Giant Molecular Clouds in Local Group GalaxiesLeo BlitzUniversity of California, BerkeleyYasuo FukuiNagoya UniversityAkiko KawamuraNagoya UniversityAdam LeroyUniversity of California, BerkeleyNorikazu MizunoNagoya UniversityErik RosolowskyHarvard-Smithsonian Center for AstrophysicsWe present the first comparative study of extragalactic GMCs using complete data setsfor entire galaxies and a uniform set of reduction and analysis techniques. We present resultsbased on CO observations for the LMC, SMC, M33, M31, IC10 and the nucleus of M64, andmake comparisons with archival Milky Way observations. Our sample includes large spiralsand dwarf irregulars with metallicities that vary by an order of magnitude. GMCs in H Irich galaxies are seen to be well-correlated with H I filaments that pervade the galactic disks,suggesting that they form from pre-existing H I structures. Virial estimates of the ratio of COline strength to H2column density, XCO, suggests that a value of 4 ×1020cm−2(K km s−1)−1is a good value to use in most galaxies (except the SMC) if the GMCs are virialized. However,if the clouds are only marginally self-gravitating, as appears to be the case judging from theirappearance, half the virial value may be more appropriate. There is no clear trend of XCOwith metallicity. The clouds within a galaxy are shown to have the about the same H2surfacedensity and differences between galaxies seem to be no more than a factor of ∼ 2. We showthat hydrostatic pressure appears to be the main factor in determining what fraction of atomicgas is turned into molecules. In the high-pressure regions often found in galactic centers, theobserved properties of GMCs appear to be different from those in the found in the Local Group.From the association of tracers of star formation with GMCs in the LMC, we find that about 1/4of the GMCs exhibit no evidence of star formation and we estimate that the lifetime of a typicalGMC in these galaxies is 20–30 Myr.1. INTRODUCTIONAlthough a great deal of progress has been made on thetopic of star and planet formation since the last Protostarsand Planets conference in Santa Barbara, little work hasbeen done to connect what we know about star formationin the Milky Way to star formation in the Universe as awhole. Fundamental limitations include only a weak un-derstanding of how the massive stars form, how clustersand associations form, and the constancy of the IMF. Af-ter all, in external galaxies, we generally observe only theeffects of massive star formation and the formation of starclusters. Furthermore, knowledge of the initial conditionsfor star formation at all masses remains elusive both withinand outside of the Milky Way.Since nearly all stars form in Giant Molecular Clouds(GMCs), one way to make progress is to examine the prop-erties of GMCs in a number of different extragalactic en-vironments to see how they differ. From the similaritiesand differences, it might be possible to make some gen-eral conclusions about how star formation varies through-out the Universe. Although individual, extragalactic GMCshad been observed previously at high enough resolution toat least marginally resolve them (e.g., Vogel et al., 1987;Lada et al., 1988), the first attempts to do this in a system-atic way were by Christine Wilson (Wilson and Scoville,1990; Wilson and Reid, 1991; Wilson and Rudolph, 1993;1TABLE 1LOCAL GROUP GMC DATAGalaxy Telescope Metallicity Spatial Resolution ReferenceLMC NANTEN 0.33 Z40 pc 1SMC NANTEN 0.1 Z48 pc 2IC10 OVRO/BIMA 0.25 Z14 – 20 pc 3M33 BIMA 0.1 – 1.0 Z20 – 30 pc 4M31 BIMA 0.5 Z26 – 36 pc 5References. — (1) Fukui et al. (2006) (2) Mizuno et al. (2006) (3) Leroy etal. (2006) (4) Engargiola et al. (2003) (5) Rosolowsky (2006)Wilson, 1994) using the OVRO and BIMA interferometers.Her efforts were hampered by small survey areas in a fewgalaxies, so general conclusions could only be made by ex-trapolation. Numerous other authors subsequently studiedone or a few extragalactic GMCs, both in the Local Groupand beyond. An exhaustive list of their efforts is beyond thescope of the present article.The situation has changed in the last five years as a re-sult of the construction of the NANTEN telescope in theSouthern Hemisphere and the completionof the 10-elementBIMA Array. The former made it possible to map the Mag-ellanic Clouds completely with high enough spatial resolu-tion and signal-to-noise to identify all of the GMCs withmasses > 3 × 104M; the completion of the BIMA in-terferometer made it possible to identify GMCs in other,more distant galaxies in the Local Group. Because of theirrelatively large fields of view, these two telescopes couldcompletely survey nearby galaxies. Thus, the first completesurvey of GMCs in any galaxy was of the LMC (Fukui etal., 1999; Mizuno et al., 2001b) and not the Milky Way(MW). Although the molecular gas in the MW has beenessentially completely mapped, velocity crowding in manydirections makes it impossible to generate a full catalogof GMCs. Similarly, the first complete CO surveys of theMagellanic Clouds were by Cohen et al. (1988) and Rubioet al. (1991), but the resolution was too poor to determinethe properties of individual molecular clouds.In this paper, we review the recent surveys of CO inLocal Group galaxies that (1) have sufficient resolution tostudy individual molecular clouds and (2) span all or mostof the target galaxy. We compare the results of observationsof GMCs in the four external Local Group galaxies thathave been mapped in their entirety in CO: the Large Mag-ellanic Cloud (LMC, Fukui et al., 2001; Fukui et al., 2006),the Small Magellanic Cloud (SMC, Miznuo et al., 2001a;Mizuno et al., 2006), IC 10 (Leroy et al., 2006), and M33(Engargiola et al., 2003). We have also made observationsin a small strip in M31 (Rosolowsky, 2006), and we com-pare the properties of the GMCs in all of these galaxiesto clouds in the outer MW (from Dame et al., 2001) us-ing a uniform set of analytic techniques. The LMC andSMC observations were made with the single-dish NAN-TEN telescope in Chile, the remaining galaxies were ob-served with the BIMA millimeter-wave interferometer atHat Creek, California (combined with obsevations from theCaltech OVRO millimeter interferometer for IC 10). A tab-ulation of the galaxies we observed, their metallicities andthe resolution used to observe them is given in Table 1.2. THE GALAXIESIn this section, we examine the distribution of CO emis-sion in
View Full Document
Unlocking...