c Pleiades Publishing, Inc., 2016. ISSN 1063-7737, Astronomy Letters, 2016, Vol. 42, No. 1, pp. 10–28.  c T.N. Tarasova, A. Skopal, 2016, published in Pis’ma v Astronomicheski˘ı Zhurnal, 2016, Vol. 42, No. 1, pp. 13–32. Original Russian Text 

Structure and Chemical Composition of the Envelope of Nova V339 Del in the Nebular Phase T. N. Tarasova1* and A. Skopal2 1

2

Crimean Astrophysical Observatory, pos. Nauchnyi, Crimea, 298409 Russia Astronomical Institute, Slovak Academy of Sciences, 05960 Tatranska´ Lomnica, Slovakia Received August 10, 2015

Abstract—Based on our spectrophotometric observations, we have investigated the envelope of Nova V339 Del in the nebular phase. Our modeling of the Hα line profiles and their comparison with the observed profiles have shown that the Nova envelope consists of circumpolar ejecta and a disk-shaped central component. The inclination of the orbital plane of the binary system, 65◦ , has been determined in the same way. We have estimated the mean electron density to be ∼106 cm−3 . Our estimates of the abundances of some chemical elements in the Nova envelope have shown that the concentrations of helium, neon, and iron are nearly solar, while the concentrations of nitrogen and oxygen exceed the solar ones by a factor of 120 ± 60 and 8 ± 1.6, respectively. The mass of the emission envelope in the nebular phase (from 253 to 382 days after the brightness maximum) has been estimated to be ≈7 × 10−5 M . DOI: 10.1134/S1063773716010060 Keywords: novae, spectroscopic observations, nova envelopes, chemical composition.

INTRODUCTION

phase was similar to the spectrum of a late-A or earlyF star (Tarasova 2013a, 2013b). We have already investigated the early spectral evolution of the Nova previously (Skopal et al. 2014). In this paper, based on our spectrophotometry obtained at later phases, we study some of the properties of the Nova envelope: its structure, elemental abundances, and mass.

Nova Del 2013 (V339 Del) was discovered on August 14, 2013, by Koichi Itagaki (Waagen 2013). The name of Nova V339 Del was assigned by Samus (2013). It was discovered before its brightness reached a maximum. At the time of its discovery, the visual brightness of the Nova was 6.8 mag. Its V magnitude at maximum light was about 4.2 mag. Denisenko and Masi (2013) identified this star before its outburst as USNO-B1.0 1107-0509795 (B = 17.2−17.4, R = 17.4−17.7). Munari and Henden (2013) and Deacon et al. (2014) confirmed this identification. Since this Nova was very bright, intensive spectroscopic observations were begun before its brightness reached a maximum (Darnley et al. 2013; Shore et al. 2013a; Munari et al. 2013a; Tomov et al. 2013; Tarasova 2013a, 2013b; Chochol et al. 2013) and were continued when the star began to fade (Tarasova and Shakhovskoi 2013; Shore et al. 2013b, 2013c; Munari et al. 2013b; Burlak et al. 2015). We took the first spectrum one day after the discovery of V339 Del, i.e., before the Nova reached its brightness maximum, and the last one more than one year later. The first spectrum showed the Nova to be in the phase of an early optically thick envelope (fireball phase). The spectrum of the Nova at this

OBSERVATIONS Our spectroscopic observations were performed at the 2.6-m Shain telescope (ZTSh). All spectra were taken with the SPEM slit spectrograph mounted at the Nasmyth focus. The detector was a SPEC-10 1340 × 100-pixel CCD camera. The dispersion with a 651 lines mm−1 grating was about 2 A˚ px−1 (the resolution was ∼1000). The log of observations containing information about the Nova spectroscopy is presented in Table 1. This table also gives information about the two high-resolution (13446–13078) Hα line spectra retrieved from the Astronomical Ring for Access to Spectroscopy (ARAS)1 database. We used these lines to analyze the envelope structure. The primary reduction of the spectra, including the bias subtraction and flat fielding, was performed with the SPERED code developed by S.I. Sergeev 1

*

E-mail: [email protected]

10

http://www.astrosurf.com/aras/Aras_DataBase/Novae/ Nova-Del-2013.htm

STRUCTURE AND CHEMICAL COMPOSITION

11

Table 1. Log of spectroscopic observations for Nova V339 Vel Date

JD (2450000+)

Days after maximum

Spectral range, A˚

Resolution λ/Δλ

May 15, 2013

6520.304

−0.5

3301–7575

1000

Tarasova

Aug. 19, 2013

6524.387

3.6

3200–7575

1000

Tarasova

Sep. 1, 2013

6537.272

16.5

3200–7575

1000

Tarasova

Sep. 13, 2013

6549.248

28.5

3253–7579

1000

Tarasova

Oct. 10, 2013

6576.377

59.6

3250–7575

1000

Tarasova

Nov. 8, 2013

6604.200

83.4

3225–7575

1000

Tarasova

Nov. 13, 2013

6610.494

89.7

6479–6646

13446

Graham

Nov. 19, 2013

6616.497

95.7

6479–6646

13078

Graham

Apr. 25, 2014

6773.490

252.7

3599–7575

1000

Tarasova

May 20, 2014

6797.502

276.7

3325–7500

1000

Tarasova

June 30, 2014

6838.530

317.8

3324–7575

1000

Tarasova

July 18, 2014

6857.493

336.7

3324–7574

1000

Tarasova

Sep. 1, 2014

6902.326

381.6

3325–7574

1000

Tarasova

at the Crimean Astrophysical Observatory. We calibrated the fluxes in the star’s spectrum using the absolute energy distribution for the spectrophotometric standard HR 7939 taken from the catalog of Burnashev (1985). We observed the spectrophotometric standard at the same date with V339 Del and with the same zenith distance; therefore, the airmass difference between the standard and the Nova was disregarded. Since the spectrograph is a slit one, for checking purposes we compared the B and V magnitudes of V339 Del with those calculated from the calibrated spectra taken at the same dates. The difference between the calculated and measured magnitudes was, on average, about 0m. 1. Figure 1 presents the light curve of the Nova constructed using the AAVSO database. The times when the spectroscopic observations were performed are marked in this light curve. SPECTRAL EVOLUTION OF NOVA V339 Del AND THE VELOCITY OF MATTER IN THE ENVELOPE All our spectroscopic observations are presented in Fig. 2. In addition to the date, the number of days from the brightness maximum is given on the right in this figure. For clarity, the spectra were shifted ASTRONOMY LETTERS

Vol. 42 No. 1 2016

Observers

relative to the first one by a constant value. The first spectrum in this figure was taken almost at the brightness maximum (within less than half a day). We analyzed the energy distribution in this spectrum and established that it corresponded to either a late-A or early-F spectral type (Tarasova 2013a, 2013b). Balmer hydrogen lines that have a P Cyg profile with deep absorption components are identified in the spectrum. The radial velocities of the absorption components in the Hα, Hβ, Hγ, Hδ, and H lines are about −1450, −1260, −1108, −1040, and −991 km s−1 , respectively. In addition, there are numerous iron lines with a P Cyg profile in the spectrum. The radial velocities of the Fe II lines were noticeably lower; in particular, the radial velocities of the strongest Fe II 4924 and 5169 lines were about −990 and −970 km s−1 , respectively. The presence of numerous Fe II lines and the P Cyg profiles of all spectral lines suggest that the spectral type of the Nova is Fe. The second spectrum was taken on the third day after the brightness maximum. The depth of the absorption components in the spectrum decreased; the Balmer jump vanished. The interstellar Ca II 3934 and 3968 lines are identified among all lines; there are also the interstellar Na I 5890 and 5896 lines in the

12

TARASOVA, SKOPAL

4 5 Spectra B V R

6

4.0 4.5 5.0

7 Magnitude

5.5

8

6.0 6.5

9

7.0

10

JDmax = 2456520.6 Vmax = 4.23 6520

6522

6524

6526

6528

6530

11 12 13 14

6550

6600

6650

6700 6750 JD 2450000+

6800

6850

6900

Fig. 1. Light curve of Nova V339 Del constructed from AAVSO data. The times when our spectra were taken are marked in the light curve.

spectrum. The energy distribution for these spectra was modeled in our previous paper (Skopal et al. 2014). The succeeding two spectra were taken on days 17 and 29 in the period when a plateau was formed on the light curve. The model energy distribution in the optical and near-infrared range showed that the radiation from the nebular component dominated at this phase (see Fig. 2 in Skopal et al. (2014)). Apart from the Balmer hydrogen lines, numerous Fe II lines and two blends of ionized carbon, C II 6727+6734 and C II 7112+71115, are identified in the Nova spectrum. In addition, forbidden oxygen lines, [O I] 6300 and 6363, appeared in the spectrum. The mean envelope expansion velocity, defined as the full width at half maximum (FWHM) of the lines, was estimated from the Hα and Hβ lines to be ∼700 ± 50 and ∼600 ± 50 km s−1 , respectively. The next two spectra were taken on days 56 and 83 after the brightness maximum, i.e., after the plateau phase, in the middle and at the end of a rapid brightness decline. The He I 5876, 6678, 7065, and N II 5679 lines, the N III 4640 blend, and forbidden lines ([O III] 4363, 4959, 5007, [O II] 7320+7330, [N II] 5755) appeared in the spectra. Thus, the

spectra suggest that the Nova is in an early nebular phase. The mean envelope expansion velocity was estimated from the Hα and Hβ lines to be ∼650 ± 50 and ∼800 ± 50 km s−1 , respectively. The last five spectra taken in 2014 from day 253 to 382 after the brightness maximum show that the Nova is in a developed nebular phase. The forbidden [O III] 5007 line is strongest in the spectra; the forbidden [Ne III] 3869, 3968 and [Ar III] 7136, [Ar IV] 7237+7263 lines also appeared. In addition, there are forbidden lines with high ionization potentials in the spectra: [Fe VI] 5176, [Ca V] 5309, [Fe VII] 6087, and [Ne V] 3346, 3426. The mean envelope expansion velocity at this phase is 750 ± 50 km s−1 from the Hβ and He I 5876 lines and ∼800 ± 50 km s−1 from the [O III] 5007 and [N II] 5755 lines (see Fig. 7 below). THE THREE-COMPONENT ENVELOPE STRUCTURE To investigate the structure of the Nova envelope, we analyzed the evolution of the spectral line profiles (Figs. 3–7) at various spectrophotometric phases of the Nova by comparing them with certain segments ASTRONOMY LETTERS

Vol. 42

No. 1

2016

13

Hα

Hβ

–10

Hγ

H9 H8 Hε Hδ

STRUCTURE AND CHEMICAL COMPOSITION

Aug. 15, 2013 –0.5

Shift –11

Aug. 19, 2013 3.6

–13 Fe II

Fe II

C II

C II

–12

Sep. 1, 2013 16.5

Fe II

–14

He I

He I

C II [O II]

[O I] [O I]

[O III]

N II [N II] He I

–16

Oct. 10, 2013 59.6

–17 Nov. 8, 2013 83.4 [Ar III] [Ar IV] [O II]

–19

[Fe VII]

[Fe VI] [Fe VII]

[Ne III] [Ne III]

–18

[Ne V]

Flux, erg cm2 s–1 Å–1

–15

N II + N III He II

Hγ + [O III]

Sep. 13, 2013 28.5

[Fe VII]

–20

–21

Apr. 25, 2014 252.7

May 19, 2014 276.7

–22

June 30, 2014 317.8

–23

July 18, 2014 336.7

–24

Sep. 1, 2014 381.6 3000

3500

4000

4500

5000

5500 6000 Wavelength, Å

6500

7000

7500

8000

Fig. 2. Low-resolution spectra of Nova V339 Del. The spectra are shifted relative to each other by a constant value starting from the first one. The date of observations and the number of days elapsed after the presumed brightness maximum are indicated to the right of each spectrum. The fluxes are given on a logarithmic scale. ASTRONOMY LETTERS

Vol. 42 No. 1 2016

14

TARASOVA, SKOPAL

Hα

2

Aug. 15, 2013

0

Aug. 19, 2013

Flux/Fluxcont-const

~ ~

~ ~

Sep. 1, 2013 –100

Sep. 13, 2013

–150

Oct. 10, 2013

–200

Nov. 8, 2013 –620 –3000

–2000

+0

+520

–1000 0 1000 Radial velocity, km s–1

2000

3000

Fig. 3. Evolution of the Hα profiles for Nova V339 Del from the phase of an optically thick envelope to the nebular one. The profiles are normalized to the continuum flux and shifted relative to each other by a constant value starting from the first one.

in the light curve. It should be noted that at the stage from pre-maximum brightness to the time of the transition to its gradual decline (Figs. 3 and 4), the Hα and Hβ profiles changed dramatically. Before and at the brightness maximum (August 15 and 19, 2013), these lines had P Cyg profiles typical for such novae. In our previous paper (Skopal et al. 2014), by analyzing the Hα profile, we showed that it consisted of two components located symmetrically relative to zero radial velocity. We explained this result by the fact that the envelope in the fireball phase is a biconical ionization structure. The Hα and Hβ profiles obtained on September 1 and 13, 2013, (Figs. 3

and 4) at the plateau phase in the light curve consists of two components with radial velocities of ∼0 and 500 km s−1 . After the plateau phase in the light curve (October 10, 2013), the profiles became threecomponent ones; a component from the red side at a radial velocity of ∼600 km s−1 was added. In our view, this component may have been veiled by a hidden P Cyg profile. The plateau phase at the early stage of outburst development (two weeks after the brightness maximum) also took place in another nova, V2491 Cyg. In addition, the line profiles for Nova V2491 Cyg were similar in shape to those for Nova V339 Del, as was their subsequent evolution. ASTRONOMY LETTERS

Vol. 42

No. 1

2016

STRUCTURE AND CHEMICAL COMPOSITION

15

2 Hβ

Aug. 15, 2013

1 0 –1

Aug. 19, 2013 ~ ~

~ ~

Flux/Fluxcont-const

–10 Sep. 1, 2013

–20 Sep. 13, 2013

–30 Oct. 10, 2013

–40 Nov. 8, 2013

–620

–50 –2000

+0

+520

0 Radial velocity, km s–1

2000

Fig. 4. Evolution of the Hβ profiles for Nova V339 Del from the phase of an optically thick envelope to the nebular one. The profiles are presented in the same way as those in Fig. 3.

In the period of a gradual brightness decline (from day 253 to 382 after the brightness maximum), the Hα and Hβ profiles changed, the central component weakened significantly, and the blue component strengthened noticeably (Fig. 6). Once the envelope had become optically thin, the hitherto veiled blue component may have finally manifested itself. The shape of the forbidden [O III] 5007 and [N II] 5755 line profiles (Fig. 5), which formed as the final threecomponent one with a more intense blue component at the phase of a rapid brightness decline, when the blue component was not dominant in the Hα and Hβ lines, can be a confirmation of this assumption. ASTRONOMY LETTERS

Vol. 42 No. 1 2016

At the beginning and after the phase of a gradual brightness decline and later, the profiles of all lines became similar (Fig. 6); they consisted of three components among which the blue component was more intense and the red one was least intense (Fig. 7). The Hα profiles, whose shape in the developed nebular phase was distorted by the strengthened forbidden [N II] 6548 and 6584 lines, constituted an exception, In addition, the [Fe VII] 6087 line profiles differed from the profiles of other lines (Fig. 8). They consisted of two components of the same intensity with radial velocities of −480 and 600 km s−1 for the blue and red components, respectively.

16

TARASOVA, SKOPAL

1.0 Oct. 10, 2013

0.5

Flux/Fluxmax-const

Hα 0 Hβ –0.5 He I 5876 –1.0 [O III] 5007

[N II] 5755

+0 520

–2.0 –4000

–620

–1.5

–2000 0 2000 Radial velocity, km s–1

4000

Fig. 5. Profiles of the strongest spectral lines in Nova V339 Del at the phase of a rapid brightness decline. For clarity, the profiles are normalized to the peak line flux and are shifted relative to each other by a constant value starting from the first one.

We assumed that the envelope initially consisted of three rather than two components: a central diskshaped component located in the equatorial plane and two conical circumpolar ejecta. Such a model for the envelope of a nova was considered by Mustel and Boyarchuk (1970) after the analysis of direct images and spectra for novae. Subsequently, Boyarchuk and Gershberg (1977) showed that a considerable variety in the shape of spectral line profiles could be observed, depending on the orientation of the symmetry axis of such an envelope relative to the direction toward the Earth. A similar model for the envelopes of novae was also calculated by Hutchings (1972) and Solf (1983). It should be noted that the “disk + circumpolar ejecta” model was proposed by Ribeiro et al. (2011) when modeling the envelope of Nova V2491 Cyg with profiles similar in shape to those for the Nova being studied. However, despite a great similarity between the profile shapes at and after the plateau phases, some difference was observed in the developed nebular phase; it consisted in the fact that the line profiles for Nova V2491 Cyg were more symmetric, and none of the components stood out in intensity. We constructed the synthetic profiles for the Hα line using the proposed model and compared them

with the observed line profiles that were retrieved from the ARAS database, because they were obtained with a higher resolution (13446–13078). We computed the synthetic Hα line profile with the SHAPE software package (Steffen and Lopez 2006; Steffen et al. 2011). A simple geometrical model was considered without investigating the physical conditions in the Nova envelope. In our modeling, we specified the expansion velocity in individual envelope components and the angle between the direction toward the observer and the equatorial plane of the envelope. In this model, we took the envelope expansion velocity to be 900 ± 50 km s−1 in the circumpolar ejecta and 800 ± 50 km s−1 in the disk. The closest coincidence between the synthetic and observed profiles was obtained for an inclination of the orbital plane of the binary system (the equatorial plane in the model) with respect to the observer equal to 65◦ . Figure 9 presents the synthetic and observed line profiles. The model profiles were compared with the observed Hα profile. We used the Hα line spectrum averaged over two dates (November 13 and 19, 2013). We associated the difference in the intensities of individual line profile components with different ejected masses in individual envelope structures. We simplified the synthetic ASTRONOMY LETTERS

Vol. 42

No. 1

2016

STRUCTURE AND CHEMICAL COMPOSITION

17

1.0 Nov. 8, 2013

Flux/Fluxmax-const

0.5

Hα 0 Hβ

–0.5

He I 5876

[O III] 5007

–4000

+0 520

[N II] 5755

–620

–1.0

–2000 0 2000 Radial velocity, km s–1

4000

Fig. 6. Profiles of the strongest spectral lines in Nova V339 Del at the beginning of the phase of a gradual brightness decline. The line profiles are presented in the same way as those in Fig. 5.

spectrum construction procedure by taking the density ratio in different envelope structures instead of the mass ratio as yet another parameter. The density ratio in the disk and circumpolar ejecta was taken to be 1:1.3:2.2, i.e., the density in the circumpolar ejecta is higher than that in the disk by a factor of 1.3 and 2.2. As can be seen from Fig. 9, the coincidence between the observed and synthetic line profiles is very close. The difference in the wings of the line profile stems from the fact that we disregarded the contribution of the forbidden [N II] 6548 and 6584 lines, which are still very weak at this phase. The remaining differences can be explained by a nonuniformity of the distribution of matter in various envelope structures and by some deviation of our model from the real one. Thus, we think that the three-component envelope structure we propose is quite admissible. An argument for the proposed model, which explains the asymmetry in the spectral lines as the result of a deviation from central symmetry in the distribution of matter, is the existence of an opposite asymmetry in other novae, in particular, in Nova Mon 2012 (Shore et al. 2013d). The difference in the intensities of the red and blue line profile components can also stem from the fact ASTRONOMY LETTERS

Vol. 42 No. 1 2016

that part of the radiation from the red circumpolar ejection is screened by dust at an inclination of 65◦ (Taranova et al. 2014), which can be concentrated in the equatorial region of the envelope. Such an asymmetry in the line profiles is observed in young stars with dust disks and with large inclinations of the equatorial region. DETERMINATION OF SOME PHYSICAL CHARACTERISTICS FOR THE ENVELOPE OF NOVA V339 Del AND DISCUSSION OF OUR RESULTS

The Abundances of Some Chemical Elements To estimate the abundances of chemical elements in the Nova envelope, we calculated the fluxes in the spectral lines that were extracted from the spectra of the Nova in the nebular phase. These are five spectra taken between days 253 and 382 after the brightness maximum. To determine the observed fluxes, it was necessary to determine the color excess E(B − V ). We determined it using the calibrations from Munari and Zwitter (1997) that relate the color excess E(B − V ) to the equivalent width of the interstellar Na I 5890 line. Based on our high-resolution

18

TARASOVA, SKOPAL

1.0 Apr. 25, 2014

Flux/Fluxmax-const

0.5

Hα 0 Hβ He I 5876 –0.5 [O III] 5007

–3000

–2000

520

+0

–620

–1.0 [N II] 5755

–1000 1000 0 Radial velocity, km s–1

2000

3000

Fig. 7. Profiles of the strongest spectral lines in Nova V339 Del in the developed nebular phase. The line profiles are presented in the same way as those in Fig. 5.

1.0 [Fe VII] 6086 Flux/Fluxmax

0.9 0.8 0.7 0.6 0.5 –3000

–2000

–480 +100 +600 –1000 0 1000 Radial velocity, km s–1

2000

3000

Fig. 8. [Fe VII] 6087 line profiles for Nova V339 Del normalized to the peak flux.

observations, we determined the equivalent width of the Na I 5890 line. It is 0.386 ± 0.003 A˚ and, according to the calibrations from Munari and Zwitter (1997), corresponds to the color excess E(B − V ) = 0.180. The inferred color excess almost coincides with the color excess that was derived by the same method by Munai et al. (2013) ((E(B −

V ) = 0.17) and Tomov et al. (2013) (E(B − V ) = 0.182) and is close to its value obtained by Chochol et al. (2013) (E(B − V ) = 0.13) from photometric observations. Using Schlegel’s interstellar extinction maps, Burlak et al. (2014) obtained EB−V = 0.18. We took 0.18 as the final E(B − V ). The calculated dereddened fluxes in selected spectral lines are presented in Table 2. ASTRONOMY LETTERS

Vol. 42

No. 1

2016

STRUCTURE AND CHEMICAL COMPOSITION

1.0

Hα

V339 Del

19

Nov. 13, 19, 2013

Radial flux

0.8

0.6

0.4

0.2

0

–4000

–3000

–2000

–1000 0 1000 Radial velocity, km s–1

2000

3000

4000

Fig. 9. Observed (solid thin line) and synthetic (solid thick line) Hα profiles in the early nebular phase on November 13 and 19, 2013.

Because of the absence of spectral lines from which the electron temperature could be determined, we took it to be 10 000 K and calculated the electron density from forbidden oxygen lines. For this purpose, we compared the observed ratio R(O III) = (I(4959) + I(5007))/I(4363) with the theoretical one. The theoretical relation between these parameters was taken from Seaton (1975). Since the [O III] 4363 line constitutes a blend with Hγ and since the contribution from the latter can distort significantly the electron density estimate, we isolated the [O III] 4363 line from the blend by scaling Hγ from the Hβ line using the theoretical line intensity ratio Hγ/Hβ = 0.47 (Osterbrock 1974) and assuming the envelope to be transparent to the emission in Balmer hydrogen lines. The abundances of ions of chemical elements relative to hydrogen for forbidden lines are determined from the relation I(λ) j(Hβ) N (X i ) (1) + = I(Hβ) j(λ) , N (H ) where I(λ) and I(Hβ) are the spectral line intensities in the observed spectrum, j(Hβ) and j(λ) are the volume emissivities for H(β) and a forbidden line with wavelength λ. The NEBULAR.IONIC code (Shaw and Dufour 1995) computes the volume emissivities for the forbidden line and Hβ for given temperature and density and then computes the abundances of ions of chemical elements relative to hydrogen using ASTRONOMY LETTERS

Vol. 42 No. 1 2016

the observed intensities of forbidden lines normalized to the Hβ intensity. Using this code, we determined the nitrogen, oxygen, neon, and argon abundances. To estimate the abundances of iron ions, we used the [Fe IV] 5176 and [Fe VII] 6087 lines. The relations for the volume emissivities for these lines were taken from Nussbaumer and Storey (1978, 1982). The abundances of ions of chemical elements for permitted lines are determined from the relation N (X i+ ) λ(X i+ ) αeff (Hβ) I(λ) = , λ(Hβ) αeff (λ) I(Hβ) N (H+ )

(2)

where αeff (Hβ) and αeff (λ) are the recombination coefficients, I(λ) and I(Hβ) are the observed intensities of a spectral line with wavelength λ(X i+ ) and Hβ. To estimate the helium abundance, we used the He I 5876 and He II 4686 lines. The relation to determine the helium ion abundance was taken from Aller (1984). When calculating the helium ion abundance from the He I 5876 line, we took into account the contribution from the collisional excitation of the metastable 2S2 3 S He I level (Clegg 1987). Below, we provide the relations used to estimate the abundances of helium ions from the He I 5876 and He II 4686 lines: N (He+ ) = −0.133 + 0.235 log(te ) (3) log N (H+ ) I(5876) , + log I(Hβ)

20

TARASOVA, SKOPAL

Table 2. Emission line fluxes from V339 Del corrected for the reddening E(B − V ) = 0.18 [10−13 erg cm−2 s−1 ] JD 2450000+ ˚ line (A)

6773.5

6797.5

6838.5

6857.5

6902.3

5.4

7.7

[Ne V] 3345.9 [Ne V] 3425.8

34.2

12.4

9.8

8.1

[Ne III] 3868.7 + H8

36.1

38.9

13.0

11.7

10.4

[Ne III] 3968.7 + H

24.6

22.8

7.3

6.1

5.2

Hδ

54.8

48.6

14.7

15.5

11.8

218.9

197.0

54.9

56.2

37.3

6.4

6.2

1.8

1.9

1.0

N II + N III 4640

51.3

48.8

17.0

16.9

12.8

He II 4685.7

23.8

23.5

9.0

8.5

6.7

Hβ

104.1

97.4

37.0

35.9

25.7

[O III] 4958.9 +

277.2

312.5

155.6

161.0

142.1

[O III] 5006.8

833.3

949.2

480.4

498.7

447.4

[Fe VI] 5177.0

6.5

6.4

3.3

2.4

2.4

[Fe VI] 5279.1 + [Ca V] 5308.9

4.0

3.8

1.4

1.1

0.9

He II 5411.5

1.6

1.6

0.8

0.8

0.6

N II 5679.6

9.3

9.2

4.2

4.9

3.0

[N II] 5754.8

123.4

124.1

56.2

68.0

42.4

He I 5875.6

10.5

10.6

5.0

5.7

3.7

3.8

3.1

1.5

1.7

1.2

14.06

12.8

7.0

7.5

6.9

7.0

6.4

3.2

3.2

2.9

[O III] 4363.2 + Hγ He I 4471.4

[Fe VII] 6085.5 [O I] 6300.2 [O I] 6363.9 + N II6379.6 Hα + [N II] 6548, 6584

395

389

214

244

223

He I 6678.2

2.9

2.9

1.6

1.7

1.4

[Ar V] 7006.3

1.1

1.0

0.3

0.4

0.3

He I 7065.7

4.8

4.2

1.7

1.8

1.6

[Ar III] 7135.8

4.0

4.6

2.1

2.2

2.2

[Ar IV] 7235 + C II 7236.19

6.3

6.3

2.8

2.3

2.2

40.3

41.6

20.8

20.3

19.4

[O II] 7319.4, 30.7

ASTRONOMY LETTERS

Vol. 42

No. 1

2016

STRUCTURE AND CHEMICAL COMPOSITION

log

N (He++ ) = −1.077 + 0.135 log(te ) N (H+ ) I(4686) 0.135 , + log + te I(Hβ)

(4)

where te = Te /104 K, I(5876) and I(4686) are the observed intensities of the He I 5876 and He II 4686 lines. The total helium abundance was derived by adding the He+ and He++ abundances. The He0 ionization state was disregarded, because it is generally believed that the entire helium in the nova envelope is ionized. To determine the abundances of nitrogen ions, apart from the forbidden [N II] 5755 line, we used the permitted N II 5680 and N III 4640 lines. We calculated the nitrogen abundance from the permitted lines using the relation I(λ) N (X i+ ) + = Xi (Te ) I(Hβ) , N (H )

and O+ (7325) + O++ (5007) O = ICF, H H+ (Hβ)

For nitrogen, N (9) H (N+ (5755) + N2+ (5680))/2 + N3+ (4640) = H+ (Hβ) and N+ (5755) N = + ICF, H H (Hβ)

(7)

(10)

where ICF =

O+ (7330) + O++ (5007) . O+ (7330)

(11)

For argon, Ar (12) H Ar++ (7136) + Ar2+ (7237) + Ar3+ (7006) = H+ (Hβ)

(5)

where Xi (Te ) = χ0 tη and t = Te /104 K. The coefficients χ0 and η were taken from Golovatyj et al. (1997) and are χ0 = 0.205, η = −0.41 for the N II 5680 line and χ0 = 0.014, η = −0.71 for the N III 4640 line. We determined the abundances of chemical elements either as the sum of the abundances of ions corresponding to different atomic ionization stages or from ions using the ionization correction factors (ICFs). The ionization correction factor is used when there is no complete set of ions with sequential atomic ionization stages to determine the abundance of a chemical element. In this case, the contribution from missing ions to the elemental abundance can be taken into account using the ICF. We derived the oxygen, nitrogen, and argon abundances by the two methods and the neon and iron abundances using only the ICF. We took the relation for the ICFs for oxygen and nitrogen from Aller (1984), for argon and neon from Saizar et al. (1992), and for iron from Andrea et al. (1994). Below, we provide the relations used to derive the total abundances of chemical elements. For oxygen, O (6) H O0 (6300) + O+ (7320 + 7330)/2 + O++ (5007) = H+ (Hβ)

21

and Ar++ (7136) Ar = ICF. H H+ (Hβ)

(13)

Here, ICF =

N

2+

N (5680)

(14)

and Ar3+ (7006) Ar = ICF, H H+ (Hβ)

(15)

where ICF =

He . He (4686) ++

(16)

For neon, Ne++ (3869) Ne = ICF, H H+ (Hβ)

(17)

where O+ (7325) + O++ (5007) . O++ (5007)

(18)

Fe5+ (5176) + Fe6+ (6087) Fe = ICF, H H+ (Hβ)

(19)

ICF = For iron,

where

where +

++

He (5876) + He (4686) . ICF = He+ (5876) ASTRONOMY LETTERS

Vol. 42 No. 1 2016

(8)

ICF =

He+ (5876) + He++ (4686) . He++ (4686)

(20)

22

TARASOVA, SKOPAL

In addition, we derived the Ca3+ 5309 abundance. For this purpose, we took the relation (Aller 1984) N (Ca3+ ) (21) N (H+ ) √ I(5309) 0 = 2.6 × 10−6 E4,2 , t × 101.14/t I(Hβ) where 0 = 1.24t−0.87 . E4,2

(22)

The total abundances of helium, nitrogen, oxygen, neon, argon, and iron ions as well as the calcium ion abundance and the electron density are given in Table 3. The oxygen abundances determined by different methods differ insignificantly. The nitrogen abundance derived from the permitted N II 5680 line differs from that inferred from the forbidden [N II] 5755 line by more than an order of magnitude. This trend in determining the nitrogen abundance from permitted and forbidden lines was pointed out by Arkhipova et al. (2002) and in our previous papers (Tarasova 2013a, 2013b, 2014a, 2014b). It was assumed in these papers that such a difference might stem from the fact that these lines are formed in different regions of the envelope. Table 3 presents the nitrogen abundances found by the two methods. The first value was obtained as the mean of all the accessible atomic ionization stages whose ion lines are present in the spectrum, while the second value is the abundance derived from the forbidden line with the ICF. The argon abundances estimated from the [Ar III] 7136 and Ar V 7006 lines using the ICF differ from those estimated as the mean of the values from the [Ar III] 7136, Ar V 7006, and [Ar IV] 7237 lines. In the second case, the estimate for Ar2+ obtained from the [Ar IV] 7237 line makes a major contribution to the argon abundance estimate. Such a difference can result from the fact that the argon abundance inferred from the [Ar IV] 7237 line may be overestimated because of the C II 7236 line, which was fairly strong in the previous phase and can still be present in the spectrum. Therefore, we took the mean of the values obtained from the [Ar III] 7136 and Ar V 7006 lines as the final argon abundance. We determined the iron abundance from the [Fe IV] 5176 and [Fe VII] 6087 lines not only for an electron temperature of 10 000 but also for 15 000 and 20 000 K. This is because we obtained an overestimated iron abundance compared to the solar one for 10 000 K. We assumed that the temperature could be higher and also calculated the iron abundance for 15 000 and 20 000 K for the same mean electron density of 3.3 × 106 cm−3 . Indeed, we obtained an abundance lower than the solar one for 20 000 K and

a nearly solar abundance for 15 000 K. Table 3 gives the iron abundance for 15 000 K. For comparison, Fig. 10 shows the abundances of chemical elements relative to the solar ones on a logarithmic scale for all Fe II novae and Nova V339 Del. The abundances for the Nova being investigated in this figure are marked by the filled circles. It can be seen from Fig. 10 that the helium abundance in Nova V339 Del is the same as that in the Sun. The nitrogen and oxygen abundances differ from the solar ones by more than two orders of magnitude and one order of magnitude, respectively. These nitrogen and oxygen abundance estimates are the mean ones among all of the novae presented in this figure. Figure 10 shows the mean of the values obtained for the forbidden and permitted lines for nitrogen. The neon and argon abundances turned out to be slightly lower than the solar ones. This may be because we underestimated the fluxes in the spectral lines from which we determined the ion abundances. These lines are weak, and the accuracy of the flux for them is about 50%. We think it necessary to note that in this paper we obtained rough estimates of the abundances of some chemical elements in the Nova envelope, because we disregarded the envelope nonuniformity in temperature and density and its complex structure. In fact, we used a single-zone model. This is because we do not have the necessary set of spectral lines at our disposal to determine the physical characteristics (density and temperature) in different envelope regions. However, in our view, such an estimate is admissible. In our previous paper (Tarasova 2014a, 2014b), we estimated the abundances of some chemical elements in Nova V2491 Cyg also using a singlezone model and compared our estimates with those from Munari et al. (2011), who took into account the envelope inhomogeneity. Our estimates virtually coincide with those from Munari et al. (2011). We associate such a coincidence with the fact that the errors in the spectral line fluxes are comparable to those arising when the envelope inhomogeneity is disregarded. The observational errors are predominantly determined by how accurately the fluxes were calibrated. As a rule, the errors related to the flux loss do not exceed 10–20%. We minimize these errors by observing the spectrophotometric standard almost simultaneously with the star. We then compare the magnitudes calculated from our spectrophotometric data with the photometric observations available in the literature. In addition, the errors in calculating the spectral line fluxes depend on how unambiguously the continuum level is determined, which is not always obvious. Usually, these errors for sufficiently strong lines, which, as a rule, are involved in determining the abundances of chemical elements, are from 5 to 20%. In addition, the accuracy of the continuum level ASTRONOMY LETTERS

Vol. 42

No. 1

2016

STRUCTURE AND CHEMICAL COMPOSITION

23

Table 3. Electron density and abundances of some chemical elements in the envelope of V339 Del Date

Ne , 106 [O III]

He+ /H 5876

He++ /H 4648

He/H

Apr. 25, 2014

6.1

0.074

0.026

0.10

May 20, 2014

4.4

0.080

0.028

0.11

June 30, 2014

1.8

0.090

0.028

0.13

July 18, 2014

1.9

0.120

0.027

0.14

Sep. 1, 2014

1.3

Date

0.110 −3

+

N /H, 10 5755

0.030 −3

+

N /H, 10 5680

N

0.14 −3

++

/H, 10 4640

N/H, 10−3

Apr. 25, 2014

0.95

18.31

6.90

16.53/2.28

May 20, 2014

0.95

19.43

7.01

17.17/2.23

June 30, 2014

1.01

23.30

6.43

18.57/1.91

July 18, 2014

1.26

28.02

6.59

21.21/2.49

Sep. 1, 2014

1.08

23.94 −3

0

6.97 −3

+

19.48/1.89 −3

++

Date

O /H, 10 6300, 6363

O /H, 10 7320+7330

O /H, 10 4363, 4959, 5007

O/H, 10−3

Apr. 25, 2014

0.120

2.08

2.92

4.08/5.34

May 20, 2014

0.093

1.98

2.66

3.74/4.93

June 30, 2014

0.078

1.96

1.73

2.79/3.47

July 18, 2014

0.088

1.99

1.93

3.01/3.56

Sep. 1, 2014 Date

0.095 Ar

++

2.46 −6

/H, 10 7136

Ar

1.85 −6

2+

/H, 10 7237

Ar

3.18/3.91 −6

3+

/H, 10 7006

Ar/H, 10−6

Apr. 25, 2014

0.80

26.59

0.26

27.6 /1.00/0.84

May 20, 2014

0.82

27.50

0.23

28.5 /0.87/0.86

June 30, 2014

0.70

32.20

0.15

33.0 /0.67/0.73

July 18, 2014

0.77

27.16

0.20

28.1 /1.11/0.80

Sep. 1, 2014

1.21 ++

37.54

0.21

Date

Ne /H 10−5 3426

Ne/H 10−5

Fe 10−6 5176

Fe 10−6 6087

Fe/H 10−5

Ca3+ 10−6 5309

Apr. 25, 2014

4.97

–

8.52

2.55

1.98

3.96

1.71

May 20, 2014

5.02

7.30

8.76

2.79

1.73

4.01

1.74

June 30, 2014

3.48

6.46

7.42

3.82

2.20

6.31

1.68

July 18, 2014

3.26

5.27

6.64

2.61

2.57

6.29

1.74

Sep. 1, 2014

3.80

6.05

8.85

3.94

2.54

6.93

1.56

Vol. 42 No. 1 2016

5+

38.9 /0.96/1.26

Ne /H 10−5 3869

ASTRONOMY LETTERS

3+

6+

24

TARASOVA, SKOPAL

V1425 Aql V443 Sct V1668 Cyg V977 Sco PW Vul V827 Her V2468 Cyg V1494 Aql V2467 Cyg V339 Del

4

log(Xi/XiSUN)

3

2

1

0

–1 He

N

O

Ne

Ar

Fe

Element Fig. 10. Abundances of chemical elements relative to the solar ones for some Fe II novae. The abundances for V339 Del are marked by the filled circles. The dotted line marks the abundances of chemical elements that do not differ from the solar ones.

determination depends on noise, which depends on the spectral region in our case. The CCD is most sensitive in the red region and loses up to 50% of its sensitivity in the blue one. Thus, the maximum observational errors can reach 50%. On the other hand, as was pointed out, in particular, by PerezMontero and Diaz (2003), the errors related to the assumption about the absence of temperature and density stratification in the envelope can reach 50%.

Estimating the Mass of the Emission Envelope We determined the envelope mass by two methods. The first method was described in our previous paper (Skopal et al. 2014). In this method, we estimated the mass of the emitting envelope material by determining its emission measure. For this purpose, we calculated the synthetic spectral energy distribution for the star and compared it with the observed one. The model and observed energy distributions for three dates (June 30, July 18, and September 1, 2014) are presented in Figs. 11a–11c. According to the synthetic energy distribution, we found that the optical continuum is dominated by the radiation from the white dwarf, while the nebular continuum is characterized by a high electron temperature,

20 000–40 000 K, and an emission measure EM ∼ 2.5 ± 0.3 × 1059 cm−3 . In this method, the envelope mass depends on its expansion velocity and the time interval elapsed from the phase of an optically thick envelope. Using a mean envelope expansion velocity of 750 km s−1 , we will obtain a mass of 1.3 ± 0.2 × 10−4 M . If the terminal velocity of 1250 km s−1 determined from the Hβ profile in the nebular phase is used, then the mass will be twice as large, 2.7 ± 0.4 × 10−4 M . This value is close to what we obtained at the phase of an optically thick envelope, when the emission measure was EM ∼ 1062 cm−3 (Skopal et al. 2014). However, given the filling factor, which, according to our envelope model, is 0.28, the mass of the emission envelope decreases by a factor of 0.53 (see Eq. (4) in Skopal et al. (2014)). Finally, the envelope mass calculated in this way lies within the range 6.9 × 10−5 −1.4 × 10−4 M . The mass of the emission hydrogen envelope The can also be determined as M = ne mH V . envelope volume is derived by comparing the observed and theoretical luminosities of the Nova in Hβ: 4πD 2 FHβ = 4πHβ V . As a result, we −1 have M = 7.99FHβ D2 t0.85 e ne M , where 4πHβ = ASTRONOMY LETTERS

Vol. 42

No. 1

2016

STRUCTURE AND CHEMICAL COMPOSITION

2

(a)

log(Flux) [erg cm–2 s–1 Å–1] + 13

V339 Del June 30, 2014 1

Spectrum

1

WD

2

Nebula

3

SED

4

1

0

2

–1

4

3

3500

4500

5500 Wavellength, Å

2

6500

(b) V339 Del July 18, 2014

log(Flux) [erg cm–2 s–1 Å–1] + 13

25

1

1

7500

Spectrum

1

WD

2

Nebula

3

SED

4

0

2

3

–1

4

3500

4500

5500 Wavellength, Å

6500

7500

Fig. 11. Observed (spectrum-solid thin line) and synthetic (SED-solid thick line) spectral energy distributions for the Nova in the nebular phase on June 30, 2014 (a), July 18, 2014 (b), and September 1, 2014 (c). The model energy distribution includes the energy distributions for the white dwarf (WD-thin solid line) and the Nova envelope (Nebula-dotted line).

ASTRONOMY LETTERS

Vol. 42 No. 1 2016

26

TARASOVA, SKOPAL

2 (c)

log(Flux) [erg cm–2 s–1 Å–1] + 13

V339 Del Sept. 1, 2014 1

1

Spectrum

1

WD

2

Nebula

3

SED

4

0

2 3

–1

4

3500

4500

5500 Wavellength, Å

6500

7500

Fig. 11. (Contd.)

1.24 × 10−25 t−0.85 ne nH+ is the volume emissivity in e the Hβ line, mH = 1.64 × 10−24 g, FHβ is the flux in the Hβ line expressed in units of 10−11 erg cm−2 s−1 , D is the distance to the Nova, te is the electron temperature of the envelope normalized to 104 , ne is the electron density, and M is the solar mass. We took the distance D = 3.2 kpc from Chochol et al. (2014). A distance of 2.7–3 kpc close to this value was obtained by Burlak et al. (2015). The last distance estimate was obtained by Schaefer et al. (2014). In this paper, the distance to the Nova is slightly larger than the previous estimates, 4.54 ± 0.59 kpc. We estimated the envelope mass for Te = 10 000 K. It is 1.6 × 10−5 . We calculated the total mass of the envelope by taking into account the fact that, according to our estimates, the hydrogen mass fraction in the Nova envelope is X = 0.594. As a result, the envelope mass is 2.7 × 10−5 M . However, if we assume, based on the model energy distribution, that the electron temperature can be higher, for example, 20 000 K, then the envelope mass will be slightly larger, ∼4.9 × 10−5 M . However, the envelope masses estimated by different methods differ as before. This difference may stem from the fact that the first method shows an overestimated electron temperature of the envelope for the nebular phase (20 000– 40 000 K). This may be because the observed shortwavelength part of the spectrum is not enough for the

proper modeling. In the second method, we assumed the envelope to be completely transparent (optically thin) to the radiation from the white dwarf, which may not quite correspond to reality. Therefore, as the final estimate we took the mean of two different estimates. It is ≈7 × 10−5 M . It should be noted that Taranova et al. (2014) found the total mass of the envelope to be 1.1 × 10−6 M . However, the mass was estimated by assuming the dust fraction in the envelope of the Nova being investigated to be 1/200 of the total mass. CONCLUSIONS The spectrophotometric evolution of the Nova showed that it belongs in its spectral characteristics to Fe II novae. However, as was shown in our previous paper (Skopal et al. 2014), a distinctive feature of this Nova was its unusual behavior at the early phase of outburst development. At this phase, the envelope luminosity exceeded the Eddington one and was variable. In addition, the envelope was expanding nonuniformly. It was also shown in this paper that the envelope could have a nonspherical shape (bipolar structure). In this paper, we continued to investigate the Nova envelope at a later stage when the Nova entered the nebular phase. The new results we obtained, namely the complex shape of the spectral line profiles and their evolution, showed that the Nova envelope is ASTRONOMY LETTERS

Vol. 42

No. 1

2016

STRUCTURE AND CHEMICAL COMPOSITION Table 4. Basic characteristics of Nova V339 Del ≈2456520.6

JDmax

4.23 mag

Vmax

−8.7

MVmax Speed class (t2 ≤10)

fast t2 = 10 days

Distance (kpc)

2.7–4.5 kpc

Color excess E(B − V )

0.18

White dwarf mass (M )

1.04 ± 0.021

Spectral type

Fe

Type of white dwarf

CO

Mean envelope expansion velocity

750 ± 50 km s−1

Elemental abundances relative to their solar values 1.2 ± 0.2

N/N

122 ± 60

O/O

7.8 ± 1.6

Ne/Ne

0.8 ± 0.4

Fe/Fe

0.9 ± 0.5

1

≈7 × 10−5

Chochol et al. (2014).

actually nonspherical. According to our model, the envelope consists of circumpolar ejecta and an equatorial disk. From our model we found the velocity of matter to be ∼900 ± 50 km s−1 in the circumpolar ejecta and ∼800 ± 50 km s−1 in the disk. We chose the inclination of the orbital plane with respect to the observer to be 65◦ . The three-component model of the emission envelope we proposed showed a close coincidence between the observed and calculated Hα profiles (Fig. 9). In this paper, using our spectrophotometric observations of Nova V339 Del, we not only investigated the structure but also estimated some of the physical characteristics of the Nova envelope: its mass and the abundances of some chemical elements. The mass of the Nova envelope was estimated by two methods. We obtained envelope masses differing between themselves; therefore, we took a mean envelope ASTRONOMY LETTERS

mass of ≈7 × 10−5 M as the final estimate. Our estimates of the abundances of chemical elements in the envelope of Nova V339 Del showed that it has a chemical composition typical for such novae. In particular, the helium abundance was the same as that in the Sun, while the nitrogen and oxygen abundances exceeded the solar ones by a factor of 122 and almost 8, respectively. The neon and iron abundances were nearly solar. We summarized the characteristic properties and parameters of the Nova in Table 4. In this table, the white dwarf mass was taken from Chochol et al. (2014). ACKNOWLEDGMENTS We are grateful to all of the variable star observers who contributed to the creation of the worldwide AAVSO database and to K. Graham, an observer of the ARAS database, for the use of his observations in our paper. This work was supported by VEGA grant no. 2/0002/13 of the Slovak Academy of Sciences. REFERENCES

He/He

Envelope mass (M )

27

Vol. 42 No. 1 2016

1. L. H. Aller, Physics of Thermal Gaseous Nebulae (Reidel, Dodrecht, 1984). 2. J. Andrea, H. Drechsel, and S. Starrfield, Astron. Astrophys. 291, 869 (1994). 3. V. P. Arhipova, M. A. Burlak, and E. F. Esipov, Astron. Lett. 28, 100 (2002). 4. V. P. Arhipova, E. F. Esipov, and G. V. Sokolov, Astron. Lett. 23, 713 (1997). 5. A. A. Boyarchuk and R. E. Gershberg, Sov. Astron. 21, 275 (1977). 6. A. M. Burlak, V. F. Esipov, G. V. Komissarova, V. I. Shenavrin, O.G Taranova, M. A. Tatarnikov, and A. A. Tatarnikova, Baltic Astron. 24, 109 (2015). 7. V. I. Burnashev, Byull. Abastumansk. Astrofiz. Observ. 59, 83 (1985). 8. D. Chochol, S. Shugarov, T. Pribulla, and I. Volkov, Contrib. Astron. Obs. Skalnate Pleso 43, 330 (2014). 9. R. E. S. Clegg, Mon. Not. R. Astron. Soc. 229, 31 (1987). 10. M. J. Darnley, M. F. Bode, R. J. Smith, and A. Evans, Astron. Telegram, No. 5279 (2013). 11. N. R. Deacon, D. W. Hoard, E. A. Magnier, Y. S. Jadhav, M. Huber, K. C. Chambers, H. Flewelling, K. W. Hodapp, N. Kaiser, R. P. Kudritzki, N. Metcalfe, and C. Waters, Astron. Astrophys. 563, A129 (2014). 12. D. Denisenko and G. Masi, IAU Circ., No. 9258 (2013). 13. V. V. Golovatyj, A. Sapar, T. Feklistova, and A. F. Kholtygin, Astron. Astrophys. Trans. 12, 85 (1997). 14. J. B. Hutchings, Mon. Not. R. Astron. Soc. 158, 177 (1972). 15. U. Munari and T. Zwitter, Astron. Astrophys. 318, 269 (1997). 16. U. Munari and A. Henden, Inform. Bull. Var. Stars 6087 (2013).

28

TARASOVA, SKOPAL

17. U. Munari, A. Siviero, S. Dallaporta, G. Cherini, P. Valisa, and L. Tomasella, New Astron. 16, 209 (2011). 18. U. Munari, P. Valisa, A. Milani, and G. Cetrulo, Astron. Telegram., No. 5297 (2013a). 19. U. Munari, S. Dallaporta, G. Cherini, P. Valisa, G. Cetrulo, A. Milani, and L. Ghirotto, Astron. Telegram, No. 5304 (2013b). 20. E. R. Mustel and A. A. Boyarchuk, Astrophys. Space Sci. 6, 183 (1970). 21. H. Nussbaumer and P. Storey, Astron. Astrophys. 70, 37 (1978). 22. H. Nussbaumer and P. Storey, Astron. Astrophys. 113, 21 (1982). 23. D. E. Osterbrock, Astrophysics of Gaseous Nebulae (W. H. Freeman, San Francisco, 1974). 24. E. Perez-Montero and A. I. Diaz, Mon. Not. R. Astron. Soc. 346, 105 (2003). 25. V. A. R. M. Ribeiro, M. J. Darnley, M. F. Bode, U. Munari, D. J. Harman, I. A. Steele, and J. Meaburn, Mon. Not. R. Astron. Soc. 412, 170 (2011). 26. P. Saizar, S. Starrfield, G. J. Ferland, R. M. Wagner, J. W. Truran, S. J. Kenyon, W. M. Sparks, R. E. Williams, and L. L. Stryker, Astrophys. J. 398, 651 (1992). 27. N. N. Samus, IAU Circ., No. 9258 (2013). 28. G. H. Schaefer, T. Ten Brummelaar, D. R. Gies, C. D. Farrington, B. Kloppenborg, O. Chesneau, J. D. Monnier, S. T. Ridgway, et al., Nature 515, 234 (2014). 29. M. J. Seaton, Mon. Not. R. Astron. Soc. 170, 475 (1975). 30. R. A. Shaw and R. J. Dufour, Publ. Astron. Soc. Pacif. 107, 896 (1995).

31. S. Shore, P. Skoda, and P. Rutsh, Astron. Telegram, No. 5282 (2013a). 32. S. N. Shore, P. Skoda, D. Korcakova, P. Ki’ek, R. Koubsky, P. Rutsch, M. Slechta, O. Garde, et al., Astron. Telegram, No. 5312 (2013b). 33. S. N. Shore, K. Alton, D. Antao, E. Barbotin, P. Berardi, P. Bohlsen, F. Boubault, D. Boyd, et al., Astron. Telegram, No. 5378 (2013c). 34. S. N. Shore, I. de Gennaro Aquino, G. J. Schwarz, T. Augusteijn, C. C. Cheung, F. M. Walter, and S. Starrfield, Astron. Astrophys. 553, 123 (2013d). 35. A. Skopal, H. Drechsel, T. Tarasova, T. Kato, M. Fujii, F. Teyssier, O. Garde, J. Guarro, J. Edlin, C. Buil, et al., Astron. Astrophys. 569, 112 (2014). 36. J. Solf, Astrophys. J. 273, 647 (1983). 37. W. Steffen and J. A. Lopez, Rev. Mex. Astron. Astrofiz. 42, 99 (2006) 38. W. Steffen, N. Koning, S. Wenger, C. Morisset, and M. Magnor, IEEE Trans. Visual. Comput. Graphics 17, 454 (2011). 39. O. G. Taranova, A. M. Tatarnikov, V. I. Shenavrin, and A. A. Tatarnikova, Astron. Lett. 40, 120 (2014). 40. T. N. Tarasova, Astron. Telegram, No. 5291 (2013a). 41. T. N. Tarasova, Astron. Rep. 57, 95 (2013b). 42. T. N. Tarasova, Astron. Rep. 58, 302 (2014a). 43. T. N. Tarasova, Astron. Lett. 40, 309 (2014b). 44. T. N. Tarasova and D. N. Shakhovskoi, Astron. Telegram, No. 5370 (2013). 45. T. Tomov, K. Ilkiewicz, E. Swierczynski, M. Belcheva, and D. Dimitrov, Astron. Telelegram, No. 5288 (2013). 46. E. O. Waagen, AAVSO Alert Not. 489 (2013).

Translated by V. Astakhov

ASTRONOMY LETTERS

Vol. 42

No. 1

2016