The celestial object NGC 2451 is a loose grouping of fairly bright stars in the southern Milky Way which was first described as a star cluster by John Herschel in 1835. In 1888, Johan Ludvig Emil Dreyer entered it in his famous sky inventory of star clusters and nebulae, the New General Catalogue (NGC), under the running number 2451.

Towards the end of the 20th century, doubts arose whether this group of stars really constitutes a cluster, i.e. a gravitationally bound collection of sibling stars of the same origin and age. These doubts were prompted by measurements of the brightness and colours of the putative members which did not quite match. In 1994, S. Röser and U. Bastian used their newly compiled Positions and Proper Motions Catalogue (PPM) to decide this question: for the PPM they had combined more than 100 years of position measurements to compute precise proper motions for almost half a million stars on the whole sky. As a little science application they checked the motions of the putative cluster members. As a cluster, those would have to move at the same speed and direction, to within the uncertainty of the measured proper motions. But they did not. Instead, in a diagram plotting the east-west motion versus the north-south motion the putative members were scattered all over the place instead of being concentrated in a tight group. The cluster was thus finally proven to be non-existent.

The plot which Röser and Bastian used to kill NGC 2451 is shown at the top left in Image 1. In its outskirts it shows some nearby stars having large apparent motions, and towards its centre the image shows an accumulation of farther-away stars which have small apparent motions - due to their large distances. Looking more closely at the image, one might imagine a barely discernible density enhancement to the top left of the central clumping. To guide the eye, it is marked by a circle in the middle panel of Image 1. Could this be a hitherto unrecognised star cluster happening to lie in the direction (and behind) the non-existent NGC 2451? This clearly cannot be decided from just the few scattered points in the diagram.

Röser and Bastian scanned through the astronomical literature to find precise colours and magnitudes of the stars belonging to the slight density enhancement. They plotted a colour-magnitude diagram (CMD) from the data that they found, and with just three exceptions the stars nicely fitted to the assumption that all of them were at the same distance of 220 parsec (700 lightyears). A real star cluster had apparently been discovered. But there were even more surprises to come: the CMD of all stars in the sky region of NGC 2451 indicated that there was another - completely unrelated - new star cluster far "behind" the one discovered from the motion diagram. It was estimated to be at about 400 parsec (1300 lightyears) from the Sun. 

The publication of this result prompted other astronomers to make additional and more precise magnitude and colour measurements in that area of the sky, and within just a few years the existence of both star clusters was general astronomical consensus. The names of NGC 2451A and NGC 2451B became customary for them.

Enter ESA's Gaia astrometry satellite, almost 20 years later. During the last few months the astronomers who are busy with the processing of the Gaia data have started to make first rough calibrations and solutions. Very preliminary trial positions, parallaxes and motions of 2 million stars were produced. Here, "trial" means computing test results, checking, improving, re-doing, and checking again. Star clusters can be helpful in this process. One of last July's trial solutions was used to produce a plot similar to the plot that Röser and Bastian produced back in 1994. It is shown (on a slightly different scale) in the bottom right panel of Image 1 where cluster NGC 2451A stands out prominently. So much more precise are the Gaia measurements of just the first 10 months of the mission compared to what more than 100 years of ground-based astronomy could achieve.

No CMD is needed anymore to confirm that NGC 2451A exists. But what about NGC 2451B, the second new cluster? Its apparent motion happens to lie right in the dense part of the galactic "background" clump in Image 1. It could thus not be discovered (nor confirmed) in the old 1994 motion data. It is also not visible in the Gaia panel in Image 1.

Enter Gaia's distance measurements. A look at the distribution of the Gaia distances (not shown) of all 1100 stars in the NGC 2451 area - among the 2 million in the whole trial solution - indicated two very slight bumps (maxima): one at about 190 parsec, and one at about 360 parsec. These agree with the supposed distances of the two clusters, within the uncertainties these distances have. Image 3 shows the same Gaia data as the right-hand panel of Image 1, but with some additional colour coding. Blue shows the stars with apparent motions consistent with membership in NGC 2451A. Magenta indicates the stars with Gaia parallaxes within the mentioned 360-parsec bump. Green are simply all other stars.

Credits: ESA/Gaia/DPAC/DPCE/DPCB/CU3/U. Bastian (ARI, Heidelberg)

Image 2: By Roberto Mura (Own work) [CC BY-SA 3.0], via Wikimedia Commons

[Published:18/12/2015]