Bayesian estimation of spectral parameters of the 6.7 - GHz methanol maser G339.884 - 1.259 from GRAO observations 2015 Bayesian estimation of spectral parameters of the 6.7 - GHz methanol maser G339.884 - 1.259 from GRAO observations – 7
Theophilus Ansah-Narh, Stephen Sottie, Nia Imara, Emmanuel Proven-Adzri · 2026
Read Bayesian estimation of spectral parameters of the 6.7 - GHz methanol maser G339.884 - 1.259 from GRAO observations 2015 Bayesian estimation of spectral parameters of the 6.7 - GHz methanol maser G339.884 - 1.259 from GRAO observations – 7 by Theophilus Ansah-Narh, Stephen Sottie, Nia Imara, Emmanuel Proven-Adzri (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-00768.txt.
Introduction
Ansah-Narh and co-authors take apart a radio signal from a star-forming cloud in this 2026 paper. The signal is a 6.7-GHz methanol maser, named G339-884-1-259. It was caught by the Ghana Radio Astronomy Observatory. Maser spectra blend many parts, and small shifts in the fit change what the cloud seems to be.
Standard Gaussian fits can miss odd shapes and give no sense of doubt. The team builds a Bayesian fit that tries Gaussian, Lorentzian, and Voigt profiles. A Markov chain Monte Carlo sampler maps the doubt in each fit. Their method finds seven parts that move as one in velocity. The Voigt model wins on the AIC and BIC scores, with the least error and an R-squared of 0.985. Pure shapes leave clear left-over patterns. High chi-squared values hint at fine parts the data cannot yet split.
Catalog record
| CREATOR | Theophilus Ansah-Narh, Stephen Sottie, Nia Imara, Emmanuel Proven-Adzri |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2606.00768
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Bayesian estimation of spectral parameters of the 6.7 - GHz methanol maser G339.884 - 1.259 from GRAO observations 2015 Bayesian estimation of spectral parameters of the 6.7 - GHz methanol maser G339.884 - 1.259 from GRAO observations – 7 Theophilus Ansah-Narh, Stephen Sottie, Nia Imara, Emmanuel Proven-Adzri Abstract Accurate decomposition of methanol maser spectra is essential for constraining the kinematics and physical conditions of high-mass star-forming regions, particularly in complex blended spectra where small differences in component structure can alter physical interpretation. Conventional profile fitting approaches often rely on fixed Gaussian decompositions that do not fully capture non-Gaussian spectral structure and provide limited statistical characterisation of parameter uncertainties. To address this, we develop a Bayesian spectral decomposition framework that models the 6.7 GHz methanol maser emission using alternative Gaussian, Lorentzian, and Voigt profile families, with parameter posteriors inferred through Markov Chain Monte Carlo sampling. This probabilistic framework enables simultaneous model comparison, uncertainty quantification, and statistically consistent estimation of spectral components. Application to the methanol maser G339.884 − - 1.259 observed with the Ghana Radio Astronomy Observatory reveals a complex multi-component spectrum composed of seven velocity-coherent features with tightly constrained parameters. Comparative analysis demonstrates that models incorporating both Doppler-like cores and Lorentzian wing structure provide the statistically preferred representation of the observed spectra, yielding the lowest information criteria (AIC ≈ 1.98 × 10 4 \approx 1.98\times 10^{4} ; BIC ≈ 1.99 × 10 4 \approx 1.99\times 10^{4} ), the smallest residual errors (RMSE ≈ 11.1 \approx 11.1 Jy), and the highest goodness-of-fit ( R 2 ≈ 0.985 R^{2}\approx 0.985 ). In contrast, purely Gaussian or Lorentzian representations leave systematic residual structure near the line wings and strongest maser components. Elevated reduced χ ν 2 \chi^{2}_{\nu} values across all tested models further indicate that unresolved spectral substructure, non-ideal noise properties, and intrinsic line-profile complexity remain important limitations in single-dish maser decomposition. These results demonstrate that Bayesian inference provides a robust and reproducible framework for analysing complex maser spectra while simultaneously quantifying parameter uncertainties and model limitations. The methodology is readily extendable to other molecular line studies and establishes a pathway toward integrating statistically rigorous spectral modelling with high-resolution interferometric observations to better constrain the dynamics and environments of massive star formation. 1 Introduction The 6.7 GHz transition of methanol (CH 3 OH) produces some of the brightest maser lines in the Galaxy, with flux densities often reaching hundreds to thousands of Janskys. Since its discovery by 26 , this transition has become a fundamental tool for identifying and characterising regions of high-mass star formation. A growing body of evidence indicates that Class II methanol masers, of which the 6.7 GHz line is the dominant representative, are exclusively associated with the early stages of massive star formation, typically tracing phases preceding the development of ultracompact H ii regions ( 3 ; 22 ; 14 ) . Their high brightness and compact emission make them ideal targets for high-resolution studies of the kinematics and physical conditions in the innermost environments of massive protostars, including accretion disks, jets, and outflows. The source G339.884 - 1.259 stands out as one of the most prominent and well-studied 6.7 GHz methanol masers in the southern sky. It is a bright and complex source, with interferometric observations revealing a linear distribution of maser spots and a monotonic velocity gradient that initially suggested the presence of a rotating circumstellar disk around a massive star ( 13 ) . Later mid-infrared imaging, however, resolved the emission into three distinct sources, challenging the single-disk interpretation and pointing instead to a more complicated environment involving a collimated jet or outflow ( 10 ) . The region also hosts the brightest known H 2 CO maser in the Milky Way and displays a rich spectrum at millimetre wavelengths ( 5 ) . This source thus offers a valuable laboratory for studying the complex interplay between accretion, outflow, and maser excitation in massive star formation. Despite the wealth of information that can be extracted from the 6.7 GHz methanol maser line, the standard approach to analysing its spectral profile has remained largely unchanged for three decades. Investigators typically inspect the spectrum manually, identify a set of visually distinct peaks, and then fit a sum of Gaussian functions to the data, often using least-squares algorithms. While widely used and often effective, such approaches can face important limitations when applied to complex blended maser spectra. First, the choice of the number of components and their initial parameters is subjective; different researchers may obtain different decompositions from the same spectrum, and small changes in component selection can substantially alter the fitted parameters. Second, the Gaussian model itself is only an approximation. In maser environments, deviations from purely Gaussian profiles may arise from several mechanisms, including unresolved velocity gradients, turbulence, partial saturation, radiative transfer effects, spectral blending of nearby maser spots, and potentially damping-related processes ( 12 ) . Consequently, a purely Gaussian representation may not always capture extended line wings or subtle asymmetries present in high signal-to-noise spectra. Restricting the decomposition to Gaussian components alone may therefore overlook non-Gaussian structure in the observed profiles and bias the inferred component parameters. Third, conventional fitting methods, whether based on Gaussian or Voigt functions, may in some cases return primarily point estimates of the parameters without fully characterising the associated uncertainties or parameter correlations. In complex blended spectra, limited uncertainty quantification can make it more difficult to assess whether a difference in, for example, central velocity between two components is physically meaningful or influenced by the fitting procedure. A flexible statistical framework for addressing these issues is provided by Bayesian inference implemented through Markov Chain Monte Carlo (MCMC) sampling. In this framework, the model parameters are treated as random variables, and the goal is to obtain their posterior probability distributions given the data and any prior information. MCMC methods have gained considerable traction in astronomy for tasks ranging from exoplanet orbit fitting to stellar population synthesis ( 32 ; 24 ; 2 ; 23 ; 1 ; 16 ) , but their application to maser spectral decomposition remains rare. A Bayesian approach offers several advantages: it enables direct estimation of parameter uncertainties in the form of posterior credible intervals, allows model complexity to be assessed via information criteria such as the Akaike Information Criterion (AIC) or the Bayesian Information Criterion (BIC), and permits the incorporation of physically motivated priors (e.g., that line widths must be positive). In the specific context of maser spectroscopy, Bayesian MCMC provides a statistically principled way to decompose a complex, blended spectrum into its constituent velocity components while simultaneously characterising the precision and covariance of the derived parameters. In this paper, we apply a Bayesian MCMC method to the 6.7 GHz methanol maser spectrum of G339.884 - 1.259, using data obtained with the 32-m radio telescope of the Ghana Radio Astronomy Observatory (GRAO). The observations were conducted in a standard position-switching mode, and the spectrum was extracted after subtraction of an off-source (blank sky) measurement. We model the spectrum as a sum of Voigt profiles, each described by an amplitude, a central velocity, a Gaussian width ( \sigma ) associated with turbulent broadening, and a Lorentzian width ( \gamma ) associated with damping. The number of components