The solar system offers a wide variety of objects to study, most of which are very poorly understood. Other parts of this chapter focus on planetary atmospheres, the gaseous emissions of comets, radar astronomy and solar physics. This section will focus quite tightly on what might be learned from observations of thermal emission from the smaller, solid bodies of the solar system using the SKA.
The SKA is proposed to have an angular resolution of 0.1
at 1.4 GHz. Table 3.1 shows how the resolution and sensitivity targets translate into
the context of solar system astronomy.
m). The LMA is intended to provide resolutions comparable to Hubble.
Thus, at 20 GHz the SKA would provide resolutions better by a factor of 3 than
the NGST and at least a factor of 10 better than the HST.
In the next few subsections we will consider in more detail how the SKA might be used to study of these smallest, and least-understood members of the solar family.
Bearing in mind that the SKA promises to provide images at 20 GHz significantly sharper than any competing instruments, the large asteroid Vesta will be used to illustrate the importance of high-quality imaging in the study of asteroids. With a diameter of 520 km, Vesta subtends 0.7 arcsec at a typical opposition distance of 1.3 AU, and is an appealing target for imaging with the best currently available telescopes.. Uniquely amongst the asteroids, Vesta has been imaged with the HST (e.g Zellner et al. 1997), from the ground with adaptive optics (Drummond et al. 1998) and at radio wavelengths with the VLA (Johnston et al. 1989). Prior to these images, several decades of lightcurve analysis had only just been able to establish that the true rotational period was 5.2 hours, not 10.4 hours, and it was widely believed that much of the shape of the lightcurve was due to albedo patches and not simply to an odd shape.
Between November 28 and December 1, 1994, the HST was used to image Vesta (Zellner et al. 1997) at a distance of roughly 1.7 AU. With a linear resolution of roughly 80 km, the visible disk was roughly 6.5 resolution elements across. These images, using roughly 8 minutes of exposure time, completely determined the rotational pole of Vesta, showed that it rotated in the prograde sense, and unambiguously demonstrated that the rotation period was 5.2 hours. Compared to the lightcurve analyses, the data gathering and analysis of the HST images was faster by a thousand-fold and was free from most of the ambiguities which plagued the earlier efforts.
The SKA at 20 GHz could have made equally valuable images of objects 10 times smaller than Vesta. There are only three main-belt asteroids with diameters as large as Vesta's, but in the IRAS FP102 database of asteroid albedoes and diameters (Tedesco et al. 1992) there are 655 with estimated diameters larger than 50 km. For those studies which do not require well-resolved images, almost all of the 1884 asteroids in the FP102 database had estimated sizes larger than 10 km, the resolution limit for the SKA at 2 AU. The true number of resolvable asteroids will be larger by one or two orders of magnitude because the FP102 database becomes seriously incomplete for asteroids smaller than 100 km.
The importance of pushing the highest operating frequency of the SKA to at least 20 GHz starts to become clear at this point. If the highest frequency is restricted to 2 GHz, the SKA will only provide a linear resolution of 85 km, no better than that already possible with the HST. Only a dozen of the largest asteroids could possibly be mapped at 2 GHz. The study of radio images of asteroids could be complete within the first year of operation, probably without any really surprising discoveries on these well-studied objects.
 |
Returning to Vesta, the HST images allow us to begin to study the geological history of this object. A later image (Fig. 3.1) shows a very large crater at the south pole of Vesta, and perhaps two other large craters near its equator. With a ten-fold improvement in the angular resolution of the raw image, we can expect that many smaller craters will be visible, allowing cratering-ages to be estimated for the larger units on the surface. Unlike Vesta, many smaller asteroids are suspected to be ``rubble piles'', collections of fragments loosely held together by their own gravity. We might expect such objects to be lumpy, much like the radar images of the asteroid Toutatis in Figure 3.5. With images of hundreds of asteroids available, we should be able to test this model rigorously.
Furthermore, the radio data will contain information not available at shorter wavelengths. As reported at the Annual General Meeting of the Canadian Astronomical Society in May 1998, (CASCA'98) the rotational lightcurve of Vesta at 2 mm wavelength, measured with the SCUBA detector on the James Clerk Maxwell Telescope on October 14, 1997, had an amplitude twice as large as the visible lightcurve (Fig. 3.2 Redman et al., 1998b). At the time of writing, it is not clear why the amplitude of this lightcurve is so large, why it was single-peaked when our previous observations of Vesta in September 1989 showed a doubly peaked curve more in accordance with our theoretical expectations, nor how many other asteroids will have similarly informative radio lightcurves.
 |
On a longer timescale, the SKA might be used to prospect for mineral deposits
on asteroids to be exploited by space-based mining ventures in the next
century. Judging from the composition of iron and stony-iron meteorites, the
M-type asteroids consist primarily of a Fe-Ni alloy with traces up to
several hundred ppm of precious metals (Kargel, 1994). Ignoring the iron and
nickel, the precious metal content of even a small (1 km) M-type asteroid
should be worth roughly
at current prices. Our
observations of the large M-type asteroids Psyche and Kleopatra with the JCMT
(Redman et al., 1998a) show that the emissivity of these metallic bodies is
roughly half that of stony or carbonaceous asteroids and that the emissivity
drops rapidly with increasing wavelength. The emissivity at 20 GHz might be as
little as a few percent. The tremendous sensitivity of the SKA should still
make these objects easy to observe, with the ore bodies cleanly distinguished
as dark patches amongst the brighter silicate rocks.
For the nonmetallic asteroids, thermal models suggest that 20 GHz is a nearly optimal observing frequency. Following the discussion in Redman et al., 1998a the deviations of an asteroid's spectral energy distribution (SED) from a Planck spectrum are dominated by different effects in different wavelength bands. At short infrared wavelengths the shape of the SED is determined by the temperature and area of the warm region around subsolar point. In the far infrared and submillimeter the shape of the SED is strongly affected by the thermal and dielectric properties of the regolith, a layer of shattered rock which is believed to cover the surfaces of most large asteroids. At centimeter wavelengths, starting around 20 GHz, the radiation emerges from layers in the regolith which do not participate in the diurnal heating and cooling cycle as the asteroid rotates. The temperature of this material is just the time average of the surface temperature, an easy quantity to model and interpret. At lower frequencies we get exactly the same information but with lower signal to noise. At higher frequencies we start to see the effects of the diurnal temperature wave which propagates down into the regolith, complicating the analysis of the data. Indeed, it is difficult to interpret the data at higher frequencies without having the centimeter data for comparison. In this sense, 20 GHz is exactly the correct frequency to observe an asteroid, providing the most easily modeled information at the highest possible signal to noise.
We note that the NGST, LMA and SKA will each be able to observe asteroids in one of the key bands mentioned above. The NGST will be able to measure the surface temperatures in the infrared, the LMA will probe the structure of the regolith and the SKA will provide the mean asteroidal temperature from which thermal models may be constructed. These will be three complementary machines, and in an ideal world we would want access to all three.
Complementing the studies of thermal emission from asteroids, radar observations offer valuable insights into the surface properties of asteroids. For example, radar might be used to prospect for metallic asteroids, since Ostro et al. (1985, 1991) have shown that they have much higher radar reflectivities at centimeter wavelengths than non-metallic asteroids. Also, the inventory of hazardous Earth-crossing asteroids is biased towards asteroids which reside primarily outside the Earth's orbit so that they can be seen at night. Radar can easily search for nearby objects in the day-time sky, eliminating this bias. The large collecting area and multiple-beam capability proposed for the SKA may make it a valuable tool for radar studies throughout the inner solar system.
The satellites of the giant planets provide a menagerie of bizarre worlds whose surfaces are unlike anything seen elsewhere in the solar system. Spacecraft have visited these places and have provided detailed maps of many of their surfaces, but have left many questions unanswered, some of which can be best addressed at radio wavelengths. The variety and complexity of each of these worlds precludes a detailed discussion here, but several interesting lines of research are worth mentioning.
Io is so strongly heated by tidal interactions with Jupiter, Europa and Ganymede that it exhibits continuous and vigorous volcanic activity, with half a dozen large vents active at any given time. Although the striking colours on Io's surface are believed to be due to sulphurous compounds the temperatures in some of the vents are so high that silicate lavas must also be involved. The vents show up brilliantly in the near infrared and are easy targets for even ground-based infrared telescopes using adaptive optics. However, to study the resulting lava flows, which run for hundreds of kilometers across the surface, will be more difficult since they are cool, faint features immediately adjacent to the bright vents. For this study the SKA will be nearly ideal. The linear resolution will be roughly 20 km at the distance of Jupiter, and in the radio the flux density will vary linearly with temperature, unlike the exponential variation in the infrared.
The other large Jovian and Saturnian satellites have icy surfaces. The emissivities of these surfaces are quite low, probably because scattering and refraction at fractures and dense lumps within the ice prevents photons from escaping the surface. Many of these objects have been tectonically active and exhibit very different colours and large-scale textures in the different terranes comprising their surfaces. It is plausible to guess that in the different terranes the emissivity of the ice might also be different. Images at 20 GHz made with the SKA would allow us to study the interplay of ice tectonics, water volcanism and impact cratering under a variety of conditions on the Jovian and Saturnian satellites.
Probably, the most mysterious world in the solar system is Titan, the only satellite with an extensive atmosphere. From images at infrared wavelengths chosen to penetrate the perpetual cloud cover, it is known that the surface of Titan is mostly dark-coloured, probably covered with an organic sludge created by photochemical reactions high in Titan's atmosphere and precipitated onto the surface. There are, however, large light-coloured patches which may be icy highlands, swept clean of the organic material by some form of organic ``rain''. It is a plausible guess that the emissivity of the light regions may be different from the dark regions. Since Titan's atmosphere is transparent at centimeter wavelengths, it may be possible to map Titan's surface using the SKA, with a linear resolution of 40 km compared to Titan's diameter of 5140 km. These maps may be combined with the radar maps which will be made by the Casini mission to better understand the chemistry and physical state of the materials on Titan's surface.
Gas giant planets do not seem to have formed beyond Neptune. Instead, the original protoplanets continue to orbit in a vast, extended disk known as the Kuiper Belt. The largest known Kuiper Belt Object (KBO) is the planet Pluto. A few KBOs have been diverted inside the orbit of Neptune. Some of these objects occupy temporary orbits between Jupiter and Neptune and are known as Centaurs, at least one of which, Chiron, shows cometary behaviour. Others, whose orbits carry them into the inner solar system, constitute the population of short-period comets.
 |
The Kuiper Belt is comparable in size to the proplids which have been observed
around newly forming stars in the Orion Nebula (Fig. 3.3 (left),
McCaughrean and O'Dell, 1996) with the HST. At the least, the proplids are
Kuiper Belts in formation, and may contain newly-forming planetary systems
closer to their central stars. Similarly-sized dust rings have
been
detected using the SCUBA submillimeter-wavelength camera on the JCMT around
several nearby main sequence stars, such as 
Eri
(Fig. 3.3 (right),
Greaves et al, 1998). Since the lifetime of these dust particles
due to Pointing-Robertson drag is short compared with the lifetime of the
central star, there must be a reservoir of larger objects just like the Sun's
Kuiper belt around each of these stars. Curiously, almost all of the dust
rings seen with SCUBA are clumpy rather than axisymmetric around the central
star. It is not clear whether the clumps are transient features or represent
structures such as density waves or swarms of particles around large objects
like planets embedded in the belts.
Since it spreads so widely across the sky, we cannot see the emission from dust in the Sun's Kuiper Belt, but we can see the larger objects which supply the dust. Observations of the solar Kuiper Belt therefore nicely complement and tie together some of the most exciting new observations in the fields of star formation and extra-solar planetary systems. At the time of writing the Minor Planet Center lists 68 KBOs, not including Pluto and Charon. Luu and Jewitt (1998) estimate there are 160,000 objects in the Kuiper Belt closer than 50 AU with diameters in excess of 100 km, hundreds of times more than similar sized objects in the main asteroid belt. Since the Kuiper Belt extends to about 100 AU, there may be comparably many objects again waiting to be discovered in the outer parts of the belt.
In spite of their large numbers, KBOs are faint, difficult sources, and it will be especially troublesome at visible and infrared wavelengths to find members of the outer Kuiper belt, beyond 50 AU. The brightness of reflected sunlight drops off with distance d as d-4, making objects in the outer belt 5 magnitudes fainter than comparable inner belt objects. Similarly, the low temperature of the KBOs, dropping from around 40 K in the inner belt to only 20 K in the outer belt, means that the thermal emission does not start to dominate the reflected sunlight until far-infrared wavelengths.
By contrast, the SKA operating at 20 GHz will be almost ideally suited to study
KBOs. The linear
scale which the SKA will resolve at a distance of 40 AU is 200 km. After an 8
hour integration it should be possible to achieve a 5 
detection on a
120 km object, as shown in Table 3.1. Even at a distance of 100 AU,
the smallest detectable object would be 350 km in diameter, and there may
be many of these objects if their size distribution is similar to that which we
have measured on the inner edge of the belt. If self-gravitating clusters of
KBOs exist, capable of generating clumps of dust emission such as are seen
around other stars, the SKA may be the tool required to find them.
Very little is yet known about the physical properties of KBOs. Luu and Jewitt (1996) have shown that they have a wide range of colours, possibly indicating collisional resurfacing. The relatively slow orbital speeds in the outer solar system suggest that collisions in general will be slower and less violent than in the asteroid belt. This should be reflected in the shapes and surface features of the KBOs, but it is not yet known whether KBOs should be modeled as rubble piles, as solid, primordial bodies, or as some other kind of structure entirely, perhaps reflecting the original, gentle growth of the protoplanets. At the moment, nothing is known about their shapes, rotation periods, masses or regolithic structures.
By imaging the larger objects it will be possible to address many of these issues with the SKA. Their shapes and rotational properties will be directly observable. With direct measurements of the thermal flux at radio wavelengths it will become possible to build reliable thermal models of the surfaces of these objects, giving insight into their densities and compositions. Curiously, with the strawman specifications for the SKA it will not be necessary to revive the old lightcurve-analysis techniques which have been used at shorter wavelengths to study asteroids. Asteroids are seen as bright point sources, but the surface brightness of KBOs will be so low that they will necessarily be extended sources. Reliable measurements will only be obtainable for objects large enough to fill at least one pixel.
To do this work it is essential that the SKA operate at 20 GHz. No machine less sensitive than this will even be able to detect these small, cold worlds. Restricting the top frequency to 2 GHz would prevent us from studying any but the largest few KBO's, comparable in size to Pluto and Charon, as shown in Table 3.1. For solar system astronomy, the unique ability of the SKA at 20 GHz to observe KBOs throughout the full extent of the belt may be the most compelling reason to build it.
AU)
2mm shown
above has twice the amplitude of the visible light curve
(Redman et al. 1998b) and has changed from a single-peaked shape in
less than 10 years.