1. Introduction
Planetary nebulae (PNe) are shells of gas that are shed during the terminal phase for the majority of stars between
${\sim}1-8$
M
$_\odot$
(Pottash Reference Pottash1983; Gathier, Pottasch, & Pel Reference Gathier, Pottasch and Pel1986), when they transition from the asymptotic giant branch (AGB) to the white dwarf stage (Kwok Reference Kwok2005). The transitioning star ionises the surrounding shells and these remain visible for thousands of years, providing crucial insight into the recent mass loss rate, mechanism and nucleosynthesis of the AGB star, as well as the evolution of the parent galaxy (particularly the chemical and star-forming evolution) (Kwok Reference Kwok2005; Gesicki, Zijlstra, & Miller Bertolami Reference Gesicki, Zijlstra and Miller Bertolami2018; Crawford Reference Crawford2015).
NGC 5189 is a well-known Galactic PN located in the southern constellation Musca at RA (J2000) 13:33:31.8 and DEC (J2000) –65:58:29.8, at a distance of
${\sim}$
1500 parsecs (Chornay & Walton Reference Chornay and Walton2021). It was discovered by Scottish astronomer James Dunlop in 1826 (Dunlop Reference Dunlop1828; Cozens, Walsh, & Orchiston Reference Cozens, Walsh and Orchiston2010) during observations from Parramatta, Australia, using a 9-inch reflector telescope.Footnote
a
However, it was not considered a PN until more recently by Evans & Thackeray (Reference Evans and Thackeray1950) (an optical study), which they describe as ‘clearly not an ordinary planetary’ and at the time they did not observe any central star. It was later determined to be a quadrupolar PN with multiple sets of ‘symmetrical condensations’ (ansae, or knots) by Sabin et al. (Reference Sabin, Vázquez, López, Garca-Daz and Ramos-Larios2012) (both an optical and infrared study), and first identified as a binary system by Manick et al. (Reference Manick, Miszalski and McBride2015), where the central Wolf–Rayet [WO1] star (Crowther, De Marco, & Barlow Reference Crowther, De Marco and Barlow1998) is in a binary orbit with potentially a main-sequence star or a white dwarf (Manick et al. Reference Manick, Miszalski and McBride2015).
NGC 5189 is renowned for its complex morphology and has been described by a number of authors (e.g. Sabin et al. Reference Sabin, Vázquez, López, Garca-Daz and Ramos-Larios2012; Danehkar et al. Reference Danehkar, Karovska, Maksym and Montez2018; Aller et al. Reference Aller, Lillo-Box, Jones, Miranda and Forteza2020) as having filamentary and knotty structures, which can be attributed to the central binary system. A subsequent study by Bear & Soker (Reference Bear and Soker2017), however, suggests the structure may result from a ternary (triple) progenitor system (Danehkar et al. Reference Danehkar, Karovska, Maksym and Montez2018). According to Phillips & Reay (Reference Phillips and Reay1983), the symmetrical condensations that NGC 5189 distinctly exhibits are the result of precessional torque induced by the companion star as it orbits the central star.
Table 1. We present the available flux density measurements for radio-detections of NGC 5189 (Column 4), in chronological order.
$^{\ast}$
Year of catalogue release, excepting ASKAP–EMU, which are years of observation.
$^{\dagger}$
Average integrated flux density measurements, based on the average of 8 or 9 scans of Infinity, conducted by the Parkes telescope (8 scans for observations at 2700 MHz and 9 scans at 1420 MHz).
$^{\ddagger}$
New integrated flux density measurements of existing radio data.
1.1. Radio Observations of PNe
Radio continuum observations of PNe are not only useful for measuring the integrated flux densities and the associated spectral index but also for tracing the thermal emission from the ionised plasma shells (Hajduk et al. Reference Hajduk2018) and determining the PN geometry and structures without the impediment of interstellar and atmospheric extinction (Zijlstra et al. Reference Zijlstra, Van Hoof, Chapman and Loup1994). While radio continuum surveys generally report the detection of weak thermal free–free emission from a PN (Asher et al. Reference Asher2024; Filipović et al. Reference Filipović2009; Gathier et al. Reference Gathier, Pottasch, Goss and van Gorkom1983; Milne & Aller Reference Milne and Aller1982; Pennock et al. Reference Pennock2021), a few studies have reported non-thermal detections in the early photoionisation phase, as AGB stars transition to the white dwarf stage (Bains et al. Reference Bains, Cohen, Chapman, Deacon and Redman2009; Cerrigone et al. Reference Cerrigone2017; Pérez-Sánchez et al. Reference Pérez-Sánchez, Vlemmings, Tafoya and Chapman2013; Suárez et al. Reference Suárez2015). A notable example is the Hajduk et al. (Reference Hajduk2024) study regarding the radio continuum emission from Sakurai’s Object (V4334 Sgr), the central star of a PN (CSPN) in the Sagittarius constellation. The authors observed the source between 2004 and 2023, finding non-thermal and highly variable emission to dominate between 2004 and 2017. This indicates Sakurai’s object is a proto-PNe (or pre-planetary nebulae [PPN], Kwok Reference Kwok2001) in the early photoionisation phase, of which the authors attribute the non-thermal emission to the shock interactions during mass-ejection events, as stellar winds from the AGB star interact with the PN shell (Hajduk et al. Reference Hajduk2024).
Slee & Orchiston (Reference Slee and Orchiston1965) were the first to observe NGC 5189 in the radio spectrum, at both 1420 and 2700 with the Parkes 64-m telescope. Subsequent Parkes radio detections (Table 1) were made between 1972 and 1982 at frequencies ranging from 2700 MHz to 14.7 GHz (Aller & Milne Reference Aller and Milne1972; Milne & Aller Reference Milne and Aller1975; Milne Reference Milne1979; Milne & Aller Reference Milne and Aller1982).
More recent detections were made in 1990 at 4850 MHz during the (PMN) radio continuum survey (Griffith & Wright Reference Griffith and Wright1993; Wright et al. Reference Wright, Griffith, Burke and Ekers1994; Condon, Griffith, & Wright Reference Condon, Griffith and Wright1993) (conducted by the Parkes telescope and using the National Radio Astronomy Observatory (NRAO)-developed seven-beam receiver, Condon, Broderick, & Seielstad Reference Condon, Broderick and Seielstad1989) and in 2007 at 843 MHz during the second epoch Molonglo Galactic Plane Survey (MGPS-2) (Murphy et al. Reference Murphy2007) – the Galactic counterpart to the Sydney University Molonglo Sky Survey (SUMSS) (Bock, Large, & Sadler Reference Bock, Large and Sadler1999; Mauch et al. Reference Mauch2003) by the Molonglo Observational Synthesis Telescope (MOST) (Mills 1981; Robertson Reference Robertson1991).
We liken the appearance of the central region of NGC 5189 to the lemniscate symbol, and hence our use of the colloquial moniker, Infinity for this object. Given the dearth of radio continuum measurements of Infinity (hereafter used interchangeably with NGC 5189) in recent years our primary goal has been to observe it and contribute up-to-date measurements using the ASKAP radio telescope array. Therefore, in this paper we present the most recent radio continuum measurements for Infinity, observed at 943 MHz during the ASKAP Evolutionary Map of the Universe (EMU) survey.
The paper is structured as follows. In Section 2 we describe the ASKAP–EMU radio data and respective processes used to determine the new radio continuum measurements, in addition to detailing the corresponding HST data used for comparative purposes. In Section 3 we present the flux density, spectral index and associated results, and we summarise our findings in Section 4.
Figure 1.
Top: ASKAP–EMU radio image of Infinity at 943 MHz with an rms noise level of 39
$\unicode{x03BC}$
Jy beam
$^{-1}$
. The contours are at levels of 10, 50, 100, 175, and 200
$\sigma$
. The synthesised beam size is shown in the bottom left corner. Bottom: RGB image of Infinity from HST images, overlaid with ASKAP–EMU radio image contours. Red is the F673N filter (a narrow-band filter centred at 676.59 nm), green is the F606W filter (a wide V band filter centred at 588.92 nm), and blue is the F502N filter (a narrow-band filter centred at 500.96 nm). All images are from the WFC3 instrument.
2. DATA
2.1. ASKAP and the Evolutionary Map of the Universe Survey
The Evolutionary Map of the Universe survey (Norris et al. Reference Norris2011; Hopkins et al. Reference Hopkins2025) is a wide-field radio continuum galaxy survey being conducted by the ASKAP radio telescope (Johnston et al. Reference Johnston2008; Hotan et al. Reference Hotan2021), observing at the central frequency of 943 MHz and a bandwidth of 288 MHz (Hopkins et al. Reference Hopkins2025). The EMU survey (ASKAP project AS201) began in May 2023 and will cover the entire southern sky by the scheduled time of completion in 2028 (Hopkins et al. Reference Hopkins2025). The ASKAP array consists of 36 12-m antennas, with baselines up to 6 km, and has a large instantaneous field of view (ranging between 15 deg
$^{2}$
at 1700 MHz and 31 deg
$^{2}$
at 800 MHz) (Hotan et al. Reference Hotan2021).
Infinity has been observed twice during the EMU survey, with both observations using all 36 antennas over a duration of 10 h. The respective scheduling blocks are SB53310 (observed on 30 September 2023) and SB62225 (observed on 7 May 2024). Both tiles are accessible via the CSIRO ASKAP Science Data Archive (CASDA).Footnote b All EMU data is processed through the ASKAPsoft pipeline (Guzman et al. Reference Guzman2019).
Both images of Infinity (see Figure 1, top) have a resolution of 15.0
$\times$
15.0 arcsec
$^2$
, as well as a Stokes I image root mean squared (rms) noise value of 39
$\unicode{x03BC}$
Jy beam
$^{-1}$
. For comparison, the median Stokes I image rms noise value of each scheduling block (across the full tile) is 40
$\unicode{x03BC}$
Jy beam
$^{-1}$
(SB53310) and 44
$\unicode{x03BC}$
Jy beam
$^{-1}$
(SB62225), of which the sky area is 41.42 and 41.57 deg
$^{2}$
respectively.
The data were predominantly analysed using the CARTA (Comrie et al. Reference Comrie2021) software, to establish the integrated flux densities for Infinity. The flux density measurements are presented in Table 1, in addition to measurements for the Parkes, PMN, SUMSS and MGPS-2 radio observations. For the PMN observation, we relied upon the MIRIAD (Sault, Teuben, & Wright Reference Sault, Teuben and Wright1995; Sault, Teuben, & Wright Reference Sault, Teuben and Wright2011) and KARMA (Gooch Reference Gooch1996) software packages to measure the integrated flux density and uncertainty level, as a consequence of statistical issues experienced when scaling the image in CARTA (discussed further in Section 3).
Numerous flux density measurements have been achieved and investigated in a number of ASKAP–related papers, which we refer the interested reader to: including the Large Magellanic Cloud (LMC) Odd Radio Circle (ORC) J0624
$-$
6948 (Filipović et al. Reference Filipović2022; Sasaki et al. Reference Sasaki2025); the Galactic supernova remnant (SNR) G278.94+1.35 (named ‘Diprotodon’) by Filipović et al. (Reference Filipović2024); the discovery of a new, young, Galactic SNR G329.9
$-$
0.5 (‘Perun’) by Smeaton et al. (Reference Smeaton2024b); the detection of a large and low surface brightness Galactic SNR G288.8
$-$
6.3 (‘Ancora’) by (Filipović et al. Reference Filipović2023) (with a follow-up study by Burger-Scheidlin et al. Reference Burger-Scheidlin2024); and others (e.g. Lazarević et al. Reference Lazarević2024; Smeaton et al. Reference Smeaton2024a; Filipović et al. Reference Filipović2025).
2.2. HST Data
We also use the available HST data from the Hubble Legacy Archive (HLA)Footnote
c
to compare the optical and radio properties. The images are from project 12812 (PI: Z. Levay), taken on 6 July 2012 using the theWide Field Camera 3 (WFC3) instrumentFootnote
d
(Leckrone et al. Reference Leckrone1998). We make use of three images to create a composite RGB image (Figure 1, bottom), specifically the F673N (red), a narrow-band filter centred at 676.59 nm with a width of 11.78 nm (which covers the [Sii] doublet), F606W (green), a wide V band filter centred at 588.92 nm with a width of 218.92 nm (which covers the [Oiii] doublet [both lines] and the H
$\alpha$
line), and F502N (blue), a narrow-band filter centred at 500.96 nm with a width of 6.53 nm (which covers the red [Oiii] line). We use these images for comparison with radio, and an in-depth optical emission line analysis using the same data can be found in Danehkar et al. (Reference Danehkar, Karovska, Maksym and Montez2018).
3. Results and Discussion
Flux density measurements of Galactic PNe are too varied to derive a suitable expectation for a single PN, which is not surprising given the variation in PN distances (the exception to this issue are the Magellanic Clouds (MC) PNe, which remain unresolved, Pennock et al. Reference Pennock2021). Based on the variation, we contribute supplementary measurements of the spectral luminosityFootnote
e
at 943 MHz, and the distance-independent radio surface brightness,
$\Sigma$
, at 1 GHz (a standard scaling frequency for determining surface brightness values, e.g. Cotton et al. Reference Cotton2024) and determine if the value falls in the expected range for Galactic PNe.
We also explore a historical comparison of flux density values. Historically, the values for Infinity have been measured between 0.215 Jy at 4850 MHz (Griffith & Wright Reference Griffith and Wright1993; Wright et al. Reference Wright, Griffith, Burke and Ekers1994) and 0.459 Jy at 14700 MHz (Milne & Aller Reference Milne and Aller1982). Using CARTA and the polygon tool to define the outline of Infinity, we measured an integrated flux density of
$S_{\rm{943\,MHz}} =$
0.333 Jy (SB53310) and
$S_{\rm{943\,MHz}} =$
0.334 Jy (SB62225). The CARTA software does not specifically provide uncertainty levels, however the MIRIAD IMFIT task does, available from the MIRIAD (Sault et al. Reference Sault, Teuben and Wright1995, Reference Sault, Teuben and Wright2011) software package developed by the Australia Telescope National Facility (ATNF). Two separate assessments of the Infinity images indicate an uncertainty level of
${\sim}$
10
$\%$
for both images. The ASKAP measurements are presented in Table 1, in addition to the historical flux density values and their respective uncertainties.
In respect to the apparent size of Infinity, using CARTA and retaining the polygon outline, we measured the angular dimensions of
$3\rlap{.}^\prime4 \times 2\rlap{.}^\prime2$
. Using the Chornay & Walton (Reference Chornay and Walton2021) distance of
${\sim}$
1500 pc in our calculation, this corresponds to a physical size of 1.48 pc
$\times$
0.96 pc. Using the same distance of
${\sim}$
1500 pc, we measure the spectral luminosity,
$L_{\rm{943\,MHz}}$
= 8.89
$\times\,10^{13}$
W m
$^{-2}$
Hz
$^{-1}$
.
We also employ the mean of the angular dimensions to determine the surface brightness of Infinity at 1 GHz, of which we measure
$6.0\times10^{-21}$
W m
$^{-2}$
Hz
$^{-1}$
sr
$^{-1}$
. This value falls within the expected range of radio surface brightnesses for Galactic PNe (between
${\sim}10^{-24}$
–
$10^{-16}$
W m
$^{-2}$
Hz
$^{-1}$
sr
$^{-1}$
, Leverenz et al. Reference Leverenz2016), again placing confidence in the measurements.
Additionally, we measured the integrated flux densities of Infinity in the SUMSS and PMN surveys. The reason for measuring SUMSS is that the integrated flux density value for Infinity is not present in the respective catalogue. For the PMN observation, there are two different measurements presented in the catalogue (where Infinity is listed in the final source catalogue, ‘Table 2’ in Wright et al. (Reference Wright, Griffith, Burke and Ekers1994) and Table 2 in the online point source catalogue,Footnote
f
as per its J2000–derived source name, J1333–6558). The two measurements are
$S_{\rm{4850\,MHz}} =$
0.309
$\pm$
0.017 Jy and
$S_{\rm{4850\,MHz}} =$
0.215 Jy, respectively. Initially, we considered that the latter measurement may have resulted from measuring a steep spectral source, as the value is not in close agreement with the majority of historical observations. However, Wright et al. (Reference Wright, Griffith, Burke and Ekers1994) explain that a few complex objects were larger than the beam size and led to multiple measurements of those sources being included in Table 2 of the PMN source catalogue for the southern survey. The measurement of 0.215 Jy was determined using a ‘General-width fit’ (Griffith & Wright Reference Griffith and Wright1993), designed for detecting and measuring extended sources, where the maximum and minimum widths of a source are normalised to the beam size of
$4\rlap{.}^\prime2$
. This method may explain the discrepancy between the two measurements. Regardless, we choose to measure the PMN Infinity detection again.
Figure 2. A comparison of Infinity observations: the cyan ellipse in both the PMN and ASKAP–EMU images represents the region that was measured during the PMN survey (Griffith & Wright Reference Griffith and Wright1993; Wright et al. Reference Wright, Griffith, Burke and Ekers1994; Condon et al. Reference Condon, Griffith and Wright1993). The central pink polygon in both images represents the Infinity PN from the recent ASKAP observations.
We accessed the respective FITS files via SkyViewFootnote
g
and analysed accordingly in CARTA, assuming a 10
$\%$
uncertainty level as described and investigated in the aforementioned ASKAP–related studies (see Section 2.1). For the 843 MHz image in the SUMSS survey, we measured an integrated flux density of 0.245
$\pm$
0.025 Jy. We note that CARTA did not generate the integrated flux density directly (due to missing information in the image file header, most notably the restoring beam and frequency), but rather we used CARTA’s Sum flux value of the selected region, 4.644 Jy, to determine the integrated flux density using the following formula (Filipović Reference Filipović1996):
\begin{align} S_\nu = \frac{\textrm{Sum}}{1.133\left(\frac{BS}{PS}\right)^{2}}\,,\end{align}
where
$S_{\nu}$
is the integrated flux density at a given frequency, Sum is the flux value of the selected region (in Jy), BS is the beam size (for SUMSS this is 45
$^{\prime\prime}\times45^{\prime\prime}$
) and PS is pixel size (for SUMSS this is 11
$^{\prime\prime}\times11^{\prime\prime}$
).
For the 4850 MHz image in the PMN survey, limited header information in the image file prevented measurements using CARTA. Therefore, we employed MIRIAD IMFIT and determined a flux density value of 0.310
$\pm$
0.004 Jy, which is in close agreement with the value listed in the final source catalogue in Wright et al. (Reference Wright, Griffith, Burke and Ekers1994). The PMNFootnote
h
and SUMSS measurements are also presented in Table 1.
Importantly, when analysing the PMN Infinity observation in CARTA, we created an ellipse region to represent the boundary of the second measurement (as per the
$4\rlap{.}^\prime2$
beam size) and applied to both the ASKAP–EMU and PMN images. In Figure 2, the normalised PMN region of
$4\rlap{.}^\prime2\,\times\,4\rlap{.}^\prime2$
is indicated by the cyan ellipse in both images (to the left is PMN and to the right is ASKAP–EMU, where both intensity scales are measured in Jy/beam). For comparison, the central region observed by ASKAP is indicated by the pink polygon. In the ASKAP–EMU image, there are a number of background sources that fall within the PMN-catalogued region. Upon assessment of the four most distinct background sources (which have a combined integrated flux value of
${\sim}$
0.0024 Jy), it is reasonable to conclude that they have a negligible effect, despite being included in the measurement. Additionally, we would expect steep spectral sources observed at 4850 MHz to be fainter than observations at
${\sim}$
1 GHz. The reason the second measurement is lower is therefore unclear.
In Figure 3, we estimate the spectral index using 11 of the 12 radio data points listed in Table 1. Only one of the PMN measurements can be included in the calculation, to correspond with one observation, of which we employ the value listed in the final source catalogue (Wright et al. Reference Wright, Griffith, Burke and Ekers1994). We estimate a spectral indexFootnote
i
of
$\alpha$
= 0.12
$\pm0.05$
. Thermal free–free emission is typically associated with spectral indices of
$\alpha = 0 - 2$
and non-thermal with
$\alpha \lt$
–0.5 (Ridpath Reference Ridpath2018). Accordingly, we infer that NGC 5189 is a thermal (free–free) emitting nebula. This finding agrees with previous observations (e.g. Aller & Milne Reference Aller and Milne1972; Milne & Aller Reference Milne and Aller1982).
Figure 3. Using 11 of the 12 available radio data points (labelled according to the respective telescope and year of the associated paper or observation), we calculated the radio spectral index for Infinity and determined
$\alpha$
= 0.12
$\pm$
0.05, represented by the dashed orange line. Both axes are log scale. For the two ASKAP-EMU data points coloured blue, the survey bandwidth of 288 MHz has been marked with a horizontal line of the same colour.
Inspection of the data in Figure 3 shows that the three data points at lower frequencies (
$\lt$
1000 MHz in this instance, i.e. SUMSS, MGPS-2, ASKAP–EMU) indicate a small degree of brightening over a 21–yr period (from 0.245 Jy at 843 MHz [SUMSS] up to 0.334 Jy at 943 MHz [ASKAP–EMU]), although this is marginal. Based on the combination of age, differing flux scales and measurement techniques, the Parkes (1965a) and PMN (1994) data points are likely to be outliers; with emphasis on the differing flux density scales, as both sets of observations were determined using the radio galaxy Hydra-A (PKS B0915–118) to calibrate the absolute flux density scale (Slee & Orchiston Reference Slee and Orchiston1965; Condon et al. Reference Condon, Griffith and Wright1993). For comparison, ASKAP employs radio galaxy PKS B1934–638 to calibrate the absolute flux density scale (Hotan et al. Reference Hotan2021).
The data point at 14.7 GHz (Milne & Aller Reference Milne and Aller1982) is slightly higher than the others, yet in agreement with an earlier Parkes observation at 5 GHz (Milne Reference Milne1979), and certainly within the uncertainty level of the 1979 observation. Milne & Aller (Reference Milne and Aller1982) compared the 14.7 GHz and earlier 5 GHz flux density measurements (Milne Reference Milne1979) for 236 confirmed PNe (including Infinity) and found there to be notable scattering in the diagram. They mostly attribute this to the fixed error when measuring flux densities, in addition to the fractional error associated with calibration.
The remaining data points at higher frequencies (
$\ge$
2700 MHz) show more consistency with little indication of variation with time. There is a marginal drop in flux density between the 1965b and 1972 Parkes observations at 2700 MHz, from 0.360 Jy to 0.330 Jy, respectively, but a similarly sized marginal increase in flux density at 5000 MHz between the 1975 and 1979 Parkes observations, although in both cases these are within the estimated uncertainties. The gap between the earliest (Parkes) and most recent (ASKAP) observations is six decades, and any temporal variation in flux density or spectral index on such a timeframe for this class of object is likely to be small.
According to Hajduk et al. (Reference Hajduk2018), flux evolution is not only dependent on factors such as age (where, for example, electron densities will generally decline monotonically as the PNe expands, Zhang et al. Reference Zhang2004) but also the central star’s evolution, which vary in their progress and can contribute to variation in flux measurements.
Given the spectral index value of Infinity lies closer to the optically thin case of
$\alpha = -0.1$
than the optically thick at
$\alpha = 0.6-2.0$
(where more light is absorbed) (Taylor, Pottasch, & Zhang Reference Taylor, Pottasch and Zhang1987; Gruenwald & Aleman Reference Gruenwald and Aleman2007), we also infer that Infinity is predominantly within the optically thin regime for observations
$\gtrsim$
1000 MHz.
To explore this result further, we rewrite the Rayleigh-Jeans approximationFootnote
j
so we can measure the optical depth,
$\tau$
, which is proportional to the emission measure (EM) (defined as the integral of the electron density,
$N_{e}$
, along the line of sight in an emission nebula, Wilson et al. Reference Wilson, Rohlfs and Hüttemeister2013):
\begin{align} S_{\nu} = \frac{2kv^{2}}{c^{2}} \; T_{B}\Delta\Omega \,,\end{align}
where S
$_{\nu}$
is the integrated flux density at a specific frequency, k is the Boltzmann constant, c is the speed of light, T
$_{B}$
is the brightness temperature and
$\Delta\Omega$
is the angular area (measured as an elliptical area in this instance, equating to 4.971
$\times$
10
$^{-7}$
sr). We rearrange Equation (2) to solve for
$T_{B}$
:
\begin{align} T_{B} = \frac{S_{\nu}c^{2}}{2k\nu^{2}\Delta\Omega} \,.\end{align}
We measure
$T_{B} = $
24.6K. Brightness temperature can also be expressed as
$T_{B} = T(1 - e^{-\tau})$
, which we rearrange to solve for
$\tau$
:
\begin{align} \tau = -ln\left(1 - \frac{T_{B}}{T}\right)\,.\end{align}
We measure
$\tau = $
0.00246. Since
$\tau \ll 1$
, this reinforces the conclusion that Infinity is optically thin at 943 MHz. To determine the EM and
$N_{e}$
values, we employ the Mezger & Henderson (Reference Mezger and Henderson1967) approximation for the free-free opacity
$\tau$
:
\begin{equation}\tau = 3.28 \times 10^{-7} \left(\frac{T}{10^{4}K}\right)^{-1.35} \left(\frac{\nu}{GHz}\right)^{-2.1} \left(\frac{EM}{\text{pc cm}^{-6}}\right) \,,\end{equation}
where for T we use the canonical value (10
$^{4}$
K) (Bojicčić et al. 2021), and
$\nu$
is the frequency of the ASKAP–EMU survey in GHz (0.943). We then rearrange Equation (5) and solve for EM specifically:
\begin{align} EM = \frac{\tau}{3.28 \times 10^{-7} \left(\frac{T}{10^{4}K}\right)^{-1.35}\left(\frac{\nu}{GHz}\right)^{-2.1}}\,.\end{align}
We measure
$EM =$
6630 pc cm
$^{-6}$
. From the following EM formula (Wilson et al. Reference Wilson, Rohlfs and Hüttemeister2013) we can now rearrange and measure
$N_{e}$
:
\begin{align} EM = \int_{0}^{s}\left(\frac{N_{e}}{cm^{3}}\right)^{2} \; d\left(\frac{s}{\text{pc}}\right)\end{align}
\begin{align}\therefore \frac{N_{e}}{cm^{3}} = \sqrt \frac{EM}{\left(\frac{s}{pc}\right)}\,.\end{align}
where overall this represents the integral of the electron density squared (N
$_{e}^{2}$
) at a depth (s) measured in pc. To determine the depth, we assume a cylindrical geometry and a path length equal to the smallest diameter of
$2\rlap{.}^\prime2$
. We measure
$N_{e} =$
138 cm
$^{-3}$
. Optically thin PNe have electron densities
$\lt$
5000 cm
$^{-3}$
, whereas for optically thick PNe the electron densities are
$\gt$
6000 cm
$^{-3}$
(Barlow Reference Barlow1987). Therefore, our result supports our earlier findings that Infinity is optically thin at 943 MHz.
Notably, from Figure 1 it is evident that the radio continuum emission closely follows the optical. Specifically, the radio emission closely traces the F606W filter. This is possibly due to the H
$\alpha$
emission line at 656.3 nm, which has a throughput close to the peak of the integrated system,
${\sim}$
29%. H
$\alpha$
is related to the distribution and density of thermally-emitting ionised gas (Tacchella et al. Reference Tacchella2022). Therefore, it provides valuable insight into the structure and morphology of PNe.
We use CARTA and the polygon tool to define the outline of the two inner envelopes in Infinity’s central region, which we arbitrarily refer to as regions R1 and R2 in Figure 4 (measured in Jy/beam). The central coordinates for R1 are RA (J2000) 13:33:37.2 and DEC (J2000) –65:58:14, where we measured an integrated flux density of
$S_{\rm{943\,MHz}} =$
0.0312 Jy (SB53310) and
$S_{\rm{943\,MHz}} =$
0.0315 Jy (SB62225). The central coordinates for R2 are RA (J2000) 13:33:30.0 and DEC (J2000) –65:58:36, where we measure an integrated flux density of
$S_{\rm{943\,MHz}} =$
0.0696 Jy (SB53310) and
$S_{\rm{943\,MHz}} =$
0.0698 Jy (SB62225).
Figure 4. We outline in red the two inner envelopes in the central region of Infinity (which we arbitrarily identify as regions R1 and R2), from which we measure the respective integrated flux densities. Additionally, we measure the apparent size of the inner region containing R1 and R2, as outlined by the central black rectangle.
We measure the apparent size of the inner region containing only the R1 and R2 envelopes (indicated by the central black rectangle in Figure 4), which is
$1\rlap{.}^\prime6\times0\rlap{.}^\prime52$
, corresponding to a physical distance of 0.70 pc
$\times$
0.23 pc. Individually, the apparent size of the R1 envelope is
$0\rlap{.}^\prime48\times0\rlap{.}^\prime37$
, corresponding to a physical distance of 0.21 pc
$\times$
0.16 pc. For the R2 envelope, we measure
$0\rlap{.}^\prime85\times0\rlap{.}^\prime42$
, corresponding to a physical distance of 0.37 pc
$\times$
0.18 pc.
Using the mean of each of the R1 and R2 integrated flux density values, we also measure the optical depth and electron density values for these two inner envelopes, using the same methodology in Equations (2)–(8) inclusive (this includes assuming a cylindrical geometry and hence measuring a path length equal to the smallest diameter, which for R1 is 0.16 pc and for R2 is 0.18 pc). This approach will determine whether Infinity becomes optically thick in the line of sight to the central region, which would not only impact the overall integrated flux density measurement but possibly explain why the spectral index value is slightly higher than the optically thin case of
$\alpha$
= –0.1. For R1 we measure
$\tau$
= 0.0098 and
$N_{e}$
= 405.82 cm
$^{-3}$
, and for R2 we measure
$\tau$
= 0.011 and
$N_{e}$
= 402.66 cm
$^{-3}$
. We can infer from both results of
$\tau$
and
$N_{e}$
that Infinity is optically thin in the R1 and R2 regions. In terms of the slightly raised spectral index value, this warrants further investigation in future research, particularly at lower frequencies where the optical thickness may not be negligible in a specific direction.
Danehkar et al. (Reference Danehkar, Karovska, Maksym and Montez2018) mapped the excitation in the inner regions using HST WFC3 imaging and describe the innermost structures as two low-ionisation envelopes (hence, spectroscopic studies predominantly detect emission lines such as [Nii] and [Oi], Mari, Akras, & Gonçalves Reference Mari, Akras and Gonçalves2023) surrounded by highly ionised environments, resulting from recent outbursts of the progenitor AGB star. According to Miszalski et al. (Reference Miszalski, Acker, Parker and Moffat2009), these low–ionisation structures are a defining feature of post common–envelope PNe that surround a Wolf-Rayet star, such as Infinity.
4. Summary
Our motivation for this study has been to contribute radio continuum measurements for the well-known Galactic planetary nebula NGC 5189 (which we refer to colloquially as Infinity), using the ASKAP radio telescope array. The ASKAP-EMU survey has observed Infinity at 943 MHz, from which we have measured the integrated flux density, spectral luminosity, surface brightness, physical diameters, optical depth, and electron column density. We also measured the integrated flux densities of Infinity in the SUMSS and PMN data, and combined those with other published radio measurements to determine the most up-to-date spectral index. We infer that Infinity is a thermal (free–free) emitting nebula, is in agreement with previous studies, and consistent with the expectations of a PNe composed of ionised plasma.
Acknowledgements
We are grateful for the useful comments and suggestions made by Sanja Lazarević, Velibor Velović, Denis Leahy, and to our referee, which have improved the manuscript. This scientific work uses data obtained from Inyarrimanha Ilgari Bundara, the CSIRO Murchison Radio-astronomy Observatory. We acknowledge the Wajarri Yamaji People as the Traditional Owners and native title holders of the Observatory site. CSIRO’s ASKAP radio telescope is part of the Australia Telescope National Facility. Operation of ASKAP is funded by the Australian Government with support from the National Collaborative Research Infrastructure Strategy. ASKAP uses the resources of the Pawsey Supercomputing Research Centre. Establishment of ASKAP, Inyarrimanha Ilgari Bundara, the CSIRO Murchison Radio-astronomy Observatory and the Pawsey Supercomputing Research Centre are initiatives of the Australian Government, with support from the Government of Western Australia and the Science and Industry Endowment Fund.
ADA is proudly supported by the Australian Government Research Training Program (RTP) Scholarship and CSIRO ATNF Space and Astronomy Student Program.
This research is also based on observations made with the NASA/ESA Hubble Space Telescope obtained from the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS 5–26555. These observations are associated with program 12812.
Data availability statement
The relevant ASKAP-EMU data (with scheduling block IDs: SB53310 and SB62225) are accessible via the CSIRO ASKAP Science Data Archive (CASDA): https://research.csiro.au/casda/.
We also use the available HST data from the Hubble Legacy Archive (HLA) to compare the optical and radio properties. The images are from project 12812, accessible at: https://hla.stsci.edu/.
Open access
