Free Access
Issue
A&A
Volume 578, June 2015
Article Number A20
Number of page(s) 7
Section Interstellar and circumstellar matter
DOI https://doi.org/10.1051/0004-6361/201526021
Published online 27 May 2015

© ESO, 2015

1. Introduction

Feedback processes associated with the collapse of protostellar envelopes at 103 − 104 AU scales limit the accretion onto the protostar and contribute to the overall low efficiency of transferring gas into stars on global scales (Offner et al. 2009; Krumholz et al. 2014). Calculating the physical conditions help to identify the most relevant phenomena and constrain their role in the star formation process (Evans 1999). Since gas in protostellar envelopes is heated to temperatures much higher than the dust temperatures, molecular transitions are the suitable tracers of physical conditions of hot (T ≳ 100 K) gas around protostars. In particular, the far-infrared (IR) lines of CO and H2O dominate the cooling of hot and dense gas (Goldsmith & Langer 1978). The excitation of CO and H2O depends on the local physical conditions (temperature, density) and thus is crucial in order to determine which physical mechanisms are responsible for the gas heating and to study whether the energetics involved in the feedback scale from low- to intermediate- to high-mass young stellar objects (YSOs).

Recent observations of CO and H2O lines with the Photodetector Array Camera and Spectrometer (PACS, Poglitsch et al. 2010) on board Herschel found large columns of dense (104 cm-3) and hot (300 K) gas towards low-mass (Lbol ≲ 102L) protostars (van Kempen et al. 2010; Herczeg et al. 2012; Manoj et al. 2013; Karska et al. 2013; Green et al. 2013; Lindberg et al. 2014) that originate largely from UV-irradiated shocks associated with jets and winds (Karska et al. 2014b). The CO and H2O line luminosities of the high-mass protostars (with Lbol ~ 104 − 106L) follow the correlations with bolometric luminosities found in the low-mass protostars (Karska et al. 2014a) and show similar velocity-resolved line profiles regardless of the mass of the protostar (Yıldız et al. 2013; San José-García et al. 2013; van der Tak et al. 2013). In contrast, rotational temperatures of H2O are lower and the H2O fraction contributed to the total cooling in lines with respect to CO is higher for the low-mass protostars (Karska et al. 2014a; Goicoechea et al. 2015), suggesting that the physical mechanisms causing the excitation in low- and high-mass protostars are different.

Table 1

CO and H2O rotational excitation.

Intermediate-mass YSOs (with Lbol ~ 102−103L1) provide a natural link between low- and high mass protostars, but their far-IR CO and H2O emission has only been studied for a single protostar position of NGC 7129 FIRS 2 (Fich et al. 2010) and the outflow position of NGC 2071 (Neufeld et al. 2014). CO emission alone was analyzed for two intermediate-mass protostars in Orion, HOPS 288 and 370 (Manoj et al. 2013). In this paper, we present the analysis of PACS spectra for the full sample of intermediate-mass protostars from the “Water in star forming regions with Herschel” (WISH) key program (van Dishoeck et al. 2011), including the maps of NGC 7129 and NGC 2071 centered on the YSO position. These results complement the work by Wampfler et al. (2013), which describes the OH excitation in our source sample and the sample of low- and high-mass protostars for which CO and H2O emission is discussed in Karska et al. (2013, 2014a). The main question addressed is whether CO and H2O rotational temperatures differ from low- to high-mass protostars.

The paper is organized as follows. Section 2 briefly introduces the observations, Sect. 3 the excitation analysis using rotational diagrams, and Sect. 4 discusses the results.

2. Observations

Our sample includes six YSOs with bolometric luminosities from 70 to 2000 L and located at an average distance of 700 pc (see Table 1). The sources were selected based on their small distances (1 kpc) and location accessible for follow-up observations from the southern hemisphere (for more details see Sect. 4.4.2. in van Dishoeck et al. 2011). Spectroscopy for all sources was obtained with PACS as part of the WISH program. For observing details see Table A.1.

With PACS, we obtained single footprint spectral maps covering a field of view of ~47′′ × 47′′ and resolved into 5 × 5 array of spatial pixels (spaxels) of ~9.4′′ × 9.4′′ each. At the distance to the sources, the full array corresponds to spatial scales of ~2−6 × 104 AU in diameter, on the order of full maps of low-mass YSOs (~104 AU) in Karska et al. (2013) and the central spaxel spectra of more distant high-mass YSOs (~3 × 104 AU) in Karska et al. (2014a).

The observations were taken in the line spectroscopy mode, which provide deep integrations of 0.5–2 μm wide spectral regions within the ~55–210 μm PACS range. Two nod positions were used for chopping 3 on each side of the source (for details of our observing strategy and basic reduction methods, see Karska et al. 2013). The full list of targeted CO and H2O lines and the calculated line fluxes are shown in Table A.2. The quality of the spectra is illustrated in Fig. A.1. We note that the simultaneous non-detections of the H2O 716–707 line at 84.7 μm and detections of the H2O 818–707 line at 63.3 μm in L 1641, NGC 2071, and NGC 7129 FIRS 2 are related to the structure of the H2O energy levels and not the differences in the sensitivities of the instrument in the second- and third-order observations. The 818–707 line is a backbone transition, with the Einstein A coefficient higher than the value for the 716–707 line.

thumbnail Fig. 1

Rotational diagrams of CO. The base-10 logarithm of the number of emitting molecules from the upper level, , divided by the degeneracy of the level, gup, is shown as a function of energy of the upper level in kelvins, Eup. Detections are shown as filled circles, whereas three-sigma upper limits are shown as empty circles. The empty upper triangle corresponds to the line flux calculated using a smaller area on the map than the rest of the lines. Blue lines show linear fits to the data and the corresponding rotational temperatures. Errors associated with the least-square linear fit are shown in brackets.

Open with DEXTER

thumbnail Fig. 2

Similar to Fig. 1 but for H2O.

Open with DEXTER

The data reduction was done using HIPE v.13 with Calibration Tree 65 and subsequent analysis with customized IDL programs (see, e.g., Karska et al. 2014b). The fluxes were calculated using the emission from the entire maps. Figure A.2 illustrates that both the line and continuum emission peaks approximately at the source position, with small shifts in continuum due to mispointing. The extent of the line emission usually follows the continuum pattern with the exception of Vela 17 where the line emission extends from NE to SW direction, while the continuum is centrally peaked. There is no contamination detected from the nearby sources or their outflows in the targeted regions.

3. Rotational diagrams

3.1. Results

thumbnail Fig. 3

Rotational temperatures of warm (Eup = 14 − 24) CO (top) and H2O (bottom) as function of bolometric luminosity. Orange circles show YSOs from Green et al. (2013), Karska et al. (2013), and Manoj et al. (2013). Dark blue crosses show intermediate-mass YSOs from the WISH program and light blue diamonds high-mass YSOs from Karska et al. (2014a). A few sources with luminosities ~ 102L shown with orange crosses are from Manoj et al. (2013) and can also be regarded as intermediate-mass YSOs. Uncertainties of the CO and H2O rotational temperatures are shown for L1641 S3 MMS1, which is representative of the sample.

Open with DEXTER

Figure 1 shows rotational diagrams of CO calculated for all sources in the same way as in Karska et al. (2014a). The corresponding rotational temperatures, Trot, and total numbers of emitting molecules, , are shown in Table 1.

All sources show a 300 K, “warm” CO component (Manoj et al. 2013; Karska et al. 2013; Green et al. 2013), with a mean temperature of ~320 K for the range of total numbers of emitting molecules, . In addition, NGC 2071 and NGC 7129 show a “hot” CO component with temperatures of ~700 K and .

The rotational diagrams for H2O, presented in Fig. 2, show a single component with a mean temperature of ~120 K. The scatter due to subthermal excitation and high opacities exceeds the uncertainties in the observed fluxes, similar to diagrams of low- and high-mass YSOs (for the discussion of both effects see Sect. 4.2.2 of Karska et al. 2014a). The corresponding numbers of emitting molecules are about 3 orders of magnitude lower than for CO, .

3.2. Comparison to low- and high-mass sources

Figure 3 shows a comparison between rotational temperatures obtained for CO and H2O for the intermediate-mass sources presented here and these quantities determined in the same way for low- and high-mass YSOs. The comparison is restricted only to the warm component seen in CO rotational diagrams owing to the low number of sources with detections of the hot CO component; those will be discussed in detail in Karska et al. (in prep.). A bolometric luminosity is used here as a proxy for the protostellar mass, but in fact some of our intermediate-mass sources may be a collection of unresolved low-mass protostars.

The median Trot of CO in low-mass protostars is 325 K, using the results from the WISH (Karska et al. 2013), “Dust, Ice, and Gas in Time” (Green et al. 2013), and “Herschel Orion Protostar Survey” (Manoj et al. 2013) programs for a total of about 50 sources. A comparable value of ~290 K was found for ten high-mass sources (Lbol ~ 104 − 106L) in Karska et al. (2014a). The range of CO Trot of 265–370 K (see Table 1) determined here for intermediate-mass YSOs is thus fully consistent with previous results. The fact that the 300 K CO component does not depend on the source bolometric luminosity over 6 orders of magnitude suggests that the origin is in a shock associated with the jet/winds impact on the envelope rather than in a photodissociation region where Trot should scale with the UV flux and luminosity (van Kempen et al. 2010; Visser et al. 2012; Manoj et al. 2013; Flower & Pineau des Forêts 2013; Kristensen et al. 2013).

The median H2O rotational temperatures (see the bottom panel of Fig. 3) for low- and high-mass YSOs are ~130 K and ~230 K, respectively (Karska et al. 2013, 2014a). The range of temperatures obtained for the intermediate-mass YSOs (90–150 K, Table 1) is thus comparable to the low-mass sources. However, the uncertainties in the rotational temperatures are high, on the order of ~100 K, and do not account for the optical depth effects, the density effects, and the possible complexity of the line profiles (absorption and emission components). It is therefore unclear if there is a true jump in rotational temperatures at Lbol ~ 104L, or a smooth trend toward higher values of Trot. In either case, these higher excitation temperatures could be due to higher densities in the more massive envelopes.

4. Summary

We analyze the excitation of far-IR CO and H2O lines in six intermediate-mass YSOs observed in the WISH survey and compare the results to low- and high-mass protostars. Rotational temperatures of CO and H2O are found to be ~320 K and ~120 K, respectively, and are consistent with low-mass and high-mass YSOs within the uncertainties. The large uncertainties in the H2O rotational temperatures, on the order of 100 K, and the order of magnitude gap in the bolometric luminosity between intermediate- and high-mass protostars does not allow us to conclude whether the changes are a smooth function of luminosity. Still, the similarities in rotational temperatures seen for sources with luminosities spanning 6 orders of magnitude and

probed at different spatial scales strongly suggest the same excitation mechanism, the UV-irradiated shocks associated with jets and winds for all sources across the luminosity range (Kristensen et al. 2013; Karska et al. 2014b; Mottram et al. 2014).


1

We use bolometric luminosity as a proxy of the protostellar mass for practical and historical reasons, but we note that Lbol changes significantly during the protostellar phase if the accretion is episodic (Young & Evans 2005; Dunham et al. 2010). Moreover, some of our intermediate-mass sources may actually be a collection of unresolved low-mass protostars.

Acknowledgments

The authors would like to thank the referee for the valuable comments which helped to improve the manuscript. Herschel is an ESA space observatory with science instruments provided by European-led Principal Investigator consortia and with important participation from NASA. A.K. acknowledges support from the Polish National Science Center grant 2013/11/N/ST9/00400. Research conducted within the scope of the HECOLS International Associated Laboratory, supported in part by the Polish NCN grant Dec-2013/08/M/ST9/00664.

References

Online material

Appendix A: Supplementary material

Table A.1 shows the observing log of PACS observations including observations identification numbers (OBSID), observation day (OD), date of observation, total integration time, and pointed coordinates (RA, Dec). Table A.2 shows molecular data and observed lines fluxes and upper limits for all sources that are analyzed in the paper. Figure A.1 shows selected spectral regions to illustrate the quality of the data. Figure A.2 illustrates the patterns of continuum emission at 145 μm and the CO 18–17 line emission at 144 μm toward all the sources.

Table A.1

Log of PACS observations.

Table A.2

Molecular dataa and observed line fluxes.

thumbnail Fig. A.1

Spectral scans covering selected H2O, CO, and OH lines in the intermediate-mass protostars from the WISH program. The rest wavelength of each line is indicated by dashed lines: blue for H2O, red for CO, and light blue for OH.

Open with DEXTER

thumbnail Fig. A.2

PACS spectral maps in the CO 18-17 line at 144 μm and the continuum emission at 145 μm in orange contours corresponding to 30%, 50%, 70%, and 90% of the peak value written in the bottom left corner of each map. Wavelengths in microns are translated to the velocity scale on the X-axis using laboratory wavelengths (see Table A.2) of the species and cover the range from –600 to 600 km s-1. The Y-axis shows fluxes in Jy normalized to the spaxel with the brightest line on the map in a range from –0.2 to 1.2.

Open with DEXTER

All Tables

Table 1

CO and H2O rotational excitation.

Table A.1

Log of PACS observations.

Table A.2

Molecular dataa and observed line fluxes.

All Figures

thumbnail Fig. 1

Rotational diagrams of CO. The base-10 logarithm of the number of emitting molecules from the upper level, , divided by the degeneracy of the level, gup, is shown as a function of energy of the upper level in kelvins, Eup. Detections are shown as filled circles, whereas three-sigma upper limits are shown as empty circles. The empty upper triangle corresponds to the line flux calculated using a smaller area on the map than the rest of the lines. Blue lines show linear fits to the data and the corresponding rotational temperatures. Errors associated with the least-square linear fit are shown in brackets.

Open with DEXTER
In the text
thumbnail Fig. 2

Similar to Fig. 1 but for H2O.

Open with DEXTER
In the text
thumbnail Fig. 3

Rotational temperatures of warm (Eup = 14 − 24) CO (top) and H2O (bottom) as function of bolometric luminosity. Orange circles show YSOs from Green et al. (2013), Karska et al. (2013), and Manoj et al. (2013). Dark blue crosses show intermediate-mass YSOs from the WISH program and light blue diamonds high-mass YSOs from Karska et al. (2014a). A few sources with luminosities ~ 102L shown with orange crosses are from Manoj et al. (2013) and can also be regarded as intermediate-mass YSOs. Uncertainties of the CO and H2O rotational temperatures are shown for L1641 S3 MMS1, which is representative of the sample.

Open with DEXTER
In the text
thumbnail Fig. A.1

Spectral scans covering selected H2O, CO, and OH lines in the intermediate-mass protostars from the WISH program. The rest wavelength of each line is indicated by dashed lines: blue for H2O, red for CO, and light blue for OH.

Open with DEXTER
In the text
thumbnail Fig. A.2

PACS spectral maps in the CO 18-17 line at 144 μm and the continuum emission at 145 μm in orange contours corresponding to 30%, 50%, 70%, and 90% of the peak value written in the bottom left corner of each map. Wavelengths in microns are translated to the velocity scale on the X-axis using laboratory wavelengths (see Table A.2) of the species and cover the range from –600 to 600 km s-1. The Y-axis shows fluxes in Jy normalized to the spaxel with the brightest line on the map in a range from –0.2 to 1.2.

Open with DEXTER
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.