Lodén 1, Part II: Proving a Negative

Last time, I described how Jason Curtis proved that the long-forgotten cluster Ruprecht 147 was actually an important benchmark in stellar astrophysics.

We wanted to know if two other entries in WEBDA — Lodén 1 and NGC 2240 — might be similarly useful.

We were pretty skeptical.  Neither “cluster” has a proper motion that distinguishes it from the field.  This means that there is no indication that the stars in those overdensities on the sky are moving as a group—the only reason to think that there might be a cluster there is that there seem to be more bright stars in the region than one would expect by chance.  But it’s a big sky—chance has lots of opportunities to fool you.

Lodén 1 was first noticed by L. O. Lodén in 1980, in his search for “hidden” clusters in the southern galactic plane.  Lodén noted an “evident concentration of late-type stars…and main sequence A–F stars” whose physical association was “not confirmed by strongly suspected.”  Kharchenko’s 2005 analysis put the cluster at 175 pc and 2.5 billion years old— very useful if true, but somewhat inconsistent with Lodén’s note that the cluster contains A stars (though they could be blue stragglers, of course).  A later paper by the Kharchenko group put the cluster at 200 million years old and over 750 pc away, further confusing the issue.  We suspected that the reason the two fits came out so differently was that the Kharchenko algorithm was trying to fit field stars, and so had a GIGO problem.

Part of a figure from Eunkyu and Jason Curtis's paper

Part of a figure from Eunkyu and Jason Curtis’s paper showing color-magnitude diagrams for Lodén’s “field 1” and a control field at the same Galactic latitude.

Jason Curtis did a lot of work putting isochrones down on color-magnitude diagrams for stars in the field.  The above figure shows the Kharchenko et al. 2005 isochrones for the “cluster” in gray, and in red the same isochrone shifted out to 500 pc.  The left panel shows Lodén’s field 1 (the location of the purported cluster), and the right shows a nearby control field at the same Galactic latitude.

The right panel shows “field stars”— the colors (x-axis) and brightness (y-axis) of the random stars along this line of sight in the Galactic plane.  The red (high J-K) stars are mostly intrinsically bright, background K giants at a variety of distances (farther away makes you fainter, moving you down in the graph).  The blue stars (J-K near 0) are mostly hot, young stars randomly strewn along the line of sight.  There is no clustering along either isochrone line because there is no coeval cluster of stars in this region—if there were, you would expect to see an overdensity of points along those or some similar isochrone.

The left hand panel shows the Lodén 1 field, and you’d be hard pressed to tell it was any different from the control.  True, there are a few more bright stars: not really any more red giants, but there are 7–10 blue stars with J < 9.25 or so, while the control field only has 3.  Those are presumably the blue stars that caught Lodén’s attention in the first place.

For reference, here’s how Ruprecht 147 looks:

Color-magnitude diagrams for Ruprecht 147 and a control field.

Color-magnitude diagrams for Ruprecht 147 and a control field.

Its control field looks very different (it’s farther from the plane, so the field stars are typically more distant) but the cluster field has the unmistakable presence of a cluster: lots of stars on the Main Sequence, a healthy but smaller red giant branch population, and even a few blue stragglers (the bright blue stars off of the red line).

If you squint, you can try to make the Lodén 1 stars fall on the red line there, but the numbers are all wrong: given the number of giants, there are way too few Main Sequence stars for this to be real.  The control field actually has a more convincing coeval sequence.

So there doesn’t appear to be anything obvious here.  But that doesn’t mean that Lodén was wrong, just that the cluster doesn’t stand out very well from the field.  To really be sure, we need to see if there are any comoving stars in the field.  We can use catalog data to see in which direction on the sky the stars are moving, and see if there is a subset of them all going in the same direction.  Ruprecht 147 stands out nicely by this metric:

Proper motion diagram for Ruprecht 147

Proper motion diagrams for Ruprecht 147 field and control. Blue cross gives representative error bars.

Here the motions in the north and east directions are plotted ((0,0) is no motion) for all of the stars in the Ruprecht 147 field (left) and its control field (right).  The red circle marks the position in proper-motion space of the cluster.  The control field reveals that field stars generally have zero proper motion — this is because they are typically very distant, so they seem to move very slowly across the sky. The more nearby field stars have large proper motion (they are farther from (0,0)), but there aren’t very many of those, so there’s not much contamination in the red circle.  The cluster itself has a bunch of stars moving southward at a bit over 20 milliarcseconds per year; the overdensity is obvious in the figure above.

How about Lodén 1?

Lodén 1 proper motion diagram

Proper motion diagrams for Lodén 1 (left) and control field (right)

Yikes.  There is no obvious overdensity.  If this cluster is real, its is not moving significantly through the plane.

At the referee’s request, Jason Curtis did some heriocs trying to see if there was an overdensity buried in there, using the control field as a guide.  Noting convincing came out.

But these things are important if they’re real, and we should give some deference to Lodén’s intuition on this.  The real test would be to measure the two missing components to these stars’ motions and positions: their radial velocities and distances.  The distances will have to wait for Gaia, but the radial velocities we could do ourselves.

Next time: Eunkyu collects radial velocities, and pronounces her verdict.

Leave a Reply

Your email address will not be published. Required fields are marked *