Elizabeth J. McGrath '01, Vassar College (elmcgrath@vassar.edu)

I. Introduction

W49 is a well known HII region, first catalogued by Westerhout in 1958. It consists of two main regions–to the east, W49B, and to the west, W49A, separated by ~12'. W49B is a supernova remnant, while W49A is an intense star-forming region, divided into a northern and a southern component. In addition, there are many ultracompact (UC) HII regions within the northern component, W49N. Our observations center on W49N, where all the maser activity is taking place. In particular, we were interested in determining which UC HII region was associated with the maser activity.

H₂0 masers are important to study for many reasons. They are both signposts of star formation and tracers of extremely dense (~10⁹ cm^-3) gas. H₂0 masers are also extremely intense–more so than any other masers–which makes them easy to observe. W49N is a unique source in that it has maser sources ranging from 1 to 4000 Jy over an extremely wide velocity range (couple hundred km/s). If the Orion Nebula, at a distance of ~0.5 kpc, was at the distance of W49 (~11.4 kpc as determined through kinematic comparisons of the measured radial velocities and the measured proper motions of the H₂0 masers by Gwinn, Moran & Reid 1992), then it would be an extremely faint continuum source–60 times fainter than W49.

Masers are the result of shocks from high density star-forming regions. Young stars have high velocity outflows that push into the dense surrounding gas region and form the shocks, which propagate through the material. The winds become entrapped behind the dense shells that they carve out, and we observe the masers that form on the edge of these shocked shells (figure 1). Masers have a population inversion which means that there are more electrons in the excited state than in the ground state. They release this energy through a rotational transition (much the same as a laser here on Earth.), which is what we see as the actual maser.

Elitzur first demonstrated his theory of a jet-driven cocoon in 1995, where masers form on the edge of the expanding shell. The shell expands quickly along the axis of the jet, slower perpendicular to it. His model gives a dynamical timescale of only 300 years which accounts for why we do not see masers in the other HII regions in W49N–they've already passed their maser lifetime. Thus we are seeing W49N at a crucial stage in its star-forming history.

Maser observations of W49N were done using the VLA in line mode using 2 intermediate frequencies (IF's), at 22 GHz at four different time periods–24 July 1998, 6 September 1998, 26 June 1999, 18 July 1999. The 1998 data was in B configuration, with a resolution of 0.30", and the 1999 data was in A configuration, with a resolution of 0.045". The first IF was centered on velocity multiples of ±30 km/s from the local standard of rest. IF 2 was centered on a velocity of +170 km/s. Phase referencing with our phase calibrator, 1923+210, was done in order to keep the phases coherent.

All data reduction was done with standard techniques using the NRAO package, AIPS (Astronomical Image Processing System). We found a strong maser feature in IF 2 in channel 31, at a velocity of 170.7 km/s, which we used to self-calibrate our data. Self-calibration destroys all absolute positioning, so all maser positions are relative to this feature. We then used the position of this maser feature in our August 1998 continuum data as a reference to all other dates and shifted the IF 1 cubes by the appropriate amount to ensure that ch 31 in IF 2 was at the correct reference coordinates. This uses the assumption that the ch 31 feature is moving by negligible amounts between observations. This allowed us to align the masers to our August continuum data. We also had continuum data from A configuration in April 1998 at 43 GHz, and continuum from the BIMA array (A and C configuration) at 90 GHz. These continuums were aligned to each other using the basic assumption that the continuum shows similar features at all frequencies. We estimate that our absolute astrometry is good to ~0.1" and our relative astrometry is good to ~0.01". This is a factor of ten better than previous observations.

We catalogued a list 312 H₂0 maser positions between velocities –260 and +270 km/s. These show strikingly similar structure as the ones previously observed by Gwinn et al. 1992, Walker et al. 1982, and Moran et al. 1973. In particular, the three main clumps that are nearly vertical north-south remain apparent in all observations. And the pattern of an outflow from a central source is apparent, as well. Masers to the west show redshifted velocities, and masers to the east show blueshifted velocities. Note that this could not be due to rotation, as the velocities are too high. Thus, Elitzur's model of a jet-driven cocoon from a young star, or cluster of young stars, is the best model for the observations. We calculate the center of outflow of the masers to be within 0.2" of the bright UC HII continuum source, G2.

I would like to thank my advisor, Miller Goss, for his constant patience and wisdom throughout the project. In addition, I would like to thank my advisors at Vassar, Debra Elmegreen and Fred Chromey, for their support. Thanks to Niruj Mohan for his expertise in AIPS, and to Teddy Cheung for his help with DIFMAP. I would also like to thank the NSF sponsored REU program for the grant that made this research possible.