Atomic Line List: Instructions

When attempting to identify an observed line usually many possible candidates can be found in the line list. In order to facilitate narrowing down the number of possible identifications a selection tool is presented which allows imposing, apart from the wavelength, several additional criteria.

All fields in the request form have default values and are therefore optional

default values always include as many identifications as possible

Each field will be discussed separately below. In the examples it is assumed that Angstrom has been chosen as the wavelength unit. For simplicity we will refer to wavelengths throughout this document, even though the term frequency or photon energy would have been more appropriate for certain wavelength unit choices.

Wavelength range:

This field allows you to supply a wavelength range. This can be done in two ways: you can supply a lower and upper limit for the region (Llow, Lhigh) or, if you are only trying to identify a single line, you can supply the central wavelength and the uncertainty (Lo, ΔLo). If the first number is smaller than the second number, this implies that the first option has been chosen, and otherwise the second option.
This tool has the capability to correct wavelengths for the Doppler effect by supplying a radial velocity or cosmological redshift, and for the wavelength shift in the earths atmosphere. Hence the user does not need to do a reverse correction for either of these effects. The wavelength of any candidate line (Lc) will have been corrected for both effects (if necessary of course) before a comparison to the values entered here will be made. If a non-zero value for the radial velocity or cosmological redshift is entered, the output will contain both the Doppler-shifted wavelength and the (unshifted) laboratory wavelength. If wavelengths in air or vacuum have been chosen, the output will only contain a single column of wavelengths of the chosen type.
Note 1: It is allowed to supply multiple wavelengths ranges to be processed simultaneously. Simply type each range on a separate line (or drop them with your mouse!). Each line must contain exactly two numbers. The first number must be ≥ 0, and the second > 0. The maximum number of output lines applies to the sum-total of all requests.
Note 2: When nothing is entered in this field, the wavelength range will default to the entire range of the line list if the option Vacuum is used, and to all air wavelengths ≥ 2000 Å if the option Air is used. Entering a range with a lower limit (defined as Lo-ΔLo in the Center/Sigma case) below 2000 Å is illegal when the option Air is used. Multiplet searches, and searches using a high radial velocity, can however result in matches with (laboratory) air wavelengths below 2000 Å. Their wavelengths will be converted if the resulting air wavelength is ≥ 1850 Å, and will be shown as < 1850 Å otherwise.

Lower/Upper Limit: When this option is used, all lines with wavelengths in the range
Llow ≤ Lc ≤ Lhigh
will be included as possible identifications.

Center/Sigma: When this option is used, all lines for which the difference between the observed central wavelength and the predicted wavelength in the table is smaller than the combined uncertainty in both wavelengths, will satisfy the condition.
The computed wavelengths (Lc) in the line list all have an estimate for the 1 sigma uncertainty (ΔLc) assigned to them. When the Center/Sigma option is used these numbers are compared to the user supplied wavelength (Lo) and its uncertainty (ΔLo) in the following way:
|Lc-Lo| ≤ (ΔLc2 + ΔLo2)½.
This can be useful when you are trying to identify high ionization lines. The wavelengths for these lines can have low accuracy in the line list and hence can appear to be "far away" from the observed wavelength. This option provides a simple way of including these lines in the query results. But also more generally, when an accurate measurement of the observed wavelength is available, this option can be useful.

Wavelength unit:

This option allows you to make a custom choice for the unit in which the wavelengths will be given (and printed in the output). The following choices are supported: Ångstrom (default), nanometer, micrometer, wavenumbers in cm-1, frequencies in gigaherz and teraherz, and photon energies in eV and keV.

Wavelength type:

Vacuum:Give wavelengths in vacuum.
Air: Give wavelengths in air. They are calculated from the vacuum wavelengths using the five-parameter formula for the refractive index of air given in:
Peck E.R., Reeder K., 1972, J. Opt. Soc. Am., 62, 958.
This expression is valid for wavelengths between 1850 Å and 17000 Å. It assumes that the air has a temperature of 15 degrees Celsius and contains 0.033% of CO2.
Note 1: This correction will only be applied if the user has chosen wavelength type units (i.e., Ångstrom, nanometer, or micrometer), and not for frequency or photon energy type units. Note that the latter also applies to wavenumbers in cm-1!
Note 2: This formula will normally only be used when the resulting air wavelength is ≥ 2000 Å. For exceptions see the section Wavelength range.
Note 3: The conversion formula will be extrapolated for wavelengths longward of 17000 Å, but its validity cannot be guaranteed in this case.

Radial velocity:

This option allows the selection tool to correct laboratory wavelengths for the Doppler shift and enable a meaningful comparison with the observed wavelengths you supplied in the Wavelength range field. You can either supply a velocity in km/s (default), or a cosmological redshift z if you click on the appropriate radio button next to this field. A positive value for the radial velocity v or redshift z means that the observed wavelengths you supplied are shifted to longer wavelengths or lower frequencies w.r.t. the rest frame (the laboratory wavelength). This option is most useful in the Center/Sigma case, but the correction will also be applied in the Lower/Upper Limit case. The code will search for possible identifications using a relativistically corrected wavelength Lc = Ll*[(1 + β)/(1 - β)]½ or Lc = Ll*(1 + z) and compare that to the observed wavelength after subsequently applying a correction for air if appropriate (Ll is the laboratory wavelength of the line in vacuum and β = v/c). The inverse formula will be used for frequencies and photon energies. The default value for the radial velocity is 0 km/s; radial velocities with an absolute value up to the speed of light can be supplied. Alternatively, any redshift > -1 can be entered. If a nonzero radial velocity is entered, the output will contain two separate columns of wavelengths. The first column will contain the Doppler-shifted wavelengths as they would have been observed on earth if present in the spectrum, the second column contains the laboratory wavelengths without any radial velocity correction (but with corrections for air if appropriate). Each column will have its own wavelength accuracy listed in the output if requested by the user.
Note: In all circumstances will the radial velocity correction be applied before the correction for air wavelengths. If a correction for air wavelengths is requested, the ratio of the wavelengths with and without the radial velocity correction shown in the output will in general not exactly match the Doppler factor since the correction for air is different at different wavelengths.

Minimum relative wavelength accuracy:

All wavelengths in the line list have relative accuracies of 5 % or better. The default is to list all lines, irrespective of their accuracy. When a relative accuracy in percent is entered in this box, only those lines with accuracies better than or equal to the prescribed value are included in the search. Values larger than 5 % are meaningless and will be ignored, values ≤ 0 will result in an error.


The default is that all ionization stages of all elements will be searched to find a possible identification. This field can be used to restrict the search to a range of elements and/or ionization stages. The elements should be entered by their usual symbolic names (e.g. Fe) and the ionization stages by the usual spectroscopic notation (e.g. I for neutral, II for singly ionized etc.).

Several lines of input can be combined, each containing entries like:

Fe I include lines of neutral Iron
Ni I-III include lines of ionization stages I thru III of Nickel
Na include lines of all ionization stages of Sodium
Mg-Sinclude lines of all elements Magnesium thru Sulphur
Sc-V I-III include lines of ionization stages I thru III of all elements Scandium thru Vanadium
Note: This entry is not case sensitive. At least one space should be typed between the element name and the ionization stage.

Minimum line strength:

The default is for all lines to be considered as a possible identification, regardless of the strength of the line. With this field it is possible to set a minimum for the strength of the line. With the drop-down menu to the right you can choose the type of line strength the minimum should be applied to. The following choices are supported: A_ki, g_k*A_ki, f_ik, log(gf), or S. Note that using this option does not automatically include the line strength in the output! See the section output format for further discussion.
Note 1: When using this option, multiplet searches will still show all members of the multiplet, regardless of the line strength.
Note 2: By default, lines with unknown line strength will be included in the search. To disable this behavior, you can use the option below.

Lines without atomic data:

The default is for all lines to be considered as a possible identification, regardless of whether the strength of the line is known or not. By using this option, you can exclude lines that have no known line strength. Note however that this may remove the correct line identification as lines with unknown line strength can still be strong lines!

Minimum abundance, Depletion factor:

With this command it is possible to impose a lower limit on the relative logarithmic number density (Alow) of elements to be considered for possible identifications. All abundances are normalized to A(H) ≡ 12. The default is to consider arbitrary low abundances. Any value ≤ 12 may be entered for the minimum abundance (higher values would exclude all elements). However, values ≤ -10 would result in all elements being included, and would therefore not impose any restriction. The elements are assumed to have standard solar abundances. The values were taken from:
Asplund M., Grevesse N., Sauval A.J., 2004, astroph/0410214 v2.

For nebular conditions using solar abundances is not a realistic assumption since most metals will be depleted in grains. To simulate this it is possible to supply a depletion factor df. Any value ≥ 0 may be entered for this factor. This factor will be used to calculate the actual abundance A from the solar abundance As using the formula:

A(elm) = max[ As(elm) - df*sd(elm), -9.99 ].

where sd is the standard depletion for each element derived using:

Savage B.D., Sembach K.R., 1996, ARA&A 34, 279 and,
Lodders K., 2003, ApJ, 591, 1220.
The page showing the value for the solar abundance and standard depletion of each element gives a more detailed account of the procedure used to derive the standard depletion. If the transition type Nebular is chosen (see section Transitions) the default value for the depletion factor df is 1.0, and otherwise it is 0.0.

Lower/Upper level energy range:

The default is to consider all values for the lower/upper level energy to find a possible identification. To restrict the search, a range of energies can be supplied as follows:
10000 selects levels between 0 cm-1 and 10000 cm-1.
10000-60000 selects levels between 10000 cm-1 and 60000 cm-1.
ground selects only levels belonging to the ground term (not case sensitive, may be abbreviated to "g" or any other input containing the letter "g").
Note that this example assumes that you use the default energy unit. However, other choices may be made using the Energy unit field. The limits imposed here will be ignored for hydrogen and helium recombination lines if you select nebular transitions. Entering values < 0 will lead to erroneous behavior.

Energy unit:

This option allows you to make a custom choice for the unit in which the energy levels will be given by you in the input and printed in the output. The following choices are currently supported: cm-1 (default), eV, keV, Rydberg, Hartree, erg, Joule, and Kelvin.

Maximum for principal quantum number n:

Default is to consider all possible values for the principal quantum number n to find possible identifications. However, transitions involving electrons with a very high quantum number n tend to be weaker and can therefore be less likely identifications. These transitions can be suppressed using this option.

Select a specific multiplet:

In an effort to make finding multiplets easier, this option is no longer part of the request form. All you need to do now is to click on the term field (if you have requested HTML output mode) and a listing of the multiplet will appear in your browser window. For hydrogenic lines the fine structure components of the line will appear (this assumes that the principal quantum number of the lower level is ≤ 15; for higher values the fine structure information is not available).

Fine structure components of transitions in hydrogen-like spectra:

In hydrogen-like ions, levels with the same principal quantum number have very small energy differences. As a result of this, all transitions between levels with the same set of principal quantum numbers have nearly identical wavelengths. Under normal astrophysical conditions these fine structure components cannot be resolved and will blend into one line. The wavelength of this blend is the average of all the components weighted by the transition probabilities and statistical weights, and is calculated assuming LTE level populations (i.e., levels are populated according to their relative statistical weights). Note that this average wavelength need not coincide with the difference between the energy levels listed. This is because the energy levels are averaged in a different way. For completeness the fine structure components of all transitions in hydrogen-like ions are included provided the principal quantum number of the lowest level is ≤ 15. The default is to suppress this information. If you click on the term field, the fine structure components will be shown in the browser window, provided the information is available for that line. Transitions between two levels with the same principal quantum number are not subject to this rule and will always be included in the output.


Default is to consider all types of transitions to find a possible identification. To alter this you first have to choose one of the following three buttons:
All:The default, consider all transition types.
Nebular: Consider only allowed (recombination) transitions of Hydrogen or Helium and only magnetic dipole/quadrupole or electric quadrupole transitions of other elements. A side effect of this choice is that the limits on the lower and upper levels will only be applied to the forbidden transitions. This allows the selection of ground state forbidden transitions only (by typing "g" for the lower level limit) while still getting all the information on the Hydrogen and Helium lines. This is very useful for identifying lines in spectra of photoionized plasmas like planetary nebulae, H II regions, etc.
Select: After clicking "Select", make a custom choice from the following five buttons (multiple selections are allowed):
E1: allowed transitions.
IC: intercombination or semi-forbidden transitions.
M1: magnetic dipole forbidden transitions.
E2: electric quadrupole forbidden transitions.
M2: magnetic quadrupole forbidden transitions.
E3: electric octopole forbidden transitions.

Transitions from auto-ionizing levels:

The default is to include transitions originating from auto-ionizing levels in the output. This option allows you to suppress these transitions. All levels with energies higher than the ionization potential going to the ground state of the next ion are considered auto-ionizing levels.

Output format:

This option allows you to check the various items you want included in your output. The wavelength of the transition is always included and need not be checked. The header will indicate if the wavelengths are valid for air or vacuum, as well as the units for the wavelength and energy fields. The following optional items can be checked:
Wavelength accuracy: Includes a rough estimate for the 1-sigma uncertainty (68% confidence interval) of the wavelength.
Spectrum: Gives Fe I for allowed transitions, Fe I] for intercombination transitions, and [Fe I] for forbidden transitions.
Transition Type: Gives E1 for allowed and intercombination transitions, M1 for magnetic dipole transitions, E2 for electric quadrupole transitions, and M2 for magnetic quadrupole transitions.
Configuration: Gives the electronic configuration of the lower and upper level. Entries like 15* denote the principal quantum number n for hydrogen-like ions. For these the wavelength of the transition and the level energies have been averaged over all allowed values of l and j assuming LTE level populations.
Term: Gives the spectroscopic term for the lower and upper level. A lowercase "o" at the end of the term indicates odd parity in plain or HTML mode.
Angular momentum: Gives the angular momentum (when "as J" is checked) or statistical weight (when "as g" is checked) for the lower and upper level. When "combine with term" is checked, the angular momentum will be given together with the term (only supported in latex mode, it will default to "as J" in any other output mode). First click on the leftmost button, then make a choice of one of these three options.
Transition probability: Gives the transition probability for that particular line (not available for all lines). First click on the leftmost button, then make a choice for the particular form you require (possible choices are the transition probability A_ki or g_k*A_ki, the oscillator strength f_ik or log(gf), and the line strength S; multiple choices are allowed). Beware that inconsistent normalizations exist for the line strength S in the case of E2, M2, and E3 transitions! This compilation adopts the conventions outlined in:
Aggarwal K.M., Keenan F.P., 2004, A&A, 427, 763.
A more detailed discussion of this point, which also lists numeric forms of all the conversion formulas used in the atomic line list, has been included here.
Transition probability flags: Gives information about the source and the reliability of the transition probability. The field consists of a number (possibly zero) of characters followed by a number. Lower case characters pertain to the lower level and upper case characters to the upper level. The characters have the following meaning:
U -- The level identification was marked uncertain in the project data.
R -- The order of the levels in the project data is reversed compared to laboratory measurements.
C -- The level identification given in the project data has been altered.
The number identifies the source of the transition probability. The references are listed in the documentation page.
Level energies: Gives the energy of the lower and upper level in the chosen wavelength unit. This unit will be indicated in the header
Literature references: The reference for the level information. If two numbers are present, the first is for the lower level and the second for the upper level. The references are listed in the documentation page.

Output mode:

PlainThe resulting output will be printed in plain ascii.
HTMLThe resulting output will be the same as in plain mode, except that it will be possible to click on the term field in the output to see multiplet information for that particular line (default).
LaTeXThe resulting output will be printed in a form that can be included directly into a LaTeX file.

Maximum no. of output lines:

This option allows you to set the maximum number of output lines. The values 50, 500 (default), and 5000 are supported.

Go back to the request form