ESO large program on Centaurs and TNOs: visible colors—final results

  • Published on

  • View

  • Download


  • 4) 15

    n CfiN. Peixinho,a,b, H. Boehnhardt,c I. Belskaya,a,d A. Doressoundiram,a M.A. Barucci,aand A. Delsanti a

    a LESIA, Observatoire de Paris, 5, pl. Jules Janssen, FR-92195 Meudon cedex, Franceb CAAUL, Observatrio Astronmico de Lisboa, Tapada da Ajuda, PT-1349-018 Lisboa, Portugal

    c Max-Planck-Institut fr Astronomie, Knigstuhl 17, DE-69117 Heidelberg, Germanyd Astronomical Observatory of Kharkov National University, Sumskaya St., 35, UA-Kharkov 61022, Ukraine

    Received 25 September 2003; revised 25 February 2004Available online 20 April 2004


    We report 43 new visible colors of Centaurs and TNOs, obtained at NTT and VLT telescopes under the ESO large program on physicalproperties of Centaurs and TNOs. Merging these new measurements with those obtained during the first part of the program (Boehnhardt etal., 2002, Astron. Astrophys. 395, 297303) and the Meudon Multicolor Survey (Doressoundiram et al., 2002, Astron. J. 124, 22792296)we have a unique dataset of 109 objects. We checked for correlations and trends between colors, physical and orbital parameters, carryingout an analysis based on Monte Carlo simulation to account for observational error bars. Centaurs show no evidence for correlation betweenV R vs. R I colors which raises the hypothesis that more than one single coloring process might be acting on their surfaces. Classicalobjects seem to be composed of two different color populations: objects with i < 4.5 display only red colors while those with i > 4.5display the whole range of colors from blue to very red. The possibility that the low inclined population is misguiding global conclusionsis analyzed. Classical objects also show a stronger colorperihelion correlation for intrinsically brighter objects, corresponding to criticalestimated sizes of different formation/evolutionary histories. Scattered disk objects show color resemblances with the classical objects ati > 12, hence surface reflectivities resemblances, pointing to a common origin. No coloraphelion trend is found for SDOs, as expectedfrom the intense irradiation by galactic cosmic-rays beyond the solar wind termination shock. Plutinos show a colorabsolute magnitudetrend, in which all the intrinsically faintest objects are blue. We see many red Plutinos in highly inclined and highly eccentric orbits, thatshould have originated in a primordial inner disk under Gomes (2003, Icarus 161, 404418) migration scenario. This seems to invalidate theassumption that objects originated in this inner disk are mainly blue. Finally, we also find six candidates for lightcurve studies: four objects(1998 WU31, 1999 OE4, 1999 OX3, and 2001 KP77) present significant short term R-magnitude variability, and two objects (1999 XX143and 2000 GP183) evidence possible color variations with rotation. 2004 Elsevier Inc. All rights reserved.

    Keywords: Kuiper belt objects; Trans-neptunian objects; Centaurs; Photometry

    1. Introduction

    Within the ten years since the discovery of the firsttrans-neptunian objects (TNOs), large and high quality colordatasets have been obtained providing significant results.In particular, interesting correlations between optical colorsand some orbital parameters (i , e, q) have been found for

    Based on observations collected at the European Southern Observa-tory, Chile, programs 167.C-0340(A), (C), and (D).

    some classes of TNOs and possible implications on theirevolution and origin have been advanced. For a completereview on these photometric surveys and their results seeDoressoundiram (2003).

    Trans-neptunian objects represent a population locatedbeyond Neptune. Many of them are in quasi-circular orbitswith semi-major axis between 35 and 48 AU and are knownas Classical objects or Cubewanos (after the discoveryof 1992 QB1). Some others are known as Resonant ob-jects located in some major mean motion resonances withIcarus 170 (200

    ESO large program ovisible colors* Corresponding author. Fax: + 33-1-45-07-28-06.E-mail address: (N. Peixinho).

    0019-1035/$ see front matter 2004 Elsevier Inc. All rights reserved.doi:10.1016/

    entaurs and TNOs:nal results Neptune (essentially the 3:1, 3:2, 2:1, 5:2, and 5:3 reso-nances). The objects on the 3:2 resonance are usually known

  • arus154 N. Peixinho et al. / Ic

    as Plutinos for the analogy of their orbit with that of Pluto.A scattered disk of objects, with highly eccentric and highlyinclined orbits, exists mainly beyond 48 AU. Moreover,Centaurs, that are located on unstable orbits whose semi-major axis fall between those of Jupiter and Neptune, aremost probably escapees from the TNO population. There-fore, Centaurs are usually added to the TNOs family.

    To better understand the physical properties and surfacecomposition of Centaurs and TNOs a large program at ESOhas been proposed to study these objects by visible andnear-infrared photometry and visible and near-infrared spec-troscopy. About 250 hours of observing time have been al-located to this project to be executed at ESO telescopes inCerro Paranal and La Silla during April 2001 to March 2003.The first results in photometry and spectroscopy have beenalready published by Boehnhardt et al. (2001, 2002), Barucciet al. (2002), Lazzarin et al. (2003), Dotto et al. (2003),and Doressoundiram et al. (2003a, 2003b). Further resultsin near-infrared photometry will be published by Delsanti etal. (paper in preparation).

    In this paper we report the results obtained for 43 ob-jects observed in the visible BVRI photometry at VLT-Paranal (Very Large Telescope) and at NTT-La Silla (NewTechnology Telescope). Adding these measurements withprevious published data obtained under the same program(Boehnhardt et al., 2002, hereafter BOE02) we obtain a ho-mogeneous data set of 71 objects. Merging our ESO largeprogram data set (hereafter, LP) with the Meudon MulticolorSurvey (Doressoundiram et al., 2002, hereafter DOR02)observed and reduced with a similar methodologyprovidesus a unique dataset of 109 objects. We analyze and discussthe colorcolor, colorHR, and colororbital parameters re-lations, as well as the dependency of some correlations onHR . A complete discussion is reported for both LP andDOR02 data sets taken together with some comparisons withour LP data set, taken alone.

    2. Observations and data reduction

    2.1. Observations

    Our photometric observations have been performed be-tween October 2001 and April 2002. The majority of thetargets have been observed in service mode at VLT (8.2 mtelescope) with the exceptions of the brightest ones that wereimaged at NTT (3.5 m telescope) in visitor mode on thenights of April 1719, 2002. See Table 1 for observationalcircumstances.

    On VLT, the FORS1 instrument is used in imaging modewith broadband Bessel BVRI filters. The detector is a TekCCD with 20482048 pixels, covering a 6.8 6.8 field ofview with 0.2/pixel. Exposures were performed under clear

    (i.e., with high extinction values) to photometric conditionswith dark skies and seeing < 0.8.170 (2004) 153166

    On NTT, the SUSI2 CCD camera at the f/11 Nasmythfocus was used with Bessel BVRI filters. The detector wasa mosaic of two EEV CCDs with 2048 4096 pixels each,covering a 5.5 5.5 field of view with 0.16/pixel. Thethree nights had seeing of 0.9 and sky conditions wereclear to photometric.

    In both cases (VLT and NTT) a standard filter expo-sure series RBVIR was executed sequentially for each object(with some exceptions) and telescopes tracked at siderealrate. Exposure times range from 180 to 1200 seconds (toavoid excessive trailing, the longer exposures are the sum oftwo frames centered on the object). The observational strat-egy adopted was the same as used for first part of the largeprogram, as described in BOE02.

    2.2. Data reduction

    2.2.1. Image processingThe images from VLT observations were processed by

    the FORS1 pipeline1 using the standard techniques of biassubtraction (bias estimated from overscan regions) and flatfield correction. NTT observations were processed usingIRAFs CCDRED package, with the same procedure butusing bias frames obtained from zero second exposures in-stead. For both cases bad pixels and bad columns were man-ually corrected only when they could affect the measurementof objects magnitudes.

    2.2.2. Aperture photometryObjects magnitudes were measured using the aperture

    correction technique, except for two bright Centaurs wherethe classical method was used.

    Most of an objects flux may be collected with an aper-ture radius of about 3.5 times the FWHM (99% of the flux).Nonetheless, in the case of faint objects, such large aperturescontain a very low S/N ratio. Therefore, a slight misestima-tion of the sky value will result in a largely misestimatedmagnitude. This effect is negligible for bright objects andalso for small apertures (where the S/N ratio is high). Theeffect is minimized by measuring the objects flux within asmall aperture (typically with a radius of the order of theseeing), the flux loss is then corrected with an analysis of theflux variation with increasing aperture radius of the brightfield stars, the so-called growth-curve (Howell, 1989).

    Corrections were computed analyzing a few tens of brightfield stars on the corresponding frames (rejecting any devi-ating profiles). Typically, corrections were done between aradius of 1 and 5 times the FWHM.

    2.2.3. Standard calibrationFor NTT observations Landolt standard stars were ob-

    served at different air masses for each night and the photo-metric calibration parameters (zeropoints, first-order extinc-


  • on C1998 WT31 Clas VLT 2001/11/18 37.615 38.593 0.21998 WU31 Plut VLT 2001/11/19 31.983 32.971 < 0.11998 WV31 Plut VLT 2001/10/23 32.724 33.681 0.51998 WW24 Plut VLT 2001/12/08 30.741 31.663 0.61998 WZ31 Plut VLT 2001/10/22 32.347 33.171 1.01999 CB119 Clas VLT 2002/01/10 40.495 40.968 1.21999 CL119 Clas NTT 2002/04/18 46.528 46.726 1.21999 CV118 Scat VLT 2002/02/11 38.092 39.049 0.41999 CX131 5:3 VLT 2002/01/10 41.612 42.512 0.5

    (38084) 1999 HB12 5:2 VLT 2002/02/19 33.960 34.654 1.2. . . VLT 2002/03/22 33.673 34.635 0.41999 HS11 Clas VLT 2002/03/18 42.881 43.735 0.71999 OE4 Clas VLT 2001/10/09 42.908 43.447 1.11999 OF4 Clas VLT 2001/10/10 44.554 45.074 1.11999 OJ4 Plut VLT 2001/10/09 37.680 38.206 1.3. . . Plut VLT 2001/10/10 37.695 38.205 1.3

    (44594) 1999 OX3 Cent NTT 2002/04/19 26.674 26.479 2.11999 RX214 Clas VLT 2001/10/23 44.966 45.888 0.51999 RY214 Clas VLT 2001/10/09 36.617 37.599 0.31999 XX143 Cent VLT 2001/12/10 24.267 24.996 1.5. . . VLT 2002/01/22 24.044 25.027 0.1

    (60454) 2000 CH105 Clas VLT 2002/01/10 43.301 43.875 1.02000 CL104 Clas NTT 2002/04/18 42.161 42.737 1.12000 CN105 Scat NTT 2002/04/17 45.349 45.912 1.02000 CQ105 Plut VLT 2002/01/22 49.299 50.245 0.3

    (60620) 2000 FD8 Clas VLT 2002/02/19 39.084 39.868 0.92000 FV53 Plut VLT 2002/03/17 29.396 30.296 0.82000 FZ53 Cent VLT 2002/03/18 16.626 17.396 2.12000 GP183 Plut NTT 2002/04/18 36.252 37.256 < 0.1. . . Plut VLT 2002/03/20 36.400 37.259 0.82000 OU69 Clas NTT 2002/04/18 41.157 41.088 1.42000 QB243 Scat NTT 2002/04/17 19.538 19.144 2.72000 YW134 Scat NTT 2002/04/17 42.869 42.894 1.3. . . NTT 2002/04/18 42.886 42.894 1.3

    (63253) 2001 BL41 Cent NTT 2002/04/19 8.329 8.550 6.72001 FP185 Scat NTT 2002/04/17 33.379 34.284 0.72001 FZ173 Scat NTT 2002/04/18 32.357 33.246 0.82001 KA77 Clas NTT 2002/04/17 48.123 48.863 0.82001 KB77 Plut NTT 2002/04/17 30.452 31.429 0.42001 KD77 Plut NTT 2002/04/17 34.582 35.300 1.22001 KP77 Clas NTT 2002/04/19 35.163 36.011 0.92002 CB249a Cent NTT 2002/04/17 13.190 13.901 3.0

    (42355) 2002 CR46 Scat NTT 2002/04/19 17.970 18.080 3.22002 DH5 Cent NTT 2002/04/19 13.933 14.499 3.3

    (55576) 2002 GB10 Cent NTT 2002/04/19 14.288 15.206 1.62002 GO9 Cent NTT 2002/04/19 13.062 14.039 1.0

    a Object classified with one opposition only.

    tion coefficients and color terms) were computed solving thenon-linear equations. For VLT data the calibration was doneusing the parameters given by the VLT Observatory, also cal-culated from Landolt stars, for each night.2


    3. Photometric results

    Photometric errors were estimated by the most probableerror resulting from the combination of the error of photo-metric measurement (phot, using the CCD equation), theESO large program

    Table 1Observational circumstances

    Object Type Telescope1998 US43 Plut VLT1998 WA31 5:2 VLT1998 WS31 Plut VLT and TNOs 155

    Date (UT) (AU) r (AU) (deg)2001/10/23 34.657 35.562 0.72001/11/20 38.793 39.741 0.42001/10/22 30.517 31.472 0.5uncertainty of aperture correction resulting from the dis-persion among measurements of the field stars (apcorr, see

  • arus156 N. Peixinho et al. / Ic

    Stetson, 1990) and calibration error (calib, taken as the rmsof the standard stars fit):

    (1) =

    2phot + 2apcorr + 2calib.Our observations were made at small phase angles where

    the opposition effect phenomena should play an importantrole. Available estimations of this effect for a few TNOsand Centaurs have shown a linear magnitude-phase depen-dence down to phase angles of 0.10.2 (Belskaya et al.,2003). Absolute magnitudes (HR) are calculated with phasecorrection taking the linear approximation phase function() = 10 :

    (2)HR R(1,1,0) = R 5 log (r) ,where R is the R-band calibrated magnitude, r is theobjects heliocentric distance (AU), is the objects geo-centric distance (AU), is the phase angle during the obser-vation (deg), and is the phase curve slope (mag/deg).

    For TNOs we take the modal value (and not the mean) ofthe measurements published by Sheppard and Jewitt (2002): = 0.14 0.03. As to Centaurs, values are not knownfor neither of our targets. Only one object (Ellatus, for-merly 1999 UG5) has a well sampled phase curve published(Bauer et al., 2002). For all Centaurs we use a value of = 0.11 0.01 obtained from a least-squares fit of the lin-ear approximation () to the published data.

    From these published values we may see how impor-tant phase correction is. Although no impact on colors isexpected from these corrections, they should always be ap-plied to estimate absolute magnitudes. Nonetheless, due tolow statistics these values should be considered as rough es-timates.

    Diameters (D, in kilometer) are estimated using Russell(1916) equation:

    (3)D = 2

    2.24 1016 100.4(RHR)pR


    where R is the Suns R-magnitude, HR the objects ab-solute magnitude and pR the geometric albedo in the R-band. All sizes are estimated using pR = 0.09 (Brown andTrujillo, 2004) and not the usual 0.04 albedo.

    The uncertainties on size estimations are quite high andthey should be taken with some caution. All HR for BOE02data were recalculated using these values. Average colors,absolute magnitudes and estimated sizes are presented inTable 2. Individual B , V , R, and I magnitudes with thecorresponding time at each observation are available online.3

    3.1. Photometric variations

    In thi...


View more >