Kuiper Belt: Dynamics 599
greater than about 30 km to form in the current Kuiper
Belt, at least by pairwise accretion, over the age of the solar
system. The current surface density of solid material is too
low to accrete bodies larger than this size. However, the
models show that objects the size of 1992 QB 1 could have
grown in a more massive Kuiper Belt (provided that, as al-
ready said in Section 3, the mean orbital eccentricities of
the accreting objects were much smaller than the current
ones). A Kuiper Belt of at least several Earth masses is re-
quired in order for 100 km sized objects to have formed.
The same applies even more strongly to the accretion
of Pluto and Charon. For those two bodies to have grown
to their current sizes in the trans-Neptunian region, there
must have originally been a far more massive Kuiper Belt.
A massive and dynamically cold primordial Kuiper Belt is
also required by the models that attempt to explain the
formation of the observed numerous binary Kuiper Belt
objects.
Therefore, the general formation picture of an initial
massive Kuiper Belt appears secure, and understanding the
ultimate fate of the 99% (or 99.9%) of the initial Kuiper Belt
mass that appears to be no longer in the Kuiper Belt is a
crucial step in reconstructing the history of the outer solar
system.
6. Ecliptic Comets
As described in Section 1, the current renaissance in Kuiper
Belt research was prompted by the suggestion that the
Jupiter-family comets originated there. We now know that
there are mainly two populations of small bodies beyond
Neptune: the Kuiper Belt and the scattered disk. Which
one is the dominant source of these comets?
To answer this question, we need to examine a few con-
siderations on the origin of the scattered disk. We have seen
in Section 3 that the bodies in the scattered disk have in-
trinsically unstable orbits. The close encounters with Nep-
tune move them in semimajor axis, until they either evolve
into the region witha<30 AU or reach the Oort cloud at
the frontier of the solar system. In both of these cases, the
bodies are removed from the scattered disk. Despite this
possibility of dynamical removal, we still observe scattered
disk bodies today. How can this be?
There are a priori two possibilities. The first one is that
the scattered disk population is sustained in a sort of steady
state by the bodies escaping from the Kuiper Belt. This
means that on a timescale comparable to that for the dy-
namical removal of scattered disk bodies, new bodies enter
the scattered disk from the Kuiper Belt. For example, a
similar situation occurs for the population of near-Earth as-
teroids (NEAs). NEA dynamical lifetimes are only of a few
million years because they intersect the orbits of the terres-
trial planets. Nevertheless, the population remains roughly
constant because new asteroids enter the NEA population
from the Main Asteroid Belt at the same rate at which old
NEAs are eliminated.
The second possibility is that the scattered disk that we
see today is only what remains of a much more numerous
population that has been decaying in number since plan-
etary formation. Numerical simulations show that roughly
1% of the scattered disk bodies can survive in the scattered
disk for the age of the solar system. Thus, the primordial
scattered disk population should have been about 100 times
more numerous.
Which of these possibilities is true? In the first case,
we would expect that the Kuiper Belt is much more pop-
ulated than the scattered disk. For instance, the Asteroid
Belt contains about 1000 times more objects than the NEA
population, at comparable sizes. However, observations in-
dicate that the scattered disk and the Kuiper Belt contain
roughly the same number of objects. Thus, the second pos-
sibility has to be true. Scattered disk objects most likely
formed in the vicinity of the current positions of Uranus
and Neptune. When these planets grew massive, they scat-
tered them away from their neighborhoods. In this way, a
massive scattered disk of about 10M⊕formed. What we see
today is just the last vestige of that primordial population,
which is still decaying in number.
The fact that the scattered disk is not sustained in steady
state by the Kuiper Belt, but it is still decaying, implies that
the scattered disk provides more objects to the giant planet
region (a<30 AU) than it receives from the Kuiper Belt.
Thus, the outflow from the scattered disk is more important
than the outflow from the Kuiper Belt. This implies that the
scattered disk, not the Kuiper Belt, is the dominant source
of Jupiter family comets.
A significant amount of research has gone into under-
standing the dynamical behavior of objects that penetrate
into thea<30 AU region from the scattered disk. These
studies show that the encounters with the planets spread
them throughout the planetary system. These objects are
usually called ecliptic comets, even if at large distances from
the Sun they typically do not show any cometary activity.
The distribution of these objects as predicted by numerical
integrations is shown in Fig. 8.
The ecliptic comets that get close to the Sun become
active. When their semi-major axis is smaller than that of
Jupiter, they are called Jupiter-family comets. It is some-
what surprising that about a third of the objects leaving
the scattered disk in the simulations spend at least some
of their time as Jupiter-family comets. The Jupiter-family
comets that we see today are, in majority, small,R
̃<10 km.
However, if our understanding of the size-distribution of
these objects is correct (see Section 5), we should expect to
see a 100 km sized Jupiter-family comet about 0.4% of the
time. What a show that would be!
Those ecliptic comets between Jupiter and Neptune are
called the Centaurs (only the largest of which are observ-
able). The simulations predict that there are∼ 106 ecliptic