I considered it, but the problem is that the terrestrial planets are small, and they live deep in the Sun's gravity well. They aren't really big enough to toss each other out completely. So after an instability they either want to collide with each other or the Sun. The problem with an excitation that leads to a solar collision, is that you leave the remaining system in a state of instability. A collision between two proto-Venii solves this by decoupling Earth from the dynamics of the precursors. I can easily end up with solar systems that dynamically resemble our own, and I'm not sure that a solar collision would do that. But I'm still looking into it. I'm also fascinated by the apparent young, yet uniform, age of the surface of Venus.

First, the Faint Young Sun Paradox is not my idea. The first paper to point out the problem was written by Carl Sagan and George Mullen back in 1972, but the idea that stars steadily rise in luminosity as they age is older than that, and it is a very well established outcome of our understanding of how stars work. Basically as a main sequence star converts H->He in its core, the density of the core increases over time, which causes the core's pressure to increase in order to keep the star in hydrostatic equilibrium, and its luminosity rises as a consequence of the higher nuclear reaction rates in response.

An approximate functional form of the luminosity as a function of time is given by Gough (1981, Solar Physics) L=1/(1+0.4*(1-t/4.5e9))*3.9e33 erg/s (where t is given in units of years). This form, while an approximation, has been well matched by computer simulations of solar evolution, and is well matched by observations of stars on the main sequence.

Based on measurements of the partitioning of the elements Hafnium and Tungsten in rocks from the Moon mantle we can date when the giant impact that formed the Moon occurred. You can read more here. Basically Earth completed its formation in less than 100 million years after the start of the solar system, about 4.57 billion years ago. The earliest minerals on the Earth are zircon crystals that form in continental crust, which are dated to at least about 4.3 billion years ago, and seem to indicate formation on a planet with liquid water oceans on the surface. The earliest whole rocks on the surface are about 3.8 billion years ago. Plate tectonics has a tendency to recycle and destroy rocks, but nevertheless geologists have been able to identify several locations where very ancient bits of the Earth are still preserved. The Faint Young Sun Paradox is about the time from about 4.3 billion years ago to the Archean/Proterozoic boundary at about 2.5 billion years ago, or about 2 billion years after the formation of the Earth.

The mechanism by which Earth gets its orbit raised is by a corresponding decrease in the orbital distance of a hypothetical Venus-precursor body. Yes, I did those calculations. That was the point of the talk this article is about.

The person you describe sounds an awful lot like Jacques Laskar. Yes, I am aware of his work, and in fact it was that result that you describe that inspired me to try my hypothesis out.

I won't

Your mistake is a common one that I find in places like Slashdot and other online communities. The bulk of scientific discussion and debate takes place in the peer reviewed literature. It's what it's designed for, and for all its flaws, it works quite well. People often mistake articles written in the popular press, such as TFA, as accurately reflecting the science. Unfortunately by their nature these kinds of articles offer only a distorted shadow of the science they are reporting on. Now that's not necessarily a criticism of science reporting, as my own attempt at distilling my work here in these comments is also a pale distorted shadow of the work. It is quite difficult to distill complex topics down into short, digestible chunks. Even a single peer-reviewed paper can't possibly contain everything about a piece of scientific work. You have to follow all the references, and all the references in those references, and so on to really *get* all the pieces of a scientific story. But we try our best.

No, the asteroid belt was never a planet. We receive pieces of the asteroid belt in the form of meteorites, and most of those meteorites basically reflect a well-mixed sample of the same kinds of things that the Sun is made out of (minus all the gas), which hints that they are samples of the solid component of the solar nebula that the Sun formed out of. If they were a part of a planet, the planet would have differentiated into a core and a mantle, so would not reflect the "primitive" composition of the solar nebula. The asteroid belt is thought of a s a place where planet formation was halted very early on (likely due to the gravitational influence of the formation of nearby Jupiter), and was cleared out of most mass before it had a chance to combine to make planets. The combined mass of the asteroid belt is quite small, only about 5% the mass of the Moon.

No, those weren't considered because their effect is so tiny. In other work I do, I consider radiation forces as they pertain to the orbits of small bodies, such as asteroids. The most important of these is the Yarkovsky effect (non-isotropic radiation of thermal photons), which can change the orbits of small asteroids, but the effect diminishes as the size of the body increase. For objects greater than about 10-20 km in diameter in the inner solar system, the Yarkovsky effect isn't important over the age of the solar system.

Hi, good questions. The time period relevant to this is the Archean. The interior of the Earth was warmer back in the Archean than it is now, and there may have been more volcanic activity, but it's difficult to know what style of tectonics was operating at the surface. Very few rocks survive from that time period. Now one proposed solution to the Faint Young Sun problem was just that there was a lot more CO2 in the atmosphere. The subject of a few talks at this workshop a couple weeks ago was constraining the abundance of atmospheric CO2 from looking at the chemistry of the few rocks we have from that epoch. There were some presentation suggesting that the atmosphere contained no more than about 20x the present abundance of CO2, but you may need more like 100-1000x in order to completely solve the problem. So people have suggested things like more CH4, NH3, and also that perhaps the Earth was somewhat darker due to different styles of cloud-making and fewer continental land masses (oceans are quite dark), meaning that the surface did not reflect back as much radiation as it does now. All of these ideas are being actively debated.

Now as to the question of meteor bombardment: that was the topic of the last 1/3 of my talk at the workshop, but was not mentioned in TFA. I am on a paper coming out in a couple of weeks that is showing that the so-called Late Heavy Bombardment persisted on the Earth all throughout the Archean, rather than ending abruptly at the end of the Hadean, as was thought from looking at lunar samples. The bombardment rate, while much higher than present-day, was not so high as to likely have had any major direct effect on the climate over geologically interesting timescales (say an impact creating a 1000 km wide basin occurring every 200-500 million year during the Archean). However, there may have been indirect effects of impact bombardment that have yet to be explored, and we find that it is an interesting coincidence that bombardment rate pretty much drops off completely by the early Proterozoic, just as Earth began to show signs of having some oxygen in the atmosphere, and the first real evidence for any kind of major glaciation events (the Huronian snowball). Could somewhat elevated impact bombardment rate be a controlling factor in the warm and anoxic Archean? I don't know the answer to that, but were studying it.

The problem is not Earth's stability, it's Mercury's. Mercury is close to a so-called secular resonance, and it's eccentricity varies more chaotically than Earth or Venus. So yes, Earth would remain bounded indefinitely as long as Mercury never attains a high enough eccentricity that it begins crossing into Venus's orbit. Once close encounters take place with Mercury, the whole inner solar system can rapidly destabilize.

That was the subject of my 2007 paper. The problem is that the present-day mass loss rate of e Sun due to solar wind and coronal mass ejections is tiny. The Sun loses more mass do to the conversion of mass to energy in the core, and it's not enough to appreciably change the mass of the Sun over the age of the solar system. Young Sun-like stars appear to have stronger stellar winds, correlated with their higher rotation rate. But the Sun would have had to sustain orders of magnitude higher mass loss than present-day winds for its first 2 1/2 billion years on the main sequence, and this does not appear to match measured mass loss rates of nearby Sun-like stars of those ages.

This work bears only a superficial resemblance to the ideas of Velikovsky (and I'm being generous here).

During the Archean, the time period relavant to this study, tidal heating was not terribly important. The larger internal heat from radioactive decay was higher, yet still dwarfed by the energy input from the Sun in setting the surface temperature of the Earth.

Ah, so here's the deal. I'm the person that this article is talking about (David Minton, professor at Purdue University). I've been reading Slashdot for a fair number of years now, though it took me a long time to sign up and comment for the first time (I've always been a lurker at heart). Because I have a soft spot for all you basement dwellers (I kid!), I'm going to give you a bit of behind the scenes regarding this article, which kind of took me by surprise, actually. This is a bit long, so TL;DR: Science sometimes happens during panicked last minute coding sessions in hotel rooms prior to delivering invited talks that were procrastinated about.

So about five years ago my graduate school advisor and I wrote what was my very first peer-reviewed paper, which was on the subject of the Faint Young Sun Paradox. The paradox goes something like this: The early Sun was fainter than it is today, so all things being equal the Earth should have spend the first half of its life frozen over. Geologists tell us it wasn't, so something wasn't equal. What was it? We investigated the idea that the Sun may have been slightly more massive (something like 2-7% more massive), and that it had to lose most of that excess mass over a few billion years, which is at odds with measurements of mass loss of Sun-like stars. So we published it, and I went on to do other things in grad school, mostly involving trying to figure out the early impact bombardment history of the solar system, which we think may have been influenced by an early period of migration of the gas giant planets.

Fast forward to a few months ago, and a fellow at the Space Telescope Science Institute (the place they run the Hubble from) contacted me to ask if I'd like to give a talk about my old mass-losing Sun paper at a workshop that was planned to bring together astrophysicists, geologists, climate scientists, and planetary dynamicists to talk about the Faint Young Sun problem. They wanted me to also talk about planet migration and how that might fit in to the problem. Sure, why not? Revisiting the problem would be fun! The thing is, I've just started a new faculty job, and part of my job is helping get a new planetary science group built up at Purdue, so I've been extremely busy. And, well, I procrastinated. Big time. There was always some pressing thing to do that took time away from getting ready for the workshop. So the next thing I know, it's a few days before the meeting and I still haven't really thought about the faint Sun in about five years. So I dust off my old files, start futzing around with a talk, and the next thing I know I'm on a plane to Baltimore.

Late the night before the workshop is about to start, I'm racking my brain trying to come up with something new to say. You see, I've been thinking about early solar system history, and planet formation. Migration is a big deal in those early days. It's easy to get planets to move around in young solar systems. But the Faint Young Sun problem is a problem for the Earth's mid-life, not it's adolescence. Then I remembered a paper I really liked that came out a couple of years ago by Jaques Laskar and Mickaël Gastineau. They showed that our own solar system could potentially destabilize after a few billion years of seeming-stability due to Mercury's proximity to a chaotic region. It's described briefly here: http://en.wikipedia.org/wiki/Stability_of_the_Solar_System#Laskar_.26_Gastineau

What if something like that had happened *already?* So I futzed around with an N-body gravitational dynamics code remotely from my hotel room, in my pajamas, playing around with plausible initial solar systems where Earth stared just a tad closer to the Sun, but close enough to solve the problem of being frozen over, and Venus started out as two separate planets and then went unstable after many billions of years, scattering Earth to its present location in the process. And, when I checked the output of the code the next morning, I found a set of conditions that worked beautifully! So I had an intriguing new result to present at this workshop that tied in to some of my more recent work. I presented it with lots of caveats that, while intriguing, was probably not likely, and I hadn't even really thought about what other constraints there were.

There was one thing that I didn't really appreciate, and that was that this workshop was being webcast. Two things happened that are now out there on the Web forever. First, the audience pointed out a typo in a plot from my old grad-school paper that I had missed. It's a critical plot, and the typo made the mass-losing Sun idea seem worse than it really is. Oops. The second thing is this article, which is ultimately about my wacky idea (but it's so crazy it might just work!), but that I haven't spent the time poking around at to see if I can kill it. But I'm okay with that. Science can sometimes resembles play, and between writing grant proposals, committee meetings, peer review, debugging code, and all the other stuff that is important but not really directly about the science it's easy to forget that. A lot of it is really about just trying out new stuff to see what sticks, sometimes by standing on giants' shoulders and seeing farther, but sometimes by procrastinating too long and scrambling around at the last minute because you feel like you need to do *something* new.

This "jumping Earth" idea is probably wrong. I'm actually working now to find out how I might kill the idea dead (I'm looking at the effect a big collision like that would have had on the cratering record of the Moon, which should be a very tight constraint for an event so late in solar system history). I'll probably write a paper for peer review on the work once I've explored all the ways I can think of that might show that the idea is wrong. If I kill it dead, then at least we can cross this off the list of possible solutions to the Faint Young Sun problem (and there are already too many). If I can't, then maybe someone smarter than me will. Or maybe Venus really did wait a few billion years to finish forming and its precursors scattered Earth out in the process. The Universe has shown itself to be stranger than was ever imagined, and it's through the process of science that we've pinned down much of the details of its particular flavors of strangeness. Science can be fun and creative, and when I'm getting a chance to do it, I realize that I have one of the best jobs in the world.

http://en.wikipedia.org/wiki/KH-11_Kennan

Actually, more than a dozen "Hubble space telescopes" were built and launched into orbit. The biggest differences are that they point at the Earth instead of away from it, and they are called KH-11 instead of HST. Oh, and their imagery data is mostly classified.

Considering that out of the 150,000 stars, there are 1200 planetary systems that are both oriented such that the planets pass directly in front of their stars as seen from our solar system, and did so over a period of about 4 months, that's a *very* good ratio. The whole point of Kepler is to gather statistics on planetary systems. There's no need to wait 20 years. Trend lines are being plotted now.

Well, to an astrophysicist "roughly 12 times" is equivalent to 13 times, but your point is taken. I've sat in with the Catalina guys (on a nearly full moon night, so they didn't discover anything while I was there), and they don't wait to submit data. They send candidate objects to a followup telescope to confirm the discovery, then publish any object with the Minor Planet Center as soon as they are confirmed. They need to act quickly, because orbit refinements often rely on followup observations (often by amateur astronomers), and many objects, especially Near Earth Asteroids, could be lost if they are not followed up quickly. The big pulses in the discovery rates at early times are because objects were only discovered in sensitive surveys that were not run very frequently (and before the mid 1990s usually relied on photographic plates). After about 1997 once LINEAR got going (and later Catalina and a couple others) asteroid surveys have more or less been continuous, with lulls arising due to full moon nights and the weather patterns of southern Arizona and New Mexcio.

You'll also notice that the discovery rate seems to "pulsate" with a period of about 12 times per year (this is most obvious in the 2000s when the discovery rate was mostly uniform throughout the year). I'll leave it as an exercise to the reader to explain why that is (hint, skies need to be very dark to observe faint asteroids).