Hi Gareth, as we promised it a few weeks back when Mario and I visited you, here is a summary of re-calibrated LINEAR photometry and a data file. Briefly, all of LINEAR's archived images were run through SExtractor by Scott Stuart from the LINEAR team. The resulting approximate astrometry was improved using astrometry.net, and the instrumental photometry was calibrated using SDSS data by Branimir Sesar at UW. Some preliminary analysis of the results is described below. We also matched the new photometry with the original LINEAR observations as they were reported to the MPC (see below for how to retrieve the resulting data file). The good news is that we now have a good understanding of what is the relationship between the original LINEAR magnitudes, L, and the standard Johnson V band magnitudes: L = V -0.65*[(B-V)-0.78] + 0.18, where B-V=0.78 is the median B-V color for main-belt asteroids. The overall median offset of 0.18 is easy to take into account, but for objects without measured colors, one cannot avoid about 0.13 mag large systematic offsets between the C type asteroids with the median B-V=0.65 and the S type asteroids at B-V=0.85. In addition, the original LINEAR magnitudes show a scatter (independent of color) of about 0.45 mag around that relation due to random calibration errors. The bad news is that we still do not fully understand the difference between the H mags taken from ASTORB file and the SDSS measurements. As per Parker et al. (2008, Icarus, 198, 138), about 2/3 of all H measurements listed in ASTORB file are systematically too bright by 0.33 mag (with an rms scatter of 0.28 mag relative to SDSS measurements; some of this scatter, could be due to rotational variability but not all). We hope that you will have further insight given the new recalibrated magnitudes, and that these new data may be useful for your dissertation! Please do not hesitate to ask for more details. We also have an early draft of a paper describing this calibration effort should you be interested in gory details. Cheers, Scott, Branimir, and Zeljko === Preliminary Analysis === 1) Recap: the original LINEAR magnitudes reported to MPC were calibrated to correspond to the standard V magnitude for a solar-type star (not Vega!), so we expect: L = V + a*[(B-V)-0.65] where 0.65 is the solar B-V color, and the a coefficient is not provided. The calibration is actually based on B magnitudes from the USNO A.2 catalog (themselves poorly calibrated, see Sesar et al. 2006, AJ 131, 2801), which are then transformed to the V band using B-R magnitudes and somewhat incorrect transformation coefficients. Due to all these factors, it is likely that the a coefficient is not a universal constant, and that an additional offset should be added to the above expression (i.e. 0.65 times the error in the a coefficient). 2) Branimir's recalibrated magnitudes, Ls, are approximately equal to (over the asteroid color range; the full definition is based on the 4th-order polynomial of g-i color): Ls = r + 0.075 -0.048*(g-i) Standard stars with their g,r,i mags from SDSS are used to calibrate instrumental magnitudes extracted using SExtractor. Note that the color term (-0.048) is very small, and induces only a 0.014 mag large systematic effect between the C type (the median g-i=0.64) and S type (g-i=0.93) asteroids with unknown color. In other words, the SDSS r band magnitude is very closely related to the LINEAR bandpass (the latter is wider, but for smooth and simple asteroid SEDs, the color term between them is not large). 3) A comparison of L reported to MPC, as obtained from Scott, and recalibrated Ls from Branimir (for the same detections!) gives: L = Ls + 0.47 - 0.08*(g-i) with an rms of 0.45 mag, and a median difference of 0.40 mag. That is, L = r + 0.55 - 0.13*(g-i), with a median = 0.45 for the median asteroid g-i=0.8. By combining this result with SDSS to Johnson transformations from Ivezic et al. (2007, AJ 134, 973; the full transformations are non-linear, but approximately, B-V = 0.945*gr + 0.21, and g-i=1.378*gr - 0.01), the best transformation between the *original* LINEAR magnitudes and the Johnson system is L = V + a0*[(B-V)-0.65] + b0 = V - 0.649*(B-V) +0.69 where a0 = -0.649 and b0=0.27. Note that b0 is not 0, as suspected under recap above, and that a0 induces a systematic offset of 0.13 mag between the median B-V colors for C and S type asteroids (that is, much larger than when the L magnitudes are transformed to the SDSS r band magnitudes). 4) The above result implies that the best transformation from the *original* LINEAR magnitudes reported to MPC to the standard V band is Vlinear = L + 0.649*(B-V) - 0.69 (for the median asteroid color of B-V=0.78, Vlinear = L - 0.18). Note that although the last expression is a good description of the original LINEAR magnitudes, they show a scatter of about 0.45 mag around that relation due to random calibration errors. 5) Instead, the MPC code interprets L magnitudes as the "B" band magnitudes, and computes corresponding V magnitudes as (we think): Vmpc = L - 0.80 where 0.80 is presumably typical B-V color for asteroids. Thus, the error in MPC interpretation is Vlinear - Vmpc = [L+0.649*(B-V)-0.69] - [L-0.80] = 0.649*(B-V) + 0.11 = 0.649*[(B-V)-0.78] +0.62 where B-V=0.78 is the median B-V color for asteroids with SDSS observations. Therefore, for a "typical" asteroid, ** Vmpc mags generated from reported LINEAR mags are 0.62 mag too bright. ** The difference ranges from 0.53 mag for the blue mode (C type) at B-V=0.65 to 0.66 mag for the red mode (S type) at B-V=0.85. This doesn't compare well with the results from Parker et al. (2008, Icarus, 198, 138) which show that = 0.33 mag for about 70% of the sample (and not 0.62 mag as derived above). On the other hand, the H mags could be only partially affected by LINEAR measurements and thus the difference could be smaller than expected. === Data File === These data are included in file (gzip-ed plain ascii) http://www.astro.washington.edu/users/ivezic/linear/MPC_LINEAR_match.dat.gz There are 157,515 data lines in this file, and 30 header lines. The header describes the data format.