A pilot study of the radio-continuum emission from MASH planetary nebulae
I. S. Bojičić, Q. A. Parker, D. J. Frew, A. E. Vaughan, M. D. Filipović, M. L. P. Gunawardhana
aa r X i v : . [ a s t r o - ph . S R ] J un Astron. Nachr. / AN , No. 88, 789 – 797 (2006) /
DOI please set DOI!
A pilot study of the radio-continuum emission from MASH planetarynebulae
I. S. Bojiˇci´c , ,⋆ , Q. A. Parker , , D. J. Frew , A. E. Vaughan , M. D. Filipovi´c , and M. L. P.Gunawardhana , Department of Physics and Astronomy, Macquarie University, Sydney, NSW 2109, Australia University of Western Sydney, Locked Bag 1797, Penrith South DC, NSW 1797, Australia Australian Astronomical Observatory, Epping, NSW 1710, Australia Sydney Institute for Astronomy, School of Physics, University of Sydney, NSW 2006, AustraliaReceived 01 Jan 2010, accepted 01 Jan 2010Published online later
Key words astronomical data bases: miscellaneous - planetary nebulae: general - radiation mechanisms: thermal - radiocontinuum: ISMWe report an Australia Telescope Compact Array (ATCA) radio-continuum observations of 26 planetary nebulae (PNe)at wavelengths of 3 and 6 cm. This sample of 26 PNe were taken from the Macquarie/AAO/Strasbourg H α PNe (MASH)catalogue and previous lists. We investigate radio detection quality including measured and derived parameters for alldetected or marginally detected PNe from this combined sample. Some 11 objects from the observed sample have beensuccessfully detected and parametrized. Except for one, all detected PNe have very low radio surface brightnesses. Weuse a statistical distance scale method to calculate distances and ionised masses of the detected objects. Nebulae fromthis sample are found to be large ( > α fluxes and interstellar extinction coefficients, either taken from the literature or derived herefrom the Balmer decrement and radio to H α ratio methods. Finally, our detected fraction of the MASH pilot sample isrelatively low compared to the non-MASH sub-sample. We conclude that future radio surveys of the MASH sample mustinvolve deeper observations with better uv coverage in order to increase the fraction of detected objects and improve thequality of the derived parameters. c (cid:13) Planetary nebulae (PNe) are the manifestation of the finalevolutionary stage of low and intermediate mass stars (1-8 M ⊙ ). For a brief period ( < × yr; Frew & Parker2010) these astrophysical phenomena evolve from compactand dusty, circumstellar envelopes detached from the parentstar to the final stage when the star reaches the cooling pathon the HR diagram, the ionised material starts to disperseinto the surrounding interstellar medium and the luminosityof the nebula drops swiftly.A significant number ( ∼ α PNe(MASH) PN catalogues (Miszalski et al. 2008; Parker et al.2006) which are a direct product of the Anglo-AustralianObservatory/UK Schmidt Telescope (AAO/UKST) H α sur-vey (Parker et al. 2005). A significant fraction of MASHPNe represent the oldest stages of PN evolution, dominatingthe known population at the faint end of the PN luminosityfunction.Based on these discoveries, a pilot observational radiostudy was conducted in early 2003. This study intended toobtain and explore the radio-continuum data of a relatively ⋆ Corresponding author: e-mail: [email protected] small sub-set of the newly discovered PNe. With the knownbenefits of the radio observational data (e.g. negligible in-terstellar extinction at cm wavelengths, absolute calibration,and a well studied emission mechanism) the pilot studyaimed to examine some elementary physical properties ofthese new PNe, to test observing strategies and to confirmthe viability of a large-scale observing program of the largenumber of MASH discoveries (Bojiˇci´c et al. 2011, in prepa-ration).In this paper, we present the observational techniques,the radio detection quality, together with measured and de-rived parameters for the detected objects. We also exam-ined the spectral energy distributions (SEDs) and the corre-lation between measured radio fluxes and Balmer line fluxesestimated from the SuperCOSMOS H-alpha Survey (SHS;Gaustad et al. 2001) H α images (the MASH part of the sam-ple) or found in the literature (the previously known fractionof the sample). We also present derived distances, radii andphysical properties of the detected objects. A group of 17 MASH PNe was randomly selected for ob-servation with the ATCA. Care was taken that no obvious c (cid:13)
90 I. S. Bojiˇci´c et al.: A pilot study of the radio-continuum emission from MASH PNe
Table 1
Field names, positions and general properties for PNe in the observed pilot sample. The listed coordinatesdesignate the targeted position from the Strasbourg-ESO Catalogue of Galactic Planetary Nebulae (Acker et al. 1992, A92),Kohoutek & K¨uhl (2002, K02) and the MASH I catalogue (Parker et al. 2006, P06). Column (4) is an object status flagas described in Parker et al. (2006). Columns (8) and (9) gives the morphological classification and optically determinedangular diameter as found in the literature. Column (10) shows detection flags at 6 cm from this project. Objects flaggedwith “y” are positively detected and those with “n” not detected.
Field Name PNG Stat. a RAJ2000 DECJ2000 Cat Morph b . θ det.(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)mA02 KeWe 2 228.5-11.4 T 06 37 39.1 -18 57 24 K02 E/B? 30 nmA12 PHR0724-2021 234.7-02.2 L 07 24 13.1 -20 21 49 P06 A 85 nmA08 PHR0724-1757 232.6-01.0 T 07 24 43.4 -17 57 51 P06 Rs 169 nmA10 PHR0726-2858 242.5-05.9 T 07 26 04.8 -28 58 23 P06 R 32 nwB11 PHR0731-2439 239.3-02.7 L 07 31 59.6 -24 39 04 P06 E 19 ymA06 PHR0732-2825 242.6-04.4 T 07 32 17.4 -28 25 18 P06 B 27 ywB06 PHR0745-3535 250.3-05.4 T 07 45 41.1 -35 35 04 P06 Es 51 nwB07 PHR0755-3346 249.8-02.7 L 07 55 55.5 -33 46 00 P06 Ea 100 nwB04 PHR0758-4243 257.8-06.9 T 07 58 26.3 -42 43 53 P06 R 25 nmA09 M 3-2 240.3-07.6 T 07 14 49.8 -27 50 23 A92 B 11 ywB03 A 23 249.3-05.4 T 07 43 18.0 -34 45 16 A92 R 65 ywB10 NGC 2452 243.3-01.0 T 07 47 26.3 -27 20 07 A92 A 15 ywB12 M 3-4 241.0+02.3 T 07 55 11.4 -23 38 13 A92 - 14 n c wB09 PHR0803-3331 250.4-01.3 T 08 03 12.5 -33 31 02 P06 B 64 nwB05 KeWe 4 257.8-05.4 T 08 05 33.7 -41 56 51 K02 R? 45 nwA08 PHR1833-2632 007.2-08.1 T 18 33 21.5 -26 32 28 P06 Rrs 28 nwA02 PHR1835-2751 006.2-09.1 T 18 35 44.6 -27 51 21 P06 Eam 22 ywA03 PHR1837-2827 005.9-09.8 T 18 37 54.6 -28 27 31 P06 Eas 41 nwA07 PHR1841-2716 007.3-10.1 T 18 41 55.0 -27 16 42 P06 Eas 14 nwA09 PHR1848-1829 016.0-07.6 T 18 48 11.3 -18 29 43 P06 Ems 19 ywA12 PHR1849-1952 014.8-08.4 T 18 49 24.2 -19 52 14 P06 Es 18 ywA04 PHR1852-2749 007.9-12.5 T 18 52 51.5 -27 49 05 P06 R 22 nwA10 PHR1857-1750 017.5-09.2 T 18 57 16.8 -17 50 53 P06 Eas 11 nwA06 Hf 2-2 005.1-08.9 T 18 32 31.0 -28 43 21 A92 - 22 ywA05 He 2-418 004.7-11.8 T 18 44 14.6 -30 19 37 A92 E 11 ywA11 A 51 017.6-10.2 T 19 01 01.6 -18 12 13 A92 R 59 y a All PNe from A92 and K02 were designated as true (T). b References for morphological classification of non-MASH PNe: KeWe 2 and KeWe 4: Kerber et al. (1998); M 3-2: Perinotto & Corradi (1998a); A 23and NGC 2452: Rauch et al. (1999); A 51: Stanghellini et al. (1993); He 2-418: Ruffle et al. (2004) c The wB12 field is excluded from the reduction process due to the insufficient visibility data (see text).
PN mimics (Frew & Parker 2010; Frew et al. 2010) were in-cluded in the sample. The selection criterion, based on an-gular diameter, was weighted toward objects with θ opt < ′′ which is approximately the size of the sky-projected shortest baseline in the EW352 ATCA configu-ration at 6 cm. A group of 10 known PNe catalogued inAcker et al. (1992) and Kohoutek & K¨uhl (2002) was ob-served along with the selected MASH sample. In order toavoid large interruptions in observation, the main selectioncriterion for this group was based on their angular proximityto the previously selected MASH sample.Selected objects were observed with the ATCA on the12th, 13th and 14th of May 2003. Fields observed on the12th and 13th are designated in the target listings with apreceding mA and wA , respectively, and fields observed onthe third day (14th) has been designated with a preceding wB . We present field names, positions, general propertiesand detection flags (see §
3) for PNe in the observed pilotsample in Table 1. Observations were conducted at 6 cm and 3 cm, using the EW352 configuration in snap-shot mode andwith a moderate total integration time on each source. Theprimary calibrator used for the preliminary antennae cali-bration and for the gain calibration was always PKS 1934-638 with an adopted flux of 2.842 Jy at 6 cm and 5.829 Jyat 3 cm. The summarised observational parameters for thispilot ATCA study are presented in Table 2.Unfortunately, for all of the observed objects, data fromantennae CA01 and CA02 (the shortest baseline) have beencorrupted at the beginning and at the end of the observingruns because of the shadowing effect . Corrupted visibili-ties were flagged-out prior to the calibration process. Afterthe initial flagging process the data-lost was approximately15-40 per cent of the total integration time in the shortestbaseline and a significant part of correlations with the CA01antenna. The wB10 field is left with only one uv cut in theshortest baseline visibility data, and the wB12 field with no For a full explanation of the shadowing effect refer toSubrahmanyan & Deshpande (2004). c (cid:13) stron. Nachr. / AN (2006) 791 Table 3
Radio parameters of the ATCA radio-detected PNe. (1) (2) (3) (4) (5) (6) (7)Name S cm S cm S NV SS θ cm θ cm θ opt [mJy] [mJy] [mJy] [arcsec] [arcsec] [arcsec]PHR0731-2439 3.5 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
15 69 ±
15 59
Table 2
ATCA observational parameters for pilot study.
Observing wavelength [cm] 3 6Observing frequency [MHz] 8640 4800Bandwidth [MHz] 128 128ATCA configuration EW352 EW352Typical size of the synthesisedBeam [arcsec] 50 ×
20 100 × a ) 35 (15 a )Typical rms noise [mJy/Beam] ∼ . ∼ . Prim. cal. flux [Jy]:PKS 1934-638 5.829 2.842Sec. cal. defect:0646-306 (mA fields) 1% 1%0736-332 (wA fields) 3% 1%1933-400 (wB fields) 1% 1% a Day three. shortest baseline visibility data at all and with only two uv cuts in the longer baselines. The wB10 field fortunately con-tained a strong and relatively compact ( θ < ′′ ) programsource. We argue that the loss of large structure informationshould not have a crucial effect on the final result. Due to theinsufficient visibility data, the wB12 field is completely ex-cluded from further analysis. In the wB02 field strong inter-ference in the CA01-CA02 correlation at 3 cm was noticed(unrelated to the shadowing effect) . Since this interferencecould not be properly flagged, this part of the data set wasalso excluded from further processing. The program sourceA 23, observed in the wB02 field, has an optically deter-mined angular size of 65 ′′ . Thus, we anticipated that the3 cm flux density could be underestimated due to a lack ofthe flagged correlations (short baseline) data.Further processing (calibration, deconvolution and pa-rameterisation) was performed using standard MIRIAD pro-cedures (Sault & Killen 2008) with no attempt to employa self-calibration due to low signal to noise ratio in the ob-served fields. The “dirty” images were created using the nat-ural weighting scheme and excluding antenna 6 (at 6 km).Detected objects were parameterised using the IMSADtask. Measured integrated flux densities are tabulated in Ta-ble 3 (columns 2 and 3). If the IMSAD task reported a proper deconvolution the angular diameters were calculatedusing the Gaussian deconvolution method as described invan Hoof (2000). The derived angular diameters from 6 cmand 3 cm maps are presented in columns (5) and (6) respec-tively. We note that because of the small declinations andincomplete (and non uniform) uv coverage the synthesisedbeams have strong eccentricities ranging from 0.87 to 0.97.Therefore, we used only a minor deconvolved diameter forthe calculation of angular diameters.We expected the influence of several systematic effectsto affect the accuracy of derived parameters due to the lim-ited integration time per field, a considerable loss of theshort baseline information, the strong elongation of the syn-thesised beam and the intrinsic low brightness of observedobjects. This pilot study gave us an excellent opportunity tomeasure the observational limits of MASH PN detectabil-ity, the extent of uncertainties and a chance to examine theobservational components which have to be improved in thefollowing full-scale experiment. We detected 11 of the 26 observed objects in our radio maps(25 if we exclude the wB12 field which was not fully pro-cessed). Six detected objects are associated with PNe po-sitions from the Acker et al. (1992) and Kohoutek & K¨uhl(2002) catalogues and five from the MASH I catalogue, withone object from the MASH catalogue positively detectedonly at 6 cm. That give us a detection quality strongly infavour of known PNe with around 60% detected, against30% of detected MASH PNe. This two-to-one detection ra-tio between the known and MASH samples will not changesignificantly even if we exclude three PNe with optical sizessignificantly larger than the synthesised beam (PHR0724-2021, PHR0724-1757, and PHR0755-3346).The only object which is detected at only one frequency(6 cm) is PHR0732-2825. The 6 cm detection is at the 5 σ threshold level and it is not a surprise that no emission isvisible at 3 cm. Even though possible non-thermal emis-sion could produce such an effect, due to its distinct bipolarmorphology we have no doubt that the object in question is c (cid:13)(cid:13)
20 100 × a ) 35 (15 a )Typical rms noise [mJy/Beam] ∼ . ∼ . Prim. cal. flux [Jy]:PKS 1934-638 5.829 2.842Sec. cal. defect:0646-306 (mA fields) 1% 1%0736-332 (wA fields) 3% 1%1933-400 (wB fields) 1% 1% a Day three. shortest baseline visibility data at all and with only two uv cuts in the longer baselines. The wB10 field fortunately con-tained a strong and relatively compact ( θ < ′′ ) programsource. We argue that the loss of large structure informationshould not have a crucial effect on the final result. Due to theinsufficient visibility data, the wB12 field is completely ex-cluded from further analysis. In the wB02 field strong inter-ference in the CA01-CA02 correlation at 3 cm was noticed(unrelated to the shadowing effect) . Since this interferencecould not be properly flagged, this part of the data set wasalso excluded from further processing. The program sourceA 23, observed in the wB02 field, has an optically deter-mined angular size of 65 ′′ . Thus, we anticipated that the3 cm flux density could be underestimated due to a lack ofthe flagged correlations (short baseline) data.Further processing (calibration, deconvolution and pa-rameterisation) was performed using standard MIRIAD pro-cedures (Sault & Killen 2008) with no attempt to employa self-calibration due to low signal to noise ratio in the ob-served fields. The “dirty” images were created using the nat-ural weighting scheme and excluding antenna 6 (at 6 km).Detected objects were parameterised using the IMSADtask. Measured integrated flux densities are tabulated in Ta-ble 3 (columns 2 and 3). If the IMSAD task reported a proper deconvolution the angular diameters were calculatedusing the Gaussian deconvolution method as described invan Hoof (2000). The derived angular diameters from 6 cmand 3 cm maps are presented in columns (5) and (6) respec-tively. We note that because of the small declinations andincomplete (and non uniform) uv coverage the synthesisedbeams have strong eccentricities ranging from 0.87 to 0.97.Therefore, we used only a minor deconvolved diameter forthe calculation of angular diameters.We expected the influence of several systematic effectsto affect the accuracy of derived parameters due to the lim-ited integration time per field, a considerable loss of theshort baseline information, the strong elongation of the syn-thesised beam and the intrinsic low brightness of observedobjects. This pilot study gave us an excellent opportunity tomeasure the observational limits of MASH PN detectabil-ity, the extent of uncertainties and a chance to examine theobservational components which have to be improved in thefollowing full-scale experiment. We detected 11 of the 26 observed objects in our radio maps(25 if we exclude the wB12 field which was not fully pro-cessed). Six detected objects are associated with PNe po-sitions from the Acker et al. (1992) and Kohoutek & K¨uhl(2002) catalogues and five from the MASH I catalogue, withone object from the MASH catalogue positively detectedonly at 6 cm. That give us a detection quality strongly infavour of known PNe with around 60% detected, against30% of detected MASH PNe. This two-to-one detection ra-tio between the known and MASH samples will not changesignificantly even if we exclude three PNe with optical sizessignificantly larger than the synthesised beam (PHR0724-2021, PHR0724-1757, and PHR0755-3346).The only object which is detected at only one frequency(6 cm) is PHR0732-2825. The 6 cm detection is at the 5 σ threshold level and it is not a surprise that no emission isvisible at 3 cm. Even though possible non-thermal emis-sion could produce such an effect, due to its distinct bipolarmorphology we have no doubt that the object in question is c (cid:13)(cid:13)
92 I. S. Bojiˇci´c et al.: A pilot study of the radio-continuum emission from MASH PNe
Fig. 1
The radio-continuum SED plots of PNe from the pilot sample. Filled circles represent the integrated flux densitiesmeasured from the new ATCA radio images (this paper). Other values are from: MW79 (Milne & Webster 1979); MA82(Milne & Aller 1982); PMN (Griffith et al. 1994); NVSS (Condon et al. 1998). The straight line represent the best fit ofthe power-law function ( S ( ν ) ∝ ν α ) to all data-points from the obtained empirical distribution. In the case of PN A 51the power law function is also fitted to a sub-set of flux densities (ATCA and NVSS; dashed line). The value of a spectralindex obtained from the best fit is given in the lower left corner.a genuine PN and that the missing 3 cm detection is solelyproduced by the insufficiently sensitive observation.Assuming that the intrinsic properties of this prelimi-nary MASH sample conform to the general characteristicsof the full MASH sample it is clear that it can be expectedthat most MASH PNe would be below or at the edge of sen-sitivity of a similar quick ATCA survey (of 1 mJy zero levelnoise). Obviously, faint and extended ( θ > ′′ ) PNe arenot detectable using this observational configuration. Evenwith much larger integration times, the measured fluxes willbe strongly affected by the missing flux effect. With a presumption of free-free emission as the primary ra-dio emission mechanism in PNe (Osterbrock 1989; Pottasch1984), a construction of the SED plots can be used as acrude test of the reliability of our results.First, a thorough literature search for published radiodata of the selected sample was undertaken. We collectedall radio data which can be used to cross-correlate with theobtained parameters. We found flux densities at 6 cm forM 3-2, A 23, NGC 2452 and A 51 catalogued in Cahn et al.(1992). However, the only genuine 6 cm flux density ob-servation is for NGC 2452 (55 ± β flux fromKaler et al. (1990). We also found a 5 GHz flux density(65 ±
11 mJy) for NGC 2452 detected in the PMN Tropicalsurvey (Griffith et al. 1994), which is in excellent agreementwith the Milne & Aller (1975) value. From Milne & Aller(1982) we found flux measurements at 14.7 GHz for A 23(6 ± ± ± ±
25 mJy(Calabretta 1982) and at 2.7 GHz we found 60 ±
10 mJy(Milne & Webster 1979) for NGC 2452. Finally, the NRAOVLA Sky Survey (NVSS: Condon et al. 1998) covers thispart of the sky at 1.4 GHz. Flux densities from NVSS, for8 detected PN, are tabulated in column (4) in Table 3. Pub-lished data, together with the data from our observations,were used for the construction of SEDs for 6 known and 2MASH PNe (Fig. 1). All plots were produced only for ob-jects where the independent data were available.Except in the case of the A 51, our measurements arein a good agreement with that expected for optically thinnebulae. The spectra observed in A 51 appear to have a sig-nature of two component emission or some additional emis-sion mechanism at frequencies above 10 GHz. We examinedthis PN in more detail in section 7. c (cid:13) stron. Nachr. / AN (2006) 793 Table 4
Analysis of the extinction coefficients calculated from different methods.
Name log F(H α ) E ( B − V ) E ( B − V ) cm E ( B − V ) cm M 3-2 -12.11 (1) (2) (3) (1) (4) (5) (4) (6) (0.22) (7) (4) (3) (0.65) (7) (8) (8) - -NGC 2452 -10.81 (1) (9) α ) (10) E ( B − V ) (11) E ( B − V ) cm E ( B − V ) cm PHR0724-2021 -12.47 - - -PHR0732-2825 -12.67 0.51 0.53 -PHR0726-2858 -12.76 0.26 - -PHR1835-2751 -13.05 - 1.26 -PHR1837-2827 -13.18 - - -PHR1841-2716 -13.72 - - -PHR1848-1829 -12.67 - 0.88 1.05PHR1849-1952 -12.50 - 0.98 -PHR1857-1750 -13.22 - - -PHR0758-4243 -13.50 0.77 - -PHR0745-3535 -12.38 - - -PHR0755-3346 -12.07 0.65 - -PHR0803-3331 -12.32 0.50 - -PHR0731-2439 -12.53 1.19 1.00 1.03References: (1)
Shaw & Kaler (1989); (2)
K¨oppen et al. (1991); (3)
Ruffle et al. (2004); (4)
Frew et al. (2011, in preparation); (5)
Acker et al. (1991); (6)
Kaler (1983); (7)
Frew (2008); (8)
Kerber et al. (1998); (9)
Pe˜na et al. (1998); (10)
Gunawardahana et al. (2011, in preparation); (11) this work. α surface brightness and interstellarextinction We found published H α and/or H β fluxes and extinctioncoefficients for seven known PNe from this sample. All butone of these (KeWe 4) have been positively detected in ourtotal intensity radio images. For the MASH part of the sam-ple the integrated H α fluxes were obtained from intensitycalibrated SHS images. A set of integrated H α fluxes for 14MASH PNe were taken from Gunawardahana et al. (PASA,in preparation). The integrated fluxes, not corrected for red-dening, are presented in Table 4. The colour excess E ( B − V ) is estimated for five objects from our MASH samplewith available flux-calibrated spectra using the standardBalmer decrement method. The flux in the Balmer lines (H α and H β ) have been fitted with Gaussians using the IRAF task SPLOT . For PHR0755-3346 the value for E ( B − V ) iscalculated from the c Hβ tabulated in Frew (2008).It is expected that radio-continuum and Balmer linesemission will be well correlated, due to the same depen-dence on the nebular density, except in the case of opticallythick radio-continuum emission. In the range of used fre-quencies (1-10 GHz), and assuming canonical electron tem-perature ( T e ∼ = 10 K) and density ratios ( n ( He + ) /n ( H + ) = 0.11, and n ( He ) /n ( H + ) = 0.013 the E ( B − V ) canbe calculated from the measured radio flux F ν at frequency IRAF is distributed by the National Optical Astronomy Observato-ries, which is operated by the Association of Universities for Research inAstronomy, Inc. (AURA) under cooperative agreement with the NationalScience Foundation ν and the measured and reddening-corrected H α flux using(Pottasch 1984): (cid:18) F ν mJy (cid:19) = (cid:18) F ( Hα ) erg cm − s − (cid:19) . × T . e ν − . (1)where the standard theoretical ratio of H α /H β =2.85 is as-sumed. We list the extinction coefficients found from thismethod in the last two columns of Table 4. Further parameterisation of the ATCA detected objects re-quires knowledge of reliable distances. For four previouslyknown PNe (M 3-2, A 51, A 23 and NGC 2452) distances,calculated using some of the statistical distance scalemethod, have been published in Cahn et al. (1992); Kerber et al.(1998); Phillips (2004).In this study, statistical distance scale methods fromPhillips (2004), Stanghellini et al. (2008) and (Frew 2008,F08) has been used to calculate distances to the observedPNe. For the F08 scale we calculated the H α surface bright-ness from the integrated H α fluxes corrected for reddening.Optically determined angular sizes are used throughout ourcalculations as they were considered more reliable. The de-rived distances ( D P /kpc) and radii ( R P /pc) are tabulated inTable 5.Considering an optically thin, ionised region of gas withapproximation of constant electron temperature ( T e /K), the c (cid:13)
94 I. S. Bojiˇci´c et al.: A pilot study of the radio-continuum emission from MASH PNe
Table 5
Derived parameters.
Name log ( T b ) D P D S D F D adopt R n e M ion (kpc) (kpc) (kpc) (kpc) (pc) (cm − ) (M ⊙ )PHR0726-2858 - - - 8.8 8.8 0.68 < < < < < < < < electron density ( n e / cm − ) and ionised mass ( M ion / M ⊙ )can be calculated from (Gathier 1987): n e = ζ × S cm T e D − θ − ε − × (cid:18) y + 3 xy y + xy (cid:19) − , (2)where ζ = 4 . × and:M ion = 1 . × − n e D θ ε y y + xy , (3)where S cm is the 6 cm flux density (mJy), D is the distance(kpc), θ is angular diameter (arcsec) of the source, y is theHe abundance, x is the fraction of He II and ε is the volumefilling factor defined as the ratio of the electron density ( n e )averaged over the volume to the electron density averagedover the mass (Daub 1982). We assumed an electron tem-perature of K, He/H ratio of 0.11, average filling factorof 0.35 (Boffi & Stanghellini 1994), and that half of the Heatoms are doubly ionised. The distance adopted for calcu-lation of n e and M ion is the average of all obtained valuesexcept when one of the used methods shows a clear devia-tion from the mean value (in which case we used an averageof the similar diameters). The estimated n e and M ion aretabulated in Table 5.We compare derived parameters with parameters for thewell known Helix Nebula (NGC 7293). The Helix Neb-ula is one of the best studied large, nearby and low surfacebrightness PN with a reliable distance estimate, based on thetrigonometric parallaxes method, of only 216 pc(Benedict et al. 2009). The angular radii (seen in the radio-continuum) is ∼ ∼ . M ⊙ . However, Rodr´ıguez et al. (2002)argue that the filling factor approximation of ε = 0 . is too large for the Helix and that a more realistic value forthe ionised mass is ∼ . M ⊙ (with ε = 0 . ) givenby Boffi & Stanghellini (1994). The electron density in thisnebula varies from 30 to 120 cm − (Henry et al. 1999) witha mean value in the main torus of about 60 cm − (O’Dell1998). Scaling down the flux at 5 GHz and the angular di-ameter of 10 arcmin to the mean value of distances to ourdetected objects of 5 kpc (see Table 5) will give S cm =2 . mJy and angular diameter of θ = 30 arcsec. Usingthe same filling factor as used in our sample ( ε = 0 . )we calculate an electron density and ionised mass of n e ≈ cm − and M ion ≈ . M ⊙ respectively. Mean valuesfor electron density, ionised mass and radius found from thedetected portion of the pilot sample (excluding NGC 2452)are 200 cm − , 0.15 M ⊙ and 0.3 pc, respectively. Clearly,the average physical properties of PNe from our pilot sam-ple are very similar to those of the Helix Nebula. In this section, we list individual notes on observed objectsfor which measured, derived or previously catalogued pa-rameters are ambiguous or in disagreement.
Hf 2-2 is a PN with a variable, close binary central star(De Marco et al. 2008; Lutz et al. 2010). Liu et al. (2006)presented an extreme abundance discrepancy factor (ADF)of about 70 for this nebula (the typical ADF is ∼ for mostPNe). The ADF is a quantification of the abundance discrep-ancy, found in PNe and HII regions, between abundancesdetermined from optical recombination lines and thosefound from collisionally excited lines (Liu 2006). A possi-ble explanation for this discrepancy could be strong temper-ature gradients within the nebula (Peimbert 1967), or inner,hydrogen deficient clumps embedded in the diffuse nebula(Zhang et al. 2009). The electron density determined from c (cid:13) stron. Nachr. / AN (2006) 795 the hydrogen recombination spectrum near the Balmer jumpregion of n e ≈
400 cm − (Zhang et al. 2004) is a factor oftwo larger than our radio-continuum determined value. This is a moderate excitation (EC = 5 ; Rauch et al. 1999),round PN with a distinct thin shell (Acker et al. 1992).Rauch et al. (1999) measured strong [O
III ] lines and did notdetect He II indicating that it might be only moderately opti-cally thin. A strong divergence from the radio flux expectedfrom the measured H β and constructed SED for this nebulaimply the possibility of mild self-absorption effects in theradio-continuum. Due to its large angular size and the smallintegration time (only three 5 min cuts or 15 min of total in-tegration time) we suggest that the integrated flux, reportedhere, should be taken as a lower limit. Another indicator thatthis object is not properly sampled is that the Gaussian fit-ting failed at 6 cm, while the estimated angular diameter at3 cm is twice as small as the one determined from opticalobservation. On the other hand, the extinction coefficient es-timated from the Balmer line ratio (Rauch et al. 1999) couldbe overestimated due to the possible internal absorption.Also, a deviation from the optical diameter could suggestthat the majority of the measured radio emission is actuallyproduced in some smaller structure. In order to positivelydistinguish between possible situations high resolution andhigh sensitivity radio observations are needed. A PN ionised by a faint but extremely hot ( ∼ × K)Wolf-Rayet central star (Pe˜na et al. 2001). The nebula con-sists of high gas density knots embedded in a diffuse body.NGC 2452 is the brightest radio object in this sample andwith most independent flux measurements. Our 5 GHz fluxappear to be in better agreement with the predicted opti-cally thin SED than measurements from the PMN survey.As stated above, the excluded shortest baselines at 3 cm im-ply that the integrated flux from this wavelength must beconsidered as a lower limit. The observed flat SED downto ∼ GHz is in contradiction to an extremely high elec-tron density of ∼ × cm − found by Feibelman (1999)from the [Ne IV ] diagnostic. On the other hand electron den-sity determined from the [S II ] doublet diagnostic( n e ≈ . × cm − ) from Pe˜na et al. (1998) are in ex-cellent agreement with our estimate ( . × cm − ). This MASH PN has an optical diameter of 28 ′′ but our radioimage shows extended structure above 2 σ larger than 1 ′ anda radio peak above 5 σ placed ∼ ′′ from the centroid ofthe nebula. In order to determine if this extended structure isreal or a possible artefact from a strong source placed ≈ ′ from the field centre, the 6 cm contour image is comparedwith the 1.4 GHz NVSS total intensity map. As can be seen Fig. 2
NVSS total intensity map centered on the positionof possible detected PN PHR1833-2632. Overlaid are con-tours from our 6 cm image. The contour levels are: -2, 2,3, 4, 5 and 6 × − . The sidebar quantifies thepixel map and its units are Jy beam − .the extended structure in Fig. 2, which is visible in our 6 cmmap, is correlated with a similar extended emission which isat the edge of the adopted zero level flux for the NVSS. Thisgives confidence that the observed emission is real. How-ever, the question if this emission is actually correlated tothe much smaller PN still stands open. The flux density found from a 14.7 GHz Parkes observation(Milne & Aller 1982) is almost three times larger than boththe NVSS and our ATCA flux densities. The spectral energydistribution, constructed from all four measurements implythat the bulk of the radio emission, up to 14.7 GHz, is com-ing from an optically thick environment. However, the cal-culated brightness temperature at 1.4 GHz of T b ≈ K andexcellent agreement with the flux predicted from H β (fromKaler (1983)) imply optically thin free-free emission at fre-quencies ν > GHz. One can argue that both NVSS andATCA flux measurements could be affected by the miss-ing flux problem due to the relatively large angular size ofthis PN ( θ opt ≈ ′′ ). However, our radio determined an-gular diameters, at both 3 and 6 cm (62 ± ′′ and 69 ± ′′ ,respectively), are in excellent agreement with the opticalvalue which strongly imply that ATCA properly sampledthis PN. Also, as stated in Condon et al. (1998), the VLA,in the used configurations, starts to be insensitive for struc-tures larger than several arcminutes. This puts A 51 well be-low the NVSS filtering limit. On the other hand, the Parkesbeamwidth at this frequency is ∼ ′ . Except for a few am-biguous field detections of order of 1-2 mJy we did notfound any additional and strong source in the ∼ ′ and ∼ ′ (at 6 cm and 3 cm respectively) radii from the observed PN.This allows us to conclude that the measured 14.7 GHz, if c (cid:13)
96 I. S. Bojiˇci´c et al.: A pilot study of the radio-continuum emission from MASH PNe not affected by some observational and/or systematic effect,flux should originate solely from this object. Consequentlywe believe that the previously published 14.7 GHz flux den-sity of 26 mJy is greatly overestimated and that the morerealistic value for the flux density at this frequency (roughlyestimated from the constructed, optically thin,radio-continuum SED) is ∼ mJy. A deeper study of A 51is required to resolve its intrinsic density structure. Using the ATCA radio telescope a representative sample offaint and extended Galactic PNe have been observed. Thesample consists of 10 previously known and 17 newly cat-alogued PNe from the MASH I catalogue. Some 11 objectsfrom the observed sample have been successfully detectedand parameterised. For nine partially resolved PNe we de-termined radio angular diameters. For six PNe these are ingood agreement with optically determined values.We examined SEDs in the cm range for eight radio-detected PNe from our sample for which we found other in-dependent radio-continuum observations. All PNe from thissub-sample appear to emit in the radio optically thin regimewhich imply a more diffuse ionised medium is present.Except for one object, all detected PNe have very lowradio surface brightness. We use several statistical distancescale methods to calculate distances, electron densities andionised masses for the detected PNe. Except for NGC 2452,all of PNe from this sample are found to be moderately large( > − to1640 cm − with a median of 180 cm − ). The ionisedmasses are found to be concentrated in a relatively narrowrange around 0.15 M ⊙ .The results presented, together with a review of previousradio detections of MASH PNe (Bojiˇci´c et al. 2010), giveus an insight into the general radio-continuum properties ofthis new MASH sample. From the derived physical prop-erties we can see that detected MASH PNe are not signifi-cantly different from the known PNe sub-sample. However,it appears that, unsurprisingly, MASH PNe are placed at thefaint end of the radio luminosity distribution.The detection rate of is rather poor. All future radiosurveys of MASH PNe should involve significantly deeperobservations with better uv coverage in order to improve thedetection rate and quality of the derived parameters. Verycareful planning of observations is necessary to reconcilelimited observational time and the size of the sample whichcan properly represent the full catalogue. Using experiencegained from this pilot study, we examined the methods andtechniques which would improve our future observations.These upcoming observations will form the core of our newdetailed investigation of the radio-continuum properties ofMASH PNe (Bojiˇci´c et al. in preparation). References
Acker A., Marcout J., Ochsenbein F., Stenholm B., TylendaR.: 1992, Strasbourg - ESO catalogue of galactic plane-tary nebulae. Part 1; Part 2. Garching: European SouthernObservatoryAcker A., Raytchev B., Koeppen J., Stenholm B.: 1991,A&AS, 89, 237Benedict, G. F., et al. 2009, AJ , 138, 1969Boffi F. R., Stanghellini L.: 1994, A&A, 284, 248Bojiˇci´c, I. S., Parker, Q. A., Filipovi´c, M. D., & Frew, D. J.2011, MNRAS , 412, 223Cahn J. H., Kaler J. B., Stanghellini L.: 1992, A&AS, 94,399Calabretta M. R.: 1982, MNRAS, 199, 141Condon J. J., Cotton W. D., Greisen E. W., Yin Q. F., PerleyR. A., Taylor G. B., Broderick J. J.: 1998, AJ, 115, 1693Costa R. D. D., Uchida M. M. M., Maciel W. J.: 2004, A&A,423, 199Daub C. T.: 1982, ApJ, 260, 612De Marco O., Hillwig T. C., Smith A. J.: 2008, AJ, 136, 323Feibelman W. A.: 1999, ApJ, 525, 863Frew D. J.: 2008, PhD thesis, Macquarie UniversityFrew, D. J., & Parker, Q. A. 2010, PASA, 27, 129Frew, D. J., Madsen, G. J., O’Toole, S. J., & Parker, Q. A.:2010, PASA, 27, 203Gathier R.: 1987, A&AS, 71, 245Gaustad J. E., McCullough P. R., Rosing W., Van Buren D.:2001, PASP, 113, 1326Griffith M. R., Wright A. E., Burke B. F., Ekers R. D.: 1994,ApJS, 90, 179Henry R. B. C., Kwitter K. B., Dufour R. J.: 1999, ApJ, 517,782Kaler J. B.: 1983, ApJ, 271, 188Kaler J. B., Shaw R. A., Kwitter K. B.: 1990, ApJ, 359, 392Kerber F., Roth M., Manchado A., Groebner H.: 1998,A&AS, 130, 501Kohoutek L., K¨uhl D.: 2002, AN, 323, 484K¨oppen J., Acker A., Stenholm B.: 1991, A&A, 248, 197Liu, X.-W. 2006, Planetary Nebulae in our Galaxy and Be-yond, 234, 219Liu X.-W., Barlow M. J., Zhang Y., Bastin R. J., Storey P. J.:2006, MNRAS, 368, 1959Lutz, J., Fraser, O., McKeever, J., & Tugaga, D.: 2010,PASP, 122, 524Milne D. K., Aller L. H.: 1975, A&A, 38, 183Milne D. K., Aller L. H.: 1982, A&AS, 50, 209Milne D. K., Webster B. L.: 1979, A&AS, 36, 169Miszalski B., Parker Q. A., Acker A., Birkby J. L., FrewD. J., Kovacevic A.: 2008, MNRAS, 384, 525O’Dell C. R.: 1998, AJ, 116, 1346Osterbrock, D. E. 1989, Research supported by the Univer-sity of California, John Simon Guggenheim MemorialFoundation, University of Minnesota, et al. Mill Valley,CA, University Science Books, 1989, 422 p.,Parker Q. A., Acker A., Frew D. J., et al.: 2006, MNRAS,373, 79 c (cid:13) stron. Nachr. / AN (2006) 797 Parker Q. A., Phillipps S., Pierce M. J., et al.: 2005, MN-RAS, 362, 689Peimbert M.: 1967, ApJ, 150, 825Pe˜na M., Stasi´nska G., Esteban C., Koesterke L., Medina S.,Kingsburgh R.: 1998, A&A, 337, 866Pe˜na M., Stasi´nska G., Medina S.: 2001, A&A, 367, 983Perinotto M., Corradi R. L. M.: 1998a, A&A, 332, 721Perinotto M., Corradi R. L. M.: 1998b, A&A, 332, 721Phillips J. P.: 2004, MNRAS, 353, 589Pottasch S. R.: 1984, Planetary nebulae - A study of latestages of stellar evolution. Vol. 107 of Astrophysics andSpace Science Library, D. ReidelRauch T., K¨oppen J., Napiwotzki R., Werner K.: 1999,A&A, 347, 169Rodr´ıguez L. F., Goss W. M., Williams R.: 2002, ApJ, 574,179Ruffle P. M. E., Zijlstra A. A., Walsh J. R., Gray M. D.,Gesicki K., Minniti D., Comeron F.: 2004, MNRAS, 353,796Sault B., Killen N.: 2008, MIRIAD User Guide. Aus. Teles.Nat. Fac. (ATNF), AustraliaShaw R. A., Kaler J. B.: 1989, ApJS, 69, 495Stanghellini L., Corradi R. L. M., Schwarz H. E.: 1993,A&A, 279, 521Stanghellini L., Shaw R. A., Villaver E.: 2008, ApJ, 689,194Subrahmanyan, R., & Deshpande, A. A.: 2004, MNRAS,349, 1365van Hoof P. A. M.: 2000, MNRAS, 314, 99Young K., Cox P., Huggins P. J., Forveille T., Bachiller R.:1999, ApJ, 522, 387Zhang Y., Liu X.-W., Wesson R., Storey P. J., Liu Y.,Danziger I. J., 2004, MNRAS, 351, 935Zhang Y., Yuan H.-B., Hua C.-T., Liu X.-W., Nakashima J.,Kwok S., 2009, ApJ, 695, 488 c (cid:13)(cid:13)