Giant planets are thought to form in two possible ways. The
first method involves accreting a solid core and once the solid core attains
~10 Earth masses, it becomes massive enough to accrete hydrogen and helium in a
runaway process to form a giant planet. This mode of giant planet formation is
known as “core accretion”. In the Solar System, the giant planets Jupiter and
Saturn are believed to have formed via core accretion. The second mode of giant
planet formation involves a protoplanetary disk becoming gravitationally
unstable and collapses directly to form a giant planet. This is known as “disk
instability”.
Figure 1: Artist’s impression of a giant planet. Giant
planets range in mass from a fraction of Jupiter’s mass to tens of Jupiter’s
mass.
In the core accretion scenario, the solid core accretes gas
through an accretion disk. This process cools the gas, causing it to lose much
of its initial entropy and forms a giant planet that has low initial entropy (i.e.
a “cold-start”). For disk instability, the gas that collapses directly to form
a giant planet retains most of it intitial entropy, resulting in high initial
entropy (i.e. a “hot-start”). If the age of a giant planet is known well
enough, it might be possible to distinguish a “cold-start” from a “hot-start”
based on observables such as the giant planet’s entropy, radius, effective
temperature and spectrum.
However, the radiating efficiency of the gas could
complicate the “cold-start” / “hot-start” idea. It can cause a “cold-start”
that might not be as cold or a “hot-start” that might not be as hot, leading to
what could be thought of as a “warm-start”. For instance, the accreting gas in
the core accretion scenario may not cool as efficiently and the giant planet
formed at the end of the accretion process would have a “cold-start” that is
not as cold. On the contrary, the gas in the disk instability scenario may cool
more efficiently than thought, resulting in a “hot-start” that is not as hot.
A giant planet formed by disk instability is expected to have
higher entropy, larger radius and higher effective temperature than if it had formed
by core accretion. Basically, disk instability corresponds to a “hot-start”,
while core accretion corresponds to a “cold-start”. Observable differences
between “cold-start” and “hot-start” scenarios can be used to distinguish
whether a giant planet formed by core accretion or disk instability. These
observational differences are substantial when the giant planet is young and
diminish after several tens of millions of years.
When a giant planet cools (i.e. loses entropy), its radius
shrinks and its effective temperature drops. More massive giant planets retain
heat longer than less massive ones, and so evolve slower. As “cold-start” and
“hot-start” giant planets of equivalent masses cool over time, their entropies,
radii, and effective temperatures gradually converge (Figures 2, 3 & 4).
For a giant planet with 10 Jupiter masses, the “memory” of the initial
conditions (i.e. whether the giant planet had a “cold-start” or “hot-start”) is
lost within a few hundred million years after formation. For lower mass giant
planets, the convergence is more rapid because they cool more quickly and the
initial differences between “cold-start” and “hot-start” models is smaller for
them.
Figure 2: Evolution of entropy for giant planets with 1, 2,
5 and 10 times Jupiter’s mass. Spiegel & Burrows (2012).
Figure 3: Evolution of radius for giant planets with 1, 2, 5
and 10 times Jupiter’s mass. Spiegel & Burrows (2012).
At early times, the difference between “cold-start” and
“hot-start” models can be quite large. For a giant planet with ~5 times
Jupiter’s mass, the differences can persist for ~100 million years. For lower
mass giant planets with ~1 to 2 times Jupiter’s mass, the differences fade
within ~10 to 30 million years. During the first few million years after
formation, giant planets that started hot (i.e. “hot-start”) can be ~10 to
1,000 times more luminous than those that started cold (i.e. “cold-start”),
depending on the giant planet’s mass and spectral band.
Reference:
Spiegel & Burrows, “Spectral and Photometric Diagnostics
of Giant Planet Formation Scenarios”, 2012 ApJ 745 174