The CRONA (Caltech-Roma-Napoli) project

a r X i v :a s t r o -p h /9706238v 1 24 J u n 1997The CRONA (Caltech-Roma-Napoli)project

S.Andreon 1,S.Zaggia 1,R.de Carvalho 2,S.Djorgovski 3,

G.Longo 1,I.Musella 1,R.Scaramella 4

1Osservatorio Astronomico di Capodimonte,Naples,Italy

2Observatorio Nacional de Rio de Janeiro,Brasil

3Palomar Observatory,Caltech,USA

4Osservatorio Astronomico di Roma,Monteporzio,Italy

Abstract The CRONA project is a joint enterprise between Caltech and the astronomical obser-vatory of Rome and Naples aimed to the photometric calibration of the Second Palomar Sky Survey (POSS-II)plates and to the construction of a complete photometric and astro-metric catalogue of all objects present on the POSS-II.The calibrated ?nal catalogues of objects will be made available to the community.We present here a comparison between

the performances of the SKICAT and DAOPHOT packages applied to POSS-II images of the globular cluster M92.

1Introduction

CRONA (Caltech-ROma-NApoli)is a recent extension of a lorg term project started at Caltech in 1993(Djorgovski et al.1992)and aimed at the construction of the so called Palomar Norris Sky Catalog (PNSC):i.e.a complete photometric and astrometric catalogue of all objects present on the Second Palomar Sky Survey (POSS-II)for an estimated number of 2×109stars and 5×107galaxies.

The data processing pipeline consists of three steps:

?digitalization of the plates,performed at ST-ScI with 15μ(=1arcsec)pixels.STScI provides also the astrometric plate solution (within 0.5arcsec accuracy);

?plate processing and catalogs construction,performed with the SKICAT software (Weir et al.1995a)developed at Caltech;

Figure1:Comparison between POSS-I(left)and POSS-II(right)images of galaxies in the Coma cluster direction

?photometric calibration of the plates through the aquisition of CCD frames of galaxy rich ?elds.

Details on the expected scienti?c results and aims are given in Djorgovski et al.(1997);in this paper we shall focus on some aspects of the data quality and on some speci?c applications. 2The data

POSS-II(Reid et al.1991)covers all the Northern Sky(including the celestial Equator)and consists of894plates(each6.5×6.5sq.deg.)in three colors(B J,R F and I N)spaced by5 degree on the sky,i.e.with an overlap(0.75deg.on each side)between adiacent?elds,much larger than for the POSS-I plates(which were taken only in two colors and with a spacing between plate centers of about6deg).The large overlap regions ensure several advantages:40 %of the sky is registered at least twice in three colors;better photometric matching between adiacent plates can be obtained,and most of the sky is imaged in the central unvignetted part of the plates.Due to the improved sensitivity and resolution of the emulsion,to the use of hypersensitization and long exposure times,magnitude limits of the POSS-II material are typically22.5,20.8and19.5in the B J,R F and I N respectively(Reid et al.1991);i.e. about1mag fainter than for the POSS-I plates.Furthermore POSS-II plates are provided with calibration spots(missing on POSS-I),thus allowing density-intensity conversion of the data. Figure1gives a visual impression of the di?erence between POSS-I and POSS-II images for the same Coma cluster?eld.

Absolute photometric calibration of the plate material is obtained through CCD photometry of at least two?elds per plate,taken in fair seeing conditions(better than2arcsec)in the g, r and i bands of the Thuan–Gunn photometric system(Thuan&Gunn1976,Kent1985). Being CRONA mainly aimed to the study of the extragalactic universe,CCD?elds are selected to contain a fairly large number of galaxies,in order to attain more accurate photometric calibration at low light levels,and to minimize color mismatch and reciprocity e?ects.Selection

of target?elds is performed either from the Abell and Zwicky catalogues(Abell1958,Zwicky 1961-1968)and/or from the CRONA catalogues themselves.

There is some mismatch between the plate and CCD photometric systems:F and N re-sponse plates are reasonably matched by the r an i,while the J is larger and bluer than the g band and some color term correction needs to be taken into account.So far,CCD observations are made at the1.5m Palomar and Loiano(Bologna)telescopes,at ESO(an equatorial zone,δ<10degrees,10nights allocated)and soon at the new1.5m TT1telescope in Southern Italy. The achieved photometric accuracy,both across and within the plates and including both pho-tometric and systematic errors,ranges from0.05mag at the bright end to0.2at r～21.5for stellar objects(Djorgovski et al.1997).For extended objects,photometric accuracy is typically 50%worse than for stellar ones.Such low values result from both the proper managing of the plate data and from the strong requirements on the quality of CCD calibration images.

Plates are currently being reduced at three center,Caltech,Naples and Rome,with identical machines and software(SKICAT).After the arrival of one of us(RdC)to the Observatorio Nacional in Rio de Janeiro,also this institute will soon join in.

Due to vignetting,non linearity of the plate response,position dependence of the point spread function and large space disk occupation of the data(1GB per plate),plate reduction is

a tricky business.For detailed discussions,see Fayyad et al.(1996)and Weir et al.(1995a and

b).Few items need however to be stressed here.The object detection algorithm in SKICAT is an improved version of FOCAS(Jarvis&Tyson1979)which allows the use of a?xed?ux detection threshold for?elds in which the noise is not uniform.Star/galaxy classi?cation is performed from the catalogues via decision trees trained on the CCD calibration frames(Weir et al.1995a).In order to take into account the variation of the PSF shape as a function of the position on the plate,the object classi?cation is performed locally-id est in each”small”portion of the plate.At20mag in r F the galaxy catalogue is90%complete with a20% contamination from stars(Weir et al1995b).

The true bottle neck of the project is in the allocation of telescope time for the aquisition of the CCD calibration frames.It is reasonable to expect,however,that the whole set of plates will be calibrated before year2000.

The main goals of the CRONA collaboration are,besides the completion of the PNSC itself, i)the study of large scale structure via the compilation of an objective catalogue of putative galaxy clusters;ii)the derivation of accurate lumnosity functions for a large sample of galaxy clusters;iii)the search for high redshift QSO’s(in the range of z3.8-4.3,Kenne?ck et al.1995); iv)cross correlation of the CRONA catalogues with catalogues obtained at other wavelenghts.

A variety of interesting results can however be obtained also in other?elds of astrophysical research.

3Color–magnitude diagram of M92

Some further indications on the quality of the POSS-II data can be inferred from the study of the color–magnitude diagram(CMD)of the globular cluster M92(NGC6341).In Fig.2 we show the uncalibrated(F,J?F)diagram for the cluster M92obtained from the J and F plates of the POSS-II?eld n.278.The left and the central panels refer to an annular region comprised between5and10arcmin from the center of the cluster(the region below 5arcmin is completely overexposed and overcrowded and no information can be extracted). The central and right panels show the CMD build from the SKICAT core magnitudes,de?ned as aperture photometry in a square box of3′′×3′′.In order to evaluate the quality of the POSS-II data and of the SKICAT outputs we also derived a CMD of M92using DAOPHOT

Figure2:Left and central panel:comparison between the(F,J?F)CMD obtained with

DAOPHOT(left)and with SKICAT(center)in the radial range5÷10arcmin.The right

(Stetson1987)aperture photometry.The left panel in Fig.2gives the DAOPHOT CMD in the same radial range of the central panel(5÷10arcmin).The DAOPHOT CMD required an interactive de?nition of the PSF for each plate after that the plates had been linearized using the density–intensity calibration curve provided by SKICAT.The positions of the objects?tted by ALLSTAR were then used to derive aperture magnitudes for each star.Since DAOPHOT calculates aperture magnitudes in circular spot,we used a circle having the same area of the square box used by SKICAT in deriving core magnitudes.

When comparing the left and central panel of Fig.2,few points are worth noticing:?rst,the two CMDs appears to be almost similar with a scatter of the data1mag below the turno?point (at F～20.5)identical for the two plots;second,there are more objects in the DAOPHOT plot;third,the SKICAT CMD appears to be cleaner than the DAOPHOT one.We have to stress that DAOPHOT magnitudes have been obtained after some interactive work and some ?ne tuning on images already linearized by SKICAT,whereas SKICAT magnitudes come from the stardard output of the automatic procedure.

The larger number of objects detected by DAOPHOT is mainly due to:contamination from galaxies(which are automatically removed by SKICAT),mismatching of stars(SKICAT matchs the plates using the ST-ScI astrometric solution of the plates whereas DAOPHOT data were matched in pixel space),a deeper detection threshold used for DAOPHOT,and a slightly better e?ectiveness of DAOPHOT in dealing with crowded?elds.The cleaner apparence of the SKICAT CMD derives,instead,mainly by the lower contamination by mismatching and by the rejection of non-stellar object(galaxies,plate scratches,etc.).

The right panel of Fig.2shows the CMD for an annular region centered on the cluster and comprised between50and60arcmin.At such large distances from the center of the cluster, most objects are likely galactic stars with only a small fraction belonging to the cluster.The large statistics allows an accurate evaluation of the background and,therefore,permits the derivation of star counts up to large clustercentric radii as well as higher accuracy in deriving the outer isoplets and the tidal radius of the cluster.

A second hint on the quality of the data comes from the thickness of the cluster MS. Accordingly to Renzini and Fusi Pecci(1988)the intrinsic thickness of the MS is expected to be negligible and therefore the observed thickness(?0.16mag at F=20.5,i.e.～1mag below the turno?point)may be attributed to galactic contamination and observational errors only.

A third and last hint on the quality of the data comes from the comparison of the SKICAT CMD with those derived by other authors who have performed similar work.The SKICAT CMD of M92is deeper and looks better de?ned and thinner than those derived for other nearby globular clusters by Grillmair et al.(1995)or by Lehmann and Scholz(1997)who used UKST and Tautenburg plates scanned with APM.

As it was to be expected the SKICAT CMD looks noiser and less deep than that of M55, measured by Zaggia et al.(1997)who made use of a mosaic of CCD frames taken at the3.5m ESO-NTT.These authors,however,could cover only a quadrant of the cluster and push their data only out to1.5times its tidal radius without reaching the background.This undermines the detection of those tidal tails and di?use extra-tidal stars which were?rst discovered,for other clusters,by Grillmair et al.(1995).

POSS-II data extracted by SKICAT allow to circumvent the present CCD?eld of view limitations and a systematic study of the outer haloes of all globular clusters present on the POSS-II plates is under progress.

References

[1]Abell G.,1958,Astroph.J.Suppl.Ser.3,211

[2]Djorgovski S.,Lasker B.,Weir N.,Potsman M.,Reid I.and Leidler W.,1992,Bul.Amer.

Astr.Soc.,24,750

[3]Djorgovski S,de Carvalho R.,Gal R.et al.1997,IAU Symp.179,in press,eds.B.McLeon

et al.(Dortrech:Kluwer),(astro-ph/9612108)

[4]Fayyad,U.,Djorgovski,S.,and Weir,N.1997,AI magazine,in press

[5]Grillmair C.,Freeman K.,Irwin M.,Quinn P.,1995,Astron.J.109,2553

[6]Jarvis J.,Tyson J.,1979,SPIE Proc.on Instrum.in Astron.172,422

[7]Kenne?ck J.,de Carvalho R.,Djorgovski S.,Wilber M.,Dickson E.,Weir N.,1995,Astron.

J.110,78

[8]Kent S.,1985,PASP97,165

[9]Lehmann I.,and Scholz R.D.,1997,Astron.&Astroph.,in pubblication

[10]Reid I.,Brewer R.,Brucato W.,et al.1991,PASP103,661

[11]Renzini A.,Fusi Pecci F.,1988,Ann.Rev.Astron.&Astroph.26,199

[12]Stetson,P.,1987,PASP,99,191

[13]Thuan T.,Gunn J.,1976,PASP88,543

[14]Weir N.,Fayyad U.,Djorgovski S.,Roden J.,1995a,PASP107,1243

[15]Weir N.,Djorgovski S.,Fayyad U.,1995b,Astron.J.110,1

[16]Zaggia S.,Piotto G.,Capaccioli M.,1997,Astron.&Astroph.,submitted

[17]Zwicky F.et al.,1961-1968,Catalogue of Galaxies and Cluster of Galaxies,

(Pasadena:Caltech)