Mike Kretlow
Home > Oldstuff (Archive) > Occultations > About Star Catalogs

About Star Catalogs

Though the major part of the prediction uncertainty comes nowadays (usually) from the SSO (= Solar System Object) ephemeris uncertainty (before the availability of Hipparcos, Tycho etc. catalogs more or less ~50% of the uncertainty came from the catalog) it is of course desirable to have as accurate star data as possible.  Going a little bit deeper into this subject it would be also the best if the prediction catalog is the same (or refers to the same system like the ICRS) as the catalog used for the astrometric reduction of asteroid observations which are again used for the computation of an (improved/updated) orbit. This ensures that both the ephemeris and the star position are within the same reference frame (zero point etc.), bypassing or at least reducing any systematic differences which lead into larger (and maybe not recognized) prediction uncertainties. But in practice the 'bias' introduced by such disagreements is one or two order of magnitudes smaller than the primary error sources.

The basic informations about these catalogs (mainly grabbed from the documantation or primary web site) are summarized below. The order of appearance reflects also (more or less) my personal recommendations using these catalogs for asteroidal occultation predictions.

Of course there might be special applications for which this 'order' could/should be changed, e.g. TNO occultations are not so frequently and the uncertainty is usually rather large. One possible strategy could be to increase dramatically the number of predicted events accepting an also increasing uncertainty because less accurate but more comprehensive catalogs (like USNO-B1.0) are used for that. This large event list could then be used for some 'brute force' observing method, i.e. dedicating an automatic telescope to observe each of this predicted events (automatic systems get not frustrated if they observe hundreds of events with an success rate of maybe 1%...).


CatalogReported OCCPositive OCC(O+)SR
UCAC4 72 10 14%
UCAC3 313 11 04%
UCAC2 690 141 20%
TYC2 1790 355 20%
HIP 292 68 23%
GSC 49 3 06%
PPM 128 18 14%
TAC 52 6 12%

 SR: Overall Success Rate = OCC(O+) / OCC(total)
 From 3386 reported asteroidal occultations 1997-2013.

This table was generated using all archived (negative and positive) occultations (events) and while it reflects somehow the experience about the accuracy of star catalogs for occultation work it is not a real 'hard' statistic and you should consider all the circumstances biasing the quality and success prospects of an occultation prediction (now and in the past) like orbit updates etc..




FK6 and ARIHIP - Released 2000 and 2001

The FK6 created and published by ARI (Astronomisches Recheninstitut) combines the results of the astrometric satellite Hipparcos with the data of the "Fifth Fundamental Catalogue (FK5)". The result contains, in particular for the proper motions, statistically significant improvements of the Hipparcos data und represents a system of unprecedented accuracy for these 4150 fundamental stars. The typical mean error in pm is 0.35 mas/year for 878 basic stars, and 0.59 mas/year for the sample of the 3272 additional stars.

As in the case of the FK6, (using ground based data provided by the FK5 to improve Hipparcos' results), the Boss  General Cat. (GC) was used to increase the number of stars with such better data: The combination of HIPPARCOS and GC data gives improved results for 20 069 stars. This catalogue - named GC+HIP - gives proper motions (pm) with a typical mean error of 0.66 mas/year (about a factor of 1.2 better than Hipparcos for this stars). Then the proper motions of the Tycho-2 Catalogue were used to improve the pm for an additional huge number of Hipparcos stars. This catalogue giving pm with a typical mean error of 0.83 mas/year (about a factor of 1.3 better than Hipparcos for this sample) is called TYC2+HIP. The FK6 Part I is providing pm with a typical mean error of 0.35 mas/year, the FK6 part III with a typical mean of 0.59 mas/year.

The catalogue ARIHIP has been constructed by selecting the 'best data' for a given star from the three combination catalogues mentioned above. It provides 'best data' for 90842 stars with a typical mean error of 0.89 mas/year (about a factor of 1.3 better than Hipparcos for this sample of stars).

Personal note: ARIHIP provides the best data currently available for occultation work. Add Tycho-2 stars to go deeper (see below).


Hipparcos and Tycho - Released 1997...2007 and 1997...2000

This catalog contains 118218 stars that were observed by the European Space Agency's Hipparcos Satellite, operational from late 1989 to 1993. It is complete to V=7.3. The positional accuracy is 1-3 mas at epoch 1991.25 and 1-2 mas/yr for proper motions (pm), respectively. Hipparcos catalog is the standard reference catalog for optical astrometry, representing the ICRF in the optical wavelengths.

Tycho-2 contains about 2.5 million stars within the HIP/ICRF system. The catalog is 99% complete to V=11.0 and 95% complete at V=11.5. Positional accuracies range from about 10 to 100 mas, depending on magnitude. Proper motion accuracies are from 1 to 3 mas.

Personal note: the original Hipparcos data should be replaced by ARIHIP. ARIHIP+Tycho-2 should give the best positional data up to V~10-11.


UCAC2 - Released 2004

The second release of the UCAC (USNO CCD Astrograph Catalog) is an astrometric and photometric catalog of 48.3 million stars covering the sky from -90° to +40° in declination and going up to +52° in some areas. Positions and proper motions are on the ICRS (International Celestial Reference System) and given at the epoch J2000.0.

The UCAC2 star catalog has an observed positional error of about 20mas for 10 mag < R < 14 mag and about 70mas for R_lim ~16, as shown in the left figure (Zacharias et al. AJ 2004).

How to improve V-mag calculation from UCAC2 magnitudes is described on the web page by Bruce L. Gary.

Personal note: still one of the primary input catalogs for occultation predictions by SSOs. Supported in my occultation predictions.

UCAC3 - Released 2009

From the catalog description:

UCAC3 is a compiled, all-sky star catalog covering mainly the 8 to 16 magnitude range in a single bandpass between V and R. Positional errors are about 15 to 20 mas for stars in the 10 to 14 mag range.   It is supplemented by proper motions and SuperCosmos and 2MASS photometric data, as well as various flags.  The proper motions of bright stars are based on about 140 catalogs,  including Hipparcos and Tycho, as well as all catalogs used for the Tycho-2 proper motion construction. Proper motions of faint stars are based on a re-reduction of early epoch SPM data (-90 to -10 deg Dec)  plus Schmidt plate data from the SuperCosmos project (down weighted  due to systematic errors of order 100 mas). The proper motions of faint stars (R >= 13.5) therefore should be used with caution. The unpublished plate measure data from the AGK2, the Hamburg Zone Astrograph, the USNO Black Birch Astrograph, and the Lick Astrograph have considerably contributed to improve proper motions for stars mainly in the 10 to 14 mag range (down to the UCAC3 limit for Lick data); however, these data do not cover all sky. 

Personal note: due to some known issues (false double stars, and distortions in its proper motion system north of -20d declination) with this catalog I do not recommend to use it neither for occultation work nor for astrometry!! Use UCAC4 instead.


UCAC4 - Released 2012

It is the final release in this series and contains over 113 million objects; over 105 million of them with proper motions (PMs). UCAC4 is an updated version of UCAC3 with about the same number of stars also covering all-sky. Bugs were fixed, Schmidt plate survey data were avoided, and precise five-band photometry was added for about half the stars. Astrograph observations have been supplemented for bright stars by FK6, Hipparcos, and Tycho-2 data to compile a UCAC4 star catalog complete from the brightest stars to about magnitude R = 16. Epoch 1998-2004 positions are obtained from observations with the 20 cm aperture USNO Astrograph's "red lens," equipped with a 4k by 4k CCD. Mean positions and PMs are derived by combining these observations with over 140 ground- and space-based catalogs, including Hipparcos/Tycho and the AC2000.2, as well as unpublished measures of over 5000 plates from other astrographs. For most of the faint stars in the southern hemisphere, the first epoch plates from the Southern Proper Motion program form the basis for PMs, while the Northern Proper Motion first epoch plates serve the same purpose for the rest of the sky. These data are supplemented by 2MASS near-IR photometry for about 110 million stars and five-band (B, V, g, r, i) APASS data for over 51 million stars. Thus the published UCAC4, as were UCAC3 and UCAC2, is a compiled catalog with the UCAC observational program being a major component. The positional accuracy of stars in UCAC4 at mean epoch is about 15-100 mas per coordinate, depending on magnitude, while the formal errors in PMs range from about 1 to 10 mas yr–1 depending on magnitude and observing history. Systematic errors in PMs are estimated to be about 1-4 mas yr–1.

Personal note: If UCAC4 has a quality comparable to that UCAC2 had in the past and no errors or distorsions are hidden (and yet not found), then UCAC4 can be considered as one of the best catalog suitable for occultation work in terms of accuracy and deepness, probably together (or in addition) to PPMX,PPMXL. Supported in my occultation predictions.


CMC14 - Released 2006

CMC14 external errors

r' ΔRA
<13 34 33 34 25
14 38 34 36 35
15 52 41 47 70
16 88 63 76 170

The Carlsberg Meridian Catalogue 14  is an astrometric and photometric catalog of 95.9 million stars in  the red (SDSS r') magnitude range 9 to 17 covering the declination range -30° to +50°.

The CMC has no proper motions, but because the observations were made in the period  between March 1999 and October 2005 the positions (on the ICRS) refer to a rather recent epoch.

For the 90842 ARIHIP stars the arithmetic mean p.m. in RA*cos(DE) and DE is 34 mas/yr and 31mas/yr. This gives an idea about the positional bias due to the lacking pm in CMC14.

V-mag estimates from catalog magnitudes has been investigated by Dymock and Miles (R.Dymock & R.Miles: J.Br.Astron.Assoc. 119,3,2009)

Personal note: actual (2010) this catalog can be considered roughly equivalent to the UCAC2 in terms of accuracy/quality. Advantage is the higher star density for the covered regions.


CMC15 - Released 2014

The Carlsberg Meridian Catalogue 15 is the successor of the CMC14 and is destined to be the last of the series "Carlsberg Meridian Catalogue, La Palma". Its is based on observations made between March 1999 and March 2011 (CMC14: 1999-2005). 

It contains more than 122 million observations (overall sky completeness about 95-98%) of right ascension, declination, and magnitude of stars in the magnitude range of 9m < r′ < 17m and declination range of –40° < δ < +50°. The catalogue internal errors in astrometry are below 30 mas in both coordinates for stars brighter than r′ = 13, reaching 60 mas for r′ = 16. The internal magnitude error is below 0.020 mag for stars brighter than r′ = 13, and about 0.090 mag for r′ = 16.

This catalogue fill the gap between 5h 30m and 10h 30m for declinations south of -15º of the CMC14 and add the band -30º
to -40º. Also some zones north of -30º has been re-observed in order to improve its internal errors.

Personal note (2014): this catalog should be used in replacement of the CMC14 in astrometry and possibly in occultation work, though because CMC does not include proper motions it is not the first choice for occultation work. Currently supported in my occultation predictions.


SPM4 - Released 2011

The Southern Proper Motion Catalog 4 contains absolute proper motions, positions, and B,V photometry for over 100 million stars and galaxies, down to a magnitude of V=17.5.  Sky coverage is from declination -20 degrees to the south celestial pole.  Cross references to the 2MASS near-infrared catalog are also included, as is the 2MASS JHK photometry. The absolute proper motions are tied to the ICRS at the bright end, via Hipparcos stars, and to external galaxies at the faint end.  The final precision of the SPM positions and absolute proper motions is approximately 30 to 150 mas and 2 to 10 mas/yr, respectively.  Systematic errors in the proper motions are still being evaluated but are estimated to be on the order of 1 mas/yr.

Personal note: This catalog is probably the best choice for positions and proper motions of southern sky stars for the current epoch and I expect that these data are superior to data from UCAC4 or CMC15. Supported in my occultation predictions.


PPMX - Released 2008

PPM-Extended (PPMX) is a catalogue of 18,088,919 stars containing astrometric and photometric information. Its limiting magnitude is about 15.2 in the GSC photometric system. PPMX consists of three parts: a) a survey complete down to R_U=12.8 in the magnitude system of UCAC2; b) additional stars of high-precision proper motions, and c) all other stars from GSC 1.2 identified in 2MASS. The typical accuracy of the proper motions is 2 mas/y for 66 percent of the survey stars (a) and the high-precision stars (b), and about 10 mas/y for all other stars. PPMX contains photometric information from ASCC-2.5 and 2MASS.

Personal note: Reasonable replacement for the original FK6, HIPPARCOS and Tycho-2 catalogs. Supported in my occultation predictions.


PPMXL - Released 2010

PPMXL contains about 900 million objects, some 410 million with 2MASS photometry, and is the largest collection of ICRS proper motions at present. As representative for the ICRS the PPMX was choosen. USNO-B1.0 and 2MASS are the most widely used full-sky surveys. However, 2MASS has no proper motions at all, and USNO-B1.0 published only relative, not absolute (i.e. on ICRS) proper motions. ARI  performed a new determination of mean positions and proper motions on the ICRS system by combining USNO-B1.0 and 2MASS astrometry. This catalog is called PPMXL , and it aims to be complete from the brightest stars down to about V ~ 20 full-sky.  The resulting typical individual mean errors of the proper motions range from 4 mas/y to more than 10 mas/y depending on observational history. The mean errors of positions at epoch 2000.0 are 80 to 120 mas, if 2MASS astrometry could be used, 150 to 300 mas else. PPMXL (ARI) also gives correction tables to convert USNO-B1.0 observations of e.g. minor planets to the ICRS system.

Personal note: I have yet no experience with this catalog but I plan to add support for this catalog to my occultation program.


XPM - Released 2010

This catalog is still very unknown. It contains absolute proper motions of 280 million stars all over the sky without gaps in the magnitude range 10 mag < V <20 mag and might be thus comparable to PPMXL in terms of deepness. I have no experience with this catalog but as the authors tried to realize with this catalog an ICRS  in the optical range it would be interesting to take a look into this work. Unfortunately this catalog gives no mean errors on the position and proper motions.

Here some papers the authors published: Paper Paper Paper Paper 

Personal note: I have yet no experience with this catalog. Unfortunately there are no error estimates for the positions and proper motions, so I do not plan to include this catalog into my predictions.


URAT1 - Released 2015

The catalog contains ~ 228 million objects (stars and non-star entries) covering about R = 3 to 18.5 mag with positional  precisions of 5 to 40 mas. URAT1 is an observational catalog, i.e. mean positions of stars at mean epoch of URAT observing (between 2012.3 and 2014.6), on the International Celestial Reference System (ICRS) by using UCAC4 as reference in conventional "plate" adjustments. It covers most of the northern hemisphere and in some areas as  far south as -24.8 deg Dec (Pluto field). Proper motions are derived from 2-epoch URAT1 and Two Micron All Sky Survey (2MASS) data (at around epoch 2000) resulting in proper motion errors of typically 5 to 8 mas/yr.

Personal note: It seems that URAT1 provides at the moment the best positions beside ARIHIP (HIP and FK6 stars) for occultation work. Supported in my occultation predictions.


Gaia-DR1 - Released 2016



HSOY - Released 2017

HSOY stands for Hot Stuff for One Year. The HSOY catalogue was compiled using the positions taken from Gaia-DR1 combined with the input data from the PPMXL catalogue, employing the same weighted least-squares technique that was used to assemble the PPMXL catalogue itself. Results. This effort resulted in a four-parameter astrometric catalogue containing 583,000,000 objects, with Gaia-DR1 quality positions and proper motions with precisions from significantly less than 1 mas/yr to 5 mas/yr, depending on the object's brightness and location on the sky. HSOY is described in this paper.

Personal note: This is now (January 2017) the main star catalog used in my occultation predictions.


The following catalogs should not be used for occultation work !

B1.0 - Released 2003

The USNO-B1.0 contains entries (position, pm and mag) for about 1 billion stars/galaxies(down to about V~20) which were detected in the digitized images of several photographic sky surveys. It is estimated that the positional error at current epoch is near 200 mas. USNO-B1.0's proper motions mean epoch is about 1975 and the proper motions are relative to the intermediate catalog YS4.0, not absolute! The present offset wrt the ICRS seems to be about 0.3".

Personal note: I cannot really recommend this catalog for occultation predictions unless systematic and/or local differences wrt the ICRS are better investigated except there is a need for a dense and deep going star catalog. For occultation work a cleaned version from astrometry.net plus B1.0-ICRS corrections (see PPMXL) should be used.


NOMAD - Released 2005

NOMAD is not a 'new' star catalog but rather datas compilation using the 'best' data from these catalogs: Hipparcos, UCAC2, Tycho-2, B1.0 (and some other data sources like 2MASS etc.) For detailed information you should consider the ReadMe file.

Personal note: Do not use this catalog for occultation work or astrometry. It might be useful for other applications, but it is no alternative to the catalogs mentioned before.



Catalog corrections / debiasing

  • Systematic differences HIPPARCOS catalog - FK5 (and older important cataloges) were published by Schwan (ARI, see here).
  • Systematic differences PPMXL (ARI) - USNO B1.0 and UCAC3 respectively were published by Roeser (ARI, see here and here).
  • Debiasing astrometric observations based on USNO A1.0, A2.0, and B1.0 is described by Chesley et al. in this paper .



If the positional accuracy has highest priority I would suggest following:

  • HSOY or Gaia-DR1
  • --- Big gap ---
  • ARIHIP or URAT1 > PPMX or UCAC4 > (PPMXL ?) > UCAC2 for the northern sky (-20 < dec < +90).
  • ARIHIP or URAT1 > SPM4 > PPMX or UCAC4 > (PPMXL ?) > UCAC2 for the southern sky (-90 < dec < -20). 
  • Don't use USNO B1.0, UCAC3 or NOMAD for occultation work.
Last edited: January 29, 2017, 20:42

Powered by CMSimple_XH