Doing Science at FIRO

[a picture w/ caption could be placed here]

FIRO Specializes in Speckle Interferometry 

The twinkling of starlight is caused by hundreds of thousands of small air pockets that lie between the ground level and the top of the atmosphere. Starlight coming to earth from space as it travels in the vacuum of space is traveling in a perfect spherical wavefront, what is known as an isoplanatic wavefront. But upon striking the atmosphere, those little turbulence pockets start refracting the starlight every which direction. By the time the light gets to the ground, it has been bent and deflected thousands of times and meaningful information in the isoplanatic wavefront seems to be irretrievably lost.

But not so. We can recover that information if we can regain the isoplanatic wave state of the light. That is where speckle interferometry enters the picture. Figure 1 shows a picture of speckles taken at Kitt Peak Observatory in 2014:

You can see several “images” of the double star splattered across the image. Each of these is an image that passes through one path of turbulence cells for a very short amount of time.
Speckle is most effective when both of the stars are within the isoplanatic patch (explain patch) which is about 5 arcseconds, and bright enough to image both stars in 40 milliseconds or so, depending on the seeing.  But will need to describe quasi speckle, because it is important.   Under these conditions, we have learned that we have a very good chance of “freezing” these speckles in such a way that important information can be gathered from the light. 
On a typical night at any Observatory, the seeing is usually so poor that most stars produce a fuzzy blob of light that is 2 arcseconds in diameter. May need to distinguish between mountaintop and lowland observatories like FIRO.  In double star astronomy, the closer the stars are together, the shorter will be the orbital period, and the more likely it will be that we can derive an orbit. But if the pair is three arcseconds apart and the scene produces two-arcsecond stars, it is virtually impossible to extract useful information From the light. 


Figure 2 shows what a typical close double star might look like on a night of marginal seeing. We can tell that there are two stars there, but exactly how far apart are they? And how do we accurately measure the angle between them relative to true north?
Speckle has the ability to recover the isoplanatic wavefront information in such a way that a given telescope on a night of at least modest seeing will achieve its diffraction limit in resolution, allowing astronomers to take a figure 2 star and clearly separate each star into a separate image that allows for precise measurement. See Figure 3, which is the same pair as seen in Figure 2 but with the isoplanatic information recovered.

Figure 1: Speckles obtained at Kitt Peak, 2014.

Figure 2: A typical "excellent" image of a close double star as seen in an eyepiece with the eye.

Figure 3: Well resolved pair at nearly the diffraction limit of the telescope.

Choose Your Target Carefully

The Holy Grail too informal of double star astronomy is the production of an orbit for a pair of stars. If we know the orbit, we can deduce the total system mass using Kepler's laws of motion. And if we know the distance to the system, we can determine the individual masses of each star and their luminosities. The relationship between a star’s mass and its luminosity is critical. In the 1920s, two astronomers , independent of each other, came up with the same relationship. The mass/luminosity diagrams were produced by Ejnar Hertzsprung (a Danish astronomer) and Henry Russell (an American astronomer), so to be fair, astronomers gave each scientist equal credit for this critical discovery and now call their diagram the H-R Diagram (for “Hertzsprung-Russell”). The H-R Diagram (see Figure 1) is the backbone of all modern astronomy and theories of stellar development, maturation, and demise. These stellar theories in turn drive galactic formation models, so in a very real sense the H-R Diagram is the lynchpin for most of modern galactic astronomy.
Figure 4: The GAIA DR2 H-R Diagram, a breakthrough of high scientific importance.

Therefore, we strongly encourage you to select stars to study that either already have demonstrated short arcs or that have a very low GB index. Such research will do more to further our knowledge of double stars and stellar physics then measuring random pairs. 
Over the years, we have seen many teams of researchers (most typically high school teams with little astronomical background) make target selections based on three main criteria:  (1) position in the sky (will the star be visible at a time when we need to observe it in order to fulfill our course time restraints?); (2) separation (is the pair under 5 arc seconds apart), and (3) a magnitude difference of 3 or less with the companion being brighter than 12.5 magnitude. Student teams often query the WDS (Washington Double Star Catalog) with these criteria in mind and often get hundreds of hits, from which they will select a star (or a few). The problem is, about 89% of the systems in the WDS are what could loosely be termed “common proper motion” pairs (CPM) and have shown little relative motion since discovery a century or two ago. CMPs or just optical doubles?  So a random search of the WDS based on criteria such as these will probably return targets that are of little interest to astronomers at this time.  I would skip the above paragraph.  No need to talk about what high school students do with astrometric eyepeices.
The hot zone informal for double star astronomy is to identify and measure stars which are starting to show relative motion, and that this relative motion seems to fit an arc. This is important because an arc is a part of an ellipse, and an ellipse is the projected orbit of a double star against the plane of the night sky.  Nice.
So how do you find “short arc binaries” (SABs)? Before GDS1 was available, all one could do was scan the WDS for stars with long histories of measurement but lacking in a recent measurement, then requesting the data on that pair and plotting the measurements using a program like Excel. As often as not, the results will not be promising. 
But GDS1 changes all of that. Drawing on the vast Gaia Data Release 2 (DR2), Dave Rowe, the author of the program, has found a way to generate what is called the “gravitational binding index ”. The  GB index is a relative measure of how likely the two stars are to be gravitationally involved with each other. The lower the index, the higher the likelihood they are gravitationally involved. 
That is not to say that the stars are in orbit around each other. It is to say, however, that they could very well be traveling throughout the Galaxy together, having experienced a common origin In a star cluster in the distant past and then been ejected from that cluster as the cluster passes through the density waves of the galaxy's spiral arms. The stars may be too far from each other to orbit their common center of gravity (barycenter) but they may be surfing informal the Galaxy as distant companions. 
Therefore, we suggest that you use GDS1 (described below) And concentrate on stars with a low GB index.  “We suggest that you use” is sort of me teacher you student, me experienced researcher, you novice.  Our researchers are professionals.  We just provide them with a facility to use and instructions on how to use it.  Of course we can describe GDS and its uses.  Just not “we suggest that you”  
Another important area of research for double stars is the ability to recognize when a given pair is not gravitationally bound in any way but are simply two stars passing each other in the Galaxy with no orbit or physical involvement. Such pairs are called “optical pairs” and are of little interest to astronomers since we cannot glean any useful mass and luminosity data from them. Once a pair has been classified as a “linear system” by the curators of the WDS, most astronomers going forward will tend to leave those stars off of their target lists since telescope time is a very precious commodity to research astronomers as they do not want to waste their time on a pair that is not likely to yield mass and luminosity data. 


Figure 4: The GAIA DR2 H-R Diagram, a breakthrough of high scientific importance.

A Question to Consider:  Speckle or Cv?  ​

Not every star returned by a query of GDS1 will qualify for speckle interferometry. Some of the returns will be farther apart than 5 arcseconds, and some will be so faint that exposures of 0.25” or longer may be required to get the stars to register on the CCD. Yet over half the stars that GDS1 can output fall into this “too wide / too faint” group. But there is still useful science that can be done with them.

The WDS has method codes in the measurements it reports, and one that caught the eye of FIRO staff astronomer Richard Harshaw was Sv. It turns out that Sv means that the US Naval Observatory measured the star with the 26-inch Clark Refractor, but it was too wide (or faint) for speckle. However, the USNO team discovered that speckle reduction software did a superb job of helping with the measurements.

Since Harshaw was doing the same thing with his CMOS camera and 11-inch telescope, he corresponded with Dr. Brian Mason, chief administrator of the WDS, and they concluded that the method Harshaw was using was essentially the same as that used by the USNO (which uses different speckle reduction software than The Speckle Toolbox). So Mason agreed to create a new observing code for amateur class measurements using speckle reduction software to measure CCD images and decided to give it the code Cv. The Cv code means that an astronomer used a small instrument to record a double star that was wider than 5 arc seconds and so faint it required more than 40 milliseconds to register.  This is really good and important stuff, but a suggestion.  I don’t think Brian made an observing code for amateurs, just for smaller telescopes.  I hope we can avoid the word “amateur” on the entire website, and also mostly avoid the word “student.”  As I recall, Sv was for USNO quasi speckle, while Cv was a non-USNO quasi speckle. 

Many of these “Cv” types of stars are already showing short arcs, and many more will start to show them as more and more measurements accumulate. So don’t rule out pursing wide and very faint pairs as part of your research program. This is a wide-open field that is ripe for harvesting.

© 2018 by Russell M. Genet and The Fairborn Institute