DTIC ADA639998: Discovery of a high velocity, spatially extended emission shell in front of the southeast lobe of the eta Carinae Homunculus

5 Pages·0.2 MB·English

by  Defense Technical Information Center

A&A 389, L65{L68 (2002) Astronomy
DOI: 10.1051/0004-6361:20020805 &
(cid:13)c ESO 2002 Astrophysics r
o
t
i
d
E
e
Discovery of a high velocity, spatially extended emission \shell" h
t
in front of the southeast lobe of the (cid:17) Carinae Homunculus? o
t
r
e
D. G. Currie1;2, B. N. Dorland1;3, and A. Kaufer4
t
t
e
1 Physics Department,Universityof Maryland, College Park, MD 20742-4111, USA L
2 European SouthernObservatory,Garching beiMuenchen, Germany
3 Astrometry Department,United States NavalObservatory,3450 Mass. Ave,NW,Washington,
DC20392-5420, USA
4 European SouthernObservatory,Alonso de Cordova 3107, Vitacura, Santiago 19, Chile
Received 2 April2002 / Accepted 31 May 2002
Abstract. We report the discovery of the (cid:17) Carinae \Ghost Shell," a high-velocity, spatially extended emission
featurethatliesinfrontofthesoutheastlobeofthe(cid:17) CarinaeHomunculus.Usingdataobtainedwith\Kueyen,"
oneoftheEuropeanSouthernObservatory’sVeryLargeTelescope8.2mtelescopesanditsUltravioletandVisible
Echelle Spectrograph instrument, we have observed a structure in velocity space of width (cid:25)35 kms−1 and with
Dopplervelocities ranging from −675(cid:20)v(cid:20)−850 kms−1. This is up to 500 kms−1 faster than the Homunculus
front wall. The structure is distinct from the front wall in velocity space, and extends beyond the Homunculus’
spatial boundaries. The Ghost Shell has been detected in emission for multiple allowed Balmer lines and in
forbidden lines of [NII], [SII], and [ArIII]. The feature is also associated with a complex absorption structure
in CaH and Klines. We propose that the Ghost Shell lies outside the Homunculus and represents the forward
shock between thefast stellar wind of theGreat Eruption epoch and theolder slow massive stellar wind.
Key words.stars: individual: (cid:17) Carinae { variables: LBV { winds, outflows { circumstellar matter
1. Introduction the (cid:12)rst use of UVES for extended objects such as the
Homunculus.
Thestar(cid:17)Carinae,generallyconsideredanextrememem-
During this e(cid:11)ort to take advantage of the capabil-
ber of the Luminous Blue Variables (LBVs) (Humphreys
ities available with UVES, we have found a new emis-
& Davidson 1994), and its surrounding nebulosities have
sion feature. This feature, which we call the \Ghost
been an ongoing subject of observation and theoretical
Shell" (GS), extends at least over the southeast lobe of
modeling for many years. Much attention has been paid
the Homunculus, with typical velocities of order hun-
to the Homunculus nebula, whose material was expelled dreds of kms−1 faster than the Homunculus front wall.
from the centralstar during the GreatEruptionof c.1842
Observationalresults describedbelowlead us to conclude
(Currie et al. 1996). Under the assumption of unacceler-
that the GS is a phenomenon associated with the Great
ated expansion of the Homunculus material, the plane-
Eruption, but distinct from the Homunculus. In this let-
of-sky and Doppler motion measurements have allowed
ter we describe the initial results relating to the GS and
us to de(cid:12)ne in detail the \Double Flask" Model of the
propose a physical explanation.
Homunculus morphology (Dowling 1996; Currie et al.
2000). In order to re(cid:12)ne the model, we are currently en-
gagedinextendingtheaboveDopplermeasurementsusing 2. Observations and reduction
the high spectral resolutionof the Ultraviolet and Visible
2.1. Collection, calibration, and initial reduction
Echelle Spectrograph(UVES), the excellentseeing condi-
tons available at Paranal and the collection capability of UVES observations of (cid:17) Carinae were made in December
the Very Large Telescope (VLT). These results represent 1999 and January 2000 using Kueyen, one of the 8.2 m
VLT telescopes. The January 2000 data set consisted
of 18 long slit spectra observations. Each slit’s long axis
Send o(cid:11)print requests to: B. N. Dorland,
e-mail: [email protected] was 1000 (in the \blue" arm) or 1200 (in the \red" arm)
? Based onobservationscollected at theEuropeanSouthern wide. Each long slit was oriented orthogonal to (i.e, at a
Observatory at Paranal, Chile (UVES commissioning II). positionangle=45deg)andcenteredontheHomunculus
Article published by EDP Sciences and available at http://www.aanda.org or http://dx.doi.org/10.1051/0004-6361:20020805
Report Documentation Page Form Approved
OMB No. 0704-0188
Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington
VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it

does not display a currently valid OMB control number.
1. REPORT DATE 3. DATES COVERED
2002 2. REPORT TYPE 00-00-2002 to 00-00-2002
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
Discovery Of A High Velocity, Spatially Extended Emission
5b. GRANT NUMBER
5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S) 5d. PROJECT NUMBER
5e. TASK NUMBER
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION
United States Naval Observatory,3450 Mass. Ave, REPORT NUMBER
NW,Washington,DC,20392
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)
11. SPONSOR/MONITOR’S REPORT
NUMBER(S)
12. DISTRIBUTION/AVAILABILITY STATEMENT
Approved for public release; distribution unlimited
13. SUPPLEMENTARY NOTES
A&A 389, L65-L68 (2002)
14. ABSTRACT

15. SUBJECT TERMS
16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF
ABSTRACT OF PAGES RESPONSIBLE PERSON
a. REPORT b. ABSTRACT c. THIS PAGE Same as 5
unclassified unclassified unclassified Report (SAR)
Standard Form 298 (Rev. 8-98)
Prescribed by ANSI Std Z39-18
L66 D. G. Currie et al.: Ghost shell of (cid:17) Carinae
r Table1.GhostShellemissionlineobservationsandstrengths.
o
\(cid:21) " indicates the zero-velocity laboratory reference wave-
t 0
i length. \2000 Data" and \1999 Data" columns indicate slit
d
o(cid:11)setsinwhichthelineisdetectable.<E >columngives
norm
E 12" GS GS H alpha themean line energy,normalized to H(cid:12)=1:00.
P-Cygni
e
h
Identi(cid:12)cation (cid:21) 2000Data 1999Data <E >
0 norm
t
6500 6550 6600 Hγ 4340.5 −3 0.62
o
t [rNef eIIr]e6n5c4e8.1 Hre faelrpehnace H(cid:12) 4861.3 −8$−4 −4, −3 1.00
r Fig.1.[NII]andH(cid:11)emissioninUVESdata,slit−8.\GS"in- [NII] 6548.1 −8$−2 −4 1.85
e H(cid:11) 6562.8 −8$−5 −4 2.57
dicates suspected Ghost Shell emission. Dashed lines indicate
t [NII] 6583.5 −7 4.23
t reference wavelengths; box indicates region shown in Fig. 2.
e [SII] 6716.4 −8$−6 −4 0.16
Abscissa is the spectral axis, ordinate is spatial.
L [SII] 6730.8 −8$−6 −4 0.19
[ArIII] 7135.7 −3 0.12
axis of symmetry (AOS). We restrict discussion in this 3. Observational results
letter to slits covering the SE lobe of the Homunculus,
numbered −2 to− 8, with the slit number correspond- The initial discovery of the GS was made when mea-
suring two correlated features observed in Slit −8 at
ing (approximately)to o(cid:11)set in arc seconds fromthe cen-
about 6530 and 6544 (cid:23)A, shown in Fig. 1. Referencing
tral star position. The spectral coveragein the 2000 data
taken in dichroic #1 mode was 3100{3866 (cid:23)A (blue) and these two features to zero-velocity wavelengths of 6548.1
4791{6798 (cid:23)A (red); resolving power (cid:21)=(cid:14)(cid:21) (cid:25)70000 (blue) and6562.8(cid:23)A(i.e,[NII]andH(cid:11)) yieldsaDopplervelocity
and(cid:21)=(cid:14)(cid:21)(cid:25)100000(red);seeingfortheobservingsession of (cid:25)−825 kms−1 for both of the features.
was(cid:25)0:600(muchgreaterthantheslitwidthsof0.400(blue) According to the current Homunculus models and ob-
and 0.300 (red)); and exposure time in each slit position servations(Dowling1996;Currieetal.2000),blue-shifted
photons must all be from wall emission, but no known
was 300 seconds. 1999 data were restricted to two slits,
Homunculus feature is present at this position in velocity
one in dichroic #1 mode and the second in dichroic #2
space. For slit −8, the front wall of the Homunculus is
mode with resultant additional spectral coverage from
3760{4973 (cid:23)A (blue) and 6716{10 401 (cid:23)A (red). 1999 ex- closer to 350 kms−1(Currie et al. 2000; Davidson et al.
2001). In orderto determine if this was ananomalousob-
posure times were 3{5 times longer than 2000 exposure
servation or an actual physical phenomenon, we searched
times. All references in this paper are to 2000 data un-
for this feature in other slits and spectral regions. Using
lessotherwisenoted.Calibrationandinitialreductionwas
the velocity space pro(cid:12)le for [NII] 6548.1 as a template,
performed at the ESO Vitacura facility with the ESO-
wehavedetected the GSfeaturein emissionineight lines
MIDAS data reduction system using modi(cid:12)ed versions of
in multiple slit positions in both the 1999 and 2000 data
the uves and feros contexts. The latter was modi(cid:12)ed in
(see Table 1).
ordertoproperlyaddressthe spatiallyextendednatureof
Representative data for (cid:12)ve of the lines present in
thesource.Theresultantcombinedorderdatahasaspec-
tral scale of 0.02 (cid:23)A and a spatial scale of 0:18200 (blue) the 2000 data are shown in Fig. 2. The data have been
and 0:24600 (red) per pixel. scaledandalignedsothatthe abscissarepresentsvelocity
withrespecttothreeofthereferencewavelengthsgivenin
Table 1. The unmistakable agreement in Doppler veloci-
2.2. Analysis reduction ties of the lines is excellent. Velocity measurements on a
spatial row-by-spatial row basis typically di(cid:11)er by <1%.
Data reduction was accomplished as follows. A 2D 3(cid:2)3 An example of this close agreement in line velocities for
median (cid:12)lter was applied to the data to reduce high fre- (cid:12)ve di(cid:11)erent lines are shown in Fig. 3 for slit −7.
quency pixel noise. These data were then (cid:12)ltered with
a4(cid:23)A widemedian(cid:12)lterinthespectraldirection.Thelat-
4. Discussion
terwassubtractedfromtheabovedatatoremovelowfre-
quency spectralfeaturesand isolateGS lines.A Gaussian
4.1. 3-D Spatial con(cid:12)guration of Ghost Shell
function was (cid:12)tted to suspected emission line positions
and centroid, width, and amplitude of these lines were The de(cid:12)nitive method to determine the 3-D spatial con-
calculated. From these results, we derived velocity, width (cid:12)guration of the GS is to combine plane-of-the-sky as-
in physical units, and \energy", de(cid:12)ned as the product trometry with Doppler astrometry (i.e., the method used
of the width of a line in kms−1 by the amplitude as mea- fortheHomunculusdescribedinDowling1996andCurrie
suredinrelativefluxunits.Incertain(noted)instanceswe etal.2000).However,wehaveessentiallyasingleepochso
summedoverthespectraldimensiontoincreasesignal-to- other methods must be used to evaluate the structure of
noise ratio (S=N). theGS.WehypothesizethattheformoftheGSisroughly
D. G. Currie et al.: Ghost shell of (cid:17) Carinae L67
N r
o
t
i
GS H beta d
12" H beta P-Cygni E
e
h
4840 4860
t
o
t
N
r
e
FW
t
GS GS [N II] 6548 t
12" [N II] H alpha P-Cygni e
H alpha L
6530 6550
N N
FW FW
[S II] 6716
12" GS [S II] GS [S II] [S II] 6731
6700 6720
Approx. velocity
1500 1000 500 0
scale (km/sec)
Fig.2.GhostShellemissionfor(cid:12)velinesforslit −6.\GS"indicatesGhostShell,\FW"Homunculusfrontwall,and\N"back-
groundnebularemissionlines.H(cid:11)andH(cid:12) P-Cygnifeaturesarenoted.Imagesandvelocityscalearealignedalongzero-velocity
reference wavelengths for H(cid:12), H(cid:11),and [SII] 6730.8. Median subtraction hasbeen applied to all images. The bottomimage has
been averaged over0.4 (cid:23)A spectral binsto increase S=N.
in velocity space for individual slits yields an equivalent
sphereofradius(cid:25)1100,and3)assumingthatthefastwind
began during the Great Eruption epoch, and given the
measured Doppler velocities for material behind the for-
ward shock, we derive a radial velocity of (cid:25)900 kms−1.
Assuming that it is a heavy shock (see discussion below)
we derive a velocity of the fast shock (cid:25)1200 kms−1. The
productofthisvelocitywiththeelapsedtimeyieldsacur-
rent radius (cid:25)1200.
The (cid:12)rst argument suggests that the GS lies outside
of the Homunculus. The remaining arguments indepen-
dently conclude that the GS has the form of a sphere of
radius 11{1200. Initial analysis of the data suggests that
this sphere is centeredalongthe AOSbut somewhat(i.e.,
Fig.3. Ghost Shell emission line Doppler velocities along
a few arcseconds) SE of the central star.
slit−7.[NII]6548measurementsareconnectedbydashedline.
We emphasize that these results apply only to the
SE lobe region beyond slit position −3. The quasi-
spherical approximation we propose is not valid near the
a sphere of radius (cid:25)1100 that surrounds the Homunculus central star and in the NW lobe region. In these other
andpossessesuniformradialvelocity.Wepresentthreear- regions, the GS shell is distorted as it is penetrated by
guments based onthe UVES observationsto support this fast moving ejecta such as the equatorial disk and the
hypothesis:1)observationsin Slits −6,−7,and −8,show north \jet" (vide Hester et al. 1991, for details of the de-
thattheGSemissionsextendbeyondtheHomunculusspa- bris (cid:12)eld). We are continuing to analyze UVES data to
tial boundary,2)analysis ofthe curvatureofthe GS lines determine the nature of the GS for these other regions.
L68 D. G. Currie et al.: Ghost shell of (cid:17) Carinae
r The details of the shape will be addressed in a publica- with the expected exception that we see no oxygen emis-
o
tion in preparation. sion lines. We have also found a related set of absorption
t
i featuresinCaKandH.TheDopplervelocity,energy,and
d
line width of the emission feature have been measured in
E 4.2. The Ghost Shell as the shock interface between
three Balmer lines, two [NII], two [SII] lines, and [ArIII].
e fast and slow winds These measurements show great consistency (e.g., (cid:25)1%
h
We propose that the Ghost Shell is the excited material for velocity measurements). We also argue that, over the
t
associated with the forward shock created as the current SE lobe, the emission nebula roughly has the form of
o
fast wind propagates into the earlier slow massive wind. a spherical shell of radius (cid:25)1100. We conclude that the
t
r The rear shock has then swept up material to form the Ghost Shell represents the forward shock as the current
e leading edge of the Homunculus. The wide separationbe- fast wind, intially ejected during the epoch of the Great
t
t tween the GS and the forward edge of the Homunculus Eruption, propagates into the earlier slow massive wind.
e
is due to the region being essentially adiabatic, a result
L
of thermalizing a portion of the fast wind. However, the
5.2. Future work
region emitting the Balmer, [NII], and [SII] lines is much
cooler, as indicated by the line widths of these features Details of these arguments, including presentation of line
and the [SII] temperature determination. This may be a strengthsinabsolutefluxunits,reddeningcorrections,dis-
result of cooling, or more likely, a mechanism similar to cussion of the S=N considerations, and extension of the
\Balmer shock", found in supernova remnants (Chevalier analysis to the central star and NW lobe regions will be
et al. 1980), although some puzzles remain when using addressedinseveralpapersnowinpreparation.Additional
this model. The approximately adiabatic conditions are observations will be necessary in order to constrain the
con(cid:12)rmed by computations of the cooling times that are fullyhydrodynamicalmodelingofthehistory.Bycombin-
muchlongerthanthe dynamicaltimes.Thelackofagree- ing observations with 3-D hydrodynamical modeling, we
ment between the shape of the GS and the flask with the will explore new details of the Great Eruption and the
\dimple" of the Homunculus (Currie et al. 2000) may be evolutionofboththeforwardshock(theGS)andtherear
due to the variationin density and/or other conditions of shock, that is, the Homunculus.
the slow wind as a function of latitude.
Acknowledgements. WewishtothanktheEuropeanSouthern
4.3. Observation of Ghost Shell features in previous Observatory for theopportunity to obtain these observations.
Ourresultsarepossibleduetotheveryhighresolutionandsta-
work
bility of UVES and the large collecting aperture of the VLT.
Although the GS has never been identi(cid:12)ed as such, we The (cid:12)rst author appreciates the support received from the
believethat somefeatures havebeenpreviouslyobserved. ESO/Chile Visiting Scientist Program, and wishes to thank
Wallerstein et al. (2001) observed an unidenti(cid:12)ed emis- N. Langer, J. Stone, and F. Bacciotti for useful discussions.
sion line at 6530.8 (cid:23)A, which coincides with the peak [NII] ThesecondauthorwishestothankA.Hajianformanyhelpful
discussion and USNO for supporting the analysis and prepa-
6548 emission we observe when the UVES data are aver-
ration activities associated with this paper.
aged over the SE lobe (i.e., slits −2 through −8). Gull &
Ishibashi (2001) reported H(cid:11) emission structures outside
the Homunculus, some of which may correlateto the GS. References
We have also detected a GS absorption complex
Chevalier,R.A.,Kirshner,R.P.,&Raymond,J.C.1980,ApJ,
in CaH and K bands for slit −3 in the 1999 data. This
235, 186C
matches anomalous Ca absorption features reported by
Currie, D. G., Dowling, D. M., Shaya, E. J., et al. 1996, AJ,
Davidson et al. (2001). The Doppler velocities Davidson
112, 115
et al. report for the Ca anomaly are generally consistent Currie, D. G., LeMignant, D., Svensson,B., et al. 2000, ESO
withtheDopplervelocitieswemeasurealongtheAOSfor Messenger, 101, 24
the GS emission lines. Davidson, K., Smith, N., Gull, T. R., et al. 2001, ApJ, 121,
1569
Dowling, D.M. 1996, Ph.D. Thesis, Physics Dept.,University
5. Conclusions and future work
of Maryland, College Park, MD, USA
Gull, T. R.,& Ishibashi, K. 2001, PASP,242, 151
5.1. Conclusions and implications
Hester, J. J., Light, R. M., Westphal, J. A., et al. 1991, AJ,
Wehavediscoveredauniquenewfeatureinthe(cid:17) Carinae 102, 654H
system, consisting of an emission nebula surrounding the Humphreys,R. M., & Davidson,K. 1994, PASP,106, 1025
Wallerstein, G., Gilroy,K. K.,Zethson, T., et al. 2001, PASP,
SE lobe of the Homunculus. It generally has the appro-
113, 1210
priateemissionlinesforafastshock(Chevalieretal.1980)