THE INNER SOLAR SYSTEM

Chapter 9 - Asteroids

Chapter 9 - All

1 2

WHO CARES? MAIN BELT e-a & i-a

9SOLAR SYSTEM FORMATION

9CONTI NUI NG EVOLUTION ν 9ASTEROID 6 STRENGTHS

9EARTH IMPACT HAZARD Eos 4

Asteroids: Gaps & Resonances Distribution and Orbits of Main-

•Astronomer Daniel Belt Asteroids Kirkwood (1886) noticed • Most asteroids found between Mars and Jupiter that the Main Belt has “gaps” in which asteroids • Distribution in the main-belt not uniform are “missing”. – strongly influenced by resonances with Jupiter (Kirkwood Gaps) • The Kirkwood Gaps are “locations” where • Collisional disruption of larger bodies long ago has left resonances with Jupiter’ s physical and dynamical asteroid “families” orbit occur; i.e. where – Eos, Hirayama, Themis and Koronis are major groups gravitational disturbances – Asteroids in each family show similar spectra by Jupiter are the strongest. • Protective zones near Lagrange points of Jupiter heavily • May explain why there is populated with Trojan asteroids no planet there: Jupiter – Mars also has five known Trojan asteroids (Neptune has six) only allowed small bodies to coalesce and prevented • Typical main-belt orbit stable on timescales of Ga a larger planet from • Separation of asteroids 1 km and larger ~5 million km! forming. 5

1 Formation and History of the Asteroids: Size Distribution Asteroid Belt • The number of asteroids of a given diameter D is • Main-belt asteroids believed to be in orbits stable for long 2 periods, but higher relative energies than original primordial proportional to 1/D orbits – Collisional distribution • For example: • Very little mass today (fraction of Earth’s mass), but likely – 3 > 500 km much more originally (comparable to Earth) – 13 > 250 km 1. Most material ejected from the solar system by Jupiter or driven into – Hundreds > 100 km the Sun (Iron overabundance observed in other stars?) – 10,000+ > 10 km (?) 2. Average mass density in SS shows discontinuity at MB Total: 1,000,000 > 1 km (?) • Asteroids are plantesimals which never were able to grow – Most of the mass is in the larger via collisional accretion because Jupiter increased largest few asteroids their average impact velocity early in the SS, turning accretional events into disruption events. – Total mass of all asteroids is only ~5-10% mass of the Moon

7

DAVIS SIZE DISTRIBUTIONS – MAIN Size Distribution BELT −ς ⎛ R ⎞ ⎜ ⎟ R is the radius, N0dR N(R) = N0 ⎜ ⎟ ⎝ R0 ⎠ is the number of asteroids between R, R + dR The power law form of the size distribution is explained by collisional evolution. Theory suggests that populations will evolve to ζ~3.5 when disruption is self-similar. The steep slope implies that most of the mass is in the large bodies, though most of the surface area is in the small ones.

9

Radar Observations of 4179 Toutatis Radar observations of Toutatis Asteroid Sizes I Determination of asteroid sizes done via: 1. Assume albedo; measure brightness and distance and get geometric surface area 2. Measure visible and IR flux at the same time; ratio gives albedo since visible light depends on albedo and IR on (1-A). Also need thermal model of asteroid 3. Occultations by Stars 4. Direct measurement by Spacecraft (NEAR, Galileo) 5. Direct imaging with Adaptive optics

6. Radar observations 11

2 13 14 Ceres as observed by Hubble, 2004

Spacecraft measurements Near-Earth Asteroids (NEAs)

• Smaller asteroids “escape” main-belt from resonances • Classes of NEAs based on orbits – Amor (()outside Earth’s orbit) – Apollo (cross Earth’s orbit, a>1 AU) – Aten (cross Earth’s orbit, a<1 AU) – Atira (inside Earth’s orbit) • Potentially Hazardous Asteroids (PHAs) : minimum orbital distance with Earth’s orbit < 0.05 A.U. 15 16

NEA TYPES NEA TYPES

AMOR APOLLO 1.017 AU < q < 1.3 AU q < 1.017 AU •Amor • Apollo a > 1.0 AU

EARTH EARTH

3 NEA TYPES Origin of NEAs

ATEN • NEAs are continually injected into NEA region Q > 0.983 AU from main-belt • Aten a < 1.0 AU • Collisions between larger asteroids produce fragments which evolve due to radiation forces and weak resonances until reaching a maj“”hth(tk10jor resonance “escape” hatch (takes 107- 8 EARTH 10 years depending on size of daughter asteroids) • Orbit then rapidly evolves – Some are ejected from solar system before becoming NEAs – Some become NEAs • May eventually hit a planet, get ejected or be driven into the sun 20

Yarkovsky Effect Yarkovsky Effect

• Asymmetric re-radiation of thermal energy for a rotating body • Uncertainties: – Thermal properties, The evening hemisphere radiates albedo extra energy and momentum because it is hotter than the morning – Spin rate (changes in spin hemisphere. For prograde rotation, rate) the net force is forward. 8 ⎛σT 4 ⎞⎛ ΔT ⎞ 2 ⎜ ⎟ FY = πs ⎜ ⎟⎜ ⎟cosζ 3 ⎝ c ⎠⎝ T ⎠ Diurnal Yarkovsky: net force opposite the Seasonal Yarkovsky: net force opposite 1/ 2 afternoon direction. Prograde rotator the “summer” hemisphere. Makes the ⎛ P ⎞ Burns et al (1979) ΔT ≈ (1−δ )γS⎜ ⎟ spirals out; retrograde rotator spirals in. object spiral inwards. ⎝ 2π ⎠ 21 22 Bottke et al. (2001)

Meteorite parents Light Curves Largest NEAs • Shape changes tend to produce double peaked curves, Tunguska-like objects whereas albedo variations tend to produce single peaked curves. • Light curves at different position angles can be used to 1. K=0.002 W/mK 2.K=0020.02 W /mK determine the pole position and sense of rotation (which 3. K=0.2 W/mK may have two components, like a badly thrown football). 4. K=2 W/mK 5. K=40 W/mk • Comets tend to have higher amplitude variations and slower rotation. • Trojan asteroids tend to show larger amplitudes, Mean change in a of inner main belt asteroids over their collisional lifetimes suggesting that they are more elongated than main belt versus radius, for 5 thermal conductivities. Low K is dominated by diurnal asteroids. Yarkovsky: high K by seasonal.

24 Bottke et al. (2001)

4 25

Asteroid Physical Structure Asteroids: Geology • Many asteroids believed to be collections of • ~12 asteroids visited up close by spacecraft: a few large fragments and smaller debris – 951 Gaspra: Galileo flyby in 1991 bound together as a “rubble-pile” – 243 Ida: Galileo flyby in 1993 Evidence: – 253 Mathilde: NEAR flyby in 1997 1. Rotation rates of most asteroids sharply cut off at critical value – 433 Eros: NEAR orbital mission in 2000-2001 (2. 2 h) imp ly ing no tens ile s treng th to o bjec ts – 25143 Itokawa : Hyabusa sample -return (?) mission in 2005 2. Densities lower than meteorites suggesting large amounts of – 5535 Annefrank: Stardust in 2002 empty space – 4 Vesta by Dawn in 2011 3. Models of early solar system formation suggest most (or all) –etc moderate-sized (<100 km) asteroids catastrophically disrupted over lifetime of the solar system • Also: Spacecraft images of Martian moons Phobos and 4. Simulations of catastrophic disruption show debris reforming Deimos: captured asteroids? into piles of flying debris under gravitational attraction. • Abundant evidence for impacts, and surprising evidence 5. Craters on some asteroids are too large – should break apart a solid body, but would be possible if body is rubble-pile for erosion and tectonism on these small bodies. 27

Itokawa

NEAR at Eros

29

5 Collisions

• Escape velocities are very small for most bodies. • Therefore most collisions are explosive. • Orbital families are collections of smaller asteroids which are apparently fragments of a larger parent – namedftthild after their larges t exi itisting mem ber. • Collisions also produce dust which is seen in infrared observations (zodiacal dust). • Some collision fragments can re-coalesce into a “rubble pile”.

Phobos and Deimos, Mars’ two moons as seen by the Mars • Binary asteroids can also be formed. Reconnaissance Orbiter. Captured asteroids? • Binaries are useful since they provide mass estimates. 31 • Some craters on Earth are pairs (10%). 32

Collisions Main Belt Evolution • Statistics of binaries, shapes, rotation periods, and even albedo variations are related to the • Shortly after formation (~of order 10 Ma) collisional history of the population. most of MB experienced large amount of • Size distributions of individual asteroid classes collisional fragmentation followed by mass or reggpions can have bumps. This could be a removal result of a remnant large body or because of – Event likely related to formation/migration of differences of material strength in different Jupiter populations. • Models (cf. Petit 2002) suggest that the • The current asteroid belt is likely only a remnant MB was heavily influenced by the final of an earlier population. However, it is difficult to stages of planet formation/nebular clearing figure out what the original population of objects was like (mass, composition, distribution). 33 34

Present Main Belt Features I Present Main Belt Features II Dynamical Excitation of MB Massive Main Belt Mass Loss • Current mass of main belt 10-4 • Most of the belt is -3 –10 Me dynamically (not • Discontinuity in surface thermally!) warm to hot density of SS suggests mass – Not consistent with loss in this region relative velocities needed • Accretion of largest asteroids for original accretion (Ceres, Vesta) over timescale comparable to range of – Present planetary meteoritic solidification ages perturbations even over Main belt requires primordial mass 4.5 Ga not sufficient to reservoir at least 100 times explain excitation current value (Duncan et al 1989) Weidenschilling, 1977 35 Surface density of material in solar system 36 Petit et al (2002)

6 Removing Mass & Mixing things Removing MB Mass via Planetary up Embryos: A different approach Model of Lecar and Franklin (1997); Franklin and Lecar (2000): • Formation of numerous moon – Mars – Earth – • As solar nebula decays, change in mass sized bodies in or near MB distribution in the nebula forces secular – Numerous models (eg. Chambers and Wetherill resonances to “sweep” through the primordial 2001) MB • S()Some portion (or all) controlled by Jupiter which – Forces asteroid e and i to larger values and “pumps” leads to dynamical excitation of the population the belt – This scattering produces gravitational effects on – Will remove mass – amount depends on how many entire MB population through close encounters with resonances sweep through the belt and how long they small MB asteroids (exciting them dynamically) remain in the belt – Critically depends on how long the embryo can – Drag from residual gas also helps mass removal – remain in MB before dynamical removal and how big small bodies decay, depopulating the outer belt Jupiter is as a function of time during this process

38

Mass in asteroid belt (2.1 < a < 4.0 AU) versus time for simulations with modern Jupiter and Saturn (upper curve), and slightly eccentric (e=0.1) Jupiter and Saturn (lower curve).

Chambers and Wetherill (2001) Chambers and Wetherill (2001)

Same dynamical processes which remove embryos also remove MB asteroids 40

Tagish Lake, measured LL

Asteroid Masses and Densities L Densities for most

H asteroids much lower

• Asteroid masses found CV than meteorite-analogs – From their gravitational effects on spacecraft paths CR CO – From their gravitational effects on each other Suggests lots of porosity CM

– Size of orbit and period of revolution of satellite “moons” CI

• If volume is computed then density can be found Phobos 3782 Celle V 4 Vesta V – Density provides insight into composition and structure 87 Sylvia X 22 Kalliope M 16 Psyche M – Densities generally found to be lower than meteorites 1999 KW4 S 2000 DP107 S 433 Eros S 243 Ida S 20 Massalia S 15 Eunomia S 11 Parthenope S 762 Pulcova C 253 Mathilde C 121Hermione C 45 Eugenia C 10 Hygiea C 2 Pallas B 1 Ceres G 41 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 -3 Bulk Density (gcm )

7 Discovery image (top) of the asteroid satellite of 45 Eugenia Orbit of the system shown to the right Inferred density (assuming 45 Eugenia is spherical) is -3 Britt et al. 2001 43 1.2 g cm

1991 VH binary asteroid detected based on light Origin of Binary Asteroids curve variations. • Binary NEAs show small separations, circular orbits, Primary rotates every 2.6 hours; eclipse/occultations round primaries with near critical rotation periods occur every 33 hours – Suggests these systems are formed due to tidal spin- up/distortion at time of planetary close approaches 1.2 km diameter primary and 0.5 km secondary • Main-belt asteroids show larger separations and seppyarated by 3 km larggper primar y/secondar y diameter ratios ;p; possible Most NEA binaries found to origins: be in circular orbits; – 1) creation of mutually co-orbiting fragments from a catastrophic primary very round and disruption of a parent body slow rotation – 2) reaccretion of ejecta from a major, oblique, sub-catastrophic collision between an impactor and the primary – 3) bifurcation of a (rubble-pile) parent body after rapid spin-up by a large impactor.

2000 DP107

Rotation Asteroid Spins and Shapes

• 80% of planetary bodies rotate with a period between 4 and 16 hours. There is a correlation between size and period. Smaller asteroids spin • Larger asteroids are spherical due to high gravity faster. • Smaller (~km) sized asteroids vary greatly in shape • Spin rates for large asteroids (>60km) are probably determined by their collision history since the distribution is not smooth. • Elongated shapes of some NEAs interpreted as tidal • Some asteroids can be tidatidallylly despun by small satellites. di st orti on of w eakl y h el d t ogeth er object s • If a body does not spin about one of its “principal” axes then the • Asteroids spin relatively slowly (hours – months) wobble is unstable. Wobble damping can be very slow if the rotation is • Fast spinning small bodies (<2.2h) in the NEA slow. • Bodies are most stable if they are rotating about their short axis. population may be monolithic rocks • Inertial stresses caused by motions in the body will slowly damp the • Spins determined by variations in light curves rotation until the asteroid is rotating about its shortest principal axis. The damping timescale depends on the density, radius and rigidity of the body. 47 48

8 Sources of Information about Asteroid Surfaces Only the smallest bodies rotate with periods less than ~2 hours. This is taken • Telescopes. Hemispherically averaged, to support the ‘rubble pile’ model for ambiguous info on particle size, asteroids, that is, larger asteroids cannot rotate faster than this limit or they will mineralogy, and not much else rotationally disrupt • Radar. Roughness on a human-scale and reflectivity (ice, metal, rock… ), especially for NEAs that pass close • Inferences from Meteorites. Better than nothing, but meteorites are not regolith samples • Spacecraft. Only a few studied to date; best data by far for Eros (NEAR) 49 50

•Example of an asteroid detection with the 3.6m CFHT. The asteroid is not resolved, the • Radar observation of 1999 JM8 taken on August 3, 1999 apparent with the Arecibo radar. size is the size of the • The asteroid is resolved and we can see features on it, point-spread but this technique can only be applied to asteroids which function pass quite close to the Earth. The radar power required drops like 1/r4 where r is the distance from Earth

51 52

Asteroid Physical Classification

• Different asteroid classes established based on reflectance spectra (colors) Tagish Lake • Shapes of spectra indicate major minerals present on the surface ONLY – No direct means to determine the mineral composition of the interior of an asteroid – Issues of space weathering make compositional links with meteorites uncertain – Do not get highly defined spectral absorption lines (like in a gas) due to reflectance from powdered surface • 16 Major defined classes and many more sub-divisions • As asteroid spectra become more refined and applied to smaller bodies, more asteroid groups are defined

9 Typical asteroid spectra lab spectra of minerals expected

Iron-nickel

Olivine

Pyroxene

Unfortunately many of the spectral features that tell Spinel you the most about the minerals present are in the infrared, which is very difficult to observe from the ground. As a result, many current asteroid classifications rest on a sparse number of observations in the IR

Asteroid classes The most widely used scheme is that of Dave Tholen proposed in 1984. There are 14 categories, but 3 main ones C-type (dark carbonaceous objects). This group contains about 75% of asteroids in general (includes subtypes B, F & G) S-type (silicaceous (or "stony" ) objects). This class contains about 17% of asteroids in general. M-type metallic objects, the third most populous group, include subtypes E and P and small classes that include just a few asteroids that don’t fit in the scheme above. A-type (446 Aeternitas) D-type (624 Hektor) T-type (96 Aegle) Q-type (1862 Apollo)

R-type (349 Dembowska) V-type (4 Vesta) 57

The puzzle of meteorite sources • No class of asteroids have reflectance spectra that resemble the reflectance spectra of ordinary chondrites. And yet OC meteorites are the most common meteorites. Does that mean that asteroids are NOT the primary source of meteorites? If not, what is?

Chapman 2004

59

10 Evidence related to formation of The puzzle of meteorite sources Asteroids • Space weathering is the leading theory here. • The idea is that constant bombardment by energetic • Significant gradient in asteroid types through particles in space has modified a veneer on the the MB asteroid’s surface so the colours of asteroids are different from those of meteorites (which have had any – Silicate/Metallic rich in the inner portion of the belt, such veneer burned off) more primitive, organic rich, water-rich(?) in outer • Experiments attempting to reproduce this modification in belt the lab have had some success but its not always clear – If asteroids are still sorted approximately in the that the processes are comparable (e.g. that a million same locations as they formed this represents years of low dose radiation can be replicated in a signature of original nebular temperature gradient reasonable length lab experiment) • Examination of meteorites – different heating

61 histories for different classes 62

Present Main Belt Features Stratification in MB Structure of the Asteroid Belt: E HOT Variation of Taxonomic Types... …with Distance from the Sun S • Asteroid taxonomies Despite voluminous data appear to be ordered by acquisition, almost no bias- solar distance corrected statistical studies M – Reflect original formation have been published since

ce (AU) locales? the 1980s… n – Clue to nature of asteroid Difficulty in bias corrections types? and uncertain assumptions C – Proxy for primordial solar mean this gradient may not nebula temperature be so clear… gradient Gradie, et al. (1988) Heliocentric Dista

P COLD 63

Structure of the Asteroid Belt: Variation of Taxonomic Types... Asteroid Heat Sources

Mothe, et al. (2003) • Thermal processing – two main sources: 1. Collisional (primordial/formation and recent) 2. Radioactive

Also (possibly): Magnetic Induction from early sun?

66

11 • Asteroids never exceeded the size of planetesimals and (in a few cases) Collisional Heating small proto-planets (Ceres) • The larger (>100 km) bodies Add shell of thickness dr contributing underwent heating which led to 4 mass dM to existing body of mass M differentiation due to: dm = πr 2 ρ dr 1. short-lived radioactivity (Al26 ?) 3 Gravitational potential energy converted 2. Collisions to KE in moving mass dm from infinity to • These differentiated asteroids were surface of body of mass M is: mini-planets, with iron cores and silicate mantles GMdm 16 dW = = π 2 ρ 2Gr 4dr • Some of these objects were broken r 3 apart by collisions (note: differentiated 4 M = πr 3ρ meteorites testify to this) 3 Starting from a small planetesimal with r ~ 0 to asteroid of radius R liberates potential energy of:

2 R 16 R 3 GM W = dW = π 2 ρ 2G r 4dr = ∫0 3 ∫0 5 R

67

Collisional Heating - III Collisional Heating - II Absolute upper limit to global thermal change can be • For collisional heating to be the main source for estimated by assuming ALL gravitational potential energy changed to KE goes to internal (thermal energy). thermal metamorphism, it has to occur very fast and the main body has to be big (large Recall C = dE / dT, where Cv = gravitational potential well) v R (km) dT (degrees) specific heat at constant volume, 1 0. 0007 • Neither condition is likely met for asteroids which E is internal energy and T is 2 0.0028 40.011 are parent bodies of meteorites temperature. Thus dT = dE / Cv and if we allow dE ~ W we have 8 0.045 • Collisional energy raises temperature locally, but 2 16 0.183 dT=0.6GM / (Cv R) global change is not more than a few degrees 32 0.735 even for big collisions 64 2.94 128 11.76 256 47.05 69 512 188.23 70

Shock Pressure % (N) T Increase Stage GPa Radioactive Heating S1 < 4 - 5 11.6% (257) 10 - 20 K • Short-lived nuclides abundant in early SS S2 5 - 10 34.0% (753) 20 - 50 K • Most important for thermal alteration in 26 S3 15 - 20 34.8% (770) 100 - 150 K early SS is Al

S4 30-35 12.9% (286) 250 - 350 K

S5 45 - 55 4.2% (94) 600 - 850 K 1500 - 1750 S6 70 - 90 2.5% (55) K

Stöffler, Keil, and Scott, GCA 55, 3845 & Grady (2000) 72

12 The Complete List

Which asteroids melt? How do they melt?

Thermal model for the asteroid belt assuming heating from decay of short-lived radioactive isotope, aluminum-26. Asteroids farther from the sun accreted • Shortest lifetime sets time scale of CAI formation relative to nucleosynthesis later, and incorporated less “live” 26Al. Those closer to the sun were heated to events. higher temperatures. Asteroids with diameters of 100 km within 2.7 AU of the • If 106 yr is “first enough”, CAIs were the first. Sun produced achondrites. Ice melted in bodies between 2.7 and 3.4 AU, • 244Pu, 129I, 182Hf, 60Fe require supernova. allowing aqueous alteration of chondrites. At great solar distances, asteroids 60 never warmed above the melting point of ice. • Fe requires supernova within 5(?) myr. 73 74 •We did form in an Orion-like cloud (?)

The End

Time-temperature curves plotted at various depths in the (a) uncompacted and (b) compacted H-chondrite parent bodies (Bennett and McSween, 1996). 75 76

13