Updated Radio Sigma-D Relation for Galactic Supernova Remnants

You are here: Home / Submitted Papers / 2015 / Updated Radio Sigma-D Relation for Galactic Supernova Remnants

Abstract

We present the updated empirical radio surface-brightness-to-diameter (Sigma – D) relation for supernova remnants (SNRs) in our Galaxy. Our original calibration sample of Galactic SNRs with independently determined distances (Pavlovic et al. 2013, hereafter Paper I) is reconsidered and updated with data which became available in the past two years. The orthogonal fitting procedure and probability-density-function-based (PDF) method are applied to the calibration sample in the log Sigma – log D plane. Non-standard orthogonal regression keeps the Sigma-D and D-Sigma relations invariant within estimated uncertainties. Our previous Monte Carlo simulations verified that the slopes of the empirical Sigma-D relation should be determined by using the orthogonal regression, because of its good performances for data sets with severe scatter. The updated calibration sample contains 65 shell SNRs. 6 new Galactic SNRs are added to the sample from Paper I, one is omitted and distances are changed for 10 SNRs. The slope derived is here slightly steeper (β ≈ 5.2) than the Sigma-D slope in Paper I (β ≈ 4.8). The PDF method relies on data points density maps which can provide more reliable calibrations that preserve more information contained in the calibration sample. We estimate distances to five new faint Galactic SNRs discovered for the first time by Canadian Galactic Plane Survey, and obtained distances of 2.3, 4.0, 1.3, 2.9 and 4.7 kiloparsecs for G108.5+11.0, G128.5+2.6, G149.5+3.2, G150.8+3.8 and G160.1-1.1, respectively. The updated empirical relation is used to estimate distances of 160 shell Galactic SNRs and new results change their distance scales up to 15 per cent, compared to the results from Paper I. The PDF calculation can provide even few times higher or lower values in comparison with the orthogonal fit, as it uses a totally different approach. However, on average, this difference is 32, 24 and 18 per cent for mode, median and mean distances.

Author

Pavlovic, M. Z.; Dobardzic, A.; Vukotic, B.; Urosevic, D.

Journal

Serbian Astronomical Journal

Paper Publication Date

December 2014

Paper Type

Astrostatistics