Research Note

Space Telescope Imaging Spectrograph Parallel Observations of the Planetary Nebula M94‐201

Philip Plait2,3 and Theodore R. Gull3  

ABSTRACT

The planetary nebula M94‐20 in the Large Magellanic Cloud was serendipitously observed with the Space Telescope Imaging Spectrograph on board the Hubble Space Telescope as part of the Hubble Space Telescope Archival Pure Parallel Program. We present spatially resolved imaging and spectral data of the nebula and compare them with ground‐based data, including detection of several emission lines from the nebula and the detection of the central star. We find that the total flux is ergs s−1 cm−2, and we estimate the magnitude of the central star to be . Many other Hα sources have been found in M31, M33, and NGC 205 as well. We discuss the use of the parallel observations as a versatile tool for planetary nebula surveys and for other fields of astronomical research.

Received 1998 March 30; accepted 1999 March 4

1. INTRODUCTION

 

The Hubble Space Telescope (HST) Archival Pure Parallel Program, started in 1997 shortly after the second Hubble servicing mission, was implemented to provide the maximum amount of science possible with HST and its primary cameras: the Wide‐Field Planetary Camera 2 (WFPC2), the Near‐Infrared Camera and Multi‐Object Spectrograph (NICMOS), and the Space Telescope Imaging Spectrograph (STIS). During observations of a target by the primary instrument, the other instruments make a preplanned series of observations in nearby fields, the positions of which are determined by the focal plane offset of the instrument and the roll angle of HST. The resulting observations are placed in the Hubble Data Archive at the Space Telescope Science Institute and are made immediately available to the astronomical community.

The STIS parallels include broadband camera images and full‐field slitless spectroscopy. Full details of the STIS parallel survey capabilities and techniques can be found in Gardner et al. (1998). The imaging mode has a very broad bandpass (FWHM ≈ 5000 Å) peaking at 5500 Å and ranging from ≈2500 to 10300 Å. The full‐field spectroscopy mode has a wavelength range of 5200–10300 Å and a dispersion of 4.9 Å pixel−1. In an imaging mode observation of 2000 s, point sources as faint as can be detected with a signal‐to‐noise ratio of 5, and the spectra can achieve a limiting magnitude of 22 under the same circumstances.

Approximately 6000 STIS parallel images and spectra were taken in the first year of the parallel program, comprising more than 1000 separate fields. Of these, over 300 images and spectra are of the Large Magellanic Cloud (LMC). These data represent a unique opportunity to peer deeply into the LMC and investigate many of its physical properties. Planetary nebulae (PNe, singular PN) provide a way to study several important properties of the LMC, including kinematics (Vassiliadis, Meatheringham, & Dopita 1992) and dynamical and chemical evolution, and can also be used as distance indicators (see, for example, Feldmeier, Ciardullo, & Jacoby 1997 and references therein). Nearly 300 PNe have been found in the LMC (Leisy et al. 1997). LMC PNe are of particular importance in the determination of the PNe luminosity function, since they all lie at approximately the same distance. Unfortunately, the great distance to the LMC means that most PNe are very faint and unresolvable by ground‐based telescopes, and the typically crowded LMC fields make studying the PNe very difficult.

The high spatial resolution of HST and low sky background of Earth orbit provide an advantage for studying LMC PNe (for example, Dopita et al. 1996). In particular, STIS can record spatially resolved spectra of the larger, fainter PNe. Here we report on the serendipitous STIS parallel observations of the planetary nebula M94‐20, first discovered by Morgan (1994) in a ground‐based emission‐line survey of the LMC. The STIS camera‐mode images clearly resolve the nebula to be ≈2 in diameter and also detect the central star. The spectra provide identification of several emission lines in the nebula. The low resolution of the STIS spectra together with the relatively large angular extent of the nebula blend the nebular diagnostic emission lines, but the images and spectra are still a useful tool in investigating extragalactic PNe and can be used to plan follow‐up observations, both ground‐ and space‐based.

2. OBSERVATIONS

 

Six images and two spectra were taken of the field containing the planetary nebula M94‐20. The total exposure times were 1100 and 1200 s for the images and spectra, respectively. The observations were taken as part of the STIS Archival Pure Parallel Program, HST Program ID 7783, S. Baum, Principal Investigator. WFPC2 was the prime instrument during the STIS parallels, observing the LMC as part of the HST program “Star Formation History of the Large Magellanic Cloud,” HST ID 7382, T. Smecker‐Hane, Principal Investigator. The primary target was LMC‐DISK1 at 92, 62. STIS was located approximately 5 north of WFPC2 during the observations. The STIS parallel images were formatted as arrays ( 051), while the spectra were automatically binned on chip in the y‐direction to ( 051 in the spectral direction and 0 102 in the spatial direction). The observations were taken over two orbits, and there was a 0 8 (15.5 STIS CCD pixels) telescope slew to the southwest between the two sets of observations. The offsets between images were determined by cross‐correlating a small subsection in each, which were then used to shift and combine the images. The offsets were applied with appropriate binning to the spectra which were shifted and combined as well. The images and full‐field spectra were processed using STIS Investigation Definition Team pipeline calibration software which performs basic data reduction steps such as bias and dark current subtraction. The observation particulars are shown in Table 1.

TABLE 1
TABLE 1 Log of STIS M94‐20 Observations (1997 October 23)

Open New Window

For the analysis using the imaging mode, the sky background was subtracted by taking a simple median of the area near the PN. The background in the full‐field spectral image was subtracted from the spectrum using a column‐by‐column median, employing sigma clipping to remove the positive bias of the stellar contamination.

3. DISCUSSION

 

3.1. The Nebula

Figure 1 shows the full‐field processed image containing M94‐20. The PN is located at the top of the image, centered roughly in the horizontal (x‐axis) direction. North is to the left and east is down; we display the image in this way so that it has the same sense as the spectrum, which disperses light along the x‐direction. Although close to the detector edge, the nebula is located fully inside the image. Note the small irregular galaxy located 9 to the north of M94‐20 and another located 22 to the south. Although both lie in the dispersion direction of the PN, neither interferes significantly with the spectrum. The bright diagonal line in the image is a diffraction spike from a star located just off the detector, and the circular features near bright stars are internal reflections in STIS. The inset shows a close‐up of just the nebula. The PN, unresolved in the discovery survey (Morgan 1994), is clearly resolved in the STIS image. The measured properties of the nebula and central star (see § 3.2) are listed in Table 2. The nebula is slightly elliptical, with a mean diameter of roughly 2 , making this one of the largest PNe in the LMC. Most LMC PNe are smaller than 1 in diameter, with the notable exception of LMC‐SMP72, a bipolar PN measuring approximately (Dopita et al. 1996). M94‐20 has a bright elliptical rim, and the outermost parts of the nebula also show some faint structure, reminiscent of the double‐elliptical structure of NGC 6543 (i.e., the “Cat’s Eye”). One of the outer ellipses is aligned with the inner rim, whereas the other has a position angle of ∼145°. The large angular extent of the PN implies that this is an evolved object.

Fig. 1.— Combined and processed camera‐mode image of M94‐20 field. The inset shows a close‐up of the nebula. The inner rim and central star can be seen clearly in the close‐up image.

Open New Window

TABLE 2
TABLE 2 Properties of M94‐20 and the Central Star

Open New Window

A subarray of the spectrum containing M94‐20 is shown in Figure 2 (top). The extracted subarray covers the full spectral range of the original spectrum, but only 32 pixels (3 3) in the spatial direction. The stars in the image appear as sources of continuum, while M94‐20 is clearly an emission‐line object. Figure 2 (middle) shows a close‐up of the nebular spectrum from 6000 to 7200 Å as well as the positions of the line images using ellipses with major and minor axes corresponding to those measured in the camera‐mode image (bottom). The bright elliptical patch corresponds to the lines of Hα and [N ii] in M94‐20. At this resolution, the [N ii] λλ6548, 6584 lines are separated from Hα by only 3 and 4 pixels, respectively. The nebula itself is about 40 pixels or 10 times that size, so the three spectral line images overlap. The blended nebular spectral image is about 4 pixels wider than the nebular image in the camera mode, also indicating that more than one line is present. The [O i] line at 6300 Å is also clearly seen. The [S ii] λλ6717, 6731 emission‐line images are barely detected in the spectrum. We also note that we may have a detection of [S iii] λ9069 and Paϵ, [S iii] at 9545 Å, which does not reproduce well in Figure 2 but can be seen very faintly in the original data.

Fig. 2.— Top: A subarray of the full‐field spectrum containing M94‐20. The subarray covers the full spectral range of the original spectrum and 32 pixels (3 3) in the spatial direction. Several emission lines can be seen in this log scale image, while stars appear as streaks. The close‐up of the region from 6000 to 7200 Å is also shown (middle) along with a schematic of the positions of the detected emission lines (bottom).

Open New Window

We flux‐calibrated the brighter emission lines using the absolute sensitivity of STIS for a point source (Collins & Bohlin 1998), correcting the sensitivity curve for a slitless spectrum of an extended line emission object, and masking out the continuum spectra of nearby stars that overlapped the PN spectrum. We find that the total observed λλ6548, 6584 flux is ergs s−1 cm−2. D. Morgan & Q. Parker (1998, private communication) made follow‐up observations of many of the PNe found by Morgan (1994), and for M94‐20 detected [O iii] λλ4959, 5007 and Hα. They report the [O iii] flux to be ergs s−1 cm−2. The value of the ratio of λλ6548, 6584 to [O iii] λλ4959, 5007 in LMC PNe is typically 3–4 (see, for example, Vassiliadis et al. 1992), which is somewhat lower than but consistent with these measurements. The typical value of the ratio of [O i] λ6300 to Hα in LMC PNe is approximately 0.07. We find the [O i] λ6300 flux is ergs s−1 cm−2, which again gives a ratio higher than, but consistent with, the typical value. Morgan & Parker also find an upper limit to the Hβ flux of ergs s−1 cm−2, making M94‐20 a relatively faint LMC PN (Vassiliadis et al. 1992). We note that our measurements may also be upper limits, since background subtraction is difficult because of the stellar spectra superposed on the PN spectrum. The detection of [S ii] λλ6717, 6731 is very weak and is complicated by their proximity to the λλ6548, 6584 lines. We subtracted a linear fit to the slope of the wing of the lines to find the total flux of the [S ii] lines and get an upper limit of ergs s−1 cm−2.

The emission lines from the PN are remarkably featureless spatially, given the obvious structure in the camera‐mode image. The spectral mode of STIS has a blue cutoff at ∼5300 Å, whereas the imaging mode goes down to ∼2000 Å. The imaging mode will therefore detect the lines of Hβ, the [O ii] doublet at 3727 and 3729 Å, and the lines of [O iii] at 4959, 5007 Å. The emission lines seen in the imaging mode but not in the spectroscopic mode must account for these features.

3.2. The Central Star

The central star, as well as two other superposed stars, are clearly visible in the camera‐mode image. Relative to the Fine Guidance Sensor guide stars, we find the J2000 coordinates of the central star are 64, 54, in good agreement with those found in Leisy et al. (1997). The spectrum of the star is too weak to determine a temperature, so we used the camera‐mode sensitivity of STIS to flux‐calibrate blackbody models of 20,000, 40,000, and 100,000 K. The calibrated models were then folded into a Johnson V‐filter bandpass to find the V magnitude of the star. We found the central star magnitude to be . The sensitivity of STIS in the imaging mode drops rapidly with decreasing wavelength, so the calculated magnitude does not depend strongly on the temperature of the star. Given the magnitude and a distance to the LMC of 55 kpc, we find that the luminosity of the central star is 0.09 solar, and for a typical white dwarf radius of cm. The low temperature indicates this may be a relatively old white dwarf, which is consistent with the evolved nature of the PN.

3.3. Other PNe

There are approximately 300 known PNe in the LMC (Leisy et al. 1997). The number density is highest near the Bar and tapers off rapidly with distance (Morgan 1994). M94‐20 lies over a degree from the Bar, near the edge of the LMC. Over 300 observations comprising 70 separate fields have been taken in the LMC by STIS. Of these, 29 also have associated full‐field spectra, enhancing the ability to detect candidate PNe (i.e., emission‐line objects) clearly. The detection limit of an observation depends most strongly on the brightness and the size of the nebula in a given emission line and the integration time. The 1200 s exposure of the M94‐20 observations is fairly typical of parallels, and noting that the [S ii] λλ6717, 6731 emission lines in the nebula are barely detected, we expect to detect any nebula the size of M94‐20 brighter than about ergs s−1 cm−2 in a line for a typical LMC PN. A smaller nebula would be detectable at fainter limits, with the brightness scaling inversely with the area. We note here again that M94‐20 is a relatively faint, large LMC PN.

No other previously cataloged LMC PNe have been observed by STIS in the parallel survey. We have processed and examined the other fields in the LMC and also 20 STIS fields in the Small Magellanic Cloud (SMC) and have found no other extended PNe to within the detection limits, cataloged or otherwise, although large H ii regions are common and easily detected. Unresolved or partially resolved PNe are difficult to identify in this case because most of the full‐field spectra in the LMC and SMC either are single observations or have one cosmic‐ray split (that is, two observations without telescope movement), making the differentiation of sharp spectral features from cosmic rays and hot pixels difficult. We note here that in many other fields, STIS observations have two or more cosmic‐ray splits, making unresolved PN detection relatively easier. We have also searched the relevant WFPC2 parallel fields; no known LMC or SMC PNe are located within these parallel survey fields. Discovering previously unrecorded PNe using WFPC2 parallels is more difficult because of the lack of unambiguous spectral information in the broad passbands used.

Although the odds of finding PNe in the random STIS parallels are low for the LMC or SMC, we are encouraged by the observations of M94‐20. The ability of STIS to get deep, spatially resolved spectra makes it an excellent instrument for tagging objects for further deep ground‐based spectroscopy. STIS parallels will continue to be taken in the Magellanic Clouds, and it is only a matter of time before additional PNe are observed. Interestingly, PNe in nearby galaxies can also be detected in STIS parallels, although this becomes a difficult process as PNe at that distance are no longer resolved spatially. However, it also means that the larger volumes of space on the scale of the host galaxy are observed as well. Ground‐based surveys using narrow passband filters centered on [O iii] λ5007 and Hα have found many PN candidates in M31 and M33 (Ciardullo et al. 1989; Bohannan, Conti, & Massey 1985). We made a preliminary search of STIS parallel fields in M31 and NGC 205 which yielded several emission‐line objects. At least one of these objects is a planetary nebula, even though only a small fraction of the surface area of those galaxies has been observed. A similar search in M33 fields has also resulted in several detections. These observations are awaiting ground‐based follow‐ups to get spectra with wavelength coverage that includes Hβ and [O iii] λ5007 to help distinguish PNe from H ii regions. Future STIS parallels will undoubtedly find more PNe, especially if the narrow [O ii] λ3727 and [O iii] λ5007 filters are used in imaging mode in conjunction with the G430L grating, which has a bandpass that includes these emission lines.

We note finally that the use of STIS parallels goes well beyond that of finding (or adding to our knowledge of preexisting) PNe. Moderate redshift galaxy counts (Gardner et al. 1998), research on stellar population counts, low‐mass stars (Plait 1999), globular clusters in nearby galaxies, and many other fields can benefit from a relatively simple search of the parallel archive. Since the data already exist and are publically accessible, the STIS parallels are a excellent tool that should be exploited to their fullest extent.


The authors wish to acknowledge the help of Nick Collins, Bob Hill, Jon Gardner, and David Morgan for giving their advice during this work.

REFERENCES

 

  • 1 Based on observations made with the NASA/ESA Hubble Space Telescope, obtained from the data archive at the Space Telescope Science Institute. STScI is operated by the Association of Universities for Research in Astronomy, Inc., under the NASA contract NAS 5‐26555.

  • 2 Advanced Computer Concepts, Inc., Potomac, MD 20854; .

  • 3 Laboratory for Astronomy and Solar Physics, Code 681, Goddard Space Flight Center, Greenbelt, MD 20771; .

© 1999. The Astronomical Society of the Pacific. All rights reserved. Printed in U.S.A.