ADAS: Asiago-DLR Asteroid Survey

Revision 19 May 2001 ADAS: Asiago-DLR Asteroid Survey Cesare Barbieri 1, Massimo Calvani 2, Gerhard Hahn 3, Stefano Mottola 3,Giuliano Pignata 1, Luciano Salvadori 1 1 Department of Astronomy, University of Padova 2 Astronomical Obervatory of Padova 3 DLR Berlin (Germany) Abstract This report covers the activities of the Asiago DLR Survey of Asteroids (ADAS) since Dec. 20, 2000 to date. Main results are provided. The report is kept regularly updated. Introduction The project to adapt a CCD camera to the S67/92 cm Schmidt telescope at Cima Ekar is a joint collaboration between the Department of Astronomy and the Astronomical Observatory of Padova on one side, and DLR Berlin on the other. The main scientific driver is the discovery and follow up of moving objects (asteroids, NEOs, NEAs, TNOs, KBOs, etc.). Hence the name ADAS: Asiago-DLR Asteroid Survey given to the project. However other scientific programs will be possible: no filter is at moment provided, but a filter wheel device is available and it will be mounted in the near future. DLR has provided the SCAM-1 camera which can be operated both in Time-Delay Integration mode and in normal mode. In TDI, the effective exposure time for each star is of 196s at the equator. Furthermore, DLR has provided the software for image acquisition and quick look, and the routines for photometry and centroiding of all stars on the frame (Sextractor), and for astrometry and automatic detection of moving objects by comparing 3 frames (Rackis), working under Windows NT. The thick front-illuminated CCD is a grade A 2048x2048 LORAL chip with a pixel size of 15x15 µm (1.437x1.437 on the sky), and covers an area of 49x49 arcmin (0.67 sq deg). The camera is equipped with a precision shutter (which in guided mode cuts the light from the very corners of the CCD; this vignetting is not apparent on the TDI frames); the shortest exposure time is 0.1 sec). The chip is refrigerated by a two-stage cooling device, where the primary stage is a Peltier cooler and the secondary one consists of a closed-circuit liquid refrigerator. The achieved CCD operational temperature is 63 C. DA/OA Pd have provided the acquisition computer and the secondary cooling stage. They are also responsible for the mechanical adaptation of the telescope, for the operations, for the data acquisition and analysis, and for the data distribution and archiving. A second computer for on-line data analysis has been installed. The system obtained useful data since the 21 st of December 2000. Till the middle of February 2001, the focal plane was folded to the CCD camera via a (slightly undersized) flat metal mirror kindly provided by Officine Galileo (Firenze); the mirror is a spin-off of the very 1

successful prototype built for the Halley Multicolour Camera on board GIOTTO, now produced in large quantities for several non-astronomical applications. See Fig. 1. A new flat mirror in glass, with larger dimensions in order to collect all the light beam, and excellent optical quality, was produced by Ottica ZEN (Venezia); it was installed at the telescope the 21 st of February 2001. The coordinates of the Asiago-Cima Ekar S67 telescope are (Barbieri et al., 1992): Astronomical coordinates (East Longitude, Latitude) +11 34 22.19 (+0 h 46 m 17 s.48) +45 50 40.18 Geodetic coordinates +11 34 07.77 (+0 h 46 m 16 s.52) +45 50 58.00 Ellipsoidic elevation: 1410.2 m Elevation above sea level: 1370 m The Minor Planet Center has attributed to ADAS the survey code 209. The old code 098 (Cima Ekar) had the following values: MPC cos = 0.69790 sin = 0.71410 11.5688 JPL 45 51 00.8 N 11 34 09.8 E h = -162716 m (notice the strange value in elevation). The new code 209 ADAS Cima Ekar has: MPC cos = 0.69790 sin = 0.71410 11.5688 JPL 45 50 58.8 N 11 34 07.7 E h = 1395.37 m much closer to the correct values. Several tools for ADAS have been adapted from available software packages. The astrometric residuals are evaluated by a comparison with the asteroids positions (MPC format) in the asteroid server developed by J. Skvarc through the web interface: http://astro.ago.uni-lj.si/asteroids/residuals.html This service uses several programs and information sources developed by different people. The asteroid database is maintained at Lowell Observatory by E. Bowell. Propagation of asteroid positions is done by a program called Orbfit part of a NEO information tool NEODyS. It is developed by the Orbfit consortium. Identification of the 2

asteroids is made using the MPC tool: MPChecker http://cfaps8.harvard.edu/~cgi/checksn?s=m Fig. 1 On the left, the SCAM camera and electronics attached to the side of the S/67-92 cm telescope at Cima Ekar, on the right the OG metal mirror used in the first phase of ADAS till Feb. 2001. 1 The First Phase, 20 Dec. 2000 20 Feb. 2001 The first phase of our work, till the installation of the glass mirror, lasted through Feb. 20, 2001. Although the optical quality had not reached its optimal value, the limiting magnitude was already sufficiently faint to give hope to have a competitive system. For instance, the faintest observed object the very first night was 116262 1998 KN45, V(JPL) =19.94 (remember, we have no filter in front of the CCD, the effective band is essentially V+R), with an exposure time of 80 sec. The night of 15 th and 16 th January images of the recently discovered TNO 2000 WR106, the brightest object of this class, were obtained; at the time of our observations its apparent V magnitude was 19.91. We could well detect it, although not by the automatic software procedure, but by manual processing. Examples of images are given here in Figs. 2 and 3, namely asteroid Citrinella and nebula M42 in Orion Fig. 2 three images for the detection of asteroid 8965 Citrinella (V=19.03), exp. time 80 s each Almost all nights were used in guided mode; the exception was one night were the scan mode was tested. The weather was generally bad, allowing only 13 useful nights from Dec. 21 to Feb. 16. Nevertheless the amount of data produced was already very high. 3

Fig. 3 M42 in Orion, metal mirror, no filter, exp. Time 15 sec. Image processing courtesy of R. Falomo. Table 1 and Fig. 4, Fig 5 provides an indication of the astrometric precision achieved in the first phase by using only numbered asteroids. Actually the intrinsic precision of these first data was higher, and the calculated values were limited by a non optimal implementation of the reduction software. A revision of the procedure with an advanced version of Rackis was made available in the second phase. Table 1 Astrometric quality of the data obtained during the first phase Residuals (arcsec) N of observations Percentage < 0.2 5 2.3 < 0.5 27 12.3 < 1.0 126 57.5 < 2.0 181 82.6 > 2.0 38 17.4 All observations 219 Average RA residual Average DE residual Average total residual -0.69 ± 1.62 arcsec -0.41 ± 0.49 arcsec 1.38 ± 1.27 arcsec 4

Fig. 4 scatter plot of residuals in RA and Dec. Fig. 5 distributions of total residuals (< 2.0 arc sec). For the greatest part of the observations of this first phase, fields near the ecliptic and the meridian were chosen. The observing sequence for the pointed frames has been: 1. expose the first field for 30 to 196 s (the exposure times were changed for testing purposes) move the telescope in Dec by 45 and acquire the new field repeat for a total of 8 or 9 steps, for a total time of 30 to 45 minutes 2. go back to first field and repeat the sequence 3. go back to first field and repeat the sequence 5

In this way each field was imaged 3 times at intervals of 30 (or 45) min. Although we tried to minimize the dead times by reading the CCD (readout time 40s) during the movement of the telescope, it was plainly evident that this procedure was inefficient in terms of square degrees/unit time, approximately 2.7 sq deg/h, although the limiting magnitude is faint. Instead, the nights of the 13 th, 14 th, and 15 th February 2001 the telescope was pointed toward known NEOs in order to test the performances of ADAS also in this application. Table 2 resumes the results. Table 2 Observations of known NEOs object Exp. time (sec) Rate RA Rate DE Rate mag (arcsec/h) (arcsec/h) (arcsec/h) 2001 CC32 100-203.3-229.3 306.4 17.6 2001 CB32 100 271-116.6 214.6 17.1 2001 YM29 100 55.8 64.6 85.3 19.0 2001 CP36 80-1063 128.7 1070.8 16.9 The rates given in Table 2 are that measured on our frames, not necessarily those provided by the various ephemeris generators. Because the fields were selected to contain known NEOs, the statistics of asteroids detected in these fields is low. The magnitude distribution of the 227 objects observed in the first phase (not necessarily different, several asteroids might well have been detected twice) is shown in Fig. 6. The faintest observed magnitude is nominally 20.6 Fig. 6 The histogram shows the magnitude distribution of the detected asteroids 6

Fig. 7 The magnitude error distribution Fig. 7 shows the distribution of the errors on the magnitudes over the 3 frames; in general the magnitude determination is fairly reliable, 96% of the differences are between ( 0.48, +0.48) but occasionally the difference can be as high as 1 mag. Fig. 8 provides the velocity distribution of all asteroids detected in Phase 1. Fig. 8 (55) Pandora is the brightest asteroid till now detected, ASI012 is an unidentified asteroid, perhaps a new discovery To summarize the results obtained in this first part of the program, where we have substantially operated in guided mode, we give the following figures: Total asteroids detections: 227 Covered square degrees: 33.3 Number of detected asteroids per square degrees: 6.8 Smallest detected angular displacement rate: 3.9 arcsec/h. for the asteroid (13407) 1999 TF4 7

Smallest detected angular displacement: 1.6 arcsec On one of the fields, we have been able to detect a known Trojan Jupiter asteroid (7152) Euneus with orbital semi major axis of 5.17 AU 2 The second phase, since 21 Feb. 2001 The second phase of ADAS started on 21 Feb 2001, when the new excellent glass mirror was mounted. The optical quality improved, and also the alignment of the CCD columns with the Hour Angle was optimized, so that the scan mode could be implemented. Fig. 9 M13 with new flat mirror, no filter, exp. Time 60 sec With the TDI technique and 30 min long scans, we can cover a field of 6.15 sq deg for 3 times in 1.7 hours, approximately 3.6 sq deg/h. In winter time (10h observing runs), the total surveyed field will be of 36.0 sq deg; in summer time (6h observing runs) the total surveyed field will be of 21.6 sq deg. A second computer was put in parallel with that for acquisition, in order to speed up the data reduction, which can now be performed in parallel with the acquisition. Here are the results obtained from Feb. 21 st to date: total number of found asteroids: 469 covered area: 48.1 sq. deg asteroids per sq. deg: 9.7 smallest angular displacement rate: 4.5 arcsec/h. smallest angular displacement: 2.5 arcsec Beyond the optical quality, a decisive factor for the improvement in efficiency has been the implementation of version 4.02 of Rackis and a better tuning of the parameters. 8

Table 3 The astrometric quality obtained in the second phase Residuals (arcsec) N of observations Percentage < 0.2 52 14.3 % < 0.5 189 52.1 % < 1.0 319 87.9 % < 2.0 358 98.6 % > 2.0 5 1.4 % All observations 363 Average RA residual Average DE residual Average total residual -0.27 ± 0.54 arcsec 0.15 ± 0.35 arcsec 0.56 ± 0.43 arcsec Table 3 and Fig.10, Fig11 provide a summary of the astrometric precision achieved in the second phase. Fig. 10 The distribution of the astrometric residuals in the second phase 9

Fig. 11 distributions of total residuals (< 2.0 arc sec). Regarding the limiting magnitude, the improvement has not been so noticeable because of the prevailing conditions of bad weather and mediocre seeing. Therefore we show in Fig. 12 the distribution of the magnitudes over the frames where the FWHM was less than 3.5 arcsec (the FWHM is due not only to seeing but to the TDI limitations; in practice already at 15 deg from the equatorial plane the image quality is sensibly deteriorated). Fig. 12 The magnitude distribution of detected asteroids on frames with FWHM < 3.5 arcsec. In this case, 17.7 % of detected asteroids has magnitude > 19.5; the faintest detected asteroid was 2001 FE168, V = 21.0. 10

Fig. 13: distribution of the velocities of asteroids detected in Phase 2 of ADAS Fig. 13 gives the velocity distribution. 2001 BY60 is a Mars crosser. The dynamical classification of these objects is given in Table 4. Table 4 Orbital classification of objects detected in Phase 2 Orbital Classification Nr. ATEN 0 AMOR 0 APOLLO 0 MARS CROSSER 2 HUNGARIAN 0 HILDA 3 TROJANS 3 Table 5 asteroids discovery by ADAS asteroid a (AU) e i (deg) Orbit Orbital clas. 2001 FF154 - - - None - 2001 FG154 3.9329812 0.1264463 3.91503 25-day arc HILDA 2001 FT167 2.6551083 0.3498727 7.39449 8-day arc MAIN BELT 2001 FU167 - - - None - 2001 FW167 - - - None - 2001 FN168 2.7746620 0.0366720 4.88842 31-day arc MAIN BELT 2001 FP168 - - - None - 2001 FR169 2.1944128 0.1509959 2.24936 4 opps 1992-2001 MAIN BELT 2001 FS169 - - - None - 2001 FT169 2.7755051 0.11407752 9.94469 7-day arc - These are only preliminary designation, In the future they could be identified with previously discovered asteroids. 11

3 Observing at small solar elongations In this period we are testing a third phase of ADAS, pointing to sky zones whose angular distance from the Sun is less than 90. This strategy has been suggested by the following considerations: 1) Inner NEOs (INEOs) can be discovered only at E<90. None however was know till Feb. 2000 (Michel et al. 2000). 2) Many large surveys observe near the opposition, therefore new discoveries with ADAS likely will be made at small elongations. Table 6 Small solar elongations areas observed till now Solar Elongations (deg) Surveyed sq deg [40,50[ 0 [50,60[ 10.3 [60,70[ 0 [70,80[ 3.8 [80,90[ 0 On the other hand, observing at small solar elongations will exclude in general the ecliptic plane, thus considerably lowering the total number of observable asteroids. This expectation is borne out by the available data: number of observed hours: total number of detected asteroids: 4 (3 main belt, 1 Hilda) surveyed area: 14.1 sq deg asteroids per sq deg: 0.28 4 - Further developments Several improvements are being sought of: 1. add a filter wheel; several filters apt for CCDs, and filter wheels built for other purposes, are available at the Observatory; one will be selected for utilization with SCAM-1 (action under way). 2. replace the obsolete electronics of telescope and dome encoders read-out with a new one, in order to have automatic filling of the frame headers (action under way) 3. complete overhauling of the electrical system of the dome, to comply with CE safety regulations (action under way) 4. On a longer time scale, we can think of a full automatization of the telescope and dome, but that action would probably imply a complete overhaul of the motors and of the mechanics of the telescope and of the dome. This long and expensive action is certainly justified per se, to have a modern and updated telescope. In the light of the asteroidal program it would however require an order of magnitude improvement in the efficiency of the system, such as replacing the SCAM-1 camera with a new one having twice the size and larger quantum efficiency. A good possibility would be to use a 4048x4048 thinned device with pixels of 13.5 micrometers (in order to better sample the seeing), LN2 refrigerated. In this case also the shutter and the filters would have to be replaced by others with larger size. 12

References Barbieri, C., Pertile, V., Rampazzi, F., Caporali, A., Dallaporta, G.F. 1992, Una Nuova Determinazione delle Coordinate Astronomiche e Geodetiche degli Osservatori di Asiago e Cima Ekar, Rapporto Interno nr.10 Michel, P., Zappalà, V., Cellino, A., Tanga, P. 2000, Estimated Abundance of Aten and asteroids evolving on orbits between Earth and Sun, Icarus, Volume 143, Issue 2, pp 421-424. 13