R Lep: Bands 8, 9, and 10, high angular resolution test observation with band-to-band phase referencing taken in High Frequency Long Baseline campaign 2021
Science Target Overview
R Lep is a Mira-type variable with a period of 445 days (Watson et al. 2006) at a distance of 471+88-64 pc, based on the estimate by Andriantsaralaza et al. (2022) using the third Gaia data release (DR3; Gaia Collaboration et al 2005). R Lep has an apparently spherical diameter of 15.20+/-0.20 mas at a pulsation phase of 0.20, in the near-infrared K-band (2.0-2.4 micron), as measured by the VLTI (Hofmann et al. 2005). The mass loss from AGB stars forms a circumstellar envelope (CSE) with a lot of gas and dust. The continuum emission of AGB stars at radio and (sub)millimeter wavelength is consistent with the “radio photosphere” model, in which the dominant opacity is caused by free-free interactions between electrons and neutral H and H^2, and the spectral index is roughly 2 in the radio domain (Reid & Menten 1997; Planesas et al. 2016). The continuum flux densities of R Lep in Bands 6 (224 GHz) and 7 (338 GHz), as measured from the images provided by Ramstedt et al. (2020) at the Strasbourg astronomical Data Center, are 16 mJy and 27 mJy, respectively, implying spectral index of ∼1.3. The flux density at Band 10 (890 GHz) is expected to be >100 mJy.
Mass loss from AGB stars builds up a circumstellar envelope (CSE) rich in molecules and dust. In using different molecular species we can probe the CSE, and therefore the mass loss process, at different radii from the central stars. Hydrogen cyanide (HCN) is one of the most abundant molecules in carbon-rich CSEs and is known to exhibit maser action in many transitions in the (sub)millimeter domain (e.g., Guilloteau et al. 1987; Bieging 2001; Smith et al. 2014; Menten et al. 2018), especially in the vibrationally excited states of the bending mode (see Table A.1 of Jeste et al. 2022, for a list of known HCN maser from C-rich CSEs to date). There are two submillimeter HCN masers at 804.75 and 890.76 GHz, first detected in C-rich CSEs by Schilke et al. (2000) and Schilke & Menten (2003), respectively, using the Caltech Submillimeter Observatory (CSO) 10.4 m telescope on Mauna Kea, that are observable in Band 10 with ALMA. The HCN maser targeted in R Lep is that at 890.7607 GHz, from the J = 10 − 9 transition between the (11^1 0) and (04^0 0) vibrationally excited states with an upper level energy above 4260 K (Hocker & Javan 1967; Barbe et al. 2014).
Wong (2019) reported the first ALMA imaging of bright HCN masers towards C-rich stars including R Lep at an angular resolution of ∼ 0.1 arcsecond, and found that the spatial structure of the HCN masers is generally not well resolved, confirming previous predictions that these masers should arise very close to the star (Schilke & Menten 2003). This gives us some a priori knowledge about the probable properties of target images in these long baseline experiments. The bright and relatively compact maser emission also makes it easy to perform self-calibration (self-cal), which is helpful in assessing the image coherence. Therefore R Lep is an excellent target for use in fully validating ALMA’s long baseline capability in Band 10.
ALMA Data Overview
The ALMA long baseline imaging capability in Bands 8, 9, and 10 was demonstrated in HF-LBC-2021 by observing R Lep, a source with complex structure. In the experiments, the quasar J0504-1512, whose separation angle from R Lep is 1.2 deg, was selected as the phase calibrator. One experiment run was performed in each of Bands 8, 9, and 10. In Band 8 two spectral windows with a bandwidth of 1.875 GHz (frequency width per channel of 0.9766 MHz) and the other two spectral windows with a bandwidth of 2 GHz (frequency width per channel of 15.625 MHz) were set up, while there were eight spectral windows with a bandwidth of 1.875 GHz (frequency width per channel of 0.9766 MHz) in Bands 9 and 10 using the 90 deg Walsh phase switching to enable sideband separation in the double sideband receiving systems. Each spectral window was observed in the two linear polarisation pairs, called XX and YY. In Band 10, one of the spectral windows covered the HCN maser at 890.7607 GHz, in the J=10-9 transition between the (11^1 0) and (04^0 0) vibrationally excited states. In Bands 9 and 10, two of the four low frequency spectral windows to observe the phase calibrator were used due to set-up and instrumental problems.
After the phase and amplitude calibration including B2B phase referencing, the images of R Lep were generated with an imaging area of 512x512 mas^2 with a pixel size of 1 mas, centered at (04h59m36s.3590, -14d48’22”.531) in the International Celestial Reference System (ICRS) at J2000. CASA tclean was used for imaging with a “Briggs” robustness parameter of 0.5 and a small cleaning box around the compact source at the centre of the image. For the continuum images in Bands 8 and 9, the entire frequency range was used, whereas for Band 10 the known frequency range of the HCN maser emission, between 890.664 and 890.860 GHz, was excluded. All the available spectral windows were used to synthesize the continuum image in each Band. To obtain the HCN maser data alone, the continuum model for Band 10 obtained was subtracted from the spectral channels containing the HCN maser emission. The frequency axis was adjusted to the Local Standard of Rest Kinematic (LSRK) frame with the rest frequency of the HCN maser line. Those two procedures were made at once using CASA mstransform. An image cube of the HCN maser emission was made covering 100 channels, centered at 890.742 GHz, with a frequency width per channel of 976.562 kHz. This corresponds to a velocity width of ~0.3 km/s. These data were then imaged channel by channel to obtain the HCN maser cube. The achieved angular resolutions are 13, 7, and 5 mas in Bands 8, 9, and 10, respectively.
The CLEANed continuum images of R Lep obtained after applying B2B phase referencing were used to provide models for phase self-cal. The target scan length (25-30 s on average) was used as the solution interval. The combination of the XX and YY polarization pairs and all of the the spectral windows increased the S/N of the solution. In Band 8, the continuum emission extends over only a few synthesized beams, so accurate phase self-cal solutions were obtained for almost all antennas and the restoring beam sizes are very similar before and after phase self-cal. The self-cal for the entire Bands 9 and 10 continuum data set was not successful, because the continuum source is significantly more resolved at the longer (u, v)-distances, compared to the Band 8 continuum data. These self-cal continuum images have lower angular resolutions, compared to the images with B2B phase referencing only.
Creating the HCN maser cube confirmed that the brightest HCN maser emission, in the spectral channel at the LSRK velocity of 9.4 km/s, was spatially compact. Its image model was used for phase self-cal of this channel with the solution interval of ~6 sec (approximately the correlator data integration period). This produced an improved HCN maser image at the LSRK velocity of 9.4 km/s, and this model was then used for amplitude self-cal with the solution interval of the scan length (25-30 s on average). These phase and amplitude self-cal corrections were then applied to the entire Band 10 HCN maser as well as the continuum data sets. This allowed improved imaging of the Band 10 R Lep HCN maser cube and the continuum image at the full angular resolution of ~5 mas.
Using the data for publication
The following statement should be included in the acknowledgement of papers using the dataset listed above:
"This paper makes use of the following ALMA data: ADS/JAO.ALMA#2011.0.00009.E. ALMA is a partnership of ESO (representing its member states), NSF (USA) and NINS (Japan), together with NRC (Canada) and MOST and ASIAA (Taiwan), and KASI (Republic of Korea), in cooperation with the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO and NAOJ.”
Obtaining the Data
The data products are contained in the following downloadable files:
BAND 8
Readme
Uncalibrated data
Calibrated data
Data reduction scripts
Reference images
BAND 9
Readme
Uncalibrated data
Calibrated data
Data reduction scripts
Reference images
BAND 10
Readme
Uncalibrated data
Calibrated data
Data reduction scripts
Reference images