The size distribution of kilometer-sized asteroids implies that the presently un-detected population of sub-kilometer asteroids far outnumber the known larger objects. Although small, sub-kilometer asteroids may provide unique insight into the evolution of the asteroid belt and the meteorite-asteroid connection. Small asteroids probably have fresh surfaces with minimal regolith cover thus presenting a clearer view of their internal structure. Streams of small asteroids connecting parent asteroids with resonances may provide insight into the transfer of material to the inner Solar System. We propose a space mission to detect and study these objects. A key feature of the mission is full onboard autonomy of both object detection and tracking, thus reducing the cost of running the mission to a minimum, while enabling science to focus on the main objectives.

1. Why study small Asteroids?

Our present understanding of asteroids and their orbits is almost entirely based on surveys of asteroids larger than 10 km in diameter. Ground based telescopes cannot detect smaller objects except for the immediate vicinity of the Earth and no space craft has, so far, detected any previously unknown asteroids. Despite the fact that several spacecrafts to date, statistically, must have passed by smaller asteroids, the technology employed in these vessels has not held the capability of autonomously detection. Therefore the encounters have gone unnoticed by. Recent development in the autonomy of space-borne imager- and computer-technology has changed this, so that it is now possible to detect, classify and observe an encounter of a smaller asteroid.

One such instrument is the advanced stellar compass, which is able to detect moving objects and autonomously and with arc second accuracy guide the science telescope to track the object. For the first time, this gives us the possibility to study smaller objects. In short we propose a mission where a fully autonomous spacecraft detects the objects, determine their orbital parameters, light curve and spectral characteristics.

Text Box: Fig. 1. Size distribution of objects larger than 1 km based on the currently known population (From [1]).

1.1. Abundance of small asteroids

We have very little information on the abundance and characteristics of objects smaller than ca. 1 km except for those that have been observed in the immediate vicinity of the Earth. The power law distribution of asteroid sizes (Fig. 1) suggests that objects smaller than 1 km should be very abundant. Furthermore, the orbits of these small objects are easily perturbed and we should therefore expect to see them more or less uniformly distributed throughout the Solar System. Within the asteroid belt we have no information at all on these objects, since they cannot be observed from Earth and no spacecraft have been actively looking for them.

Since any fragmentation process tends to generate power law size distributions of the fragments we should not be surprised to see that the size distribution of asteroids follow a power law distribution. There is, however, reason to believe that the smallest asteroids are less abundant than a simple extrapolation of the data in Fig. 1 would suggest. Smaller asteroids are more easily influenced by the Yarkovsky effect [2] and may thus be removed from the belt on a shorter time scale than the larger asteroids. Still smaller fragments may be removed as a consequence of the Poynting-Robertson effect. A direct measurement of the size distribution would allow us to assess the efficiency of the two effects.

1.2. Hazards to the Earth

Although impacts of small asteroids will only cause regional damage their higher abundance than kilometer-sized objects will make such impacts much more frequent and therefore a serious threat that needs to be addressed. Impacts of 50 m objects like Tunguska are expected to happen a few times every millenium and has an explosive power equivalent to 20 megatons TNT [3]. Earth-based telescopes can only detect such objects in the immediate vicinity of the Earth.

1.3. Current collision rate

The orbits of small objects are perturbed due to the Yarkovsky effect and Poynting-Robertson drag leading to shorter lifetimes than regular asteroids. We should expect a continuos flux of small asteroids into the resonances. The abundances of small objects close to and within the resonances are therefore measures of the current production rate of smaller fragments and of the transfer rate of fragments from the main belt.

Earth based observations have discovered two cases where clusters of small asteroids and a possible parent with similar spectral characteristics suggest a recent large impact event [4, 5]. Since such fragmentation events are expected to produce much higher numbers of sub-kilometer asteroids a survey of small asteroids may be able to detect more recent impact events.

1.4. Asteroid-Meteorite relationship

Meteorites are fragments of approximately 150 different main belt asteroids. Since meteorites are well studied representatives of the abundant low mass tail of the objects that impact the Earth a better understanding of their origin in the asteroid belt and their subsequent orbital evolution will allow us to better understand the transfer of larger objects. Cosmic ray exposure ages of meteorites show that they have spend 10⁶-10⁹ years as meter-sized fragments before falling on Earth. Detailed studies of meteorites allow us to determine age constraints for the disruptions of their parent asteroids and the subsequent transfer of fragments to the inner Solar System. Only in one case has it been possible to establish a reasonably good case for a specific asteroid-meteorite relationship. The unique spectrum of the basaltic surface of 4 Vesta makes it the prime candidate for about 400 igneous meteorites known as the HED meteorites. There is considerable interest in establishing links between other groups of meteorites and their parent asteroids.

The Vestoids is an example of a group of small objects for which a combination of their spectral characteristics and orbit parameters have been used to establish a link between the members and their parent asteroid, Vesta [5]. Vestoids are 5-10 km V-type asteroids that form a stream in orbit space from Vesta toward the 3:1 mean motion resonance. Vestoids are believed to represent large fragments that were ejected from Vesta in a major impact event. Those fragments that made it to the 3:1 resonance are believed to be the parent bodies of HED meteorites. We infer that sub-kilometer objects that are considerably more abundant than 5-10 km objects will be even better to establish links between fragments and their parent asteroid. For the Vestoids, detection of sub-kilometer V-type fragments will help constrain the efficiency of the Yarkovsky effect that operate faster on smaller objects.

Meteorites also provide evidence of catastrophic events in the asteroid belt that may have dramatic consequences for the impact flux on Earth. Today, 38% of all meteorites falling on Earth are L-chondrites. This may be a result of a catastrophic disruption of their parent body 500 My ago [6]. Abundant fossil chondritic meteorites in 480 My old Ordovician limestone from Sweden suggest that the meteorite flux following the event was one order of magnitude higher than the present day flux [7].

Another example is the IIIAB iron meteorites that are the most common type of iron meteorites falling on Earth. IIIAB irons probably come from the catastrophic disruption of a large differentiated asteroid 650 My ago [8].

Some of the small objects come from the same asteroidal source as the meteorites that have already fallen on Earth. Data on the orbital distribution of objects with spectral characteristics similar to a group of meteorites may provide new constraints on the meteorite-asteroid relationship.

1.5. Surface characteristics

Unlike the larger asteroids studied from space so far, small objects are unlikely to be regolith covered, and their surfaces are therefore representative of their interior. A longstanding debate has been the relationship between S-type asteroids and ordinary chondrites. Differences in spectral characteristics have been attributed to a poorly constrained space weathering process. Since small asteroids probably have smaller life times and less gravity we should expect them to have younger surfaces that have been exposed to space weathering for a shorter period of time. Also the lower gravity should reduce the build-up of a regolith cover on their surfaces that may hide geologic units underneath. Both of these effects will make a comparison with spectral characteristics of meteorites and other materials easier.

1.6. Crater densities on planetary surfaces

Furthermore, constraints on the abundance of small objects will allow much better age determinations of cratered planetary surfaces such as the very young lava flows (< 10 My) on the northern plain on Mars [9]. Better ages for such surfaces will provide new insight on the climatic and geological evolution of Mars and other planets and moons.

2. The bering mission

For the reasons outlined above, the main goal of the proposed Bering mission is to detect and characterize a sizeable amount of sub-kilometer objects from space. This will be the first systematic survey of sub-kilometer objects in the Solar System. The numbers detected need to be sufficiently high that the distribution of small objects with similar spectral characteristics and therefore potentially identical parent asteroid may be established throughout the main belt. The resonances are expected to be sparsely populated and we therefore require good statistics to constrain the distribution of objects within and close to the resonances.

Since the distribution of the objects to be studied is unknown the probe must detect the objects and guide the science instruments in a fully automated process. It is estimated that a reasonable detection rate (i.e. several detections/day) requires that the detection limit of Bering should be better than m_v = 9.

In order to determine the distribution and dynamics of small objects and their links to the NEO population we need to detect objects in the main asteroid belt as well as inside the Earth’s orbit. The objects within the resonances that are either already NEOs or are becoming NEOs within a short time frame, generally have aphelion within the main belt and spend most of their time outside the Earth orbit. In order to study the distribution of these very important members of the NEO population it is therefore important that the probe also detect object within the main belt.

Although the probe could be directed either solely inside the Earth orbit or solely outside the Earth orbit we propose a mission profile that would give us data on the distribution of small objects all the way from 0.7 AU to 3.5 AU. Alternatively, it should be considered to launch several identical probes on different trajectories.

For each detected object we propose to automatically determine:

· The position, and velocity vector of the object.

· Multicolor photometry of the reflected light from the object in the 440 – 2200 nm range.

· The light curve of the object and thereby its rotation period.

For a few selected larger objects we furthermore propose to:

· Record multicolor images of the surface

· Determine the mass and magnetic moment of the object.

The data will allow us to determine a current orbit for the object. With a detection limit of m_V =9 for the advanced stellar compass and a detection limit of m_V = 25 for the multicolor imager we will be able to follow the object out to a distance of approximately 1500 times the distance where it was detected. Depending on the geometry and the size of the object this will typically allow us to follow the object for weeks to months and determine high precision orbital parameters for a large fraction of the orbital arc. The orbital data will also allow us to determine the objects position in orbital space and its proximity to resonance’s and/or other asteroids or asteroid families with similar spectral characteristics and dynamical characteristics.

The photometry will allow us to determine the spectral type of the object. This will make it possible for us to determine its relationship with other asteroids and/or groups of meteorites with similar mineralogy. Ultimately, we will attempt to backtrack the object to its parent asteroid

The ability to detect objects down to m_V = 25 from within the asteroid belt with the multicolor imager may also be utilized to further constrain the density of small objects. In a few campaigns we propose to make a series of frames with either subsequent onboard processings or data processing on Earth. This will allow us to determine the number of asteroidal objects in each frame, going to very faint objects. Although the size-distance relationship cannot be determined on the basis of a few frames these data may be used to check predictions based on models of the distribution of asteroids.

3. Main Science Payload

The core autonomy of the proposed mission is centered on the Autonomous Stellar Compass (ASC), a fully autonomously operation star tracker. The ASC consist of a powerful microcomputer with search engine and star catalogue, with several camera heads attached, typically two to four. Each camera head will acquire an image of the night sky in the Field Of View FOV two times per second and pass on the digital image to the data processor.

Fig. 2. Preliminary design of Bering.

The data processor will then sift the image for luminous objects, so as to extract all stars in the FOV. Typically, the detection limit is set to m_v 9, but it may be lowered, at the cost of additional process time, to m_v 11 in special occasions where deeper objects are sought. Based on the star patterns thus found, a match is established against the onboard star catalogue. This process takes about 50 ms. Based on this match, the ASC determines the attitude, i.e. the pointing direction of the camera, with an accuracy better than 1 arc second. This process is extremely robust and reliable operations have been demonstrated on multiple satellites presently in orbit. Similar solutions are generated for the other camera heads.

Since all luminous objects above the selected threshold are detected, also non stellar such as galaxies, nebulae, other satellites and asteroids are also autonomously picked out. They are identified as luminous objects brighter than the threshold, with no matching object in the star catalogue. Since the ASC has established the attitude of the image in question, the apparent position of the object is established with high accuracy. Typically the instant position determination is in the range of 3 arcseconds, but since the ASC update the measurement per camera twice per second, averaging will bring the accuracy to the sub-arcsecond level in the matter of seconds. In this way, the entire FOV is searched for non-stellar objects. Since the FOV covers 1% of the entire night sky, a scanning operation of the spacecraft will have to be employed. We plan to arrange 7 camera heads onboard the spacecraft, whereby almost 100% sky coverage is guaranteed at a spin rate of once per hour.

The non-stellar objects are entered into a database. When the same area of the night sky is revisited after one rotation period of the spacecraft, the apparent position of the objects are established again. Any object thus showing a proper motion above the measurement noise is moved to an asteroid candidate list for further investigations.

The main science instrument on the proposed mission is a multi-band imaging telescope. The telescope has a focal length of 1

m, and an entrance pupil of 0.4m. To compensate for the spacecraft rotation and to allow for fast tracking, the telescope is fitted with a folding mirror. To enable fast and accurate tracking of a target, the telescope is furthermore equipped with an eight ASC camera, mounted on the back of the secondary mirror. The ASC camera and the telescope optical axes are approximately parallel. From simultaneous images of the night sky, the relative orientation of the telescope to the ASC camera is easily established with high accuracy.

When an object in the asteroid candidate list is in the part of the night sky that can be viewed by the telescope a three-step procedure is initiated. 1) The folding mirror is commanded to move, open loop, to a position close to the object. 2) Using the ASC camera, the folding mirror is used to bring the target into the FOV. 3) The telescope takes over and tracks the object. This procedure allows the Bering spacecraft to track more than 1000 objects at any give time of the mission.

4. References

1. Davis, D. R. Farinella, P., Small bodies tale: a comparison among three populations, Lunar Planet. Sci. Conf. 28, Abstract # 1765, 1997.

2. Farinella, P. Vokrouhlicky, D. Hartmann, W. K., Meteorite delivery via the Yarkovsky orbital drift, Icarus, Vol. 132, 378-387, 1998.

3. Morrison, D. A. Clark, C. R. Slovic, P. The impact hazard, in Asteroids II, University of Arizona Press, Tucson, AZ, pp. 59-91, 1991.

4. Nesvorny, D. Bottke, W. F. Dones, L. Levison, H. F., The recent breakup of an asteroid in the main-belt region, Nature Vol. 417, 720-722, 2002.

5. Binzel, R. P. Xu, S., Chips off asteroid 4-Vesta - Evidence for the parent body of basaltic achondrite meteorites, Science Vol. 260, 186-191, 1993.

6. Haack, H. Farinella, P. Scott, E. R. D. Keil, K., Meteoritic, asteroidal, and theoretical evidence for the 500 Ma disruption of the L chondrite parent body, Icarus Vol. 119, 182-191, 1996.

7. Schmitz, B. Peucker-Ehrenbrink, B. Lindstrom, M. Tassinari, M., Accretion rates of meteorites and cosmic dust in the Early Ordovician, Science Vol. 278, 88-90, 1997.

8. Keil, K. Haack, H. Scott, E. R. D., Catastrophic fragmentation of asteroids - Evidence from meteorites. Planetary and Space Science Vol. 42, 1109-1122, 1994.

9. Hartmann, W. K. Berman, D. C., Elysium Planitia lava flows: Crater count chronology and geological implications. J. Geophys. Res.-Planets Vol. 105, 15011-15025.