Time evolution of the line emission from the inner circumstellar ring of SN 1987A and its hot spots
Per Groeningsson, Claes Fransson, Bruno Leibundgut, Peter Lundqvist, Peter Challis, Roger A. Chevalier, Jason Spyromilio
aa r X i v : . [ a s t r o - ph ] O c t Astronomy & Astrophysics manuscript no. 10551 November 2, 2018(DOI: will be inserted by hand later)
Time evolution of the line emission from the inner circumstellarring of SN 1987A and its hot spots ⋆ Per Gr¨oningsson , Claes Fransson , Bruno Leibundgut , Peter Lundqvist , Peter Challis , Roger A.Chevalier , and Jason Spyromilio Stockholm Observatory, Stockholm University, AlbaNova University Center, SE-106 91 Stockholm, Swedene-mail: [email protected], [email protected], [email protected] European Southern Observatory, Karl-Schwarzschild-Strasse 2, D-85748 Garching, Germany Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, MS-19, Cambridge, MA 02138. Department of Astronomy, University of Virginia, P.O. Box 400325, Charlottesville, VA 22904, U.S.A.Received; accepted
Abstract.
We present seven epochs between October 1999 and November 2007 of high resolution VLT/UVESechelle spectra of the ejecta-ring collision of SN 1987A.The fluxes of most of the narrow lines from the unshocked gas decreased by a factor of 2 − ∼ ∼ α lineprofile can be traced up to ∼
500 km s − at the latest epoch. This is consistent with the cooling time of shockspropagating into a density of (1 − × cm − . This means that these shocks are among the highest velocityradiative shocks observed. Key words. supernovae: individual: SN 1987A – circumstellar matter, shocks
1. Introduction
Since the explosion more than two decades ago, supernova(SN) 1987A in the Large Magellanic Cloud (LMC) hasbeen extensively observed in multiple wavelength ranges.It is now by far the most studied of all supernovae. Afew months after outburst, the first evidence of the sta-tionary circumstellar matter around the SN came fromdetection of narrow optical lines (Wampler et al. 1988)
Send offprint requests to : P. Gr¨oningsson ⋆ Based on observations made with ESO telescopes at theLa Silla Paranal Observatory under programme IDs 60.A-9022,66.D-0589, 70.D-0379, 074.D-0761, 078.D-0521, 080.D-0727. and UV emission lines by the International UltravioletExplorer (IUE) (Fransson et al. 1989). Resolved imagingobservations later disclosed the now classical ring nebulaaround the SN consisting of three approximately plane-parallel rings; the radioactive SN debris is located atthe center of the inner equatorial ring (ER). The ERis intrinsically close to circular in shape and has a tiltangle of 43 ◦ (Sugerman et al. 2005; Kjær et al. 2007).The two outer rings (ORs) are displaced by ∼ . ∼ . Gr¨oningsson et al.: Line emission from the inner ring of SN 1987A ities ( ∼ . − for the ER), indicate that the ringsystem was ejected about 20 ,
000 years ago, if constant ex-pansion velocities of the rings are assumed. The formationmechanism of the three rings is still not fully understood,although a scenario is that the structure was formed as aresult of a merger with a binary companion in a commonenvelope phase (Morris & Podsiadlowski 2007). However,single star models for the progenitor have also been pro-posed (e.g., Eriguchi et al. 1992; Woosley et al. 1997).The ring structure was photoionized by the EUV andsoft X-ray flash emitted from the SN explosion, and sincethen the gas has been cooling, recombining and fadingroughly linearly with time (e.g., Lundqvist & Fransson1996). This gas can be traced by narrow emission linesand the fading of these lines is consistent with ringgas densities in the range 3 × − × cm − (Lundqvist & Fransson 1996; Pun 2007). Early lightechoes revealed, however, that the observed photoionizedgas is probably only a small fraction of the total massejected from the progenitor system (Gouiffes et al. 1988;Crotts et al. 1989).A new phase was entered, marking the birth of the SNremnant, as the SN debris drives a blast wave into its cir-cumstellar medium. The first indication of the encounterbetween this blast wave and the ER came in 1997 fromHST/STIS observations where broad ( ∼
250 km s − )blue-shifted emission components were detected in H α (Sonneborn et al. 1998). Later it was found that thisinteraction could be detected in HST images already in1995 (Lawrence et al. 2000). The interaction took placein a small region located in the northeastern part ofthe ER (P.A. 29 ◦ ) (Pun et al. 1997). This “hot spot” isusually referred to as Spot 1. Hence, Spot 1 is believedto mark the place were the blast wave first strikes aninward-pointing protrusion of the ER surrounded byan H II region inside the ring (Pun et al. 2002). Theprotrusions are presumably a result of Rayleigh-Taylorinstabilities caused by the interaction of the stellarprogenitor wind with the ER. Since 1997, Spot 1 hassteadily increased in flux and many more clumps havebeen shocked giving rise to more “hot spots” in differentlocations around the ring. Ten years later, there are “hotspots” distributed all around the ER. It is, however, notclear if the ring is really homogeneous, or if it may consistof isolated clumps. Time will tell.In the interaction between the supernova ejecta andthe circumstellar matter a multi-shock system develops.A forward shock (blast wave) propagates into the CSMand a reverse shock is driven backward (in the Lagrangianframe of reference) into the ejecta (Chevalier 1982). Acomplication is that the circumstellar structure is proba-bly complex, with a constant density H II region inside theER (Chevalier & Dwarkadas 1995). When the forwardshock, propagating with a velocity of ∼ − through the circumstellar gas (with density ∼ cm − ),reaches the denser ER ( ∼ cm − ), slower shocks are transmitted into the ring. Because the transmitted shockvelocities depend on both the density profile of the ERand the incident angle, a wide range of shock velocities isexpected ( ∼ − km s − ) (Michael et al. 2002). Inaddition, the interaction with a density jump may createreflected shocks between the ring and the reverse shock.The interaction has been observed in virtually allwavelength ranges, from radio to X-rays (e.g. McCray2007). The high resolution X-ray spectrum observed byChandra shows a number of H and He-like emission linesin the wavelength range 5 −
25 ˚A such as Si XIII-XIV, MgXI-XII, Ne IX-X, O VII-VIII, Fe XVII and N VII emittedby shocks produced by the interaction of the blast waveand the ER. The line profiles indicate gas velocities inthe range ∼ − − (Zhekov et al. 2005,2006; Dewey et al. 2008). As the expanding ejecta collidewith the CSM, the luminosity increases rapidly in radiosynchrotron emission (Gaensler et al. 2007), as well as inX-rays originating from the hot shocked gas (Park et al.2007). The ring evolution has also been studied in mid-IR(Bouchet et al. 2006), near-IR (Kjær et al. 2007), as wellas optical and UV (Pun et al. 2002; Sugerman et al.2002, 2005; Gr¨oningsson et al. 2006, 2008).For this paper, the X-rays are of special interest. Up to1999 both the soft and hard X-rays correlated well withthe radio emission. Since then, however, the flux of softX-rays began to increase much more rapidly than bothradio and hard X-ray fluxes (Park et al. 2007). Instead, inGr¨oningsson et al. (2006) we found that the evolution ofthe flux of the optical high ionization lines from the hotspots correlated well with the soft X-ray emission. Thissuggests that much of the soft X-ray emission arises fromthe ejecta-ring collision, while the hard X-rays mainlycome from the hot gas between the reverse and the for-ward shock. The radio emission continues to correlate wellwith hard X-rays, and is most likely caused by relativisticelectrons accelerated by the reverse shock interior to theER (Zhekov et al. 2006).A complementary view of the interaction is offeredby the IR emission from the dust (Bouchet et al. 2006;Dwek et al. 2008). Dwek et al. find a temperature of thesilicate rich dust of ∼
180 K, heated either by the shockedgas or by the radiation from the shocks. The dust emissioncoincides roughly with the optical emission. A surprisingresult is the comparatively low IR to X-ray ratio, indicat-ing destruction of the dust by the shock. The fact thatthis ratio decreased with time is also consistent with dustdestruction.In this paper we discuss the evolution of the opticalemission lines from high resolution spectroscopic datataken with VLT/UVES. In the first paper in this series(Gr¨oningsson et al. 2006) we discussed the implications ofthe coronal lines from [Fe X-XIV] found in these spectra.In the second paper (Gr¨oningsson et al. 2008, in thefollowing referred to as G08) we discussed a full analysisof the spectrum from one epoch, 2002 October. These r¨oningsson et al.: Line emission from the inner ring of SN 1987A 3 spectra showed narrow emission line components withFWHM ∼ −
30 km s − for most of the lines, arisingfrom the unshocked ring gas. The interaction between theejecta and the ER can be seen as intermediate velocitylines coming from shocked ring material, extending toapproximately ±
300 km s − , with clearly asymmetric lineprofiles. In addition, a broad ( ∼ ,
000 km s − ) velocitycomponent in H α and [Ca II], originating from the reverseshock in the outer ejecta (Heng 2007), is clearly visible.In contrast to the analysis in G08, we focus in this paperon the time evolution of the line emission. Hence, in thispaper, we present data taken with VLT at seven epochsranging from 1999 October to 2007 November.The observations and data reductions are discussed inSect. 2. Analysis of the data is presented in Sect. 3 fol-lowed by discussions of the results in Sect. 4, and finallya summary in Sect. 5.
2. Observations
High resolution spectra of SN 1987A were obtained usingthe Ultraviolet and Visual Echelle Spectrograph (UVES)at ESO/VLT at Paranal, Chile (Dekker et al. 2000).The UVES spectrograph disperses the light beam in twoseparate arms covering different wavelength ranges of thespectrum. The blue arm covering the shorter wavelengths( ∼ − . ′′ / pixel while the redarm ( ∼ −
10 600 ˚A) is covered by a mosaic of twoCCDs having a resolution of 0 . ′′ / pixel. Thus withtwo different dichroic settings the wavelength coverageis ∼ −
10 600 ˚A. Due to the CCD mosaic in thered arm there are, however, gaps at 5770 − − . ′′ . ′′ ◦ (see Fig. 1). Hence, thespectral resolution for these epochs was λ/ ∆ λ ∼
50 000which corresponds to a velocity of 6 km s − . For Epoch 1the slit width was 1 . ′′ ◦ .For the data reduction we made use of the MIDAS im-plementation of the UVES pipeline version 2.0 for Epochs1–3 and version 2.2 for Epochs 4–7. For a detailed de-scription of the steps involved in the reductions we referto G08.For the time evolution of the fluxes an accurateabsolute flux calibration is necessary. As discussed inGr¨oningsson et al. (2006) and G08, we estimated theaccuracy of the flux calibration by comparing the spectraof flux-calibrated spectrophotometric standard stars with their tabulated physical fluxes. In addition, we comparedemission line fluxes from different exposures (and hencewith different atmospheric seeing conditions) within thesame epochs. From these measurements, together withthe results from HST photometry in Appendix A, we con-clude that the accuracy in relative fluxes should be within10 − −
3. Analysis
The systemic velocity of SN 1987A can be estimated di-rectly from the peak velocities of the unshocked gas. Thebest accuracy should in principle be obtained for thestrongest emission lines with well defined intensity peaks,such as the [O II], [S II] and [N II] doublets. However,since there are uncertainties in the rest wavelengths and inthe wavelength calibration (amounting to ∼ .
01 ˚A, cor-responding to 0 . − at 5000 ˚A), we have chosen toinclude all lines in Tables B.1–B.6 in the velocity estimate.Taking the average of the peak velocities of the northern(281 . ± .
07 km s − ) and southern (291 . ± .
07 km s − )parts of the ER in the data set we find a center of mass Gr¨oningsson et al.: Line emission from the inner ring of SN 1987A
Table 1.
VLT/UVES observations of SN 1987A and its rings.
Epoch Date Days after Setting λ range Slit width Resolution Exposure Seeing Airmassexplosion (nm) (arcsec) ( λ/ ∆ λ ) (s) (arcsec)1 1999 Oct 16 4618 346+580 303–388 1.0 40,000 1,200 1.0 1.4476–6842 2000 Dec 9–14 5039–5043 346+580 303–388 0.8 50,000 10,200 0.4–0.8 1.4476–6845038–5039 390+860 326–445 9,360 0.4 1.4–1.6660–10603 2002 Oct 4–7 5704–5705 346+580 303–388 0.8 50,000 10,200 0.7–1.0 1.5–1.6476–6845702–5703 437+860 373–499 9,360 0.4–1.1 1.4–1.5660–10604 2005 Mar 21 6601–6623 346+580 303–388 0.8 50,000 9,200 0.6–0.9 1.5–1.8Apr 8-12 476–6846621–6623 437+860 373–499 4,600 0.5 1.6–1.7660–10605 2005 Oct 20 6826 346+580 303–388 0.8 50,000 2,300 0.9 1.4Nov 1–12 476–6846814–6837 437+860 373–499 9,200 0.5–1.0 1.4–1.6660–10606 2006 Oct 1–29 7160–7180 346+580 303–388 0.8 50,000 9,000 0.5–0.9 1.4–1.5Nov 10–15 476–6847188–7205 437+860 373–499 9,000 0.5–1.0 1.4660–10607 2007 Oct 23–24 7547–7583 346+580 303–388 0.8 50,000 11,250 0.8–1.4 1.4–1.6Nov 27–28 476–6847547 437+860 373–499 9,000 1.1–1.4 1.4–1.6660–1060 velocity of 286 . ± .
05 km s − (1 σ errors). This re-sult is in agreement with Wampler & Richichi (1989) andMeaburn et al. (1995) who obtained 286 . ± . − and 286 . ± − , respectively, but lower than 287 . ± . − and 289 . − as reported by Cumming(1994) and Crotts & Heathcote (1991), respectively.The expansion velocity of the ER is more difficult tomeasure directly from the data. One method to estimatethe expansion velocity would be to assume a luminosityand a temperature distribution of the ER and then modelthe resulting line profiles of either side of the ER by takinginto account the instrumental broadening, spectral resolu-tion and atmospheric seeing (e.g., Meaburn et al. 1995).The best fit model to the narrow emission componentswould then give the expansion velocity. A more direct ap-proach is to measure differences between the peak veloc-ities of the emission lines from the unshocked gas at thenorth and south parts of the ER and correct for the ring inclination and the slit orientation. From geometrical con-siderations we would ideally have for the radial velocity V r = | ∆ V peak | i )cos( ψ ) . (1)The angle ψ is given by tan( ψ ) = tan(PA + φ )cos( i ) where i is the inclination angle of the ring, PA the position angleof the slit, and φ the angle between the minor axis of thering and north. However, due to the slit width and the see-ing, different parts of the ring contribute to the differentcomponents along the line of sight. Since the total broad-ening is a convolution of the macroscopic motion and thethermal motion, the peak of the emission line will be veloc-ity shifted with an amount and a direction that dependson the skewness of the macroscopic velocity distribution,temperature and the atomic weight of the element. Thevelocity distribution of the northern part of the ring is ex-pected to be positively skewed and the distribution at thesouthern part is expected to be negatively skewed. Hence,in this case the measured ∆ V peak would be systematicallyunderestimated and, as a consequence, Eq. (1) only pro-vides a lower limit to the radial velocity. To account for r¨oningsson et al.: Line emission from the inner ring of SN 1987A 5 WFPC2/F656NEpoch 2
ACS/F658NEpoch 3
ACS/F658NEpoch 4
ACS/F658NEpoch 6
Fig. 1.
Ring images taken with HST at four different epochs (obtained by the SAINTS team; PI: R.P. Kirshner). Theslit position of the VLT/UVES observations is superimposed (the slit width is 0 . ′′ ◦ ). North is up andeast is to the left. All images except the earliest epoch (upper left panel) have the same scaling in flux. The epochslabeled in the panels correspond roughly to the time of the HST observations. Note the evolution and distribution of“hot spots” around the ring. The evolution is most prominent on the southwest part of the ring.the fact that the emission lines sample a range of ψ ’s, Eq.(1) can be multiplied by the factor (∆ ψ/ / sin(∆ ψ/ ψ to be in the range ∼ ◦ − ◦ . Hence, it fol-lows that Eq. (1) would underestimate the real expansionvelocity by ∼ − λ , λ λ λ λ V peak . Ifwe adopt the geometrical values of the ER suggested bySugerman et al. (2002, 2005), i.e., i = 43 ◦ and φ = 9 . ◦ ,we derive for PA = 30 ◦ the radial expansion velocity V r = 10 . ± . − (1 σ error). This result agreewith earlier estimates by Cumming (1994) who found10 . ± . − and Crotts & Heathcote (2000) whoreported 10 . ± . − . Gr¨oningsson et al.: Line emission from the inner ring of SN 1987A
As discussed in detail in G08 (see Table 2 in that paper),we detect narrow components for a large range of ioniza-tion stages, from neutral up to Ne V and Fe VII. The onlyexceptions are the coronal lines from Fe X-XIV.Figure 2 shows the flux evolution of the narrow linesfor the northern and southern parts of the ER (see TablesB.1–B.6). In order to compare the trends of the differentlines more clearly we have normalized these to the flux onOctober 2002 (Epoch 3). Because of the low S/N, we donot include Epoch 1 in the analysis.The Balmer line fluxes appear to be constant in timeup to day ∼ ∼ λ ∼ − ), the bulk motion of the ER and the temper-ature of the unshocked gas. The instrumental broadeningcan be considered constant from epoch to epoch. The ex-pansion velocity of the ring should also be fairly constant.However, due to the evolution of the spatial ring profile,different regions of the ring may be sampled in the extrac-tion process causing slight changes in the macroscopic ve-locity distribution from epoch to epoch. This effect wouldin such case broaden the line and become more importantfor later epochs. Nevertheless, for the light elements, suchas hydrogen and helium, any differences should be smallrelative to the thermal broadening, and the FWHM fromepoch to epoch should mainly reflect changes in tempera-tures for these lines.As shown in Fig. 3, the widths are fairly constant formost of the lines. Only the Balmer lines show a tendencyof increasing line widths, which may indicate an increasein temperature of the emitting gas. Note, however, thatlines with higher ionization stages are slightly wider, likelydue to higher temperature in the emitting gas (see G08).Figure 4 shows the line ratios for a number of diag-nostic lines for the narrow component. As can be seen,the flux ratio of [N II] j λλ , /j λ decreases withtime, which implies an increase in temperature. For [S II]and [O I] we see roughly the same behavior as for [N II].However, there is no indication of an increase of the linewidths. Instead, the widths appear rather constant (Fig.3). The [O II] j λλ − /j λλ , ratio, on the otherhand, is rather constant except for a lower value at Epoch N o r m a li z ed f l u x Days after explosion N o r m a li z ed f l u x Days after explosion H α H β H γ [O I] λ λ λ λ λ λ λ N o r m a li z ed f l u x H α H β H γ [O I] λ λ λ Days after explosion N o r m a li z ed f l u x [Ne III] λ λ Days after explosion [O II] λ λ Fig. 2.
Upper panels: Fluxes for the unshocked, narrowcomponent from the northern part of the ring, normalizedto the flux at Oct. 2002 (day 5703). Lower panels: Fluxesfor the southern part of the ring.2 than at later epochs. By contrast, the [O III] ratio j λλ , /j λ increases with time. In general, thereare only modest differences between the flux ratios at eachepoch for the northern and southern sides. We return toa discussion of the narrow lines in Section 4.1. The shocked gas is dominated by emission from the hotspots. A complete list of lines from the shocked compo-nent is given in Table 2 in G08. The fluxes of a selec-tion of representative lines from the intermediate velocitycomponent are shown in Fig. 5 and Fig. 6 for the north-ern and southern components, respectively. As with thenarrow lines, we have normalized the fluxes, but in thiscase to Epoch 4 (2005 April) to facilitate a comparisonof the different lines more directly. The absolute valuesof the fluxes can be found in Tables B.7 and B.8 in the r¨oningsson et al.: Line emission from the inner ring of SN 1987A 7 F W H M ( k m s − ) H α H β H γ H δ [O I] λ λ λ Days after explosion F W H M ( k m s − ) [Ne III] λ λ Days after explosion [O II] λ λ Fig. 3.
FWHM of the unshocked components as a functionof time for the northern part of the ring. F l u x r a t i o [O I] NorthSouth
NorthSouth F l u x r a t i o [S II] NorthSouth
NorthSouth F l u x r a t i o [O II]Days after explosion NorthSouth
Fig. 4.
Diagnostic line ratios from the unshocked north-ern and southern components. The ratios shown are [O I]( λ λ /λ λ λ /λ λ λ / ( λ λ λλ − / ( λ λ λ λ /λ N o r m a li z ed f l u x N o r m a li z ed f l u x N o r m a li z ed f l u x H α H β H γ H δ He I λ λ λ λ λ λ λ λ λ λ λ λ λ Fig. 5.
Fluxes for the shocked component from the north-ern part of the ring for selected lines, normalized to theflux at Apr. 2005 (day 6618). The lower right panel showsthe soft X-ray flux normalized to the same date (fromPark et al. 2007).ture for all lines. The rate of increase is, however, differentfor the different lines (Fig. 5). The low ionization lines ofHe I, [N II], [O I], [S II] and [Fe II] increase similar to eachother and to the Balmer lines. This is also the case for the[Fe XIV] line. The intermediate ionization lines, [O III],[Ne III], [Ne V], [Fe III] and [Fe VII], however, all show abreak between day 6500 and day 7000. This is probablyalso the case for the [Fe X] and [Fe XI] lines, althoughthis result is close to the systematic uncertainties. Thesame pattern is seen for the southern component (Fig. 6).The Balmer decrement, as seen in H α , H β , H γ , and H δ ,is constant within the systematic errors.The flux ratios for the shocked components, plottedin Fig. 7, are useful diagnostics for the shocked gas, be-ing sensitive to both electron density and temperature. InG08 we found that the densities are consistent with thecompressions from radiative shocks. The electron densityin the [O III] region was constrained to be in the range ∼ − cm − and with a temperature in the interval ∼ (1 − × K for both the northern and southern partsof the ring (see also Pun et al. 2002).For [S II] we find a density of n e > ∼ cm − anda temperature of the order of T e ∼ T e ∼ Gr¨oningsson et al.: Line emission from the inner ring of SN 1987A N o r m a li z ed f l u x H α H β H γ H δ [He I] λ λ λ λ N o r m a li z ed f l u x [Ne III] λ λ λ [Fe II] λ λ λ N o r m a li z ed f l u x [Fe X] λ λ λ X−ray (0.5−2 keV)
Fig. 6.
Same as Fig. 5, but for the southern part of thering. Note the change of flux scale compared to Fig. 5.line ratios in Fig. 7, consistent with an increasing temper-ature for [S II], [N II] and [O I].The [O III] flux ratio j λλ , /j λ is temperaturesensitive for the considered density interval. The data sug-gest that the emission comes from a region with an elec-tron density of a few times 10 cm − and a temperaturearound 15 , − ,
000 K. It is difficult to find a cleartrend for the evolution of the flux ratio, perhaps becausethe line [O III] λ λ Figure 8 illustrates the evolution of the H α , [N II] λ λ λ α from the reverse shock to determine the profiles of theemission from the shocked gas for H α and [N II] λ F l u x r a t i o [O I] NorthSouth
NorthSouth F l u x r a t i o [S II] NorthSouth
NorthSouth F l u x r a t i o [N I]Days after explosion NorthSouth
Fig. 7.
Diagnostic line ratios from the shocked northernand southern components. The ratios shown are [O I]( λ λ /λ λ λ /λ λ λ / ( λ λ λ λ /λ λ λ /λ λ λ α ,with the high ionization coronal lines, like [Fe XIV] (inthe same way as in G08). The latter has a considerablynoisier profile due to the low flux (only < ∼ α )and disappears in the noise at levels less than ∼
5% of thepeak emission. We discuss the velocity at lower levels inSect. 4.2.The peak velocity in Fig. 9 only marginally evolveswith time in both the southern and northern regions.There are, however, noticeable differences between the ve-locity values for the different lines. The low ionizationlines, H α , [O I] and [N II], have all similar velocities, whilethe [O III] and [Fe XIV] lines have lower and higher peakvelocities, respectively, showing that contribution to thedifferent lines probably come from different regions cov-ered by the slit.In addition, there are clear differences in peak veloci-ties of the lines at the northern part of the ring comparedto those in the southern part (Fig. 9). The velocities are r¨oningsson et al.: Line emission from the inner ring of SN 1987A 9 −500 −250 0 250 50000.20.40.60.81 N o r m a li z ed f l u x H α −500 −250 0 250 50000.20.40.60.81 [N II] λ −500 −250 0 250 50000.20.40.60.81 Velocity (km s −1 ) N o r m a li z ed f l u x [O III] λ −500 −250 0 250 50000.20.40.60.81 Velocity (km s −1 ) [Fe XIV] λ −500 −400 −300 0.01 −450 −350 −250 0.05−450 −350 −250 0.1 −500 −400 0.2 Fig. 8.
Evolution of the H α , [N II] λ λ λ λ −
150 km s − in the dif-ferent bands, and only gives an average, bulk velocity ateach point on the ring. As found in G08, there is goodagreement at the epoch of the first SINFONI observationin November 2004.Turning now to the maximum velocity in Fig. 9 wesee an interesting difference between the lines, as well asan evolution in time. While the low and intermediate ion-ization lines, H α , [O I], [N II] and [O III], show similarvelocities, 250 −
300 km s − , the coronal lines have con-siderably higher velocities, ∼
350 km s − for [Fe X] and ∼
450 km s − for [Fe XIV]. Again, this shows that the twosets of lines come from regions of quite different physicalconditions. We will discuss this in more detail in Sect. 4.2.There is an indication that the coronal line profiles, suchas [Fe XI-XIV], become less blueshifted over time for thenorthern part of the ring. The maximum velocity (FWZI)does, however, not decrease with time, as seen in Fig. 8.
4. Discussion
In this section we discuss the implications of the resultsin Sect. 3. The general picture we here have in mind forthe optical emission is that of a radiative shock, movinginto the dense ER, with a pre-shock density of 3 × − × cm − and a temperature of ∼ (1 − × K (seebelow). The narrow line emission comes from this gas.The soft X-rays, as well as the coronal lines, then comefrom the immediate post-shock gas, with a temperature
North H α [N II] λ λ λ λ λ Days after explosion V e l o c i t y a t pea k l e v e l ( k m s − ) South H α [N II] λ λ λ V e l o c i t y a t % l e v e l ( k m s − ) H α [N II] λ λ λ λ λ Fig. 9.
Upper panels: Velocities of the peak of the linesfor the shocked component. Lower panel: Velocities at 5%flux level for the shocked component from the northernpart of the ring.of ∼ (1 − × K and a density four times that ofthe pre-shock density. As the gas cools to ∼ K, thedensity increases by a similar factor to 10 − cm − .This is where most of the optical, intermediate velocitylines arise. We now discuss these regions, one by one. In connection with the shock breakout the rings wereionized by the soft X-rays during the first hour(Lundqvist & Fransson 1996). After this they have re-combined and cooled at a rate determined mainly by thedensity of the different parts of the rings. However, be-cause of the strong X-ray flux from the shocks propagat-ing into the dense blobs one expects re-heating and re-ionization of this gas, as well as gas in the inner ring andsurrounding gas that was never ionized by the supernova flash. The decrease in the line fluxes after the flash ion-ization is therefore expected to turn into an increase, orat least a slower decay, when preionization by the shocksstarts to dominate the conditions in the ring. The evo-lution of the narrow components of the different lines istherefore of great interest.As shown in Fig. 2, we do indeed find that the evolutionof the fluxes of the narrow lines is different for differentlines. The [O I], [O II], [N II], [S II] and [Fe II] lines alldecrease with time (Fig. 2). These lines are all collisionallyexcited and therefore sensitive to both temperature anddegree of ionization. The hydrogen lines arise mainly byrecombination and therefore decay slower than the colli-sionally excited lines, as is observed.The most interesting result is, however, the observed increase of the [O III] and [Ne III] lines up to day ∼ − × K immediately in front of theshock (Allen et al. 2008). The absorbed X-rays are therere-emitted as UV and optical lines. The interesting pointis now that these shock calculations show that the [O III] λλ , n e > ∼ cm − ) hasbeen ionized by the X-rays, which will result in a nearlyconstant flux after this epoch. It could, of course, also bethe result of a decreasing X-ray flux from the shocks, orfinally from sweeping-up of the dense gas by the shocks.Similarly, the fast decrease of the low ionization linesmay either be a result of ionization or sweeping-up of theunshocked gas. Future HST imaging may distinguish be-tween these scenarios.A more detailed picture of the physical conditions ofthe narrow-line emitting gas can be obtained from lineratios (see Fig. 4). The declining diagnostic ratios of [N II],[O I] and [S II] all suggest an increasing temperature inthis zone, which may be indicative of pre-ionization.For the [O II] and [O III] lines we see a rather dra-matic change between Epochs 2 and 3 compared to atlater epochs. The low ratios for [O III] j λλ , /j λ at Epoch 2 indicate much higher temperatures in boththe northern and southern parts of the ring than at laterepochs. Using nebular analysis as in G08, the [O III]temperature appears to decline from ∼ × K to ∼ . × K between Epoch 2 and 3, and further down to ∼ × K at later epochs. The Epoch 2 temperature isclearly much higher than expected from a shock precursor(cf. above, see also Allen et al. 2008). r¨oningsson et al.: Line emission from the inner ring of SN 1987A 11
The most likely reason for this is that the narrow-line emission at the earliest epochs comes mainly fromlow-density gas that was flash-ionized by the supernova.Mattila et al. (2003, and paper in preparation) show thatthe density of the flash-ionized gas dominating the [O III]emission at epochs around 5000 days is ∼ × cm − .The [O III] temperature in these models is ∼ × K,i.e., close to what we observe. From an inspection of theHST images (see Fig. 1) it is also clear that by day ∼ α +[N II] emission shown in Fig. 1. Note alsothat only a fraction of the emission from the blobs seenin the HST images stems from narrow-line emitting shockprecursors, as the HST images cannot distinguish betweenthat emission and the line emission from the shocked gas.At Epoch 2, we estimate that < ∼
30% of thenarrow-line [O III] λλ , λ < ∼
15% to obtain the observed j λλ , /j λ ratio. However, already by Epoch 3, thenarrow [O III] λλ , λ ∼ × cm − component which would then still have a temperatureof ∼ (4 − × K, according to the models dis-cussed in Mattila et al. (2003). The higher temperaturefor [O III] λ λλ , ∼ . ∼ .
3. However, after Epoch 3, when the precursor emis-sion is likely to dominate [O III], the increase of the narrowcomponent is similar to the intermediate velocity flux.The extent of the pre-ionized zone depends on thenumber of ionizing photons from the cooling shock, whichshould scale as the total flux ρV s . The number of recom-binations per unit area is α B n e ∆ x (HII), where ∆ x (HII)is the thickness of the ionized region and α B is the CaseB recombination rate. Therefore, ∆ x (HII) ∝ V s /α B n e .This describes well the shock calculations by Allen et al.(2008), which gives the normalization, and we find∆ x (HII) ≈ . × (cid:18) V s
200 km s − (cid:19) (cid:16) n e cm − (cid:17) − cm . (2)The pre-ionized region is therefore sensitive to boththe shock velocity, which probably spans a range 100 −
400 km s − , as well as the density. It is clear that a largefraction of the ring may be pre-ionized by the shocks, aswell as material outside of the ER, which is likely to havea lower density. From Fig. 5 and Fig. 6 we found that the lines from theshocked gas increase by a factor of ∼ α , [N II], [O I] and [O III], now at a 1 % level ofthe peak flux. We here see the same increase in maximumvelocity with time as in Fig. 9, although the velocities areconsiderably higher. From Fig. 8 we see that H α can infact be traced to ∼
500 km s − in the last spectrum. Thesame is true for the southern part of the ring.As can be seen from Figs. 9 and 10, the velocities fromthe southern part of the ring are consistently lower by ∼
40 km s − , but do also show a similar increase withtime as those from the northern part. The lower velocities North H α [N II] λ λ λ Days after explosion V e l o c i t y a t % f l u x l e v e l ( k m s − ) South H α [N II] λ λ λ Fig. 10.
Velocities at 1% flux level for the shocked gasfrom the northern and southern components.may either be a projection effect, an effect of the later im-pact in this region, or a higher density. Even though thereis clearly a time delay between the northern and southernparts of the ring ( ∼
300 days), the most plausible explana-tion is a difference in the projection of velocities betweenthe two regions. The flux from the northern part mainlycomes from “hot spots” between ∼ ◦ − ◦ while spotsbetween ∼ ◦ − ◦ dominate the flux from the south-ern part (see Fig. 1). This asymmetric distribution of “hotspots” between the northern and southern parts of the ringcould well account for a difference in projection such thatthe line-of-sight velocities from the southern part are con-sistently lower than the velocities from the northern part,consistent with the observed line profiles. This is also sup-ported by the similar difference in peak velocity, which is40 −
50 km s − (see also Kjær et al. 2007).For a shock to become radiative it must cool from itspostshock temperature to ∼ K from the time it wasshocked. From shock calculations we found that the timeit takes for a shock to become radiative, i.e., the coolingtime, t cool , is related to the preshock density and the shockvelocity as t cool ≈ . (cid:16) n spot cm − (cid:17) − (cid:18) V s
300 km s − (cid:19) . years (3)(Gr¨oningsson et al. 2006), where n spot is the pre-shockdensity of the ring.At Epoch 3 (October 2002) the shocks had ∼ < ∼
300 km s − have had enoughtime to cool. This places a lower limit on the preshock den-sity from Eq. (3) and we estimate n spot > ∼ . × cm − .At our last epoch the maximum velocity of H α is ∼
500 km s − . With this velocity Eq. (3) implies a densityof n spot > ∼ . × cm − . These densities can be compared to the densities de-rived from the analysis of the ring during the first fewyears after outbreak. Lundqvist & Fransson (1996) findthat the light curves of the ring emission require a rangein densities from 6 × cm − to 3 . × cm − . Further,Mattila et al. (2003, and paper in preparation) show thatlater observations of the narrow optical lines require den-sities down to ∼ cm − . This density agrees with thelimit on the ionization time n e t i > ∼ cm − s foundfrom the soft X–rays by Park et al. (2007). Using this,and the assumed impact times between 4000 − . − . × cm − , which should be seen as a lowerlimit to the density of the pre-shock gas. It is likely thatwe now see the densest blobs cooling down first, and theagreement with the densities determined for the ring be-fore the collision makes this scenario likely. As time goes,gas shocked by higher velocity shocks and/or lower densi-ties will gradually cool down.In G08 we showed that the line width of the low ion-ization lines were considerably smaller than the coronallines in 2002 (see also Fig. 9). A likely interpretation isthat part, or all, of the gas emitting the coronal lines atthat time had not yet had time to cool and be seen as op-tical low ionization emission lines. The fact that the H α velocities from the northern part are now comparable tothose of the coronal lines is consistent with the estimatesof the cooling time above. In Fig. 5 we see that all linesincreased in flux up to day ∼ r¨oningsson et al.: Line emission from the inner ring of SN 1987A 13 [Fe XIV] and other coronal lines are special in thatthey may come from both radiative and adiabatic shocks,as long as the temperature behind the shock is < ∼ × K. At least part of their emission may therefore befrom non-radiative shocks, which is consistent with themore extended line profiles, especially at early epochs.In a recent paper Dewey et al. (2008) find that (adia-batic) shock model fits of the X-ray line fluxes indicate abimodial distribution of temperature. The dominant com-ponent has a temperature of ∼ .
55 keV, nearly constantbetween 2004 and 2007. The hard component, on the otherhand, decreases from ∼ . ∼ . −
450 km s − . This iscomparable to our optical velocities and strengthens theidentification of the soft X-ray component with the op-tical lines, as was already indicated from the flux evolu-tion. The harder X-ray component does, however, not haveany corresponding optical component. Further, a temper-ature of ∼ . ∼ − . Such a component should be adiabaticfor a very long time. From the fact that such a high ve-locity is not seen in the line widths, Dewey et al. suggestthat the high velocity component may originate from areflected shock, moving back into the previously shockedgas.An interesting additional fact is that the Chandraimaging shows an expansion of the X-ray emitting plasmacorresponding to a velocity of 1400 − − (e.g.,Park et al. 2007). Also the extent of the optical emissionincreases with a similar velocity (P. Challis priv.comm.).In ∼
10 years the emitting region has therefore expandedby ∼ × cm, which is a considerable fraction of thethickness of the visible ring. As Dewey et al. (2008) pointout, this expansion is, however, likely to take place in alow density component, and not in the dense clumps. Weexpect this expansion to continue as long as there is lowdensity gas in the vicinity of the clumps.
5. Summary
We have presented high spectral resolution UVES obser-vations at multiple epochs of the inner circumstellar ringof SN 1987A. Due to the high spectral resolution, we areable to separate the shocked from the unshocked ring emis-sion, as well as to study the evolution of the fluxes andline profiles of the different lines.For the narrow lines we find strong evidence for pre-ionization of the unshocked ring by the soft X-rays fromcooling gas behind the shocks. This is indicated both bythe increasing [O III] flux and from the temperature in-ferred from the low ionization lines [N II], [O I] and [S II].The increasing [O III] line ratios at the early epochs (untilday ∼ ∼ × cm − ) ionized bythe SN flash in connection with the explosion. Later on,however, the emission is dominated by gas pre-ionized bythe X-rays. For the intermediate velocity low ionization lines, wefind evidence for increasing line widths for the shockedcomponents. This is to be expected as faster shocks willhave enough time to become radiative and cool. At the lat-est epoch the H α line profile extends to ∼
500 km s − forboth the northern and southern part of the ring, which iscomparable to the maximum velocity of the coronal lines.The ring collision in SN 1987A is therefore one of thebest examples of high velocity radiative shocks we have.Together with the X-ray observations these observationstherefore offer a unique opportunity to study this class ofshocks in detail.The difference in shapes for the line profiles betweenthe northern and southern components cannot solely beexplained by a time delay effect, but is most likely ex-plained by a difference of projection of the shock velocitiescaused by the locations of “hot spots” around the ring.We find a correlation of our optical light curves to thatof the soft X-rays. This suggests that most of the opticallines and soft X-rays arise from the same region. The dif-ference in line widths of the coronal lines and the low ion-ization lines at early epochs indicates that at least some ofthe coronal emission comes from adiabatic shocks. Shortof day ∼ Acknowledgements.
We are grateful to the referee, PatriceBouchet, for his constructive comments, which have improvedthe paper and to the observers as well as the staff at Paranalfor performing the observations at ESO/VLT. This work hasbeen supported by grants from the Swedish Research Council,the Swedish National Space Board and STScI for the SAINTSproject (GO-11181).
References
Allen, M., Groves, B., Dopita, M., Sutherland, R., &Kewley, L. 2008, astro-ph/0805.0204Bouchet, P., Dwek, E., Danziger, J., et al. 2006, ApJ, 650,212Burrows, C. J., Krist, J., Hester, J. J., et al. 1995, ApJ,452, 680Chevalier, R. A. 1982, ApJ, 258, 790Chevalier, R. A. & Dwarkadas, V. V. 1995, ApJ, 452, L45Crotts, A. P. S., & Heathcote, S. R. 2000, ApJ, 528, 426Crotts, A. P. S., & Heathcote, S. R. 1991, Nature, 350,683Crotts, A. P. S., Kunkel, W. E., & McCarthy, P. J. 1989,ApJ, 347, 61
Cumming, R. J. 1994, Ph.D. thesis, Imperial College,LondonDekker, H. et al. 2000, SPIE, 4008, 534.Dewey, D., Zhekov, S. A., McCray, R., & Canizares, C. R.2008, ApJ, 676, 131Dwek, E., Arendt, R. G., Bouchet, P., et al. 2008, ApJ,676, 1029Eriguchi, Y., Yamaoka, H., Nomoto, K., & Hashimoto, M.1992, ApJ, 392, 243Fransson, C., Cassatella, A., Gilmozzi, R. Kirshner, R. P.,Panagia, N., Sonneborn, G., & Wamsteker, W. 1989,ApJ, 336, 429Gaensler, B. M., Staveley-Smith, L., Manchester, R. N.,Kesteven, M. J., Ball, L., & Tzioumis, A. K. 2007, AIPConf., 937, 86Gouiffes, C., et al., 1988, A&A, 198, 9Graves, Genevieve J. M., et al., 2005, ApJ, 629, 944Gr¨oningsson, P., Fransson, C., Lundqvist, P., Lundqvist,N., Leibundgut, B., Spyromilio, J., Chevalier, R.,Gilmozzi, R., Kjær, K., Mattila, S., & Sollerman, J.2008, A&A, 479, 761Gr¨oningsson, P., Fransson, C., Lundqvist, P., Nymark,T., Lundqvist, N., Chevalier, R., Leibundgut, B., &Spyromilio, J. 2006, A&A, 456, 581Heng, K. 2007, AIP Conf., 937, 51Kjær, K., Leibundgut, B., Fransson, C., Gr¨oningsson, P.,Spyromilio, J., & Kissler-Patig, M. 2007, A&A, 471, 617Lawrence, S. S., Sugerman, B. E., Bouchet, P., Crotts,A. P. S., Uglesich, R., & Heathcote, S. 2000, ApJ, 537,L123Lundqvist, P. & Fransson, C. 1996, ApJ, 464, 924Mattila, S., Lundqvist, P., Meikle, P., Stathakis, R., &Cannon, R. 2003, astro-ph/0308533McCray, R. 2007, AIP Conf., 937, 3Meaburn, J., Bryce, M., & Holloway, A. J. 1995, A&A,299, 1Michael, E., et al. 2002, ApJ, 574, 166Morris, T., & Podsiadlowski, P. 2007, Science, 315, 1103Park, S., Burrows, D. N., Garmire, G. P., McCray, R.,Racusin, J. L., & Zhekov, S. A. 2007, AIP Conf., 937,43Pun, C.S.J. 2007, AIP Conf., 937, 171Pun, C. S. J., Michael, E., Zhekov, S. A., et al. 2002, ApJ,572, 906Pun, C. S. J., Sonneborn, G., Bowers, C., et al. 1997, IAUCirc., 6665, 1Sonneborn, G., Pun, C. S. J., Kimble, R. A., et al., 1998,ApJ, 492, 139Sugerman, B. E. K., Crotts, A. P. S., Kunkel, W. E.,Heathcote, S. R., & Lawrence, S. S. 2005, ApJ, 627,888Sugerman, B. E. K., Lawrence, S. S., Crotts, A. P. S.,Bouchet, P., & Heathcote, S. R. 2002, ApJ, 572, 209Wampler, E. J., & Richichi, A. 1989, A&A, 217, 31Wampler, J., Schwarz, H. E., Andersen, J., & Beresford,A. C. 1988, IAUC, 4541, 2Wampler, E. J., Wang, L., Baade, D., et al. 1990, ApJ,362, 13 Woosley, S. E., Heger, A., Weaver, T. A., & Langer, N.1997, astro-ph/9705146Zhekov, S. A., McCray, R., Borkowski, K. J., Burrows,D. N., & Park, S. 2005, ApJ, 628, 127Zhekov, S. A., McCray, R., Borkowski, K. J., Burrows,D. N., & Park, S. 2006, ApJ, 645, 293 r¨oningsson et al.: Line emission from the inner ring of SN 1987A 15
Appendix A: Comparison of fluxes from HST andUVES.
Pipeline-processed HST imaging have been obtained bythe SAINTS team (PI: R.P. Kirshner) at similar epochsas for the UVES spectra. The instruments were WFPC2with narrow-band filters F502N and F656N at day 5013and ACS with filters F502N and F658N (at days 5795-7229).In order to compare the HST fluxes with thefluxes of UVES we first converted the counts per elec-trons in the HST images to flux density units (givenin erg s − cm − ˚A − ). Thus, the images were multipliedby the PHOTFLAM header keyword and, in the case ofWFPC2 images, also divided by the exposure time. Thenext step was to rotate the images to 30 ◦ in accordancewith the slit rotation for the UVES spectra. To accountfor the moderate seeing of the ground-based UVES ex-posures (typically ∼ . ′′
8) we artificially added this seeingto the HST images by convolving each image with a two-dimensional Gaussian kernel.All the pixel columns in the images within 0 . ′′ ∼ ∼ ∼ ∼ −
499 nm) and two in the red part( ∼ −
577 nm and ∼ −
684 nm respectively forthe standard setting 346+580). Hence, in this analysis weare only probing the fluxes for two out of the three CCDsand the uncertainty in the relative fluxes can thereforepossibly be larger for emission lines situated in the bluepart of the spectrum. The relative fluxes over the wholespectral range have, however, been checked against UVESspectrophotometric standard star spectra (see Sect. 2).Moreover, to check the robustness of these results wevaried the artificially imposed seeing on the HST imagesbetween 0 . ′′ . ′′
0. This showed that the total flux cov-ered by the slit is fairly insensitive for this seeing range,and differs by less than a few percent. The redistributioneffect of the fluxes from different parts of the ring is, ofcourse, still important as the seeing is varied. The net ef-fect here is that the flux on the northern part of the ringtends to decrease with increasing seeing, while the oppo-site effect holds for the southern part. This is explainedby the fact that the “hot spots” are distributed unevenlyaround the ring (see Fig. 1) and this effect will thereforevary with time. We find that for some epochs, the individ-ual fluxes from north and south respectively could shift byan amount of ∼
10% due to this effect.In addition, from the HST images we can estimate therelation between the ring flux encapsulated by the slit andthe total flux from the ring. For a simulated seeing of 0 . ′′ . ′′ . Days since SN F l u x ( × − e r g s − c m − A ng − ) HST/WFPC2 (F656N)HST/ACS (F658N)UVES/WFPC2 (F656N)UVES/ACS (F658N)
Fig. A.1.
HST narrow-band filter fluxes for WFPC2 andACS together with the corresponding UVES bandpassfluxes for the F656N and F658N filters for the northernpart (solid lines) and the southern part (dashed lines) ofthe ring.
Days since SN F l u x ( × − e r g s − c m − A ng − ) HST/WFPC2 (F502N)HST/ACS (F502N)UVES/WFPC2 (F502N)UVES/ACS (F502N)
Fig. A.2.
The same as Fig. A.1 but for WFPC2 and ACSnarrow-band filters F502N. r¨oningsson et al.: Line emission from the inner ring of SN 1987A 17
Appendix B: Line fluxes
Table B.1.
Emission lines from the unshocked gas of the ER at Epoch 2.
North SouthRest wavel. V cpeak V FWHM V peak V FWHM
ExtinctionEmission ˚A Relative flux a km s − km s − Relative flux a km s − km s − correction b [Ne V] 3425.86 19 . ± . . ± . . ± .
72 17 . ± . . ± . . ± .
31 2.06[O II] 3726.03 162 . ± . . ± . . ± .
10 152 . ± . . ± . . ± .
16 1.98[O II] 3728.82 97 . ± . . ± . . ± .
22 101 . ± . . ± . . ± .
17 1.98[Ne III] 3868.75 39 . ± . . ± . . ± .
34 29 . ± . . ± . . ± .
58 1.94[S II] 4068.60 23 . ± . . ± . . ± .
19 22 . ± . . ± . . ± .
29 1.90[S II] 4076.35 7 . ± . . ± . . ± .
55 8 . ± . . ± . . ± .
59 1.90H δ . ± . . ± . . ± .
45 26 . ± . . ± . . ± .
48 1.89H γ . ± . . ± . . ± .
28 48 . ± . . ± . . ± .
25 1.84[O III] 4363.21 11 . ± . . ± . . ± .
71 8 . ± . . ± . . ± .
19 1.84He II 4685.7 - - - - - - 1.75H β . . ± . . ± .
12 105 . ± . . ± . . ± .
07 1.71[O III] 4958.91 58 . ± . . ± . . ± .
17 41 . ± . . ± . . ± .
25 1.68[O III] 5006.84 176 . ± . . ± . . ± .
09 124 . ± . . ± . . ± .
12 1.67[O I] 5577.34 0 . ± . . ± . . ± .
17 0 . ± . . ± . . ± .
38 1.57[N II] 5754.59 21 . ± . . ± . . ± .
08 20 . ± . . ± . . ± .
16 1.54He I 5875.63 20 . ± . . ± . . ± .
14 19 . ± . . ± . . ± .
18 1.53[O I] 6300.30 52 . ± . . ± . . ± .
07 52 . ± . . ± . . ± .
08 1.47[S III] 6312.06 1 . ± . . ± . . ± .
13 1 . ± . . ± . . ± .
59 1.47[O I] 6363.78 18 . ± . . ± . . ± .
05 18 . ± . . ± . . ± .
09 1.47[N II] 6548.05 394 . ± . . ± . . ± .
04 397 . ± . . ± . . ± .
04 1.45H α . ± . . ± . . ± .
04 380 . ± . . ± . . ± .
04 1.45[N II] 6583.45 1206 . ± . . ± . . ± .
04 1216 . ± . . ± . . ± .
03 1.45He I 6678.15 5 . ± . . ± . . ± .
24 5 . ± . . ± . . ± .
44 1.44[S II] 6716.44 72 . ± . . ± . . ± .
07 76 . ± . . ± . . ± .
06 1.43[S II] 6730.82 121 . ± . . ± . . ± .
05 128 . ± . . ± . . ± .
05 1.43[Ar III] 7135.79 6 . ± . . ± . . ± .
26 4 . ± . . ± . . ± .
62 1.39[Fe II] 7155.16 5 . ± . . ± . . ± .
11 5 . ± . . ± . . ± .
23 1.39[O II] 7319 d . ± . . ± . . ± . . ± . . ± .
08 11 . ± . . ± . . ± .
15 1.37[O II] 7330 e . ± . . ± . . ± . . ± .
24 8 . ± . . ± .
25 1.24[S III] 9530.6+31.1 30 . ± . . ± .
24 18 . ± . . ± .
22 1.22 a Fluxes are relative to H β : “100” corresponds to (29 . ± . × − erg s − cm − . b E ( B − V ) = 0 .
16, with E ( B − V ) = 0 .
10 for LMC and E ( B − V ) = 0 .
06 for the Milky Way. c The recession velocity of the peak flux. d Applies to both lines at 7318 .
92 ˚A and 7319 .
99 ˚A. e Applies to both lines at 7329 .
67 ˚A and 7330 .
73 ˚A.r¨oningsson et al.: Line emission from the inner ring of SN 1987A 19
Table B.2.
Emission lines from the unshocked gas of the ER at Epoch 3.
North SouthRest wavel. V cpeak V FWHM V peak V FWHM
ExtinctionEmission ˚A Relative flux a km s − km s − Relative flux a km s − km s − correction b [Ne V] 3425.86 17 . ± . . ± . . ± .
58 13 . ± . . ± . . ± .
57 2.06[O II] 3726.03 109 . ± . . ± . . ± .
13 101 . ± . . ± . . ± .
23 1.98[O II] 3728.82 66 . ± . . ± . . ± .
20 63 . ± . . ± . . ± .
37 1.98[Ne III] 3868.75 44 . ± . . ± . . ± .
18 24 . ± . . ± . . ± .
28 1.94[S II] 4068.60 17 . ± . . ± . . ± .
15 15 . ± . . ± . . ± .
22 1.90[S II] 4076.35 5 . ± . . ± . . ± .
30 5 . ± . . ± . . ± .
44 1.90H δ . ± . . ± . . ± .
36 15 . ± . . ± . . ± .
29 1.89H γ . ± . . ± . . ± .
26 39 . ± . . ± . . ± .
24 1.84[O III] 4363.21 10 . ± . . ± . . ± .
31 6 . ± . . ± . . ± .
64 1.84He II 4685.7 10 . ± . . ± . . ± .
27 7 . ± . . ± . . ± .
41 1.75H β . . ± . . ± .
11 97 . ± . . ± . . ± .
16 1.71[O III] 4958.91 81 . ± . . ± . . ± .
11 50 . ± . . ± . . ± .
14 1.68[O III] 5006.84 247 . ± . . ± . . ± .
06 154 . ± . . ± . . ± .
10 1.67[O I] 5577.34 0 . ± . . ± . . ± .
17 0 . ± . . ± . . ± .
23 1.57[N II] 5754.59 20 . ± . . ± . . ± .
09 19 . ± . . ± . . ± .
16 1.54He I 5875.63 17 . ± . . ± . . ± .
15 16 . ± . . ± . . ± .
16 1.53[O I] 6300.30 48 . ± . . ± . . ± .
12 47 . ± . . ± . . ± .
16 1.47[S III] 6312.06 2 . ± . . ± . . ± .
47 1 . ± . . ± . . ± .
85 1.47[O I] 6363.78 16 . ± . . ± . . ± .
09 16 . ± . . ± . . ± .
12 1.47[N II] 6548.05 359 . ± . . ± . . ± .
05 352 . ± . . ± . . ± .
05 1.45H α . ± . . ± . . ± .
07 369 . ± . . ± . . ± .
05 1.45[N II] 6583.45 1112 . ± . . ± . . ± .
03 1096 . ± . . ± . . ± .
04 1.45He I 6678.15 5 . ± . . ± . . ± .
35 4 . ± . . ± . . ± .
32 1.44[S II] 6716.44 62 . ± . . ± . . ± .
06 69 . ± . . ± . . ± .
07 1.43[S II] 6730.82 103 . ± . . ± . . ± .
06 115 . ± . . ± . . ± .
08 1.43[Ar III] 7135.79 8 . ± . . ± . . ± .
15 5 . ± . . ± . . ± .
29 1.39[Fe II] 7155.16 4 . ± . . ± . . ± .
13 4 . ± . . ± . . ± .
28 1.39[O II] 7319 d . ± . . ± . . ± . . ± . . ± .
13 9 . ± . . ± . . ± .
21 1.37[O II] 7330 e . ± . . ± . . ± . . ± .
10 11 . ± . . ± .
17 1.24[S III] 9530.6+31.1 55 . ± . . ± .
12 27 . ± . . ± .
22 1.22 a Fluxes are relative to H β : “100” corresponds to (31 . ± . × − erg s − cm − . b E ( B − V ) = 0 .
16, with E ( B − V ) = 0 .
10 for LMC and E ( B − V ) = 0 .
06 for the Milky Way. c The recession velocity of the peak flux. d Applies to both lines at 7318 .
92 ˚A and 7319 .
99 ˚A. e Applies to both lines at 7329 .
67 ˚A and 7330 .
73 ˚A.0 Gr¨oningsson et al.: Line emission from the inner ring of SN 1987A
Table B.3.
Emission lines from the unshocked gas of the ER at Epoch 4.
North SouthRest wavel. V cpeak V FWHM V peak V FWHM
ExtinctionEmission ˚A Relative flux a km s − km s − Relative flux a km s − km s − correction b [Ne V] 3425.86 18 . ± . . ± . . ± .
76 15 . ± . . ± . . ± .
71 2.06[O II] 3726.03 98 . ± . . ± . . ± .
11 99 . ± . . ± . . ± .
18 1.98[O II] 3728.82 61 . ± . . ± . . ± .
22 63 . ± . . ± . . ± .
21 1.98[Ne III] 3868.75 59 . ± . . ± . . ± .
20 49 . ± . . ± . . ± .
23 1.94[S II] 4068.60 11 . ± . . ± . . ± .
65 11 . ± . . ± . . ± .
66 1.90[S II] 4076.35 4 . ± . . ± . . ± .
02 3 . ± . . ± . . ± .
61 1.90H δ . ± . . ± . . ± .
71 18 . ± . . ± . . ± .
68 1.89H γ . ± . . ± . . ± .
45 39 . ± . . ± . . ± .
53 1.84[O III] 4363.21 10 . ± . . ± . . ± .
52 8 . ± . . ± . . ± .
64 1.84He II 4685.7 10 . ± . . ± . . ± .
34 7 . ± . . ± . . ± .
20 1.75H β . . ± . . ± .
25 100 . ± . . ± . . ± .
18 1.71[O III] 4958.91 94 . ± . . ± . . ± .
07 75 . ± . . ± . . ± .
14 1.68[O III] 5006.84 278 . ± . . ± . . ± .
04 229 . ± . . ± . . ± .
05 1.67[O I] 5577.34 0 . ± . . ± . . ± .
30 0 . ± . . ± . . ± .
48 1.57[N II] 5754.59 18 . ± . . ± . . ± .
15 17 . ± . . ± . . ± .
21 1.54He I 5875.63 11 . ± . . ± . . ± .
29 12 . ± . . ± . . ± .
27 1.53[O I] 6300.30 34 . ± . . ± . . ± . . ± . . ± . . ± .
27 1.47[S III] 6312.06 2 . ± . . ± . . ± .
20 2 . ± . . ± . . ± .
83 1.47[O I] 6363.78 11 . ± . . ± . . ± .
10 13 . ± . . ± . . ± .
10 1.47[N II] 6548.05 263 . ± . . ± . . ± .
03 281 . ± . . ± . . ± .
03 1.45H α . ± . . ± . . ± .
07 380 . ± . . ± . . ± .
07 1.45[N II] 6583.45 812 . ± . . ± . . ± .
02 866 . ± . . ± . . ± .
03 1.45He I 6678.15 3 . ± . . ± . . ± .
47 3 . ± . . ± . . ± .
38 1.44[S II] 6716.44 39 . ± . . ± . . ± .
05 48 . ± . . ± . . ± .
07 1.43[S II] 6730.82 63 . ± . . ± . . ± .
04 74 . ± . . ± . . ± .
09 1.43[Ar III] 7135.79 9 . ± . . ± . . ± .
12 7 . ± . . ± . . ± .
37 1.39[Fe II] 7155.16 3 . ± . . ± . . ± .
29 3 . ± . . ± . . ± .
29 1.39[O II] 7319 d . ± . . ± . . ± . . ± . . ± .
22 3 . ± . . ± . . ± .
23 1.37[O II] 7330 e . ± . . ± . a Fluxes are relative to H β : “100” corresponds to (32 . ± . × − erg s − cm − . b E ( B − V ) = 0 .
16, with E ( B − V ) = 0 .
10 for LMC and E ( B − V ) = 0 .
06 for the Milky Way. c The recession velocity of the peak flux. d Applies to both lines at 7318 .
92 ˚A and 7319 .
99 ˚A. e Applies to both lines at 7329 .
67 ˚A and 7330 .
73 ˚A.r¨oningsson et al.: Line emission from the inner ring of SN 1987A 21
Table B.4.
Emission lines from the unshocked gas of the ER at Epoch 5.
North SouthRest wavel. V cpeak V FWHM V peak V FWHM
ExtinctionEmission ˚A Relative flux a km s − km s − Relative flux a km s − km s − correction b [Ne V] 3425.86 18 . ± . . ± . . ± .
98 18 . ± . . ± . . ± .
02 2.06[O II] 3726.03 94 . ± . . ± . . ± .
21 108 . ± . . ± . . ± .
19 1.98[O II] 3728.82 59 . ± . . ± . . ± .
30 68 . ± . . ± . . ± .
39 1.98[Ne III] 3868.75 61 . ± . . ± . . ± .
17 50 . ± . . ± . . ± .
14 1.94[S II] 4068.60 10 . ± . . ± . . ± .
41 11 . ± . . ± . . ± .
54 1.90[S II] 4076.35 3 . ± . . ± . . ± .
74 3 . ± . . ± . . ± .
71 1.90H δ . ± . . ± . . ± .
64 20 . ± . . ± . . ± .
78 1.89H γ . ± . . ± . . ± .
35 42 . ± . . ± . . ± .
30 1.84[O III] 4363.21 9 . ± . . ± . . ± .
31 9 . ± . . ± . . ± .
34 1.84He II 4685.7 9 . ± . . ± . . ± .
86 9 . ± . . ± . . ± .
80 1.75H β . . ± . . ± .
25 122 . ± . . ± . . ± .
20 1.71[O III] 4958.91 100 . ± . . ± . . ± .
12 97 . ± . . ± . . ± .
14 1.68[O III] 5006.84 303 . ± . . ± . . ± .
11 287 . ± . . ± . . ± .
08 1.67[O I] 5577.34 0 . ± . . ± . . ± .
94 0 . ± . . ± . . ± .
64 1.57[N II] 5754.59 18 . ± . . ± . . ± .
29 21 . ± . . ± . . ± .
38 1.54He I 5875.63 12 . ± . . ± . . ± .
65 14 . ± . . ± . . ± .
52 1.53[O I] 6300.30 33 . ± . . ± . . ± .
42 43 . ± . . ± . . ± .
33 1.47[S III] 6312.06 3 . ± . . ± . . ± .
65 3 . ± . . ± . . ± .
82 1.47[O I] 6363.78 13 . ± . . ± . . ± .
18 16 . ± . . ± . . ± .
22 1.47[N II] 6548.05 271 . ± . . ± . . ± .
06 326 . ± . . ± . . ± .
05 1.45H α . ± . . ± . . ± .
13 500 . ± . . ± . . ± .
11 1.45[N II] 6583.45 838 . ± . . ± . . ± .
04 992 . ± . . ± . . ± .
05 1.45He I 6678.15 4 . ± . . ± . . ± .
27 4 . ± . . ± . . ± .
83 1.44[S II] 6716.44 38 . ± . . ± . . ± .
10 52 . ± . . ± . . ± .
14 1.43[S II] 6730.82 59 . ± . . ± . . ± .
10 80 . ± . . ± . . ± .
12 1.43[Ar III] 7135.79 9 . ± . . ± . . ± .
14 10 . ± . . ± . . ± .
15 1.39[Fe II] 7155.16 3 . ± . . ± . . ± .
14 3 . ± . . ± . . ± .
27 1.39[O II] 7319 d . ± . . ± . . ± . . ± . . ± .
40 2 . ± . . ± . . ± .
13 1.37[O II] 7330 e . ± . . ± . a Fluxes are relative to H β : “100” corresponds to (31 . ± . × − erg s − cm − . b E ( B − V ) = 0 .
16, with E ( B − V ) = 0 .
10 for LMC and E ( B − V ) = 0 .
06 for the Milky Way. c The recession velocity of the peak flux. d Applies to both lines at 7318 .
92 ˚A and 7319 .
99 ˚A. e Applies to both lines at 7329 .
67 ˚A and 7330 .
73 ˚A.2 Gr¨oningsson et al.: Line emission from the inner ring of SN 1987A
Table B.5.
Emission lines from the unshocked gas of the ER at Epoch 6.
North SouthRest wavel. V cpeak V FWHM V peak V FWHM
ExtinctionEmission ˚A Relative flux a km s − km s − Relative flux a km s − km s − correction b [Ne V] 3425.86 20 . ± . . ± . . ± .
88 20 . ± . . ± . . ± .
53 2.06[O II] 3726.03 107 . ± . . ± . . ± .
13 119 . ± . . ± . . ± .
13 1.98[O II] 3728.82 67 . ± . . ± . . ± .
18 74 . ± . . ± . . ± .
14 1.98[Ne III] 3868.75 69 . ± . . ± . . ± .
17 66 . ± . . ± . . ± .
19 1.94[S II] 4068.60 8 . ± . . ± . . ± .
39 10 . ± . . ± . . ± .
33 1.90[S II] 4076.35 3 . ± . . ± . . ± .
77 3 . ± . . ± . . ± .
57 1.90H δ . ± . . ± . . ± .
60 22 . ± . . ± . . ± .
70 1.89H γ . ± . . ± . . ± .
48 52 . ± . . ± . . ± .
34 1.84[O III] 4363.21 13 . ± . . ± . . ± .
41 12 . ± . . ± . . ± .
36 1.84He II 4685.7 10 . ± . . ± . . ± .
99 13 . ± . . ± . . ± .
19 1.75H β . ± . . ± .
28 138 . ± . . ± . . ± .
15 1.71[O III] 4958.91 122 . ± . . ± . . ± .
06 129 . ± . . ± . . ± .
06 1.68[O III] 5006.84 365 . ± . . ± . . ± .
09 375 . ± . . ± . . ± .
12 1.67[O I] 5577.34 0 . ± . . ± . . ± .
28 0 . ± . . ± . . ± .
53 1.57[N II] 5754.59 19 . ± . . ± . . ± .
16 24 . ± . . ± . . ± .
54 1.54He I 5875.63 11 . ± . . ± . . ± .
44 14 . ± . . ± . . ± .
42 1.53[O I] 6300.30 35 . ± . . ± . . ± .
63 47 . ± . . ± . . ± .
43 1.47[S III] 6312.06 4 . ± . . ± . . ± .
39 4 . ± . . ± . . ± .
42 1.47[O I] 6363.78 13 . ± . . ± . . ± .
18 17 . ± . . ± . . ± .
33 1.47[N II] 6548.05 285 . ± . . ± . . ± .
03 359 . ± . . ± . . ± .
03 1.45H α . ± . . ± . . ± .
16 621 . ± . . ± . . ± .
10 1.45[N II] 6583.45 880 . ± . . ± . . ± .
03 1109 . ± . . ± . . ± .
03 1.45He I 6678.15 2 . ± . . ± . . ± .
58 4 . ± . . ± . . ± .
34 1.44[S II] 6716.44 38 . ± . . ± . . ± .
08 53 . ± . . ± . . ± .
09 1.43[S II] 6730.82 58 . ± . . ± . . ± .
06 79 . ± . . ± . . ± .
12 1.43[Ar III] 7135.79 12 . ± . . ± . . ± .
14 13 . ± . . ± . . ± .
11 1.39[Fe II] 7155.16 3 . ± . . ± . . ± .
30 4 . ± . . ± . . ± .
26 1.39[O II] 7319 d . ± . . ± . . ± . . ± . . ± .
37 2 . ± . . ± . . ± .
52 1.37[O II] 7330 e . ± . . ± . a Fluxes are relative to H β : “100” corresponds to (23 . ± . × − erg s − cm − . b E ( B − V ) = 0 .
16, with E ( B − V ) = 0 .
10 for LMC and E ( B − V ) = 0 .
06 for the Milky Way. c The recession velocity of the peak flux. d Applies to both lines at 7318 .
92 ˚A and 7319 .
99 ˚A. e Applies to both lines at 7329 .
67 ˚A and 7330 .
73 ˚A.r¨oningsson et al.: Line emission from the inner ring of SN 1987A 23
Table B.6.
Emission lines from the unshocked gas of the ER at Epoch 7.
North SouthRest wavel. V cpeak V FWHM V peak V FWHM
ExtinctionEmission ˚A Relative flux a km s − km s − Relative flux a km s − km s − correction b [Ne V] 3425.86 23 . ± . . ± . . ± .
71 23 . ± . . ± . . ± .
68 2.06[O II] 3726.03 105 . ± . . ± . . ± .
17 123 . ± . . ± . . ± .
18 1.98[O II] 3728.82 66 . ± . . ± . . ± .
32 76 . ± . . ± . . ± .
38 1.98[Ne III] 3868.75 64 . ± . . ± . . ± .
41 63 . ± . . ± . . ± .
18 1.94[S II] 4068.60 6 . ± . . ± . . ± .
64 6 . ± . . ± . . ± .
78 1.90[S II] 4076.35 3 . ± . . ± . . ± .
30 3 . ± . . ± . . ± .
79 1.90H δ . ± . . ± . . ± .
85 21 . ± . . ± . . ± .
38 1.89H γ . ± . . ± . . ± .
48 55 . ± . . ± . . ± .
34 1.84[O III] 4363.21 11 . ± . . ± . . ± .
54 10 . ± . . ± . . ± .
36 1.84He II 4685.7 10 . ± . . ± . . ± .
74 12 . ± . . ± . . ± .
11 1.75H β . ± . . ± .
31 132 . ± . . ± . . ± .
21 1.71[O III] 4958.91 123 . ± . . ± . . ± .
06 146 . ± . . ± . . ± .
08 1.68[O III] 5006.84 367 . ± . . ± . . ± .
11 453 . ± . . ± . . ± .
13 1.67[O I] 5577.34 0 . ± . . ± . . ± .
43 1 . ± . . ± . . ± .
01 1.57[N II] 5754.59 23 . ± . . ± . . ± .
13 29 . ± . . ± . . ± .
34 1.54He I 5875.63 11 . ± . . ± . . ± .
39 14 . ± . . ± . . ± .
58 1.53[O I] 6300.30 33 . ± . . ± . . ± .
31 47 . ± . . ± . . ± .
25 1.47[S III] 6312.06 3 . ± . . ± . . ± .
60 4 . ± . . ± . . ± .
40 1.47[O I] 6363.78 14 . ± . . ± . . ± .
13 18 . ± . . ± . . ± .
18 1.47[N II] 6548.05 266 . ± . . ± . . ± .
04 357 . ± . . ± . . ± .
04 1.45H α . ± . . ± . . ± .
16 627 . ± . . ± . . ± .
10 1.45[N II] 6583.45 816 . ± . . ± . . ± .
03 1093 . ± . . ± . . ± .
03 1.45He I 6678.15 3 . ± . . ± . . ± .
85 4 . ± . . ± . . ± .
61 1.44[S II] 6716.44 32 . ± . . ± . . ± .
14 48 . ± . . ± . . ± .
20 1.43[S II] 6730.82 48 . ± . . ± . . ± .
09 71 . ± . . ± . . ± .
16 1.43[Ar III] 7135.79 11 . ± . . ± . . ± .
24 11 . ± . . ± . . ± .
26 1.39[Fe II] 7155.16 3 . ± . . ± . . ± .
45 3 . ± . . ± . . ± .
21 1.39[O II] 7319 d . ± . . ± . . ± . . ± . . ± . . ± . . ± .
63 1.37[O II] 7330 e . ± . . ± . a Fluxes are relative to H β : “100” corresponds to (22 . ± . × − erg s − cm − . b E ( B − V ) = 0 .
16, with E ( B − V ) = 0 .
10 for LMC and E ( B − V ) = 0 .
06 for the Milky Way. c The recession velocity of the peak flux. d Applies to both lines at 7318 .
92 ˚A and 7319 .
99 ˚A. e Applies to both lines at 7329 .
67 ˚A and 7330 .
73 ˚A.4 Gr¨oningsson et al.: Line emission from the inner ring of SN 1987A
Table B.7.
Fluxes of emission lines from the shocked gas from the northern ER a . Ion λ rest Extinct. 5039 d 5703 d 6618 d 6826 d 7183 d 7559 d[Ne V] 3425.86 2.06 - 2 . ± .
67 1 . ± .
28 2 . ± .
36 1 . ± .
30 1 . ± . . ± .
35 5 . ± .
92 5 . ± .
64 6 . ± .
08 6 . ± .
38 7 . ± . . ± .
64 21 . ± .
62 17 . ± .
48 16 . ± .
35 14 . ± .
34 12 . ± . . ± .
60 32 . ± .
81 35 . ± .
58 33 . ± .
49 36 . ± .
29 36 . ± . . ± .
99 8 . ± .
38 8 . ± .
29 8 . ± .
13 8 . ± .
17 7 . ± . δ . ± .
11 20 . ± .
43 21 . ± .
64 19 . ± .
38 21 . ± .
28 21 . ± . γ . ± .
30 42 . ± .
77 41 . ± .
55 40 . ± .
58 41 . ± .
57 40 . ± . . ± .
07 5 . ± .
35 3 . ± .
32 3 . ± .
30 2 . ± .
19 2 . ± . . ± .
20 1 . ± .
07 1 . ± .
05 1 . ± .
08 1 . ± . . ± .
30 7 . ± .
22 6 . ± .
11 6 . ± .
14 5 . ± . . ± .
55 8 . ± .
28 4 . ± .
11 4 . ± .
11 3 . ± .
05 2 . ± . . ± .
66 27 . ± .
39 14 . ± .
13 14 . ± .
28 12 . ± .
13 10 . ± . c . ± .
30 0 . ± .
16 0 . ± .
07 0 . ± .
06 1 . ± .
07 1 . ± . . ± .
24 2 . ± .
19 2 . ± .
09 2 . ± .
13 3 . ± .
07 3 . ± . . ± .
55 3 . ± .
17 3 . ± .
09 3 . ± .
14 3 . ± .
08 4 . ± . . ± .
72 42 . ± .
44 32 . ± .
34 34 . ± .
31 34 . ± .
28 36 . ± . . ± .
69 29 . ± .
32 24 . ± .
28 27 . ± .
42 30 . ± .
26 30 . ± . . ± .
06 0 . ± .
03 0 . ± .
05 0 . ± .
03 0 . ± . . ± .
95 45 . ± .
41 42 . ± .
67 50 . ± .
68 55 . ± .
50 59 . ± . . ± .
15 - - - -[O I] 6363.78 1.47 16 . ± .
68 15 . ± .
43 14 . ± .
41 17 . ± .
49 18 . ± .
24 20 . ± . . ± .
21 1 . ± .
14 1 . ± .
08 1 . ± .
13 1 . ± .
11 1 . ± . . ± .
53 14 . ± .
26 9 . ± .
17 9 . ± .
33 10 . ± .
12 11 . ± . α . ± .
42 403 . ± .
39 344 . ± .
26 387 . ± .
92 393 . ± .
58 406 . ± . . ± .
94 43 . ± .
43 31 . ± .
26 33 . ± .
40 31 . ± .
20 32 . ± . . ± .
57 7 . ± .
19 6 . ± .
09 6 . ± .
15 7 . ± .
19 7 . ± . . ± .
16 0 . ± .
05 0 . ± .
05 0 . ± .
08 0 . ± .
05 0 . ± . . ± .
21 1 . ± .
16 1 . ± .
06 1 . ± .
13 1 . ± .
05 2 . ± . . ± .
05 0 . ± .
02 0 . ± .
02 0 . ± .
02 0 . ± . . ± .
38 3 . ± .
16 2 . ± .
09 2 . ± .
04 1 . ± .
07 1 . ± . . ± .
36 8 . ± .
13 10 . ± .
11 10 . ± .
08 11 . ± .
11 13 . ± . . ± .
35 1 . ± .
16 1 . ± .
09 1 . ± .
07 1 . ± .
07 1 . ± . . ± .
16 - - - -[S III] 9531.10 1.22 7 . ± .
37 10 . ± .
28 - - - -Flux(H β ) b . ± .
61 164 . ± .
06 468 . ± .
76 515 . ± .
75 556 . ± .
04 564 . ± . a Fluxes are relative to the shocked component of H β × b H β fluxes are in 10 − erg s − cm − Applies to both lines at 5197 .
90 ˚A and 5200 ..
90 ˚A and 5200 ..
26 ˚A.r¨oningsson et al.: Line emission from the inner ring of SN 1987A 25
Table B.8.
Fluxes of emission lines from the shocked gas from the southern ER a . Ion λ rest Extinct. 5039 d 5703 d 6618 d 6826 d 7183 d 7559 d[Ne V] 3425.86 2.06 - 4 . ± .
13 2 . ± .
48 3 . ± .
65 1 . ± .
41 1 . ± . . ± .
24 4 . ± .
55 4 . ± .
19 4 . ± .
55 5 . ± . . ± .
30 19 . ± .
00 19 . ± .
80 18 . ± .
50 15 . ± .
50 13 . ± . . ± .
78 27 . ± .
37 34 . ± .
93 32 . ± .
52 32 . ± .
45 31 . ± . . ± .
78 8 . ± .
60 7 . ± .
43 7 . ± .
22 7 . ± .
20 7 . ± . δ . ± .
53 17 . ± .
18 20 . ± .
61 19 . ± .
30 19 . ± .
33 18 . ± . γ . ± .
22 37 . ± .
28 39 . ± .
69 39 . ± .
52 39 . ± .
47 37 . ± . . ± .
55 6 . ± .
42 4 . ± .
33 4 . ± .
16 3 . ± .
14 2 . ± . . ± .
35 1 . ± .
12 1 . ± .
07 1 . ± .
13 1 . ± . . ± .
80 6 . ± .
34 7 . ± .
17 6 . ± .
14 6 . ± . . ± .
69 12 . ± .
47 7 . ± .
20 7 . ± .
32 5 . ± .
23 4 . ± . . ± .
99 36 . ± .
96 24 . ± .
26 24 . ± .
31 18 . ± .
20 17 . ± . c . ± .
30 0 . ± .
13 0 . ± .
09 0 . ± .
09 1 . ± . . ± .
26 2 . ± .
14 2 . ± .
21 2 . ± .
09 3 . ± . . ± .
62 2 . ± .
22 2 . ± .
07 2 . ± .
22 3 . ± .
15 4 . ± . . ± .
41 42 . ± .
90 37 . ± .
40 41 . ± .
49 37 . ± .
31 42 . ± . . ± .
89 33 . ± .
72 28 . ± .
33 30 . ± .
53 30 . ± .
23 31 . ± . . ± .
13 0 . ± .
06 0 . ± .
08 0 . ± .
05 0 . ± . . ± .
36 46 . ± .
91 43 . ± .
48 48 . ± .
82 47 . ± .
60 54 . ± . . ± .
22 - - - -[O I] 6363.78 1.47 6 . ± .
66 15 . ± .
59 14 . ± .
53 16 . ± .
43 16 . ± .
17 18 . ± . . ± .
30 1 . ± .
10 2 . ± .
19 1 . ± .
06 2 . ± . . ± .
79 16 . ± .
51 12 . ± .
23 13 . ± .
38 12 . ± .
16 12 . ± . α . ± .
63 388 . ± .
30 364 . ± .
92 384 . ± .
62 366 . ± .
92 411 . ± . . ± .
88 50 . ± .
87 39 . ± .
36 42 . ± .
53 36 . ± .
26 37 . ± . . ± .
85 6 . ± .
32 6 . ± .
17 7 . ± .
26 7 . ± .
10 8 . ± . . ± .
91 1 . ± .
15 0 . ± .
09 0 . ± .
06 0 . ± .
03 0 . ± . . ± .
21 2 . ± .
22 1 . ± .
13 1 . ± .
18 1 . ± .
04 1 . ± . . ± .
02 0 . ± . . ± .
65 4 . ± .
38 3 . ± .
16 3 . ± .
10 2 . ± .
08 2 . ± . . ± .
59 7 . ± .
34 8 . ± .
13 8 . ± .
10 9 . ± .
08 9 . ± . . ± .
21 1 . ± .
07 1 . ± .
05 1 . ± .
07 1 . ± . . ± .
52 - - - -[S III] 9531.10 1.22 4 . ± .
84 11 . ± .
83 - - - -Flux(H β ) b . ± .
41 46 . ± .
60 247 . ± .
56 329 . ± .
93 429 . ± .
75 505 . ± . a Fluxes are relative to the shocked component of H β × b H β fluxes are in 10 − erg s − cm − Applies to both lines at 5197 .
90 ˚A and 5200 ..