Asteroids, comets, and dinosaur-killing impacts

This is chapter 9 of the Solar System’s story. It’s about those little shrimpy rocks in the Solar System — the asteroids and comets — that sometimes crash into us…

Asteroids and comets are the leftovers from the planets’ formation. They’re like the potato peels that end up on the floor instead of in the mashed potatoes. So, what have those leftovers been up to the past few billion years?

The first thing that may come to mind is that sometimes they crash into the planets.  The Moon is covered in craters, and each represents the impact of an asteroid or comet.  And for every object that hits the Moon, about twenty hit the Earth! Earth is not covered in craters, because they are covered up by erosion and plate tectonics such that only a small number are visible today. Our planet’s surface belies its violent history.

While they are thought of as separate beasts, it’s worth pointing out that asteroids and comets are closely related.  Asteroids are generally thought of as rocky and comets as icy, but that distinction does not always hold true.  There exist ‘active asteroids’ in the main asteroid belt that exhibit comet-like activity, as well as purely rocky objects on orbits similar to comets’.  

The main division between asteroids and comets is their orbits.  In this chapter I’ll discuss their rough properties and evolution and explain what causes them to – on rare occasions – crash into a planet. 


The asteroid belt is a giant swath of Solar System real estate, extending from Mars’ orbit out to Jupiter’s.  The vast majority of the asteroids are found in the main belt, located between about 2.1 and 3.2 astronomical units.  The belt contains less than one one-thousandth of Earth’s mass, although there are still more than a million stadium-sized asteroids (larger that 1 kilometer in diameter).  The belt has a lot of substructure in terms of compositional classes of asteroids and asteroid ‘families’.  Of particular note, the inner parts of the main belt are dominated by relatively dry asteroids and the outer belt by more volatile-rich and water-rich asteroids.

The asteroid belt (including Jupiter’s co-orbitals). Credit: Wikimedia Commons.

Ideas for the origin of the belt are divided. Some (in particular the classical modal) envision a very massive primordial belt that was strongly depleted during planet formation. Others (including the Low-mass or empty asteroid belt models) envision an empty primordial belt that was populated during planet formation (discussed in detail in chapter 6).

The asteroid belt is full of holes.  In 1866, Daniel Kirkwood noticed gaps in the distribution of the asteroids’ orbital semimajor axes and eccentricities.  These gaps are located at orbital resonances with Jupiter, and are called the Kirkwood gaps. The image below shows the main Kirkwood gaps such as the 3:1 resonance (where an asteroid completes exactly three orbits for every orbit of Jupiter) at 2.5 astronomical units. There also exist many narrower gaps that are not visible in this image associated with weaker resonances.

Credit: Wikimedia commons.

An asteroid in a Kirkwood gap is unstable. That is why the gaps are empty: any asteroids that started off in those gaps are long gone.

The asteroids in the present-day belt are, quite simply, the ones that survived.  This is analogous to a famous example of survivorship bias.  Bomber planes returning from their runs under enemy fire showed an interesting pattern of bullet holes. They had bullets all over their fuselage except in certain specific locations; planes that were hit in those locations presumably did not survive. Likewise, asteroids that entered the Kirkwood gaps did not survive.  (It’s not a perfect analogy, because when an asteroid enters a Kirkwood gap it doesn’t destroy the whole asteroid belt).

Simulated locations of bullet holes on bomber planes returning to base. Credit: Wikipedia.

What exactly happens to an asteroid in an unstable resonance (a Kirkwood gap)? On a timescale of about a hundred thousand years, the asteroid’s orbit becomes stretched out (its eccentricity increases). Eventually the asteroid’s orbit crosses the orbit of either Jupiter or Mars. If it encounters Jupiter, the asteroid will wind up being strongly gravitationally scattered and ejected into interstellar space.

If the asteroid encounters Mars, the story is a little more interesting. It can collide with Mars, but more likely it will be gravitationally deflected by Mars.  The asteroid’s orbital path will be quite chaotic, as it will be continually gravitationally scattered by Earth and Venus, and it may end up colliding with one of the rocky planets (or the Moon).  Sometimes asteroids’ orbits become stretched-out so quickly that they never have a close encounter with Mars and are instead driven into the Sun. 

Within about ten million years, any asteroid in a Kirkwood gap is removed – it either collides with a planet or the Sun, or is ejected from the Solar System. But the planets and asteroids finished forming 4.6 billion years ago. So, why do we still have objects crashing into the planets in recent times (astronomically speaking)? Resolving this conundrum requires understanding the Yarkovsky effect. 

The Yarkovsky effect starts from two pretty obvious facts. First, asteroids are heated by the Sun.  Second, asteroids spin. The part of an asteroid that gets most heated is the part facing the Sun. Asteroids take some time to cool down, and, given their spin, they radiate away their heat in a different direction – not toward the Sun but a little to one side of the Sun, depending on the direction of the asteroid’s rotation.

Radiation from an asteroid’s surface is like a teeny little rocket thrusting in that direction.  Depending on the direction of this tiny little thrust, the asteroid can either speed up (for prograde spin) or slow down (for retrograde spin) along its orbit. This causes the asteroid to gain or lose orbital energy; its orbit grows or shrinks.  [Technical note: a separate but interconnected thermal effect, the YORP effect, affects asteroids’ spin vectors.]

Illustration of the Yarkovsky effect, which widens or shrinks asteroids’ orbits. Credit: R. Binzel/Nature.

The Yarkovsky effect has been observed in two ways.  First, it has been directly measured using radar observations of asteroids.  Second, its effects have been seen on the orbits of fragments of collisions between asteroids.  Since the asteroid’s radiation only provides a tiny amount of thrust, the Yarkovksy effect is much stronger for smaller asteroids. A collision between two asteroids produces fragments of all sizes, whose orbits remain clumped and are identified as asteroid ‘families’.  The Yarkovsky effect causes small fragments drift away from the point of impact faster than larger ones, in a distinctive pattern that can be used to quite precisely date the timing of the asteroid collision itself. 

Collisions between asteroids are therefore responsible for the continued bombardment of asteroids on the planets.  The fragments from an asteroid collision spread out due to the Yarkovsky effect, with the smallest fragments spreading fastest.  In time, some fragments may run into an unstable resonance (a Kirkwood gap).  When a fragment enters the gap, its orbit becomes unstable and it may be kicked into the rocky planet region, to eventually collide with a planet. From an Earth impact point of view, the most dangerous asteroid collisions are those that take place close to a strong unstable resonance (so that fragments, including larger ones, can become unstable) in the inner belt (so that, when fragments’ orbits become unstable, they are kicked inward toward the rocky planets rather than being ejected by Jupiter).


Comets are just so beautiful and charming.  They are the belles of the ball – when they enter the room, they get everyone’s attention! 

Their brightness comes from the release of gases as the comet heats up and enters the inner Solar System.  The actual solid part (the nucleus) of most comets is usually pretty shrimpy, less than 1 kilometer in size.  But the bright part of a comet (the coma) can be a hundred thousand to a million times larger, and the tails extend much farther still.

Credit: Roen Kelly/Discover.

Comets get so bright because they have been holding on to their ices (the source of outgassing) far from the Sun, out in the cold refrigerator of space.  It’s when they enter the inner Solar System that they start to vaporize and look so bright and charming.  If comets stuck around in the inner Solar System for a long time, their ices would all vaporize and they would just look like asteroids (like D- or C-types, to be unnecessarily technical).

So where do comets come from anyway? 

There are three main classes of comet orbits, which can be separated by their orbital periods (the time to complete one orbit):

  • “Jupiter-family” comets, with orbital periods less than 20 years, and usually with orbits close to the plane of the planets’ orbits.  The target of the Rosetta space mission, comet 67P/Churyumov-Gerasimenko, is perhaps the most famous example.
  • “Halley-type” comets, with orbital periods of 20-200 years and orbits that can point significantly out of the plane of the planets’ orbits. Halley’s comet, which returns every 86 years (next stop: 2061), is the most famous example.
  • “Long-period” comets, with orbital periods of up to millions of years, have orbital planes that are totally uncorrelated with the plane of the planets’ orbits.  A famous example – and my personal favorite comet – is Hale-Bopp, an extremely bright comet that zoomed by while I was in college and helped kindle my interest in astronomy.

Here is a very simple view of the breakdown between the classes of comets:

The three classes of comets reflect their source regions.  Jupiter-family comets originate in the Kuiper belt, the population of icy bodies beyond Neptune with orbits close to the plane of the planets.  Jupiter-family comets become unstable and then gravitationally interact with Neptune.  Neptune gives them a series of gravitational kicks that ends up shrinking their orbits to interact with Uranus.  The process repeats with Saturn and Jupiter. During the phase while a comet is interacting with Jupiter, it can pass close to the Sun (near Earth’s orbit in some cases) and shine brightly with tails and a come.  Eventually, Jupiter-family comets are ejected from the Solar System by Jupiter (no fairy-tale ending, sadly).

Halley-type comets originate in the scattered disk, a population of icy bodies on orbits beyond Neptune that are both more distant than the Kuiper belt and have more stretched-out orbits that are inclined with respect to the plane of the planets.  After strong perturbations by a planet, Halley-type comets directly enter the inner Solar System without being traded off between the planets like Jupiter-family comets. 

Long-period comets originate in the Oort cloud, which extends to the outer limits of the Solar System more than 100,000 astronomical units from the Sun. The Oort cloud is truly spherical such that comets can originate in any direction.  Comets’ orbits are so wide that they are just barely gravitationally bound to the Sun.  Small perturbations from either a star passing nearby or the galactic gravitational field can strongly change a comet’s orbital shape, sending it plunging into the inner Solar System.

You might be wondering, why are icy bodies in the Solar System spread between the Kuiper belt, scattered disk, and Oort cloud?

A large fraction of the planetesimals that formed in the Sun’s gaseous disk were probably icy, including all of those that originated past the snow line (see chapter 2).  Of course, most of those ended up as food for the cores of the giant planets.  The planetesimals that avoided being eaten by the growing planets were preferentially the ones that formed farthest out and latest.  Those planetesimals may have had nice, calm existences … until the giant planet instability (see chapter 7).

When the giant planets’ orbits became unstable, almost all of the icy outer planetesimals were destabilized.  Planetesimals were scattered by the planets in all directions.  Most were ejected from the Solar System completely.  Some were scattered onto extremely stretched-out, wide orbits and almost ejected, only to have their orbits changed by external forces. A small gravitational kick from a passing star or gentle nudging from the gravitational field of the galaxy protected these objects from being scattered by the planets, and randomized their orbital orientations.  This was the origin of the Oort cloud.  Some planetesimals were scattered by Neptune onto stretched-out orbits and then left stranded in place; this was the origin of the scattered disk.  The Kuiper belt, meanwhile, is thought to represent the outermost, thin slice of icy planetesimals that was never scattered by the planets.

Comets rarely hit the planets. A comet has a chance to hit a planet once each orbit, as long as its orbit is crossing the planets’.  Yet each comet has a finite (quite short) lifetime on that particular orbit.  Combining these factors, comets’ impact probabilities are much smaller than asteroids’.  The most likely setup for impact is when a long-period comet passes close to the Sun and is tidally disrupted into fragments, each of which has a chance to collide with a planet. Comet impacts do happen on astronomical timescales and can have dramatic consequences, since the typical impact speed of a comet (especially a long-period or Halley-type comet) is much higher than for an asteroid.

The dinosaur-killing impact (and others like it)

The most famous impact of all is the one that killed the dinosaurs. There is abundant evidence from the fossil record for a global extinction event, and it is estimated that 75% of all species on Earth were rendered extinct.  The official name is the K/Pg extinction because it separates the Cretaceous (K) and Paleogene (Pg) geological periods.  (Note: this is sometimes called the Cretaceous-Tertiary, or K-T extinction).

The evidence for an impact origin of the K/Pg comes from the geological record. A globally-deposited layer of rocks rich in Iridium provided one hint, because Iridium is highly siderophile and very rare in Earth’s crust but much more common in asteroids. The boundary layer also contained beds of spherules, crystallized from molten rock, as well as shocked quartz.  The Chicxulub crater on the Yucatan peninsula in Mexico was proposed as the K/Pg impact crater in a famous study by Alvarez et al in 1980

Published in the New Yorker August 23, 1999 (credit: Gahan Wilson)

The source of the K/Pg impactor is an interesting whodunit. To explain the amount of Iridium measured, the K/Pg impactor must have been about 10 kilometers in diameter. A study in 2007 showed that the breakup of the asteroid Baptistina in the inner part of the asteroid belt about 160 million years ago would, through the Yarkovsky effect, have significantly increased the chances of a 10-km impactor on Earth. However, comparing astronomical and geological data suggests that Baptistina was not the source.  Rather, the impactor was likely volatile-rich, similar to meteorites from asteroids in the outer parts of the asteroid belt. A highly contested study in 2020 proposed that a fragment a long-period comet could have been the impactor, but that idea was quickly and convincingly falsified. A study in 2021 demonstrated, instead, that the Yarkovsky effect can naturally explain the impactor as an asteroid from the outer, volatile-rich parts of the belt.

Earth continues to be hit by asteroids and comets.  Meteors (shooting stars) are the smallest of impacts, created by the burnup of grain-of-sand-sized pieces of cometary space dust burning up in the atmosphere. The rate of impacts scales with the impactor size: small impacts are far more common that the larger, devastating ones portrayed in movies like Armaggedon and Deep Impact (which I love despite their scientific shortcomings). Events like the 12 megaton Tunguska impact in 1908 that flattened almost 100 million trees in Russia only happen every few hundred years (usually over the oceans).  Luckily for us, dinosaur-killing collisions only happen every hundred million years on average. 


The TL;DR version of this post: Asteroids and comets are the leftovers of planet formation.  Impacts on the planets happen when objects are pushed into unstable resonances with the giant planets (Kirkwood gaps).  The dinosaur-killer was a volatile-rich asteroid.

Additional Resources


2 thoughts on “Asteroids, comets, and dinosaur-killing impacts

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s