S. Mottola, G. De Angelis, M. Di Martino, A. Erikson, G. Hahn, G. Neukum
Submitted to Icarus
Revised version
Apr 25, 1995
In the framework of our project for the establishment of a follow-up station to be integrated into the future European network for the search for Near Earth Objects (EUNEASO), we have carried out photometry and astrometry of 8 Near Earth Asteroids and two Mars crossers.
From a total of 42 single-night lightcurves we have determined for the first time accurate rotation periods and amplitudes for eight objects (1981, 2062, 4953, 5653, 5836, 1993 BX3, 1994 JF1, 1994 LC1) . In one case (5797) we have refined the previous determinations. For the asteroid 1994 AW1 we find an indication for a possible complex rotation state.
Interplanetary bodies traveling through near-Earth space (near-Earth objects - NEOs) are of particular interest for several reasons: in the first place their close approaches to the Earth make it possible to study them in more detail than asteroids of comparable size located in the main-belt; in the second place, over geologic times they play an important role in the evolution of the Earth's surface and biosphere because of impact-triggered events like mass extinctions (e.g. Alvarez et al. 1980); furthermore, as potential sources for raw material, they constitute a major asset for future space exploration, and in the short term represent easily accessible targets for reconnaissance space missions.
Currently some 300 NEO have been discovered, defined as asteroids and comets the orbits of which cross or approach relatively closely that of the Earth (see e.g. Shoemaker et al. (1979) and Milani et al. (1989) for the definition of various classes of NEOs). According to the current estimates (Morrison, 1992) there are around 2000 Earth-crossing asteroids larger than 1 km in diameter, of which about 5% have been found, and several hundred thousand in the size range down to 100 m. Realizing their importance, dedicated search programs have been performed since the 1970's (Helin and Dunbar, 1990), and intensified in recent years using CCD detectors (Scotti, 1994). Massive survey programs are proposed and planned for the near future (Morrison 1992, Hahn and Maury 1992).
The rapidly increasing number of new discoveries urges also intensified observing programs in order to determine the basic physical parameters of NEOs and to classify them and select particularly interesting ones for further scrutiny. Photometric lightcurve measurements allow the determination of the rotation period, the spin axis orientation, and give us information on the overall shape of the object. Broadband color data can be used for surface classification and give an indication of the mineralogical properties.
We recently initiated a project which is aimed at monitoring NEOs on a long-term basis, by observing both objects for which the orbits are already well established, and, as target of opportunity, objects which are newly discovered. The goal is to use observational techniques, mainly CCD photometry and astrometry, to contribute to the build-up of a data base of physical properties of NEOs, and to the determination and/or improvement of their orbital parameters.
This paper reports the first results of the project. We obtained lightcurve parameters for 8 NEOs and two Mars crossers, for most of which we also derived astrometric positions.
The main goal of this first series of observations was to test the suitability of various site/telescope/instrumentation combinations (see Table I) and to develop observing strategies suited to the usually fast-moving NEOs. Many of the results presented here have been derived from observations of targets of opportunity, during campaigns originally planned for other objects.
Different observing strategies have been explored in order to maximize the scientific return. In particular the use of differential photometry at poorer observing sites has proved to be fully satisfactory, allowing to derive rotational periods and amplitudes even under moderate weather conditions (cf. Wisniewski and McMillan, 1987). Furthermore, the use of a large-area CCD allowed us to measure astrometric positions on the photometric frames, by using astrometric standard stars present in the same field as the asteroid.
Particular care has been paid during the planning and realization of the observations in order to avoid any observational bias that could favourably select asteroids with short rotational periods and large amplitudes. Instrumental to that was the possibility offered by the acquisition system to perform the photometric reduction on-line or right after the observation, in order to determine as soon as possible the characteristics of the lightcurve and tune consequently the observing strategy. Asteroids which showed a long rotation period were given a higher priority, and were observed (compatibly with the allotted telescope time) during the following nights until the complete coverage of the rotational period was achieved. The sampling frequency was also reduced, in order to avoid unnecessary oversampling and allow the observation of several objects in the same night by cycling between them.
Furthermore we decided to observe only asteroids for which it was possible to achieve a signal to noise ratio of at least 100 with the longest possible exposure time. This would allow us to reliably detect lightcurve variations of only a few hundredths of a magnitude.
The measurements were performed in the Johnson V or R band. The photoelectric observations were performed setting the diaphragm to an aperture size of 15 arcsec. Five asteroid measurements were interspersed with a sky background measurement and the extinction was determined by observing every night groups of standard stars taken from the E-regions of Graham (1982) at different airmasses. The reduction of the photoelectric measurements was performed using the SNOPY program developed at ESO.
In the case of the CCD observations, the adopted exposure times were chosen according to the brightness of the object, and ranged typically between 3 and 8 minutes. A large number of suitable comparison stars was typically present every night in the same CCD field of the asteroid. The telescope was usually tracked at the asteroid rate, thereby reducing the image profile for the asteroid. This allowed us to reduce the aperture size, thus minimizing the background noise in the asteroid aperture. The elongation resulting on the star images was taken into account by increasing the size of the integration aperture during the data reduction. During nights of good photometric quality, stars from Landolt (1983) and from the GSPC-1 catalog (Lasker, Sturch, et al. 1988) have been measured to tie the observations to the Johnson system.
Dark fields acquired through the night and flat fields of the sky taken at dusk and dawn have been used for the frame calibration. The photometric reduction of the CCD frames was performed by using ASTPHOT, a synthetic aperture photometry package developed at DLR (Mottola priv. comm.).
Differential photometry between the asteroid and the field stars available in the frame resulted in formal 1-s errors typically of the order of 0.01 mag. The uncertainties in the determination of the extinction and zero-point coefficients however, dominate the error budget for the absolute photometry, resulting in final error bars of about 0.03 mag. No attempt was made to measure photometric standard stars in nights that were less than photometric.
The individual lightcurves have been corrected for light-travel time and the magnitudes have been reduced to unit distances from the Earth and from the Sun. The use of Fourier analysis, in the way described by Harris et al. (1989), allowed us to determine the rotational periods of the observed objects, which have been used to generate the composite lightcurves.
The ASTPHOT reduction package also allows us to interactively measure positions of objects within a CCD frame with typical accuracies of one tenth of a pixel. By using astrometric standard stars present in the CCD field it is then possible to calculate the celestial coordinates of the observed asteroid. A least square fit of the measured positions of the standard stars is used to derive plate constants for the CCD frame, which in turn allow us to calculate the celestial coordinates of the asteroid.
As source for the reference stars, the Guide Star Catalog (Lasker et al. 1990) was taken. Although this catalog contains no information about the star proper motion, it yields a fairly good internal accuracy (less than one arcsecond), although somewhat varying in different areas of the sky. The big advantage of the GSC lies in its large number of (faint) stars which makes it possible to find a suitable number of stars even on the comparatively small CCD frames (see also Pravec et al., 1994).
In various campaigns we have been observing 10 objects. The observational circumstances and aspect data for each night of observation are listed in Table II. The table gives the geocentric ecliptic coordinates, the phase angle, the heliocentric and geocentric distances, and the telescope and instrumentation used. The V(1,a) column represents the observed V magnitude reduced to unit distances from the Earth and from the Sun and averaged over one rotational cycle. As mentioned in the prevoius section, typical errors on V(1,a) are of the order of 0.03 mag. The date of observation is given to the nearest day to the mid-time of the observed lightcurve. The observational results are summarized in Table III. The first two columns contain the number and name of the asteroid or its provisional designation. The third column gives the orbit classification by Shoemaker et al. (1979). The absolute magnitude H has been calculated by assuming a nominal value G=0.15, as conventionally done by the Minor Planet Center for every new discovery. Only for 2062 Aten, for which the taxonomic class is known, we assumed a phase coefficient G=0.22, which is typical for an S-type asteroid. The diameter of the asteroid has been estimated by using an albedo p=0.20 for 2062 and p=0.12 for all the others (chosen as an unweighted mean of typical C and S-type albedos). Since this estimate is very tentative, the presented diameters should be considered only indicative, and therefore are presented without error bars. The derived rotational periods are given together with their estimated uncertanties, and are follwed by the maximum lightcurve amplitudes. For those lightcurves which do not cover the whole rotational cycle, a lower limit for the amplitude is given. The last column gives the quality code of the period as defined by Lagerkvist et al. (1989). In the following section the individual objects are discussed.
1981 Midas
This high-inclination (i = 40º) Apollo-type asteroid made its most recent close approach to the Earth in March 1992. During this period we obtained photoelectric photometry observations during five consecutive nights. As can be seen from Table II, both the solar phase angle and the geocentric longitude changed by almost 20º during the observations. We have therefore chosen to show three different composite lightcurves (figures 1, 2 and 3), all based on a rotation period of 5.22 hours, to illustrate the rapid change in the shape of the lightcurve. During the course of a few days the amplitude more than doubled, indicating that Midas must have quite an elongated shape, which has been observed under varying viewing and illumination conditions.
On the night of March 7th, UBVRI photometry was performed, producing the following color indices: U-B= 0.48±0.07; B-V=0.97±0.03; V-R=0.49±0.03; V-I= 0.32±0.03. The B-V and V-B color indices fall within the range of taxonomic class S.
Figure 1. Composite lightcurve of 1981 Midas at an average phase angle of 73 deg
Figure 2. Composite lightcurve of 1981 Midas at an average
phase angle of 78 deg
Figure 3. Composite lightcurve of 1981 Midas at an average
phase angle of 85 deg
2062 Aten
The particular geometry of the orbit of this object causes close encounters with the Earth to happen roughly every twenty years and last about five years. We are now in such a period, the previous having been around the time of the discovery of Aten in 1976. Since it was then the first object of the newly defined Aten-class, a series of papers have been published, reporting on the orbital and some physical characteristics (Helin and Shoemaker, 1977; Marsden and Williams, 1977; Veeder et al., 1977; Cruikshank and Jones, 1977). Despite several attempts (Gradie, 1976) no lightcurve variations were observed during the discovery apparition.
We observed Aten during five nights in Nov. 1993 when the asteroid was some 0.3 AU away and about 16.5 mag. in V. A considerable lightcurve amplitude was detected, amounting to at least 0.25 mag in the night-to-night variations, and a clear indication for Aten being a slow rotator. A Fourier analysis search for the most likely rotation period yielded a value of 40.77 hours. In Fig. 4 a composite lightcurve based on such a period is presented. This composite gives the best fit to our data points, although there is only moderate overlap between the different nights. Shorter periods have been ruled out since those would imply a single maximum-minimum lightcurve. Rotation periods in eccess of 2 days produce a much poorer fit to the observed data.
Figure 4. Composite lightcurve of 2062 Aten
4953 1990 MU
This Apollo asteroid has been observed with the Bochum telescope about one month after the June 1994 close approach with the Earth. Four nights of observations have allowed us to reliably determine a unique rotational period of 14.218 hours and an amplitude of 0.68 mag, despite the lack of coverage of one of the lightcurve maxima. A composite based on this period is shown in Fig. 5.
Figure 5. Composite lightcurve of 4953 1990 MU
5653 1992 WD5
This Amor-type asteroid has been observed shortly after discovery as a target of opportunity during a test campaign at Kvistaberg, Sweden. Both astrometric and photometric good-quality data could be obtained despite the prevailing poor wheather conditions at the telescope. From three nights of observations an unambiguous rotation period of 4.8341 hours was derived, with an amplitude of 0.85 mag., as can be seen in the composite lightcurve (Fig. 6). Astrometric positions were determined from CCD frames containing 4 to 10 GSC-stars. These positions, published elsewhere (MPC 22127), were used to improve the orbit of the asteroid.
Figure 6. Composite lightcurve of 5653 1992 WD5
5797 1980 AA
This long-lost Amor asteroid has been recovered at the time of our observing run in Kvistaberg, and was therefore chosen as a target of opportunity. Actually it was independently discovered as 1993 BC2 (IAU Circ. 5695) and shortly thereafter found to be identical with 1980 AA (IAU Circ. 5697). Observations published by Harris and Young (1989), collected during the discovery apparition, yielded a rotation period Psyn = 2.697 hours. Our observations during two nights on Jan 26 and 28, 1993 confirmed the rather short rotation period and also revealed a similar lightcurve amplitude to that observed in 1980. This was to be expected since the aspect conditions were about the same.
When combining our data into a composite, we have been searching for a rotation period that provided a good match for both the old and new data sets. This resulted in a best-fit period of 2.706 hours, which compares well with the one from Harris and Young (1989). This rotation period has been used to plot the composite in Fig. 7.
Figure 7. Composite lightcurve of 5797 1980 AA
5836 1993 MF
This rather large Amor-type NEO was discovered in July 1993 and because of its brightness (V=13.5) and very favourable viewing geometry it was observable over a period of several months. We obtained two nights of observations with the 90cm Dutch telescope at ESO in Nov. 1993, when the object had faded to about 16.5 mag. During one night we covered more than one rotation cycle, and in the following night we could confirm the period. We found a value for Psyn = 4.959 hours and a lightcurve amplitude of 0.76 mag., as can be seen in the composite shown in Fig. 8.
Figure 8. Composite lightcurve of 5836 1993 MF
1993 BX 3
This border-line Apollo-type object (perihelion distance q = 1.003 AU) has been observed as a target of opportunity shortly after its discovery in Feb. 1993. During five consecutive nights we obtained lightcurves which showed large nightly variations of up to 0.5 mag. A period search indicated a value slightly larger than 20 hours. The composite lightcurve shown in Fig. 9 presents the best fit for our data, using a synodic rotation period of 20.463 hours, with an amplitude of 0.91 mag. Again we discarded shorter periods on the assumption of two maxima-minima model.
Figure 9. Composite lightcurve of 1993 BX3
1994 AW1
We observed this Amor asteroid shortly after discovery for a total of about 33 hours over 8 nights. After a few nights of observation it clearly appeared that this object exhibited a complex lightcurve. In fact on some nights the asteroid displayed light varations characterized by low amplitude (~0.10 mag) and a periodicity of about 2.5 hours, and on some other nights it showed much larger variations with apparent periodicities of the order of about 11 hours (or multiple of that). The best match to all lightcurves we were able to produce is with a period of 11.194 hours (Fig. 10). However we do not believe that this is the correct period, since the fine structure of the individual lightcurves is not matched and the resulting scatter of the data points is is about 5 times larger than one would expect from signal-to-noise estimations. (We have to emphasize here that signal-to-noise estimations for all of the other lightcurve observations presented here are in excellent agreement with the actual scatter of the data points measured on the composite lightcurves). We were also able to produce a composite resulting in a good fit for the nights of the 12, 13,16 and 19 Feb. 1994, based on a rotation period of 2.52 hours (see Fig. 11). Even though this it is not likely to be the right rotation period, since it is not compatible with the whole data set, the composite matches the four lightcurves in a accurate fashion, giving us additional confidence that the high frequency variations we observe in the individual lightcurves are real and not due to spurious effects. Given the impossibility to reconcile the whole data set with a single rotation period, we suggest that this asteroid is in a complex rotation state with a fundamental period of about 2.5 hours, with a superimposed modulation of about 11 hours. Following Harris (1994) we can derive an order-of-magnitude estimation of the damping time necessary for an asteroid of this size, in order to relax into a simple rotation around the axis of principal momentum of inertia. Assuming a period of 2.5 hours and a diameter of 1.2 km we obtain a damping time of the order of 3 million years. Although this time is very short if compared with the age of the Solar System, it could represent a significant fraction of the average time between disrupting collisions for an asteroid of this size (Harris 1994, and references therein). This would imply that it is not impossible that we are really observing a case of excited rotation state.
Figure 10. Composite lightcurve of 1994 with AW1 with a period of 11.194 hours. Although this is the best match to the whole data set, we believe that this is not the correct period (see discussion in the text).
Figure 11. A rotation period of 2.5206 hours produces a good fit to a set of four single-night lightcurves. We interpret this fact as the probable indication of a complex rotation state, with a fundamental period around 2.5 hours modulated by a period around 11 hours.
1994 JF1
We observed this Mars-crossing asteroid in June from the Dutch telescope at ESO. Already from the first night of observation it was clear that the rotation period was very long. For this reason we continued to observe it as long as possible, compatibly with the allotted telescope time. Even so, five nights of observations were not enough to cover the whole rotation cycle. However, by assuming a two-maxima and two-minima shape for the lightcurve we were able to derive a unique rotational period of 50.6 hours and set a lower limit for the lightcurve amplitude (see Fig. 12).
Figure 12. Composite lightcurve of 1994 JF1
1994 LC1
This Mars-crossing asteroid was observed during a close approach to the Earth, about two months after discovery. Three nights of observations allowed us to determine a unique rotation period of 2.406 hours and an amplitude of 0.13 mag as shown in Fig. 13. This asteroid is one of the fastest rotators ever observed, following closely 1566 Icarus (Gehrels et al. 1970) and 1866 Sisyphus (Schober et al. 1993).
Figure 13. Composite lightcurve of 1994 LC1
We would like to thank the Uppsala Astronomical Observatory for their provision of the Kvistaberg Schmidt telescope, and in particular Prof. Tarmo Oja and Per-Olow Karlsson for their kind help and assistance. We would also like to thank the assistance of Egon Braatz and of the ESO staff during the observations. We are grateful to Per Magnusson for useful discussions.
Alvarez, L.W., W. Alvarez, F. Asaro, and H.V. Michel 1980. Extraterrestrial Cause for the Cretaceous-Tertiary Extinction. Science 208, 1095-1108.
Cruikshank, D. P. and T. J. Jones 1977. The diameter and albedo of asteroid 1976 AA. Icarus 31, 427-429.
Gehrels T., E. Roemer, R.C. Taylor, and B. H. Zellner 1970. Minor Planets and related objects. IV. Asteroid (1566) Icarus. Astron. J. 75, 186-195.
Gradie, J. C. 1976. Physical observations of object 1976 AA. Bull. Amer. Astron. Soc. 8, 458-459 (abstract).
Graham, J. A. 1982. UBVRI Standard stars in the E-Regions. PASP 94, 244-265.
Hahn, G., and A. Maury 1992. Project EUNEASO: an initiative for a EUropean Near-Earth Asteroid Search Observatory. In Observations and Physical Properties of Small Solar System Bodies. Proceedings of the 30th Liège International Astrophysical Colloquium (A. Brahic, J.-C. Gerard, and J. Surdej, Eds.), pp. 239-242. Université de Liège, Liège (Belgique)
Harris, A.W. J.W. Young, E. Bowell, L. J. Martin, R. L. Millis, M. Poutanen, F. Scaltriti, V. Zappalà, H.-J. Schober, H. Debehogne, and K. W. Zeigler 1989. Photoelectric observations of asteroids 3, 24, 60, 261, and 863. Icarus 77, 171-186.
Harris, A.W., and J.W. Young 1989. Asteroid lightcurves: 1979 - 1981 observations. Icarus 81, 314-364.
Harris, A.W. 1994. Tumbling asteroids. Icarus 107, 209-211.
Helin, E.F., and R.S. Dunbar 1990. Search Techniques for Near-Earth Asteroids. Vistas in Astronomy, 33, 21-37.
Helin, E.F. and E.M. Shoemaker 1977. Discovery of asteroid 1976 AA. Icarus 31, 415-419.
Lagerkvist C-I., A. W. Harris, and V. Zappalà. Asteroid lightcurve parameters. In Asteroids II
(R. P. Binzel, T. Gehrels, M. S. Matthews, Eds.), pp. 1162-1179. Univ. of Arizona Press, Tucson.
Landolt, A. U. 1983. UBVRI photometric standard stars around the celestial equator. Astron. J. 88, 439-460.
Lasker, B. M., C. R. Sturch, C. Lopez, A. D. Mallama, S. F. McLaughlin, J. L. Russel, W. Z. Wisniewski, B. A. Gillespie, H. Jenkner, E. D. Siciliano, D. Kenny, J. H. Baumert, A. M. Goldberg, G. W. Henry, E. Kemper, and M. J. Siegel 1988. The Guide Star Photometric Catalog. I. Ap. J. Suppl. 68, 1-90.
Lasker, B. M., C. R. Sturch, B. McLean, J. L. Russel, H. Jenkner, and M. Shara 1990. The Guide Star Catalog. I. Astronomical foundations and image processing. Astron. J. 99, 2019-2058.
Marsden, B.M. and J.G. Williams 1977. Orbit of 1976 AA. Icarus 31, 420-423.
Milani, A., M. Carpino, G. Hahn, and A.M. Nobili 1989. Project SPACEGUARD: dynamics of planet-crossing asteroids. Classes of orbital behavior. Icarus 78, 212-269.
Morrison, D., Ed. 1992. The Spaceguard Survey. Report of the NASA International Near-Earth Object Detection Workshop, NASA, Washington, D.C.
Pravec, P., M. Tichý, J. Tichá 1994, Z. Moravec, Z. Vávrová, M. Velen 1994. CCD astrometry of asteroids and comets using the Guide Star Catalog. Planet. Space Sci. 42, 345-348.
Scotti, J.V. 1994. Computer aided Near Earth Object detection. In Asteroids, Comets, Meteors 1993 (A. Milani, M. Di Martino, A. Cellino, Eds.), pp. 17-30. Kluwer Academic Publishers, Dordrecht.
Schober, H.-J., A. Erikson, G. Hahn, and C.-I. Lagerkvist 1993. Physical studies of asteroids XXVI. Rotation Periods and photoelectric photometry of asteroids 323, 350, 582, 1021 and 1866. Astron. Astrophys. Suppl. 101, 499-505.
Shoemaker, E.M., J.G. Williams, E.F. Helin, and R.F. Wolfe 1979. Earth-crossing asteroids: Orbital classes, collision rates with Earth and origin. In Asteroids (T. Gehrels, Ed.), pp. 253-282. Univ. of Arizona Press, Tucson.
Veeder, G.J., D.L. Matson, and O.L. Hansen 1977. Photometry of the asteroid 1976 AA at 0.56 and 2.2 µm. Icarus 31, 424-426.
Wisniewski, W.Z. and R.S. McMillan 1987. Differential CCD photometry of faint asteroids in crowded star fields and nonphotometric sky conditions. Astron. J. 93, 1264-1267.