VLBI study of maser kinematics in high-mass SFRs. II. G23.01-0.41
A. Sanna, L. Moscadelli, R. Cesaroni, A. Tarchi, R. S. Furuya, C. Goddi
aa r X i v : . [ a s t r o - ph . GA ] A p r Astronomy&Astrophysicsmanuscript no. SANNApaperG23 c (cid:13)
ESO 2018November 13, 2018
VLBI study of maser kinematics in high-mass SFRs. II . G23.01–0.41 A. Sanna , , L. Moscadelli , R. Cesaroni , A. Tarchi , R. S. Furuya , and C. Goddi , Dipartimento di Fisica, Universit`a degli Studi di Cagliari, S.P. Monserrato-Sestu km 0.7, I-09042 Cagliari, Italy INAF, Osservatorio Astronomico di Cagliari, Loc. Poggio dei Pini, Strada 54, 09012 Capoterra (CA), Italy INAF, Osservatorio Astrofisico di Arcetri, Largo E. Fermi 5, 50125 Firenze, Italy Subaru Telescope, National Astronomical Observatory of Japan, 650 North A’ohoku Place, Hilo, HI 96720, USA European Southern Observatory, Karl-Schwarzschild-Strasse 2, D-85748 Garching bei M¨unchen, Germany Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA 02138, USAPreprint online version: November 13, 2018
ABSTRACT
Aims.
We performed a detailed study of maser and radio continuum emission toward the high-mass star-forming region G23.01–0.41.This study aims at improving our knowledge of the high-mass star-forming process by comparing the gas kinematics near a newlyborn young stellar object (YSO), analyzed through high spatial resolution maser data, with the large-scale environment of its nativehot molecular core (HMC), identified in previous interferometric observations of thermal continuum and molecular lines.
Methods.
Using the VLBA and the EVN arrays, we conducted phase-referenced observations of the three most powerfulmaser species in G23.01–0.41: H O at 22.2 GHz (4 epochs), CH OH at 6.7 GHz (3 epochs), and OH at 1.665 GHz (1 epoch).In addition, we performed high-resolution ( ≥ . ′′ ), high-sensitivity ( < . mJy) VLA observations of the radio continuumemission from the HMC at 1.3 and 3.6 cm. Results.
We have detected H O, CH OH, and OH maser emission clustered within 2000 AU from the center of a flattened HMC,oriented SE-NW, from which emerges a massive CO outflow, elongated NE-SW, extended up to the pc-scale. Although the threemaser species show a clearly di ff erent spatial and velocity distribution and sample distinct environments around the massive YSO,the spatial symmetry and velocity field of each maser specie can be explained in terms of expansion from a common center, whichpossibly denotes the position of the YSO driving the maser motion. Water masers trace both a fast shock (up to 50 km s − ) closerto the YSO, powered by a wide-angle wind, and a slower (20 km s − ) bipolar jet, at the base of the large-scale outflow. Since thecompact free-free emission is found o ff set from the putative location of the YSO along a direction consistent with that of the maserjet axis, we interpret the radio continuum in terms of a thermal jet. The velocity field of methanol masers can be explained in termsof a composition of slow (4 km s − in amplitude) motions of radial expansion and rotation about an axis approximately parallel tothe maser jet. Finally, the distribution of line of sight velocities of the hydroxyl masers suggests that they can trace gas less dense(n H ≤ cm − ) and more distant from the YSO than that traced by the water and methanol masers, which is expanding toward theobserver. A few pairs of OH masers, with di ff erent circular polarization, are well aligned in position on the sky and we interpret themas Zeeman pairs. From Zeeman splitting, the derived typical values of the magnetic field are of a few mG. Key words.
Masers – Techniques: high angular resolution – ISM: kinematics and dynamics – Stars: formation – Stars: individual:G23.01–0.41
1. Introduction
Hot, dense, molecular cores (HMCs) of dust and gas locatedin giant molecular clouds (GMCs) are the birth sites of high-mass young stellar objects (YSOs). Ultraviolet light from thenewly formed stars ionizes the surrounding gas, creating a com-pact H ii region (e.g., Hoare et al. 2007), excites several molecu-lar transitions (e.g., Sridharan et al. 2002), and heats the dust,which reradiates the energy in the far infrared (FIR) band(e.g., Molinari et al. 2008). At the radio wavelengths, maseremissions of several molecular transitions are observed duringthe earliest evolutionary phases of high-mass (proto)stars (e.g.,Szymczak et al. 2005), even before the appearance of an ul-tra compact H ii region (UCH ii ). From an observational pointof view, however, the study of high-mass star-forming regions(HMSFRs) is challenging because of three main limitations: thedusty environment is obscured at optical / NIR frequencies; timescales of massive star formation are short, and hence, the chanceof observing massive YSOs is small; massive YSOs are rare and
Send o ff print requests to : A. Sanna, e-mail: [email protected] statistically found far away from the observer, at distances of afew kpc, clustered in tight associations.Since a few years, we have started an observational cam-paign to study the high-mass star-forming process by compar-ing interferometric thermal data, tracing the large-scale environ-ment (e.g., Codella et al. 1997; Furuya et al. 2008), with VeryLong Baseline Interferometry (VLBI) measurements of masertransitions, tracing the inner kinematics of the (proto)stellar co-coon. Details about our VLBI observational program to mea-sure molecular masers in HMSFRs are extensively presented inSanna et al. (2010, hereafter Paper I). The present paper focuseson our observations and analysis of the HMSFR G23.01 − O), 6.7 GHzmethanol (CH OH) and 1.665 GHz hydroxyl (OH) maser tran-sitions, together with the new Very Large Array (VLA) obser-vations of the radio continuum emission at 1.3 and 3.6 cm. InSect. 4, we illustrate the spatial morphology, kinematics, andtime-variability of individual maser species, and present results
A. Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 from our VLA observations, constraining the properties of theradio continuum observed associated with the masers. Section 5discusses the spatial association of the maser species and theiroverall kinematics, and draws a comprehensive picture of thephenomena observed in the HMSFR G23.01–0.41 on angularscales from a few mas to tens of arcsec. Main conclusions aresummarized in Sect. 6.
2. The HMSFR G23.01–0.41
The HMSFR G23.01–0.41 is located at a distance of4.59 + . − . kpc (Brunthaler et al. 2009), has a large-scale clumpmass of about 4 × M ⊙ (Furuya et al. 2008), and a bolometricluminosity of about 10 L ⊙ (upper limit) inferred from IR data(Araya et al. 2008) . From observations of NH and CH CNtransitions, the systemic velocity of the region (V sys ) with re-spect to the local standard of rest (LSR) was determined to be77.4 km s − (Codella et al. 1997; Furuya et al. 2008).On an arcminute scale, Spitzer IRAC GLIMPSE observa-tions show strong 4.5 µ m excess indicative of shocked gas(Araya et al. 2008 and references therein), at the position ofa molecular clump extended over an area of about 30 ′′ × ′′ , imaged with the Nobeyama Millimeter Array (NMA)in the CO (1 −
0) and C O (1 −
0) lines (Furuya et al.2008). Associated with the extended mid-IR emission, a mas-sive CO (1 −
0) bipolar outflow was observed, with a promi-nent red-wing emission, elongated in the NE–SW direction.The same outflow was also detected in the CO (1 −
0) andHNCO (5 − ) lines (Furuya et al. 2008). Single-dish ob-servations toward G23.01–0.41 detected several, both thermaland maser, molecular transitions. Anglada et al. (1996) reportedsimilar spatial and velocity distribution of the CS (1 − (1 ,
1) lines tracing the turbulent, high-density (10 − cm − ) gas in the region (see also Larionov et al. 1999).Thermal, broad line (FWHM >
20 km s − ) SiO emission inthe ground vibrational state (J = − −
2) was de-tected by Harju et al. (1998). Caswell et al. (2000) detected ther-mal CH OH emission from the 156.6 GHz (see also Slysh et al.1999) and 107.0 GHz transitions, this latter superimposed on amaser component. Methanol maser emission was also reportedat 6.7 GHz (e.g., Menten 1991; Goedhart et al. 2004), 12.2 GHz(e.g., Caswell et al. 1993; Błaszkiewicz & Kus 2004), 44.1 GHz(Slysh et al. 1994), and 95.2 GHz (Val’tts et al. 2000). First at-tempts to measure Zeeman splitting of the 6.7 GHz maser com-ponents suggest a magnetic field strength of a few tens of mG(Vlemmings 2008). The G23.01–0.41 region hosts both strong22.2 GHz H O (e.g., Caswell et al. 1983; Szymczak et al. 2005)and 1.665, 1.667 (main-lines), and 1.612 GHz (satellite-line)OH masers (e.g., Caswell & Haynes 1983; Szymczak & G´erard2004). Through Zeeman splitting of the 1.667 GHz maser com-ponents, Szymczak & G´erard (2004) estimated a magnetic fieldstrength of a few tenths of mG. Interestingly, in the spec-tra of the 1.665, 1.667, and 1.720 GHz OH transitions, ab-sorption features were also noted (Caswell & Haynes 1983;Szymczak & G´erard 2004). A similar absorption pattern was ob-served in the 4.8 GHz H CO transition by Downes et al. (1980),who associated G23.01–0.41 to a nearby di ff use H ii region,identified through the H110 α recombination line emission.On angular scales of a few arcsec, NH (3 ,
3) VLAand CH CN (6 −
5) NMA observations (Codella et al. 1997; The reported values of mass and luminosity have been cor-rected taking into account the accurate distance recently measured byBrunthaler et al. (2009).
Furuya et al. 2008) revealed the presence of an HMC, with anelongated shape and a velocity gradient oriented in the NW–SE direction. From the 3 mm dust continuum emission, theHMC mass was estimated to be of about 70 M ⊙ (Furuya et al.2008). These observations suggest that the HMC may be aflattened structure rotating about the axis of the CO (1 −
0) bipolar outflow detected at larger scales. The VLA posi-tions of the H O, OH (Forster & Caswell 1999), and 4.8 GHzH CO (Araya et al. 2008) masers, as well as the positions ofthe 6.7 GHz (J.L. Caswell unpublished; Caswell et al. 2000)and 12.2 GHz (Brunthaler et al. 2009) CH OH masers mea-sured with the Australia Telescope Compact Array (ATCA) andVery Long Baseline Array (VLBA), respectively, agree well withthe position of the HMC. From the large collection of multi-wavelength data reported here, the G23.01–0.41 region can beconvincingly depicted as an active site of massive star forma-tion.
3. Observations and Calibration
The source was observed with the VLA at X and K bands inboth the C-array (in April 2008) and A-array configurations (inOctober 2008). At 3.6 cm in both configurations and at 1.3 cmin the VLA–C, the continuum mode of the correlator was used,resulting in an e ff ective bandwidth of 172 MHz. At 1.3 cm inthe VLA–A we used mode “4” of the correlator, with a pair of3.125 MHz bandwidths (64 channels) centered on the strongestH O maser line and a pair of 25 MHz bandwidths (8 channels)su ffi ciently o ff set from the maser lines to obtain a measurementof the continuum emission. The two bandwidths centered at thesame frequency (measuring the two circular polarizations) wereaveraged.At X band, 3C 286 (5.2 Jy) and 3C 48 (3.1 Jy) were used asprimary flux calibrators, while 1832–105 (1.4 Jy) was the phasecalibrator. For the K-band observations, the primary flux calibra-tor was 3C 286 (2.5 Jy), the phase calibrator 1832–105 (1.0 Jy),and the bandpass calibrator (for the VLA–A data only) 1733-130(3.7 Jy).The data were calibrated with the NRAO AIPS softwarepackage using standard procedures. Only for the VLA–A dataat 1.3 cm, several cycles of self-calibration were applied to thestrongest maser channel, and the resulting phase and amplitudecorrections were eventually transferred to all the other line chan-nels and to the K-band continuum data. This procedure resultedin a significant (at least a factor 2) improvement of the signal-to-noise ratio (SNR).Natural-weighted maps were made with task IMAGR ofAIPS both for the continuum and the line data. Simultaneousobservation of the line and continuum emission at K band madeit possible to obtain the relative position of the continuum im-age with respect to the H O maser spots with great precision( ∼ . ′′ <
20 mas in each coordinate. Inconclusion, we believe that the absolute astrometrical precisionof the continuum maps presented in this paper is of ∼ . ′′ The VLA is operated by the National Radio AstronomyObservatory (NRAO). The NRAO is a facility of the National ScienceFoundation operated under cooperative agreement by AssociatedUniversities, Inc.. Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 3
We conducted VLBI observations of the H O and CH OHmasers (at several epochs), and of the OH maser (at a singleepoch) toward G23.01–0.41 in the K, C, and L bands, respec-tively. In order to derive the maser absolute position, we usedphase-referencing by fast switching between the maser targetand the calibrator J1825 − ff set from the maser source of 2 . ◦ ± − , respectively (Fomalont et al. 2003).Five fringe finders (J1642 + + + + + ff set calibration. Data werereduced with AIPS following the VLBI spectral line procedures.Paper I describes the general data calibration procedures. O masers
We observed the HMSFR G23.01–0.41 (tracking center: R . A . (J2000) = h m . s
39 and
Dec . (J2000) = − ◦ ′ . ′′ in the 6 − H O transition (rest frequency22.235079 GHz). The observations (program code: BM244)consisted of 4 epochs: April 17, June 29, and September 22,2006, and January 17, 2007. During a run of about 6 h per epoch,we recorded the dual circular polarization through a 16 MHzbandwidth centered on a LSR velocity (V
LSR ) of 77.0 km s − .The data were processed with the VLBA FX correlator inSocorro (New Mexico) using an averaging time of 1 s and 1024spectral channels. The total-power spectrum of the 22.2 GHzmasers toward G23.01–0.41 is shown in Fig. 1 (top panel). Thisprofile was obtained by averaging the total-power spectra of allVLBA antennas, weighting each spectrum with the antenna sys-tem temperature (T sys ).The natural CLEAN beam was an elliptical Gaussian witha FWHM size of about 1 . × . − ◦ (east of north), with little variations from epoch to epoch. Theinterferometer instantaneous field of view was limited to about2 . ′′
7. At each observing epoch, using an on-source integrationtime of about 2.5 h, the e ff ective rms noise level of the channelmaps ( σ ) varied in the range 0.006–0.03 Jy beam − . The spectralresolution was 0.2 km s − . We observed the HMSFR G23.01–0.41 (tracking center: R . A . (J2000) = h m . s
27 and
Dec . (J2000) = − ◦ ′ . ′′ Π / J = / − . The 4 MHzbandwidth was used to increase the SNR of the weak L-bandsignal of the continuum calibrator. The data were processed withthe VLBA FX correlator in two correlation passes using 1024and 512 spectral channels for the 1 MHz and 4 MHz bands, re-spectively. In each correlator pass, the data averaging time was2 s. The T sys -weighted mean of antenna total-power spectra forthe right and left circular polarizations at 1.665 GHz are shownin Fig. 1 (bottom panel). After removing a first order baseline, The VLBA is operated by the NRAO. the 1.665 GHz OH spectra of both circular polarizations presentsimilar characteristics, i.e. several maser emission componentssuperposed on a broad absorption feature (see Sect. 4.3).The natural CLEAN beam was an elliptical Gaussian with aFWHM size of 19 mas ×
10 mas at a P.A. of 3 ◦ . The interfer-ometer instantaneous field of view was limited to about 18 . ′′ ff ectiverms noise level of the channel maps was about 0.02 Jy beam − .The 1 MHz band spectral resolution was 0.2 km s − . The visibil-ity phase of both circular polarizations was calibrated using asphase-reference the brightest maser channel of the right circularpolarization. OH masers
We observed the HMSFR G23.01–0.41 (tracking center: R . A . (J2000) = h m . s
39 and
Dec . (J2000) = − ◦ ′ . ′′ in the 5 − A + CH OH transition (rest frequency 6.668519 GHz). This workis based on 3 epochs (program codes: EM061, EM069), sep-arated by about 1 yr, observed on February 27, 2006, onMarch 17, 2007, and on March 16, 2008. At the first twoepochs, antennas involved in the observations were Cambridge,Jodrell2, E ff elsberg, Hartebeesthoek, Medicina, Noto, Torun,and Westerbork. Since the longest baselines involving theHartebeesthoek antenna (e.g., Ef-Hh baseline about 8042 km)heavily resolve the maser emission and do not produce fringe-fit solutions, the Hartebeesthoek antenna was replaced with theOnsala antenna in the third epoch. During a run of about 6 h perepoch, we recorded the dual circular polarization through twobandwidths of 2 MHz and 16 MHz, both centered on a LSR ve-locity of 77.0 km s − . The 16 MHz bandwidth was useful toincrease the SNR of the weak continuum calibrator. The datawere processed with the MKIV correlator at the Joint Institutefor VLBI in Europe (JIVE - Dwingeloo, The Netherlands) us-ing an averaging time of 1 s and 1024 spectral channels for bothobserving bandwidths. The E ff elsberg total-power spectrum at6.7 GHz toward G23.01–0.41 is shown in Fig. 1 (middle panel).The natural CLEAN beam was an elliptical Gaussian with aFWHM size of about 13 mas × ◦ , slightlyvarying from epoch to epoch. The interferometer instantaneousfield of view was limited to about 9 . ′′
2. At each observing epoch,using an on-source integration time of about 2.2 h, the e ff ec-tive rms noise level of the channel maps varied in the range0.008–0.3 Jy beam − . The 2 MHz band spectral resolution was0.09 km s − .
4. Results
The present section reports our main results for the radio contin-uum and individual maser species in G23.01–0.41. We describein Paper I the criteria used to identify, derive parameters (posi-tion, intensity, flux and size), and measure proper motions of in-dividual masing-clouds. In the following, the term “spot” refersto maser emission on a single velocity channel, whereas the term“feature” indicates a collection of spots emitting at similar posi-tion over contiguous channels (i.e. an individual masing-cloud). The European VLBI Network is a joint facility of European,Chinese and other radio astronomy institutes founded by their nationalresearch councils. A. Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41
Our measurements of the 1.3 and 3.6 cm continuum emissionin both VLA–A and VLA–C configurations are summarized inTable 1. While the most extended VLA configuration partly re-solved the radio continuum at 1.3 cm (missing about half of theflux measured in the compact configuration), the fluxes at 3.6 cmof both configurations are comparable. Taking into account theshape of the VLA–A beam (HPBW of 0 . ′′ × . ′′
09 at a P.A.of − ◦ ), the 1.3 cm continuum emission appears slightly elon-gated in the E–W direction (see Fig. 3). Fig. 2 shows that a spec-tral index ( α ) of 0.9 is derived from our highest-flux measure-ments.Previous observations at 3 and 6 cm were performed us-ing the ATCA and the VLA–BnA configuration, respectively(Forster & Caswell 2000; Araya et al. 2008). Forster & Caswell(2000) reported a flux at 3 cm of 0.37 ± .
12 mJy with an HPBWof 1 . ′′ × . ′′
5, and Araya et al. (2008) derived a 5 σ upper limit of2.4 mJy in the 6 cm band (HPBW of 1 . ′′ × . ′′ − ◦ ). O masers
We imaged the whole range of LSR velocities, from 68 to87 km s − , where Forster & Caswell (1999) detected watermaser emission. Using the VLA–C they identified two regionsof maser activities, each elongated in the north-south direc-tion by about 1 ′′ , and separated by about 40 ′′ . Our field ofview covers the region associated with their brightest spotsand the absolute position we have determined for the referencemaser spot is o ff set by about 1 . ′′ ′′ × . ′′
5. They are mainlygrouped in three clusters to the NE, SW, and towards the cen-ter of the plotted area, which are identified in Fig. 3a with labels“A”, “B”, and “C”, respectively. The individual features proper-ties are presented in Table 2. Their intensities range from 0.07 to202.89 Jy beam − . Only 28 features (33% of the total) persistedover at least 3 epochs, 15 of which lasting 4 epochs. The spreadin LSR velocities ranges from 89.4 km s − , for the most red-shifted feature ( − , for the most blueshiftedone ( . ′′ × AU) from the cluster“C” and both have a threadlike N–S morphology extending upto 117 mas. Water maser emission from cluster “A” (15 features)and “B” (6 features) presents di ff erent properties: the cluster “B”consists of faint (weaker than 0.49 Jy beam − ), short-living (oneepoch) features, whereas the cluster “A” consists of bright (upto 22.21 Jy beam − ) and persistent emission (67% of the fea-tures lasted at least 3 epochs and 8 features persisted 4 epochs).The cluster “C” has an arc-like morphology, of about 200 masin size, and extends on top of the 1.3 cm continuum emission.Water maser emission from this region (61 features) is the mostvariable: only 25% of the observed features persisted for at least3 epochs and only 4 features lasted 4 epochs. This region hasalso undergone a powerful maser flare (feature ± − , for feature ± − , forfeature µ x = − µ y = −
90 km s − (where the X and Y indices refer to themotion components toward the east and the north, respectively)representing the Galactic peculiar motion of the HMSFR as mea-sured by Brunthaler et al. (2009). Mean uncertainties in the mag-nitude of the absolute motions of water maser features are about40%. The relative and absolute proper motions of water masersare plotted in Figs. 5a and 5b, respectively. Absolute velocitiesof maser features belonging to cluster “C” show a regular vari-ation of orientation with position, rotating from north to souththrough the west across the arc-like distribution of features, andit appears that they are diverging from a point to the east of themaser cluster. We imaged the whole range of LSR velocities, from 65 to81 km s − , where 1.665 GHz OH maser emission was detectedby Forster & Caswell (1999). Using the VLA A-B hybrid con-figuration they found OH maser emission scattered over a re-gion of about 2 ′′ × ′′ . The absolute position of our referencemaser spot is o ff set by about 0 . ′′ . ′′ × . ′′
4, half of maser featuresare clustered into a small area, 0 . ′′ × . ′′
12, eastward of the1.3 cm continuum peak (Fig. 3c). We have identified 24 dis-tinct OH maser features and individual features properties arepresented in Table 3. Positions are relative to the isolated andbright feature − , for the most redshifted feature ( − , for the most blueshifted one ( − .The l.o.s. velocities are blueshifted toward the radio continuumand closer to V sys at larger distances. An ordered velocity shiftfrom 73.0 to 62.6 km s − (features with label number 8, 11,15, 12, 17, 20, 16, 22) is observed in an elongated structure ex-tending N–S for 33 mas ( ≈
150 AU) over the central region.Maser intensities range from 0.12 to 2.39 Jy beam − with thebrightest features showing blueshifted l.o.s. velocities. We haveidentified 5 Zeeman pairs satisfying the condition that the circu-lar polarization components coincide positionally within 3 σ andare separated in velocity by more than their typical line width of0.3 km s − , corresponding to magnetic fields less than 0.5 mG(e.g., inset in Fig. 3c; see also Fish & Reid 2006). The l.o.s. mag-netic field strength ranges from -5.8 to + − , from about 50 to 90 km s − ,with two deeps at the LSR velocities of 58.5 and 77.3 km s − .Absorption is not visible on the cross-power spectra of the short-est VLBA baseline (KP–PT ≈
400 km). Assuming the absorp-tion is resolved out, this allows to set a lower limit of about 0 . ′′ OH masers
We imaged the whole range of LSR velocities, from 69 to84 km s − , where 6.7 GHz maser emission was previously de-tected towards G23.01–0.41 by Caswell et al. (1995) using theParkes 64-m telescope. The ATCA position of the 6.7 GHz . Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 5 maser peak reported by Caswell et al. (2000) falls inside the areaof our 6.7 GHz maser detections.Methanol maser emission at 6.7 GHz is distributed within aregion of about 0 . ′′ × . ′′ − , for the most red-shifted feature ( − , for the most blueshiftedone ( ff sets in l.o.s. velocities (with respect toV sys ) of the maser features belonging to the southern and north-ern regions of the radio continuum are + . − . − ,respectively. Maser intensities are on average very bright (30%of the features have peak emission greater than 10 Jy beam − ),but range from 0.10 to 146.03 Jy beam − . 67 features (83% ofthe total) persisted over the 2 yr interval covered by our 3 ob-serving epochs. For each maser feature, Table 4 reports the meanbrightness variability defined as the ratio between the variationof the brightness (of the strongest spot) and the average of theminimum and maximum brightness. The mean brightness vari-ability of methanol maser features was about 45% (compared toonly 20% for the source G16.59–0.05, Paper I), varying froma minimum of 5%, for feature ff ects fromthe slightly varying beam shape among di ff erent epochs as wellas amplitude calibration uncertainties. Positions are relative tothe structurally-stable, compact, and bright feature all the detected features (hereafter “barycen-ter”, identified with label S in Table 4), whose position is indi-cated by a star in Fig. 3b. In deriving the “barycenter” position, toreduce the weight of clustered maser features, uniform weight-ing has been used, weighting feature positions with the recipro-cal of the local feature density. Note that, since only a subset of6.7 GHz maser features had a stable (spatial and spectral) struc-ture over time, the “barycenter” is not a good point to refer theinternal motions. We have also calculated the geometric center(hereafter “center of motion”, identified with label ± − , for feature ± − , for fea-ture − , about 2–3 times the spread of l.o.s. ve-locities. We observe a cluster of maser features connected by abridge of extended weaker emission (110 mas ×
60 mas), emit-ting over a narrow velocity range of a few km s − (Fig. 6b).Since in this region spatial blending of nearby maser centerslimits the precision of positions derived with Gaussian fitting,proper motions from this area were su ffi ciently accurate only forthe brightest, more isolated features (
5. Discussion
The continuum source appears slightly resolved at the VLA–A1.3 cm angular resolution of ∼ . ′′ ii region with ra-dius of about 450 AU ionized by a Lyman continuum of 1–3 × s − , corresponding to a ZAMS B1 star (Schraml & Mezger1969; Panagia 1973). The spectral index α = λ < ∼ = . ( Ω / π ) ˙P, where F is themeasured continuum flux in mJy, ˙P is the outflow momentumrate in M ⊙ yr − km s − , Ω is the jet solid angle in sr, and dis the source distance in kpc. Using the flux of 1.4 mJy mea-sured at 1.3 cm with the VLA–C, and the accurate distance of4.6 kpc to G23.01–0.41 by Brunthaler et al. (2009), one derives:˙P = − ( Ω / π ) − M ⊙ yr − km s − . The momentum rate thusdepends on the estimate of the jet collimation factor ( Ω / π ) − .If the jet structure is not resolved but only a bright knot alongthe jet axis is observed, the jet solid angle can be estimated fromthe angular size of the knot if the position of the YSO power-ing the jet is known. In the next section we discuss the maserinformation with the aim to constrain the YSO position. Current models explain H O maser excitation by collisionalpumping with H molecules within hot ( &
400 K) shocked lay-ers of gas behind both high-velocity ( ≥
50 km s − ) dissocia-tive (J-type; Elitzur et al. 1989) and slow ( ≤
45 km s − ) non-dissociative (C-type; Kaufman & Neufeld 1996) shocks, propa-gating in dense regions (H pre-shock density in the range 10 –10 cm − ). Following these models, the cluster “C” of watermasers witnesses the presence of dense shocked gas near thecontinuum source. The higher variability of the bright watermasers from this cluster, compared with clusters “A” and “B”,could indicate that the cluster “C” pinpoints the most active siteof the region. The arc-like distribution of maser features in thisregion, together with their fast (as high as 55 km s − ) and diverg-ing proper motions (see Fig. 5b), lead us to think that this watermaser emission traces a shock front expanding from a center lo-cated to the east of the 1.3 cm continuum peak. The barycenterof the funnel-like structure of the 6.7 GHz masers (the star inFig. 3) falls also in this area, and we speculate that is the placeof the massive YSO(s), possibly a multiple system, exciting anddriving both maser emissions. In the following discussion, wewill use the 6.7 GHz maser barycenter as the best guess for theYSO(s) position.If water maser emission is excited in a jet from a YSO,knowing the average distance of water masers from the YSOand the average maser velocity, one can estimate the momen-tum rate of the jet assuming that is momentum driven. Using A. Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 the model-predicted H pre-shock density of 10 cm − , the mo-mentum rate in the water maser jet is given by the expression:˙P = . × − V R ( Ω / π ) M ⊙ yr − km s − , where V is the average maser velocity in units of 10 km s − , R isthe average distance of water masers from the YSO in units of100 AU, and Ω is the solid angle of the jet. This expressionhas been calculated by multiplying the momentum rate per unitsurface transferred to the ambient gas (n H m H V ), by a factor Ω R , assuming that the jet is emitted from a source at a distanceR from the masers within a beaming angle Ω .We hypothesize that a wind emitted from the YSO impactsagainst the dense circumstellar gas, and excites and drives themotion of water masers of cluster “C”. Water masers of cluster“C” have an average (sky-projected) distance from the YSO of ≈
200 AU, and their average absolute velocity is 40 km s − .Assuming an isotropic wind ( Ω = π ), and considering that thederived sky-projected distance should be taken as a lower limitto the true distance, we derive a value for the momentum rate˙P ≥ . ⊙ yr − km s − . This value is one order of magnitudehigher than the typical momentum rate of the stellar wind froma ZAMS O star of 10 − M ⊙ yr − km s − (Markova et al. 2005),and might be explained if the massive YSO was still accretingmass and a fraction of mass approaching the YSO was deflectedinto the wind, as predicted by X-wind (Shu et al. 2000) and disk-wind (K¨onigl & Pudritz 2000) models.Assuming that the position of the YSO is close to the geo-metric center of the (methanol and water) maser distributions,the 1.3 cm continuum source is found o ff set from the YSO byabout 60 mas to the SW. Such an o ff set is significantly largerthan the accuracy ( ≈
10 mas) of the absolute position of thecontinuum source, which leads us to think that the best interpre-tation is in terms of a thermal jet rather than an UCH ii region.We describe a case similar to what observed for the well-studiedsource IRAS 20126 + ff set of the peaks of the VLA 3.6 cm continuum emission,tracing shock induced ionization in a thermal jet (Hofner et al.2007). Towards G23.01–0.41, the VLA–A slightly resolves the1.3 cm emission, which indicates that the source size is less thanthe VLA–A beam at 1.3 cm of ≈
110 mas. Then, at the dis-tance of the 1.3 cm continuum peak from the YSO of ≈
60 mas,the jet has to be collimated within a cone of semi-aperture θ ≤ . × / = . Ω ≤ . ≥ × − M ⊙ yr − km s − ,which is consistent with the value of the momentum rate derivedabove for the YSO wind.Figure 3a shows that water masers not belonging to cluster“C” are distributed to the NE (mainly in cluster “A”) and to theSW (mainly in cluster “B”) of the putative location of the YSO.They could trace shocked gas along the path of the bipolar jet,emerging from the YSO and interacting with the surroundingmolecular core. Such a jet could be the engine of the massive CO (1 −
0) bipolar outflow, elongated in the NE–SW direc-tion, observed on arcsec scales (see the sketch in Fig. 8). Thisinterpretation explains the absolute velocities of water masersin cluster “A” oriented towards NE (see Fig. 5b). The same jetshould be responsible for the excitation of the continuum sourceand all the water maser features separated from the YSO by morethan ≈ . ′′
2. To intercept all the water masers found at larger dis-tance from the YSO, the jet semi-aperture has to be larger than θ ≥ .
25 rad, which is consistent with the condition θ ≤ . − and are separated from the YSO by ≈ = . Ω / π ) M ⊙ yr − km s − . If thesame jet is responsible for exciting the continuum source anddriving the water masers of cluster “A”, the jet solid angle ( Ω )can be determined by requiring that the momentum rate in themaser jet equals that from the continuum emission:3 . Ω / π ) = − ( Ω / π ) − From this equation one derives
Ω = . θ = . . ≤ θ ≤ . = . ⊙ yr − km s − .This value for the jet momentum rate agrees with that(0.1 M ⊙ yr − km s − ) of the dominant (SW redshifted) lobe ofthe large-scale CO (1 −
0) outflow reported by Furuya et al.(2008). These evidence support the interpretation that the NE–SW distribution of water masers traces the jet at the base of themassive molecular outflow observed at arcsec scales. Note alsothat the momentum rate in the jet is comparable with the mo-mentum rate (0.1 M ⊙ yr − km s − ) derived for the wide-anglewind driving the water masers of cluster “C”, which presentsthe case that the jet itself is powered by the YSO wind. At adistance of ≈
300 AU from the YSO, i.e. at the position of the1.3 cm source, most of the momentum of the wind would bee ffi ciently collimated within the jet, with a collimation factor( Ω / π ) − ≈
20. For comparison, values of collimation factorsof the order of 10 are typical for well-studied radio jets in low-and intermediate-mass star-forming regions (Anglada 1996). Ifthe high momentum rate of the wind could hint to the YSO be-ing still accreting mass, the e ffi cient collimation of the wind intoa jet could indicate that the accretion onto the YSO is mediatedthrough an accretion disk. In Sect. 5.2, we have used the symmetries in the spatial distribu-tions of the 6.7 GHz methanol and 22 GHz water masers to con-strain the YSO position, and postulated that the YSO is locatedclose to the geometric center of both maser species distributions.Figure 4 shows that 6.7 GHz and 22.2 GHz masers emerge fromnearby but di ff erent positions around the putative location ofthe massive YSO. Figures 5b and 6a show that water masersmove significantly faster than methanol masers, and the averagedirection of motion of the two maser species is also di ff erent.Therefore, the comparison of the spatial and velocity distribu-tion suggests that the two maser species, although associatedwith the same massive YSO, originate from distinct environ-ments, characterized by a di ff erent kinematics and, likely, as pre-dicted by water and methanol maser excitation models, by di ff er-ent physical conditions (e.g., Moscadelli et al. 2007; Goddi et al.2007). CH OH excitation models (Cragg et al. 2005) predictthat Class II methanol maser emission is produced by radia-tive pumping in cool ( ≈
30 K), or moderately warm regions( <
200 K), with H densities in the range 10 –10 cm − . Eventhough comparably high densities are required for both waterand Class II methanol maser action, strong 6.7 GHz methanolmasers can be produced at gas temperature significantly lowerthan that of the shocked layers of gas emitting water masers.Looking at Fig. 6a, one notes that the distribution of 6.7 GHzmaser velocities is not isotropic about the maser barycenter. . Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 7 Taking the maser barycenter as the origin of a system of polar co-ordinates, (sky-projected) maser velocities with a dominant ra-dial component appear to concentrate at a P.A. di ff erent from thatwhere maser velocities are predominantly azimuthal. Figure 7shows the distribution of the angle between the maser propermotion and the corresponding position vectors, as a function ofboth the radial distance and the P.A. of the maser feature. Most ofmaser velocities form an angle with the radius (outward from themaser barycenter) varying in the range from -20 ◦ to 120 ◦ (evalu-ated positive counterclockwise for the observer). While the dis-tribution of the velocity-position angle appears to be rather uni-form with the maser radial distance, all the azimuthal velocities(forming with the radius an angle within 90 ◦ ± ◦ ) concentratein a well defined P.A. range, 20 ◦ ≤ P.A. ≤ ◦ .The observed distribution of 6.7 GHz maser velocities canbe interpreted in terms of a combination of expanding and rotat-ing motions. Let us consider a rotating structure about an axisinclined with respect to the l.o.s., and indicate with x ax and y ax the projections on the plane of the sky of the rotation axis and theline perpendicular to that, respectively (see the sketch in Fig. 8).Points of the rotating structure belonging to the plane contain-ing the rotation axis and the l.o.s. lie on x ax and have rotationvelocities parallel to y ax . Therefore, it is possible to identify theprojection onto the plane of the sky of the rotation axis lookingfor the direction in the sky to which the proper motions are per-pendicular. The upper plot of Fig. 7 suggests that the azimuthal6.7 GHz maser velocities can result from rotation around an axiswhose x ax crosses the maser barycenter and is oriented at P.A.within 45 ◦ ± ◦ . Note that this x ax crosses also the 1.3 cm con-tinuum peak (at a P.A. of 225 ◦ ) and agrees with the orientationof the 22 GHz water maser jet discussed in Sect. 5.2. Then, thereare indications that 6.7 GHz masers can emerge from a structurerotating around the jet / outflow axis.Many of the observed 6.7 GHz maser velocities show a com-ponent directed radially outward from the maser barycenter, sug-gesting that the gas traced by 6.7 GHz masers is also expandingaway from the YSO. Looking at Fig. 6a and the upper plot ofFig. 7, one notes that a group of approximately radial velocities(directed within 10 ◦ from the local radius) concentrates at P.A.between 150 ◦ − ◦ and 330 ◦ − ◦ , that is along a directionapproximately perpendicular to x ax . If the rotation axis is closeto the plane of the sky, the rotation velocities of the spots be-longing to the plane containing the rotation and y ax axes shouldbe almost parallel to the l.o.s.. Then, the sky-projected velocitiesalong the y ax axis are expected to show small azimuthal compo-nents and to be mainly radial if the gas is also flowing away fromthe YSO, in agreement with what we actually observe.The same assumptions that the 6.7 GHz masers trace gasflowing away from a center and rotating about an axis at smallinclination with the plane of the sky, allows also to easily inter-pret the observed distribution of l.o.s. velocities. Material on thenear and far side of the maser structure would move towards andaway from the observer, respectively. Along the sky-projectedrotation axis (i.e. x ax ), the observed blue- and redshifted maservelocities towards the NE and SW, respectively, can be explainedif we admit that maser emission to the NE and SW originatesfrom material on the near (approaching) and far (receding) sideof the structure, respectively. Along the direction perpendicularto the sky-projected rotation axis (i.e. y ax ), the distribution ofl.o.s. velocities would reflect the sense of rotation, with blue-and redshifted maser velocities to the NW and SE, respectively,in agreement with the observations (Fig. 8).If the rotation axis is slightly inclined with respect to theplane of the sky, the small group of maser features along the y ax axis with approximately radial velocities (negligible azimuthalcomponents), should be found close to the sky. Then, their dis-tribution of l.o.s. velocities can be used to constrain the centralmass, in the hypothesis of centrifugal equilibrium. Their aver-age distance from the maser barycenter (Fig. 7, lower plot) is ≈ ≈ − . Using these values, a central mass of ≈
20 M ⊙ is derived, corresponding to an early B – late O typeZAMS star.This discussion suggests that the geometrical properties andthe amplitude of the observed 6.7 GHz maser velocities canbe explained in terms of rotation and expansion about a mas-sive YSO. The idea that the methanol-masing gas can partici-pate in an overall expanding motion has also been proposed byBartkiewicz et al. (2009) on the basis of single-epoch VLBI ob-servations of a sample of 31 6.7 GHz maser sites. These authorshave noted that most maser regions coincide with a 4.5 µ m emis-sion excess (as it is also the case for G23.01–0.41), which iscommonly interpreted as a tracer of expanding, shocked, molec-ular gas. In the G23.01–0.41 region, the derived orientation ofthe rotation axis (at P.A. within 45 ◦ ± ◦ ) is consistent withthe scenario where 6.7 GHz CH OH masers can be tracing theinternal portions (within a radius of about 2000 AU from theYSO) of the rotating core, observed in the NH and CH CNlines on arcsec scale along the SE–NW direction (Codella et al.1997; Furuya et al. 2008). The pattern of CH CN l.o.s. velocitiesobserved by Furuya et al. (2008, Fig. 11), mainly blueshifted tothe NW and redshifted to the SE, agrees with the distribution ofl.o.s. velocities of 6.7 GHz CH OH masers at much smaller an-gular scale. We note that the 12.2 GHz maser emission measuredin the G23.01–0.41 region (Brunthaler et al. 2009) has both po-sition and LSR velocity (V
LSR = . − ) consistent withthe 6.7 GHz masers (Fig. 4). In order to produce bright 6.7 and12.2 GHz masers (T b & K), the Class II methanol maserpumping models by Cragg et al. (2005) prescribe a combinationof warm dust (T d =
175 and 125 K) and cool gas (T k =
30 and50 K), for a gas density of n H = –10 cm − . These valuesagree well with the parameters of the ammonia core (T k =
58 Kand n H = . × cm − ) derived by Codella et al. (1997), andwould marginally support the idea of an association of methanolmasers with the NH core.The measured pattern of 6.7 GHz maser proper motions of-fers a simple explanation that the CH CN LSR velocity gradientin the toroid mapped by Furuya et al. (2008) presents a not so-well defined orientation. Gas close to the equatorial plane wouldindeed undertake a complex motion resulting from a combina-tion of rotation and expansion. Taking all the 6.7 GHz maserfeatures with measured proper motions, the mean absolute value( ≈ − ) of the radial velocity components is basicallythe same as that of the azimuthal velocity components, indicat-ing that the masing gas expands and rotates at similar speed.Assuming that the expansion of the gas traced by the 6.7 GHzmasers is momentum driven, the momentum rate to acceleratethe gas can be estimated with the expression ˙P = M V / R,where M is the total mass of the expanding gas, V the av-erage expansion velocity and R the radius of the expandingsphere (working in spherical symmetry, for simplicity). Takingan upper limit to the gas density of n H = cm − , a velocityV = − , and an upper limit to the observed (sky-projected)distance of 6.7 GHz masers from the YSO of R = = × − M ⊙ yr − km s − is derived. Thisvalue is about two orders of magnitude smaller than the momen-tum rate of the YSO wind supposed to be driving the expansionof water masers of cluster “C”, indicating that a negligible frac- A. Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 tion of the wind, non-collimated into the jet and blowing acrossthe equatorial plane, could be responsible for the expansion ofthe gas nearby the YSO. The YSO wind is only one of the possi-ble causes for the observed 6.7 GHz maser outward motion. Thespatial sampling of the measured maser velocities is not su ffi -ciently complete for a quantitative analysis to e ff ectively con-strain the geometry of the masing gas and determine which por-tions of it is expanding. Figures 3c and 4 show that the spatial distribution of the1.665 GHz OH masers di ff ers from that of 6.7 GHz CH OH and22.2 GHz H O masers. The N–S elongated structure traced bythe OH masers over the 1.3 cm continuum is slightly o ff set tothe east of the water maser cluster “C”, and is found near tothe (putative) YSO location. To explain the strongly blueshiftedl.o.s. velocities (from − −
14 km s − ) of this cluster of OHmasers, a possible interpretation is that they are seen in the fore-ground of the YSO and the radio continuum emission, and thatthey are originating in a layer of gas moving towards the ob-server. The same wind, which we have conjectured to be driv-ing the expansion of water masers in cluster “C”, might be re-sponsible for accelerating the gas hosting the OH maser struc-ture. The OH and CH OH Class II maser excitation models byCragg et al. (2002, Fig. 4) predict strong 1.665 GHz maser ac-tion and inhibition of the 6.7 GHz masers when n H ≤ cm − .This suggests that the 1.665 GHz OH masers towards the ra-dio continuum are tracing lower densities than those requiredfor intense 6.7 GHz CH OH masers. Thus, the properties of theelongated structure observed (only) in the 1.665 GHz OH maseremission might be explained if the OH masers trace a gas layerwith n H ≤ cm − , rapidly expanding away from the YSOtowards the observer. From Zeeman splitting, the l.o.s. magneticfield strength in this OH maser layer is estimated to be about1 mG (see Table 3, features ii regions byFish & Reid (2007). These authors found that OH masers tracethe expansion of ionized gas and also established a correspon-dence between the gas density and the magnetic field, deriving avalue of the order of 1 mG for the l.o.s. magnetic field strengthif n H = cm − .We propose that the OH absorption is spatially unre-lated to the OH maser emission and may instead be con-nected with the large scale dynamics of the region. In theNRAO VLA Sky Survey, the G23.01–0.41 region shows dif-fuse 1.4 GHz radio continuum emission extended on scales ofabout 1 ′ . Toward G23.01–0.41, using the E ff elsberg radiote-lescope Downes et al. (1980) reported both di ff use H110 α re-combination line (4.874 GHz) emission and absorption in the4.8 GHz H CO line at various velocities ranging from 55 to103 km s − . Furthermore, a lower limit of a few arcminutes tothe size of the absorption region is set by the H CO VLA–D ob-servations by Araya et al. (2008), who resolved the absorption.The 1.6 GHz OH absorption feature visible in our VLBA total-power spectra is mainly blueshifted and that may suggest thatthe absorbing molecular gas stays in front of the extended radiocontinuum emission and is expanding towards the observer (e.g.,Szymczak & G´erard 2004).
6. Summary and Conclusions
Using the VLBI technique, we observed the HMSFR G23.01–0.41 in the three most powerful maser transitions: 22.2 GHzH O, 6.7 GHz CH OH, and 1.665 GHz OH. The source G23.01–0.41 was also observed with the VLA, detecting faint ( ≈ mJy)radio continuum emission with 0 . ′′ O, CH OH, and OH maser emissions are distributedwithin ≈ ff erent, althoughcomplementary, spatial distributions and kinematical prop-erties.2. Water masers trace fast (20–50 km s − ) outflows emittedfrom the YSO. An arc-like structure of water masers super-posed on the radio continuum marks a fast shock propagatingthrough dense gas, and is probably driven by a YSO wind.The 1.3 cm continuum source and the two clusters of wa-ter masers, aligned along the NE–SW direction detected atlarger distance from the YSO, are likely tracing the jet driv-ing the massive CO outflow observed at larger scales.3. Methanol masers present a N–S oriented, funnel-like spa-tial distribution with red- and blueshifted features located tothe south and the north, respectively. Observing 3 di ff erentEVN epochs spanning 2 yr, we have measured accurate (rel-ative errors < − . The pattern of 6.7 GHz maser propermotions can be interpreted in term of a composition of ex-pansion and rotation around a YSO of about 20 M ⊙ ) , withthe rotation axis oriented on the sky at similar P.A. to theaxis of the jet / outflow system traced by the water masers.It is then plausible that the CH OH masers trace the internalportions of the toroid, elongated along the SE–NW direction,observed in the CH CN and NH lines on arcsec scale.4. Hydroxyl masers superposed on the radio continuum areprobably seen in the foreground and expand outward fromthe central source tracing a lower density environment thanthat harboring the methanol and water masers.This study demonstrates that multi-epoch VLBI observa-tions of di ff erent maser species provide information useful toexplore the complex phenomena occurring at distances of tensup to thousands of AU around massive YSOs. The best scien-tific return from VLBI maser observations can be expected whensuch observations will be compared to data of new generation(sub)millimeter interferometers (such as ALMA). Clearly, ther-mal line observations of both outflow(s) and core(s) with an an-gular resolution closer to that of VLBI maser data will help tosolve ambiguities in the interpretation of kinematic structures,thanks to a better sampling of the (proto)stellar environment andits physical properties. Acknowledgements.
This work is partially supported by a Grant-in-Aid from theMinistry of Education, Culture, Sports, Science and Technology of Japan (No.20740113).
References
Anglada, G. 1996, in Astronomical Society of the Pacific Conference Series,Vol. 93, Radio Emission from the Stars and the Sun, ed. A. R. Taylor &J. M. Paredes, 3–7 . Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 9
Anglada, G., Estalella, R., Pastor, J., Rodriguez, L. F., & Haschick, A. D. 1996,ApJ, 463, 205Araya, E. D., Hofner, P., Goss, W. M., et al. 2008, ApJS, 178, 330Bartkiewicz, A., Szymczak, M., van Langevelde, H. J., Richards, A. M. S., &Pihlstr¨om, Y. M. 2009, A&A, 502, 155Błaszkiewicz, L. & Kus, A. J. 2004, A&A, 413, 233Brunthaler, A., Reid, M. J., Menten, K. M., et al. 2009, ApJ, 693, 424Caswell, J. L., Batchelor, R. A., Forster, J. R., & Wellington, K. J. 1983,Australian Journal of Physics, 36, 443Caswell, J. L., Gardner, F. F., Norris, R. P., et al. 1993, MNRAS, 260, 425Caswell, J. L. & Haynes, R. F. 1983, Australian Journal of Physics, 36, 417Caswell, J. L., Vaile, R. A., Ellingsen, S. P., Whiteoak, J. B., & Norris, R. P.1995, MNRAS, 272, 96Caswell, J. L., Yi, J., Booth, R. S., & Cragg, D. M. 2000, MNRAS, 313, 599Codella, C., Testi, L., & Cesaroni, R. 1997, A&A, 325, 282Cragg, D. M., Sobolev, A. M., & Godfrey, P. D. 2002, MNRAS, 331, 521Cragg, D. M., Sobolev, A. M., & Godfrey, P. D. 2005, MNRAS, 360, 533Downes, D., Wilson, T. L., Bieging, J., & Wink, J. 1980, A&AS, 40, 379Elitzur, M., Hollenbach, D. J., & McKee, C. F. 1989, ApJ, 346, 983Fish, V. L. & Reid, M. J. 2006, ApJS, 164, 99Fish, V. L. & Reid, M. J. 2007, ApJ, 670, 1159Fomalont, E. B., Petrov, L., MacMillan, D. S., Gordon, D., & Ma, C. 2003, AJ,126, 2562Forster, J. R. & Caswell, J. L. 1999, A&AS, 137, 43Forster, J. R. & Caswell, J. L. 2000, ApJ, 530, 371Furuya, R. S., Cesaroni, R., Takahashi, S., et al. 2008, ApJ, 673, 363Goddi, C., Moscadelli, L., Sanna, A., Cesaroni, R., & Minier, V. 2007, A&A,461, 1027Goedhart, S., Gaylard, M. J., & van der Walt, D. J. 2004, MNRAS, 355, 553Harju, J., Lehtinen, K., Booth, R. S., & Zinchenko, I. 1998, A&AS, 132, 211Hoare, M. G., Kurtz, S. E., Lizano, S., Keto, E., & Hofner, P. 2007, Protostarsand Planets V, 181Hofner, P., Cesaroni, R., Olmi, L., et al. 2007, A&A, 465, 197Kaufman, M. J. & Neufeld, D. A. 1996, ApJ, 456, 250K¨onigl, A. & Pudritz, R. E. 2000, Protostars and Planets IV, 759Larionov, G. M., Val’tts, I. E., Winnberg, A., et al. 1999, A&AS, 139, 257Markova, N., Puls, J., Scuderi, S., & Markov, H. 2005, A&A, 440, 1133Menten, K. M. 1991, ApJ, 380, L75Molinari, S., Pezzuto, S., Cesaroni, R., et al. 2008, A&A, 481, 345Moscadelli, L., Goddi, C., Cesaroni, R., Beltr´an, M. T., & Furuya, R. S. 2007,A&A, 472, 867Panagia, N. 1973, AJ, 78, 929Sanna, A., Moscadelli, L., Cesaroni, R., et al. 2010, ArXiv e-printsSchraml, J. & Mezger, P. G. 1969, ApJ, 156, 269Shu, F. H., Najita, J. R., Shang, H., & Li, Z.-Y. 2000, Protostars and Planets IV,789Slysh, V. I., Kalenskii, S. V., Val’tts, I. E., Golubev, V. V., & Mead, K. 1999,ApJS, 123, 515Slysh, V. I., Kalenskii, S. V., Valtts, I. E., & Otrupcek, R. 1994, MNRAS, 268,464Sridharan, T. K., Beuther, H., Schilke, P., Menten, K. M., & Wyrowski, F. 2002,ApJ, 566, 931Szymczak, M. & G´erard, E. 2004, A&A, 414, 235Szymczak, M., Pillai, T., & Menten, K. M. 2005, A&A, 434, 613Val’tts, I. E., Ellingsen, S. P., Slysh, V. I., et al. 2000, MNRAS, 317, 315Vlemmings, W. H. T. 2008, A&A, 484, 773
Fig. 1.
Total-power spectra of the H O, CH OH and OH masers toward G23.01–0.41.
Upper panel: system-temperature (T sys )weighted average of the 22.2 GHz total-power spectra of the 9 VLBA antennas observing on 2007 January 17 (BR, FD, KP, LA,MK, NL, OV, PT, SC). Inset: zoom of the water maser peak.
Middle panel: E ff elsberg total-power spectrum of the 6.7 GHz methanolmaser emission on 2007 March 17. Inset: zoom of the methanol maser peak. Lower panel: T sys -weighted average of the 1.665 GHztotal-power spectra of the 9 VLBA antennas observing on 2007 April 27 (BR, FD, HN, KP, MK, NL, OV, PT, SC). Continuous(black) and dashed (red) lines are used to distinguish between right (RCP) and left (LCP) circular polarizations, respectively. Thedotted line crossing the spectra represents the systemic velocity (V sys ) inferred from CH CN measurements. . Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 11
Table 1.
G23.01–0.41: Radio continuum emission.
Peak position a Telescope λ HPBW P.A. Image rms R.A.(J2000) Dec.(J2000) F peak F int (cm) ( ′′ × ′′ ) ( ◦ ) (mJy beam − ) (h m s) ( ◦′ ′′ ) (mJy beam − ) (mJy)VLA–A 1.3 0 . × .
095 -13 0.06 18 34 40.284 -09 00 38.31 0.72 0.98VLA–C 1.3 1 . × . . × .
24 -5 0.03 ... ... 0.47 0.59VLA–C 3.6 2 . × . ± . ′′ Fig. 2.
Spectral energy distribution of the radio continuum in the HMSFR G23.01–0.41. Dots and errorbars report the values andthe associated errors (1 σ ) of our measurements at 1.3 and 3.6 cm using both the VLA–A and VLA–C configurations (see Table 1).Taking at each observing frequency the highest-flux measurement, the derived spectral index, indicated by the dotted line, is α = Fig. 3.
Absolute positions and LSR velocities of the three maser species observed in G23.01–0.41: a) H O (triangles), b) CH OH(dots), and c) OH (squares). Di ff erent colors are used to indicate the maser LSR velocities, according to the color scale on theright-hand side of the plot, with green representing the systemic velocity of the HMC. The VLA 1.3 cm continuum emission isplotted with dashed contours. Contour levels range from 30 to 90% of the peak emission (0.72 mJy beam − ) at multiples of 10%.The restoring beam is shown in the lower left corner of the upper panel. The linear scale of the plots is shown in the lower leftcorner of the middle panel. For water masers, letters A, B, and C indicate the clusters defined in Sect. 4.2. For methanol masers, thestar marks the position of the barycenter of the methanol maser distribution, as defined in Sect. 4.4. For hydroxyl masers, full andempty squares denote right and left circularly polarized features, respectively, and numbers close to (some) maser features indicatethe inferred local value of the magnetic field strength (in mG). Inset: example of Zeeman pair for the feature . Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 13
Fig. 4.
Collection of the sub-arcsec observations toward G23.01–0.41.
Left panel: map in the NH (3,3) line from Codella et al.(1997). Contour levels range from 20 to 90% of the peak emission (46 mJy beam − ) at multiples of 10%. The restoring beam isshown in the lower left corner of the panel. The star marks the peak position of the 3 mm continuum emission as determined byFuruya et al. (2008). Right panel: enlargement of the region over which we have detected maser emission. Red triangles, blue dots,and green squares represent respectively H O, CH OH, and OH maser positions from our VLBI measurements. The empty circlemarks the position of the brightest 12.2 GHz CH OH maser feature (at V
LSR = . − ) measured by Brunthaler et al. (2009)with an uncertainty of a few mas. The empty triangle with the errorbars indicates the position (and the associated uncertainty)of the 4.8 GHz H CO maser feature (at V
LSR = . − ) derived by Araya et al. (2008). Dashed contours show the 1.3 cmVLA continuum emission. Contour levels range from 30 to 90% of the peak emission (0.72 mJy beam − ) at multiples of 10%. Therestoring beam is shown in the lower left corner of the panel. Fig. 5. O maser kinematics toward G23.01–0.41. a) positions (triangles) and transverse velocities of the H O maserfeatures relative to the feature ff erent colors are used to indicate the maser LSR velocities, according to the color scale on the right-hand side ofthe panel, with green denoting the systemic velocity of the HMC. b) absolute positions and transverse velocities of the water maserfeatures. The potted field of view is the same as in the upper panel and symbols have the same meaning as in the upper panel. Dashedcontours show the VLA 1.3 cm continuum emission. Contour levels range from 30 to 90% of the peak emission (0.72 mJy beam − )at multiples of 10%. The restoring beam is shown in the lower left corner of the panel. . Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 15 Fig. 6. OH maser kinematics toward G23.01–0.41. a) absolute positions (dots) and transverse velocities of the CH OHmaser features relative to the center of motion (as defined in Sect. 4.4) of the methanol maser distribution (indicated by the cross).Colored cones are used to show both the direction and the uncertainty (cone aperture) of the proper motion of maser features. Theproper motion amplitude scale is given by the black arrow on the bottom right corner of the panel. Di ff erent colors are used toindicate the maser LSR velocities, according to the color scale on the right-hand side of the plot, with green denoting the systemicvelocity of the HMC. Dashed contours show the VLA 1.3 cm continuum emission. Contour levels range from 30 to 90% of the peakemission (0.72 mJy beam − ) at multiples of 10%. The restoring beam is shown in the lower left corner of the panel. b) Zoom of thecrowdest region of 6.7 GHz maser emission. The map is obtained from data of the second EVN epoch (2007 March 17), summingthe maser emission over the velocity range from 73.40 to 74.63 km s − . Plotted contour levels are spaced by factor of 2, with thelowest level corresponding to the 3 σ rms of 3.0 Jy beam − . The restoring beam is given in the lower left corner of the panel. Thenumbers near emission peaks are the maser feature labels given in Table 4. Fig. 7.
Upper and lower panels show the distribution of the angle between the proper motion and the position vectors of 6.7 GHzmasers as a function of the maser P.A. and radial distance, respectively. Approximately azimuthal and radial velocities are indicatedwith empty squares and triangles, respectively (velocity-position angle in the range from 60 ◦ to 120 ◦ and from -10 ◦ to 10 ◦ , respec-tively), whereas filled dots are used for other values of the velocity-position angle. The 6.7 GHz maser barycenter (see Table 4) istaken to be the origin of the (sky-projected) maser positions. The angle between the velocity and the position vectors of a givenmaser feature is evaluated positive counterclockwise and is equal to the di ff erence between the P.A. of the feature velocity andposition vectors. Maser features with P.A. in the range 180 ◦ − ◦ have been folded into the P.A. interval 0 ◦ − ◦ (by subtracting180 ◦ to their true P.A.). In the upper plot, horizontal dotted and dashed lines mark the boundary and the center, respectively, of theinterval over which maser velocities are approximately azimuthal. . Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 17 NSE Wx ax y ax CO, HNCO outflow 3 mm, NH ,CH CN toroid CO, C O clump
CH3OHCH3OHradio free-freeH2O H2O
Fig. 8.
Schematic cartoon of the main components in the molecular cloud associated with G23.01–0.41. The drawing is not to scale.The components represented are (i) the outflow / jet structure traced by CO and H O maser emission (black dots) from pc-scale toa few thousand of AU; (ii) the toroid structure seen on a large scale in CH CN and highly excited NH molecules, and its innerportion traced by CH OH maser emission (blue and red dots according to the l.o.s. velocity); (iii) the radio continuum emissionsuperposed on the arc-like H O maser structure. The star marks the assumed YSO(s) location. . Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 , Online Material p 1
Online Material . Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 , Online Material p 2
Table 2.
Parameters of VLBA 22.2 GHz water maser features.
For each identified feature, the label (given in Col. 1) increases with decreas-ing brightness. Cols. 2 and 3 report the LSR velocity and brightness of the brightest spot (for the epoch specified by the number in brackets).Cols. 4 and 5 give the position o ff set relative to the feature Feature V LSR F peak ∆ x ∆ y V x V y ∆ | V | / | V | (km s − ) (Jy beam − ) (mas) (mas) (km s − ) (km s − )1 72.58 202.89 (4) − . ± . − . ± . − . ± . . ± . . ± .
11 74 . ± .
28 11 . ± . − . ± . . ± .
10 73 . ± .
14 10 . ± . − . ± . . ± .
10 159 . ± .
12 9 . ± . − . ± . − . ± . − . ± . − . ± . − . ± . − . ± . − . ± . − . ± . − . ± . . ± .
10 158 . ± .
13 10 . ± . − . ± . − . ± . − . ± .
12 ... ... ...9 66.47 5.14 (4) − . ± . − . ± .
12 3 . ± . − . ± . − . ± . − . ± . − . ± . . ± . − . ± . − . ± .
12 1 . ± . − . ± . − . ± . − . ± .
14 ... ... ...13 76.58 4.82 (4) 425 . ± .
10 158 . ± .
11 5 . ± . − . ± . − . ± . − . ± . − . ± . . ± . − . ± . − . ± .
13 ... ... ...16 72.58 3.86 (1) − . ± . − . ± .
11 ... ... ...17 77.21 3.06 (4) − . ± . − . ± .
09 ... ... ...18 74.26 2.80 (2) − . ± . − . ± . − . ± . − . ± . − . ± . − . ± .
13 ... ... ...20 75.10 2.37 (3) 32 . ± . − . ± .
09 ... ... ...21 79.74 2.27 (3) 420 . ± .
10 126 . ± .
12 5 . ± . − . ± . − . ± . − . ± .
13 ... ... ...23 78.69 2.10 (1) − . ± . − . ± .
18 ... ... ...24 71.52 1.95 (2) 0 0 0 0 ...25 78.26 1.72 (4) 435 . ± .
10 77 . ± .
13 6 . ± . − . ± . − . ± . − . ± . − . ± . . ± . − . ± . − . ± .
12 ... ... ...28 81.21 1.43 (3) − . ± . − . ± .
09 ... ... ...29 79.95 1.41 (4) − . ± . − . ± . − . ± . − . ± . − . ± . − . ± . − . ± . − . ± . − . ± . − . ± .
11 ... ... ...32 80.37 0.76 (4) − . ± . − . ± .
11 ... ... ...33 79.53 0.75 (3) − . ± . − . ± . − . ± . − . ± . . ± .
10 158 . ± .
11 ... ... ...35 72.16 0.67 (1) − . ± . − . ± .
11 ... ... ...36 78.48 0.59 (2) − . ± . − . ± .
13 ... ... ...37 66.89 0.56 (1) − . ± . − . ± .
11 ... ... ...38 80.58 0.55 (1) 437 . ± .
10 67 . ± .
15 11 . ± . − . ± . − . ± . − . ± .
11 ... ... ...40 79.53 0.48 (2) − . ± . − . ± .
17 ... ... ...41 76.79 0.44 (2) − . ± . − . ± .
09 ... ... ...42 79.32 0.43 (4) − . ± . − . ± .
08 ... ... ...43 67.94 0.42 (1) − . ± . − . ± .
11 ... ... ...44 80.58 0.39 (1) − . ± . − . ± . − . ± . − . ± . − . ± . − . ± . − . ± . − . ± . − . ± . − . ± .
09 ... ... ...47 80.16 0.36 (1) 440 . ± .
10 58 . ± .
16 18 . ± . − . ± . − . ± . − . ± .
09 ... ... ...49 79.32 0.33 (3) − . ± . − . ± . − . ± . − . ± . − . ± . − . ± .
17 ... ... ...51 62.25 0.31 (3) − . ± . − . ± .
10 ... ... ...52 69.42 0.31 (1) − . ± . − . ± .
11 ... ... ...53 79.74 0.31 (1) 284 . ± . − . ± .
12 0 . ± . − . ± . . Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 , Online Material p 3
Table 2. continued.
Feature V LSR F peak ∆ x ∆ y V x V y ∆ | V | / | V | (km s − ) (Jy beam − ) (mas) (mas) (km s − ) (km s − )54 77.21 0.30 (1) − . ± . − . ± .
13 ... ... ...55 80.16 0.28 (1) 437 . ± .
10 67 . ± .
20 ... ... ...56 76.16 0.27 (1) − . ± . − . ± .
12 ... ... ...57 89.43 0.26 (1) − . ± . − . ± .
12 ... ... ...58 71.95 0.23 (1) − . ± . − . ± .
11 ... ... ...59 80.58 0.22 (4) − . ± . − . ± .
09 ... ... ...60 77.42 0.21 (2) − . ± . − . ± .
10 ... ... ...61 70.68 0.21 (2) − . ± . − . ± .
09 ... ... ...62 77.21 0.20 (1) 418 . ± .
10 92 . ± .
12 5 . ± . − . ± . . ± . − . ± .
12 ... ... ...64 59.51 0.19 (3) − . ± . − . ± .
11 ... ... ...65 75.74 0.17 (3) − . ± . − . ± .
11 ... ... ...66 81.63 0.16 (1) − . ± . − . ± .
13 ... ... ...67 79.74 0.16 (3) − . ± . − . ± .
10 ... ... ...68 70.89 0.16 (4) − . ± . − . ± .
10 ... ... ...69 74.68 0.15 (4) 31 . ± . − . ± .
09 ... ... ...70 70.89 0.14 (4) − . ± . − . ± .
09 ... ... ...71 68.57 0.14 (4) − . ± . − . ± .
09 ... ... ...72 67.31 0.13 (1) − . ± . − . ± .
14 ... ... ...73 75.10 0.13 (1) 442 . ± .
10 98 . ± .
14 ... ... ...74 75.74 0.12 (1) − . ± . − . ± .
16 ... ... ...75 66.05 0.11 (3) − . ± . − . ± .
12 ... ... ...76 79.53 0.11 (4) − . ± . − . ± .
10 ... ... ...77 73.21 0.10 (1) − . ± . − . ± .
15 ... ... ...78 81.00 0.09 (4) 412 . ± .
09 171 . ± .
10 ... ... ...79 73.84 0.09 (2) − . ± . − . ± .
15 ... ... ...80 76.16 0.09 (3) − . ± . − . ± .
14 ... ... ...81 77.21 0.09 (1) 419 . ± .
10 168 . ± .
13 ... ... ...82 71.94 0.09 (3) − . ± . − . ± .
11 ... ... ...83 69.42 0.09 (1) − . ± . − . ± . − . ± . . ± . − . ± . − . ± .
16 ... ... ...85 63.73 0.08 (2) − . ± . − . ± .
14 ... ... ...86 85.85 0.07 (4) − . ± . − . ± .
12 ... ... ...Reference Feature: Absolute position & Proper motionFeature R.A.(J2000) Dec.(J2000) V x V y ∆ | V | / | V | (h m s) ( ◦′ ′′ ) (km s − ) (km s − )24 18:34:40.28950 ± ± . ± . . ± . . Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 , Online Material p 4
Table 3.
Parameters of VLBA 1.665 GHz hydroxyl maser features.
For each identified feature, the label (given in Col. 1) increaseswith decreasing brightness. Cols. 2(3) and 4(5) report the LSR velocity and brightness of the brightest spot for the right (left) circularpolarization. Cols. 6 and 7 give the position o ff set relative to the feature Feature V LSR F peak ∆ x ∆ y ∆ V z a B z (km s − ) (Jy beam − ) (mas) (mas) (km s − ) (mG)RCP LCP RCP LCP1 70.88 71.76 2.39 0.40 0 0 − . − .
52 67.89 ... 2.02 ... − . ± .
07 178 . ± .
77 ... ...3 66.84 66.84 1.36 0.84 14 . ± .
62 18 . ± .
77 ... ...4 73.87 73.87 1.01 0.16 − . ± .
78 88 . ± .
47 ... ...5 ... 74.75 ... 0.97 − . ± .
46 203 . ± .
00 ... ...6 ... 70.57 ... 0.89 199 . ± .
45 174 . ± .
13 ... ...7 68.07 ... 0.74 ... − . ± .
67 191 . ± .
29 ... ...8 72.99 72.99 0.67 0.26 57 . ± .
71 119 . ± .
94 ... ...9 ... 75.10 ... 0.60 − . ± .
46 130 . ± .
14 ... ...10 74.92 ... 0.55 ... − . ± . − . ± .
14 ... ...11 70.35 69.82 0.52 0.23 56 . ± .
94 114 . ± . + . + .
012 67.19 ... 0.46 ... 45 . ± .
66 111 . ± .
98 ... ...13 76.33 79.67 0.25 0.43 − . ± .
45 71 . ± . − . − .
814 ... 76.33 ... 0.41 366 . ± . − . ± .
13 ... ...15 68.77 ... 0.38 ... 52 . ± .
18 113 . ± .
27 ... ...16 ... 63.50 ... 0.31 32 . ± .
56 108 . ± .
73 ... ...17 64.90 ... 0.30 ... 39 . ± .
10 111 . ± .
12 ... ...18 70.18 ... 0.25 ... 192 . ± .
33 175 . ± .
53 ... ...19 79.32 77.21 0.25 0.17 − . ± .
90 130 . ± . + . + .
620 64.55 65.08 0.23 0.13 37 . ± .
92 108 . ± . − . − .
821 66.84 ... 0.22 ... − . ± .
91 147 . ± .
29 ... ...22 62.62 62.79 0.16 0.20 26 . ± .
98 106 . ± .
81 ... ...23 69.65 69.30 0.18 0.19 17 . ± .
78 18 . ± .
17 ... ...24 60.68 60.51 0.12 0.14 20 . ± .
70 70 . ± .
47 ... ...Reference Feature: Absolute positionFeature R.A.(J2000) Dec.(J2000)(h m s) ( ◦′ ′′ )1 18:34:40.28515 ± ± − . Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 , Online Material p 5
Table 4.
Parameters of EVN 6.7 GHz methanol maser features.
For each identified feature, the label (given in Col. 1) increases with decreasingbrightness. Cols. 2, 3 and 4 report the LSR velocity and brightness of the brightest spot, and its percent brightness variability. Cols. 5 and 6 give theposition o ff set relative to the feature Feature V LSR F peak Var. a ∆ x ∆ y V x V y ∆ | V | / | V | (km s − ) (Jy beam − ) (mas) (mas) (km s − ) (km s − )1 74.72 146.03 52% 144 . ± .
08 310 . ± .
09 1 . ± . . ± . . ± .
07 233 . ± .
08 ... ... ...3 74.28 103.74 76% 346 . ± .
08 425 . ± .
08 3 . ± . − . ± . . ± .
08 257 . ± .
08 0 . ± . . ± . . ± .
08 241 . ± .
09 ... ... ...6 80.69 53.26 62% − . ± .
08 43 . ± . − . ± . . ± . . ± .
08 239 . ± .
08 ... ... ...8 81.74 37.47 18% 1 . ± .
07 1 . ± . − . ± . − . ± . . ± .
08 304 . ± .
09 4 . ± . . ± . − . ± . − . ± . . ± .
08 277 . ± .
08 3 . ± . − . ± . − . ± .
07 61 . ± . − . ± . − . ± . . ± .
09 278 . ± .
11 ... ... ...14 74.28 22.88 40% 220 . ± .
08 275 . ± .
10 ... ... ...15 74.63 21.54 70% 309 . ± .
12 425 . ± .
16 15 . ± . − . ± . − . ± .
08 47 . ± . − . ± . . ± . . ± .
08 272 . ± .
09 0 . ± . − . ± . . ± .
09 276 . ± .
10 12 . ± . . ± . . ± .
08 281 . ± . − . ± . − . ± . . ± .
09 272 . ± .
13 ... ... ...21 76.74 12.77 19% 190 . ± .
07 229 . ± .
08 10 . ± . − . ± . . ± .
19 418 . ± .
21 14 . ± . − . ± . . ± .
12 267 . ± .
19 ... ... ...24 74.02 10.06 9% 236 . ± .
09 289 . ± .
11 ... ... ...25 80.60 10.01 36% 331 . ± . − . ± .
09 4 . ± . − . ± . . ± .
08 341 . ± . − . ± . . ± . . ± . − . ± .
09 3 . ± . − . ± . − . ± .
08 103 . ± . − . ± . . ± . . ± .
08 223 . ± .
10 3 . ± . − . ± . . ± .
21 279 . ± .
22 ... ... ...31 73.23 6.23 50% 315 . ± .
08 325 . ± .
09 1 . ± . − . ± . . ± .
11 284 . ± .
16 ... ... ...33 80.86 5.15 101% 4 . ± .
11 8 . ± .
15 ... ... ...34 77.26 4.35 31% 196 . ± .
08 231 . ± .
09 ... ... ...35 74.28 4.33 ... 191 . ± .
14 289 . ± .
25 ... ... ...36 74.10 4.30 ... 183 . ± .
11 258 . ± .
16 ... ... ...37 82.62 3.23 15% − . ± .
08 40 . ± . − . ± . . ± . − . ± .
08 97 . ± . − . ± . . ± . . ± . − . ± .
09 0 . ± . − . ± . . ± .
09 209 . ± .
11 ... ... ...41 73.93 2.58 ... 176 . ± .
13 260 . ± .
29 ... ... ...42 72.79 2.27 29% 262 . ± .
10 221 . ± .
15 ... ... ...43 81.48 2.06 77% − . ± .
09 39 . ± .
12 ... ... ...44 78.93 1.94 24% 78 . ± . − . ± . − . ± . . ± . . ± .
11 281 . ± .
19 ... ... ...46 81.04 1.79 11% − . ± . − . ± . − . ± . . ± . − . ± .
15 294 . ± . − . ± . . ± . . ± .
26 315 . ± .
44 ... ... ...49 72.87 1.71 28% 331 . ± .
12 326 . ± .
17 2 . ± . − . ± . . ± . − . ± . − . ± . − . ± . . ± .
09 215 . ± .
11 1 . ± . − . ± . . ± . − . ± .
17 ... ... ...53 78.58 1.08 76% 361 . ± . − . ± .
21 ... ... ... . Sanna et al.: VLBI study of maser kinematics in high-mass SFRs. II. G23.01–0.41 , Online Material p 6
Table 4. continued.
Feature V LSR F peak Var. a ∆ x ∆ y V x V y ∆ | V | / | V | (km s − ) (Jy beam − ) (mas) (mas) (km s − ) (km s − )54 79.90 1.04 58% 148 . ± . − . ± . − . ± . − . ± . . ± .
09 47 . ± . − . ± . − . ± . . ± . − . ± .
13 1 . ± . − . ± . . ± .
13 263 . ± .
18 ... ... ...58 82.44 0.83 18% 211 . ± . − . ± .
33 ... ... ...59 78.49 0.78 32% 365 . ± . − . ± .
16 ... ... ...60 69.80 0.76 45% − . ± .
11 303 . ± . − . ± . . ± . . ± .
24 381 . ± .
26 ... ... ...62 79.81 0.60 57% 110 . ± . − . ± .
24 ... ... ...63 80.16 0.55 40% 109 . ± . − . ± .
26 ... ... ...64 80.16 0.54 28% 369 . ± . − . ± . − . ± . − . ± . . ± .
14 298 . ± .
22 1 . ± . . ± . . ± . − . ± .
28 ... ... ...67 79.72 0.46 42% 160 . ± . − . ± .
27 ... ... ...68 78.23 0.41 18% 83 . ± . − . ± . − . ± . − . ± . . ± . − . ± .
38 ... ... ...70 77.44 0.36 65% − . ± .
14 69 . ± .
22 ... ... ...71 77.70 0.29 85% − . ± .
26 302 . ± . − . ± . . ± . − . ± .
26 313 . ± .
38 ... ... ...73 78.23 0.25 23% 167 . ± . − . ± .
33 ... ... ...74 79.20 0.24 46% − . ± .
19 102 . ± .
32 ... ... ...75 77.88 0.22 45% − . ± .
33 103 . ± .
44 ... ... ...76 78.14 0.19 40% 200 . ± .
35 178 . ± .
46 ... ... ...77 78.14 0.17 54% − . ± .
37 103 . ± .
62 ... ... ...78 77.35 0.16 47% − . ± .
28 303 . ± .
43 ... ... ...79 72.35 0.13 28% 178 . ± .
38 159 . ± .
56 ... ... ...80 78.32 0.12 5% 50 . ± .
32 257 . ± .
50 ... ... ...81 67.87 0.10 ... 178 . ± .
48 49 . ± .
55 ... ... ...0 131 . ± .
07 149 . ± .
08 0 0S 155 143Reference Feature: Absolute positionFeature R.A.(J2000) Dec.(J2000)(h m s) ( ◦′ ′′ )10 18:34:40.27658 ± ± (a) Variability (%) = (F MAX − F MIN of the brightest spot) / ((F MAX + F MIN ) //