Performance of VERA in 10 micro-arcsecond astrometry

Abstract

Abstract Very Long Baseline Interferometry (VLBI) astrometry using the phase-referencing technique remains an open issue for the quantitative characterization of the observing conditions to achieve a feasible parallax precision of 10 micro-arcseconds (μas). To address this issue, we evaluated the astrometric performance of the VLBI Exploration of Radio Astrometry (VERA) through the parallax measurements of five distant star-forming regions under good observing conditions of close separations (${0{^{\circ}_{.}}5}$–${1{_{.}^{\circ}}3}$) and high elevations (≥50°). Their parallaxes measured 89–200 μas, corresponding to distances of 5–11 kpc with an error of 11–20 μas. Furthermore, we investigated the contributions to the position error budget and concluded that the tropospheric residual contribution is the dominant error source. We also confirmed that the astrometric error propagation strongly depends on the term $\Delta \sec Z$, which stands for the difference between $\sec Z$ of the target and its reference source, where Z is the zenith angle during the observations. We found that for a source pair with a $\Delta \sec Z$ less than 0.01 (for example, a set of a close separation of $\le {{0{^{\circ}_{.}}5}}$ and a high elevation of ≥50°), we can achieve the parallax precision of 10 μas using a typical monitoring program comprising 10 observing epochs over a span of two years.