Browsed by
Tag: science!

Another Interesting Result

Another Interesting Result

Comments 2 comments
Artist’s conception of the TRAPPIST-1 planetary system (NASA/JPL-CalTech)

Sometimes a bit of research comes across my desk that leads to big changes in one of my creative projects. Today we have a case in point.

At the beginning of December, Ken Burnside was kind enough to bring this article from phys.org to my attention: The solar system follows the galactic standard – but it is a rare breed.

The article was from the Niels Bohr Institute, summarizing some research done there by Nanna Bach-Møller and Uffe G. Jørgensen. The upshot is that, based on our extensive sample of detected exoplanets, we can conclude that there’s a fairly strong correlation between the number of planets in a planetary system and the average eccentricity of the orbits of those planets. “Just a few planets” seems to correlate to highly elliptical orbits, while “more planets” means closer-to-circular orbits.

It makes sense. We know a lot more about the process of planetary formation than we did even twenty years ago. That process appears to be pretty chaotic. Planets sometimes interact a lot while they’re forming, with unpredictable results. Sometimes that interaction leads to some of the young planets getting “pumped” into highly eccentric orbits, but that also leads to more of them being “ejected” from the planetary system entirely. So it makes sense that planetary systems that end up with fewer planets might also see those planets line up into more eccentric orbits.

The article claims that our own planetary system is unusual in that we ended up with more planets than the average. As a corollary, it shouldn’t be a surprise that the planets we still see have settled into a well-behaved stack of nearly circular orbits.

Okay. The article was interesting enough. The problem was that the actual research paper behind it was sitting behind a paywall. I put off reading that until after I had finished the rough draft of the Architect of Worlds design sequence. Maybe then I would track down a copy, and it might suggest a way to improve the step in which I assign orbital eccentricities. Not a big deal.

Well, I finished the rough draft just before Christmas, and yesterday I found a way to get a copy of the paper for a reasonable fee. I sat down to read it, and . . .

Whoa. This is a lot bigger than I thought.

Here’s a link to the abstract for the paper in question: Orbital eccentricity–multiplicity correlation for planetary systems and comparison to the Solar system. From there you can get to the whole paper, assuming that you have an Oxford Academic account or can work through DeepDyve.

Bach-Møller and Jørgensen have done something a little more remarkable than I expected. They haven’t just derived a strong correlation between planetary multiplicity and eccentricity of orbits. They’ve demonstrated that we can derive a clear power law for how many total planets a given system has, including the ones we can’t detect yet, just based on the observed eccentricity of the ones we can detect.

Applying this result to my models in Architect of Worlds, I find that I can do a lot more than just superficial improvement to one step of the system design process. I can actually rework several of the steps in the sequence, making them simpler and easier to use, and also making them line up a lot better with the current state of exoplanetary science.

The executive summary is that instead of laying down planets until you run into any of several limiting conditions, you randomly generate the total number of planets first, and then place that many. Much simpler, and it fixes the problem that the current version seems to generate too many planets.

This isn’t a small improvement. We’re talking about eliminating several of the most cumbersome computations and procedures, while also forcing the outcome to match observed results much more closely. I can’t really let that sit in the idle stack, especially since I have a couple of other projects that are dependent on having a complete draft here.

One complication is that one of those other projects was something I was planning to put together for my patrons, as a charged release, before the end of December. Although I think I see how to make all the necessary changes to the Architect of Worlds draft, that’s going to take a day or two of work, and I have a pile of other things to get finished over the next few weeks as well.

So here’s a revision to my creative plan for the next couple of weeks:

  • The top priority right now is to revise the partial Architect of Worlds draft to fit these new results. This should be complete no later than 28 December. At that point, I will release a revised version of all of the completed sections of the draft, for my patrons only (with one or two exceptions for non-patrons who have been helping out with extensive comments on the draft). That will constitute my charged release on Patreon for December 2020.
  • The PDFs that are already on the Architect of Worlds page will remain there. Those won’t constitute the most up-to-date version, but they are certainly “playable” for anyone who wants to experiment with them. I won’t be updating those PDFs for at least three months. During that time, I’ll continue to polish and tweak the system with input from my patrons, and possibly work on some additional material. I’ll reassess the situation in early April. By then I may be within striking distance of starting to prepare a publication-ready draft of the entire book. If not, then I’ll create and post new PDFs at that point.
  • By the end of December, I need to write a new book review, and also finish and release another piece of short fiction. Those will be posted here and released to my patrons for free. I also need to get started on a new piece of short fiction (more about that later).
  • Once all of the above is finished – probably over the New Year’s holiday – I’ll take stock. That’s a traditional time for such things anyway.

Dependencies in a Model

Dependencies in a Model

Comments 5 comments

Still working on the next sections of Architect of Worlds.

What’s interesting is that we’re coming to a number of items that fall into a web of dependencies. Some items affect the most likely outcome of others, and it’s not a nicely linear process. Just to give you a sample, here’s some storyboarding I’ve been doing using the Miro application online:

Where this mostly comes into play is with the order that the steps need to come in the design sequence. Fortunately, I have yet to come across any dependency loops. As long as the sequence moves more or less left to right, everything should work properly . . .

Alone in a Crowded Milky Way

Alone in a Crowded Milky Way

Comments 0 Comment
Enrico Fermi

I don’t usually post just to link to articles, but this one was particularly intriguing: Alone in a Crowded Milky Way.

Executive summary: the author and his colleagues did some modeling of the expansion of interstellar civilizations through a segment of the galaxy. They made some fairly conservative assumptions – STL travel only, how many planets would be worth colonizing, how long a given planetary civilization would be likely to survive. The result was that space-faring cultures tended to form “archipelagos” in interstellar space, leaving vast regions unvisited for many millions of years at a time. The limiting factor was imposed by stellar cartography – the arrangement of interesting planets among a population of stars that move over time.

This strikes me as a similar approach to Geoffrey Landis’ percolation theory, and yields similar results. In either case, we get a galaxy that could be full of interstellar-capable civilizations, and yet one in which many habitable worlds (like ours) might appear unvisited at any given time. It’s a sophisticated solution to the Fermi Paradox, and one which avoids nightmare scenarios.

One insight the authors of this article mention, which I hadn’t considered, is that even if Earth had been visited – or even colonized – at some point in the distant past, we might have no way of knowing that. My own models for putative galactic civilization might stand to be relaxed a little. That might go on the list for some deeper research, if and when I get back to working on space-opera projects.

An Interesting Result

An Interesting Result

Comments 2 comments

Just a short note, to call your attention to an interesting result in recent astrophysics that’s quite relevant to the Architect of Worlds project.

It’s well known that we’ve discovered thousands of exoplanets in the last couple of decades. Now the state of the art is approaching the point where we can get clues about the environment on those planets. For example, one recent result (here’s an article in Scientific American) is the first indication we have of the kind of atmosphere that exists around an Earth-sized rocky exoplanet. In this case, the planet is in its primary star’s habitable zone, and it’s more than large enough to retain a significant atmosphere against thermal or Jeans escape. Yet there doesn’t seem to be much if any atmosphere there.

I’ve long since worked out a model for Jeans escape; that much was in the world-design system I wrote for GURPS Space back in the day. (If you’ve used that system, you may recall a “minimum molecular weight retained” or MMWR calculation. That’s specifically relevant to Jeans escape.)

The problem is that thermal loss isn’t the only way a planet’s atmosphere can get stripped off. If the primary star is prone to flares and has an energetic stellar wind, that will do the trick too. This is specifically relevant to red dwarfs, like the star LHS 3844 which has the planet mentioned above. Red dwarf stars punch well over their weight in the stellar-wind and flare department, especially early in their lives. Thus, any rocky planet close enough to be in the liquid-water zone will probably get a serious sandblasting early on.

This is kind of a new area in astrophysics, and there are a lot of competing models out there. Some scientists are predicting that Earth-sized planets should be able to retain their atmospheres in at least some cases, others are much less sanguine. The LHS 3844 result certainly seems to support the pessimistic case.

For Architect of Worlds, I’ve been thinking in terms of assigning each planet a “volatiles budget” from its formation and early years, modified by things like the planet’s MMWR, whether it formed inside the snow line, whether there’s a dominant gas giant to fling comets in-system, and so on. A big random factor as well, since it looks as if this feature is strongly subject to chance. Then we would reduce that volatiles budget to reflect non-thermal processes of atmospheric escape, photodissasociation of water molecules, and so on. (Hmm. Maybe have a separate budget just for water, since that goes through some significantly different processes and might not be correlated with atmospheric volatiles.)

The devil’s in the details, of course, and for all my tinkering I have yet to come up with a model that satisfies me (or even fits all the cases we know about). This might actually be the biggest obstacle to getting the third section of the design sequence hammered out.

Architect of Worlds: Reality Ensues

Architect of Worlds: Reality Ensues

Comments 3 comments

One element of work on the Architect of Worlds project is that real-world science in this area has been advancing by leaps and bounds for years. Not once but several times, I’ve settled on a method for modeling some feature of planetary formation, only to see some new result published that seems to demand a different model. On one occasion, while I was testing the current model by generating planetary systems for nearby stars, I literally saw the first publication of new exoplanets in a star system I had just done randomly a few days before. This can be frustrating, although I have to admit it’s also rather exciting!

Now it seems to have happened again. The current model I use for planetary system formation implements a combination of the Nice model and the Grand Tack hypothesis, which together describe how Jupiter may have formed and migrated through the early solar system before settling down in its current location. The assumption is that a planetary system’s primary gas giant will normally form near the “snow line” and then migrate inward (and possibly outward) to its final position. This is all implemented in Step 10 of the current sequence.

Now there seems to be evidence that these models aren’t telling the whole story. In a new paper, “The consequences of planetary migration on the minor bodies of the early Solar System,” computer simulation seems to suggest that the core of Jupiter must have formed much further out than the Nice model suggests. It may have formed as far out as four or five times the snow-line radius, and then migrated inward very quickly.

I need to read the Pirani paper in more detail, and see if there’s been any discussion as to how to reconcile those results with the Nice model. To be honest, the Nice model isn’t entirely consensus in the community, so a new model that fills in some of its problematic details might work.

Still, if the Pirani paper seems supportable, it may be necessary to do a significant rewrite of the second chunk of the design sequence. It’s possible that the result may actually be simpler than what I’ve got now. I’ve thought of a way to cut out some of the current sub-steps and computation, making the process a little smoother, that might work. We’ll see what develops.