Early Planet Formation in Embedded Disks (eDisk). VII. Keplerian Disk, Disk Substructure, and Accretion Streamers in the Class 0 Protostar IRAS 16544–1604 in CB

We present observations of the Class 0 protostar IRAS 16544 – 1604 in CB 68 from the “ Early Planet Formation in Embedded Disks ( eDisk ) ” ALMA Large program. The ALMA observations target continuum and lines at 1.3 mm with an angular resolution of ∼ 5 au. The continuum image reveals a dusty protostellar disk with a radius of ∼ 30 au seen close to edge-on and asymmetric structures along both the major and minor axes. While the asymmetry along the minor axis can be interpreted as the effect of the dust ﬂ aring, the asymmetry along the major axis comes from a real nonaxisymmetric structure. The C 18 O image cubes clearly show the gas in the disk that follows a Keplerian rotation pattern around a ∼ 0.14 M e central protostar. Furthermore, there are ∼ 1500 au scale streamer-like features of gas connecting from northeast, north – northwest, and northwest to the disk, as well as the bending out ﬂ ow as seen in the 12 CO ( 2 – 1 ) emission. At the apparent landing point of the NE streamer, there is SO ( 6 5 – 5 4 ) and SiO ( 5 – 4 ) emission detected. The spatial and velocity structure of the NE streamer can be interpreted as a free-falling gas with a conserved speci ﬁ c angular momentum, and the detection of the SO and SiO emission at the tip of the streamer implies the presence of accretion shocks. Our eDisk observations have unveiled that the Class 0 protostar in CB 68 has a Keplerian-rotating disk with a ﬂ aring and nonaxisymmetric structure associated with accretion streamers and out ﬂ ows.


INTRODUCTION
There is now growing observational evidence that planet formation is initiated in circumstellar disks around protostars in their deeply embedded (Class 0/I) stages.For example, the mass of the disks around Class II young stellar objects (YSOs), so-called protoplanetary disks, is low compared to the total mass of solids in exoplanet systems, which suggests that planet formation should happen earlier than the Class II stage (Tychoniec et al. 2018;Ward-Duong et al. 2018;Tychoniec et al. 2020).The concentric ring and gap features found in dust disks around Class II YSOs are thought be related to planet formation processes; e.g., the gaps reflect the orbits of protoplanets and rings are possible growth sites of planetesimals (ALMA Partnership et al. 2015;Pinte et al. 2016;Andrews et al. 2018;Long et al. 2018).Consequently, it is of great interest to characterize the structures of disks around deeply embedded protostars, and to investigate whether they also show ring-like structures as those seen around Class II YSOs -or perhaps precursors to these.
The purpose of the Atacama Large Millimeter/submillimeter Array (ALMA) Large Program "Early Planet Formation in Embedded Disks (eDisk)" is to systematically characterize the structures of disks around a sample of 19 protostars in Class 0/I stages through high spatial resolution images of 1.3 mm continuum and line emission (Ohashi et al. 2023).Specific interests of the target sources of eDisk include: (1) the presence or absence of a Keplerian rotating disk, as well as its size and internal structure (Tobin et al. 2012(Tobin et al. , 2015(Tobin et al. , 2020;;Sheehan et al. 2022); (2) spatial and velocity structure of the protostellar envelope, including the accretion streamers onto the disk, which are found in recent ALMA studies (Yen et al. 2014(Yen et al. , 2019a;;Pineda et al. 2020;Thieme et al. 2022); (3) evolutionary stage of the sources in the context of planet formation.In this paper we present the eDisk results for the Class 0 protostar IRAS 16544-1604 embedded in the Bok globule CB 68, located in the southwestern outskirt of the Ophiuchus North region (Launhardt & Henning 1997;Launhardt et al. 1998Launhardt et al. , 2010)).In our eDisk papers, we simply refer to the Class 0 protostar IRAS 16544-1604 as IRAS 16544.IRAS 16544, one of the Class 0 sources among the eDisk sample, has a bolometric temperature T bol = 50 K and bolometric luminosity L bol = 0.89 L , which are reestimated using the archival photometric data by the eDisk team (see Ohashi et al. 2023, for details).Distance estimates to the ρ Ophiuchus region range from 160 pc (Chini 1981) to 119±6 pc (Lombardi et al. 2008); the latter based on Hipparcos and Tycho parallax measurements.A recent VLBA parallax measurement yielded a distance of 137.3±1.2 pc (Ortiz-León et al. 2017).The distance to the cloud region closest to IRAS 16544 in Ophiuchus North was reported to be 151 +3 −5 pc by Zucker et al. (2020), who adopted an advanced method that combines stellar photometric data with parallax measurements.In this paper we adopt this distance of 151 pc determined within a ∼5 % accuracy.IRAS 16544 drives an expansive bipolar outflow along the northwest to southeast direction at a position angle of ∼140 • , where the northwestern outflow lobe is redshifted and southeastern lobe blueshifted (Wu et al. 1996;Vallée et al. 2000;Vallée & Fiege 2007).Recently, Imai et al. (2022) reported results for IRAS 16544 at a spatial resolution of ∼70 au from the ALMA Large Program "Fifty AU Study of the chemistry in the disk/envelope system of Solar-like protostars (FAUST)".They revealed the presence of a ∼1000-au scale infalling and rotating protostellar envelope by C 18 O and the existence of a velocity gradient on a smaller scale (∼100 au), potentially related to disk rotation, by C 18 O, CH 3 OH, and OCS.Compact emission of various complex organic molecules are also found, suggesting that IRAS 16544 harbors a hot corino.
We here report the ∼5 au resolution observations for IRAS 16544 from eDisk revealing the presence of a protostellar disk in Keplerian rotation as well as accretion streamers from the larger scale cloud to this disk for the first time.The paper is organized as follows: Section 2 describes eDisk observations of IRAS 16544, and parameters of the imaging.Section 3.1 shows the results of the 1.3-mm dust-continuum emission.Images and velocity structures of the 12 CO (2-1) emission are shown in Section 3.2, while those of the 13 CO (2-1), C 18 O (2-1) emission are in Section 3.3.In section 4.1 and 4.2, we analyze the Keplerian disk and streamers in the envelope, respectively.We further discuss the nature of the streamers in section 5.1, and an overall physical picture of the Class 0 protostellar system IRAS 16544 in section 5.2.Section 6 summarizes our main results.Appendix A show the moment maps of the SO, c-C 3 H 2 , and H 2 CO lines and the channel maps of the SiO, CH 3 OH, and DCN lines.The full set of the velocity channel maps of the CO isotopologue lines is presented in Appendix B. 1 Jy beam −1 =24836 K Noise level (continuum, robust=0.0)21 µJy beam −1 =0.5 K

ALMA OBSERVATIONS AND DATA REDUCTION
The IRAS 16544 data presented in this paper were taken from two observing projects; one from the ALMA large program eDisk (project code: 2019.1.00261.L, PI: N. Ohashi) and the other from the Director's Discretionary Time (DDT) program (project code: 2019.A.00034.S, PI: J. J. Tobin).The eDisk observations were conducted to achieve highresolution (<0. 1) imaging.The DDT observation was made to supplement the eDisk observations with a more compact antenna configuration C-4, which is sensitive to extended structures of ∼5 .Details of these observations are summarized in Table 1.Hereafter, we call the eDisk observations/data LB (long baseline) observations/data, and the DDT observation/data SB (short baseline) observation/data.The correlator setup included seven spectral windows (spws).Two spws have a bandwidth of 1.875 GHz to increase the continuum sensitivity, and four spws have a bandwidth of 58.594 MHz to ensure a high enough velocity resolution of the main target lines, and the other one has a 937.5 MHz bandwidth for the 12 CO (2-1) line.Between the LB and SB observations a slight change of the correlator setting was made for the 13 CO (2-1) emission.Table 2 lists the detected molecular lines, along with angular and velocity resolutions and rms noise levels.
The delivered visibility data after the standard pipeline calibrations were further edited and self-calibrated with the Common Astronomy Software Applications (CASA) version 6.2.1.The continuum-only visibility data were extracted from the channels without spectral emission lines detected.The SB continuum visibility data was Fourier-transformed and CLEANed with the CASA task tclean, and the SBonly continuum image was made.Then, self-calibrations of the SB continuum data only were conducted.The phaseonly self-calibration was first conducted with progressively shorter solution intervals, which improved the S/N until the solution interval of "int".The total number of the iteration was six.Then a single amplitude and phase self-calibration was conducted.Next, from the self-calibrated SB visibility data and the three LB visibility data individual 1.3-mm continuum images were made.Through the 2-dimensional Gaussian fitting to the images, the centroid positions of the continuum images were derived, and the alignment of the image center was applied to the visibilities.Then, phase-only self-calibrations of the combined SB+LB data were conducted.We found that while the 1st and 2nd round phaseonly self-calibration improved the S/N, the third round of the phase-only calibration at the intermediate solution interval worsened the S/N.Amplitude and phase self-calibration was thus not performed for the SB+LB data.The calibration tables and the image centering were applied to the line visibility data.A more comprehensive description of the data reduction process can be found in Ohashi et al. (2023), where the standard eDisk data reduction script for each source is also linked.
With the self-calibrated SB+LB datasets, the final 1.3-mm continuum image was made with the Briggs robust parameter of 0.0, which gives the best compromise between the spatial resolution and signal-to-noise ratio (Figure 1).For the line imaging robust=2.0,i.e., Natural weighting, and uv-tapering at 2000 kλ are adopted, except for the LB-only P-V diagram shown in Figure 8b which adopts robust=1.5 and uv-tapering of 2000 kλ.We also made line images with robust=0.5 and the same uv-tapering.We found, however, that those higherresolution line images show patchy features presumably due to the more severe effect of the missing flux.We thus adopted robust=2.0and uv-tapering at 2000 kλ for the line imaging.
The C-coefficient of the HCN line is adopted.
the vicinity of the protostar (Imai et al. 2022).Our highresolution eDisk observations have resolved the elongated disk feature around the protostar IRAS 16544 for the first time.From a two-dimensional Gaussian fitting with the CASA task imf it, the deconvolved full width at half maximum (FWHM) of the 1.3-mm continuum emission along the major × minor axes is derived to be 0. 207 × 0. 060 (31.3 au × 9.1 au) at a position angle of 45 • , where the fitting region is set to include the entire emission extent down to 3σ.Assuming that the dust grains are settled to the midplane, the aspect ratio equals cos i, where i is the inclination angle (i = 0 • means face-on), and i is estimated to be 73 • .Since the grains are not completely settled, this serves as a lower limit.The location of the emission centroid derived from the two-dimensional Gaussian fitting is (α ICRS , δ ICRS )=(16 h 57 m 19.6428 s , −16 d 09 m 24.016 s ), which we regard as the position of the central protostar (crosses in Figure 1).The centroid position of the Gaussian is slightly (∼0.01) offset from the position where the peak intensity of the map is seen.This reflects that the 2D Gaussian structure does not completely represent the actual intensity distribution as also illustrated in the residual map (Figure 1b) after subtracting the fitted Gaussian.
From the 1.3-mm flux density of 50.63 mJy (≡ S ν ) as derived from the integration over the emission area above 3σ, the dust mass of the disk (≡ M dust ) is calculated by the following formula, where ν is the frequency (= 225 GHz), d is the distance (= 151 pc), κ ν is the dust mass opacity, and B ν (T dust ) is the Planck function at dust temperature T dust .The dust opacity of κ 225GHz = 2.3 cm 2 g −1 is adopted (Beckwith et al. 1990).
Adopting a typical value of the dust temperature in Class II disks, i.e., T dust = 20 K (Pascucci et al. 2016), M dust is estimated to be ∼ 34.1 M ⊕ .Tobin et al. (2020) derived the scaling relation between the bolometric luminosity (≡ L bol ) and the dust temperature as; For IRAS 16544, T dust is estimated to be 42 K with L bol = 0.89 L .This dust temperature gives M dust ∼ 14.1 M ⊕ .On the other hand, the peak brightness temperature of the dust emission is as high as 90 K (see Figure 13), suggesting that the dust temperature in the central region should be higher than 90 K. T dust = 100 K gives the dust mass of M dust ∼5.4 M ⊕ .Thus, with a fixed value of κ 225GHz = 2.3 cm 2 g −1 , the range of the dust mass is M dust ∼5.4-34.1 M ⊕ .Assuming the gas-to-dust mass ratio of 100, this dust mass yields the total gas + dust mass of 1.63×10 −3 -1.02×10 −2 M .Note that all of these estimates assume optically-thin 1.3-mm dust emission.Since the high brightness temperature implies the optically thick 1.3-mm dust-continuum emission at least partially, these mass estimates should be regarded as lower limits.

12 CO (J=2-1) Emission
Figure 2 shows various maps of the 12 CO (2-1) emission in IRAS 16544.From the analyses of the Position-Velocity diagrams of the C 18 O emission with SLAM (see Section 4.1), the systemic velocity is derived to be 4.96 km s −1 , which is consistent with that in previous studies (Codella & Muders 1997;Imai et al. 2022).Hereafter in this paper we use V LSR = 5.0 km s −1 as a systemic velocity of IRAS 16544.Figure 2a and 2b compare the integrated-intensity maps of the blueshifted and redshifted 12 CO emission, while Figure 2c shows the moment 1 map.The 12 CO emission is predominantly distributed in a bipolar structure, normal to the elongated continuum structure, with the blueshifted and the redshifted emissions located to the southeast and northwest, respectively.This distribution suggests that the 12 CO emission toward IRAS 16544 primarily traces the molecular out-flow associated with the Class 0 protostar.There are also blueshifted emission toward the northwest and redshifted emission toward the southeast.It indicates that the outflow axis is close to the plane of the sky, which is consistent with the geometry of the dust disk close to edge-on (i 73 • ).
The redshifted outflow lobe to the northwest appears to consist of a number of jet-like emission components, which are most clearly visible in the peak intensity map (moment 8 map; Figure 2d).These emission components point toward a variety of directions, suggesting that the outflow direction may have changed during the protostar's lifetime.From the visual inspection of the moment 0 map (Figure 2b), the mean position angle of the redshifted outflow lobe is derived to be ∼−17 • .On the other hand, from the investigation of the 12 CO velocity channel maps we found that the 12 CO map at V LSR = 6.68 km s −1 shows the V -shaped feature of the outflow cavity most unambiguously (Figure 19).From this map the opening angle of the V -shape is inferred to be ∼60 • for a spatial extent of ∼300 au at the outer intensity threshold of ∼50 σ, which corresponds the lowest contour level of the V -shaped cavity feature.The mean central axis of the redshifted outflow lobe appears to be tilted with respect to the minor axis of the dust disk (dashed line from northwest to southeast in Figure 2).
Fan-shaped feature Jet-like Comp.The blueshifted outflow lobe in contrast does not clearly exhibit such jet-like components toward different directions.The mean position angle of the blueshifted outflow is ∼135 • from the visual inspection of Figure 2a, which is not that of the redshifted lobe plus 180 • .This suggests the presence of the outflow bending, which is also seen in other protostellar sources (e.g., Aso et al. 2018Aso et al. , 2019)).If the magnetic-field structure in a protostellar system is not orthogonal to the disk rotational axis, misalignments between the outflow and disk rotational axes could be produced, and the observed bending outflow structure could reflect such a misalignment (e.g., Matsumoto & Tomisaka 2004;Hirano & Machida 2019;Hirano et al. 2020).
In addition to these outflow components, the 12 CO (2-1) emission also traces distinct components.There is an inverse J-shaped blueshifted component (dashed curve in Figure 2a) starting from northeast from the protostar toward the southeastern end of the protostar.There is another, fan-shaped blueshifted component to the west of the protostar.Whereas the spatial distributions of these components appear to trace the shell surrounding the northern redshifted outflow, these components are blueshifted, and the velocity features of these components appear distinct from those of the outflows.In addition, these components have 13 CO and C 18 O counterparts.We will discuss the natures of these components in the subsequent sections.

13 CO and C 18 O (J=2-1) Emission
Figures 3 and 4 show the 13 CO and C 18 O counterparts of Figure 2.While the extended blueshifted and redshifted 13 CO and C 18 O emission to the southeast and northwest, respectively, resemble the bipolar molecular outflow as traced by the 12 CO emission, the velocity structures of these components are distinct from those of the 12 CO outflows, as discussed below.Note that the velocity range of the 13 CO and C 18 O emission is much narrower than that of the 12 CO emission.Furthermore, there is a blueshifted, inverse J-shaped structure extending to the northeast (dashed curves in Fig- ures 3 and 4).The same component is also seen in the 12 CO emission (Figure 2).There is another 13 CO and C 18 O emission component located to the west of the protostar, which exhibits a blueshifted, fan-shaped feature.This component also has the 12 CO counterpart.
The moment 0 and 1 maps of the CO isotopologue lines in the vicinity of the dust disk are shown in Figure 5.The 13 CO and C 18 O emission associated with the dust disk are visible, and in particular, the C 18 O emission is elongated to the same direction as that of the dust emission (Figure 5b and  c).This suggests that the C 18 O emission traces the molecular gas in the protostellar disk.On the other hand, the 13 CO and particularly the 12 CO emission show more extended structures along the outflow direction, which may suggest that the 13 CO and 12 CO emissions have more contamination from the outflows.Inside the dust disk these CO isotopologue emission are redshifted to the southwest and blueshifted to the northeast.This velocity gradient along the disk major axis can be interpreted as the rotational motion in the disk.On the contrary, outside the dust disk the CO isotopologue emission is red-and blueshifted to the northwest and southeast, respectively.In the 12 CO outflow map, the northwestern side is redshifted and the southeastern side blueshifted.Thus, the northwestern and southeastern sides of the disk correspond to the near-and far-sides of the disk plane, on the assumption that the disk plane is perpendicular to the outflow axis.If the 13 CO and C 18 O components originate from the disk plane, the redshifted and blueshifted emission arise from the nearand far-sides, and such a velocity feature can be interpreted as an infalling motion.
To investigate the velocity features of the molecular gas surrounding the protostar in more detail, Figures 6 and 7 show the velocity channel maps of the C 18 O emission in the high-and low-velocity ranges, respectively.The full 13 CO and C 18 O velocity channel maps for the entire and zoom-in regions are Figure 20, 21, 23, and 24.Note that the channel maps in the high-velocity range are presented in the close vicinity of the dust disk (white contours in Figure 6).The high-velocity blueshifted C 18 O emission is located to the northeastern part of the dust disk while the high-velocity redshifted emission to the southwest, suggesting the presence of the velocity gradient along the disk major axis and rotation in the disk.In the blueshifted velocity range of 3.52-3.85km s −1 , the location of the C 18 O emission is shifted toward southeast, which connects to the emission component in the lower blueshifted velocities shown in Figure 7. Similarly, in the redshifted velocity range of 6.19-6.52km s −1 , the C 18 O emission protrusion toward the west emerges, which connects to the lower redshifted emission in Figure 7.
The C 18 O velocity channel maps at lower velocity ranges over the entire emission area (Figure 7) present gas components distinct from the central disk.At V LSR = 4.02 km s −1 , a blueshifted component to the southeast of the disk is seen.From V LSR = 4.19 km s −1 , a blueshifted, inverse J-shaped structure emerges.This component curls from the southeast of the disk to the northeast, and extends to the northeast progressively until V LSR = 4.69 km s −1 .This velocity feature, that is, higher velocity components located closer to the protostar and lower velocity components further, appears to be distinct from that of the outflows (see also Figure 9b).As we described above, this component is seen in all the CO isotopologue lines.In the velocity range of 4.19-4.69km s −1 , there is another blueshifted component to the west of the protostellar disk.This component is the origin of the fanshaped blueshifted signature seen in Figure 4.This western component also appears to curl to northwest, and presents Inverse J-shaped Comp.
a similar spatial-velocity feature to that of the northeastern blueshifted component, again opposite to that expected from the outflows.In the velocity range of 4.35-4.69km s −1 the C 18 O emission to the south is present.The extent of this blueshifted component is largest at the lowest velocity.
In the redshifted velocity of 6.02 km s −1 , the C 18 O emission is located to the west of the protostar.In the lower redshifted velocities (5.35-5.86km s −1 ), this component appears to be divided into two components; one to the southwest of the disk and the other extending to the northnorthwest (NNW).While the spatial location of the NNW component matches with that of the redshifted outflow lobe as seen in the 12 CO emission, this component extends further to the NNW at lower velocities.A similar velocity feature of the redshifted 13 CO emission to the NNW is identified.The sense of this velocity structure seen in the C 18 O and 13 CO Inverse J-shaped Comp., respectively.5σ clipping is adopted to make the moment 1 map (1σ = 1.9 mJy beam −1 ).

Fan-shaped feature
emission is opposite to that of the Hubble flow of molecular outflows (Arce et al. 2007).

Keplerian Protostellar Disk
The CO isotopologue emission directly associated with the dust disk and most probably tracing the gas disk exhibits a velocity gradient along the northeast (blueshifted) to southwest (redshifted) direction.The velocity channel maps of the C 18 O (2-1) emission at high velocities (Figure 6) show archetypal signatures of the rotation.Figure 5. Moment 0 and 1 maps of the 12 CO, 13 CO, and the C 18 O (2-1) emission in the vicinity of the protostar as labeled.The integrated velocity ranges of the moment 0 maps are −12.37-23.19km s −1 , 0.01-10.03km s −1 , and 1.01-9.36km s −1 for the 12 CO, 13 CO, and, the C 18 O emission, respectively.5σ clipping is adopted to make moment 1 maps (1σ = 1.2 mJy beam −1 , 2.5 mJy beam −1 , and 1.9 mJy beam −1 for the 12 CO, 13 CO, and the C 18 O maps, respectively).Contours denote the 5σ and 150σ levels of the 1.3-mm continuum emission.White and black filled ellipses at the bottom-right corners denote the synthesized beams of the line and 1.3-mm continuum images, respectively.
forth quadrants.This result implies that the SB+LB P-V diagram exhibits velocity structures of the protostellar disk as well as those of the protostellar envelope.On the other hand, the LB only P-V diagram shows the emission predominantly in the first and third quadrants, suggesting that it traces the disk component only.Such a difference between the SB+LB and LB-only P-V diagrams can be attributed to the different degree of the missing flux.In the high-velocity blueshifted and redshifted components (1 km s −1 ≤ |V LSR −V sys | ≤ 3 km s −1 ), the C 18 O emission are well separated to the northeast and southwest, respectively.On the other hand, at lower velocities (|V LSR −V sys | 1 km s −1 ) the SB+LB P-V diagram shows overlaps of the northeastern and southwestern emission components at the same velocity.These results suggest that the high-velocity components trace the rotating disk, while the lower-velocity components correspond to the rotating and infalling protostellar envelope surrounding the disk as already demonstrated by the FAUST results (Imai et al. 2022).
Fitting of rotation curves to the C 18 O P-V diagrams along the disk major axis was performed, using the Spectral Line Analysis/Modeling (SLAM; Aso & Sai (2023)) code (Ohashi et al. 2023).SLAM identifies an emission peak along each velocity or positional axis with the Gaussian fitting, and derives the best parameters of the rotational profile, i.e., the central stellar mass (≡ M ), the power-law index of the rotation velocity (≡ p, where v rot ∝ r −p ), and the outermost radius of the disk (≡ R disk ), through the Markov Chain Monte Carlo (MCMC) algorithm.The SLAM fitting to the emission ridge in the LB-only P-V diagram with an intensity threshold of 3σ (1σ = 2.0 mJy beam −1 ) indicates p = 0.451±0.015and R disk = 25.69±0.50au (solid orange curves in Figure 8b).The rotational power-law index is close to 0.5 of Keplerian rotation, which implies that the high-velocity C 18 O emission associated with the dust disk indeed traces the Keplerian rotation.The central protostellar mass is then derived to be M = 0.158±0.003M .We note that these quoted errors are only statistical and arise from the MCMC process, and that SLAM defines 68% of the posterior distribution as   the range of the 1σ error.The actual error of the protostellar mass should be larger.For example, Czekala et al. (2015) adopted a more sophisticated fitting to the visibilities in the 3-dimensional space and quoted 4% error bars on dynamical mass measurements of the pre-main sequence spectroscopic binary AK Sco.Measurement of the dynamical mass of IRAS 16544 through such a full 3-dimensional fitting is deferred to our next eDisk papers.
The SB+LB P-V diagram, on the other hand, should include both the central Keplerian disk and the outer envelope.We thus adopted two rotational profiles using a double power-law function, i.e., the outer rotational profile in the envelope and the inner disk rotational profile.To avoid contamination from extended low-velocity gas, the fitting velocity range is restricted to 0.5 km s −1 ≤ |V LSR −V sys |≤ 4 km s −1 , and the intensity threshold of 5σ (1σ = 1.3 mJy beam −1 ) is adopted.The double power-law fitting to the P-V diagram with SLAM yields the power-law indices in the inner and outer regions of p = 0.514 ± 0.019 and p = 1.014 ± 0.039, respectively.The central protostellar mass is derived to be M = 0.137±0.003M , consistent with that from the LBonly fitting.The outer most disk radius, or the radius of the breaking point of the rotational profile, is derived to be R disk = 54.55±0.50au.The larger disk radius as derived from the SB+LB data than that derived from the LB data only is likely attributed to the effect of the missing flux.Previous FAUST observations of IRAS 16544 have also derived the central stellar mass in the range of M = 0.08-0.3M , from the comparison of the P-V diagram of a model infalling-rotating envelope to that of the observed C 18 O (2-1), CH 3 OH (4 2 -3 1 E), and OCS (19-18) emission.While this mass estimate is consistent with that of the present work, the FAUST estimate does not pinpoint the mass range and assumes that the protostellar envelope is free-falling, which is not necessarily the case (Takakuwa et al. 2013;Ohashi et al. 2014;Aso et al. 2015Aso et al. , 2017)).Our higher-resolution eDisk observations have succeeded to resolve the disk structure and to identify the Keplerian rotation in the protostellar disk, which enables us to refine the value of the central protostellar mass.

Modeling of the Elongated Molecular Structure
In addition to the Keplerian protostellar disk, we have also identified intriguing elongated features at ∼1500 au scales seen in the lines of the CO isotopologues.The velocity channel maps of the C 18 O emission at low velocities (Figure 7) exhibit at least three such features, inverse J-shaped strucrure, blueshifted component to the northwest of the disk, and redshifted one extending to the NNW.The highervelocity parts in these features are located closer to the protostar, which is opposite to the molecular outflow.Those highvelocity parts apparently deviate from the disk components as shown in Figures 6 and 7 at velocities of 3.52-3.85km s −1 and 6.19-6.52km s −1 .These features are reminiscent of accretion streamers recently observed in young, Class 0 protostars (e.g., Thieme et al. 2022).
We here discuss the spatial and velocity structure of the inverse J-shaped structure to the northeast (hereafter we call this component "NE streamer").Figure 9a shows the moment 0 map of NE streamer as seen in the C 18 O emission.A prominent inverse J-shaped feature is evident in the moment 0 map.The P-V diagram along the model curve that traces the emission ridge (see below) is shown in Figure 9b.There appears to be a long chain of the C 18 O emission to the northeast of the source that is blueshifted, which we denote as NE streamer.NE streamer exhibits a gradual increase of the line-of-sight velocity toward the central disk.
To reproduce these observational results, we calculated spatial and velocity trails of ballistic accretion based on the formulae by Ulrich (1976) and Cassen & Moosman (1981) (hereafter CMU model).The radial (≡ v r ) and azimuthal velocities (≡ v φ ) and the trail of the accretion are expressed as; where G is the gravitational constant, M is the mass of the central protostar, r is the radius, φ is the azimuth angle on the plane of the streamer, and r d is the centrifugal radius.The mass of the central protostar is adopted as M = 0.14 M , which is derived from the SLAM fitting (see section 4.1).We then varied r d and φ 0 , the incident azimuthal angle, as well as the inclination i s and position angles θ s of the rotational axis of the streamer, to match the observed spatial and velocity structures with the calculated trail by visual inspection.i s is defined as the angle between the rotational axis and the line-of-sight.φ 0 is the angle of the streamer to which the parabolic trajectory opens, and φ 0 = 0 • indicates west when θ s is 0 degree and increases counter-clockwise.The visually determined parameter set is φ 0 = 64 • , r d = 100 au, i s = 73 • , and θ s = 60 • , with M = 0.14 M (red curves in Figure 9).Illustration of the derived streamer line in the 3-dimensional space is shown in Figure 10.Note that r d is the parameter of the rotational angular momentum of the infalling material, and thus this can be different from the present radius of the disk toward which the materials are accreting.The calculated curve reproduces both the observed inverse J-shaped feature and the gradual increase of the velocity as the position gets closer to the disk.This result implies that NE streamer could be interpreted as an accretion streamer toward the central disk, although the curve in the P-V diagram does not perfectly trace the observed spatial/velocity locations of the emission ridge.

Accretion Streamer to the Protostellar Disk
From our model fitting in section 4.2, it is likely that the NE elongated structure traces the trajectory of the material accreting to the disk.Figure 11 compares the spatial distribution of NE streamer and the moment 0 maps of a) the SO (6 5 -5 4 ) and b) SiO (5-4) emission (contours) at the common velocity range.Figure 11 shows that the SO and SiO emission appear to trace the tip of NE streamer.Furthermore, the SO and SiO emission appears to curl toward the northeast, following the trail of NE streamer.Since the SO and SiO lines are known to be shocked gas tracers (Bachiller et al. 1998(Bachiller et al. , 2001;;Hirano et al. 2006;Sakai et al. 2014;Oya et al. 2018;Okoda et al. 2021), one of the possible interpretations for these emission distributions is that these emission trace the accretion shock at the landing point of the streamer.The centrifugal radius of NE streamer is estimated to be ∼100 au (section 4.2).This radius is larger than the radius of the Keplerian rotating disk as inferred from the SLAM fitting (section 4.1).Thus NE streamer is likely to accrete onto the envelope outside the Keplerian rotating disk.A schematic picture of the streamer plus disk system is shown in Figure 12.
Previous ALMA observations of protostellar envelopes have also found similar accretion streamers.Yen et al. (2014) have revealed blueshifted (∼2000 au in length) and redshifted (∼5000 au) gas streamers in the C 18 O (2-1) emission toward the Class I protostar L1489 IRS.The blueshifted and redshifted streamers are found to accrete onto the plane of the large (r ∼300 au) Keplerian protostellar disk above and below the disk plane, respectively.The velocity structures in these two steamers are consistent with free-falling gas flows with the centrifugal radius of 300 au and the mass accretion rate of 4-7×10 −7 M yr −1 .In HL Tau, Yen et al. (2019a) have found an intriguing one-arm spiral with a length of 520 au in the HCO + (3-2) emission, which extends from southwest to northwest of the planet-forming disk, and curls toward northwestern vicinity of the disk center.Kinematical analyses of this HCO + component reveals that the spiral is a rotating and infalling flow above the disk surface.In the Class 0 protostar Lupus 3-MMS, Thieme et al. (2022) found multiple extended accretion flows along the outflow cavities in the C 18 O (2-1) emission.The flows matched well with the CMU model with the mass accretion rates of 0.5-1.1×10−6 M yr −1 .While those flows follow the edges of the outflows, they have revealed that those structures are not outflow cavities but accretion flows, with the help of their kinematical model.Note that the streamer found in IRAS 16544 also resembles the outflow cavity at a first glance.However, the channel map shows that the high-velocity component is de-tected near the central star and the low-velocity component is away from the IRAS 16544, suggesting that this component is not outflow-related.
These growing pieces of evidence imply that accretion streamers in protostellar envelopes are not rare, but could be a common astrophysical phenomenon.Recent high-resolution, high dynamic-range observations of protostellar envelopes with ALMA have been finding these accretion streamers, which are in contrast with the classical picture that protostellar envelopes are continuous gas structures with rotation and infalling motions.If non-uniform, filamentary or fiber-like structures surround the natal dense cores (Hacar & Tafalla 2011;Hacar et al. 2013Hacar et al. , 2017)), it is natural that these gas structures exhibit accretion streamers in the course of protostellar formation.Numerical simulations of magnetized turbulent cloud cores indeed show such kinds of filamentary structures (Kuffmeier et al. 2017).Furthermore, a flattened envelope formed in the magnetized turbulent core is warped and exhibits spiral structures around a centrifugallysupported rotating disk (Li et al. 2014).Non-ideal MHD simulations with ambipolar diffusion also predict infalling spirals connecting to the central disk (Zhao et al. 2016(Zhao et al. , 2018)).

Physical Properties of the Disk Associated with the
Class 0 Protostar IRAS 16544 Our high-resolution eDisk observations have succeeded to identify the Keplerian disk around the Class 0 protostar IRAS 16544.Previous interferometric observations have been finding a number of Keplerian rotating disks around Class I protostars (e.g., Takakuwa et al. 2012;Yen et al. 2014;Aso et al. 2015;Yen et al. 2017), but only a handful of Class 0 protostars associated with the Keplerian disks are identified (e.g., Tobin et al. 2012;Murillo et al. 2013;Ohashi et al. 2014;Aso et al. 2017).A statistical study of Class 0 disks by CA-LYPSO argues that disk sizes around Class 0 sources could be smaller than those around Class I sources (see e.g., Maury et al. 2019).ALMA observations of a Class 0 protostar B335 identified the upper limit of the Keplerian disk of 5 au (Yen et al. 2015(Yen et al. , 2019b;;Bjerkeli et al. 2019;Imai et al. 2019).Systematic studies of disk sizes as a function of the protostellar evolutionary sequence will be the subject to the forthcoming eDisk papers.
The dust disk exhibits non-Gaussian, asymmetric structures, as shown in the residual image after the subtraction of the fitted Gaussian (Figure 1b).To investigate the distribution of the dust emission more closely, the intensity profiles of the 1.3-mm dust-continuum emission along the major and minor axes are shown in Figure 13.The origin of the profiles is set to be the centroid position as derived from the Gaussian fitting.It is obvious that the Gaussian centroid does not match with the position of the emission peaks both along the  5-4) emission (contours).These shock tracers all peak to the southeast of the disk and might be the streamer's landing point.The integrated velocity ranges of the C 18 O, SO, and SiO emission are 3.184-4.520km s −1 , 3.184-4.520km s −1 , and 3.18-4.52km s −1 , respectively.Note that the SiO map is made from the integration over the two velocity channels at VLSR = 3.18 and 4.52 km s −1 , because of the coarser velocity resolution of the SiO data (= 1.34 km s −1 ).Contour levels are 5σ, 8σ, 12σ, 15σ, 20σ, 30σ, 40σ, 50σ, and 60σ.1σ = 1.2 and 1.3 mJy beam −1 km s −1 in panels (a) and (b), respectively.Blue and black open ellipses at the bottom-right corners denote the beam sizes of the C 18 O and the relevant molecular lines in the panels.The gray dashed lines are the the same as in Figure 2.
major and minor axes.Along the major axis, the northeastern and southwestern parts are not mirror-symmetric, and there is a possible "shoulder" around ∼0. 11 in the northeastern side.This may imply that the disk is not azimuthally symmetric.Along the minor axis, the peak location is ∼0.01 offset toward southeast from the Gaussian centroid, and the northwestern profile is shallower than the southeastern profile within ∼0.05.
The observed peak brightness temperature of the 1.3-mm dust-continuum emission exceeds 90 K.While detailed radiative transfer modeling is required, such a high brightness temperature of the 1.3-mm dust-continuum emission likely indicates that the 1.3-mm emission is optically thick.The observed outflows are redshifted to the northwest and blueshifted to the southeast, and thus the northwestern side of the dust disk is on the near side while the southeastern side is the far side.If the 1.3-mm dust-continuum emission is optically thick and the dust distribution is flared, toward the southeastern, far side, the flared, warm disk surface is directly seen without the cold dust component intervening along the line of sight.On the other hand, toward the northwestern, near side, we should see the colder portion of the disk close to the midplane along the line of sight.Thus the observed shift of the peak position of the 1.3-mm dustcontinuum emission along the minor axis can be interpreted as the disk flaring.Such a dust flaring has been directly imaged by the VLA observations of the optically-thin 7-mm dust-continuum emission toward the Class 0 protostar L1527 IRS by Sheehan et al. (2022).
Stability of a protostellar disk can be estimated with the Toomre Q parameter as, where M disk is the mass of the disk, M is the mass of the star, R is the radius, and H is the scale height at a radius R (e.g., Kratter & Lodato 2016;Tobin et al. 2020).From the results of the dust continuum emission and the SLAM fitting of IRAS 16544, we derived that M disk is 1.63×10 −3 -1.02×10 −2 M and M is ∼ 0.14 M .Assuming the typical value of H R = 0.1, the range of Toomre Q is 17-2.7.These nominal range of the Q value suggests that the protostellar disk is gravitationally stable.On the other hand, the 1.3-mm dust emission is probably optically thick as discussed above and the derived disk mass should be regarded as a lower limit.If the true disk mass is several times higher, the protostellar disk should be gravitationally unstable.Observations at  longer wavelengths, such ALMA Band 1-3, are required to properly estimate the disk mass and the stability of the disk.If the true Q value is less than ∼1, the major axis asymmetry and the detected shoulder-feature could be due to spirals induced by the gravitational instability.
Whereas the gravitational instability in a massive disk is a possible cause of the observed non-axisymmetric dust feature, there are other physical mechanisms to produce such a disk asymmetry.In the case of IRAS 16544, the accretion streamers are observed, and the shock could affect the protostellar disk.The accretion shock can generate strong spiral density waves, which could exhibit the non-axisymmetric feature in the disks (Lesur et al. 2015).Kuznetsova et al. (2022) have incorporated heterogeneous infall based on the CMU model and an embedded disk, and performed hydrodynamic simulations to investigate the effect of the anisotropic accretion steamers on the disk structure.Their results demonstrate that the anisotropic infall induces the Rossby Wave Instability (RWI) in the disk, and forms a vortex and azimuthally asymmetric feature in the disk.Recent growing number of observational evidence for accretion streamers, combined with this hydrodynamic simulation, im-plies that the streamers could be one of the main physical mechanisms to form the asymmetric structure in the disks.This in turn could affect future evolution of disks and planet formation therein.Within their parameter space, Kuznetsova et al. (2022) also argued that disk self-gravity does not play an important role, with the Toomre Q parameter above the marginal stability criterion.Another physical mechanism of formation of azimuthal asymmetric structure in the disk incorporates presence of planetary and substellar companions.The companions in the disk can also induce RWI, vortex and gas horseshoes, and spiral density waves, which show asymmetric millimeter dust-continuum emission (van der Marel et al. 2021).
IRAS 16544 is associated with an active, prominent molecular outflow.A rotating and infalling protostellar envelope has also been identified in IRAS 16544 (Imai et al. 2022), and we have found a possible accretion streamers.These results suggest that IRAS 16544 is in the active mass accretion phase, typical of Class 0 sources.Our high-resolution eDisk observations have unveiled that such an active Class 0 protostar also has a well-developed Keplerian rotating disk with shoulder East West hints of flaring and non-axissymmetric substructure, which could be related to future planet formation in the disk.

SUMMARY
We have carried out high-resolution (0. 036 × 0. 027 ∼5 au) and high-sensitivity ALMA observations of the young Class 0 protostar IRAS 16544-1604 embedded in Bok Globule CB 68 with the 1.3-mm dust-continuum emission, 12 CO, 13 CO, C 18 O (J = 2-1), SO (J N = 6 5 -5 4 ), and other Band 6 lines as a part of the ALMA Large Program, eDisk.The main results are as follows: 1.The 1.3-mm dust-continuum emission reveals a r ∼30 au protostellar disk along the northeast to southwest direction at a position angle of ∼45 • .The aspect ratio indicates that the disk is near to edge-on, with an inclination angle of ∼73 • .Along the minor axis, the emission peak is skewed toward southeast, and beyond the peak the emission profile is steeper in the southeastern side than that in the northwestern side.The skewed intensity profile implies that the 1.3-mm dustcontinuum emission is optically thick and the dust distribution is flared, i.e., dusts are yet to be settled onto the midplane.The intensity profile along the major axis is not mirror-symmetric but asymmetric, with a possible shoulder to the ∼17 au northeast.This suggests the presence of non-axissymetric structure in this Class 0 protostellar disk.
4. In the outer region, the C 18 O emission shows multiple streamer-like features.The most prominent one is an inverse J-shaped, blueshifted component elongated toward the northeast (NE streamer), which has 12 CO and 13 CO counterparts.There are other such features; redshifted component to the north-northwest and blueshifted one to the northwest.In these features, the higher-velocity components are located closer to the protostellar disk, suggesting accretion motion.We searched for the trajectory of the ballistic infalling flow which reproduces the moment 0 map and the P-V diagram of NE streamer.We found that a centrifugal barrier of 100 au reasonably reproduces the spatial and velocity structure of NE streamer, with a central protostellar mass of 0.14 M derived from the P-V analysis of the Keplerian disk.Furthermore, the SO (6 5 -5 4 ) and SiO (5-4) emission, shock tracers, are seen at the tip of NE streamer.Since the centrifugal radius of NE streamer is larger than the radius of the Keplerian disk (∼50 au), NE streamer is landing onto the envelope outside the Keplerian rotating disk, where the accretion shock takes place.
Our high-resolution eDisk observations of the Class 0 protostar IRAS 16544 have identified a compact Keplerianrotating disk associated with flared and non-axisymmetric dust distribution.At the same time, active mass outflow and mass accretion are ongoing through the powerful molecular outflows and accretion streamers, respectively.The nonaxisymmetric signature in the dust disk could reflect the spiral features induced by the gravitational instability in the disk, or interaction with the accretion streamers, although the optical thickness of the 1.3-mm dust-continuum emission prevents us from directly investigating the disk mass and its gravitational stability.These results present an updated physical picture of the Class 0 stage, when formation of substructures often seen in Class II disks has started.The integrated velocity range of the moment 0 map is from 1.514 km s −1 to 9.196 km s −1 .5σ clipping is adopted to make the moment 1 map (1σ = 2.4 mJy beam −1 ).The white and gray dashed lines are the the same as in Figure 2.

A. SPATIAL AND VELOCITY STRUCTURES OF THE DETECTED MOLECULAR LINES
The spectral setting of eDisk enables us to observe a number of ancillary molecular lines simultaneously (see Table 2).In this appendix, we present spatial and velocity structures of these molecular lines.
Figure 14 shows the moment 0, 1, and moment 8 maps of the SO (J N = 6 5 -5 4 ) emission in IRAS 16544.The primary SO emission peak is located to southeast of the dust disk, and the secondary peak is located to the northwest.Additional emission peaks are seen to the east and northeast of the primary peak, and these emission components likely comprise the blueshifted, NE streamer seen in the CO isotopologue lines (see Figure 14b).On the other hand, the secondary SO peak to the northwest is redshifted.To the west of the protostar, an extended, fan-shaped redshifted SO emission is present.The blueshifted component to the west of the protostar as seen in the CO isotopologue lines (see Figure 7) is not seen in the SO emission, and difference of the distributions between the SO and CO isotopologue emission in the 3-dimensional space is present.This is the reason why the moment 1 map of the SO emission shows redshifted to the west while that of the CO isotopologue emission blueshifted.
Figures 15, 16, and 17 show velocity channel maps of the SiO (5-4), CH 3 OH (4 2 -3 1 E), and DCN (3-2) emission, respectively.In the blueshifted velocity of 3.18-4.52km s −1 , the SiO emission peaks toward the southeast of the dust disk.At V LSR = 4.52 km s −1 , the SiO emission distribution curls toward the northeast, consistent with our interpretation that the SiO emission traces the tip of NE streamer.The SiO emission is also present in the southeastern side at the redshifted velocity V LSR = 5.86 km s −1 .The CH 3 OH emission is located toward the southeast of the disk, but the peak location is closer to the disk than that of the SiO emission.In addition, the CH 3 OH emission shows a clear velocity gradient along the disk major axis, consistent with the disk rotation.Thus the CH 3 OH emission could be originated from the protostellar disk.The DCN emission shows a similar emission component to the southeast with a velocity gradient along the disk major axis, while another redshifted component to the northwest is also seen.
Figure 18 compares the moment 0 maps of the three c-C 3 H 2 and H 2 CO lines.The c-C 3 H 2 emission are extended (∼2000 au), and a number of patchy c-C 3 H 2 emission components are present.The 6 0,6 -5 1,5 and 5 1,4 -4 2,3 transitions are weak the protostellar disk, while the 5 2,4 -4 1,3 transition peaks at the disk location.On the other hand, all the three H 2 CO transitions exhibit strong peaks toward the protostar.H 2 CO lines also trace the extended component surrounding the disk, and the 3 0,3 -2 0,2 transition also appears to trace NE streamer.Multi-transitional analysis using these three transitions of c-C 3 2 and H 2 CO should provide us with important insights on the physical conditions of the molecular gas, which should be subject to the subsequent papers.

Figure 1
Figure 1.a) 1.3-mm dust-continuum image of IRAS 16544 with a robust parameter 0.0.Contour levels are 5σ, 20σ, 40σ, 60σ, 80σ, 100σ, 120σ, 150σ, and 180σ (1σ = 21 µJy beam −1 ).A cross at the center denotes the centroid position of the 1.3-mm dust-continuum emission as derived from the two-dimensional Gaussian fitting, which is regarded as the protostellar position.A filled ellipse at the bottom-right corner shows the synthesized beam (0. 036 × 0. 027; P.A. = 69 • ).b) Residual 1.3-mm dust-continuum image after subtracting the fitted two-dimensional Gaussian (color), overlaid with the observed image in white contours.Contour levels are the same as those in a).

Figure 2 .
Figure 2. Various maps of the 12 CO (2-1) emission in IRAS 16544.a), b) Moment 0 maps in the blueshifted (VLSR = −2.21-4.78km s −1 ) and redshifted (VLSR = 5.41-23.19km s −1 ) velocity ranges.The color ranges from 4.0×10 −3 to 4.0×10 −1 Jy beam −1 km s −1 .The apparent redshifted components to the southeastern and southwestern ends are interferometric artifacts.c) Intensity-weighted mean velocity map (Moment 1 map).10σ clipping is adopted to make this map (1σ = 1.2 mJy beam −1 ).d) Map of the peak intensity in the spectra (Moment 8 map).In all the panels, the gray dashed lines show the major and minor axes of the dust-continuum image, and the filled ellipse at the bottom-right corner the synthesized beam.A blue dashed curve in panel a) delineates the inverse J-shaped feature.
Figure 8 shows Position-Velocity (P-V) diagrams of the C 18 O (2-1) emission along the major axis of the dust disk (P.A.= 45 • ) using the SB+LB data (a) and the LB data only (b).The SB+LB and LB-only P-V diagrams are made by averaging the spectra over the transverse width of 0. 15 and 0. 05, respectively.The SB+LB P-V diagram shows the molecular emission not only in the first and third quadrants but also in the second and

Figure 6 .
Figure 6.Channel maps of the C 18 O (2-1) emission in the high-velocity blueshifted (upper panels) and redshifted ranges (lower panels).Numbers in the upper-left corners denote the LSR velocities.The systemic velocity is 5.0 km s −1 .Contours denote the distribution of the 1.3-mm continuum emission at 5σ and 150σ levels (1σ = 21 µJy beam −1 ).

Figure 7 .Figure 8 .
Figure 7. Channel maps of the C 18 O (2-1) emission in the low-velocity blueshifted (upper panels) and redshifted ranges (lower panels).Crosses show the position of the protostar, and numbers in the upper-left corners denote the LSR velocities.

Figure 9 .
Figure 9. a) Comparison of the solution of an accretion streamer to the spatial distribution of NE streamer as seen in the moment 0 map of the C 18 O emission in the blueshifted velocity range (3.5-4.7 km s −1 ).A red curve represents the solution at φ0 = 64 • , is = 73 • , and θs = 60 • , with a central stellar mass of 0.14 M and a centrifugal radius of 100 au.b) P-V diagram of the C 18 O emission along the red curve shown in panel a).A red curve shows a spatial and velocity trail of the accretion streamer shown in panel (a).

Figure 11 .
Figure 11.Comparison of the spatial distributions of NE streamer as seen in the C 18 O (2-1) emission (colors) and the shock tracers of a) SO (65-54) and b) SiO (5-4) emission (contours).These shock tracers all peak to the southeast of the disk and might be the streamer's landing point.The integrated velocity ranges of the C 18 O, SO, and SiO emission are 3.184-4.520km s −1 , 3.184-4.520km s −1 , and 3.18-4.52km s −1 , respectively.Note that the SiO map is made from the integration over the two velocity channels at VLSR = 3.18 and 4.52 km s −1 , because of the coarser velocity resolution of the SiO data (= 1.34 km s −1 ).Contour levels are 5σ, 8σ, 12σ, 15σ, 20σ, 30σ, 40σ, 50σ, and 60σ.1σ = 1.2 and 1.3 mJy beam −1 km s −1 in panels (a) and (b), respectively.Blue and black open ellipses at the bottom-right corners denote the beam sizes of the C 18 O and the relevant molecular lines in the panels.The gray dashed lines are the the same as in Figure 2.

Figure 12 .
Figure12.Schematic picture of the protostellar disk and NE streamer in IRAS 16544 projected onto the plane of the sky obtained with the eDisk observations.For comparison, a zoom-in view of Figure11b, with the distribution of the dust-continuum emission in white contours overlaid, is also shown aside.The contour levels of the continuum emission are 5σ and 150σ (1σ = 21 µJy beam −1 ).The white ellipse at the lower right corner shows the beam size of the dust continuum emission.

Figure 13 .
Figure13.Intensity profiles of the 1.3-mm dust-continuum emission along the major (light green curve) and minor (orange) axes in IRAS 16544.The origin of the coordinates, denoted by a vertical dotted line, is set to be the centroid position of the continuum emission as derived from the two-dimensional Gaussian fitting.The positive direction of the major axis corresponds to the northeast, and the positive direction of the minor axis corresponds to southeast.The right vertical axis indicates the brightness temperature.Shaded areas in the profiles denote the ±3σ ranges.For comparison, the profile of the geometrically-averaged beam is shown in a gray curve.

Figure 14 .
Figure14.Moment 0, 1, and 8 maps of the SO (65-54) emission in IRAS 16544.The integrated velocity range of the moment 0 map is from 1.514 km s −1 to 9.196 km s −1 .5σ clipping is adopted to make the moment 1 map (1σ = 2.4 mJy beam −1 ).The white and gray dashed lines are the the same as in Figure2.

Figure 20 .
Figure 20.Velocity channel maps of the 13 CO (2-1) emission toward IRAS 16544 over the entire region.Crosses indicate the position of the protostar.A white ellipse at the lower-left corner in the lowest-left panel shows the synthesized beam.

Figure 23 .
Figure 23.Velocity channel maps of the 13 CO (2-1) emission in the high-velocity blue-(upper panels) and red-shifted ranges (lower panels) toward IRAS 16544 in the zoom-in region.Contours denote the distribution of the 1.3-mm dust-continuum emission, and the contour levels are 5σ and 150σ (1σ = 21 µJy beam −1 ).

Figure 24 .
Figure 24.Velocity channel maps of the C 18 O (2-1) emission in the high-velocity blue-(upper panels) and red-shifted ranges (lower panels) toward IRAS 16544 in the zoom-in region.Contours denote the distribution of the 1.3-mm dust-continuum emission, and the contour levels are 5σ and 150σ (1σ = 21 µJy beam −1 ).

Table 2 .
Observed Molecular Lines Critical density at TK = 20 K, except for that of the SO and c-C3H2 lines which adopt 60 K and 30 K, respectively.Calculated from the Einstein Aand C-coefficients from the LAMDA database.
(Schöier et al. 2005base(Schöier et al. 2005).c Einstein A coefficient.d e Velocity resolution.f Eu and A values are from the CDMS database