We compare the CO J =(1-0) and HI emission in the Large Magellanic Cloud (LMC) in three dimensions, i.e. including a velocity axis in addition to the two spatial axes, with the aim of elucidating the physical connection between giant molecular clouds
(GMCs) and their surrounding HI gas. The CO J =1-0 dataset is from the second NANTEN CO survey and the HI dataset is from the merged Australia Telescope Compact Array (ATCA) and Parkes Telescope surveys. The major findings of our analysis are: 1) GMCs are associated with an envelope of HI emission, 2) in GMCs [average CO intensity] is proportional to [average HI intensity]^[1.1+-0.1] and 3) the HI intensity tends to increase with the star formation activity within GMCs, from Type I to Type III. An analysis of the HI envelopes associated with GMCs shows that their average linewidth is 14 km s-1 and the mean density in the envelope is 10 cm-3. We argue that the HI envelopes are gravitationally bound by GMCs. These findings are consistent with a continual increase in the mass of GMCs via HI accretion at an accretion rate of 0.05 Msun/yr over a time scale of 10 Myr. The growth of GMCs is terminated via dissipative ionization and/or stellar-wind disruption in the final stage of GMC evolution.
We studied star formation activities in the molecular clouds in the Large Magellanic Cloud. We have utilized the second catalog of 272 molecular clouds obtained by NANTEN to compare the cloud distribution with signatures of massive star formation inc
luding stellar clusters, and optical and radio HII regions. We find that the molecular clouds are classified into three types according to the activities of massive star formation; Type I shows no signature of massive star formation, Type II is associated with relatively small HII region(s) and Type III with both HII region(s) and young stellar cluster(s). The radio continuum sources were used to confirm that Type I GMCs do not host optically hidden HII regions. These signatures of massive star formation show a good spatial correlation with the molecular clouds in a sense they are located within ~100 pc of the molecular clouds. Among possible ideas to explain the GMC Types, we favor that the Types indicate an evolutionary sequence; i.e., the youngest phase is Type I, followed by Type II and the last phase is Type III, where the most active star formation takes place leading to cloud dispersal. The number of the three types of GMCs should be proportional to the time scale of each evolutionary stage if a steady state of massive star and cluster formation is a good approximation. By adopting the time scale of the youngest stellar clusters, 10 Myrs, we roughly estimate the timescales of Types I, II and III to be 6 Myrs, 13 Myrs and 7 Myrs, respectively, corresponding to a lifetime of 20-30 Myrs for the GMCs with a mass above the completeness limit, 5 x 10^4 Msun.
The second survey of the molecular clouds in 12CO (J = 1-0) was carried out in the Large Magellanic Cloud by NANTEN. The sensitivity of this survey is twice as high as that of the previous NANTEN survey, leading to a detection of molecular clouds wit
h M_CO > 2 x 10^4 M_sun. We identified 272 molecular clouds, 230 of which are detected at three or more observed positions. We derived the physical properties, such as size, line width, virial mass, of the 164 GMCs which have an extent more than the beam size of NANTEN in both the major and minor axes. The CO luminosity and virial mass of the clouds show a good correlation of M_VIR propto L_CO^{1.1 +- 0.1} with a Spearman rank correlation of 0.8 suggesting that the clouds are in nearly virial equilibrium. Assuming the clouds are in virial equilibrium, we derived an X_CO-factor to be ~ 7 x 10^20 cm^-2 (K km s^-1)^-1. The mass spectrum of the clouds is fitted well by a power law of N_cloud(>M_CO) proportional to M_CO^{-0.75 +- 0.06} above the completeness limit of 5 x 10^4 M_sun. The slope of the mass spectrum becomes steeper if we fit only the massive clouds; e.g., N_cloud (>M_CO) is proportional to M_CO^{-1.2 +- 0.2} for M_CO > 3 x 10^5 M_sun.
