https://njuastrometry.github.io/en/author/hua-feng-yu/Hua-Feng YuHugo Blox Builder (https://hugoblox.com)en-USWed, 01 Apr 2026 00:00:00 +0000https://njuastrometry.github.io/en/author/hua-feng-yu/avatar_hu17376295067228175092.jpghttps://njuastrometry.github.io/en/author/hua-feng-yu/https://njuastrometry.github.io/en/publication/2026-aa-708-a83/Wed, 01 Apr 2026 00:00:00 +0000https://njuastrometry.github.io/en/publication/2026-aa-708-a83/<p><strong>Sample and analysis</strong></p> <p>Starting from 21,949 compact radio sources in RFC release <code>rfc_2025c</code>, we compile redshifts for 9907 objects using DESI DR1, SDSS DR17/DR19, and literature data. After spectroscopic quality checks and cross-matching with Gaia DR3, the final sample contains 4068 sources, including 934 ICRF3 sources. Spectral classifications identify 784 galaxies, 3151 quasi-stellar objects (QSOs), and 133 sources with stellar or inconsistent classifications.</p> <p><strong>Main results</strong></p> <ul> <li>VLBI positional uncertainties, radio structure indices, and compactness show no significant redshift dependence within the current precision.</li> <li>Median absolute radio-optical offsets decrease markedly over $0&lt;z&lt;0.5$, decline more gradually over $0.5&lt;z&lt;1.3$, and show a mild increase at $z&gt;1.3$.</li> <li>The optical trends largely reflect Gaia&rsquo;s magnitude-dependent astrometric performance and the changing mixture of galaxies and QSOs. Galaxies have larger optical astrometric uncertainties even at a fixed $G$ magnitude, indicating an additional contribution from extended morphology.</li> <li>The nearly constant median error-normalised proper motion of approximately 1.2 at $z&gt;0.5$ is consistent with the observed proper motions being dominated by random measurement uncertainties.</li> </ul> <p>These results help distinguish observational and source-population effects from an intrinsic dependence on redshift when evaluating sources for the optical–radio reference frame connection.</p>https://njuastrometry.github.io/en/publication/2025-aa-704-a136/Mon, 01 Dec 2025 00:00:00 +0000https://njuastrometry.github.io/en/publication/2025-aa-704-a136/<h2 id="an-infrared-extension-of-the-celestial-reference-frame">An infrared extension of the celestial reference frame</h2> <p>The International Celestial Reference System is currently realized primarily in the radio and optical bands. An infrared realization would extend its wavelength coverage and provide a denser reference grid in regions affected by strong interstellar extinction, such as the Galactic plane and bulge.</p> <p>In this study, we cross-matched CatWISE2020 with Gaia DR3, identifying approximately 404 million common sources, including 1.38 million Gaia-CRF3 quasars. We investigated their positional and proper-motion differences using vector spherical harmonics, with separate analyses for stellar and extragalactic sources.</p> <h2 id="main-results">Main results</h2> <ul> <li><strong>Large-scale systematic offsets:</strong> For stellar sources, the rotation and glide amplitudes relative to Gaia DR3 are approximately 6.6 and 6.9 mas in position, and 1.9 and $2.7,\mathrm{mas,yr^{-1}}$ in proper motion.</li> <li><strong>Internal frame differences:</strong> The stellar and extragalactic reference frames within CatWISE2020 differ by approximately 6 mas in position and $1,\mathrm{mas,yr^{-1}}$ in proper motion.</li> <li><strong>Magnitude-dependent systematics:</strong> The offsets vary significantly with magnitude, indicating that a single global correction cannot fully describe the systematic differences. No significant color dependence was found.</li> <li><strong>Underestimated uncertainties:</strong> The reported positional uncertainties are underestimated by factors of approximately 1.5–2.5.</li> </ul> <p>The featured figure illustrates the spatial distribution of the positional and proper-motion offsets. Averaging the offsets within 192 equal-area sky pixels reveals coherent patterns, demonstrating the need to account for large-scale systematics when linking CatWISE2020 to Gaia.</p> <h2 id="implications">Implications</h2> <p>After correcting systematic offsets relative to Gaia, CatWISE2020 can serve as an infrared realization of the ICRS. Its deeper coverage and improved proper motions offer substantial advantages over AllWISE. However, global corrections do not remove all tile-dependent systematics, and the comparatively large proper-motion uncertainties limit the long-term stability of the frame. Further improvements will require refined calibration and, ultimately, higher-precision infrared astrometry.</p>