Saturday, October 4, 2014

A New Star Catalog is Always a Cause for Celebration

A new star catalog is always a cause for celebration. Anyone who does more than dabble at astronomy visualization realizes that it is getting ahold of the data that is the sine qua non of serious work.

And yet getting the data is just the first of your many challenges. Very soon after downloading the data you will realize that one needs to have done some advanced study in astrophysics and/or have the services and advice of an astrophysicist or two in order to get anywhere. Without one or the other or both, the would-be visualizer will soon feel that he or she is lost in the forest on a moonless night.

But getting the data is the first necessary step, so it was with great pleasure that I read about the newly announced star survey with the catchy title of The second data release of the INT Photometric Hα Survey of the Northern Galactic Plane (IPHAS DR2)

And what is in this fabulous new survey?
The INT/WFC Photometric Hα Survey of the Northern Galactic Plane (IPHAS) is a 1800 deg2 imaging survey covering Galactic latitudes |b| < 5° and longitudes ℓ = 30°–215° in the r, i, and Hα filters using the Wide Field Camera (WFC) on the 2.5-m Isaac Newton Telescope (INT) in La Palma. We present the first quality-controlled and globally calibrated source catalogue derived from the survey, providing single-epoch photometry for 219 million unique sources across 92 per cent of the footprint. The observations were carried out between 2003 and 2012 at a median seeing of 1.1 arcsec (sampled at 0.33 arcsec pixel−1) and to a mean 5σ depth of 21.2 (r), 20.0 (i), and 20.3 (Hα) in the Vega magnitude system.

And so forth.

There are 219,000,000 unique sources in this survey. That is quite a bit more than the 10K star Yale Bright Star Catalog that we all grew up on. (1)

And yet there are certain similarities when you look a bit closer.   Certain subtle indications that suggest that the compilers of this fabulous expanse of data are adepts of the esoteric knowledge of astrophysics and that this data is intended for other members of this elite group.

In order to brace you for the tasks to come, should you decide to take on the challenge of visualizing this august data set, this post will present some first principles that are important for someone from the world of 3D as they step into the galaxy of creative decisions to be made and the arcane knowledge to master before the images latent in this esoteric compendium will be revealed.

I myself was first introduced to this knowledge during my brief but rewarding tenure on the Digital Galaxy Project portion of the rebuild of the Hayden Planetarium at the American Museum of Natural History in New York City.   Everyone on that project was a dedicated idealist with stars in their eyes.(2)

Because, you see, this dataset, like most astronomy datasets, is not intended to be for computer animation people. It is first and foremost to help scientists in the field, both those associated with the survey itself and those in the larger field of astrophysics, do their research. If people want to use it to “visualize the data” or make interesting pictures, that is all well and good, but that purpose is much further down the list of intended uses.

If we want to visualize the night sky what information would we ideally want? Well, we might want the 3D position of the stars, the brightness of the star, the color and presumably the type of star, as well as information about other objects in the sky that are not stars. This is a short list, there are actually many other things one might want to know for visualization purposes, but lets stop here. Ironically, a normal star survey or catalog does not generally have any of this information. At least not directly.

That is because what scientists are looking for in a star survey are the facts as they were observed. They may also be interested in your interpretations of the facts, but first and foremost they are interested in the data and how it was collected.

That means that instead of the 3D position of the “star” which we do not observe directly, except in a few exceptional cases, what the star catalog has is the position of the star as observed from earth on an imaginary 2D spherical coordinate system, a virtual sphere around the earth. In other words, the catalog does not say where the star is, but rather what direction it was observed to be from earth. The survey does not contain how bright the star was, or even if it was a star, but rather the spectra and intensity of the energy source as observed with the specific instrument. From that spectra, and from other information about the source of energy, one can deduce the type of star it probably is, but that is an interpretation of the data, not the data itself.

The star catalog or survey will not report the “color” as one normally thinks of such things, whether in RGB values, or CIE or other systems, but rather what was observed by the specific instrument or instruments used to collect the information. From the information in the catalog and the operating characteristics of the sensor one can derive spectra and from that a handy “color” in the visible bands if one so desires.

Nor is there any particular guarantee that the catalog is comprehensive, in some normal sense of the word. The authors of the catalog are not guaranteeing you that this survey contains all possible “stars” in the sky, or part of the sky. Instead what they are reporting here is the information about what they observed, and often some measure or estimate of how complete it is. The famous Messier catalog is not a complete list of anything, except the list of objects that Messier did not want to be bothered with anymore in his search for comets.

In a similar fashion, the data does not tell you how bright the star actually is, if it is a star, but tells you the intensity of the source as observed from earth. Again, an astrophysicist has various techniques to convert the apparent magnitude of the source to the absolute magnitude, if that is what you need.

The new star catalog has 219 million unique sources which may or may not be stars, and a variety of information about each, including some derived information (e.g. a probability that the specific source is indeed a star).

But it won't tell you anything about ISM (interstellar media), H2 regions and so forth.

When downloading the data, you can choose to download prepared datasets of the survey, or you can choose which data fields you want to review.  There are 99 different fields, and a dozen or so different formats you can get them in.  Just getting the data will probably take several days.

This post is the tip of the iceberg of what you need to know to do visualization from astronomy datasets but we will reserve these other topics for later posts.

The announcement can be found here

The database can be found here



1. The Yale Bright Star Catalog is a famous dataset of the 10,000 or so brightest stars visible from earth. These are the stars that one could typically see from earth on a clear night with the naked eye from somewhere on the planet. It is a famous catalog in part because it was widely distributed even in days before the Internet or the Arpanet.

2. There were many, many people on this project which was part of a much larger project to rebuild the Hayden Planetarium and the entire north side of the American Museum of Natural History. Many people thought we were from Mars.  The cast of characters included such luminaries as Dennis Davison, Neil deGrasse Tyson, Carter Emmart, Frank Summers, Aram Friedman, Loretta Skeddle, Julio Morano, Benjy Benjamin, Ron Drimmel and many others.  And of course, your humble author, was also briefly a consultant on this project. 

No comments:

Post a Comment