Variations of absolute source positions determined from quad-band VLBI observations

TL;DR

This paper examines the accuracy and stability of AGN positions measured by quad-band VLBI observations, identifying key factors affecting positional variations and emphasizing the need for a dedicated quad-band catalog.

Contribution

It provides a detailed analysis of source position variations in quad-band VLBI data and highlights the importance of a specialized catalog for improved astrometric precision.

Findings

Source structure is the main factor for sources above 20° declination.

Position offsets are significant for 15% of sources compared to 2.2/8.6 GHz.

Visible and invisible source structures impact position measurements and offsets.

Abstract

Active Galactic Nuclei (AGNs) observed with the technique of very long baseline interferometry (VLBI) are used as fiducial references on the sky to precisely measure the shape and orientation of the Earth. Their positions form a celestial reference frame that plays an important role in both astronomy and geodesy. This study investigates the accuracy and stability of the positions of the AGNs that are measured by simultaneous VLBI observations at 3.3, 5.5, 6.6, and 10.5 GHz. Based on position time series from dedicated geodetic solutions, we characterize the observed source position variations and identify the possible factors causing such variations. We find that the primary contributor is source structure for sources above 20-degree declination while the sensitivity of the observations to the declination coordinate predominates for sources below 20-degree declination. The position time series are further explored to derive more realistic uncertainties for the quad-band positions. Significant position offsets with respect to the positions at 2.2/8.6 GHz are found for 15% of the sources. For 6% of the sources, the offsets are larger than 0.8 milli-arcseconds. Source structure may be divided into two parts: the invisible structure (within the beam size) and the visible structure (on larger scales). The latter causes closure delays enlarging post-fit delay residuals in geodetic solutions whereas the former causes source position changes. Such position changes will contribute significantly to the offsets between radio and optical positions. Overall, this work highlights the necessity to have a specific quad-band catalog for processing operational quad-band observations.