[Space] Various Star System Generation Thoughts and Questions

[Nemoricus]

So, I'm presently working on an automated star system generator based on the rules in GURPS Space, primarily for the purpose of getting numbers on the expected results of that generator. In the process, I've run across a number of issues and would like some thoughts on my solutions.

Please note that I am attempting to remain as faithful to the rules as written in the book as possible while avoiding absurdities and fixing ambiguities.

1. White dwarves have no hard information given for their temperature, luminosity, and radius. As the book suggests that they have a luminosity of no more than 0.001, I have arbitrarily assigned them that luminosity and a surface temperature of 20000 K based on the information I've found in various online sources.

Given that the book states that white dwarves are Earth sized, I may change this so that this radius and a luminosity of 0.001 are used to find the temperature of the star.

I have no interest at this point in devising a more realistic system that takes into account the white dwarf's age and progenitor star mass, as the rules as written produce white dwarves that have no correlation with the progenitor star anyway. I will revisit this when I have finished the rest of the generator.

2. The third star of a trinary system must orbit at a distance greater than that of the second star, but no hard rules are given for how much greater. I have decided that the minimum separation of the third star must be at least three times the maximum separation of the third star. If it is not, the minimum separation is assigned a value equal to 3.1 times the maximum separation of the second star and its orbital radius and maximum separation are adjusted accordingly. A factor of 3.1 is used instead of 3 to give more margin between the stars and to avoid potential problems with floating point comparisons later on.

The rationale for this is that the limits for stable planetary orbits apply to stable orbits for stars, too. This rule is effectively that stars may be no closer than the edges of their respective forbidden zones.

3. The maximum separation of a subcompanion must be no more than one-third its closest approach to any other star. If it exceeds that, its maximum separation is multiplied by 1/3.1 and its other orbital parameters adjusted accordingly, for the same reasons as in 2.

4. Planetary orbital zones are placed wherever the forbidden zones of stars permit. The primary of a trinary system may have as many as three orbital zones associated with it, for example. The first is inside the orbit of the the first companion, the second is between the first and second companion stars, and the last is outside the orbit of the third companion star. A five star system may have as many as seven orbital zones associated with it. One for the primary, and three each for the binary pairs.

Now, I have some areas that I have not yet been able to resolve or have not yet had the time to find a solution.

1. The rules for stellar orbits combined with those for stellar evolution occasionally produce a star that is inside another. In all cases I've examined so far, this occurs when one star grows into the red giant stage, and especially one of the cooler ones with correspondingly larger radius. It's a rather rare occurrence, though, and only happens 1-5 times per 100,000 star systems, so I haven't worried about fixing it.

From the tables on page 105, the closest one star may approach its primary is 0.04 AU, and so I think that it would be appropriate to boost the minimum separation of a star to this value, plus the radius of the star and its primary. The reason for this rather than simply destroying the star is that the GURPS Space rules seem to favor taking action to preserve a body once generated rather than let it be destroyed.

2. The luminosity based inner limit can be inside the star for cool giant stars, specifically for giant stars cooler than about 3950 K. This can result in a planetary orbit being placed inside the star, especially if it is an epistellar gas giant. Contrary to my general inclination to preserve objects when possible, in this case I think it most convenient to declare that the planet is destroyed.

3. The rules presently do not have much guidance on what to do if a star has split orbital zones. In fact, there appears to be an assumption that the planetary system being generated is always between the either the outer limit of a star/the inner edge of a forbidden zone and the inner limit of a star.

However, this does work for the inner orbital zone for a star, and so the rules as written may be used for it. The question then becomes what to do for outer orbital zones. In particular, what should be done if the snow line exists between the limits of the zone or if it exists inside of the zone's inner limit?

If the snow line exists outside the zone's outer limit, then naturally planetary generation proceeds as with the No Gas Giant arrangement.

[Agemegos]

Originally Posted by Nemoricus

2. The third star of a trinary system must orbit at a distance greater than that of the second star, but no hard rules are given for how much greater.

It is the +6 modifier on the roll for orbital separation for second companions that is supposed to handle this.

The maximum separation of a subcompanion must be no more than one-third its closest approach to any other star. If it exceeds that, its maximum separation is multiplied by 1/3.1 and its other orbital parameters adjusted accordingly, for the same reasons as in 2.

It is the -6 modifier to the roll for orbital separation of sub-companions that is supposed to handle this.

From the tables on page 105, the closest one star may approach its primary is 0.04 AU, and so I think that it would be appropriate to boost the minimum separation of a star to this value

It would better reflect the physical processes involved to remove the companion and add its mass* to the primary. There is no effect in Nature that moves companion stars outwards as their primaries swell.

(* Caution! An astronomer might come along and tell you that something more violent would in fact occur.)

2. The luminosity based inner limit can be inside the star for cool giant stars, specifically for giant stars cooler than about 3950 K. This can result in a planetary orbit being placed inside the star, especially if it is an epistellar gas giant. Contrary to my general inclination to preserve objects when possible, in this case I think it most convenient to declare that the planet is destroyed.

Indeed. Even if the outer parts of the red giant aren't actually hot enough to evaporate silicates such as enstatite drag from the stellar material will cause the planet to lose orbital speed and it will spiral in to the centre of the red giant.

In particular, what should be done if the snow line exists between the limits of the zone or if it exists inside of the zone's inner limit?

Start placing orbits just inside the outer limit and work your way inwards? That's how I read the fifthe paragraph of Step 22 (p.108).

[Nemoricus]

Originally Posted by Brett

It is the +6 modifier on the roll for orbital separation for second companions that is supposed to handle this.

It is the -6 modifier to the roll for orbital separation of sub-companions that is supposed to handle this.

Supposed to, yes, but it will fail often enough that I have to ensure that it won't happen. For two stars at distant separation, it is entirely possible that the second star will roll an orbital radius smaller than that of the first companion, for example.

Before I instituted those rules, 3 star systems in every hundred had orbital conflict problems

It would better reflect the physical processes involved to remove the companion and add its mass* to the primary. There is no effect in Nature that moves companion stars outwards as their primaries swell.

Indeed not, but physical laws are not the motivation behind my choice. Instead, it is an attempt to remain faithful to the tendency in the rules to favor preserving an object over having catastrophe befall it. See the rules for star orbits, planets that end up placed in forbidden zones, and moon orbits for examples.

Start placing orbits just inside the outer limit and work your way inwards? That's how I read the fifthe paragraph of Step 22 (p.108).

That is what I intend to use for placing the orbits within a zone if the snow line exists beyond its outer limit or if it has the No Gas Giant arrangement. The question is more one of what to do for the gas giant arrangement for each orbital zone.

My present inclination is to randomly determine the gas giant arrangement for each zone and treat results of epistellar gas giants as eccentric for outer zones. If the snow line exists within the orbital zone, it is used to place the gas giant as normal, except that it is not be inside of the inner edge of a zone. If the snow line exists inside of the zone's inner limit, the gas giant is placed at the inner limit. The rest of the orbits are then generated as normal.

[Nakawaros]

That is what I intend to use for placing the orbits within a zone if the snow line exists beyond its outer limit or if it has the No Gas Giant arrangement. The question is more one of what to do for the gas giant arrangement for each orbital zone.

My present inclination is to randomly determine the gas giant arrangement for each zone and treat results of epistellar gas giants as eccentric for outer zones. If the snow line exists within the orbital zone, it is used to place the gas giant as normal, except that it is not be inside of the inner edge of a zone. If the snow line exists inside of the zone's inner limit, the gas giant is placed at the inner limit. The rest of the orbits are then generated as normal.

I did it a bit more simply: I test to see if any of the "range" for the arrangements are available, if not, it generates as per the flags and simply places no first gas giant.

[Anthony]

Originally Posted by Nemoricus

1. The rules for stellar orbits combined with those for stellar evolution occasionally produce a star that is inside another. In all cases I've examined so far, this occurs when one star grows into the red giant stage, and especially one of the cooler ones with correspondingly larger radius.

I believe this situation is realistic, but results in nonstandard stellar evolution when it occurs (it looks it's probably a common envelope. Actually, this looks like it's a fairly direct study of the situation.