Deriving chemical abundances in ionised gas

nebulae from the optical spectrum

Enrique Pérez-Montero
Instituto de Astrofı́sica de Andalucı́a - CSIC. Apdo. 3004, 18080, Granada, Spain

This is a brief manual of instructions used to calculate electronic densities and temperatures,
ionic and total abundances of the main chemical species found in gaseous nebulae showing an
emission line spectrum.
Our main objective is to present some of the fittings obtained by means of the IRAF tasks
TEMDEN and IONIC in the STSDAS package. These can then be used inside programs in order
to achieve a complete diagnostic for a large number of objects (e.g. Pérez-Montero & Dı́az 2005).

1. Introduction
(From Pérez-Montero & Dı́az, 2005)
HII regions, from faint ones on the Galactic disk to Giant Extragalactic ones (GEHR)
and HII galaxies, constitute the main source of information about the metallicity of
all kind of objects, including the most distant ones. Their bright spectra of emission
lines can be seen in all those regions that have gone through a recent star formation
episode. The analysis of such spectra provides the best method (if not the only one)
to determine the chemical abundances of elements like helium, nitrogen, oxygen, neon,
argon, sulfur, displaying prominent emission lines in the optical range corresponding to
several ionization states. Detailed knowledge of these abundances is essential for fully
understanding gaseous and stellar evolution, and has helped us towards the solution of
some yet unsolved questions regarding the chemical evolution of galaxies in the local
Universe. These questions are becoming more and more relevant for a more distant –
hence younger – Universe, as we are able to obtain data on fainter objects.
Recombination lines, weakly dependent on the nebular internal temperature, provide
more precise abundance determinations. He abundance, for instance, can be derived
with precision better than 5%. Unfortunately, many of the emission lines are collisionally
excited and their intensities depend exponentially on the temperature. In principle, the
temperature can be determined from appropriate line intensity ratios which however
require the detection and measurement of intrinsically faint (or absent) auroral lines.
This is in particular the case for regions with high metal content – for which the cooling
produced by the metals is very efficient and the lines are no longer detected – and also
for distant HII regions and regions with low surface brightness. For these cases, empirical
methods based on the strong nebular line intensities are used.
The main sources of uncertainty associated to the derived physical properties are re-
lated with the error of the fluxes of the lines. The measure of these fluxes is usually done
by means of tasks like SPLOT of IRAF, that integers the intensity of each line over a
local continuum. The errors of the fluxes measured in this way can be calculated with
the expression:
r
EW
σl = σc · N +
∆
where σl is the error of flux of the line, σc represents the standard deviation in a box near
of the measured emission line and represents the error in the position of the continuum.
1
2 E. Pérez-Montero: Ionised gas nebulae optical diagnostics

Figure 1. Energy levels and main transitions for np2 , np3 and np4 type ions.

, N is the number of pixels in the measure of the flux of the line, EW is the equivalent
width of the line and ∆ is the dispersion of the wavelength in angstroms per pixel.
Each one of the intensities measured in this way must be reddening corrected. The
reddening constant c(Hβ) can be calculated from the decrement of Balmer of the stronger
recombination lines of hydrogen. The measurement of these lines must be done, generally,
taking into the account the absorption of the underlying stellar population. The physical
law that allows to calculate the reddening constant from this decrement and therefore to
correct the other emission lines from reddening takes the form:
I0 (λ) I(λ)
= · 10−c(Hβ)·[f (λ)−f (Hβ)]
I0 (Hβ) I(Hβ)
where I(λ) is the intensity of the line we want to correct and I0 (λ) is the corrected value in
relation to Hβ, c(Hβ) is the reddening constant and f (λ) is the extinction function. The
theoretical values of the hydrogen recombination lines that allow to calculater C(Hβ)
are a function of the temperature and density so this imply the necesity of an iterative
proccess to calculate them. This calculation can be done using the tabuled coefficients by
Storey & Hummer (1995) and the corresponding program availabe en the CDS database
†. Regarding the extinction function one can assume the law of Miller & Matthews (1972),
which normalized to a value of RV = 3.2, that is the value for the Galaxy takes the next
form:

n 0.477λ−1 − 1.209, λ−1 ≤ 2.29

f (λ) − f (Hβ) =
0.342λ−1 − 0.645, λ−1 > 2.29

† CDS webpage (Centre de données astronomiques de Strasbourg) is

http://cdsweb.u-strasbg.fr/
 E. Pérez-Montero: Ionised gas nebulae optical diagnostics 3
where λ is the wavelength of the line we want to correct in micron unities. We can verify
that this function takes a value equal to 0 for λ = 0.4861 µ.
The uncertainty associated with the reddening correction must be taken into the ac-
count in the errors of the emission lines and conviently propagated into all the physical
magnitudes we calculate. One must take into account, however, that in many cases, cal-
ibrations are used obtained from photoionization models for which the uncertainties are
difficult to quantify. Even when the electronic temperature has been determined with
high precision, there are some problems limiting the confidence on the results attained.
Those problems include: (1) the effect of the internal ionization structure on multiple
zone models (PMD03); (2) temperature fluctuations across the nebula (Peimbert, 2003);
(3) collisional and density effects on the ionic temperatures (Luridiana, Peimbert & Pe-
imbert, 1999); (4) neutral gas zones affecting the determination of ionization correction
factors (ICFs) (Peimbert, Peimbert & Luridiana, 2002). ICFs are used to calculate total
abundances from the abundances of the observed species
N (X) N (Xobs )
= ICF (Xobs ) ·
N (H) N (H + )
(5) the ionization structure is not adequately described by present models (PMD03); (6)
possible photon escape affecting low ionization lines in the outer regions of the nebula
(Castellanos, Dı́az & Tenorio-Tagle, 2002).
The first three effects can introduce uncertainties regarding the derived O abundance
of some 0.2, 0.3 and 0.4 dex respectively, depending on the degree of excitation. The
uncertainties associated to the rest of the enumerated problems have not yet been quan-
tified.

2. Electron temperature and density

2.1. Oxygen
[O iii] temperature is calculated from the ratio:
I(4959Å) + I(5007Å)
RO3 =
I(4363Å)
given that, according to Osterbrock (1989), temperature can be obtained from the ratio
of collisional transitions that have a similar energy but occupy different levels. The fitting
between the ratio and the electronic temperature was obtained using the task TEMDEN
from the package STSDAS inside IRAF. This task is based on the program FIVEL (De
Robertis, Dufour & Hunt 1987; Shaw & Dufour 1995) for a five level atom. In this case,
and likewise for the rest of the ions unless stated differently, the transition probability,
energy levels and collision coefficients are those given in such program by default. The
obtained fitting, as for the rest of the np2 ions, is independent of the electronic density
and is:

47.77
t([OIII]) = 0.8254 − 0.0002415RO3 +
RO3
where t is the temperature in units of 104 K.
For [O ii] the quotient for the electron temperature is calculated from
I(3726Å) + I(3729Å)
RO2 =
I(7319Å) + I(7330Å)
4 E. Pérez-Montero: Ionised gas nebulae optical diagnostics
2,5

2,0

1,5
t(OIII)

1,0

0,5

0,0
0 100 200 300 400 500 600 700 800

I(4959 + 5007)/I(4363 )
Figure 2. Relation between RO3 and t[O iii]

One has to be careful in this case, because the [O ii] auroral lines might be contaminated
by recombination emission. Such emission however, can be quantified and corrected for.
According to Liu et al. (2000), such contribution can be fitted (for 0.5 ≤ t ≤ 1.0) by the
function:
IR (7319 + 7330) O2+
= 9.36 · t0.44 · +
I(Hβ) H
Moreover, the ratio of the [O ii] lines is strongly dependent on the electron density.
Ideally, one should know the [O ii] density from the ratio I(3726Å)/I(3729Å) but very
frequently we lack resolution to separate the doublet in which case one has to resort to
the [Siii] density, also representing the low excitation zone. The fitting obtained is:
a2 (n)
t([OII]) = a0 (n) + a1 (n) · RO2 +
RO2
where the coefficients are respectively:
0.17
a0 (n) = 0.23 − 0.0005 · n −
n
0.0064
a1 (n) = 0.0017 + 0.000009 · n +
n
16.4
a2 (n) = 38.3 − 0.021 · n −
n
in this case, the atomic coefficients vary. PMD03 and PMD05 take the collisional strengths
from McLaughlin & Bell (1998), as the default ones in CLOUDY 96. It is necessary to
note, though, that Wang et al. (2004) underline that the coefficients fail to fit the ob-
servations (contrary to those of Pradham, 1976). For the fitting presented here then, we
have used Pradham’s.
Occasionaly the auroral [O ii] lines are not observed with good S/N or they are outside
our observed spectral range. In that case it is practice to use some relation based on
 E. Pérez-Montero: Ionised gas nebulae optical diagnostics 5

2,00
ne = 10
ne = 100
1,75
ne = 500

1,50
t(OII)

1,25

1,00

0,75

0,50
20 40 60 80 100 120

I(3727+3729)/I(7319+7330 )
Figure 3. Relation between RO2 and t[O iii]

photoionization models in order to infer t[O ii] from t[O iii]. For instance, the relation
2
t([OII]) =
t([OIII])−1 + 0.8
based on Stasińska (1990) models is frequently accepted. However, such expression ne-
glects the dependence of t[O ii] on the density, consistent with the observed dispersion
for the objects for which both temperatures have been derived from observations (see
Figure 4). Also in Figure 4 we show a different set of models that do take this dependence
into account (presented in PMD03). The fitting obtained in this case is:

1.2 + 0.002 · n + 4.2

n
t([OII]) = 2.5
t([OIII])−1 + 0.08 + 0.003 · n + n

2.2. Sulfur
The [Siii] lines ratio is commonly used to determine electron density for the low excitation
zone. It is generally assumed that the nebula has constant density, although there is
growing evidence for the existence of a density profile instead. Fortunately, the high
excitation species have diagnostic ratios that are not density sensitive.
I(6717Å)
RS2 =
I(6731Å)
Density is then calculated as:
RS2 · a0 (t) + a1 (t)
n([SII]) = 103 ·
RS2 · b0 (t) + b1 (t)
where
a0 (t) = 2.21 − 1.3/t − 1.25t + 0.23t2
a1 (t) = −3.35 + 1.94/t + 1.93t − 0.36t2
6 E. Pérez-Montero: Ionised gas nebulae optical diagnostics
2,4

-3
2,2 ne = 10 cm

2,0
-3
ne = 100 cm
1,8
t([OII])

1,6

1,4

1,2
-3
1,0 ne = 500 cm

0,8

0,6
0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,0 2,2 2,4 2,6

t([OIII])
Figure 4. Relation between t[O iii] and t[O ii] for model sequences and for direct observations
of HII galaxies

b0 (t) = −4.33 + 2.33/t + 2.72t − 0.57t2

b1 (t) = 1.84 − 1/t − 1.14t + 0.24t2
here t is generally t[O iii], although an iterative process could be used to calculate it with
t[Siii] given that this temperature, like t[O ii], a type np3 ion, is density dependent. The
ratio used in this case is:
0 I(6717Å) + I(6731Å)
RS2 =
I(4068Å) + I(4076Å)
For the [Siii] auroral lines it is enough to measure one of them, as they are related by
a fixed theoretical ratio, I(4068Å) ≈ 3· I(4076Å). We can calculate in this way the [Siii]
temperature.

0 a2 (n) a3 (n)
t([SII]) = a0 (n) + a1 (n) · RS2 + 0 + 02
RS2 RS2
where
0.848
a0 (n) = 1.92 − 0.0017 · n +
n
0.0185
a1 (n) = −0.0375 + 4.038 · 10−5 · n −
n
10.4
a2 (n) = −14.15 + 0.019 · n −
n
58.52
a3 (n) = 105.64 − 0.109 · n +
n
When the [Siii] auroral lines are not available, it is usually assumed that t[Siii] ≈ t[O ii].
There is evidence however, suggesting a somewhat lower value. From the models used
 E. Pérez-Montero: Ionised gas nebulae optical diagnostics 7

2,25
-3
10 cm
-3
2,00 100 cm
-3
500 cm
1,75
t(SII)

1,50

1,25

1,00

0,75

0,50
8 12 16 20

I(6717+6731)/I(4069+4076 )
Figure 5. Relation between RS2 and t[Siii]

2,0

1,8

1,6

1,4
-3
ne = 10 cm
t([SII])

1,2

1,0
1:1
0,8

-3
0,6 ne = 100 cm

0,4
0,8 1,0 1,2 1,4 1,6

t([OII])
Figure 6. Relation between t[O ii] and t[Siii] for model sequences and for a sample of HII
galaxies.

for my PhD thesis, I obtained a lineal fitting:

t[Siii] = 0.71 · t[Oii] + 0.12
for a 100 cm−3 number density. For lower densities, the first expresion seems to be valid.
In any case, for the few objects for which we have a simultaneous measurement of both
temperatures, the dispersion is quite large as can be seen in Figure 6 for HII galaxies.
8 E. Pérez-Montero: Ionised gas nebulae optical diagnostics
Direct measurements of the [S iii] temperature have become possible with the avail-
ability of the colissional lines in the near IR.

I(9069Å) + I(9532Å)
RS3 =
I(6312Å)
expression that can be simplified in case of lacking one of the near IR lines, knowing that
I(9532Å) ≈ 2.44 · I(9069Å). In that case, the [S iii] temperature can be calculated by
the fit:
RS3 + 36.4
t([SIII]) =
1.8 · RS3 − 3.01
There is also work relating the [O iii] and the [S iii] temperatures. As discussed in Garnett
(1992), t[S iii] is in between the temperatures of [O iii] and of [O ii] and allows to calculate
the S2+ abundance from just the 6312 Å line. The importance of this relation has reversed
recently, given that in very high metallicity objects one can more easily detect the [S iii]
auroral line than the [O iii] one (e.g. S5 en M101; Kinkel & Rosa, 1994 or CDT1 in
NGC1232; Castellanos et al., 2002). The relation given by Garnett is:
t([SIII]) = 0.83 · t([OIII]) + 0.17
although this relation underestimates t[S iii] given new atomic coefficients for sulphur(Tayal
& Gupta, 1999). From my models, instead, I can derive:
t([SIII]) = 1.05 · t([OIII]) − 0.08
plotted as a thin solid line in Figure 7. Since the relation between electron temperatures
does not allow to obtain a reallistic estimation of the associated uncertainties to the
models and since there is growing number of observations in the spectral range where it
is possible to obtain this temperatures, it is possible now to give an empirical fit:
t([SIII]) = (1.19 ± 0.08) · t([OIII]) − (0.32 ± 0.10)
(Hägele et al. en prep.) that is plotted as thick solid line in Figure 7.

2.3. Nitrogen
The temperature of [Nii] can be calculated from the quotient:
I(6548Å) + I(6584Å)
RN 2 =
I(5755Å)
The nebular lines of [Nii] are very close to Hα so they appear sometimes blended to this
line and therefore it is not possible sometimes to measure them or, at least, to measure
the weakest of them. In this case it is often used the theoretical relation between them,
in such a way that I(6584) ≈ 2.9 · I(6548). Besides, the auroral line of [Nii] is affected
by recombination emission, that can be corrected using the next expression proposed by
Liu et al. (2000):
IR (5755) N 2+
= 3.19 · t0.30 · +
I(Hβ) H
in the range between 5000 and 20000 K.
In these conditions it is possible to calculate the temperature of [Nii] by means of the
expression:
42.126
t([N II]) = 0.537 + 0.000253RN 2 +
RN 2
 E. Pérez-Montero: Ionised gas nebulae optical diagnostics 9
2,6

2,4

2,2

2,0

1,8
t([SIII])

1,6

1,4

1,2

1,0

0,8

0,6
0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,0 2,2
t([OIII])

Figure 7. Relation between t[O iii] and t[S iii] for model sequences and for a sample of
different objects.

that does not depend on density due to the np2 nature of N+ .

Unfortunately, the auroral line has very low signal-to-noise ratio, so it is usually con-
sidered the approximation t([Nii]) ≈ t([Oii]) as valid. This relation is confirmed by pho-
toionization models but is quite sensitive to density and the inner ionization structure
of the nebula, and it is possible to reach values closer to T([Siii]) in some cases, so in all
case it can’t be taken without any sort of uncertainty.
Other possibility is to calculate t([Nii]) directly from t([Oiii]) using the expression
derived using photoionization models by Pérez-Montero & Contini (2009):
1.85
t([N II]) =
t([OIII])−1 + 0.72

2.4. Balmer temperature

The Balmer tempeature depends on the value of the Balmer jump (BJ) in emission.
In order to measure this value, it is necessary to fit the continuum in both sides of
the discontinuity (λB = 3646 Å). The contribution of the underlying population affects,
between other factors, to he emission of hydrogen lines near from the BJ. The increment
in the number of lines at shorter wavelengths produces blends between them that trend
to reduce the level of the continuum at right of the discontinuity so it is necessary to
take into the account all this in the final uncertainty. Once measured the BJ, the Balmer
temperature (T(Bac)) is measured from the quotient of the flux of the jump and the
emission of the line H11 by means of the expression:

³ BJ ´−3/2
T (Bac) = 368 × (1 + 0.259y + + 3.409y ++ )K
H11
where y + and y ++ are the ionic abundances of hellium once and twice ionized, respec-
tively, and BJ is in ergs cm−2 s−1 Å−1 .
10 E. Pérez-Montero: Ionised gas nebulae optical diagnostics
2.5. Other temperatures and densities
There are other line quotients for other ions, but they are not necessarily easily available.
This is the case for [ArIV] for instance, for which we can deduce the density in the inner
nebula, or the temperature of the rest of ions for which the corresponding auroral line is
not available in the observed spectral range or there is not enough signal-to-noise ratio
to detect the recombination lines, but whose nebular emission lines are detectable in the
spectrum, allowing the determination of their ionic abundances. In this case, they are
usually taken some assumptions about the gas structure and the position of each species
is associated to another whose temperature is known. This is the case of the next ions:
t([N eIII]) ≈ t([F eIII]) ≈ t(HeII) ≈ t([ArIV ]) ≈ t([OIII])

t([ArIII]) ≈ t([SIII)]

3. Ionic abundances and ICFs

In what follows I will describe the fittings derived for each ion once we know the avail-
able line fluxes and the appropriate electron densities and temperatures. The relations
are listed in Table 1. Note however that the fittings have been obtained with the atomic
coefficients as in STSDAS except those for O+ , and changes in them will produce changes
in the equations. Most of the fittings follow equations as proposed by Pagel et al. (1992).

3.1. Hellium
Hellium lines, as equal as hydrogen ones in the visible spectrum have a recombination
nature and they are generally strong and numerous, although many of them are usually
blended with other lines. Besides they are affected by absorption of underlying stellar
population, by fluorescence or by collisional contribution. They are generally used the
lines of Hei λλ 4471, 5876, 6678 and 7065Å, and Heii λ 4686 Å to estimate the hellium
abundances once and twice ionized respectively. In order to determinate the contributions
to these lines, it exists a very deep study in Olive & Skillman (2001) who deduce the
equations from the theoretical emissivities done by Smits (1996) In these conditions it is
considered as valid to approximate that:
He He+ + He2+
=
H H+

3.2. Oxygen
+
The chemical abundance of O can be determined from the intensity of the 3727 Å line,
although recently Kniazev et al.(2003) suggest the use of the 7319,7330 Å doublet in the
low redshift SLOAN spectra without 3727. For the O2+ abundance, both strong lines at
4959 and 5007 Å are used.
The total oxygen abundance can be approximated by
O O+ + O2+
=
H H+
given that due to the charge exchange reaction the relative fractions of neutral oxygen
and hydrogen are similar:
O0 H0
=
O H
 E. Pérez-Montero: Ionised gas nebulae optical diagnostics 11

7,4

7,2
12+log(N /H )
+
+

6,8

6,6

6,4
0,8 1 1,2 1,4 1,6 1,8 2
t([NII])

Figure 8. N+ abundance as a function of electron temperature, for a fixed line value.

Table 1. Equations for abundance calculations.

Ion† Equation
³ ´
I(3727)
O+ log + 5.992 + 1.583 − 0.681 · log t + log(1 + 0.00023 · n)
³ I(Hβ) ´ t

log I(7325) + 6.895 + 2.44

t − 0.58 · log t − log(1 + 0.0047 · n)
³ I(Hβ) ´
I(4959)+I(5007)
O 2+
log I(Hβ) + 6.144 + 1.251
t − 0.55 · log t
³ ´
S+ log I(6717)+I(6731)
I(Hβ) + 5.423 + 0.929
t − 0.28 · log t + 0.0001 · n)
³ ´
2+ I(9069)+I(9532) 0.771
S log I(Hβ) + 5.8 + t − 0.22 · log t
³ ´
log I(6312) + 6.74 + 1.672 − 0.595 · log t
³ I(Hβ) ´ t
I(6548)+I(6584)
N +
log I(Hβ) + 6.273 + 0.894
t − 0.592 · log t
³ ´
Ne2+ log I(3868) + 6.486 + 1.558 − 0.504 · log t
³ I(Hβ) ´ t
2+ I(7137) 0.808
Ar log I(Hβ) + 6.157 + t − 0.508 · log t
³ ´
Ar3+ log I(4740)
I(Hβ) + 4.705 + 1.246
t − 0.156 · log t

† As 12+log(X/H+ )

3.3. Sulphur
The abundances are obtained from the 6717, 6731 Å lines for S+ and by the 9069, 9532
Å lines for S2+ , though for the latter, also the 6312 Å line can be used.
The ICF for sulphur takes into account the S3+ abundance which cannot be determined
12 E. Pérez-Montero: Ionised gas nebulae optical diagnostics
in the optical range. A good approximation is given by Barker (1980) as:
· µ ¶α ¸−1/α
+ 2+ O2+
ICF (S + S ) = 1 −
O+ + O2+
Although it is customary to write Barker’s expression as a function of the O+ /(O+ +O2+ )
ionic fraction, we have reformulated it in terms of O2+ /(O+ +O2+ ) since the errors as-
sociated to O2+ are considerably smaller than for O+ . and for a sample of objects with
observed [SIV] line at 10.5 µm, it was derived α ≈ 2.5 (Pérez-Montero, Dı́az, Vı́lchez &
Kehrig, 2006).

3.4. Nitrogen
+
We can calculate N abundance from the 6548 and 6584 Å lines. Given their proximity
to Hα, if one of them cannot be measured it doesn’t matter as there is a theoretical
relation between them as I(6584) ≈ 2.9 · I(6548).
Starting from:
N+ O+
=
N O
one can calculate quite precisely the total abundance of Nitrogen.

3.5. Neon and Argon

The [Ne iii] line at 3868 Å is used for neon,
The ionisation correction factor for neon can be calculated according to the expression
given by Pérez-Montero et al. (2007):
0.171
ICF (N e2+ ) = 0.142 x + 0.753 +
x
where x = O2+ /(O+ +O2+ ).
This expression deviates from the classical approximation Ne/O≈Ne2+ /O2+ used to
derive total neon abundances.
For argon, we use the [ArIII] 7137 Å line. It is possible to measure as well the lines of
[ArIV] at 4713 and 4740 Å. Nevertheless, the first of them usually appears blenden with
another line of Heii at 4711Åthat is difficult to correct, so it is better to use the second
and brighter to calculate the ionic abundance of Ar3+ .
As in the case of neon, the total abundance of argon has been calculated using the ion-
ization correction factors (ICF(Ar2+ ) and the ICF(Ar2+ +Ar3+ )) given by Pérez-Montero
et al. (2007). We use the first one only when we cannot derive a value for Ar3+ . The
expressions for these ICFs are:
0.064
ICF (Ar2+ ) = 0.507 (1 − x) + 0.749 +
(1 − x)
0.006
ICF (Ar2+ + Ar3+ ) = 0.364 (1 − x) + 0.928 +
(1 − x)
where x = O2+ /(O+ +O2+ ).

REFERENCES
Barker, T. 1980, ApJ, 240, 99.
Castellanos, M., Dı́az, A.I. & Terlevich, E. 2002, MNRAS, 329, 315.
De Robertis, M.M., Dufour, R.J. & Hunt, R.W. 1987, JRASC, 81, 195.
 E. Pérez-Montero: Ionised gas nebulae optical diagnostics 13
Garnett, D.R. 1992, AJ, 103, 1330.
Kingsburgh, R.L. & Barlow, M.J. 1994, MNRAS, 271, 257.
Kinkel, U. & Rosa, M.R. 1994, A&A, 282, 37.
Kniazev, A.Y., Grebel, E.K., Hao, L., Strauss, M.A., Brinkmann & Fukugita, M. 2003, ApJ,
593L, 73.
Liu, X.-W., Storey, P.J., Barlow, M.J., Danzinger, L.J., Cohen, M. & Bryce, M. 2000, MNRAS,
312, 585.
McLaughlin, B.M. & Bell, K.I. 1998, J.Phys.B, 31, 4317.
Pagel, B.E.J., Simonson, E.A., Terlevich, R.J. & Edmunds, M.G. 1992, MNRAS, 255, 325.
Pérez-Montero, E., Contini, T., Hägele, G.F., & Dı́az, A.I:, 2007, MNRAS, 381, 125
Pérez-Montero, E. & Contini, T., 2009, MNRAS, 398, 949
Pérez-Montero, E. PhD thesis. 2003. Universidad Autónoma de Madrid.
Pérez-Montero, E. & Dı́az, A.I. 2003, MNRAS, 346, 105. (PMD03)
Pérez-Montero, E. & Dı́az, A.I. 2005, MNRAS, 361, 1063 (PMD05)
Pérez-Montero, E. Dı́az, A.I., Vı́lchez, J.M. & Kehrig, C. 2006, A&A, 449, 163. bibitem[] Prad-
ham, A.K. 1976, MNRAS, 177, 31.
Shaw, R.A. & Dufour, R.J. 1995, PASP, 107, 896.
Tayal, S.S. & Gupta, G.P. 1999, ApJ, 526, 544.
Wang, W., Liu, X.-W., Zhang, Y. & Barlow, J. 2004, A&A, 427, 873.