by Kevin Krisciunas, Cerro Tololo Inter-American Observatory
From a 9 night run in late September/early October 1997 with the USNO 1-m telescope I devised a set of 17 secondary star values. These were derived from the instrumental magnitudes and the nightly extinctions, and reduced to BD +17 4708 from observations on 4 photometric nights when this primary standard was observed. Two other Landolt stars (SA 110-441 and -450) were observed on two nights when BD +17 4708 was not observed; enough of the secondary stars were observed that SA 110-441 and -450 could be tied to BD +17 4708. Another star, BD +21 607, intended to become a primary standard, was also observed on three nights, one of which when BD +17 4708 was observed. So I have put 20 stars on the AB magnitude system, calibrated from spectrophotometry of the primary star BD +17 4708.
From 58 sets of 5-filter observations of 17 secondary stars observed on photometric nights when the data could be tied directly to BD +17 4708 we obtain the transformations given below. (Two stars, Ross 374 and BD -21 910, had no R-I color, so had to be left out of some of the regressions.) Of course, I could obtain slightly different regressions by weighting the points by the different numbers of observations, but for now here are the equally weighted solutions. Since these transformations are based on a relatively small number of stars, and many more stars have been observed since I observed these in Sept/Oct of 1997, the parameters on the left hand sides could be given a subscript of "KK" to indicate that they are not official Sloan transformations.
RMS residual of fit N
(1) r' = V - 0.4222 (B-V) + 0.0965 +/- 0.0269 17
(0.0146) (0.0120)
(2) (u'-g') = 1.3290 (U-B) + 1.1786 0.0632 17
(0.0239) (0.0179)
(3) (g'-r') = 0.9578 (B-V) - 0.1803 0.0299 17
(0.0163) (0.0134)
(4) (r'-i') = 0.9950 (R-I) - 0.2157 0.0136 15
(0.0147) (0.0067)
(5) (r'-z') = 1.6660 (R-I) - 0.4049 0.0255 15
(0.0276) (0.0115)
(6) (i'-z') = 0.6710 (R-I) - 0.1893 0.0232 15
(0.0251) (0.0113)
(7) V = r' + 0.4425 (g'-r') - 0.0178 0.0156 17
(0.0088) (0.0057)
Above I simply took (r'-z') = (r'-i') + (i'-z').
Rounded off versions of Eqs. 1, 2, 3, 4, and 6 are to be found in our
paper in the November 1998 issue of Publications of the Astronomical
Society of the Pacific, vol. 110, p. 1342.
Update (November 2001)... John Tonry has taken the data in Table 1 of our
1998 PASP paper and derived the following transformations:
U = 1.002*u' + -0.939 rms = 0.029 U = 1.000*u' + 0.008*g' + -1.015 rms = 0.029 B = 0.266*u' + 0.767*g' + -0.463 rms = 0.049 B = 1.487*g' + -0.490*r' + 0.201 rms = 0.026 V = 0.467*g' + 0.545*r' + -0.178 rms = 0.010 R = -0.098*g' + 1.105*r' + -0.225 rms = 0.009 R = 0.830*r' + 0.180*i' + -0.286 rms = 0.010 I = 1.044*i' + -0.927 rms = 0.026 I = -0.173*r' + 1.180*i' + -0.469 rms = 0.018 I = 0.770*i' + 0.240*z' + -0.521 rms = 0.018 Z = 0.273*i' + 0.745*z' + -0.680 rms = 0.024 However, see the comments immediately below about a larger database of Sloan magnitudes and a systematic error in our (u'-g') colors from Eq. 2 above.There are magnitudes and/or colors of 164 Sloan standards given here and here. A preprint of 10 January 2002 on the SDSS standard system is now available by clicking here.
From a comparison of my values for stars in common, it seems that my griz magnitudes or colors are accurate at the 0.015 mag level, but looking at the (u'-g') reveals a systematic error that is a function of color. I find the following subsequent transformation is necessary:
(u'-g')_KK = 1.0264 (u'-g')_SDSS + 0.0208 (RMS resid = +/- 0.030 mag) or (u'-g')_SDSS = 0.9743 (u'-g')_KK - 0.0203Email me by clicking here .
Go back to Kevin Krisciunas home page by clicking here .
This document last revised on 21 January 2002.