1	Bye-bye to UBVRI? Mark Kidger
2	Just what DOES an astronomer do? All observational astronomy is basically a matter of recording photons, be they visible, infrared, gamma rays, or radio. It all reduces to one of two things: Seeing how the photons are distributed over a two-dimensional space (taking an image), or... Measuring how many photons arrive per unit of time (taking photometry). It does not matter if your detector is your eye, a CCD, or a heterodyne receiver; you still are counting photons.
3	"The concept of astronomical photometry" The concept of astronomical photometry dates from Ptolemy in the 2nd Century A.D. Ptolemy divided the stars into 6 degrees of magnitude from the brightest – magnitude 1 – to the faintest visible with the naked eye – magnitude 6. It is Ptolemy that we must thank/blame for introducing photometry and the concept of photometric calibration.
4	"All the systems of photometry..." All the systems of photometry that exist today are based on Ptolemy’s original system. Remarkably, with just a few exceptions, Ptolemy’s magnitudes agree well with modern photometry (although to low precision). We now define the magnitude scale mathematically such that a difference of 5 magnitudes is a factor of 100 in brightness. Or, a difference of 1 magnitude is a factor of 10^0.4 in the “irradiance of the detector”.
5	"2000 years ago everything was..." 2000 years ago everything was far simpler. The only detector in use was the naked eye. The magnitude scale only had to deal with a range of less than three orders of magnitude in brightness – the difference between Sirius and the faintest star visible to the naked eye. Now the magnitude scale must cover »55 magnitudes (11 orders of magnitude in brightness) from the Sun – magnitude –26.7 – to the faintest stars detectable with the Hubble. And it has extended to other ranges than the visible.
6	First Steps By the 1950s it had become obvious that it was necessary to have both a list of stars apt for the calibration of photoelectric instruments and a common photometric system for astronomers to use. A series of defined photometric bands. Stars with a known magnitude in each band.
7	How the first system was defined... The basis of the modern “Johnson” system that is still the most common today (Johnson & Morgan, 1951, Astrophysical Journal, 114, 522) is: A photometric criterion such that, measured OUTSIDE the atmosphere... For a normal A0V star: (B-V)=(U-B)=0.0 For a giant K0 (K0III) star: (B-V)=(U-B)=+1.0 The spectral types were defined in the revised Yerkes system (Morgan, Keenan & Kellman, 1943, "An Atlas of Stellar Spectra with an Outline of Spectral Classification", Astrophysics Monographs, University of Chicago Press).
8	What was defined... Three standard photometric bands: U (ultraviolet), B (blue), & Y (yellow-green, now “V”). These were the basis of modern broad-band UBV filters. The UBV as defined above was refined by Johnson & Harris (1954, Ap.J., 120, 196) and by Johnson (1955, Ann. Astrophysics, 18, 292), becoming the standard system for all visible photoelectric photometry and, by extension, all spectrophotometry.
9	Bringing it up to date... In parallel, other workers defined the red bands: First defined in 1951, principally due to the work of Kron. A new R filter was defined in 1951 by Lynn Smith (1951, PASP, 63, 91). Johnson attempted to define an alternative system in 1966, but it was little accepted. The current system was defined by Cousins (1976, MNRAS, 81, 25). This gave rise to the current Kron-Cousins system in RI. The system that all astronomers use, amateur, professional, visual, or CCD is thus... “Johnson-Kron-Cousins”. This system was the result of more than 30 years of work.
10	But things start to get complicated After 50 years, UBVRI is showing its age. It was designed for photoelectric photometers, which ceased to be used some 20 years ago. U has fallen into disuse. I comes in almost as many varieties as ice cream. The UBVRI system was ideal for observing galaxies but, is far from ideal for photometry of stars. The bands are too wide for separating the main spectral lines. Makes measuring the physical properties of stars extremely difficult. Many different alternative systems were thus defined (Strömgren, narrow band, Gunn, etc). Very confusing. Difficult to compare data. And…
11	Worse Still... There is a calibration crisis. There are few well-calibrated stars, mainly around the equator. The faintest well-calibrated stars in UBVRI are around magnitude 15, but astronomers are regularly observing objects that are magnitude 25 and fainter. A 10-m telescope will saturate in 1s at magnitude 19!!! Even the “ultimate” photometric catalogue – Tycho – has major problems Like Claudia Schiffer, it looks good on the outside. But, the discerning astronomer really needs something that goes a lot deeper. It comes with a lot of baggage and style issues. What looks good, may not be apt for normal, daily use.
12	New Surveys. New System. A series of new, large sky surveys are coming on line, or are imminent: Sloan Digital Sky Survey (SDSS) – in the image. Panoramic Survey Telescope & Rapid Response System (Pan-STARRS) The Large Synoptic Survey Telescope (LSST) None of these use the traditional BVRI filters. Observatories are not replacing their old UBVRI filter sets. Little or no new work is being done to support UBVRI despite its known inadequacies.
13	Meet ugriz! Increasingly the new standard system is the Sloan Digital Sky Survey filter system… ugriz. The ugriz system is optimised for stellar photometry. It is being used increasing widely and is the new “buzz word” in photometry. UBVRI is likely to be extinct in professional astronomy in 10 years or less.
14
15
16	Other Advantages of ugriz H-alpha falls in “r” and H-beta and H-gamma both fall in “g”. The “u” filter is broader than Johnson U. The two main metal lines in stellar spectra (from calcium) fall inside “u”. So, photometry in ugr allows the metal content of stars to be determined with great precision. ugriz photometry allows the physical parameters of stars to be determined without spectroscopy (from CCD images in large surveys).
17	The bad news for amateurs BVRI filters are simple to mass produce (coloured glass) and thus cheap – a reasonable set can be bought for a few hundred dollars. Sloan ugriz filters are special, interference filters and made to order. They are not catalogue items. A typical price is $3000-$10000 per filter. Smaller professional observatories cannot afford them. They are totally out of reach of 99.99% of amateurs at present. They have a limited lifetime and need replacing.
18	The bad news for astronomy There is a real danger of creating a two-speed structure in astronomy: Big observatories carrying out standardised photometry using the new all-sky photometric catalogues in ugriz and producing the faint all-sky calibration stars that CCDs need. Small observatories and amateurs marginalised by using an antiquated system that cannot easily be homologated with new professional photometry.
19	The way forward? One solution is to persuade filter manufacturers to offer ugriz filters as catalogue items. By mass producing them, on the grounds that these will be the standard filters that everyone will need to buy in the future, the cost could be greatly reduced. IAU Commission 25 is charged with photometry and calibration. The progressive change to ugriz seems to be a “fait accompli”. But, Commission 25 will hopefully attempt to negotiate with manufacturers to persuade them to produce cheaper ugriz. Astronomers will need to get used to the idea that UBVRI is on its way out.