wavelength_range)
      nii_rc_data_wave = atomneb.read_aeff_n_ii_fsl13(atom_rc_file, atom, ion, wavelength_range, wavelength=True)
      n_line = len(hei_rc_data_wave['wavelength'])
      for i in range(0, n_line):
         print(nii_rc_data_wave['wavelength'][i], nii_rc_data_wave['tr'][i], nii_rc_data_wave['trans'][i])

      which gives:

      255.000      79.5000      47.3000      12.5000      6.20000      4.00000      2.86000
      258.000      54.4000      29.7000      7.92000      4.11000      2.72000      2.00000
      310.000      48.1000      23.7000      5.19000      2.55000      1.65000      1.21000
      434.000      50.3000      23.2000      4.71000      2.26000      1.45000      1.05000
      6413.23 6g - 4f2p6g G[9/2]o4 - 2p4f F[7/2]e3
      6556.32 6g - 4f2p6g G[9/2]o5 - 2p4f G[7/2]e4
      6456.97 6g - 4f2p6g G[9/2]o5 - 2p4f F[7/2]e4
      6446.53 6g - 4f2p6g F[7/2]o3 - 2p4f D[5/2]e2
      6445.34 6g - 4f2p6g F[7/2]o4 - 2p4f D[5/2]e3
    • SSB17 O II data: You first need to unpack rc_o_iii_SSB17_orl_case_b.fits.tar.gz, e.g.:

      tar -xvf rc_o_iii_SSB17_orl_case_b.fits.tar.gz

      If you need to have access to the full dataset (entire wavelengths, case A and B):

      tar -xvf rc_o_iii_SSB17.fits.tar.gz

      Please note that using the entire atomic data will make your program very slow and you may need to have a higher memory on your system. Without the above comment, as default, the cose uses rc_o_iii_SSB17_orl_case_b.fits. You can also unpack them using tarfile shown below:

      import atomneb
      import numpy as np
      import os
      import tarfile
      base_dir = '.'
      data_rc_dir = os.path.join('atomic-data-rc')
      atom_rc_file = os.path.join(base_dir,data_rc_dir, 'rc_o_iii_SSB17_orl_case_b.fits')
      atom_rc_file_tar_gz = os.path.join(base_dir,data_rc_dir, 'rc_o_iii_SSB17_orl_case_b.fits.tar.gz')
      atom_rc_path = os.path.join(base_dir,data_rc_dir)
      tar =, "r:gz")
      atom = 'o'
      ion = 'iii' # O II
      case1 = 'B'
      wavelength_range=[5320.0, 5330.0]
      oii_rc_data = atomneb.read_aeff_o_ii_ssb17(atom_rc_file, atom, ion, case1, wavelength_range)
      oii_rc_data_wave = atomneb.read_aeff_o_ii_ssb17(atom_rc_file, atom, ion, case1, wavelength_range, wavelength=True)
      n_line = len(oii_rc_data_wave['wavelength'])
      for i in range(0, n_line):
         print(oii_rc_data_wave['wavelength'][i], oii_rc_data_wave['lower_term'][i], oii_rc_data_wave['upper_term'][i])

      which gives:

      1.64100e-30  1.60000e-30  1.56400e-30  1.54100e-30  1.52100e-30  1.50900e-30
      5327.17 2s22p2(1S)3p 2Po
      5325.42 2s22p2(1S)3p 2Po
      5327.18 2s22p2(1D)3d 2Ge
      5326.84 2s22p2(1D)3d 2Ge


For more information on how to use the API functions from the AtomNeb Python package, please read the API Documentation published on



Using the AtomNeb Python package in a scholarly publication? Please cite these papers:

  author = {{Danehkar}, Ashkbiz},
  title = {AtomNeb Python Package, an addendum to AtomNeb: IDL Library
           for Atomic Data of Ionized Nebulae},
  journal = {Journal of Open Source Software},
  volume = {5},
  number = {55},
  pages = {2797},
  year = {2020},
  doi = {10.21105/joss.02797}

and if you use the AtomNeb IDL library:

  author = {{Danehkar}, Ashkbiz},
  title = {AtomNeb: IDL Library for Atomic Data of Ionized Nebulae},
  journal = {Journal of Open Source Software},
  volume = {4},
  number = {35},
  pages = {898},
  year = {2019},
  doi = {10.21105/joss.00898}