Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.mso.anu.edu.au/~bessell/FTP/Hipparcos/bessell_hip_PASP.pdf Äàòà èçìåíåíèÿ: Wed Aug 22 04:09:58 2007 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 11:00:32 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: þæíàÿ àòëàíòè÷åñêàÿ àíîìàëèÿ

PUBLICATIONS OF THE ASTRONOMICAL SOCIETY OF THE PACIFIC, 112 : 961 õ 965, 2000 July ( 2000. The Astronomical Society of the Paciïc. All rights reserved. Printed in U.S.A.

The Hipparcos and Tycho Photometric System Passbands
MICHAEL S. BESSELL
Research School of Astronomy and Astrophysics, Institute of Advanced Studies, The Australian National University, Private Bag, Weston Creek P.O., ACT 2611, Australia ; bessell=mso.anu.edu.au Received 2000 March 6 ; accepted 2000 April 17

ABSTRACT. Hipparcos and T ycho magnitudes have been extracted for many of the E-region standard stars. Dierences between B, V , and the Hipparcos and T ycho magnitudes have been derived and regressed against color. Fiducial lines and polynomial ïts have been derived and compared with values in the Hipparcos and T ycho Catalogues. Synthetic photometry has been made using the Vilnius spectra and compared with the observations. As a result, it has been necessary to modify some of the published bandpasses.

1. INTRODUCTION The Hipparcos mission of the European Space Agency was a remarkable achievement. As well as providing parallaxes of unprecedented precision and accuracy for the nearby stars, it also produced exceedingly precise magnitudes for hundreds of thousands of stars. All this is described in the Hipparcos and T ycho Catalogues (Perryman et al. 1997, hereafter HTC). It is these magnitudes and the photometric system that they deïne that are the subject of this paper. The main Hipparcos detector was an unïltered S20 image dissector scanner which provided the H magnitudes. Most stars brighter than 8.5 were meaP sured with a precision of a few tenths of a millimagnitude. In addition, light from the star mapper area was divided by a dichroic beam splitter onto two photomultiplier tubes, providing simultaneously measured B and V magnitudes. T T The T ycho catalog provides magnitudes for a larger number of stars, but for the fainter stars the precision is lower than the Hipparcos catalog. For the brighter stars it is comparable. Never before has such a wealth of accurately calibrated and precisely measured photometric data been obtained, data covering the whole sky, north and south of the equator. This enables intercomparison of many of the ground-based standard photometric systems and a search to be made for systematic dierences. Amongst the most precisely deïned ground-based standard systems are the UBV RI E-region standards set up originally by Cousins (1974, 1976) and continued by Menzies et al. (1989). These standards comprise samples of stars with spectral types between B and early M and apparent magnitudes between m \ 4 and 9 in 10 regions with a V declination of about S35¡ and at roughly 3 hour intervals around the sky. Additional very blue and red standards on 961

the same system have been provided by Kilkenny et al. (1998). During an investigation of the zero points of the various sets of UBV RI standard-star photometry (Cousins & Bessell 2000), it became apparent that the dierences between the South African Astronomical Observatory (SAAO) B and V magnitudes (E-region : Menzies et al. 1989 ; E-region second 10 year means : J. W. Menzies 1999, private communication ; supplementary blue and red stars : Kilkenny et al. 1998) and the H , B , and V magnitudes PT T (HTC) diered slightly from those described in tables and equations provided in the HTC. These precisely deïned dierences are interesting in that they indicate qualitatively and quantitatively the dierences to be expected when different passbands are used, dierences that have been masked in the past by the lower precision of most groundbased photometry systems. Such systematic dierences are important to consider also in the context of new magnitude data from new and broader band photometric systems of the gravitational lensing programs.1 In addition, when synthetic photometry was carried out on the Vilnius spectra (Straizys & Sviderskiene 1972) using the published H , B , and V passbands (HTC) and the PT T standard BV passbands (Bessell 1990), some of the synthetic relations did not agree well with the observed relations. It is important that the passbands of the HTC and other new systems be well understood so that calibrations of temõõõõõõõõõõõõõõõ 1 MACHO (http ://wwwmacho.anu.edu.au), EROS (http :// www.lal.in2p3.fr/recherche/eros/), and OGLE (http ://www.astrouw.edu.pl/ Dftp/ogle/)õmonitoring the LMC Bar and Galactic bulgeõand wide-ïeld surveys, such as the Sloan Digital Sky Survey (http :// www.sdss.org).


962

BESSELL
TABLE 1 RELATION BETWEEN V [I, B[V , AND Hipparcos/T ycho DATA FOR MAIN-SEQUENCE BõG STARS AND KõM GIANTS V [I [0.250 [0.200 [0.150 [0.100 [0.050 [0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 0.600 0.650 0.700 0.750 0.800 0.850 0.900 0.950 1.000 1.050 1.100 1.150 1.200 1.250 1.300 1.350 1.400 1.450 1.500 1.600 1.700 1.800 1.900 2.000 2.100 2.200 2.300 2.400 2.500 ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ ............ V [H P 0.089 0.068 0.046 0.026 0.006 [0.015 [0.030 [0.041 [0.051 [0.059 [0.067 [0.075 [0.082 [0.089 [0.095 [0.101 [0.110 [0.116 [0.125 [0.132 [0.135 [0.140 [0.146 [0.153 [0.160 [0.160 [0.163 [0.166 [0.165 [0.163 [0.160 [0.156 [0.152 [0.147 [0.143 [0.138 [0.127 [0.116 [0.106 [0.097 [0.086 [0.076 [0.066 [0.056 [0.046 [0.036 B[V [0.244 [0.198 [0.152 [0.104 [0.050 0.009 0.060 0.102 0.145 0.185 0.223 0.262 0.300 0.339 0.381 0.425 0.476 0.535 0.595 0.654 0.704 0.761 0.825 0.893 0.957 1.007 1.076 1.133 1.192 1.238 1.285 1.326 1.364 1.401 1.437 1.468 1.504 1.522 1.545 1.569 1.592 1.615 1.638 1.661 1.684 1.707 *(B[V ) 0.021 0.016 0.010 0.006 0.003 [0.004 [0.013 [0.019 [0.021 [0.020 [0.022 [0.025 [0.026 [0.026 [0.028 [0.030 [0.038 [0.049 [0.061 [0.072 [0.081 [0.096 [0.111 [0.128 [0.142 [0.157 [0.181 [0.194 [0.210 [0.225 [0.241 [0.253 [0.263 [0.273 [0.283 [0.291 [0.300 [0.305 [0.311 [0.318 [0.324 [0.330 [0.336 [0.342 [0.348 [0.354

FIG. 1.õThe V [H vs. V [I regression for individual E-region stars P (10 year mean data). The thin lines represent the empirical relations for main-sequence stars and for red giants given in the HTC. The plus signs are the synthetic photometry derived from the set of Vilnius spectra using the new Hipparcos passband. The thick line is the locus of the synthetic photometry derived from the Vilnius spectra using the original Hipparcos passband given in the HTC.

perature and luminosity can be made from model atmosphere ÿuxes for stars, such as metal-deïcient stars and supergiants, that are poorly represented in standard-star photometry lists and spectrophotometric atlases. In this paper are presented some mean empirical relations between the Cousins-Johnson B and V magnitudes and Hipparcos-T ycho H , B , and V magnitudes as deïned PT T by the E-region stars and the sample of bluer sdOB stars and redder KõM dwarf stars of Kilkenny et al. (1998) together with modiïed H , B , and V bandpasses that PT T better match the observed interrelations.

2. DISCUSSION A list of HD numbers for the E-region stars was submitted to the VizieR Catalogue interface to the HTC pro vided by the Centre de Donnees Astronomiques de Strasbourg. Where available the HTC data were returned and combined with the UBV RI data. The two sets of UBV RI data used were the original E- and F-region 1975õ 1989 data set and an "" independent îî subset of E-region 1989õ1999 data. There was no systematic dierence between the two data sets, but the 10 year (1989õ1999) set had a higher internal precision of a few millimagnitudes and was mainly for stars fainter than V \ 6. There were 419 stars in the ïrst list and 275 in the second. After rejecting some more highly reddened stars, some "" ÿiers,îî and those few stars with HTC errors exceeding 0.02 mag, there remained 389 stars in the ïrst list and 253 (230 for T ycho)in the second. Tables 14.1õ14.4 in HTC volume 3 list relationships between V [I and H [V , and V [I and B[V , for giants P

and dwarfs. The H , B , and V passbands are given in ° 1.3 PT T of volume 1. The number of photons counted in passband R for a X star with ÿux f (j) is given by

P

[ f (j)/hl]R (j)dj \ (1/hc) X

P

f (j)[jR (j)]dj . X

2000 PASP, 112 : 961õ965


HIPPARCOS AND T Y CHO PHOTOMETRIC SYSTEM
TABLE 2 RELATION BETWEEN B [V AND Hipparcos/Tycho T T DATA FOR BõG MAIN-SEQUENCE STARS AND KõM GIANTS B[ T [0.250 [0.200 [0.150 [0.100 [0.050 [0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 0.450 0.500 0.550 0.600 0.650 0.700 0.750 0.800 0.850 0.900 0.950 1.000 1.050 1.100 1.150 1.200 1.250 1.300 1.350 1.400 1.450 1.500 1.550 1.600 1.650 1.700 1.750 1.800 1.850 1.900 1.950 2.000 V T ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... T 0.038 0.030 0.022 0.015 0.008 0.001 [0.005 [0.012 [0.018 [0.024 [0.029 [0.035 [0.040 [0.045 [0.050 [0.054 [0.059 [0.064 [0.068 [0.072 [0.077 [0.081 [0.085 [0.089 [0.093 [0.098 [0.102 [0.106 [0.110 [0.115 [0.119 [0.124 [0.128 [0.133 [0.138 [0.143 [0.148 [0.154 [0.160 [0.165 [0.172 [0.178 [0.185 [0.191 [0.199 [0.206 V [V *(B[V ) 0.031 0.021 0.011 0.005 0.002 [0.005 [0.010 [0.017 [0.020 [0.021 [0.023 [0.025 [0.025 [0.026 [0.030 [0.035 [0.045 [0.051 [0.060 [0.068 [0.076 [0.085 [0.094 [0.104 [0.113 [0.122 [0.131 [0.142 [0.154 [0.166 [0.178 [0.189 [0.199 [0.210 [0.222 [0.234 [0.245 [0.256 [0.266 [0.277 [0.288 [0.299 [0.309 [0.320 [0.331 [0.342 V [H P 0.066 0.051 0.036 0.021 0.006 [0.011 [0.025 [0.038 [0.048 [0.058 [0.069 [0.079 [0.087 [0.094 [0.101 [0.108 [0.114 [0.120 [0.127 [0.131 [0.134 [0.137 [0.142 [0.147 [0.151 [0.155 [0.158 [0.157 [0.160 [0.162 [0.164 [0.166 [0.166 [0.165 [0.164 [0.161 [0.157 [0.153 [0.148 [0.143 [0.137 [0.131 [0.125 [0.119 [0.112 [0.106

963

FIG. 2.õThe new adopted jR(j) response for H (thick line) compared P with the original passband (medium line) and the standard V passband (thin line).

the red. This can be better understood in terms of its requiring more red photons than blue photons to yield the same (energy) ÿux. The B[V versus V [I relation for early-type stars and red giants given below was derived from the less reddened E-region stars. 2.1. The H and V Magnitudes P Figure 1 plots the V [H versus V [I relation from the P 10 year E-region data. Note the extremely small scatter for the hotter stars and the increasing scatter for the redder stars where presumably the metal-line dierences become more obvious. However, it is very interesting that the scatter is so small given the large dierence in the width of the V and H passbands. The ïducial lines from HTC P

This rearrangement shows how one can then use the same ÿuxes with the jR (j) responses to compute the synthetic X photometry. That is, in order to convert quantum efficiency (QE) based passbands into photon-counting passbands, the QE-based passbands are multiplied by the wavelength, then renormalized. This has the eect of weighting the ÿux by the wavelength and essentially shifts the passbands slightly to

FIG. 3.õThe V [V vs. B [V regression for the E-region stars. The T T T straight line is the suggested approximate empirical relation given in the HTC ; the curve is a polynomial that better ïts the data. The plus signs are the synthetic photometry derived from the Vilnius spectra using the new passband.

2000 PASP, 112 : 961õ965


964

BESSELL
TABLE 3 NORMALIZED jR (j) RESPONSES X j P 0.000 0.015 0.032 0.052 0.103 0.155 0.227 0.300 0.400 0.530 0.700 0.845 0.928 0.970 1.000 0.997 0.988 0.973 0.956 0.935 0.911 0.887 0.858 0.825 0.789 0.752 0.714 0.675 0.637 0.599 0.562 0.525 0.487 0.449 0.411 0.372 0.335 0.299 0.264 0.232 0.203 0.176 0.152 0.131 0.112 0.093 0.074 0.057 0.043 0.033 0.026 0.020 0.013 0.005 0.000 H j 350 355 360 365 370 375 380 385 390 395 400 405 410 415 420 425 430 435 440 445 450 455 460 465 470 475 480 485 490 495 500 505 ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... T 0.000 0.011 0.048 0.103 0.175 0.262 0.363 0.468 0.569 0.656 0.722 0.771 0.810 0.846 0.886 0.929 0.968 1.000 0.985 0.879 0.707 0.510 0.334 0.215 0.147 0.110 0.088 0.066 0.040 0.018 0.003 0.000 B j 455 460 465 470 475 480 485 490 495 500 505 510 515 520 525 530 535 540 545 550 555 560 565 570 575 580 585 590 595 600 605 610 615 620 625 630 635 640 645 650 655 660 665 670 675 ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... V T 0.000 0.020 0.106 0.280 0.498 0.700 0.835 0.912 0.953 0.980 0.996 1.000 0.994 0.977 0.952 0.921 0.886 0.849 0.808 0.766 0.723 0.678 0.632 0.584 0.536 0.487 0.439 0.391 0.345 0.302 0.261 0.225 0.194 0.166 0.141 0.121 0.103 0.087 0.074 0.060 0.049 0.037 0.024 0.011 0.000

340 350 360 370 380 390 400 410 420 430 440 450 460 470 480 490 500 510 520 530 540 550 560 570 580 590 600 610 620 630 640 650 660 670 680 690 700 710 720 730 740 750 760 770 780 790 800 810 820 830 840 850 860 870 880

...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ......

FIG. 4.õThe (B[V )[(B [V ) vs. B [V regression for the E-region T T T T stars. The straight line is the suggested approximate empirical relation given in the HTC ; the curve is a polynomial that better ïts the data. The plus signs are the synthetic photometry derived from the Vilnius spectra using the new passband.

volume 3 Tables 14.1 and 14.2 are also plotted ; the line for early-type stars and red giants lies within 0.01 mag of the E-region observations. In Table 1 are given the mean observational lines from the E-region data together with the interpolated B[V values from the mean B[V versus V [I relation deïned by the E-region stars. The H passband, modiïed as described above, was used P with the Vilnius spectra (Straizys & Sviderskiene 1972) to compute synthetic H magnitudes. When combined with P the synthetic V magnitude and plotted against the synthetic V [I colors for early-type dwarfs and late-type giants, it diered greatly from the observed relation as seen in Figure 1 (thick line). The passband was adjusted in various ways, and eventually a passband with the original red edge but with a blue edge shifted redward by about 30 nm gave quite good agreement with the observations. The colors computed for the individual Vilnius spectra using the revised Hipparcos passband are shown in Figure 1 as plus signs ; the three upper plus signs redward of V [I \ 1.0 are for K-dwarf spectra. Although the agreement is not perfect, given the uncertainties involved, it is not really worthwhile to attempt further modiïcations of the band on the basis of these data. As it stands, the adopted passband is adequate to produce useful synthetic magnitudes from grids of model atmosphere ÿuxes. Such computations are being done by Bessell & Castelli (2000). In Figure 2 the original and revised H passbands are shown with the standard V for P comparison. 2.2. The V and V Magnitudes T In Figure 3 are plotted the observed dierences between standard V and T ycho V for the E- and F-region stanT dards. Also shown is the straight line suggested in the HTC

2000 PASP, 112 : 961õ965


HIPPARCOS AND T Y CHO PHOTOMETRIC SYSTEM

965

Although better than the straight line, there are still systematic deviations from the curve resulting mainly from the Balmer jump and the conÿuence of the hydrogen lines. The mean empirical relation has been derived and is given in Table 2. The heading *(B[V ) in Tables 1 and 2 refers to (B[V )[(B [V ). The plus signs show the synthetic phoT T tometry which now agrees well. In Figure 5 are shown the adopted B and V passbands in comparison with the stanT T dard B and V passbands. Table 3 lists the adopted passbands for H , B , and V . PT T

3. SUMMARY
FIG. 5.õjR(j) responses of the B and V passbands (thick lines) comT T pared with the B and V passbands (thin lines).

as approximating the relationship, namely, V [V \[0.09 T (B [V ), and a cubic that better ïts the data. T T Although a cubic is better than a straight line, the data deviate systematically from the cubic and are best represented by a spline. In Table 2 the observed mean relation between V [V and B [V is listed. T T T The V [V and B [V colors were synthesized from the T T T Vilnius spectra and compared with the observations. Here the agreement was much better. The modiïed (photon counting) B and V passbands reproduced the observed T T relations quite well. The synthetic data are plotted in Figure 3 as plus signs. They ït well except for the late M giants. Figure 4 shows the dierence between B[V and B [V T T as a function of B [V color. The straight line is the T T approximate relation suggested in the HTC for stars other than M stars. The curve is a fourth-order polynomial ït.

Comparison between the E-region standard UBV RI magnitudes and the H , B , and V magnitudes from the PT T Hipparcos and T ycho Catalogues for the same stars has yielded improved relations between the systems. Synthetic photometry carried out with the Vilnius spectra for H , B , PT V , B, and V were also compared with the observations, and T as a result a large modiïcation was necessary to the H P bandpass. The revised relations and passbands are shown in a series of ïgures and tables. These should permit better calibration of the Hipparcos and T ycho photometry using model atmosphere ÿuxes. I would like to thank John Menzies and Dave Kilkenny for providing the digital data of the E-region stars and the blue and red supplementary stars, Alan Cousins for discussions and the stimulus for this work, and Bob Stobie for making possible my visit to the SAAO. I would also like to thank Catherine Turon and the referee Michael Richmond for useful comments on the paper.

REFERENCES
Bessell, M. S. 1990, PASP, 102, 1181 Bessell, M. S., & Castelli, F. 2000, A&A, submitted Cousins, A. W. J. 1974, MNRAS, 166, 711 õõõ. 1976, MmRAS, 81, 25 Cousins, A. W. J., & Bessell, M. S. 2000, MNRAS, submitted Kilkenny, D., van Wyk, F., Roberts, G., Marang, F., & Cooper, D. 1998, MNRAS, 294, 93 Menzies, J. W., Cousins, A. W. J, Banïeld, R. M., & Laing, J. D. 1989, SAAO Circ., 13, 1 Perryman, M. A. C., et al. 1997, The Hipparcos and Tycho Catalogues (SP-1200 ; Noordwjk : ESA) (HTC) Straizys, V., & Sviderskiene, Z 1972, Astron. Obs. Bull. Vilnius, 35, 1

2000 PASP, 112 : 961õ965