Astrometric positions of Eta and surrounding stars

This star's precise position is unusually hard to measure. Its HST image is slightly fuzzy because of the diffuse stellar wind. Worse, the wind may be asymmetric. And even worse than that, its shape may vary with time and it moves in a 5.5-year orbit. These effects can cause errors of the order of 0.01 arcsec even with a modern instrument such as GAIA (see below). Eta's position in pre-2000 star catalogs had errors exceeding 0.2", because ground-based measurements included asymmetric ejecta extending out to r > 1". Radio interferometry has the same problem.

Position using HST: In 2003 we measured the location of the central star in some Hubble Space Telescope images, and then calibrated those images' coordinate grids relative to many stars seen in ground-based CCD images and old photographic plates. The result, expressed in the ICRS (Hipparcos/Tycho) reference system, was:
RA = 10h 45m 03.563s, dec = -59° 41' 04.14",
with an expected uncertainty of roughly ±0.2 arcsec in each coordinate. This matched the Tycho Catalog position quite well, but Eta has a poor quality rating in the latter. We did not attempt to estimate proper motions, which are typically of the order of 0.002" per year at Eta's distance.

GAIA position: The second Gaia data release (May 2018) gives
RA = 10h 45m 03.5362s, dec = -59° 41' 04.053",
about 0.2 arcsec different from our 2003 estimate. The Gaia result is presumably superior, but its precision is uncertain. Given the systematic effects noted above, we do not entirely trust the formal uncertainty estimate which is smaller than 0.001". For similar reasons, Gaia has not produced a parallax for Eta Car. (Other evidence indicates a distance of 2300 pc or 7500 lightyears.)

Most HST images of Eta Car show two stars about 14" NE of Eta. They're 12th-magnitude B-type stars, and their GAIA DR2 positions are
10h 45m 04.7480s, -59° 40' 53.514", and
10h 45m 05.1250s, -59° 40' 57.204".
Their GAIA parallaxes indicate distances of roughly 3400 pc or 11,000 lightyears, farther from us than Eta is.

A list of stellar positions used for HST acquisition of Eta Car

In order to point at an object ("acquire the target"), the Hubble Space Telescope needs accurate positions of some stars around the object. In 1989, when we first planned to observe Eta Car with HST, the Guide Star Catalog was unsatisfactory in that crowded part of the Milky Way. Therefore we prepared "special catalog ZZZQ" of suitable acquisition and guide stars around Eta, using the Minnesota Automated Plate Scanner (APS) to measure about 2200 stars on photographic plates. We selected stars with good circular images, thereby excluding visual binaries, nebulosity, etc. ZZZQ was used to point HST at Eta Car dozens of times from 1991 to 2018, without a single failure.

Because the 1989 measurements used a different astrometric coordinate system, Eta's position in catalog ZZZQ is
RA = 10h 45m 03.643s, dec = -59° 41' 04.22" (J2000).
Most star positions in the list have uncertainties of 0.1 to 0.3 arcsec.

Partial list of ZZZQ stellar positions (ascii text file with explanations).

Automated Plate Scanner (APS) www site

It would be fairly easy to improve the accuracies in ZZZQ by using Gaia positions, but we have not had time to do so. Note also that proper motions were not necessary for HST purposes.

A strange historical possibility

Figure showing Eta's equatorial coordinates as functions of time (line drawing, B&W GIF file).

Taking precession into account, in ancient times this object was visible much farther north than it is today. If Eta had any major eruptions then like the protracted 19th-century outburst, it would have been noticed — and remembered — in ancient Egypt, Mesopotamia, and China. Unfortunately those cultures generally recorded planets' movements rather than stellar outbursts. Is it possible that any records of such an event might survive somewhere?

For references to this idea, see the end of section 6 in [].