Mem. S.A.It. Vol.

76, 247
c SAIt 2005
Memorie della

The cores of ρ Ophiuchus

Michael D. Smith1 , Roland Gredel2 , Tigran Khanzadyan2 and Thomas Stanke3

1
Armagh Observatory, College Hill, Armagh, U.K. e-mail: [email protected]
2
Max-Planck-Institut für Astronomie, Königsstuhl 17, 69117 Heidelberg, Germany e-
mail: [email protected], [email protected]
3
Institute for Astronomy, University of Hawaii e-mail: [email protected]

Abstract. We search for dense cores and the molecular outflows which accompany stellar
birth in the rho Ophiuchi cloud L1688 through two unbiased wide-field surveys. A solution
to the problem of how to avoid the dissolution of brown dwarf binaries is suggested: binaries
forming in the widely scattered low-mass cores are not susceptible to disruption since the
region is dynamically relaxed. We also find weak H2 outflows which may be driven by
proto-brown-dwarfs.
The dense cores are detected through continuum emission from associated dust grains at
1.2 mm. Covering over 1 square degree, we detect many previously unknown sources, rang-
ing from extended cores to those harbouring Class II young stars. We analyse the mass and
spatial distributions. The core mass function resembles the stellar initial mass function, both
within the tightly-packed clumps as well as within the less-crowded surroundings. The cores
display a hierarchical spatial distribution with no preferred separation scale length. The ori-
entations of the major axes of cores are consistent with an isotropic distribution whereas
the orientations of core pairs possess a preferential direction on all separation scales, con-
sistent with the filamentary cloud appearance. Our near-infrared survey for molecular hy-
drogen emission covers 350 × 350 . We detect several new H2 flows but the total number of
detected outflows is low and is consistent with the paucity of Class 0 and Class 1 sources in
the molecular cloud. Most of the candidate driving sources are deeply embedded in dense
cores. A very young outflow arises from the newly discovered Class 0 source MMS 126.
Flow directions are generally NE–SW, perpendicular to the above preferred direction. The
apparent extents of molecular flows are related to either the widths or the separation between
cloud filaments.

Key words. ISM – Stars: formation – ISM: clouds – ISM: individual objects: Rho
Ophiuchus – ISM: structure

1. Introduction to large scales, regulation mechanisms in the

form of triggers and inhibitors, in addition to
A large number of processes are operating in unpredictable turbulent magnetohydrodynamic
star formation regions. These work on a wide flow (see Mac Low & Klessen 2004. Yet, a
range of scales, typical of Complex Systems. complex system, however complicated, can be
There are several feedback effects from small analysed. We can first take single entities and
understand the particular processes at work.
Send offprint requests to: M. Smith
248 Michael Smith et al.: Cores of rho Ophiuchus

We can then put together the components to (2004), mainly extending to the north/north-
gain an understanding of the wider picture. west of the main cloud cores.
This requires a combination of high resolu- Our 1.2 mm survey covers an area of
tion and wide field observations with sufficient 4,600 arcmin2 with a resolution of ∼2400 . The
sensitivity and uniformity to provide reliable SIMBA bolometer array at the SEST telescope
statistics. on La Silla/Chile was used during an observ-
We report here on two such surveys of ing run lasting from 2002 July 7 to 12. Similar
one of the nearest star-forming regions. The to other (sub)millimetre surveys, our new map
ρ Ophiuchi molecular cloud complex contains is sensitive only to sufficiently small structures
a number of distinct dark clouds spread over (see also J00 for a discussion of this issue). We
several degrees on the sky. The filamentary detect cores over the entire area (Fig. 1).
and clumpy distribution of the molecular gas From the perspective of our study of
has been revealed through large-scale CO sur- molecular outflows, the protostellar stages of
veys (e.g. Loren 1989). Embedded in the fil- star formation are critical. From the near-
aments are dense clumps (Benson & Myers infrared spectra, Luhman & Rieke (1999) con-
1989; Loren et al. 1990; Tachihara et al. 2000) clude that roughly 17% of the YSOs are Class I
with typical masses in the range 10 – 30 M . sources, which implies a lifetime of about
In turn, these clumps contain fragments termed 0.1 Myr for this stage and the availability of
cores of mass 0.10 M – 10 M , some of which over 20 potential driving sources for powerful
already contain protostars and young stars molecular outflows.
(Motte et al. 1998; Johnstone et al. 2000). In order to detect the outflows in ρ Oph, a
The densest part of the L 1688 cloud also range of surveys at millimetre, infrared and op-
harbours a large number of young stellar ob- tical wavelengths have been undertaken. Some
jects, as studied at mid-infrared wavelengths 15 CO outflows have so far been recorded,
with ISOCAM (Bontemps et al. 2001). To gain where the most prominent examples arise
evidence on how these stars are conceived, we from IRS 44 (Terebey et al. 1989), VLA 1623
need to relate their properties to those of the (Andre et al. 1990; Dent et al. 1995) and
ambient medium. GSS 30 (Tamura et al. 1990). The optical stud-
ies suggest that the outflows extend over scales
up to one degree. Atomic jets and Herbig-Haro
2. Surveys (HH) objects have been discovered through op-
tical searches at the perimeter of the cloud
The ‘main cloud’ of ρ Oph covers an area (Wilking et al. 1997; Gómez et al. 1998; Phelps
of roughly 480 arcmin2 and was surveyed & Barsony 2004; Wu et al. 2002). HH 550 and
at 1.3 mm by Motte et al. (1998,hereafter HH 551 are located more than 1 pc away from
MAN98), who uncovered 62 starless cores and the cloud and possibly form a parsec-scale HH
41 circumstellar structures. These observations flow. It is plausible that optical emission lines
had a resolution of just 11 00 , corresponding to from embedded flows in the denser clumps are
1,400 AU. The inferred distribution of masses not observed because of the large extinction of
of these cores was found to be comparable to AV = 50 − 100 mag in the clumps (Wilking &
the stellar initial mass function, suggesting that Lada 1983).
stellar masses are determined at conception. In contrast, near-infrared (H2 2.12 µm)
Subsequently, the results of a larger, some- imaging observations have detected and ex-
what more sensitive survey at 0.85 mm were plored shock-excited knots and jets located
presented by Johnstone et al. (2000,hereafter in highly obscured areas (Davis & Eisloeffel
J00). This survey covered ∼ 700 arcmin2 with 1995; Dent et al. 1995; Davis et al. 1999;
a resolution of 14 00 , identifying 55 cores. This Grosso et al. 2001). These searches have been
survey was recently followed up by a much confined to small regions, typically 50 × 50 .
more extensive (4 square degrees) but signifi- However, Gómez et al. (2003) presented the re-
cantly more shallow survey by Johnstone et al. sults from near-infrared observations of three
 Michael Smith et al.: Cores of rho Ophiuchus 249

Fig. 1. The locations of all the cores detected Fig. 2. The locations of all the starless cores de-
here (squares) compared to the all the cores tected here (squares) compared to the the star-
detected by Johnstone et al. (2000) (asterisks). less cores detected by Motte et al. (1998) (as-
The coordinates are centred on -16:26:58.41, - terisks). The circle is that employed to separate
24:45:36.2 (2000). A, B and E mark the location inner cores from the outer regions.
of clumps.

partly adjoining regions with a combined area 3. Results: spatial distribution

of 480 square arcminutes.With an integration
time of 480 seconds, many new H2 emission The cores we detect within the inner core-
knots which could constitute 13 distinct H2 crowded region are generally coincident with
outflows were detected. those found in previous surveys. Figs. 1 and 2
show that this is true for separate samples of
all cores and starless cores. This confirms the
reliability of our core-finding technique.
At first sight, the identified cores are
strongly clustered. Following J00, we calculate
However, the close distance to the ρ Oph the two-point correlation function, determining
molecular cloud suggests that typical flow ex- the number of core pairs with given separation.
tensions in ρ Oph are of the order of one de- This is divided by the model distribution for a
gree (i.e. a linear flow extent of 2 pc at a dis- random sample of cores spread over the same
tance of 130 pc). Therefore, we require an un- apparent area. The two-point correlation func-
biased search over a field of 1 square degree tion Φ is shown in Fig. 3.
in order to carry out a complete census of the The correlation function is positive on
molecular outflows. Our near-infrared survey scales under ∼ 105 AU. On small scales. this
was conducted in August 2001 using SOFI at result is consistent with the results obtained by
the NTT on La Silla. Additional data were se- J00, also plotted on the Figure. The inferred
cured at high airmass using the OmegaPrime power law index of -0.63 is somewhat flatter
camera at the Calar Alto 3.5m telescope. The than found by J00,In contrast to J00, however,
result is an unbiased 600 second H2 survey of we exclude the existence of a preferential scale
an area of 1,320 square arcminutes. near to 3 × 104 AU, which could correspond to
250 Michael Smith et al.: Cores of rho Ophiuchus

a Jeans length. Instead, the hierarchical cluster- jet impact, (iii) early stellar and disc winds and
ing is consistent with the generation of cores in (iv) other stars, brown dwarves and planets.
turbulence. The mass functions within and external to
the circle are similar. Thus, while the compact-
ness of a core varies in space (Stanke et al.
2005) the mass distribution does not. This must
be reconciled with the fact that the inferred
surface pressures are, on average, lower in the
external region which would suggest a higher
Jeans mass.
An IMF has been inferred for Class II
young stellar objects in ρ Oph by Bontemps
et al. (2001) from ISOCAM data. They found
an upper break in the IMF at 0.55 M separat-
ing power law slopes with indices −0.35± 0.25
and −1.7 below and above the break, respec-
tively. This IMF is ‘statistically indistinguish-
able’ from the core mass function derived by
MAN98 and is also clearly consistent with the
present data sets.
Fig. 3. The two point correlation function for
the ρ Ophiuchus cores, as defined
√ in the text. 5. Results: outflows
The error bars correspond to N statistics. The
dotted straight line corresponds to a power-law We find that 10 outflows are sufficient to ac-
of the form ω ∝ ∆r−0.63 . count for all the H2 emission. Even so, there
are alignments which could reduce the num-
ber. In fact, if the average size of a protostellar
outflow were 0.6 pc, (e.g. Stanke 2003), then
4. Results: mass distribution it would extend 15 arcmin. In other words,
the driving protostar may be located far from
Fig. 4 displays the mass functions for the 118 the molecular hydrogen features. On the other
starless cores. These results make no presump- hand, quite compact H2 outflows are also iden-
tion about the stratification of the envelopes or tified. One of these appears to stem from what
their subsequent evolution. we suspect is a Class 0 protostar within the
We find that the mass function is similar core MMS 126. Fig. 5 shows the three ex-
to that found by MAN98 and J00 for the in- tended H2 emission knots f05-04a–c which
ner zone. The power law fit with two breaks is is one of the most intriguing flows discovered
comparable to the approximation for the stellar in our survey.
initial mass function (IMF) derived by Kroupa The minimum jet power required to drive
(2001), where the power law indices are 0.7, the H2 shocks can be estimated here. The inte-
−0.3, and −1.3. Note that the flatter power law grated 1–0 S(1) H2 fluxes for the flows f05-4,
holds for masses below ∼ 0.14 M , in compari- f08-01 and f10-04 are 3.4, 12.9 and 29.5
son to the value of 0.08 M quoted for the IMF × 10−17 W m−2 . These convert to the very
(Kroupa 2001). The derived masses are, how- small line luminosities of 1.8, 6.8 and 15.6
ever, sensitive to the assumed dust temperature, × 10−5 L , respectively, for the distance of
opacity and distance. If we assume the break in 130 pc. The total H2 luminosity from the ra-
Φ and the opacity are accurate, we can specu- diating shocked layers is 10–20 times these
late that just about one half of a core mass ends values (Smith 1995). Jet powers are probably
up constructing the star. The other half is dis- 1000 times the observed 1-0 S(1) luminosity.
persed in (i) jets (up to 30%), (ii) dispersal by We thus conclude that driving jet powers are
 Michael Smith et al.: Cores of rho Ophiuchus 251

Fig. 4. The mass function (number distribution) by mass and location of the cores in ρ Ophiuchus
for 118 starless cores. The dotted circle drawn in the inset divides the cores into an inner zone of
61 objects (diamond symbols) and an outer area of 57 objects (triangles).

in the range 0.02–0.2 L . We thus suggest

that the weaker outflows are driven by proto-
brown-dwarves.

6. Conclusions
Further results and implications are discussed
in detail by Khanzadyan et al. (2004) and
Stanke et al. (2005). Deeper and wider surveys
will prove revealing. Unbiased surveys of all
cores in tracers of molecular outflow and in-
flow would provide the type of statistical infor-
mation still lacking.
Low mass objects may form in the low
mass cores recorded here. It is interesting
to note that the core distribution is probably
dynamically relaxed (Belloche et al. 2001).
Fig. 5. The f05-04 outflow. The main grayscale Hence, the low mass products of the cores
image is H2 1-0 S(1) line + Continuum and the probably remain bound to the cloud. This is
inserts are smoothed pure H2 1-0 S(1) line im- the basis for a solution to the problem of how
ages. to avoid the dissolution of brown dwarf bina-
ries. Binaries forming in the widely scattered
252 Michael Smith et al.: Cores of rho Ophiuchus

low-mass cores are not dismantled through dy- Johnstone, D., Wilson, C. D., Moriarty-
namical interactions. Schieven, G., et al. 2000, ApJ, 545, 327
Acknowledgements. We used the SIMBAD Khanzadyan, T., Gredel, R., Smith, M. D., &
database, the Digitized Sky Survey, the Two Micron Stanke, T. 2004, A&A, 426, 171
All Sky Survey and the Starlink Project (UK). Kroupa, P. 2001, MNRAS, 322, 231
Funding was provided by PPARC and DCAL (NI). Loren, R. B. 1989, ApJ, 338, 902
Loren, R. B., Wootten, A., & Wilking, B. A.
References 1990, ApJ, 365, 269
Luhman, K. L. & Rieke, G. H. 1999, ApJ, 525,
Andre, P., Martin-Pintado, J., Despois, D., & 440
Montmerle, T. 1990, A&A, 236, 180 Mac Low, M. & Klessen, R. S. 2004, Reviews
Belloche, A., André, P., & Motte, F. 2001, of Modern Physics, 76, 125
in ASP Conf. Ser. 243: From Darkness to Motte, F., Andre, P., & Neri, R. 1998, A&A,
Light: Origin and Evolution of Young Stellar 336, 150
Clusters, 313–+ Phelps, R. L. & Barsony, M. 2004, AJ, 127,
Benson, P. J. & Myers, P. C. 1989, ApJS, 71, 420
89 Smith, M. D. 1995, A&A, 296, 789
Bontemps, S., André, P., Kaas, A. A., et al. Stanke, T. 2003, Ap&SS, 287, 149
2001, A&A, 372, 173 Stanke, T., Smith, M., Gredel, R., &
Davis, C. J. & Eisloeffel, J. 1995, A&A, 300, Khanzadyan, T. 2005, A&A, submitted
851 Tachihara, K., Mizuno, A., & Fukui, Y. 2000,
Davis, C. J., Smith, M. D., Eislöffel, J., & ApJ, 528, 817
Davies, J. K. 1999, MNRAS, 308, 539 Tamura, M., Sato, S., Kaifu, N., Hough, J. H.,
Dent, W. R. F., Matthews, H. E., & Walther, & Suzuki, H. 1990, ApJ, 350, 728
D. M. 1995, MNRAS, 277, 193 Terebey, S., Vogel, S. N., & Myers, P. C. 1989,
Gómez, M., Stark, D. P., Whitney, B. A., & ApJ, 340, 472
Churchwell, E. 2003, AJ, 126, 863 Wilking, B. A. & Lada, C. J. 1983, ApJ, 274,
Gómez, M., Whitney, B. A., & Wood, K. 1998, 698
AJ, 115, 2018 Wilking, B. A., Schwartz, R. D., Fanetti, T. M.,
Grosso, N., Alves, J., Neuhäuser, R., & & Friel, E. D. 1997, PASP, 109, 549
Montmerle, T. 2001, A&A, 380, L1 Wu, J., Wang, M., Yang, J., Deng, L., & Chen,
Johnstone, D., Di Francesco, J., & Kirk, H. J. 2002, AJ, 123, 1986
2004, ApJ, 611, L45