A Revision of the Exploded Planet Hypothesis

The Problem.

In network discussions (on sci.astro on the Internet) of the exploded planet hypothesis, respondents have continued to raise objections about associating the explosion of a major planet in the asteroid belt – astronomically dated at 3.2 million years ago (Ma) – with the K/T boundary in stratigraphic layers on Earth – geologically dated at 65 Ma – because of the date discrepancy. Extensive evidence for the exploded planet hypothesis (eph) was presented in Dark Matter, Missing Planets and New Comets. This date discrepancy is one of only two lines of evidence, out of over one hundred, that does not fit the hypothesis.

As readers will be aware, the eph predicts a priori almost all the phenomena known to be associated with the K/T boundary event. For example (thanks to S Krueger for some items on this list):

• far more iridium than can be explained by terrestrial processes or slow accretion from space
• other siderophile elements, consistent with one or more major impacts
• microtectites in the boundary clay
• a verified global extent and discreteness
• shocked quartz well beyond what volcanism can produce
• abundant carbon ash
• mass extinctions occurring mainly within inches below the boundary layer
• “event beds” around the Caribbean Sea
• numerous “hot zones” of radioactivity at the top of the Cretaceous
• the Deccan Traps, and the onset of an extended period of unparalleled global volcanism
• atmospheric and ocean compositional changes
• a single global fire

However, the dating of the eph and the K/T boundary event continue to differ by an amount too large for a reconciliation through any known mechanism. The geological dating of the K/T event at 65 Ma by radiometry is confirmed by multiple independent methods yielding similar dates, which makes calibration errors in using the method unlikely. I considered that radiation from the explosion might have falsified the radiometric clocks in rocks used for this dating. But it is difficult to invent a realistic scenario in which the stratigraphic layers on Earth both below and above the K/T boundary layer, typically 5 km below the Earth’s present surface, could still show a continuous pattern of radiometric ages from zero near the present surface to hundreds of millions of years at depths greater than the K/T boundary. So I then considered the possibility that all rocks suitable for radiometric dating originated on the planet that exploded, and simply mixed with a large amount of inert terrestrial matter on impact. However, the predicted discontinuity of radiometric dates at the K/T boundary under that scenario is not observed.

Meanwhile, the astronomical dating of the eph event has proved equally resistant to change. Especially to be noted:

• The measured periods of new comets, the principal astronomical method of dating the eph at 3.2 Ma, are uncertain in the aggregate by less than 5%.
• Undiscovered mass in the outer solar system, such as the hypothetical Planet X, would alter the inferred date, but by only a minor amount.
• Galactic tidal forces and passing stars would have completely eliminated visible comets that originated from an eph-type event unless that event occurred within the past 9-10 million years.
• The cosmic ray exposure ages of many stony meteorites are only millions to tens of millions of years, but not clustered near 65 Ma.
• The thousands of asteroids over 1 km diameter and millions of smaller ones in Earth-crossing orbits require a recent source, since the Earth tends to clear out such objects with a scale time scale of 30 million years.
• The puzzle that the current impact rate on Earth and Moon exceeds the fresh crater counts by an order of magnitude is resolved only if the impact rate has been anomalously high for just the past few million years, but not much longer.
• The morning/evening asymmetry in meteorite falls on Earth seems to imply a source younger than 10 million years.
• There is evidence that the local interstellar medium has been altered by a nearby explosion within the past 4 million years.

Arguing against the astronomical date is the inferred radioactive Aluminum-26 abundances in certain meteorites. If this had been deposited just 3.2 million years ago, a larger amount of Al-26 would remain live in those meteorites today, only about 4½ half-lives later. Instead, nearly all of it has already decayed into Magnesium-26. But the overwhelming bulk of the evidence requires the indicated young age.

A Revision of the Hypothesis

However it may have formed, the main asteroid belt is an existing real feature of our solar system. Among the many lines of evidence that an exploded planet did happen are these:

• Asteroids occupy the entire volume of phase space (the full range of positions and velocities) between Mars and Jupiter that is stable against planetary perturbations over millions of years. And their mean relative velocities, averaging 5 km/s, are too high to result from collisions, fragmentation, or planetary perturbations.
• The distribution of asteroid orbital elements contains “explosion signatures” similar to those first catalogued for fragments of artificial Earth satellites that blew up in orbit.
• The cosmic ray exposure ages of stony meteorites are generally only some millions of years, not billions – distinctly shorter than the mean time between collisions.
• Comets, which are spectrally, photometrically, and chemically similar to asteroids, have all their major properties explained better by an explosion origin than by the solar nebula model. An orbital “traceback” indicates origin at a common time and place in the inner solar system.
• Comets and asteroids seem to have orbiting debris clouds. No other known method can produce such clouds except the explosion of a larger parent body.
• There is a pattern by black residue on moons and planets in the outer solar system consistent with an explosion blast wave traveling through it.

There are many other lines of evidence for the eph mentioned in Dark Matter, …, including a few that, although not so easy to state in one sentence, nonetheless seem unambiguous and exempt from questions of interpretation or systematic error of observation. The evidence from comet fragment split velocities is particularly diagnostic, and favors the eph over other possibilities at the four sigma level.

But however strong the case, when one or more lines of evidence does not fit a hypothesis, the need for ad hoc explanations is usually a sign of a theory in trouble. The planetary explosion hypothesis still fits the totality of evidence far better than the solar nebula hypothesis, but the eph in its form up to now is perhaps not quite correct.

It was mentioned already in Dark Matter, … that there may have been an even earlier planet (“Planet V” -- the original fifth planet) that also exploded. For example, the main belt consists of two dominant spectral classes of asteroids. Roughly 80% of them are class “C” because they resemble carbonaceous meteorites found on Earth. Most of the remaining 20% are class “S”, somewhat reminiscent of the silicaceous rocks found commonly on the Moon. The C-type asteroids are found predominantly in the middle and outer main belt, whereas the S-types are concentrated toward the inner main belt. This is one of a few lines of evidence mentioned in earlier writing that suggests two different origin events for the main belt.

The recent discovery of another belt of asteroid-like objects orbiting the Sun beyond Neptune (see accompanying article) has forcibly brought back to our attention the possibility that planetary explosions may be relatively common events, as uncomfortable as that thought is. My prejudice until now was to treat the eph event as unique, thinking that was the simplest form of the hypothesis in the sense of Occam’s Razor. Upon further reflection, the hypothesis is, if anything, harder to understand if the planetary explosion was a unique event than if it refers to a common process in the galaxy. Therefore, I now consider that there may have been at least two bodies orbiting between Mars and Jupiter (e.g., Planet V and Planet K, as described in Dark Matter, …) that exploded at different epochs. Indeed, if it turns out that the trans-Neptunian belt is a third example of an exploded body in our solar system, it might even be suggested that explosion is a not-infrequent end state for masses of lunar or planetary dimensions everywhere in our galaxy.

To gain anything from this multiple exploded planets hypothesis (“meph”), we must associate the K/T boundary event in the geological record at 65 Ma with an earlier explosion of a planet-size mass. Then the explosion at 3.2 Ma would have been of a smaller mass, small enough to leave only minimal evidence in the geological record on Earth, most of which was then extinguished by the onset of glaciation that the explosion most probably triggered. The principal evidence seemingly requiring at least two explosion events at different epochs is as follows:

• No major geological layer occurs at the astronomical date of the most recent explosion, 3.2 Ma as measured by comet orbits, although the current set of glaciation cycles on Earth began at about that time. A major geological layer does appear at 65 Ma, too long ago for comets from it to return to the planetary region.
• Achondritic, stony-iron, and iron meteorites found on Earth are from a chemically differentiated source body depleted in volatiles. Chondritic meteorites are from an undifferentiated source body with solar-like abundances of volatiles. It is not easy to get both types of meteorites from a single parent body.
• Oxygen isotope ratios suggest at least two distinct parent bodies of meteorites if pure oxygen-16 is created in the explosion, and four (possibly more) if it is not.
• Large asteroids come in two main classes, C (“carbonaceous”) and S (“silicaceous”). Other properties such as albedo and mean distance from the Sun are likewise divided into two major groupings along class lines.
• Cosmic ray exposure ages of chondritic meteorites have a peak very roughly at 4 Ma, and a tail out to 60 Ma or so. Iron meteorites at the other extreme have exposure ages in the hundreds of millions of years range.
• Meteors smaller than 50 meters are too numerous to be part of the same population as main belt asteroids. [See Remarkable Paper 95.15 in this Bulletin.]

Therefore, we consider that there may have been more than one explosion of a significant mass in the solar system during the past half billion years.

Details of the Revised Hypothesis

In tying up loose ends it should not be forgotten that, in the geological record on Earth, by far the largest mass extinction event occurred at about 250 Ma at the Permian/Triassic boundary. The K/T boundary event is only the second largest extinction event in geological history.pacerun: This earlier P/T event included major, long-term volcanism in Siberia, and the extinction of perhaps 90% of all marine species and 70 percent of all land vertebrates, along with most of the terrestrial plant life. A 100-meter drop in sea levels occurred at around the same period. Other lesser extinction events are seen in stratigraphic layers dated at about 500 Ma, 400 Ma, and 200 Ma. Volcanism is often suggested as the cause of these events; but what precipitates the volcanism on such a globe-altering scale? Perhaps geologists should not forget to consider the possibility of exogenous events as triggers, just as a single asteroid impact is now thought by many mainstream geologists to have triggered the K/T boundary event at 65 Ma. Major impacts of single bodies, planetary explosions, nearby supernovae, giant solar flares, and passages of the solar system through giant molecular clouds in the galaxy are obvious candidates that should be considered in each instance.

Was the K/T boundary event the result of a single asteroid impact (causing the 200-300 km diameter Chicxulub crater in Mexico) or something more? We note the following points as evidence that it was something more:

The global set of craters Manson, Karn, Kamensk, Gusev, and another impact in the Pacific Ocean apparently all date from close to the same epoch. However, the diameter and abundance of quartz grains are larger in western North America than elsewhere in the world, suggesting that the single largest impact was the Chicxulub event. [See Remarkable Papers 95.13 and 95.14 in this Bulletin.]

• The K/T boundary consists of two distinct claystone layers, the upper (soot, iridium) one with shocked grains, the lower one without.
• Gorceixite (altered tektites with swirl patterns) is segregated within each layer, suggesting that different impact events formed these glassy beads.
• A single bolide impact cannot simultaneously explain the pattern of major floral extinctions on land and other extinctions at sea.
• Sediments in Cuba range from 5 to 450 m thick, probably from a giant wave. The (upper) ejecta layer is 50 cm thick in nearby Haiti, far more than at any other site, suggesting a major impact within 1000 km, which would be far from the Chicxulub crater in Mexico.
• The K/T boundary layer is apparently absent from the Antarctic regions. Just as for the Sun, planets spend up to six months continually below the horizon as seen from each polar region alternately. So the boundary event apparently affected the entire globe except for the south polar regions. This pattern suggests multiple impacts from an exogenous source over of period of at least one day.

Considering these factors arguing against a single impact, noting the strong evidence for at least one planetary explosion event, and remembering the earlier list of predictions made by the eph that are fulfilled at the K/T boundary on Earth, we conclude that the explosion of a solar system body was the most probable cause of the K/T boundary event at 65 Ma. The earlier P/T boundary event at 250 Ma may also have been caused by the explosion of another solar system body, either larger or much closer than the K/T boundary source body. Other geological extinction events may have been caused by single asteroid impacts, or by other types of cosmic catastrophes.

So tentatively, I associate the earlier, larger mass-extinction event at 250 Ma with the explosion of Planet K in the main asteroid belt, with iron meteorites (because of their long cosmic ray exposure ages), and with most cataloged main-belt asteroids. This event occurred so long ago that it gives Mars enough time to clear out most Mars-crossing asteroids from the main belt – thereby neatly explaining the one other line of evidence that did not fit the original hypothesis. Likewise, only a minimal number of such iron bodies should remain in Earth-crossing orbits today because of planetary perturbations, consistent with observations. The carbonaceous coating over most of these asteroids presumably resulted from a later explosion event. The finding of magnetic fields in two asteroids by Voyager shows that such bodies can contain iron.

And I tentatively associate the smaller event at 65 Ma with the explosion of Planet V in the inner asteroid belt, with achondritic and stony-iron meteorites (which have younger exposure ages than iron meteorites, but are also differentiated and apparently came from a planet-sized body), and with many inner-belt asteroids. Comets created by any explosion older than 10 Ma would have long since vanished because of galactic tides and passing stars. The population of Earth-crossing asteroids from this event would now be reduced to a fraction of its original numbers. And this may have been the event that delivered large quantities of water to the inner solar system, most notably to Mars (which may have still been one of Planet V’s moons at the time of the explosion).

The event at 3.2 Ma that resulted in all comets that survive to the present must have been the explosion of a much smaller body in the main asteroid belt. I associate this latest explosion with chondritic meteorites (which are undifferentiated, and must therefore have come from a smaller parent body), with most present-day Earth-crossing asteroids, and with many main-belt asteroids smaller than 10 km. The parent body was apparently a moon of another planet (presumably, of the former Planet K) that never became hot enough during formation to melt, lose volatiles, and separate elements. This last explosion probably also produced the carbonaceous chondrite meteorites, and the carbonaceous dust that now coats most other main-belt asteroids and most other airless bodies in the solar system. In many respects, the body that exploded perhaps resembled the largest asteroid, Ceres, which we have previously suspected of being a former moon of Planet K as well. If this line of conjecture is correct, Ceres should be examined closely by spacecraft landers as a possible candidate for a future explosion event. (For example, it might radiate more heat than it takes in from the Sun.)

Two additional lines of evidence support the idea that the latest explosion was of a body smaller than a major planet. For one, the number and size of Earth-crossing asteroids suggest a parent-body of only perhaps 1000 or so kilometers diameter, even allowing for most of the material being blown out of the solar system by the explosion. For another, one can use the number and size of observed new comets to estimate the original pre-explosion mass. We can do this by computing the approximate fraction of fragments from the original mass that could end up in orbits with the very narrow range of periods we observe as new comets. That fraction is about 10-6 when the spread in periods is one century. The estimated mass of all new comets discovered in the past century is of order a few times 10¹⁸ g. So allowing a bit for discovery incompleteness, the original mass that exploded to produce those comets would have been 0.001 Earth-mass or so, about the size of Ceres, or perhaps of a typical planetary moon.

This multiple explosion variant of the hypothesis answers many questions that have come up in discussions of the topic. For example, the K/T boundary layers are sometimes buried 5 km deep on Earth, yet carbonaceous material supposedly from the same explosion can be found on the surfaces of bodies such as Saturn’s moon Iapetus, Neptune’s moon Triton, and Pluto. We now see that evidence for earlier explosion events will have generally become buried on most bodies by accretion of cosmic dust, small impacts, and resurfacing from internal processes. Only debris from the latest explosion at 3.2 Ma will still lie on the surfaces of most solar system bodies.

Similar material must reside on the surface of the Moon as well. A likely candidate may be the so-called KREEP (from the chemical symbols for potassium, rare earth elements, and phosphorus) component in the lunar soil because of its uranium, thorium, and rare-earth-elements content.

The following is a more complete description of the effects of explosion events in general on solar system members in general. A planet explodes in the asteroid belt. Ejected with highest speeds is what I call the “blast wave”, consisting of completely vaporized material mixed with ash and soot. This wave spreads radially as it travels outward. As a consequence of this spreading, the blast wave should take several days to pass Earth, and about two weeks to pass Saturn. It therefore coats planets and moons only on their faces exposed to the explosion over the period of arrival of this blast wave -- days to weeks. In some cases such as Iapetus, nearly an entire hemisphere remains shielded from the blast wave. In other cases such as Triton, one polar region is shielded.

Following behind the blast wave is a wave of chunks of matter -- material that was fragmented but not vaporized. Some of these fragments become today's asteroids, meteoroids, and comets. The explosion should tend to propel fragments with equal energy per unit area. So the smallest chunks typically get propelled the fastest, and the largest chunks the slowest, since area goes with radius squared but mass with radius cubed.

Most fragments then proceed to escape the solar system. Meanwhile, all planets and moons are exposed to just one major bombardment shortly after the explosion as these high-velocity chunks sweep past on their way out. Let's call this one-time wave of fragments the “debris wave”.

Thereafter, only the chunks that stay in orbit around the Sun can continue to strike the planets. The field density of those fragments is highest by far in the first 100,000 years following the explosion, which is the average time it takes Jupiter to eliminate almost everything in planet‑crossing or unstable orbits. Let's call the first 100,000 years after the debris wave passes the “high impact period”.

By 100,000 years after the explosion, the field density of impactors would tend to be close to what we see today. This will last until the Earth, for example, cleans out fragments in earth‑crossing orbits – about 30 million years. Let's call this the “normal impact period”.

So in summary, following the explosion we have:

• a blast wave, requiring days to weeks to pass
• a debris wave, requiring weeks to months to pass
• a high impact period, lasting about 100,000 years (limit set by Jupiter)
• a normal impact period, lasting 30-60 million years for asteroids in earth crossing orbits

Previously, I confused the blast wave and debris wave periods, leading for example to an erroneous prediction that cratering on Venus would show a strong hemispheric asymmetry. But the spread of arrival times for the debris wave might be expected to be almost an order of magnitude longer than for the blast wave, canceling much of the asymmetry for cratering that is predicted to occur with the carbonaceous soot deposits in the blast wave. The high and normal impact periods will further smooth out any asymmetry; and the effect of multiple explosions should effectively eliminate any small remaining asymmetry. If an asymmetry did still exist, this scenario predicts that it might show up only in the largest impacts, which would have come mainly from the single largest explosion event. Curiously, the one reported cratering asymmetry on Venus appears only for “volcanic craters” over 20 km in diameter [Science 261, 591-595 (1993)].

Although many issues remain, more than can be addressed here, in the light of this hypothesis revision, it clearly answers many questions and challenges, and seems to reconcile with the principal lines of evidence that did not fit the original hypothesis. It also points up the need for experts in many fields to examine the various lines of evidence without preconceptions or bias, and to contribute their knowledge and experience to this issue. No one person can be expected to have enough expertise in so many fields to get all the important details right, such as the tentative associations proposed here. For example, we read in Nature 363, 704-706 (1993) of the existence of many newly discovered small Earth-crossing asteroids with little orbital eccentricity and different colors than main belt asteroids. Yet another population of objects may be implied – perhaps a mini-asteroid belt in which the Earth is emerged. We still have much to learn about our own solar system.

I conclude by asking if I have yielded to the temptation that so many working scientists succumb to, and patched a hypothesis to keep it consistent with observations and therefore viable? Or did I originally overlook an easily anticipated feature of the hypothesis that should have been there all along? In answer I suggest that the meph is a more natural form of the hypothesis even if there were not a shred of observational evidence to support it. However, any such answer is surely subject to experimenter bias; so it is up to others to judge, and future observations to decide.