The Synergy between VLBI and Gaia astrometry
Huib van Langevelde, Luis Henry Quiroga-Nuñez, Wouter Vlemmings, Laurent Loinard, Mareki Honma, Akiharu Nakagawa, Katharina Immer, Ross Burns, Ylva Pihlström, Lorant Sjouwerman, R. Michael Rich, Iniyan Natarajan, Roger Deane
TThe Synergy between VLBI and Gaia astrometry
Huib van Langevelde ∗ ab , Luis Henry Quiroga − Nuñez ba Wouter Vlemmings c , LaurentLoinard d , Mareki Honma e , Akiharu Nakagawa f , Katharina Immer a , Ross Burns agh ,Ylva Pihlström i , Lorant Sjouwerman j , R. Michael Rich k , Iniyan Natarajan l , RogerDeane lm a JIVE, Oude Hoogeveensedijk 4, 7991 PD Dwingeloo, the Netherlands b Sterrewacht Leiden, Leiden University, Postbus 2300, 9513 RA Leiden, the Netherlands c Dept. of SEE, Chalmers Univ., Onsala Space Observatory, SE-439 92 Onsala, Sweden d Instituto de Radioastronomía y Astrofísica, Universidad Nacional Autónoma de México, 58089Morelia, Michoacán, México e Mizusawa VLBI Obs., NAOJ, 2-12 Hoshigaoka-cho, Mizusawa, Oshu, Iwate 023-0861, Japan f Graduate School of Science and Engineering, Kagoshima University, 1-21-35 Korimoto,Kagoshima-shi, Kagoshima 890-0065, Japan g Mizusawa VLBI Observatory, National Astronomical Observatory of Japan, 2-21-1 Osawa,Mitaka, Tokyo 181-8588, Japan h Korea Astronomy and Space Science Institute 776, Daedeokdae-ro, Yuseong-gu, Daejeon,34055, Republic of Korea i Dept. of Phys. & Astronomy, Univ. of New Mexico, MSC07 4220, Albuquerque, NM 87131 USA j NRAO, P.O. Box 0, Lopezville Rd 1003, Socorro, NM 87801 USA k Dept. of Phys. & Astronomy, Univ. of California, Los Angeles, CA 90095-1547 USA l CfRAT&T, Dept. of Phys. & Elect., Rhodes University,- Grahamstown 6140, South Africa m Department of Physics, University of Pretoria, Hatfield, Pretoria, 0028, South AfricaE-mail: [email protected]
With the publication of Gaia DR2, 1.3 billion stars now have public parallax and proper motionmeasurements. In this contribution, we compare the results for sources that have both opticaland radio measurements, focusing on circumstellar masers. For these large, variable and brightAGB stars, the VLBI astrometry results can be more robust. Moreover, there are a number ofapplications where VLBI astrometry provides unique data for studying stellar populations andGalactic structure. The BeSSel project not only provides parallax and proper motions at muchlarger distances than Gaia can reach, but it also uniquely samples the spiral arms of the Galaxy.The evolved stars in the BAaDE sample can potentially constrain the dynamics and stellar contentof the inner bulge and bar of the Milky Way, not reachable in the optical. ∗ Speaker. c (cid:13) Copyright owned by the author(s) under the terms of the Creative CommonsAttribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). http://pos.sissa.it/ a r X i v : . [ a s t r o - ph . S R ] J a n he Synergy between VLBI and Gaia astrometry Huib van Langevelde
1. Introduction
Accurate, milli-arcsecond astrometry at radio wavelengths is key for addressing a number ofastrophysical problems. Recent contributions with the EVN range from lining up a radio detectionof FRBs with a distant dwarf galaxies[1] and gravitational wave NS-NS mergers (e.g. [2]), to thedynamics of High Mass Star Formation (HMSF) regions (e.g. [3]). At the same time, by providing20 −
50 mirco-arcsecond accurate positions for over a billion stars, the Gaia mission is revolution-ising astrometry in the optical. It is providing parallaxes and proper motions for extremely largesamples of stars, complemented with photometry and variability data. Not only has this resultedin new calibrations of the properties of specific stellar populations, but also in novel insights intothe distribution of these stellar populations across the Galaxy (e.g. [4, 5]). These results contributein a fundamental way to our understanding of how the Milky Way was built up over cosmologicalages. Disentangling its merger history is an exciting way to observe the structure formation in theUniverse. But even with this (DR2) deluge of Gaia results, doing stellar astrometry with VLBIremains of fundamental interest and has important synergy with these recent optical astrometrystudies. Although doing astrometry with VLBI is very elaborate compared to querying the Gaiadatabases, we note that the best VLBI measurements are still more accurate than those obtained byGaia.Admittedly, VLBI can only be done on sources that harbour bright, non-thermal radio emis-sion. This implies that additional assumptions must be made about the origin of the radio syn-chrotron or molecular maser emission that is used as a beacon, where Gaia usually (but not always!)
Figure 1:
A direct comparison of parallaxes obtained in the radio with VLBI techniques and in the optical aspublished in the Gaia DR2. Continuum sources are pre-main sequence sources from [6, 8, 7], VERA resultsfor water and SiO results from [9, 10, 11, 12, 13, 14, 15], VLBA OH masers from [16, 17], and VLBA watermasers from [18, 19], pulsar results from references in [20] he Synergy between VLBI and Gaia astrometry Huib van Langevelde
Figure 2:
The residuals for Gaia versus VLBI parallaxes, allowing for a Gaia parallax zero point of − µ as.The distribution has been normalised by the quadratically summed errors, including the Gaia estimate of"excess noise". In blue are AGB stars, grey radio continuum from pre-main sequence stars, black are binarypulsars (see Fig. 1 for the data references). directly measures the position of the mass centre of a star. Nevertheless, the most important reasonfor doing astrometry with VLBI is that it is unaffected by interstellar (or circumstellar) extinction.It can therefore probe the inner Galaxy, its bar and bulge, at Galactic latitudes where Gaia’s rangeis restricted to ≤
2. VLBI-Gaia comparison
For a number of objects the Gaia DR2 allows a direct comparison with VLBI parallax andproper motion results [21]. This includes a number of active stars with non-thermal radio contin-uum, for example active pre-main sequence stars (Fig. 1). Also, some pulsars have optical com-panions. The analysis presented here, however, focuses on evolved stars with circumstellar masers.Because the masers occur —depending on molecular species— at considerable distance from theoptically visible star, the proper motions will generally not be exactly the same, except when thisis a stationary offset. However, maser components with a linear motion with respect to the star willstill show the same parallax. In Fig. 1, we show a plot of VLBI parallax versus Gaia parallax fora number of evolved stars combined with a number of pre-main sequence stars and binary pulsarsfor comparison. In general it is reassuring to see that there is a correspondence between VLBI andGaia results (see also [22]. But when trying to understand the statistical properties, we have foundthat a Gaussian error distribution can represent the distribution quite well (Fig. 2), provided weallow for an offset of the Gaia parallax and take the excess noise of the Gaia results into account[23]. The reason for the large residuals in the Gaia parallaxes is understood to originate from thenature of the AGB stars; their surface brightness distribution and colour is variable, resulting in a2 he Synergy between VLBI and Gaia astrometry
Huib van Langevelde C a r i na l = 0 o l = 45 o l = 90 o l = 135 o l = 180 o l = 225 o l = 270 o l = 315 o kpc kpc GC O u t e r P e r s e u s L o c a l C r u x Figure 3:
An analysis of Outer arm star forming regions may suggest that this segment is closer thanpreviously thought [30] shift of centroid that adds significantly to the positions that Gaia measures [24]. Clearly, for de-termining the distances and deriving quantitative properties of Mira variables and other AGB stars,VLBI astrometry still provides very valuable data.
3. The distribution of high mass star formation
The precise location and motion of high HMSF regions throughout the Galaxy can be mea-sured with VLBI observations of methanol (at 6.7 and 12 GHz ) and water masers at 22 GHz. TheBeSSeL (Bar and Spiral Structure Legacy) survey has produced a census of the size and rotationof the Galaxy, including the location of its spiral structure [25]. The effort continues by measuringmore sources with the new 6.7 GHz capability on the VLBA (e.g. [26]). Moreover, interestingresults can be expected with the new capabilities in Australia to monitor Southern targets [27]. Thedevelopment of the SKA will eventually allow one to survey much weaker sources [28, 29].The maser sites provide a unique way to outline the spiral arms of the Galaxy, as they are madeof GMCs in which massive stars form. An example of a recent result is presented in Fig. 3, wherethe masers are used to constrain the Outer spiral arm [30]. For isolated star forming regions, it maybe possible to find associated members in the Gaia database, which could provide additional con-straint on distances, motions and stellar properties. Although it is expected that optically identifiedOB associations also map out the spiral arms, Gaia will probably only be able to do that for localstructures [31].
4. The distribution of evolved stars
Maybe the most intriguing dynamical structure of the Galaxy is the bar, for which the evidencefrom IR photometry is convincing (e.g. [32]). This structure is hardly accessible for Gaia as mostof the low latitude range is blocked by interstellar extinction (Fig. 4). However, the BAaDE (BulgeAsymmetries and Dynamic Evolution) project provides a clever way to define a sample of up to ≈ he Synergy between VLBI and Gaia astrometry Huib van Langevelde G a l a c t i c L a t i t u d e ( d e g r ee s ) BAaDE targets without Gaia counterparts
Figure 4:
The location of IR-selected BAaDE targets, evolved stars that likely have SiO masers, plottedover the distribution of stars in the inner Galaxy as found the Gaia database. Clearly many targets are foundin the obscured, dark regions [37]. their circumstellar shells [33]. Indeed, a preliminary analysis has already demonstrated that adistinct population seems to make up the Mira stars associated with the bar [34].If it can be demonstrated that VLBI monitoring can provide proper motions (and maybe par-allaxes) this would provide a unique way to determine the orbits of stars that make up the Galacticbar and bulge. VLBI astrometry of SiO masers, however, is hard because of the short coherencetimes for phase referencing at 43 or 86 GHz. This is particularly true for the inner Galaxy, if onetries to observe from the Northern hemisphere using the currently available networks at 7mm. Inaddition, the calibrators at high frequency are sparse and weak. Finally, the special survey modethat is used for the BAaDE survey results a-priori maser positions that are poor by VLBI standards[35]. Therefore, carrying out astrometry for the BAaDE targets may in fact require new VLBItechniques [36].Although optical studies cannot deliver the astrometry for stars in the bar, Gaia results areextremely important to characterise the Mira population that is present in the radio sample. We arecarrying out a study of the Mira and related variables that are in the BAaDE catalogue and haveGaia counterparts. Combining with IR surveys we can derive intrinsic properties, like luminosity,IR colours and eventually variability period. These properties can be used to estimate progenitormass and age. As stars with a wide range of mass go through the AGB phase, this will allow us todistinguish between relatively young stars and much older populations [37].
5. Conclusions
We argue that VLBI astrometry will continue to deliver unique astrometry of evolved stars andobscured high mass star forming sites. This will be important for providing direct distances to themost obscured objects, which are the most extreme in terms of stellar evolution. Moreover, VLBIis not hampered by interstellar extinction and will sample the hidden parts of the Galaxy, whichappear to be key in understanding its assembly.
References [1] B. Marcote, Z. Paragi, J.W.T. Hessels, et al. 2017, ApJ 834 L8[2] G. Ghirlanda, O.S. Salafia, Z. Paragi, et al. 2018, arXiv:1808.00469 he Synergy between VLBI and Gaia astrometry Huib van Langevelde[3] C. Goddi, G. Surcis, L. Moscadelli, et al. 2017, A&A 597 43[4] J. Bland–Hawthorn, S. Sharma, T. Tepper–Garcia, et al. 2018, arXiv:1809.02658[5] A. Helmi, C. Babusiaux, H.H. Koppelman, et al., 2018, Nature 563 85[6] G.N. Ortiz-León, L. Loinard, M.A. Kounkel, et al. 2017, ApJ 834 141[7] G.N. Ortiz-León, S.A. Dzib, M.A. Kounkel, et al., 2017ApJ 834 143[8] M.A. Kounkel, L. Hartmann, L. Loinard, et al., 2017, ApJ 834 142[9] T, Kamezaki, A. Nakagawa, T. Omodaka, et al., 2012, PASJ 64 7[10] T. Kamezaki, A. Nakagawa, T. Omodaka, et al., 2016, PASJ 68 75[11] A. Nakagawa, M. Tsushima, K. Ando, et al., 2008, PASJ 60 1013[12] A. Nakagawa, T. Omodaka, T. Handa, et al., 2014, PASJ 66 101[13] A. Nakagawa, T. Kurayama, M. Matsui, et al., 2016 PASJ 68 78[14] D. Nyu, A. Nakagawa, M. Matsui, et al., 2011, PASJ 63 63[15] C. Min, N. Matsumoto, M.K. Kim, et al., 2014, PASJ 66 38[16] W.H.T. Vlemmings, H.J. van Langevelde, P.J. Diamond, et al., 2003, A&A 407 213[17] W.H.T. Vlemmings, H.J. van Langevelde, 2007, A&A 553 15[18] B. Zhang, X. Zheng, M.J. Reid, et al., 2017 ApJ 849 99[19] T. Kurayama, T. Sasao, H. Kobayashi, 2005, ApJ 627 L49[20] R.J. Jennings, D.L. Kaplan, S. Chatterjee, et al., 2018 ApJ 864 26[21] A.G.A. Brown, A. Vallenari, T. Prusti, et al. 2018, A&A 616 A1[22] G.N. Ortiz-León, L. Loinard, S.A. Dzib, et al. 2018, ApJ 869 L33[23] L. Lindegren, J. Hernandez, A. Bombrun, et al., 2018. A&A 616 A2[24] A. Chiavassa, B. Freytag, M. Schultheis, 2018 A& A 617 L1[25] M.J. Reid, K.M. Menten, A. Brunthaler, et al., 2014. ApJ 783 130[26] K. Immer, 2018,
PoS(EVN2018)055 [27] V. Krishnan, S.P. Ellingsen, M.J. Reid, et al., 2017 MNRAS 465 1095[28] M.J. Rioja, 2018, (EVN2018)054 [29] L.H. Quiroga-Nuñez, H.J. van Langevelde, M.J. Reid, J.A. Green, 2017 A&A 604 A72[30] L.H. Quiroga-Nuñez, K. Immer, H.J. van Langevelde, et al., 2018. Submitt. to A&A,[31] Y. Xu, S.B. Bian, M.J. Reid, J.J. Li, B. Zhang, et al., 2018 A&A 616 L15[32] C. Babusiaux G. Gilmore, 2005. MNRAS 358 1309[33] L.O. Sjouwerman, S.M. Capen, M.J. Claussen, 2009, ApJ 705 1554[34] A.C. Trapp, R.M. Rich, M.R. Morris, et al., 2018 ApJ 861 75[35] Y.M. Pihlström, L.O. Sjouwerman, M.J. Claussen, et al., 2018, ApJ 868 72[36] I. Natarajan, 2018, (EVN2018)078 [37] L.H. Quiroga-Nuñez, H.J. van Langevelde, L.O. Sjouwerman, et al., 2017, Proc. IAU S336 184[37] L.H. Quiroga-Nuñez, H.J. van Langevelde, L.O. Sjouwerman, et al., 2017, Proc. IAU S336 184