1: \documentclass[12pt,preprint]{aastex}
2: %IRAS4AB paper
3: \newcommand{\kms}{km s$^{-1}$\hspace{0.25em}}
4: \newcommand{\etal}{et al.\,~}
5: %\newcommand{\deg}{$^{\circ}$}
6: %
7: %\received{}
8: %\revised{}
9: %\accepted{}
10: %
11: \slugcomment{Submitted to Astrophysical Journal}
12: %
13: \shorttitle{VLBI Water Maser Observations of IRAS4A/B}
14: \shortauthors{K.B. Marvel, \etal}
15: %
16: \begin{document}
17: %
18: \title{Time-Resolved AU-Scale Jets Traced by Masers in the IRAS~4A/B Regions of NGC1333}
19: %
20: \author{Kevin B. Marvel}
21: \affil{American Astronomical Society}
22: \affil{2000 Florida Avenue, NW, Suite 400, Washington, DC 20009}
23: \email{marvel@aas.org}
24: %
25: \author{Bruce A. Wilking}
26: \affil{Department of Physics and Astronomy}
27: \affil{University of Missouri at St. Louis, 1 University Boulevard, St. Louis, MO 63121}
28: \email{bwilking@umsl.edu}
29: %
30: \author{Mark J. Claussen}
31: \affil{National Radio Astronomy Observatory (NRAO) Array Operations Center}
32: \affil{P.O. Box 0}
33: \affil{1003 Lopezville Road, Socorro, NM 87801}
34: \email{mclausse@nrao.edu}
35: %
36: \author{Alwyn Wootten}
37: \affil{NRAO}
38: \affil{520 Edgemont Road, Charlottesville, VA 22903-2475}
39: \email{awootten@nrao.edu}
40: %
41: %
42: \begin{abstract}
43: %
44: We present results of VLBI observations of the water masers associated
45: with IRAS 4A and IRAS 4B in the NGC 1333 star-forming region taken in four
46: epochs over a two month period.
47: Both objects have
48: %%Suggest we use very young here rather than Class 0, which may not be known to all abstract
49: %%readers
50: been classified as extremely young sources and each source is known to be a multiple system.
51: Using the Very Long Baseline Array, we detected 35 masers in Epoch I, 40 masers in Epoch II,
52: 35 in Epoch III, and 24 in Epoch IV. Only one identified source in each system associates with these masers. These data are used to calculate proper motions for the masers and trace the jet outflows within 100 AU of IRAS 4A2 and IRAS 4BW.
53: In IRAS 4A2 there are two groups of masers, one near the systemic cloud velocity
54: and one red-shifted.
55: %%made more direct
56: They expand linearly away from each other at velocities of 53 \kms.
57: In IRAS 4BW, masers are observed in two groups that are blue-shifted and red-shifted
58: relative to the cloud velocity. They form complex linear structures with a thickness of 3 mas
59: (1 AU at a distance of 320 pc)
60: that expand linearly away from each other at velocities of 78 \kms.
61: Neither of the jet outflows traced by the maser groups align with the larger scale outflows.
62: %%changed slightly as it wasn't clear what was suggesting.
63: We suggest the presence of unresolved companions to both IRAS 4A2 and 4BW.
64: \end{abstract}
65: %
66: \keywords{ISM:jets and outflows --- masers --- shock waves ---
67: stars:mass loss --- stars:pre-main sequence}
68: %
69: \section{Introduction}
70:
71: Water masers are excellent probes of astrophysical flows. Their large flux
72: densities and exceedingly compact sizes make them perfect Very Long Baseline
73: Interferometer (VLBI) targets.
74: Many masers observed in star-forming regions are thought to form in shocks
75: either in or along the outflow of material commonly seen associated with
76: young stellar objects (YSOs). They provide a high resolution probe of the
77: base of the stellar wind that is unaffected by extinction from dust
78: or extensive interaction with ambient material.
79: The milliarcsecond resolution of VLBI is critical to resolve the individual
80: maser components and outflows, particularly in high stellar density environments.
81:
82: Long term total power monitoring has shown that masers around low mass YSOs
83: are more episodic than those around higher mass YSOs, however, their
84: detectable phases are sufficiently long for proper motion measurements if the observations
85: are spaced by no more than two to three weeks \citep{Furuya:2003,Brand:2003,
86: CLAUSSEN:1996,WILKING:1994}. Furthermore, if they arise in the warm (400K)
87: spatially confined post-shock region, as postulated in various models
88: \citep{MODEL1, MODEL2, MODEL3, MODEL4}, they should have space velocities
89: sufficient to produce measurable proper motions over one to three weeks.
90: However, regular monitoring of low mass YSOs is necessary to ensure that
91: VLBI observations are conducted during periods of high maser activity.
92:
93: Using the NRAO's \footnote{The National Radio Astronomy Observatory is a facility
94: of the National Science Foundation operated under cooperative agreement
95: by Associated Universities, Inc.}
96: Very Long Baseline Array (VLBA), proper motion studies have been
97: made of water masers associated with high mass and intermediate mass YSOs
98: such as Cepheus A HW2 \citep{TORRELLES:2001}, S~106 FIR \citep{FURUYA:2000}, NGC 2071 \citep{SETH:2002}, and IRAS~20050+2720 MMS1 \citep{FURUYA:2005} as well as with lower mass YSOs such as Serpens SMM1, RNO 15-FIR, and IRAS~05413-0104 \citep{Moscadelli:2006,CLAUSSEN:1998}.
99: %%Added 20050, fixed 05413 misplaced tilde
100: For example, water masers in IRAS~05413-0104 (associated with HH~212 at
101: a distance of 450 pc), were
102: found to lie in a structure of 10 AU in length and less than 0.5 AU thick.
103: The masers appeared to arise in shock-related structures which showed proper motions
104: along the axis of the outflow of 60 \kms and displayed coherent structures
105: over time scales of two to three weeks (but which was less discernible
106: over time scales as long as several months). Comparison with molecular
107: and infrared observations of IRAS~05413-0104 clearly demonstrated that
108: proper motion studies of water masers in the region (within 40 AU) of the
109: central source are among the best tools for studying the kinematics
110: of the jets emanating from embedded YSOs.
111:
112: IRAS~4, comprised of multiple sources whose submillimeter-dominated 'Class 0' spectral energy distributions suggest extreme youth, is located in the NGC 1333 star-forming
113: region. As a nearby active and multiple young object, it presents a prime candidate for VLBI observations of its associated water masers.
114: Located at a distance of 320 pc \citep{HIPPARCOS}, the NGC 1333 molecular
115: cloud hosts a double infrared cluster of about 200 YSOs identified in
116: near-infrared surveys \citep{STROM:1976,ASPIN:1994,Lada:1996,Wilking:2004}
117: and x-ray surveys \citep{PREIBISCH:1997,Getman:2002,PREIBISCH:2003}.
118: Far-infrared observations have not only revealed the higher luminosity
119: sources in NGC 1333, but also YSOs in the earliest phase of evolution
120: such as the Class 0 sources IRAS~4 and IRAS~2 \citep{Harvey:1984,JENNINGS:1987}.
121: IRAS~4 was found to be a binary at submillimeter wavelengths \citep{SANDELL:1991};
122: hence IRAS~4A and 4B. IRAS~4A is located 31$\arcsec$ to the
123: NW of 4B at a position angle of -45$^{\circ}$.
124: Subsequently, both IRAS~4A and 4B were found to be binary systems.
125: \citet{LAY:1995} performed single baseline interferometry in the submillimeter
126: and found that 4A was a binary with an angular separation of 1.8$\arcsec$
127: (or 580 AU at 320 pc). Interferometric observations in the millimeter
128: and radio continuum confirmed the separation and determined that the
129: position angle of separation was $\approx -20^{\circ}$ \citep{LOONEY:2000,REIPURTH:2002}.
130: While the easternmost source (IRAS~4A1) dominates the millimeter and radio continuum emission,
131: the westernmost source IRAS~4A2) appears to be relatively more evolved with a warm
132: ammonia core \citep{WOOTTEN:1993,Shah:2000} and a wealth of complex organic molecules
133: (Bottinellii 2007, priv. comm.). Submillimeter and millimeter wave continuum observations
134: have confirmed the prediction of \citet{LAY:1995} that IRAS~4B is also a
135: binary with components (4BE and 4BW) separated by about 10$\arcsec$ (3200 AU in
136: projection) along an E-W direction \citep{SMITH:2000,LOONEY:2000,Sandell:2001}.
137: \footnote{We adopt the naming convention introduced by \citet{Sandell:2001}.
138: The western (eastern) source is referred to as IRAS~4BI (IRAS~4BII) by
139: \citet{SMITH:2000}, as IRAS~4B (IRAS~4C) by \citet{LOONEY:2000}, and as
140: IRAS~4B (4B$\arcmin$) by \citep{DIFRANCESCO:2001}.} As in the case of IRAS~4A2,
141: complex organic molecules have been detected toward IRAS~4B \citep{Bottinelli:2007},
142: presumably from a hot core associated with 4BW.
143: We note that IRAS~4BE has yet to be detected in the radio continuum \citep{REIPURTH:2002}
144: or demonstrate any sign of outflow activity. Infall motions have been detected toward IRAS~4A and 4B
145: \citep{DIFRANCESCO:2001}, unambiguously toward the former, and strongly indicated toward the latter.
146: %could reference table of properties here
147:
148: The detection of water masers toward IRAS~4A and 4B suggests that they are YSOs
149: also associated with mass outflow. Water masers were first detected in single
150: dish observations by \citet{HASCHICK:1980} and later monitored by
151: \citet{CLAUSSEN:1996}. These observations established the maser emission
152: as highly variable on monthly time scales, sometimes being completely
153: absent while at other times reaching peaks of 10 Jy. Detected emission
154: was found from $-$10 to $+$15 \kms but the angular resolution was not sufficient
155: to determine which masers were associated with 4A and 4B. Beginning in 1983,
156: VLA observations have shown water maser activity associated only with IRAS~4A2
157: and/or 4BW \citep{RODRIGUEZ:2002,Wootten:1998,Furuya:2003}. VLBA observations
158: of the water masers in IRAS~4A and 4B, acquired in 2003, have been presented
159: by \citet{Desmurs:2006}. While masers were detected in only two of the four
160: epochs observed, they confirmed the association of masers with IRAS~4A2
161: %%subject-verb agreement fixed
162: and expansion motion between the two spots was detected.
163: In IRAS~4BW, the five maser spots detected were red-shifted and formed a chain
164: 80 mas in extent with some masers displaying proper motion toward the north.
165:
166: Molecular outflows have been mapped toward both IRAS~4A and 4B.
167: IRAS~4A is associated with a highly collimated outflow about 20,000 AU
168: in extent seen in HCN and SiO with a position angle close to 20$^{\circ}$
169: \citep{Girart:1999,Choi:2001,Choi:2005}. The origin of the outflow is likely
170: IRAS~4A2 \citep{Choi:2005}. On a larger scale ($\approx 1 \times 10^{5}$ AU),
171: the outflow defined by CO and molecular hydrogen has a position angle of
172: 45$^{\circ}$ \citep{BLAKE:1995,Choi:2006}. The shift in position angle from
173: small to large scale is perhaps due to a combination of a shift in the direction
174: of the magnetic field, an encounter with denser ambient gas, and precession of
175: the outflow axis \citep{Choi:2006}. A more compact outflow is associated with
176: IRAS~4BW, with a position angle close to 0$^{\circ}$ \citep{BLAKE:1995,Choi:2001}
177: with perhaps a second outflow with a position angle of -35$^{\circ}$ \citep{DIFRANCESCO:2001}.
178:
179: In this paper we present VLBA observations of water masers associated with the
180: IRAS~4 region obtained over four epochs in 1998 and spaced by about one month.
181: We also describe Very Large Array (VLA) observations, taken one month in advance of
182: the start of the VLBA observations, that showed IRAS~4 to be a very active
183: maser source and yielded absolute positions for the masers and the 1.3 cm
184: continuum sources IRAS~4A1 and IRAS~4BW. With the VLBA, we detected 35 masers in
185: Epoch I, 40 masers in Epoch II, 35 in Epoch III, and 24 in Epoch IV
186: associated with IRAS~4A2 and IRAS~4BW. We use these data to reveal
187: the structure of shock fronts associated with stellar winds from these YSOs
188: and to estimate proper motions for masers detected in all four epochs.
189: The origin of the maser emission and its relationship to the stellar winds
190: and larger scale molecular outflows from these YSOs is discussed.
191:
192: \section{Observations and Reductions}
193:
194: \subsection{VLA Observations}
195:
196: The VLA was used on August 12, 1998 to observe the water masers
197: toward IRAS~4A and IRAS~4B to check that the masers were strong
198: enough to observe subsequently with the VLBA. The array was in the
199: {\bf B} configuration. The phase tracking center for the array was
200: set to be midway between the IRAS~4A and IRAS~4B positions.
201: spread over a range of hour angles to optimize the {\it u, v} coverage.
202: Two observation modes were used. In the first, observations were made
203: with both circular polarizations, a total bandwidth of 1.5625 MHz
204: (total velocity extent of 21 \kms) and a correlator setup which gave
205: 127 spectral channels each with width of 12.2 kHz (0.16 \kms). In this
206: mode the water masers were observed with reasonably high spectral
207: resolution in order to spectrally resolve the masers, and
208: the observations lasted for $\sim$42 minutes (on source),
209:
210: In the second mode, the correlator was configured for continuum
211: observations with one polarization pair (of IFs) set to a center
212: frequency of 22264.9 MHz and a bandwidth of 50 MHz, while the other
213: polarization pair were set to a bandwidth of 0.781 kHz, and the
214: center frequency adjusted so that the strongest masers were placed
215: at the center of the narrow band. This setup allowed the phases
216: to be tracked using the strong signal in the narrow band, in order
217: to calibrate tropospheric phase changes to apply to the 50 MHz
218: continuum band (away from the maser emission). In this mode the
219: continuum emission from the IRAS~4 region was observed for $\sim$3.5
220: hours (on source).
221:
222: The data were edited, calibrated, and imaged with the NRAO Astronomical
223: Image Processing System (AIPS, Greisen 2003) in the standard manner. In the high
224: spectral resolution data, Stokes {\bf I} images were made using uniform weighting
225: of the {\it u,v} data, of each spectral channel, and were CLEANed
226: to an rms noise (in channels without strong maser features) of $\sim$20 mJy beam$^{-1}$.
227: The synthesized beam was 300 by 275 milliarcseconds (mas) at a position angle
228: of $-$83$^{\circ}$. The strong masers in SSV13, some 196 arcseconds to the
229: north-northwest, contributed sidelobe emission in some spectral channels.
230: In general, however, those channels affected did not have maser emission
231: from IRAS~4A or 4B. With this angular resolution, many spectral features
232: mapped to the same position, although two groups each of spatially separated
233: masers were seen toward both IRAS~4A and 4B. The positions of these
234: four maser groups, as estimated with Gaussian fits to the strongest
235: maser in each group using the AIPS task JMFIT, are listed in Table 1.
236:
237: For the continuum observations, after the standard calibration, the narrow
238: band data were imaged and used in an iterative self-calibration procedure.
239: The self-calibration solutions obtained were applied to the broad band
240: data, which were imaged with natural weighting of the {\it u,v} data
241: (which increases the sensitivity relative to uniform weighting). The
242: synthesized beam for this mode was 360 by 325 mas at a position angle
243: of 77$\deg$. The rms noise in this image was $\sim$120 $\mu$Jy beam$^{-1}$.
244: Unfortunately, the weather was not particularly good, and the system temperatures
245: were elevated by about a factor 1.5 - 2.0 over good weather conditions. Integrated flux densities
246: at $\lambda$=1.3 cm
247: for IRAS~4A1 and 4BW were 1.33 mJy and 0.54 mJy, respectively.
248: IRAS~4A2 and 4BE were not detected to a 3$\sigma$ peak flux density limit of 0.36 mJy beam$^{-1}$.
249:
250:
251: \subsection{VLBA Observations}
252:
253: We have used the VLBA to obtain four epochs of observation
254: of the water maser emission associated with a number of low-mass YSOs.
255: Two subsequent papers will present results on IRAS~16293-2422 and SVS 13.
256: A previous paper \citep{CLAUSSEN:1998} presented results for IRAS~05413-0104.
257: Preliminary results of the observations presented here have appeared as \citet{MARVEL:2002},
258: while those of IRAS~16293-2422 and SVS 13 have appeared as \citet{WOOTTEN:1999}
259: and \citet{WOOTTEN:2002}.
260:
261: The observations were made with one polarization pair (both right and left circular
262: polarization) of baseband channels, each with 4 MHz of bandwidth. This provides
263: 54 \kms of total velocity coverage. Correlation was made with 256 spectral channels
264: per baseband channel, yielding a velocity width in each spectral channel of 0.21 \kms.
265: Although the velocity coverage is then only $\pm$27 \kms from the systemic, no
266: published higher velocity water masers have been detected from the IRAS~4 sources
267: \citep{CLAUSSEN:1996, Brand:2003, Furuya:2003}.
268:
269: Since IRAS~4A and 4B are located close together on the sky (within the primary beam of the
270: VLBA antennas), we were able to direct
271: the antennas of the VLBA to a position between the two sources and then perform
272: two separate correlation passes with the VLBA correlator at the precise
273: positions of the sources without loss of sensitivity.
274: The correlation positions coincide with the centimeter continuum positions
275: (see Table 1).
276: Since the data were obtained contemporaneously, simply
277: being correlated in separate passes, the two data sets share
278: identical time ranges and mid-times of observation.
279: Each of the four observations were separated in time by approximately three weeks.
280: This separation was chosen based on past single dish monitoring observations
281: \citep{CLAUSSEN:1996, WILKING:1994}, which indicated significant flux
282: changes on this timescale. The dates of the observations, mid-times
283: of each epoch and Julian date of the mid-times are
284: given in Table 2 along with the number of days
285: between each pair of the observations.
286:
287: The data from each epoch were reduced using AIPS. Strong
288: continuum sources were observed during each epoch to calibrate both the
289: bandpass response of the receiver systems of each antenna as well as allowing
290: determination of any residual delays uncompensated for by the correlator
291: model. After residual delay removal, a strong maser feature displaying
292: simple structure was selected for determination of delay changes during the
293: observations. After the removal of these residual delay rates, the same
294: channel was split from the data set for iterative self-calibration. The
295: amplitude and phase self-calibration solutions were then applied to all
296: spectral channels and the data were edited for poorly calibrated amplitudes.
297:
298: After all calibrations were applied, the data were mapped using the AIPS task
299: IMAGR. Spectral cubes were formed with pixel cell sizes of 85 $\mu$arcseconds.
300: The restoring beam varied in size, but was typically about 800 $\mu$arcseconds
301: by 400 $\mu$arcseconds with a position angle very close to
302: $-$10.0$^{\circ}$ east of north. The resolution varied for each epoch
303: depending on the details of the self-calibration process and whether or not
304: good self-calibration solutions were obtained for given antennas at given
305: times. The exact imaging parameters for each epoch
306: and rms noises for typical channels (those channels whose noise floor is
307: not dictated by dynamic range constraints) are listed in Table 3.
308:
309: To aid in the analysis of the maser motions observed, two-dimensional Gaussian
310: components were fit to each maser spot in each channel where the emission
311: from an individual spot exceeded five times the RMS noise for that channel.
312: This was carried out using an automated routine in AIPS (SAD) in an iterative
313: process that first fit the strongest Gaussians in a given channel, subtracted
314: the resulting component from the map and then reiterated the search for Gaussian
315: components above the noise cutoff in the map. This produces a fairly reliable
316: set of Gaussian components without the very tedious process of fitting each component
317: above the limit in each channel manually.
318: The resulting positions were then grouped together in velocity space and position
319: in tabular form.
320: Only masers detected in two or more channels above the cutoff and no more than one
321: beamwidth offset in position were retained. Fewer than 10\% of the components were excluded
322: in this process, pointing to the reliability of the iterative fitting process.
323:
324: The surviving component fits were then visually inspected using the 3-dimensional data analysis
325: tool XGOBI\footnote{XGOBI, a multivariate data analysis tool that operates under the X11 system
326: is freely available for download at http://www.research.att.com/areas/stat/xgobi/}.
327: Using XGOBI, it is possible to excise spurious Gaussian fits that result from
328: the automated Gaussian fitting routine we utilized. If a fit was
329: called into question by the XGOBI inspection process, the original maps were inspected
330: prior to deleting the data point. Usually, spurious fits were caused by residual sidelobes
331: being fit in several channels by the SAD routine in AIPS and occurred in channels that were dynamic-range
332: limited. The components were then used for all subsequent fitting and analysis as described in Appendix A.
333:
334: \section{Description and Analysis of the Maser Regions\label{descriptions}}
335:
336: \subsection{IRAS~4A\label{descriptions-4a}}
337:
338: At the time of our observations, the water masers of IRAS~4A were
339: found in two dominant regions, separated by 251 milliarcseconds (Epoch I)
340: or 80.3 AU. A line connecting the two maser regions lies at a position
341: angle of about $-$50$^{\circ}$ (north through east)
342: roughly
343: perpendicular to the CO outflow axis. Comparison with our VLA observations
344: show that the masers are associated with IRAS~4A2.
345: Figure 1 shows
346: the spatial and kinematic distribution of the masers for all four epochs,
347: each with a closeup of the SE maser component.
348:
349: The NW region exhibited only a single maser spot. Its peak flux decreased
350: in time from 0.97 Jy at the first epoch to 0.12 Jy at the final epoch.
351: The maser is at or slightly red-shifted relative to the molecular core velocity of $+$6.7 \kms.
352: The SE region contains more highly red-shifted maser spots in Epochs I and II,
353: including the
354: maser used to self-calibrate each epoch and another weaker maser
355: just west of the reference spot. In Epochs III and IV, another maser
356: appeared $\sim$10 mas north of the reference feature at a velocity of $\sim$12.5 \kms (Figure 1).
357: Because we self-calibrated the data using the brightest spot as the reference,
358: its position remains constant over time and appears at the origin.
359:
360: The separation vector from the reference feature to the NW feature
361: increased in magnitude over time, expanding from 80.4 AU in the first
362: epoch to 82.3 AU in the final epoch.
363: %The separation vector also exhibited a
364: %small, monotonic change in position angle over the course of our observations.
365: %I don't see evidence for this in my analysis (BW).
366: %Figure \ref{4ASEP} shows the separation vector for each of the four epochs.
367: The projected speed of expansion is 53.0 \kms given the assumed distance of
368: 320 pc to the source. Since the line of sight velocity difference between
369: the two regions is only 2 \kms and assuming that this velocity difference
370: and the proper motion result from the projected velocity of an outflow jet,
371: we can calculate an inclination for the outflow of 2$^{\circ}$. The true
372: space velocity is therefore nearly equal to that measured in the plane of the
373: sky, namely 53 \kms. As shown in Table 4,
374: the proper motion vectors for the NW feature relative
375: to the reference feature has a position angle of $-$49.8$^{\circ}$, identical to
376: the position angle of the separation vector.
377:
378: \subsection{IRAS~4B\label{descriptions-4b}}
379:
380: Similar to IRAS~4A, IRAS~4B exhibits water maser emission in two dominant
381: well-separated regions associated with IRAS~4BW: one to the NW and another to the SE.
382: Masers in the NW region are red-shifted relative to the cloud velocity of
383: 7 \kms with velocities ranging from 13 to 20 \kms. Masers in the SE regions
384: are blue-shifted with velocities ranging from -3 to 4 \kms. The radial
385: velocities for the NW and SE groups are consistent with those in the
386: northern and southern lobes of the bipolar HCN outflow observed by \citet{Choi:2001}
387: and the bipolar H$_2$CO outflow mapped by \citet{DIFRANCESCO:2001}.
388: A complete list of the maser component velocities, fluxes,
389: and positions can be found in Appendix A.
390:
391: Figures 2 and 3 show the overall spatial and kinematic
392: distribution of the maser emission as observed in each epoch for the two maser groups in IRAS~4B.
393: %redo figure as 4 panel close-ups of each region
394: Over time, the masers in the NW expand away from the reference feature at (0,0)
395: in the SE maser group. A line connecting the two regions lies at a position
396: angle of roughly $-$29$^{\circ}$ and the two regions are separated
397: by 488 milliarcseconds (mas) or 156 AU. The SE grouping is
398: roughly arc-like with an overall extent of 38 mas (12 AU) and a
399: thickness of 2 mas (0.6 AU), while the NW clump is more linear
400: and extends about 26 mas (8.3 AU) with a thickness roughly the
401: same as the SE clump. The highest velocity maser emission lies at the
402: northern tip of the NW group (+20 \kms) and at the southern
403: tip of the SE arc ($-$3 \kms). In addition, a small maser group
404: is located approximately 100 mas (32 AU) to the NNW of the main SE arc.
405: This region's structure changed during the epochs, but usually consisted of
406: two or three well-separated maser features (by several mas or about 0.3 AU).
407: This distinct region is located close to the line connecting the
408: NW and SW maser groups. If the active maser regions mark the interaction of a
409: bipolar jet from IRAS~4BW with ambient gas, then the opening angle of the jet
410: would be $\sim$10$^{\circ}$.
411:
412: Beginning in the second epoch, but much more clearly shown in the third
413: and fourth epochs, a new arc-like ridge of lower velocity emission
414: became visible to the south of the other masers in the NW region.
415: This region is not well-characterized by point-like emission, but is
416: more diffuse and seems to be an extended ridge of maser emission.
417: By the fourth epoch, the entire NW feature had grown to roughly 40 mas (12.8 AU) in extent.
418:
419: We compared the positions of the reference feature, which is
420: present in all epochs in the SE clump (B-REF in Table 4),
421: with a maser in the NW clump that persisted through all four epochs (B-5 in Table 4).
422: The separation vector for the two masers
423: increased over 59.81262 days
424: from 499.59 mas (159.87 AU) to 508.25 mas (162.64 AU). A linear
425: regression fit to the magnitude of the separation vector as a
426: function of time yields an expansion velocity of 78.2 \kms
427: with a formal error of $\pm$ 3.9 \kms. The radial velocity difference
428: between the two masers of 18.5 \kms suggests an inclination of the outflow
429: of about 13$^{\circ}$ from the plane of the sky.
430: %The position angle between the two components changes with time as well since
431: %proper motion of the maser does not coincide with the separation vector.
432:
433: To investigate the proper motions of the masers associated with IRAS~4BW,
434: we identified all components that were present in all four epochs of observation.
435: Three components were identified in the NW and two were identified in the SE.
436: We averaged the mean
437: positions of the NW and SE components and averaged these two positions to
438: estimate an average geometric fiducial position for each epoch. This
439: position was then subtracted from each maser component. These referenced
440: positions are presented along with the fluxes and original positions in
441: Table 4. Finally, we fit straight lines to the positions
442: of the components as a function of time. The resulting tangential
443: velocities and position angles from these fits are also presented in
444: Table 4 and are very similar for all five maser components.
445: Their motions are very uniform over the four epochs with average values of
446: 43 $\pm$ 2 \kms at a position angle of $-$38$^{\circ}$ $\pm$ 2$^{\circ}$.
447: Figure 4 shows all of the maser components as observed in
448: Epoch 1 and their corresponding proper motion vectors. Again, an overall
449: expansion is apparent along the axis connecting the two regions.
450:
451: \section{Discussion}
452:
453: \subsection{IRAS~4A\label{discussion=4a}}
454:
455: Since the water masers toward IRAS~4A have always been found coincident
456: with the position of A2 (when they were detected), we consider
457: three alternatives for the origin of the water maser emission seen using the
458: VLBA: 1) the water masers are in an outflow which is the close-in extension
459: of the large-scale molecular outflow, driven by IRAS~4A2; 2) the water masers
460: are excited by an interaction of the close-in flow and the accretion disk
461: itself and 3) the water maser outflow is driven by a companion to IRAS~4A2,
462: as yet unresolved. We consider each of these alternatives in turn.
463:
464: \subsubsection{Alignment with the large scale outflow}
465:
466: The alignment of the water maser outflow is curious in that it does not
467: have the same position angle as any of the larger-scale outflows emanating from the
468: two sources in IRAS~4A. The axis of the large-scale outflow defined by the
469: SiO, CS, CO, and H$_2$ emission \citep{Choi:2006,Choi:2005, BLAKE:1995},
470: and driven by A2, may be drifting with a rate of 0.011 degrees/yr
471: (counterclockwise as seen on the sky) and a zero-point angle of around $-$166$^{\circ}$
472: (i.e., presumably the jet-injection angle), depending upon the
473: fit of the large-scale emission (see Figure 3a of Choi et al. 2006).
474: For the large-scale flow, the blue-shifted emission is generally in the
475: south to southwest and the red-shifted emission is generally in the north and northeast.
476: The maser outflow, however, on a scale of ~200 mas, has a position
477: angle of $-$50$^{\circ}$, and has the more highly red-shifted emission to the southeast.
478: %and blue-shifted emission to the northwest.
479: Thus, if the masers were
480: marking the close-in extension to the flow, the position angle of the
481: flow would have to change from $+$130$^{\circ}$ to $-$165$^{\circ}$, in a counterclockwise
482: direction, from 40 to 300 AU from the star. We consider that such a shift in
483: jet direction is very unlikely, if not unphysical.
484:
485: \subsubsection{Interaction with the A2 accretion disk}
486:
487: The jet direction near the base of the flow, as indicated by \citet{Choi:2006},
488: and also by the CS J$=$7$\rightarrow$6 emission
489: \citep{BLAKE:1995} appears to be about 0 degrees (i.e. north-south).
490: If the masers were excited by an interaction of the outflow with the
491: A2 accretion disk itself, then the opening half-angle of the jet at its
492: base would have to encompass the direction of the maser axis. Thus the
493: opening half-angle of the jet would be $\sim50\deg$. Such a large opening
494: angle would seem inconsistent with the Class 0 phase and the high degree of collimation seen
495: in the larger scale outflow.
496:
497: \subsubsection{An unresolved companion to IRAS~4A2}
498:
499: Perhaps the simplest explanation for the direction of the axis of the
500: water maser emission is that it is not related directly at all to the
501: large-scale molecular outflow, but that the masers are located in an
502: independent outflow from another young stellar object. A large
503: fraction of the water masers around low-luminosity YSOs, which have been
504: observed and mapped with very high angular resolution, show bipolar flows
505: and structural evidence for bow shocks as the jet impinges on and sweeps up
506: ambient molecular material. Although the structural evidence is sparse
507: (because of the small number of maser spots) for IRAS~4A, the bipolar
508: velocity separation strongly suggests that the masers are tracing a jet
509: outflow as, for example, in IRAS~05413-0104 \citep{CLAUSSEN:1998}. It is
510: clear from VLA observations (e.g., Furuya et al. 2003; this paper) that the
511: water masers toward IRAS~4A are always found toward A2 rather than A1,
512: so A1 cannot drive the maser outflow. Therefore we suggest, that,
513: in this scenario, there must be a companion to A2 that drives
514: the water masers. Since the masers are clearly associated with the
515: position of A2, and not A1, it is clear that the maser outflow, in this
516: scenario, is {\bf not} associated with A1. Indeed,
517: the presence of a companion to A2 separated by 30-80 AU has been proposed
518: by \citet{Choi:2006} to explain the drifting of the 4A2 outflow axis. A
519: companion within 100 mas (32 AU) of A2 would be consistent with it not
520: being resolved in $\lambda$=3.6 cm continuum observations with the VLA
521: \citep{REIPURTH:2000}.
522: % Indeed, we think that the bulk of observational evidence points to
523: % the conclusion that A1 is not currently a star, and this cannot
524: % drive such an outflow.
525:
526: \subsection{IRAS~4B\label{discussion=4b}}
527:
528: While the masers trace a jet similar in velocity to the HCN outflow,
529: the position angle of $-$29$^{\circ}$ is significantly different from the north-south
530: orientation of the outflow. \citet{Choi:2001} suggests a dynamical lifetime
531: for the molecular outflow of $\sim$200 years (corrected to a distance of 320 pc).
532: This is much longer than the dynamical lifetime of the maser emission
533: which is on the order of a year. We conclude that the jet must be precessing
534: in such a way as to move the projected outflow axis from a position angle
535: of 0$^{\circ}$ to $-$29$^{\circ}$ in several hundred years. In this picture, the
536: jet would be impacting the walls of the outflow cavity as it precessed
537: ``clockwise" in the plane of the sky. The NW maser group would represent
538: the western wall of the pre-existing cavity and the SE maser group the
539: eastern wall. This would explain the linear, arc-like structure of the
540: maser groups and the appearance of new masers at the southern extent of the NW group.
541: This model can only be reconciled with observations of dense outflowing gas traced
542: %subjunctive inserted
543: in H$_2$CO by \citet{DIFRANCESCO:2001} if there were a second outflow from an
544: unresolved companion to IRAS~4BW.
545: %predictions
546: If our picture is correct, one would expect future VLBI observations to
547: reveal new masers south (north) of the NW (SE) group. One would also
548: expect that on a larger scale, tracers of the jet would display a
549: characteristic S-shape, bending back to a north-south orientation.
550: Indeed, slow precessing jets
551: have been proposed for a number of YSO outflows \citep{Matthews:2006,Hodapp:2005,
552: %%slight word change made.
553: Terquem:1999} that suggest the presence of a nearby companion \citep{Terquem:1998}.
554: %required precession rate, cause of precession
555:
556: The proper motion vectors for the five persistent maser spots relative to a
557: fiducial point (Figure 4) suggests a larger position angle than that of the outflow axis
558: defined by the relative motion of SE and NW masers ($-$38$^{\circ}$ vs. $-$29$^{\circ}$, see Table 4).
559: This is reminiscent of the proper motions seen in the triple radio continuum source
560: in Serpens \citep{Curiel:1993}, where the outer components of the radio jet have the
561: same tangential velocity but are moving in slightly different directions. Although the
562: maser emission in IRAS~4B traces the jet at scales 40 times smaller than that traced by
563: the radio continuum in the Serpens jet, this difference in relative motions may be
564: suggestive of similar physical processes in the two jets.
565:
566: The larger position angle defined by the proper motion vectors in IRAS~4B suggest that,
567: in addition to the tangential velocity imparted to the
568: masing gas by the outflow jet at a position angle of $-$29$^{\circ}$, there is an
569: additional component of about 6 \kms perpendicular to the outflow axis.
570: It is possible that this motion is related to the expansion of the outflow cavity into the ambient cloud.
571:
572: \subsection{Maser Geometry}
573:
574: We estimate that the maser outflows in both IRAS~4A and 4B are nearly in the plane of the sky
575: (inclination of only 2$^{\circ}$ for IRAS~4A and about 13$^{\circ}$ for IRAS~4B). In
576: Claussen et al. (1998) we estimated that the inclination of the maser outflow for IRAS~05413-0104
577: was only 4$^{\circ}$. Models of maser emission (e.g. Hollenbach 1997) suggest that the
578: masers trace shocks produced by the interaction of the jet and ambient material along the
579: jet. Shocks that propagate close to the plane of the sky provide much longer maser coherence
580: lengths than those that propagate at larger inclination angles. Thus outflows that have
581: inclinations close to the plane of the sky provide a more favorable maser geometry, perhaps
582: explaining why some well-known outflows from low-mass YSOs do not show maser emission.
583:
584: \section{Conclusions}
585:
586: We have observed the water masers associated with IRAS~4A and 4B at VLBI resolutions in four epochs
587: over three months. We have determined that the masers are related to the jets emanating from these
588: YSOs due to their spatio-kinematic distribution. In both sources the masers are found associated with known single components of multiple systems, with total separation velocities between 53 and 78 km s$^{-1}$.
589:
590: This is further confirmed by the large proper motions measured
591: for both sources, which clearly rules out rotation due to the mass constraints placed on the
592: central objects by other observations. The water masers of IRAS~4B form arc-like structures
593: roughly 10 AU in length and less than 0.6 AU in thickness. The structure of these structures changes
594: rapidly with time, with many new maser components appearing and disappearing in just one month. The orientation of the structures in the plane of the sky does not agree with larger scale outflow angle, which we attribute to a possible unseen very close companion.
595: Future observations will have to sample the source more frequently than once every three weeks, with
596: once every 3-5 days probably providing the best results.
597:
598:
599: \acknowledgments
600:
601: KBM wishes to thank the AAS Council and AAS Executive Officer,
602: Dr. Robert Milkey for allowing him to continue research work in
603: his current position. KBM thanks TEK for her continued support.
604: And special thanks to T. Beasley for exposing KBM to the rigors of
605: Outback observing and a grueling exercise regimen, during which
606: time a portion of this paper was written. Finally, KBM thanks
607: the NRAO for hosting him during a research leave from the AAS, during
608: which time the bulk of this paper was completed.
609:
610: \appendix
611: \section{Table of Maser Components for IRAS~4B}
612: Maser component fits were averaged using a flux-squared weighting scheme, that
613: emphasizes the strongest portion of the maser emission and produces a reliable location for
614: the maser. As some masers are present for only three channels, a full Gaussian fit to the
615: maser as a function of velocity was deemed impractical, although it could have been used for
616: the masers that existed over many channels. We opted to use the same position-determining
617: scheme for all features. These averaged component fits and their corresponding
618: errors are given in Table 5 for IRAS~4B. Note that the uncertainty in position for the reference
619: channel is simply the error on the mean position values determined for the spot and is
620: indicative of the reliability of the Gaussian fitting routines in the presence of noise
621: in the images.
622: %{\bf NEED TO ADDRESS TRUE UNCERTAINTY IN POSTIONS HERE }
623:
624: \clearpage
625:
626: \begin{thebibliography}{}
627: %
628: \bibitem[Andr\'e, et al. (2000)] {Andre:2000} Andr\'e, P., Ward-Thompson, D., \& Barsony, M.
629: \ 2000, in Protostars and Planets IV, eds. V. Mannings, A. P. Boss, \& S. S. Russell, Univ.
630: of Arizona Press: Tucson, p. 59
631:
632: \bibitem[Aspin et al.(1994)]{ASPIN:1994} Aspin, C., Sandell, G.,
633: \& Russell, A.~P.~G.\ 1994, \aaps, 106, 165
634: %
635: \bibitem[Blake et al.(1995)]{BLAKE:1995} Blake, G.~A., Sandell,
636: G., van Dishoeck, E.~F., Groesbeck, T.~D., Mundy, L.~G., \& Aspin, C.\
637: 1995, \apj, 441, 689
638: %
639: % \bibitem[Bottinelli et al.(2004)]{Bottinelli:2004} Bottinelli, S., Ceccarelli, C., Lefloch, B., Williams,
640: % J. P., Castets, A., caux, E., Cazaux, S., MAret, S., Parise, B., and Tielens, A. G. G. M.\
641: % 2004, \apj, 615, 354
642: %
643: \bibitem[Bottinelli et al.(2007)]{Bottinelli:2007} Bottinelli, S., Ceccarelli, C., Williams, J. P., \& Lefloch, B.\
644: 2007, \aap, 463, 601
645: %
646: \bibitem[Brand et al.(2003)]{Brand:2003} Brand, J., Cesaroni, R., Comoretto, G.,
647: Felli, M., Palagi, F., Palla, F., \& Valdettaro, R.\ 2003,
648: \aap, 407, 573
649: %
650: \bibitem[Choi(2001)]{Choi:2001} Choi, M.\ 2001, \apj, 553, 219
651: %
652: % \bibitem[Choi et al.(1999)]{CHOI:1999} Choi, M., Panis, J., \&
653: % Evans, N.~J.\ 1999, \apjs, 122, 519
654: %
655: \bibitem[Choi(2005)]{Choi:2005} Choi, M.\ 2005, \apj, 630, 976
656: %
657: \bibitem[Choi et al.(2006)]{Choi:2006} Choi, M., Hodapp, K., Hayashi, M., Motohara, K., Pak, S., \& Pyo, T.-S\ 2006, \apj, 646, 1050
658: %
659: \bibitem[Claussen et al.(1998)]{CLAUSSEN:1998}
660: Claussen, M.~J., Marvel, K.~B., Wootten, A., \& Wilking, B.~A.\ 1998,
661: \apjl, 507, L79
662: %
663: \bibitem[Claussen et al.(1996)]{CLAUSSEN:1996} Claussen, M.~J.,
664: Wilking, B.~A., Benson, P.~J., Wootten, A., Myers, P.~C., \& Terebey, S.\
665: 1996, \apjs, 106, 111
666: %
667: \bibitem[Curiel et al.(1993)]{Curiel:1993} Curiel, S., Rodriguez, L. F., Moran, J. M.,
668: \& Canto, J. \ 1993, \apj, 415, 191
669: %
670: \bibitem[Desmurs et al.(2006)]{Desmurs:2006} Desmurs, J.-F., Codella, C., Santiago-Garcia, J., Tafalla, M., \& Bachiller, R. \ 2006, in Proceedings of the 8th European VLBI Network Symposium, ed. A. Marecki, p.65
671: %
672: \bibitem[de Zeeuw et al.(1999)]{HIPPARCOS} de Zeeuw, P.~T.,
673: Hoogerwerf, R., de Bruijne, J.~H.~J., Brown, A.~G.~A., \& Blaauw, A.\ 1999,
674: \aj, 117, 354
675: %
676: \bibitem[Di Francesco et al.(2001)]{DIFRANCESCO:2001} Di Francesco, J.,
677: Myers, P.~C., Wilner, D.~J., Ohashi, N., \& Mardones, D.\ 2001, \apj, 562,
678: 770
679: %
680: \bibitem[Elitzur, Hollenbach, \& McKee(1992)]{MODEL1} Elitzur,
681: M., Hollenbach, D.~J., \& McKee, C.~F.\ 1992, \apj, 394, 221
682: %
683: % \bibitem[Elitzur(1992)]{ELITZURBOOK} Elitzur, M.\ 1992,
684: % Astronomical Masers, Kluwer Academic Publishers (Astrophysics and Space
685: % Science Library.~Vol.~170), 365 p.
686: %
687: \bibitem[Elitzur, Hollenbach, \& McKee(1989)]{MODEL2} Elitzur,
688: M., Hollenbach, D.~J., \& McKee, C.~F.\ 1989, \apj, 346, 983
689: %
690: \bibitem[Furuya et al.(2003)]{Furuya:2003} Furuya, R.~S., Kitamura,
691: Y., Wootten, H.~A., Claussen, M.~J., \& Kawabe,
692: R.\ 2003, \apjs, 144, 71
693: %
694: \bibitem[Furuya et al.(2000)]{FURUYA:2000} Furuya, R.~S., Kitamura,
695: Y., Wootten, H.~A., Claussen, M.~J., Saito, M., Marvel, K.~B., \& Kawabe,
696: R.\ 2000, \apjl, 542, L135
697: %
698: % \bibitem[Furuya(2002)]{FURUYA:PRIV} Furuya, R.~S.,\ 2002 {\it private communication}
699: %
700: \bibitem[Furuya et al.(2005)]{FURUYA:2005} Furuya, R.~S., Kitamura,
701: Y., Wootten, A., Claussen, M.~J., \& Kawabe, R.\ 2005, \aap, 438, 571
702: %
703: \bibitem[Getman et al.(2002)]{Getman:2002} Getman, K. V., Feigelson, E. D.,
704: Townsley, L., Bally, J., Lada, C. J., \& Reipurth, B.\ 2002, \apj, 575, 354
705: %
706: \bibitem[Girart, Crutcher, \& Rao(1999)]{Girart:1999} Girart, J., Crutcher, R., \&
707: Rao, R.\ 1999, \apj, 525, L109
708: %
709: \bibitem[Greisen (2003)]{Greisen:2003} Greisen, E. W. 2003, in Information
710: Handling in Astronomy, ed. A. Heck (Dordrecht: Kluwer), 109
711: %
712: \bibitem[Harvey, Wilking, \& Joy(1984)]{Harvey:1984} Harvey, P., Wilking,
713: B., \& Joy, M.\ 1984, \apj, 278, 156
714: %
715: \bibitem[Haschick et al.(1980)]{HASCHICK:1980} Haschick, A.~D.,
716: Moran, J.~M., Rodriguez, L.~F., Burke, B.~F., Greenfield, P., \&
717: Garcia-Barreto, J.~A.\ 1980, \apj, 237, 26
718: %
719: % \bibitem[Herbig \& Jones(1983)]{HERBIG:1983} Herbig, G.~H., \&
720: % Jones, B.~F.\ 1983, \aj, 88, 1040
721: %
722: \bibitem[Hodapp et al.(2005)]{Hodapp:2005} Hodapp, K., Bally, J., Eisloffel, J. \& Davis,
723: C.\ 2005, \aj, 129, 1580
724: %
725: % \bibitem[Hodapp \& Ladd(1995)]{HODAPP:1995} Hodapp, K., \& Ladd,
726: % E.~F.\ 1995, \apj, 453, 715
727: %
728: \bibitem[Hollenbach(1997)]{Hollenbach:1997} Hollenbach, D. 1997, in IAU Symp. 182,
729: Herbig-Haro Flows and the Birth of Low Mass Stars, ed. B. Reipurth \& C. Bertout
730: (Dordrecht: Kluwer), 181
731: %
732: \bibitem[Jennings et al.(1987)]{JENNINGS:1987} Jennings, R.~E.,
733: Cameron, D.~H.~M., Cudlip, W., \& Hirst, C.~J.\ 1987, \mnras, 226, 461
734: %
735: \bibitem[Lada, Alves, \& Lada(1996)]{Lada:1996} Lada, C.~J.,
736: Alves, J., \& Lada, E.~A.\ 1996, \aj, 111, 1964
737: %
738: \bibitem[Lay, Carlstrom, \& Hills(1995)]{LAY:1995} Lay, O.~P.,
739: Carlstrom, J.~E., \& Hills, R.~E.\ 1995, \apjl, 452, L73
740: %
741: \bibitem[Looney, Mundy, \& Welch(2000)]{LOONEY:2000} Looney, L.~W.,
742: Mundy, L.~G., \& Welch, W.~J.\ 2000, \apj, 529, 477
743: %
744: \bibitem[MacLow et al.(1994)]{MODEL3} MacLow
745: M., Elitzur, M., Stone, J.~M., \& Konigl, A.\ 1994, \apj, 427, 914
746: %
747: \bibitem[MacLow \& Elitzur(1992)]{MODEL4} MacLow, M.~\&
748: Elitzur, M.\ 1992, \apjl, 393, L33
749: %
750: \bibitem[Marvel et al.(2002)]{MARVEL:2002} Marvel, K., Claussen,
751: M., \& Wootten, A.\ 2002, IAU Symposium, 206, 63
752: %
753: \bibitem[Matthews et al.(2006)]{Matthews:2006} Matthews, B., Hogerheijde, M., Jorgensen, J. \& Bergin,
754: E.\ 2006, \apj, 652, 1374
755: %
756: \bibitem[Moscadelli et al.(2006)]{Moscadelli:2006} Moscadelli, L., Testi,
757: L., Furuya, R., Goddi, C., Claussen, M., Kitamura, Y., \& Wootten, A.\ 2006, \aap, 446, 985
758: %
759: % \bibitem[Mundy et al.(1993)]{MUNDY:1993}
760: % Mundy, L.~G., McMullin, J.~P., Grossman, A.~W., \& Sandell, G.\ 1993,
761: % Icarus, 106, 11
762: %
763: \bibitem[Preibisch(1997)]{PREIBISCH:1997} Preibisch, T.\ 1997, \aap,
764: 324, 690
765: %
766: \bibitem[Preibisch(2003)]{PREIBISCH:2003} Preibisch, T.\ 2003, \aap,
767: 401, 543
768: %
769: \bibitem[Reipurth et al.(2002)]{REIPURTH:2002} Reipurth, B.,
770: Rodr{\'{\i}}guez, L.~F., Anglada, G., \& Bally, J.\ 2002, \aj, 124, 1045
771: %
772: \bibitem[Reipurth(2000)]{REIPURTH:2000} Reipurth, B.\ 2000, \aj, 120,
773: 3177
774: %
775: % \bibitem[Rodr{\'{\i}}guez et al.(1999)]{RODRIGUEZ:1999}
776: % Rodr{\'{\i}}guez, L.~F., Anglada, G., \& Curiel, S.\ 1999, \apjs, 125, 427
777: %
778: \bibitem[Rodr{\'{\i}}guez et al.(2002)]{RODRIGUEZ:2002}
779: Rodr{\'{\i}}guez, L.~F., Anglada, G., Torrelles, J.~M., Mendoza-Torres,
780: J.~E., Haschick, A.~D., \& Ho, P.~T.~P.\ 2002, \aap, 389, 572
781: %
782: % \bibitem[Rogers \& Gottschalk(1993)]{ROGERS:1993} Rogers, C.~B.~\&
783: % Gottschalk, J.\ 1993, Bulletin of the American Astronomical Society, 25,
784: % 1367
785: %
786: \bibitem[Sandell et al.(1991)]{SANDELL:1991} Sandell, G., Aspin, C.,
787: Duncan, W.~D., Russell, A.~P.~G., \& Robson, E.~I.\ 1991, \apjl, 376, L17
788: %
789: \bibitem[Sandell \& Knee(2001)]{Sandell:2001} Sandell, G. \& Knee, L. B. G.\ 2001, \apj, 546, L49
790: %
791: \bibitem[Strom et al.(1976)]{STROM:1976} Strom, S.~E., Vrba,
792: F.~J., \& Strom, K.~M.\ 1976, \aj, 81, 314
793: %
794: \bibitem[Terquem(1998)]{Terquem:1998} Terquem, C.\
795: 1998, \apj, 509, 819
796: %
797: \bibitem[Terquem et al.(1999)]{Terquem:1999} Terquem, C., Eisloffel, J., Papaloizou, J. \&
798: Nelson, R.\ 1999, \apj, 512, L131
799: %
800: \bibitem[Torrelles et al.(2001)]{TORRELLES:2001} Torrelles, J.~M., et
801: al.\ 2001, \apj, 560, 853
802: %
803: \bibitem[Seth et al.(2002)]{SETH:2002} Seth, A.~C., Greenhill,
804: L.~J., \& Holder, B.~P.\ 2002, \apj, 581, 325
805: %
806: \bibitem[Shah(2000)]{Shah:2000} Shah, R. \ 2000, Ph.D. Thesis, University of Virginia
807: %
808: \bibitem[Smith et al.(2000)]{SMITH:2000} Smith, K.~W., Bonnell,
809: I.~A., Emerson, J.~P., \& Jenness, T.\ 2000, \mnras, 319, 991
810: %
811: \bibitem[Wilking et al.(1994)]{WILKING:1994} Wilking, B.~A.,
812: Claussen, M.~J., Benson, P.~J., Myers, P.~C., Terebey, S., \& Wootten, A.\
813: 1994, \apjl, 431, L119
814: %
815: \bibitem[Wilking et al.(2004)]{Wilking:2004} Wilking, B.~A.,
816: Meyer, M.~R., Greene, T.~P., Mikhail, A., \& Carlson, G.\
817: 2004, \aj, 127, 1131
818: %
819: \bibitem[Wootten \& Mangum(1993)]{WOOTTEN:1993} Wootten, A.~\&
820: Mangum, J.\ 1993, Bulletin of the American Astronomical Society, 25, 1367
821: %
822: \bibitem[Wootten(1998)]{Wootten:1998} Wootten, H. A., 1998, {\it unpublished observations}
823: %
824: \bibitem[Wootten et al.(1999)]{WOOTTEN:1999} Wootten, A., Claussen,
825: M., Marvel, K., \& Wilking, B.\ 1999, The Physics and Chemistry of the
826: Interstellar Medium, Proceedings of the 3rd Cologne-Zermatt Symposium, held
827: in Zermatt, September 22-25, 1998, Eds.: V.~Ossenkopf, J.~Stutzki, and
828: G.~Winnewisser, GCA-Verlag Herdecke, ISBN 3-928973-95-9, 295
829: %
830: \bibitem[Wootten et al.(2002)]{WOOTTEN:2002} Wootten, A., Claussen,
831: M., Marvel, K., \& Wilking, B.\ 2002, IAU Symposium, 206, 100
832:
833: %
834: \end{thebibliography}
835: %
836: % tables
837: %
838: \clearpage
839: %
840: %Table 1 - IRAS~4A and 4B Properties
841: \include{IRAS4AB.tab1}
842: %
843: \clearpage
844: %
845: %Table 2 - Dates/times of observation
846: \include{IRAS4AB.tab2}
847: %
848: \clearpage
849: %
850: %Table 3 - Imaging parameters for each epoch
851: \include{IRAS4AB.tab3}
852: %
853: %Table 4 - Proper Motions for Masers Detected in All Four Epochs
854: \include{IRAS4AB.tab4}
855: \clearpage
856: %
857: %Table 5 - Guassian Fits for All Maser Components
858: \include{IRAS4AB.tab5}
859: \clearpage
860: %
861: % The Figures
862: %
863: %
864: %
865: \begin{figure}
866: \includegraphics[scale=0.25]{fig1a.eps}
867: \includegraphics[scale=0.25]{fig1b.eps}
868: \includegraphics[scale=0.25]{fig1c.eps}
869: \includegraphics[scale=0.25]{fig1d.eps}
870:
871: \caption{
872: The spatial and kinematic distribution of the water masers associated with IRAS~4A2 for the
873: four epochs observed. The top frame in each panel shows the velocity and flux of the detected
874: features, the lower frame shows the
875: spatial distribution. The
876: maser to the NW is colored blue (at the cloud velocity) and those to the SE are coded red
877: (red-shifted relative to the cloud velocity). The size of the circles
878: in the figure is proportional to the flux of that maser component. All components measured in
879: each channel are presented (e.g., no component averaging has been performed). The error in the
880: position of the spots (typically less than 20 $\mu$arcseconds) cannot be plotted on the large scale
881: figure.
882: % My figure had no vertical bar. The vertical bar represents a size of 16 AU at the assumed distance of IRAS~4A of 320 pc.
883: }
884: \notetoeditor{This four-panel color figure should appear on a single page and
885: each of the four panels should be legible.}
886: \end{figure}
887: %
888: \clearpage
889: \begin{figure}
890: \includegraphics[scale=0.43]{fig2a.eps}
891: \includegraphics[scale=0.43]{fig2b.eps}
892: \includegraphics[scale=0.43]{fig2c.eps}
893: \includegraphics[scale=0.43]{fig2d.eps}
894: \caption{
895: Spatial and kinematic distribution of the blueshifted masers southeast of IRAS~4BW for all four epochs.
896: In each panel, the top frame shows the velocity (color-coded) and the flux (symbol size) of the detected
897: features, and the lower frame shows the spatial distribution and the velocity distribution as a function
898: of declination. The uncertainty in the fitted position of each feature circle cannot be displayed at this
899: scale, but in the worst case is of order 20 $\mu$arcseconds.
900: }
901: \notetoeditor{This four-panel color figure should appear on a single page and
902: each of the four panels should be legible.}
903: \end{figure}
904: %
905: \clearpage
906:
907: %
908: \begin{figure}
909: \includegraphics[scale=0.43]{fig3a.eps}
910: \includegraphics[scale=0.43]{fig3b.eps}
911: \includegraphics[scale=0.43]{fig3c.eps}
912: \includegraphics[scale=0.43]{fig3d.eps}
913: \caption{
914: Spatial and kinematic distribution of the redshifted masers northwest of IRAS~4BW for all four epochs.
915: In each panel, the top frame shows the velocity (color-coded) and the flux (symbol size) of the detected
916: features, and the lower frame shows the spatial distribution and the velocity distribution as a function
917: of declination. The uncertainty in the fitted position of each feature circle cannot be displayed at this
918: scale, but in the worst case is of order 20 $\mu$arcseconds.
919: }
920: \notetoeditor{This four-panel color figure should appear on a single page and
921: each of the four panels should be legible.}
922: \end{figure}
923: %
924: %
925: \clearpage
926: \begin{figure}
927: \includegraphics[scale=0.90]{fig4.eps}
928: \caption{
929: A plot of the proper motions for the five maser components present in all four epochs for IRAS~4BW relative
930: to their positions in Epoch 1 (see Table 4). The length of the proper motion vectors are 2 \kms per plot unit.
931: Note that the proper motions are not at the same position angle as a line connecting the two maser groups
932: (see text).
933: }
934: \end{figure}
935:
936: \end{document}
937:
938: