arXiv:1108.4809v1 [astro-ph.GA] 24 Aug 2011
Mon. Not. R. Astron. Soc. 000, 1–27 (2009) Printed 25 August 2011(MN LATEX style file v2.2)
Multi-line spectral imaging of dense cores in the Lupus
M. Benedettini1⋆, S. Pezzuto1, M. G. Burton2, S. Viti3, S. Molinari1, P. Caselli4, L. Tes
1INAF – Istituto di Fisica dello Spazio Interplanetario, Area di Ricerca di Tor Vergata, via Fosso del Cavaliere 100, 00133 Roma, Italy
2School of Physics, University of New South Wales, Sydney NSW 2052, Australia
3Department of Physics and Astronomy, University College London, Grower Street, London WC1E 6BT, UK
4School of Physics and Astronomy, University of Leeds, LS2 9JT Leeds, UK
5INAF – Osservatorio Astrofisico di Arcetri, Largo E. Fermi 5, 50125 Firenze, Italy
6ESO, Karl Schwarschild Strasse 2, 85748 Garching bei Mnchen, Germany
Accepted 2011 August 24. Received 2011 August 23; in original form 2011 February 17
The molecular clouds Lupus 1, 3 and 4 were mapped with the Mopra
telescope at 3 and 12 mm. Emission lines from high density molecular tracers
were detected, i.e. NH3(1,1), NH3(2,2), N2H+(1-0), HC3N (3-2), HC3N (10-
9), CS (2-1), CH3OH (20-10)A+and CH3OH (2−1-1−1)E. Velocity gradients
of more than 1 km s−1are present in Lupus 1 and 3 and multiple gas com-
ponents are present in these clouds along some lines of sight. Lupus 1 is the
cloud richest in high density cores, 8 cores were detected in it, 5 cores were
detected in Lupus 3 and only 2 in Lupus 4. The intensity of the three species
HC3N, NH3and N2H+changes significantly in the various cores: cores that
are brighter in HC3N are fainter or undetected in NH3and N2H+and vice
versa. We found that the column density ratios HC3N/N2H+and HC3N/NH3
change by one order of magnitude between the cores, indicating that also the
chemical abundance of these species is different. The time dependent chem-
ical code that we used to model our cores shows that the HC3N/N2H+and
HC3N/NH3ratios decrease with time therefore the observed column density
of these species can be used as an indicator of the chemical evolution of dense
cores. On this base we classified 5 out of 8 cores in Lupus 1 and 1 out of 5
cores in Lupus 3 as very young protostars or prestellar cores. Comparing the
millimetre cores population with the population of the more evolved young
M. Benedettini et al.
stellar objects identified in the Spitzer surveys, we conclude that in Lupus 3
the bulk of the star formation activity has already passed and only a moderate
number of stars are still forming. On the contrary, in Lupus 1 star formation
is on-going and several dense cores are still in the pre–/proto–stellar phase.
Lupus 4 is at an intermediate stage, with a smaller number of individual
Key words: ISM: molecules – ISM: abundances – radio lines: ISM.
One of the critical open question in star formation is the accurate determination of the
stellar Initial Mass Function (IMF), especially in the low-mass regime, in order to under-
stand its origin and particularly how it is related to the mass distribution of the dense
cores where stars form, i.e. the Core Mass Function (CMF). This fundamental question is
investigated by the means of surveying dense condensations in molecular clouds. One of the
classical tool for detecting dense cores in star forming regions is the search for dust conden-
sations using continuum measurement in the millimetre range (e.g. Testi & Sargent 1998;
Johnstone et al. 2000; Motte et al. 2001). Alternatively, one can use spectroscopic surveys
of dense gas molecular tracers. Ammonia is one of the best molecules for studying the cool,
dense molecular cores where stars form (e.g. Myers & Benson 1983; Benson & Myers 1989).
High-density condensations are also mapped in other molecular tracers such as N2H+and
CS. In particular, N2H+is known to be a good tracer of the dense centre of the cores, while
CS is depleted from the gas phase in the very centre of prestellar cores and preferentially
samples the core edge (Caselli et al. 2002b; Tafalla et al. 2002). CS is also a good tracer of
extended high density gas and is useful to probe the kinematics of the gas (Testi et al. 2000;
Olmi & Testi 2002).
The poorly studied Lupus molecular cloud is an interesting target for investigating the
low mass star formation process because its star formation regime, in terms of star forma-
tion rate and stellar clustering, represents an intermediate case between the heavily clustered
sites such as Serpens and Ophiuchus and the more isolated and quiescent sites such as Tau-
rus. Because of its location in the Southern hemisphere (declination from -33◦to -43◦), this
Multiline spectral imaging of dense cores in Lupus
extended molecular cloud has been less investigated with respect to more famous Northern
The distance of the Lupus star forming region is still subject of debate even if it is clear
that it is one of the nearest star forming regions.The most recent distance measurement
is from Lombardi, Lada & Alves (2008) that estimated a distance of (155±8) pc. Comer´ on
(2008) reviewed all the works about the distance determination for the Lupus complex
concluding that it has a depth of the same order as its angular extent on the plane of the
sky, with varying distances of the different individual structures in the 140 to 200 pc range.
He concluded that a a distance of 150 pc is adequate for Lupus 1 and 4 while a value of 200
pc is more appropriate for Lupus 3.
Up to now only a few surveys at poor spatial resolution of a few arcminutes have been car-
ried out in the region in order to study the dense molecular gas distribution (e.g. Hara et al.
1999; Vilas-Boas, Myers & Fuller 2000; Tachihara et al. 2001; Tothill et al. 2009). Maps in
the (J=1-0) transition of12CO and its isotopologues13CO and C18O have shown that the
extended Lupus complex is actually split into nine subgroups. Evidences of on-going star
formation have been found in three subgroups, namely Lupus 1, 3 and 4. Lupus 1, with a
mass of ∼1200 M⊙, is the most massive subgroup. About ten C18O cores have been identi-
fied in Lupus 1 with column densities N(C18O)=(5–10)×1014cm−2(Hara et al. 1999) that
indicate potential sites of star formation. Lupus 3 has a mass of about 300 M⊙ and it hosts
a rich cluster of T-Tauri stars. Tachihara et al. (2007) have mapped the cloud in H13CO+,
showing that no more star formation is expected at the west edge where the T association
is located, whereas there are potential sites of star formation at the eastern edge where
H13CO+emission has been detected. Lupus 4 is the third cloud of the complex that shows
evidence of star formation activity, hosting nine C18O dense cores with column densities
N(C18O)=(4–10)×1014cm−2and three H13CO+cores (Hara et al. 1999). The Lupus 1, 3
and 4 clouds have been mapped with IRAC and MIPS on board Spitzer as part of the “From
molecular clouds to planet-forming disks” (c2d) Legacy Program (Merin et al. (2008) for the
IRAC data and Chapman et al. (2007) for the MIPS data). These infrared surveys allowed
the identification of the population of Young Stellar Objects (YSOs) in the clouds. Adding
also Pre Main Sequence (PMS) objects previously known from others studies, Merin et al.
(2008) found that the total number of PMS is 17, 124 and 18 in Lupus 1, 3 and 4, respec-
tively and that the Star Formation Rate (SFR) is 4.3, 31.0 and 4.5 M⊙Myrs−1in Lupus 1, 3
M. Benedettini et al.
and 4, respectively, indicating that Lupus 3 has a higher star formation activity then Lupus
1 and 4.
The Lupus clouds are included in the extended surveys of nearby molecular clouds that
are being carried out with both ground and space facilities, namely: 70, 100, 160, 250, 350
and 500 µm Gould Belt Herschel survey (Andr´ e et al. 2010); JCMT Gould’s Belt Legacy
Survey with SCUBA2 and HARP-B (Atchell et al. 2005; Johnstone, Di Francesco & Kirk
In order to identify the population of pre- and proto-stellar cores, we carried out a
molecular survey of the three Lupus subgroups where there is evidence of star formation
activity, i.e. Lupus 1, 3 and 4, at millimetre wavelengths in several molecular species that
are good tracers of dense gas. The millimetre data are complementary to the other surveys
of the region and will facilitate their interpretation. The complete set of data will allow to
understand the star formation activity in the Lupus region and more in general the low-mass
star formation process from the cores condensation to the protostellar phase.
2 OBSERVATIONS AND DATA REDUCTION
Molecular line surveys of the Lupus 1 and 3 molecular clouds at 3 and 12 mm were carried
out with the Mopra telescope. Moreover, the Lupus 4 cloud were also mapped at 12 mm.
The observations were executed in two periods: from 17 to 19 July 2008 and from 20 to 26
The observations were carried out in the On The Fly observing mode with the nar-
row band mode of the UNSW-Mopra Spectrometer(UNSW-MOPS) digital filterbank back-
end, and the Monolithic Microwave Integrated Circuit (MMIC) 77 to 116 GHz receiver.
UNSW-MOPS has a 8-GHz bandwidth with four overlapping 2.2-GHz subbands, each sub-
band having four dual-polarization 137.5-MHz-wide windows giving a total of sixteen dual-
polarization windows. Each window has 4096 channels providing a velocity resolution of
0.11 km s−1at 94 GHz and 0.41 km s−1at 22 GHz. We selected the 16 zoom bands in the
range between 19.5 and 27.5 GHz and at 12 mm and between 90 and 98 GHz at 3 mm. The
selected frequencies are listed in Table 1.
Since the beam of the telescope at 12 mm is 2.5′, at this frequency we were able to map
all the zones of the Lupus 1, 3 and 4 clouds with visual extinction larger than 3 mag (see
Fig. 1), i.e. a region of 130′×40′in Lupus 1, 70′×20′in Lupus 3 and 30′×30′in Lupus 4. On
Multiline spectral imaging of dense cores in Lupus
Table 1. List of of the transitions selected for the observation. The transitions that have been detected are in boldface.
the other hand, at 3 mm the telescope beam is significantly smaller (35′′) and we mapped
only the regions were NH3(1,1) emission was detected. Note that the Lupus 4 cloud was not
mapped at 3 mm because we did not have enough time. The scanned regions were covered
by mini maps of 5′×5′and 18′×18′for the observations at 3 and 12 mm, respectively. Each
mini map was scanned twice in orthogonal directions in order to minimize artificial stripes
and reduce noise level.
Data reduction was performed using the ATNF dedicated packages Livedata and Gridzilla1.
Livedata performs a bandpass calibration and baseline fitting while Gridzilla regrids and
combines the data from multiple scanning directions and mini maps onto a single data cube.
The data was Gaussian smoothed so that the effective spatial resolution of the final maps
is 46′′at 3 mm and 2′at 12 mm.
At 12 mm, we detected the NH3 (1,1) transition at 23.694 GHz with the 4 satellites
and the main component of the higher excitation transition NH3(2,2) at 23.722 GHz (only
towards the sources with bright (1,1) emission), no NH3(3,3) emission was detected. At 12