The Green Bank Telescope Galactic H II Region Discovery Survey
aa r X i v : . [ a s t r o - ph . GA ] J un The Green Bank Telescope Galactic H II Region Discovery Survey
T. M. Bania , L. D. Anderson , , Dana S. Balser , & R. T. Rood ABSTRACT
We discovered a large population of previously unknown Galactic H II regionsby using the Green Bank Telescope to detect their hydrogen radio recombinationline emission. Since recombination lines are optically thin at 3 cm wavelength,we can detect H II regions across the entire Galactic disk. Our targets were se-lected based on spatially coincident 24 µ m and 21 cm continuum emission. Forthe Galactic zone − ◦ ≤ ℓ ≤ ◦ and | b | ≤ ◦ we detected 602 discrete re-combination line components from 448 lines of sight, 95% of the sample targets,which more than doubles the number of known H II regions in this part of theMilky Way. We found 25 new first quadrant nebulae with negative LSR veloc-ities, placing them beyond the Solar orbit. Because we can detect all nebulaeinside the Solar orbit that are ionized by O-stars, the Discovery Survey targets,when combined with existing H II region catalogs, give a more accurate census ofGalactic H II regions and their properties. The distribution of H II regions acrossthe Galactic disk shows strong, narrow ( ∼ II regions now givesunambiguous evidence for Galactic structure, including the kinematic signaturesof the radial peaks in the spatial distribution, a concentration of nebulae at theend of the Galactic Bar, and nebulae located on the kinematic locus of the 3 kpcArm. Subject headings:
Galaxy: structure — HII regions — radio lines: ISM — surveys
1. INTRODUCTION H II regions are the formation sites of massive OB stars. Because the main sequencelifetimes of OB stars are .
10 Myr, H II regions are zero age objects compared to the Astronomy Department, 725 Commonwealth Ave., Boston University, Boston MA 02215, USA. Current Address: Laboratoire d’Astrophysique de Marseille (UMR 6110 CNRS & Universit´e deProvence), 38 rue F. Joliot-Curie, 13388 Marseille Cedex 13, France. National Radio Astronomy Observatory, 520 Edgemont Road, Charlottesville VA, 22903-2475, USA. Astronomy Department, University of Virginia, P.O. Box 3818, Charlottesville VA 22903-0818, USA. II regions are astrophysically important objects that reveal details ofthe impact of the star formation process on the ISM. Knowing the physical properties of H II region/photo-dissociation region/molecular cloud complexes provides important constraintson the physics of star formation and the evolution of the ISM.Modern Galactic H II region surveys began with studies of the Palomar optical sur-vey plates (Sharpless 1953, 1959). Radio recombination lines (RRLs) from Galactic H II regions were discovered in 1965 during the commissioning of the NRAO 140 Foot telescope;H¨oglund & Mezger (1965) detected H 109 α emission at 6 cm wavelength from M 17 andOri A. Because the Galactic ISM is optically thin at centimeter wavelengths, RRL surveyswere able to discover large numbers of H II regions distributed throughout the entire Galac-tic disk. Pioneering RRL surveys were done by, e.g., Reifenstein et al. 1970, Wilson etal. 1970, Downes et al. 1980, Caswell & Haynes 1987, and Lockman 1989. These sur-veys gave important insights into Galactic structure and the spatial pattern of massive starformation. Particularly noteworthy was the discovery of a metallicity gradient across theGalactic disk, made apparent by RRL measurements of H II region electron temperatures(Wink, Wilson, & Bieging 1983; Shaver et al. 1983; Quireza et al. 2006). This Galactic-scalemetallicity gradient placed strong constraints on GCE. By the time of the Lockman (1989)survey, however, almost all of the reasonably strong radio continuum sources, as revealedby contour maps often drawn by hand, had been observed; large angular scale surveys fordiscrete H II regions using cm-wave RRLs as tracers ceased. Despite these efforts, the cen-sus of Galactic H II regions was clearly incomplete. The advent of modern high-resolution,Galactic-scale infrared and radio surveys, coupled with the unprecedented spectral sensitiv-ity of the NRAO Green Bank Telescope (GBT), has made the GBT H II Region DiscoverySurvey (HRDS) possible.
2. TARGET SAMPLE
We assembled our target list from the following multi-frequency, large solid angle Galac-tic surveys: the infrared
Spitzer
Galactic Legacy Infrared Mid-Plane Survey Extraordinaire(GLIMPSE: Benjamin et al. 2003) and MIPSGAL (Carey et al. 2009), the NRAO VLAGalactic Plane Survey made in 21 cm H I and continuum (VGPS: Stil et al. 2006), the VLA 3 –MAGPIS at 20 cm continuum (Helfand et al. 2006), and the NRAO VLA Sky Survey (NVSS:Condon et al. 1998). Our targets were selected by finding objects that showed spatially coin-cident 24 µ m Spitzer
MIPSGAL and ∼
20 cm continuum emission, either from the VGPS orthe NVSS. Our method was not automated, but instead relied on visual inspection of radioand IR emission maps. We removed known H II regions, planetary nebulae (PNe), luminousblue variables, and supernova remnants (SNRs) from the target sample using the SIMBADdatabase.This criterion is a strong indication that a target is emitting thermally (Haslam & Osborne1987; Broadbent, Osborne, & Haslam 1989), and therefore is likely an H II region or a PNe.Warm dust absorbs the far ultraviolet radiation from the exciting star(s) and re-emits in theIR, whereas plasma ionized by the same star(s) gives rise to free-free thermal emission atcm-wavelengths. It is certainly true that the radio continuum can result from a mixture offree-free (thermal) and synchrotron (non-thermal) emission. F¨urst, Reich, & Sofue (1987)showed that one can discriminate thermally from non-thermally emitting objects by usingthe ratio of the infrared to radio fluxes. The IR/radio flux ratio for H II regions is typically ∼
100 times larger than that for non-thermally emitting SNRs, so it is easy to to differenti-ate between the two by eye. Our visual inspection of the IR and radio images should haveeliminated all the SNRs from our target sample. Sources showing coincident mid-IR andcm-wave continuum emission almost invariably are thermally emitting: 95% of our sampletargets show hydrogen RRL emission with line to continuum ratios of ∼ − which togethersuggests that our targets are emitting thermally.
3. GREEN BANK TELESCOPE OBSERVATIONS
Figure 1 shows some representative GBT HRDS RRL detection spectra together with
Spitzer
MIPSGAL 24 µ m images of the nebulae with contours of the VGPS 20 cm con-tinuum emission superimposed. (The MIPS detector is saturated in the central region ofG32.928+0.607.) The sensitivity of the GBT and the power of its Autocorrelation Spec-trometer together made the HRDS possible. To achieve high sensitivity, we used techniquespioneered by Balser (2006), who realized that there are 8 H n α RRL transitions, H 86 α to H 93 α , that can be measured simultaneously by the GBT with the ACS at 3 cm wave-length (X-band). (The H 86 α transition, however, is spectrally compromised by confusing,higher order RRL transitions.) Balser (2006) showed that all these transitions can be av-eraged (after they are re-sampled onto the same velocity scale) to significantly improve theRRL signal-to-noise ratio, thus giving an extremely sensitive X-band H n α average nebularspectrum. 4 –This observing technique, coupled with the sensitivity afforded by the GBT’s aperture,gave us unprecedented spectral sensitivity per unit observing time advantage compared withall previous cm-wavelength RRL surveys of Galactic H II regions. The vast majority of ourdetections took only a single OffOn total power observation. After Gaussian smoothing over5 channels to a resolution of 1.86 km s − chan − (to be compared with the ∼
25 km s − RRLtypical line width), the r.m.s. noise for a single OffOn observation was typically ∼ &
120 mjy. Be-low this the fitted power law overestimates the observed distribution which implies thatthe HRDS is complete to a flux limit of &
120 mJy. Using the stellar fluxes given bySternberg, Hoffmann, & Pauldrach (2003), we estimate that optically thin H II regions ion-ized by single O9 stars within the Solar orbit have flux densities &
120 mJy at 9 GHz. TheHRDS thus should be finding all such H II regions and, furthermore, is capable of detectingO3 stars at heliocentric distances of d ⊙ ∼
20 kpc (Rubin 1968; Anderson 2009).
4. DISCUSSION
The GBT HRDS has doubled the number of known H II regions in the Galactic zone − ◦ ≤ ℓ ≤ ◦ and | b | ≤ ◦ . Each discovery gives a position and LSR velocity forthe nebula. We detected 602 discrete Hydrogen RRL components from 448 target lines ofsight, including ∼
65 infrared bubbles found in the
Spitzer
GLIMPSE survey. We found129 multiple velocity component sources (29% of the target sample). Figure 2 shows thelongitude-velocity distribution of the HRDS nebulae, projected onto the Galactic plane.Although the HRDS found 448 H II regions, the number of physically distinct nebulaerepresented by the 602 discrete Hydrogen RRL components is not well known. Thermal radiosources can often be resolved into several apparently physically distinct emission regions, eachhaving a somewhat different position and RRL velocity. Furthermore, the most massive star-forming complexes, W 43 for example, are extended and fragmented into many sub-clumpsof localized star formation, which together can ionize a very large zone. Many of our multiplevelocity component targets may be detecting such low density ionized gas in addition to RRLemission from another, physically distinct nebula. Here we follow the convention establishedby Lockman (1989) and assume that each of our 602 RRLs is produced by a distinct object.One should keep in mind, however, that the concept of a “discrete H II region” may notapply to many of the complex nebulae seen in the inner Galaxy. 5 –Here we assume that all the HRDS sources are H II regions. Anderson et al. (2010,in preparation) show that the level of contamination in the HRDS sample by heretoforeunknown SNRs, PNe, luminous blue variables, etc., is very small. This conclusion is basedon considerations of galactic structure, scale height, RRL line widths, and the RRL line-to-continuum ratios (i.e. nebulae electron temperatures).Here we use the Paladini et al. (2003) catalog as a proxy for the sample of previouslyknown H II regions. The Paladini et al. (2003) catalog of Galactic H II regions is a com-pilation of 24 Galactic H II region single-dish, medium resolution ( ∼ few arcmin) surveyscovering the entire Galactic plane. There is, however, no definitive, complete compilationof all previously detected Galactic H II regions. Our HRDS nebulae are new, previouslyunknown H II regions because they are not listed in the Paladini et al. (2003) compila-tion, the Lockman, Pisano, & Howard (1996) survey, the Ultra-Compact nebulae studiedby Araya et al. (2002), Watson et al. (2003), and Sewilo et al. (2004), nor are they in theSIMBAD database.The LSR velocity of each HRDS H II region maps into a unique Galactocentric ra-dius, R G . Using the Brand (1986) rotation curve, Figure 3 shows the azimuthally averageddistribution of Galactic H II regions as a function of R G . The filled histogram shows theHRDS nebulae; the open histogram is the total H II region sample (1276 nebulae: HRDS& Paladini et al. 2003). The distribution of H II regions across the Galactic disk showsstrong, narrow ( R G . II regions within 4 kpc radius (Burton 1976; Lockman 1981;Bronfman et al. 2000). In the R G = 2 to 4 kpc zone, however, the HRDS has found 75new nebulae, whereas Paladini et al. (2003) list only 56. The majority of these HRDS RRLcomponents appear to be associated with the Galactic Bar at R G ∼ R G determinations that assume circular rotation uncertain inside4 kpc radius (Burton & Liszt 1993).Together, Figs. 2 and 3 show that the level of PNe contamination in the HRDS samplemust be minimal. Because PNe are an old stellar population their Galactic orbits are well-mixed. PNe show, therefore, no structure in their Galactocentric radial distribution andtheir Galactic longitude-velocity distribution is a scatter plot constrained only by velocitiespermitted by Galactic rotation. Any PNe contamination of the HRDS sample must thereforebe very small, otherwise these interlopers would suppress the unambiguous signal of Galacticstructure seen in the HRDS Galactocentric radial and Longitude-Velocity distributions.Figure 4 shows that the longitude-velocity distribution of the new HRDS and previouslyknown H II region sample together now give unambiguous evidence for an ordered pattern 6 –of Galactic structure. Doubling the census makes the contrast of the ℓ – V features striking.The empty zones in Fig. 4 are just as important in this regard as are the features seen. Thesample of 1276 nebulae clearly shows the kinematic signatures of the radial peaks in thespatial distribution (Fig. 3), a concentration of nebulae at the end of the Galactic Bar, at ℓ ∼ ◦ and V LSR ∼ +100 km s − (Benjamin et al. 2005; Churchwell et al. 2009), and nebulaelocated on the kinematic locus of the 3 kpc Arm.The H II region Galactocentric radial distribution peaks at 4.3 kpc and 6.0 kpc havetraditionally been associated with the Scutum-Centaurus and Sagittarius spiral arms. Inthe first Galactic quadrant they imply tangent point longitudes of 30 ◦ and 45 ◦ , respectively,for R = 8.5 kpc. The Brand (1986) rotation curve terminal velocities for these directionsare 107 km s − and 65 km s − , respectively. The straight lines in Fig. 4 are the solid body locidefined by these tangent points and terminal velocities. These loci trace the over-densitiesin the H II region ℓ – V distribution for ℓ ≥ ◦ quite well. These Northern tangent pointlongitudes show evidence for non-circular streaming motions of ∼
10 km s − .Over 50 years after its discovery (van Woerden, Rougoor, & Oort 1957), the preciseastrophysical nature of the Milky Way’s 3 Kpc Arm remains enigmatic. The ellipse drawn inFig. 4 shows the locus of the Cohen & Davies (1976) 3 Kpc Arm kinematic expanding ringmodel ( R G = 4 kpc, V exp = 53 km s − , V rot = 210 km s − ). Discovered by the very first 21 cmH I surveys, the 3 Kpc Arm also contains considerable amounts of molecular gas (Bania1977, 1980, 1986). Although the near side of the Arm, the segment between the Sun and theGalactic Center, is quite prominent, the far side of the Arm was only recently discovered byDame & Thaddeus (2008). Because of this spatial symmetry and its extreme non-circularvelocities, the 3 Kpc Arm provides strong evidence that the Milky Way is a m=2 barredspiral galaxy (Fux 1999).Knowing that on-going star formation is occurring in the 3 Kpc Arm may help constraintheories of its dynamical origin. The 3 Kpc Arm lacks large numbers of H II regions (Lockman1980, 1981), but there are some in it (Bania 1980). Although the HRDS found a few morenebulae ( ∼
10 more in the near side Arm, see Fig. 2), it has not, however, found a substantialnew population of H II regions along the 3 Kpc Arm ℓ – V locus. The far side of the Arm,in the ( ℓ ≤ ◦ , V LSR ≥ − ) quadrant, remains almost devoid of nebulae along the Armlocus. Because the velocities predicted by the expanding ring model will blend with normalGalactic rotation velocities in the ( ℓ ≥ ◦ , V LSR ≥ − ) and ( ℓ ≤ ◦ , V LSR ≤ − )quadrants, the H II region clusterings seen along the Cohen & Davies (1976) ring locus aredifficult to interpret.We were able to determine the distances to G38 . − .
140 and G48 . − .
001 in Fig. 1by using H I emission/absorption experiments to resolve the kinematic distance ambiguity 7 –(As Anderson & Bania 2009, did for the sample of previously known first Galactic quadrantH II regions.) Each nebula is at the far kinematic distance. The majority of our HRDS H II regions are unresolved with our 82 ′′ survey resolution. We are finding that many of thesesmall angular diameter nebulae lie at the far kinematic distance. Earlier RRL surveys missedthese nebulae because their weak continuum made them poor target choices.The HRDS found 25 first quadrant nebulae with negative LSR velocities. In the firstGalactic longitude quadrant, a negative RRL LSR velocity unambiguously places the H II region beyond the Solar orbit, at large distances from the Sun, d ⊙ &
12 kpc, in the outerGalactic disk, R G & II regions known in the 18 ◦ ≤ ℓ ≤ ◦ , | b | ≤ ◦ zone. The newly discovered Fig. 1 sourcesG38 .
651 + 0 .
087 at − − and G32 .
928 + 0 .
607 at − − already match the sizeof the previous census. The ℓ – V distribution of these negative velocity sources shows goodagreement with CO maps made by Dame, Hartmann, & Thaddeus (2001). This region in ℓ , V -space has traditionally been termed the “Outer Arm”.Because of their location in the critical region beyond the Solar orbit at R G ∼ −
12 kpc,these nebulae will provide new GCE constraints. The HRDS and follow-up GBT observationswill allow us to derive the nebular electron temperature, T e , and helium abundances (Y = He/H). Because metals are the main coolants in the photo-ionized gas, both T e and Y aredirectly related to the distribution of heavy elements in the Milky Way. There are relativelyfew H II regions with accurately derived T e values, especially at the critical R G ∼
10 kpcregion. In the first Galactic quadrant, our 25 new HDRS nebulae can increase the T e samplesize by a factor of 10.We detected RRL emission from 65 H II regions that are surrounded by mid-infraredbubbles. The Spitzer
GLIMPSE survey found almost 600 objects in the inner Galaxy thathave a ring-shaped morphology (Churchwell et al. 2006, 2007). These are presumably bub-bles that are viewed in projection. These objects have simple morphologies and “swept-up”neutral material; they may be sites of triggered star formation (see, e.g., Deharveng et al.2009). Churchwell et al. (2006) argue that ∼
75% of GLIMPSE bubbles are caused by B-stars without detectable H II regions. Because of the large number of bubbles that we findenclosing HRDS sources, we speculate that nearly all GLIMPSE bubbles are caused by H II regions. 8 –
5. SUMMARY
The advent of modern high-resolution, Galactic-scale infrared and radio surveys, e.g.
Spitzer
GLIMPSE/MIPSGAL and the VLA VGPS, coupled with the unprecedented spectralsensitivity of the NRAO Green Bank Telescope allowed us to make a major new discoverysurvey of Galactic H II regions. The GBT HRDS has doubled the number of known H II regions in the Galactic zone − ◦ ≤ ℓ ≤ ◦ and | b | ≤ ◦ . The census of H II regions, whenenhanced by the HRDS, now shows in this Galactic zone a longitude-velocity distributionthat gives unambiguous evidence for Galactic structure, including the kinematic signaturesof peaks in the radial spatial distribution of nebulae, a concentration of nebulae at the endof the Galactic Bar, and nebulae located on the kinematic locus of the 3 kpc Arm. Doublingthe Galactic H II region census makes the contrast of the ℓ – V features striking. The emptyzones in Fig. 4 are just as important in this regard as are the concentrations of nebulae in ℓ – V space.We found 25 new nebulae located beyond the Solar orbit, at large distances from theSun, d ⊙ &
12 kpc, in the outer Galactic disk, R G &
10 kpc. Because of their location,these nebulae will be important for future studies of the radial metallicity gradient in theGalaxy. Many of our new nebulae are seen as bubbles in
Spitzer
GLIMPSE images. Wefound 65 such objects and speculate that nearly all the
Spitzer
GLIMPSE mid-IR identifiedbubbles are H II regions.The HRDS nebular distances will be determined by using H I emission/absorption ex-periments to resolve the kinematic distance ambiguity (see Anderson & Bania 2009). Be-cause we can detect all nebulae inside the Solar orbit that are ionized by O-stars, the GBTHRDS sources, when combined with existing H II region catalogs, will provide a more com-plete census of Galactic H II regions with known distances and physical properties, whichis the fundamental database needed for ISM evolution studies of the Molecular Cloud/H II Region/Star Cluster/Supernova Bubble life-cycle. This will enhance our ability to studyGalactic structure using spatial distributions and to constrain Galactic chemical evolutionusing spatial patterns of nebular metallicity.We thank those visionaries who came before us for the support to continue this sortof fundamental survey science. The National Radio Astronomy Observatory is a facilityof the National Science Foundation operated under cooperative agreement by AssociatedUniversities, Inc. LDA was partially supported by the NSF through GSSP awards 08-0030and 09-005 from the NRAO. 9 –
REFERENCES
Anderson, L. D. 2009, PhD Thesis, Boston Univ. (USA)Anderson, L. D., & Bania, T. M. 2009, ApJ, 690, 706Araya, E., Hofner, P., Churchwell, E., & Kurtz, S. 2002, ApJS, 138, 63Bania, T. M. 1977, ApJ, 216, 381Bania, T. M. 1980, ApJ, 242, 95Bania, T. M. 1986, ApJ, 308, 868Balser, D.S. 2006, AJ, 132, 2326Benjamin, R. A., et al. 2003, PASP, 115, 953Benjamin, R. A., et al. 2005, ApJ, 630, L149Brand, J. 1986, PhD Thesis, Leiden Univ. (Netherlands)Broadbent, A., Osborne, J. L., & Haslam, C. G. T. 1989, MNRAS, 237, 381Bronfman, L., Casassus, S., May, J., & Nyman, L. 2000, A&A, 358, 521Burton, W. B. 1976, ARAA, 14, 275Burton, W. B., & Liszt, H. S. 1993, A&A, 274, 765Carey, S.J., et al. 2009, PASP, 121, 76Caswell, J. L., & Haynes, R. F. 1987, A&A, 171, 261Churchwell, E., et al. 2007, ApJ, 670, 428Churchwell, E., et al. 2006, ApJ, 649, 759Churchwell, E., et al. 2009, PASP, 121, 213Cohen, R. J. & Davies, R. D. 1976, MNRAS, 175, 1Condon, J. J., Cotton, W. D., Greisen, E. W., Yin, Q. F., Perley, R. A., Taylor, G. B., &Broderick, J. J. 1998, AJ, 115, 1693Dame, T. M., Hartmann, D., & Thaddeus, P. 2001, ApJ, 547, 792 10 –Dame, T. M. & Thaddeus, P. 2008, ApJ, 683, L143Deharveng, L., Zavagno, A., Schuller, F., Caplan, J., Pomar`es, M., & De Breuck, C. 2009,A&A, 496, 177Downes, D., Wilson, T. L., Bieging, J., & Wink, J. 1980, A&AS, 40, 379Fux, R. 1999, ˚a, 345, 787F¨urst, E., Reich, W., & Sofue, Y. 1987, A&AS, 71, 63Haslam, C. G. T., & Osborne, J. L. 1987, Nature, 327, 211Helfand, D. J., Becker, R. H., White, R. L., Fallon, A., & Tuttle, S. 2006, AJ, 131, 2525H¨oglund, B., & Mezger, P. G. 1965, Science, 150, 339Lockman, F. J. 1980, ApJ, 241, 200Lockman, F. J. 1981, ApJ, 245, 459Lockman, F. J. 1989, ApJS, 71, 469Lockman, F. J, Pisano, D. J., & Howard, G. J., ApJ, 472, 173Paladini R., Burigana C., Davies R. D., Maino D., Bersanelli M., Cappellini B., Platania P.,& Smoot G. 2003, A&A, 397, 213Quireza. C., Rood, R. T., Bania, T. M., & Balser, D. S. 2006, ApJ, 653, 1126Reifenstein, E. C., Wilson, T. L., Burke, B. F., Mezger, P. G., & Altenhoff, W. J. 1970,A&A, 4, 357Rubin, R. H. 1968, ApJ, 154, 391Sewilo, M., Watson, C., Araya, E., Churchwell, E., Hofner, P., & Kurtz, S. 2004, ApJS, 154,553Sharpless, S. 1953, ApJ, 118, 362Sharpless, S. 1959, ApJS, 4, 257Shaver, P. A., McGee, R. X., Newton, L. M., Danks, A. C., & Pottasch, S. R. 1983, MNRAS,204, 53Sternberg, A., Hoffmann, T. L., & Pauldrach, A. W. A. 2003, ApJ, 599, 1333 11 –Stil, J. M., et al. 2006, AJ, 132, 1158van Woerden, H., Rougoor, G. W., & Oort, J. H. 1957, Comp. Rend., 249, 1961Watson, C., Araya, E., Sewilo, M., Churchwell, E., Hofner, P., & Kurtz, S. 2003, ApJ, 587,714Wilson, T. L., Mezger, P. G., Gardner, F. F., & Milne, D. K. 1970, A&A, 6, 364Wink, J. E., Wilson, T. L., & Bieging, J. H. 1983, A&A, 127, 211
This preprint was prepared with the AAS L A TEX macros v5.2.
12 – -200 -100 0Velocity (km s -1 ) -200204060 A n t enna T e m pe r a t u r e ( m K ) H-38.3HeC
G32.928+0.607 d = 19.2 kpc G a l a c t i c La t i t ude -200 -100 0 100Velocity (km s -1 ) -505101520 A n t enna T e m pe r a t u r e ( m K ) H-40.0HeC
G38.651+0.087 d = 17.8 kpc G a l a c t i c La t i t ude -100 0 100Velocity (km s -1 ) -20020406080100120 A n t enna T e m pe r a t u r e ( m K ) H 19.2HeC
G48.551-0.001 d = 9.9 kpc G a l a c t i c La t i t ude -100 0 100Velocity (km s -1 ) -505101520 A n t enna T e m pe r a t u r e ( m K ) H 60.9HeC
G38.738-0.140 d = 9.2 kpc G a l a c t i c La t i t ude Fig. 1.—
Left:
GBT HRDS RRL spectra for four new H II regions. Shown are the averagespectra for seven RRL transitions, H 87 α to H 93 α , smoothed to 1.86 km s − resolution.Kinematic distances to these nebulae, d, were resolved using the RRL velocity and H I mapsto study H I absorption of the H II region’s continuum emission. The brightest H II regionsshow He and C recombination lines. Right:
Spitzer µ m MIPSGAL infrared images forthe same sources, together with contours of VGPS 20 cm continuum emission (1 ′ resolution).Contours are drawn at 80%, 60%, and 40% of the peak emission. All images are 5 ′ squares;each scale bar is 5 pc long. The GBT 82 ′′ (HPBW) beam is shown as a hatched circle. 13 – -200 -100 0 100 200LSR Velocity (km s -1 )-200204060 G a l a c t i c Long i t ude ( deg ) Fig. 2.— Longitude-Velocity distribution of the 602 radio recombination lines found by theGBT H II Region Discovery Survey projected onto the Galactic plane. 14 – N u m be r Fig. 3.— Radial distribution of Galactic H II regions. The Galactocentric radius of eachnebula, R G , is derived from the observed RRL LSR velocity using the Brand (1986) rotationcurve. The filled histogram shows the 602 new nebulae found by the GBT HRDS. Theopen histogram shows the distribution of the 1,276 H II regions in the combined HRDS andPaladini et al. (2003) samples. These histograms are averaged in Galactic azimuth, yet theyshow two significant, narrow ( R G . R G = 4.25 kpc and 6.0 kpc. 15 – -150 -100 -50 0 50 100 150LSR Velocity (km s -1 )-60-40-200204060 G a l a c t i c Long i t ude ( deg ) Fig. 4.— Longitude-Velocity distribution of Galactic H II regions (the combined GBT HRDS& Paladini et al. 2003 samples) with kinematic Galactic structure loci. For clarity we havenot plotted nebulae in the Nuclear Disk at large LSR velocities, | V | ≥