First results from the combination of the AAO/UKST and Marseille H $\alpha$ surveys
D. Russeil , Q. A. Parker ,, PASA, 18 (1), in press.

Next Section: Some preliminary results
Title/Abstract Page: First results from the
Previous Section: Introduction
Contents Page: Volume 18, Number 1

The data: the AAO/UKST and Marseille H $\alpha$ Surveys

The main characteristics of the AAO/UKST and Marseille H $\alpha$ surveys are summarized in table 1 and described briefly below.

The UKST/AAO H $\alpha$ survey

The AAO is undertaking a photographic H $\alpha$ survey of the southern GP and Magellanic Clouds using the 1.8m/1.2m Schmidt Telescope (UKST). A monolithic 1% H $\alpha$ interference filter with a 5 $^{\circ }$ diameter circular field of view on a 6.5 $^{\circ }$ square glass substrate is used. It is believed to be the largest astronomical single-element, narrow-band filter. Deep 3 hour exposures are taken on fine grained Kodak Tech-Pan film on 4 $^{\circ }$ field centres (necessary to achieve contiguous coverage of the GP). Detailed descriptions of the filter and survey are given by Parker and Bland-Hawthorn (1998), and Parker and Phillipps (1998a,b). This survey combines wide area coverage, good ( $\sim2$ arcsec) spatial resolution and decent ( $\sim5$ Rayleighs) sensitivity to optical gaseous emission. H $\alpha$ emission structures can be simultaneously examined at very small (few arcseconds) and very large (several degree) scales. Many kinds of extended Galactic objects are detectable including discrete HII regions, diffuse emission patches, planetary nebulae (Parker, Russeil, Hartley, in preparation), large and small scale filaments, bubbles, shells, dark zones, supernovae remnants (Walker et al., 1999), Herbig-Haro objects (Mader et al., 1999), etc. Although morphological analysis of such emission structures can help to reveal connections between them, velocity information is crucial to unambiguously establish any true physical relationship. However, there is no velocity information in the UKST H $\alpha$ survey given the 70Å FWHM of the filter ( $\sim 3200$ km $\rm s^{-1}$ ). This is where the Marseille H $\alpha$ survey comes into its own.

The Marseille H $\alpha$ survey

The Marseille H $\alpha$ survey (MHS) is based on a scanning Fabry-Perot interferometer (Amram et al., 1991, le Coarer et al., 1992). The instrument, based at ESO (La Silla), provides spectral information for each pixel in the image plane. This allows a kinematic view of H $\alpha$ emission to be built up as a datacube. Such information is essential to differentiate different emission components along the line of sight. The observed line profiles are a superposition of these different emission contributions and the night sky lines (geocoronal H $\alpha$ and OH lines). After decomposition and subtraction of the sky, individual nebular components can be determined. This allows unbiased velocity estimates for the detected ionised gas in the field. Both the intensity and velocity variation of each resolved H $\alpha$ emission component can be determined across each field. Unfortunately, the MHS coverage of the GP is not continuous, the field of view is modest (0.66 $^{\circ }$ ) and the resolution is relatively low (the pixel size itself is 9 arcseconds). Only the richest areas in HII regions have so far been covered. Field selection was based on previous radio and photographic H $\alpha$ survey data (Georgelin & Georgelin, 1970, Haynes et al., 1978). More recent radio surveys (Wright et al., 1994, Griffith & Wright , 1993, Griffith et al., 1995, Gregory et al., 1996) and the AAO/UKST H $\alpha$ survey reveal many more interesting zones to select for further follow-up with the MHS instrument.

**Table:** The main characteristics of the two H $\alpha$ surveys.
	UKST	MHS
Telescope diameter (cm)	120/180	36
Field size (degrees)	5.5	0.66
Filter FWHM (Å)	70	10
Central wavelength (Å)	6590	6562.78
Detector	TechPan film	IPCS+FP $^\ddagger$
Velocity information	none	All over the field
		FSR $^\ast$ : 115 km $\rm s^{-1}$
		$\Delta$ v $^\star$ : 5 km $\rm s^{-1}$
Typical resolution (arcsec.)	$\sim2$	>9
Approximate sensitivity	5 R $^\dagger$	<1 R $^\dagger$
Observing time (hours)	3	2

notes: $^\ddagger$ IPCS+FP:Image Photon Counting System+Fabry Perot $^\ast$ FSR is the Free Spectral Range of the interferometer $^\star$ $\Delta$ v is the spectral resolution $^\dagger$ At $\lambda$ (H $\alpha$ )=6563Å, 1R

$=2.41\times10^{-7}~erg~cm^{-2}~s^{-1}~ster^{-1}$

Survey detection limits

The detection limits of each survey depend on the different characteristic parameters of the instrumentation and the type of observation. The MHS provides H $\alpha$ spectra and channel maps while the AAO/UKST survey provides H $\alpha$ intensity maps integrated along the line of sight.

The high spatial resolution AAO/UKST survey exhibits high contrast between emission regions and adjacent dark zones whilst also revealing fine detail in faint H $\alpha$ emission structures. For the MHS an emission is detected if an associated component in the line profile is observed. The narrower filter bandpass (10Å) decreases contamination from parasitic night sky lines and continuum emission. This allows detection and discrimination of diffuse and uniform H $\alpha$ emissions (e.g. the Warm Ionised Medium - WIM) as well as more discrete and extended objects. However, due to the limited 9 arcsec pixel size, fine structures (e.g. filaments) will be diluted or unresolved. Although such detailed structural information is lost, velocity information exists so that such faint structures can be detected as faint components in the observed H $\alpha$ profiles at different velocities from other diffuse emissions met along the line of sight.

At present there are no absolute comparisons of the depths of the two surveys available. The MHS has not yet been systematically flux calibrated (an estimation of 0.2R for the faintest diffuse emission detected from a circular 30' area by MHS can be found in Marcelin et al., 1998) and the AAO/UKST survey has not yet been properly calibrated. Work is currently underway to address these issues. For the AAO/UKST survey, preliminary results (J.Precious, private communication) reveal limits of about 5.1 Rayleighs for two different UKST H $\alpha$ survey fields. For the purposes of this paper we illustrate the detection power via simple qualitative comparison. For overlapping areas of coverage between the two surveys we looked for common detections in the Caswell and Haynes (1987) radio source catalogue. This is done because an important step in the study of our Galaxy's structure concerns the optical detection of radio sources. The two surveys detected the same radio sources down to the limit of 0.7 Jy.

The importance of wide area coverage and high resolution

The AAO/UKST H $\alpha$ survey allows detection of features of very large angular size while also providing fine detail on arcsecond scales. Such large features include bubbles, cavities, arcs, filaments, loops, superbubbles, supershells, and supergiant shells (e.g. Tenorio-Tagle & Bodenheimer 1988). Some of these induce the velocity departures of the triggered HII regions, identification of which is crucial to studies of Galactic structure. In parallel, HII regions which are apparently isolated or well detached on the plane of the sky can be grouped into complexes by taking their velocities into account.

The high AAO/UKST survey resolution permits the precise morphology of HII regions to be seen. This is vital information in determining the location of the exciting stars via orientation of the observed rims and dust "elephant trunks" with respect to the HII region as a whole (Pottash, 1965; Herbig, 1974). More precise determinations of the extent of HII regions can also be made (for example RCW105, Rodgers et al., 1960). This additional structural information can account for the different diffuse emission components and their spatial intensity variations observed in the lower resolution MHS mosaics.

The importance of the combination of high resolution and large field of view coupled with velocity information is illustrated by the following examples.

Next Section: Some preliminary results
Title/Abstract Page: First results from the
Previous Section: Introduction
Contents Page: Volume 18, Number 1