Data¶
Public data releases will be hosted on CANFAR and linked here as they become available. The esction below summarizes the survey design and observational setup.
Observations¶
- Telescope: Karl G. Jansky Very Large Array (VLA)
- Receiver / band: K band (18–26.5 GHz)
- Configuration: D
- Project code: VLA/26A-517 (Large Program)
- Total time: 342.5 hours, divided between semesters 26A (~174 hr) and 27B (~169 hr)
- Angular resolution: \(\approx 4.4''\) (Natural weighting), giving linear scales of \(\sim 600\) au for the nearest clouds (Taurus, Ophiuchus, Pipe) to \(\sim 1700\) au for Orion B
- Mosaic footprint: ~447 arcmin² across ~35 fields and ~1000 raster pointings
Spectral setup¶
The WIDAR correlator is configured with a hybrid 8-bit / 3-bit sampler scheme:
- 8-bit baseband (A/C, centered at 23.7 GHz) — primary science lines: NH₃ (1,1), (2,2), (3,3), (4,4); HC₅N (9–8); SO₂; CH₃OH
- 3-bit basebands (B/D, centered at 19.07 and 21.64 GHz) — secondary lines: HC₃N (2–1), HC₅N (7–6), CCS, c-C₃H₂ (8–7), HNCO
NH₃ (1,1) and (2,2) hyperfine components are covered at 1024 × 8 MHz (~50 km/s window) for the broadest expected line widths; other species use 256 × 2 MHz subbands at \(\sim 0.05\) km/s resolution.
Sensitivity¶
The on-source integration is set to deliver an RMS of 0.3 K per 0.3 km/s channel — sufficient to reach \(N(\mathrm{NH_3}) \approx 10^{14}\) cm⁻² at 3σ and to detect NH₃ (2,2) at 3–5σ for kinetic-temperature determination.
Combination with GBT¶
ProPStarK is designed as a follow-up to the GBT Ammonia Survey (GAS) (Friesen, Pineda et al. 2017). All VLA cubes will be jointly deconvolved with the matching GBT single-dish data to recover extended emission at all spatial scales between ~4″ and several arcminutes.
Reduction and analysis¶
- Calibration: CASA, building on the standard NRAO calibration pipeline
- Imaging: PHANGS-style pipeline (phangs_imaging_scripts), adapted to VLA K-band, with comparison across
tp2vis,feather, start-model, andsdintimagingfor single-dish/interferometer combination - NH₃ line fitting:
pyspeckit(Ginsburg et al. 2022) hyperfine model, with multi-component and HPC-accelerated fits vianestfit - Compute: NRAO and MPE clusters; ~19 TB raw data volume
Data products¶
- Calibrated visibilities (NRAO archive)
- VLA-only and joint VLA+GBT image cubes for each targeted line
- Moment maps and derived physical-condition maps: \(T_K\), \(N(\mathrm{NH_3})\), \(v_\mathrm{LSR}\), \(\sigma_v\), excitation temperature
- Streamer and fragment catalogs