A low-velocity bipolar outflow from a deeply embedded object in Taurus revealed by the Atacama Compact Array
Kakeru Fujishiro, Kazuki Tokuda, Kengo Tachihara, Tatsuyuki Takashima, Yasuo Fukui, Sarolta Zahorecz, Kazuya Saigo, Tomoaki Matsumoto, Kengo Tomida, Masahiro N. Machida, Shu-ichiro Inutsuka, Philippe André, Akiko Kawamura, Toshikazu Onishi
aa r X i v : . [ a s t r o - ph . GA ] J u l Draft version July 21, 2020
Typeset using L A TEX default style in AASTeX63
A low-velocity bipolar outflow from a deeply embedded object in Taurus revealed by the AtacamaCompact Array
Kakeru Fujishiro, Kazuki Tokuda,
2, 3
Kengo Tachihara, Tatsuyuki Takashima, Yasuo Fukui,
Sarolta Zahorecz,
Kazuya Saigo, Tomoaki Matsumoto, Kengo Tomida, Masahiro N. Machida, Shu-ichiro Inutsuka, Philippe Andr´e, Akiko Kawamura, and Toshikazu Onishi Department of Physics, Nagoya University, Chikusa-ku, Nagoya 464-8602, Japan Department of Physical Science, Graduate School of Science, Osaka Prefecture University, 1-1 Gakuen-cho, Naka-ku, Sakai, Osaka599-8531, Japan National Astronomical Observatory of Japan, National Institutes of Natural Science, 2-21-1 Osawa, Mitaka, Tokyo 181-8588, Japan Institute for Advanced Research, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8601, Japan Faculty of Sustainability Studies, Hosei University, Fujimi, Chiyoda-ku, Tokyo 102-8160, Japan Astronomical Institute, Tohoku University, 6-3, Aramaki, Aoba-ku, Sendai, Miyagi 980-8578, Japan Department of Earth and Planetary Sciences, Faculty of Sciences, Kyushu University, Nishi-ku, Fukuoka 819-0395, Japan Laboratoire d’Astrophysique (AIM), CEA, CNRS, Universit´e Paris-Saclay, Universit´e Paris Diderot, Sorbonne Paris Cit´e, F-91191Gif-sur-Yvette, France (Received; Revised; Accepted)
Submitted to ApJLABSTRACTThe first hydrostatic core, the first quasi-hydrostatic object formed during the star formation process,is still the observational missing link between the prestellar and protostellar phases, mainly due toits short lifetime. Although we have not established a clear method to identify this rare object,recent theoretical studies predict that the first core has millimeter continuum emission and low-velocityoutflow with a wide opening angle. An extensive continuum/outflow survey toward a large numberof “starless” cores in nearby star-forming regions works as a pathfinder. We observed 32 prestellarcores in Taurus with an average density of & cm − in 1.3 mm continuum and molecular lines usingthe Atacama Large Millimeter/submillimeter Array–Atacama Compact Array (ALMA–ACA) stand-alone mode. Among the targets, MC35-mm centered at one of the densest “starless” cores in Taurushas blueshifted/redshifted wings in the CO (2–1) line, indicating that there is deeply embeddedobject driving molecular outflow. The observed velocities and sizes of the possible outflow lobesare 2–4 km s − , and ∼ × au, respectively, and the dynamical time is calculated to be ∼ yr. Inaddition to this, the core is one of the strongest N D + (3–2) emitters in our sample. All of the observedsignatures do not conflict with any of the theoretical predictions about the first hydrostatic core sofar, and thus MC35-mm is unique as the only first-core candidate in the Taurus molecular cloud. Keywords: stars: formation — ISM: clouds— ISM: kinematics and dynamics — ISM: individual objects(L1535-NE/MC35) INTRODUCTIONUnderstanding of the low-mass star formation process has been intensively studied from decades ago, mainly bytheoretical work at the beginning (e.g., Larson 1969; Shu 1977; Shu et al. 1987; Inutsuka 2012). They explained that thefragmentation and condensation of molecular clouds result in forming dense cores, which undergo gravitational collapseto form stars. According to the theoretical studies, dense cores eventually harbor the first protostellar cores, the first
Corresponding author: Kazuki [email protected]
Fujishiro et al. quasi-hydrostatic object during the star formation process (hereafter, the first core; e.g., Larson 1969; Masunaga et al.1998; Tomida et al. 2013, ), which provide the initial condition of star formation. Recent magnetohydrodynamic(MHD) simulations demonstrated that the first core with a size of 1–100 au is formed when the central density exceeds ∼ cm − via the gravitational collapse. The first core is suggested to have a low-velocity (1–10 km s − ) molecularoutflow with a wide opening angle (Machida et al. 2008), which is qualitatively different from the collimated jet drivenby a mature protostar. However, it is difficult to identify such an object observationally because the first core has ashort lifetime, 10 –10 yr, depending on the physical condition of the parental core (Tomida et al. 2010), and does notshow bright infrared emission. Although some candidates of the first core were already reported in the past decade(e.g., Chen et al. 2010, 2012; Enoch et al. 2010; Pineda et al. 2011; Pezzuto et al. 2012; Hirano & Liu 2014), the first-core phase is not fully explored observationally and it is supposed to still be the missing link between the isothermaland adiabatic collapse (i.e., prestellar and protostellar core). To search for candidates of the first core, it is essential toperform a survey-type observation toward a large number of starless cores. According to the early dense core surveywith an average density of & cm − (Onishi et al. 2002), the lifetime of the starless phase is ∼ × yr (see alsoWard-Thompson et al. 2007). The simple calculation tells us that only one out of a few × OBSERVATIONS AND DESCRIPTIONS OF THE TARGET OBJECT, L1535-NE/MC35As the detailed observational properties of this project are given in Paper I, and thus we summarize the dataqualities of the target briefly. The angular resolution is 6 . ′′ × . ′′
0, corresponding to 850 au ×
730 au at a distance of126.6 pc (Galli et al. 2018). The frequency settings contain the 1.3 mm continuum emission and molecular lines of CO, CO, C O (2–1), and N D + (3–2). The rms sensitivities of the continuum and lines are ∼ − and ∼ − ( ∼ ∼ − , respectively. We use the combined 7 m + TotalPower (TP) array data to recover the extended emission of CO (Sect. 3.1), while for N D + we use the 7 m arraydata alone to be compared with the compact 1.3 mm source (Sect. 3.2).The target object, L1535-NE (also know as MC35; see Onishi et al. 2002), is located in the Barnard18 regionin Taurus. Hogerheijde & Sandell (2000) found an intensity enhancement in the submillimeter continuum emission,observed by JCMT/SCUBA at 850 µ m and with IRAM 30 m/MAMBO at 1.2 mm by Motte & Andr´e (2001) next to thepreviously know protostar IRAS 04325+2402, which is a Class I source with multiplicity (Hartmann et al. 1999). Thenondetection of infrared point sources at the center of the dense core MC35 by the IRAS deep survey (Beichman et al.1992) means that there is no bright source with an upper limit of ∼ L ⊙ . We checked the archival data of the Herscheland Spitzer and confirmed the non-detection of point sources at wavelengths shorter than 70 µ m. The derived dusttemperature is as cold as 10–15 K (Hogerheijde & Sandell 2000) and the previous molecular gas survey in CO(1–0)with FCRAO could not find an outflow from this object (Narayanan et al. 2012). These observational signatures areconsistent with the fact that there is no associated mature YSO. The single-dish observations found that this core hasstrong molecular line emission of high-density gas tracers, H CO + , N H + , and N D + (Onishi et al. 2002; Tobin et al.2013), indicating that there is a density enhancement at the dust continuum peak. Figure 1 illustrates the observedarea, the 1.2 mm continuum emission by the IRAM 30 m telescope, and infrared images by the Spitzer. The observedcentral coordinate was ( α J . , δ J . )=(4 h m . s
5, +24 ◦ ′ . ′′ RESULTS3.1.
Possible compact molecular outflow at the dust continuum peak
Figure 2 (a) shows the 1.3 mm continuum emission toward MC35. We call the millimeter source “MC35-mm”hereafter. Based on the two-dimensional Gaussian fitting to it, the central coordinate and deconvolved size are derivedto be ( α J . , δ J . ) = (4 h m . s
87, +24 ◦ ′ . ′′
8) and 10 . ′′ × . ′′ ∼ . ◦
1, respectively.The peak position coincides with that obtained by the IRAM 30 m (Motte & Andr´e 2001). This result means that the
LMA observations of dense cores in Taurus Figure 2
Figure 1. (a) The color-scale image and white contour show the 1.2 mm dust continuum intensity toward L1535-NE/MC35and IRAS 04325+2402 taken by IRAM/MAMBO-2 (Kauffmann et al. 2008). The dashed contour shows the observation fieldwith the 7 m array. The black cross denotes the position of 1.3 mm peak observed with the 7 m array (see the text in Sect. 3.1).(b) White contours are the same as panel (a). Color is the R(8 µ m)G(4.5 µ m)B(3.6 µ m) image with Spitzer. The black and redarrows are the directions of the scattered-light nebula and the redshifted parsec-scale CO outflow (see the text in Sect. 4.1). ACA observations traced the innermost part of the mass distribution. The peak and total continuum fluxes obtainedby the ACA are 2.8 mJy beam − and 7.8 mJy, respectively. Although the continuum flux is not the strongest one amongour targets, we detect the strongest N D + emission with a peak brightness temperature of ∼ CO data toward the continuum-detectedprestellar cores. Thus, we searched for high-velocity wing components more than a few km s − apart from the systemicvelocity as the excess emission to the Gaussian-like spectral profile by eye. As a result, we found that MC35-mm hasindications of wing components. Figure 2 (a) shows the distributions of the high-velocity wing components in CO.As one can see, there are bipolar conical-shaped features of the blue/redshifted components, and they seem to beconnected to the millimeter continuum peak. In the channel maps (Figure 3), this nature is also well represented,especially at 4.2–4.6 km s − for the blue component and the 7.9–8.3 km s − for the red one. Figure 2 (b) illustratesthe velocity profiles of the CO emission. The selected velocity ranges of the blueshifted and redshifted componentsare 3.5–4.7 km s − , and 7.5–10.0 km s − , respectively. We could not detect significant CO emission in these velocityranges. We can see clear line broadening up to ∼ − with respect to the systemic velocity. Because the turbulenceof this source (∆ V ∼ − , see Tobin et al. 2013) cannot produce such a relatively high-velocity emission, thereshould be a driving source. The wing components are likely molecular outflow from a deeply embedded object insideMC35-mm. We found the blueshifted gas at the south side of the continuum peak, while there are two redshiftedcomponents at both the north and south sides. If the outflow is viewed away from pole-on, and it has a wide openingangle, the geometric effect can explain the present configuration. We could not find the blueshifted component at thenorthern side, possibly because the emission is very close to the systemic velocity with serious contaminations of theextended emission.We characterize the possible molecular outflow components. We defined the velocity difference between the centralvelocity of the N D + profile (5.9 km s − ) and the edges of the wing emission in CO (Figure 2(b)) as its maximumvelocities. The sizes of the wing components are the distances between the peak position of MC35-mm and the peakof the blue/red contours in Figure 2 (a). Table 1 summarizes the derived parameters from the observed values withan assumption of the inclination angles of the outflows as 30 ◦ and 70 ◦ with respect to the line of sight. The resultantdynamical time (=size/velocity) of the wing is ∼ (1–6) × yr. We estimated the masses of the wing components Fujishiro et al.
Figure 2. (a) Integrated intensity distributions of the blueshifted and redshifted velocity components of the CO(2–1) emissionare shown by the contours superposed on the 1.3 mm image in the pseudo-color and black contours. The black lowest andsubsequent contour levels are 3 σ and 6 σ , respectively. The contour sequences of blue and red are [0.4, 0.6, 0.8] (K km s − ) and[0.6, 0.8, 1.0, 1.2] (K km s − ), respectively. The dashed black line indicates where the mosaic sensitivity falls to 50%. Notethat the primary beam attenuation is not corrected for the display purpose. The filled ellipse at the lower-left corner shows thesynthesized beam of the ACA observation. (b) Average spectra of the redshifted component in CO(2–1) and CO(2–1) withinthe lowest red (north) contour showing in (a). The CO one is offset by +1 K for the visualization. The red and blue dashedlines represent the integrated velocity ranges of the contours in (a). The green solid and dashed lines show the N D + (3–2)profile toward MC35-mm and the systemic velocity of 5.9 km s − derived from the Gaussian fitting. (c) Same as panel (b), butfor the blueshifted component. assuming the local thermo-dynamical equilibrium with a uniform excitation temperature of 20 K and [ CO/H ] =10 − , following the equations written by Pineda et al. (2011). We further discuss these parameters in Sect. 4.1. Table 1.
Outflow propertiesSouth NorthBlue Red RedInclination angle (deg.) obs. 30 70 obs. 30 70 obs. 30 70Maximum velocity (km s − ) 2.3 2.7 6.7 4.2 4.8 12.3 4.2 4.8 12.3Size (10 au) 1.5 3.0 1.6 2.1 4.2 2.2 1.6 3.1 1.6Dynamical time (10 yr) · · · · · · · · · − M ⊙ ) 0.6 · · · · · · · · · · · · · · · · · · Dense gas distributions traced by the N D + emission Figure 4 shows the distributions of N D + (3–2). The peak position of the velocity-integrated intensity (moment 0)image corresponds to that in 1.3 mm, indicating MC35-mm is in a cold/dense state, which is similar to evolvedprestellar cores (e.g., Caselli et al. 2002). The intensity-weighted mean velocity (moment 1) map of N D + (Figure 4(b)) marginally shows a velocity gradient from the northeast to the southwest, which is roughly perpendicular to thatof the wing components (Sect. 3.1). This feature may represent that there is a rotating component at MC35-mm. DISCUSSIONS
LMA observations of dense cores in Taurus Bl ue Red
Figure 3.
Color-scale images and white contours show velocity-channel maps of the 7 m + TP CO (2-1) data toward MC35-mm. The lowest contour level and subsequent steps are 0.06 K km s − . The black contours of each panel are the same as thosein Figure 2 (a). The upper-left corner of each panel gives the integrated velocity ranges. The filled ellipses in the lower-leftcorner of the upper-left panel show the synthesized beam size. Is MC35-mm a candidate of the first core?
We found possible outflow lobes at one of the densest “starless” cores in Taurus. Before discussing the nature of thesource, we mention the possibility of contamination emission by the nearby protostar, IRAS 04325+2402. As shownin Figure 1, there is a bipolar scattered-light nebula centered at the protostar. Based on the Spitzer image, the denseregion of MC35 seems to be illuminated by the scattered light. Although some previous studies reported a parsec-scale
Fujishiro et al. D e c ( J ) (a) (b) Figure 4. (a) The pseudo-color image shows the moment 0 map of N D + (3–2) with the 7 m array. The black contours in eachpanel are the same as those in Figure 2. The filled ellipses at lower-left corner in each panel show the synthesized beam size.(b) The pseudo-color image shows the moment 1 map of N D + (3–2) ( > σ ). redshifted CO outflow from the protostellar system (e.g., Heyer et al. 1987; Narayanan et al. 2012), its direction isinconsistent with that of the scattered-light bipolar nebula (see the red arrow in Figure 1(b)). The Submillimeter Array(SMA) observation by Scholz et al. (2010) has detected compact high-velocity CO components with a size of a fewhundred au, whose elongation is roughly consistent with the scattered-light nebula. The scattered-light nebula tracedby mid-infrared observations and molecular outflow distribution do not necessarily match completely, as discussed byTobin et al. (2010) and Tokuda et al. (2016). Moreover, the bipolar feature of the newly detected wing components ishard to explain as an outflow lobe from another protostar. In summary, we conclude that the wing components arenot just arising from the contamination of the protostellar activity in the southern part.We estimate that the dynamical time of the outflow is approximately (1–6) × yr, depending on its inclination angle(see Sect. 3.1). This outflow is possibly the youngest one among other very low-luminosity objects (VeLLOs) or first-core candidates (Dunham et al. 2011). The outflow velocity is not high compared to the typical low-mass protostar( &
10 km s − , e.g., Hogerheijde et al. 1998), and the distribution shows a wide opening angle. These observationalproperties are consistent with that of another first-core candidate, L1451-mm (Pineda et al. 2011), and the theoreticalsimulations, as mentioned in the introduction. The presence of the N D + emission (Sect. 3.2) can be a key pieceof evidence as a young phase. If there is a mature protostar, the N D + abundance rapidly decreases as the gastemperature increases. Tobin et al. (2013) found two N D + peaks away from the protostar position and a drasticabundance decrease at the inner 1000 au region in L1157. Unlike their observation, we do not find a clear positiondiscrepancy between the N D + and continuum peaks. These features strongly indicate the absence of a matureprotostar here, and thus we conclude that the MC35 core harbors a “candidate” of the first core. The HerschelGould Belt survey (Andr´e et al. 2014) detected the 100 µ m emission with an integrated luminosity of ∼ µ m peak does not completely match the ACA 1.3 mmpeak and is ∼ ′′ closer to the IRAS source.We compare the evolutionary stage of MC35 with that of other condensations just before/after star formation inTaurus. IRAM 041911+1522 is a low-luminosity ( ∼ L ⊙ ) protostar (Dunham et al. 2006). Andr´e et al. (1999) found LMA observations of dense cores in Taurus ∼ L ⊙ (Bourke et al. 2006). The subsequent ALMA studies found that the protostellar mass is ∼ M ⊙ without currentmass accretion activity (Tokuda et al. 2017), and thus the protostar is more evolved than the first-core stage. L1544,which shares several similar properties with MC27/L1521F (Crapsi et al. 2004), is famous for a prototypical prestellarcore. Although recent ALMA 12 m array observations (Caselli et al. 2019) confirmed the centrally concentrated natureof the source, there is no report on the outflow so far. Among our starless sample in the ACA survey, MC5N hasthe strongest continuum emission (Paper I), and we suggested that this core is a highly evolved stage on the verge ofbrown dwarf/very low-mass star formation (Tokuda et al. 2019). We could not detect the outflow toward MC5N aswell, and the evolutionary stage is considered to be younger than MC35 and close to L1544. In summary, we suggestthat MC35-mm is a unique source as the only first-core candidate in the Taurus region.Note that it is observationally difficult to distinguish between the first core and VeLLO (see the review byDunham et al. 2014) so far; we need follow-up high-resolution studies to clarify the evolutionary stages, for exam-ple, by estimating the mass of the central object, as demonstrated by recent ALMA observations toward VeLLOs(Tokuda et al. 2017; Lee et al. 2018). 4.2. Lifetime of the first-core candidate
The single-dish survey in Taurus found that there are ∼
50 prestellar cores with an average density of ∼ cm − ,and their lifetime is (4–5) × yr (Onishi et al. 2002; Paper I). Because the theoretically predicted lifetime of thefirst core is typically ∼ yr (e.g., Larson 1969; Saigo & Tomisaka 2006), it is no wonder that no candidates are foundwithin dozens of targets. However, our survey toward ∼
30 prestellar cores in Taurus detected at least one promisingcandidate of the first core. This result observationally gives us a lifetime of the first core, ∼ yr, which is tentativelyconsistent with the dynamical time of the outflow lobes. Some theoretical studies suggested that the statistical lifetimeof the first core can be longer than 10 yr in the case of the gravitational collapse of a sub-solar-mass condensation (e.g.,Saigo & Tomisaka 2006; Tomida et al. 2010; Stamer & Inutsuka 2018). Although the core MC35 itself is as massiveas ∼ M ⊙ (Motte & Andr´e 2001; Onishi et al. 2002), there is a possibility that the collapsing region is small. Infact, the parental core already forms the protostellar system, IRAS 04325+2402, suggesting that the core fragmentedinto two pieces by some mechanism, and then the southern part collapsed locally.We discuss the possibility of a regional difference in the lifetime of the first core. Recent systematic investigationstoward Perseus by Stephens et al. (2019) using SMA reported that 6 of the 74 observed protostellar targets weresuggested to be the first-core candidates. Note that not all of the 74 targets were confirmed to be protostars becausethey could not detect 1.3 and/or 0.85 mm continuum emission toward several sources (see their Table 7). They alsodiscussed that at least two sources are relatively unlikely as the first-core candidates. The total number of the protostarscan be revised into ∼
70, and four of them are the first-core candidate. If we assume that the Class 0/I phase lasts0.5 Myr (e.g., Dunham et al. 2014), the inferred lifetime of the first-core (candidate) phase is ∼ × yr (=0.5 Myr*4/70), which is significantly longer than the theoretically expected time. We thus speculate that some of the samplesshould contain more mature protostars or be in a prestellar phase. If only one source in Perseus is the first core, theinferred lifetime is less than 10 yr, which is consistent with that of our Taurus result. As suggested by Stephens et al.(2019), L1451-mm is the best candidate in Perseus.If the Perseus candidates are all genuinely a first core, the environment may play a vital role in extending thefirst-core phase. In general, low-mass cluster-forming regions show fragmented core distributions and highly turbulentgas kinematics (see Tachihara et al. 2002; Ward-Thompson et al. 2007) compared to those in isolated star-formingregion, like Taurus. From a theoretical perspective, rotation of a parental core can work to extend the first-corephase (Saigo & Tomisaka 2006). If the more turbulent environment forms a larger number of sub-solar-mass cores(Enoch et al. 2006) with rapid rotation, this can explain the lifetime difference in the first-core phase between thetwo regions. However, our current observational understanding regarding the first-core candidates is still incomplete.Detailed follow-up observations using millimeter/submillimeter interferometric facilities will allow us to understandthe true nature of such candidates further. Fujishiro et al.
ACKNOWLEDGMENTSThis Letter makes use of the following ALMA data: ADS/ JAO.ALMA
Andr´e, P., Motte, F., & Bacmann, A. 1999, ApJL, 513, L57Andr´e, P., Di Francesco, J., Ward-Thompson, D., et al.2014, Protostars and Planets VI, 27Beichman, C. A., Boulanger, F., & Moshir, M. 1992, ApJ,386, 248Bourke, T. L., Myers, P. C., Evans, N. J., et al. 2006,ApJL, 649, L37Caselli, P., Walmsley, C. M., Zucconi, A., et al. 2002, ApJ,565, 331Caselli, P., Pineda, J. E., Zhao, B., et al. 2019, ApJ, 874, 89Chen, X., Arce, H. G., Zhang, Q., et al. 2010, ApJ, 715,1344Chen, X., Arce, H. G., Dunham, M. M., et al. 2012, ApJ,751, 89Crapsi, A., Caselli, P., Walmsley, C. M., et al. 2004, A&A,420, 957Dunham, M. M., Stutz, A. M., Allen, L. E., et al. 2014, inProtostars and Planets VI, ed. H. Beuther et al. (Tucson,AZ: Univ. Arizona Press), 195Dunham, M. M., Chen, X., Arce, H. G., et al. 2011, ApJ,742, 1Dunham, M. M., Evans, N. J., Bourke, T. L., et al. 2006,ApJ, 651, 945Enoch, M. L., Young, K. E., Glenn, J., et al. 2006, ApJ,638, 293Enoch, M. L., Lee, J.-E., Harvey, P., et al. 2010, ApJL, 722,L33Galli, P. A. B., Loinard, L., Ortiz-L´eon, G. N., et al. 2018,ApJ, 859, 33Hartmann, L., Calvet, N., Allen, L., et al. 1999, AJ, 118,1784Heyer, M. H., Snell, R. L., Goldsmith, P. F., et al. 1987,ApJ, 321, 370Hirano, N., & Liu, F.-. chun . 2014, ApJ, 789, 50Hogerheijde, M. R., & Sandell, G. 2000, ApJ, 534, 880Hogerheijde, M. R., van Dishoeck, E. F., Blake, G. A., etal. 1998, ApJ, 502, 315Inutsuka, S. 2012, Progress of Theoretical andExperimental Physics, 2012, 01A307 Kauffmann, J., Bertoldi, F., Bourke, T. L., et al. 2008,A&A, 487, 993Kenyon, S. J., G´omez, M., & Whitney, B. A. 2008, inHandbook of Star Forming Regions, Volume I: TheNorthern Sky, Vol. 4, ed. B. Reipurth(San Francisco, CA:ASP), 405Lee, C. W., Kim, G., Myers, P. C., et al. 2018, ApJ, 865,131Larson, R. B. 1969, MNRAS, 145, 271Masunaga, H., Miyama, S. M., & Inutsuka, S. 1998, ApJ,495, 346Machida, M. N., Inutsuka, S., & Matsumoto, T. 2008, ApJ,676, 1088Mizuno, A., Onishi, T., Hayashi, M., et al. 1994, Nature,368, 719Motte, F., & Andr´e, P. 2001, A&A, 365, 440Narayanan, G., Snell, R., & Bemis, A. 2012, MNRAS, 425,2641Onishi, T., Mizuno, A., & Fukui, Y. 1999, PASJ, 51, 257Onishi, T., Mizuno, A., Kawamura, A., et al. 2002, ApJ,575, 950Pezzuto, S., Elia, D., Schisano, E., et al. 2012, A&A, 547,A54Pineda, J. E., Arce, H. G., Schnee, S., et al. 2011, ApJ, 743,201Saigo, K., & Tomisaka, K. 2006, ApJ, 645, 381Scholz, A., Wood, K., Wilner, D., et al. 2010, MNRAS, 409,1557Shu, F. H. 1977, ApJ, 214, 488Shu, F. H., Adams, F. C., & Lizano, S. 1987, ARA&A, 25,23Stamer, T., &, Inutsuka, S. 2018, ApJ, 869, 179Stephens, I. W., Bourke, T. L., Dunham, M. M., et al.2019, ApJS, 245, 21Tachihara, K., Onishi, T., Mizuno, A., et al. 2002, A&A,385, 909Tobin, J. J., Hartmann, L., & Loinard, L. 2010, ApJL, 722,L12
LMA observations of dense cores in Taurus9