Just The Caffeine Talking

Notes on Worldbuilding, Part 8: Planetary Characteristics

This is where we figure out the details that actually determine what it's like on the planet. Some of these can be figured pretty precisely, others are up to the creator. (For gamers I've included random generators.)

Size: We know the mass (in Earths), and we've determined the density relative to Earth. Mass divided by density gives us volume (in Earths), and the cube root of volume gives us the planet's radius in Earths. Multiply that by 6,400 to get the planet's radius in kilometers. (Double the radius to get the diameter.)

Some comparisons: Mars's radius is 3400 kilometers, and Venus's is 6000 km. Note that even a "mega-Earth" with a mass of 10 times Earth's own mass would be only about twice Earth's diameter.   

Gravity: This is pretty straightforward. Relative mass (in Earths) divided by the square of radius (in Earth equivalents) gives you the surface gravity in Earth "gees" (often abbreviated g). Multiply that by 9.8 meters per second squared for the actual force, but keeping it in multiples of Earth gravity is probably more useful.

Mars's surface gravity is 0.38 g, Venus's is 0.9, and that hypothetical mega-Earth would have a crushing 2.8 gsurface gravity — about the same as at the cloud tops of Jupiter.

Escape Velocity: This is how fast you have to be going to get off the planet. It's useful to figure out how practical space travel is from that world, and for the atmospheric composition (see below). The simplest way to figure it is to divide the planet's mass (in Earths) by its radius (in Earth radii), take the square root of that, and multiply the result by Earth's escape velocity of 11 kilometers per second. 

Rotation: Planetary rotation is anyone's guess, really. Close to the Sun, tidal forces slow a planet's rotation. Mercury's rotation is locked into a 3:2 resonance with its orbital period (which fooled astronomers on Earth for years into thinking the planet kept one face permanently to the Sun). And there seems to be some vague connection between mass and spin rate: Jupiter's day is only 10 hours long, and Saturn's is about the same. But . . . Earth is ten times the size of Mars and yet they have almost identical day lengths. Meanwhile Venus takes 243 days to turn on its axis, and does it backwards, to boot.

So you can basically pick whatever day length you want. The absolute minimum would be somewhere around 3 or 4 hours for an Earthlike world, which would create a perceptible reduction of surface gravity at the equator due to centrifugal force effects. Slow the rotation in close orbits to some small multiple of the orbital period, and if the planet has a big moon you should aim for something in the 20-30 hour range. (In the case of an actual double planet, or the moon of a bigger planet, rotation and orbital period will be the same. We'll cover that later.)

If  you want to generate rotation randomly, I suggest simply rolling two 20-sided dice to get the day length in hours for an Earth-sized or larger rocky planet. If either die comes up 20, multiply by 10, and if both come up 10, re-roll and multiply the new result by 100.

Axial Tilt: This is another wild card. Most planets have their axis of rotation aligned with their axis of revolution around the Sun, within twenty or thirty degrees, at least. Except . . . Uranus's axis is deflected 90 degrees so it's lying on its side, and Venus seems to be entirely flipped over. Astronomers make vague noises about giant impacts during planet formation, but nobody really knows for sure why these weird exceptions exist.

So for "normal" planets pick an axial tilt in the range of about 20 to 30 degrees. To determine randomly roll a 20-sided die plus 10 degrees. If the die comes up 20, roll two 20-sided dice and multiply the sum by 10 for the tilt in degrees.

Atmosphere: The atmosphere composition depends on gravity, temperature and whether or not life exists on the planet. For now, assume that life is Earth-type life and can only exist in the Goldilocks zone.

I started to work out an enormous overcomplicated spreadsheet to provide data for different gases at different temperatures, but in the course of researching it I found this diagram showing what gases can exist at a given combination of temperature and escape velocity: https://en.wikipedia.org/wiki/File:Solar_system_escape_velocity_vs_surface_temperature.svg#file.

There are a few gases I would add to the list. Hydrogen cyanide (HCN) has approximately the same molar mass as nitrogen, and hydrogen sulfide (H2S) is close to oxygen.

Find the point on the chart where your planet's temperature and escape velocity meet. Any gases above that on the chart will be lost to space.

At low temperatures some gases condense and freeze out. Water freezes at 273 Kelvins, hydrogen cyanide at 260 Kelvins, ammonia at 195, carbon dioxide at 194, hydrogen sulfide at 191, xenon at 160, methane at 90, nitrogen at 63, oxygen at 54, and hydrogen at 14. Higher atmosphere pressures can make this happen at warmer temperatures, but you'll have to work that out on your own. I'm already at the limits of my chemistry knowledge.

The relative abundance of the gases in an atmosphere can vary quite a bit, based on the composition of the stellar nebula when the planet was forming. The simplest rule of thumb is to take the abundance of the elements in the universe as your guide: hydrogen is the most abundant, followed by helium, carbon, nitrogen, oxygen, neon, sulfur, and argon.

Oxygen won't occur as a free gas without life. It's rocket fuel and wants to burn the universe. On a lifeless world it will be locked up in water, carbon dioxide, and in the rocks of the crust.

Effects of Life: If a planet has plants (or plant-like organisms) doing photosynthesis, they may produce oxygen from carbon dioxide and water. As noted above, oxygen plays all too well with others: a planet with oxygen in its atmosphere won't have more than trace amounts of methane, ammonia, hydrogen sulfide, or hydrogen cyanide.

Atmospheric Pressure: There's an important concept in chemistry called "partial pressure." This is the percentage of a given gas in an atmosphere, multiplied by the atmospheric pressure. Earth's atmospheric pressure at sea level is 1 bar (by definition), or 1000 millibars. The atmosphere is 20 percent oxygen, so the partial pressure of oxygen at sea level is 200 millibars. Humans seem to need about 160 millibars of oxygen for normal metabolism. This means we could survive in a lower-pressure atmosphere if its oxygen content was higher, so that the partial pressure comes out the same — or conversely in a higher-pressure environment with a low oxygen concentration.

Pressure is a function of the atmosphere's actual mass, times the planet's surface gravity. The actual quantity of atmosphere around a planet can vary widely — Venus has a much more massive atmosphere than Earth does, even though it's hotter and has little magnetic protection, so its surface pressure is nearly 100 times what you and I are breathing. For rocky planets it does seem to be an all-or-nothing situation: small bodies have no pressure to speak of, Earth has 1 bar, and Venus has 100 bars.

To generate atmosphere pressure for rocky worlds randomly, roll a 6-sided die: 1 means vacuum, 2 means a trace atmosphere, 3 means a thin atmosphere, 4 means a medium atmosphere like Earth, 5 means dense, and 6 means massive. Pressure for a trace atmosphere is 1-10 millibars (use a 10-sided die to determine randomly) times the surface gravity (in Earth gravities). For a thin atmosphere it's 1-100 millibars (use percentile dice) times gravity. For a medium atmosphere, it's 100-1000 millibars times gravity (use a 10-sided die and multiply by 100). For a dense atmosphere it's 1000-10,000 millibars times gravity (roll a 10-sided die times 1000). And for a massive atmosphere it's 10 to 200 bars (roll a 20-sided die and multiply by 10 bars). 

If there's oxygen in the atmosphere, calculate the partial pressure to see if it can support humans and Earthly animal life. Presumably any native species are evolved to fit the local conditions. Note that since pressure drops off at high altitudes, a world with pressures too high at sea level might still have high plateaus with bearable conditions for humans. This involves a concept called "scale height" which is explained here: https://en.wikipedia.org/wiki/Scale_height. Pressure drops logarithmically as you increase altitude (that's the natural log, not log-10).

Next time: Moons!

Notes on Worldbuilding, Part 7: Temperature and Composition

From here on out I'm going to concentrate on solid planets which might be home to life we can interact with. These are bodies with a mass in the range of 0.1 Earth to 10 Earths, though I will include occasional nods to "Mega-Earth" type worlds even bigger than that. My plan is to work my way up the temperature scale to give some ideas about composition, density, and what atmospheric gases the world can retain.

Very Cold: These are planets with a temperature less than 100 Kelvins. They may have a little internal heat if they're big enough, but get almost no energy from the parent star. If they're small — a mass less than 0.1 Earths — they're likely low-density agglomerations of rock and ice. The very small ones may not even be differentiated inside (i.e. no distinct core, mantle, or crust). Density for those little ones will be low, from 0.2 times that of Earth (which is to say, about the density of ice) to perhaps 0.75 Earth (like Mars). Atmosphere will be very thin, a wisp of hydrogen and nitrogen with perhaps some helium.

Bigger worlds comparable to Earth (0.1 to 2 Earths) in that cold region can generate enough internal heat to have a distinct core and interior layers. I'm going to assume they are primarily rock and metal, so density will be close to that of Earth, 0.75 to 1. However, at the big end one might see "baby ice giants" which have swept up a lot of lighter elements and so have a low density of 0.3 or 0.4. Either way, the atmosphere will be mostly hydrogen with significant amounts of helium and nitrogen.

"Super-Earth" sized planets at those low temperatures are simply small ice giants like Uranus and Neptune. Density of 0.3 times that of Earth, with a massive atmosphere of hydrogen and helium, seas of liquid hydrogen, ice below that, then rock, then metal at the very center.

Cold: This is the temperature range from 100 to 200 Kelvins. Tiny planets less than 0.1 Earths are similar to their Very Cold counterparts — rock and ice, low density. The main difference is that methane and carbon dioxide become important atmosphere gases. Rocky worlds in the 0.1 to 2 Earth size can have hydrocarbon atmospheres like Saturn's moon Titan. Densities range from 0.6 to 1 for those worlds. The slightly warmer temperature makes "baby ice giant" planets less likely at this size.

Big "super-Earth" rocky planets in the 2 to 10 Earth size range may exist, with thick atmospheres and deep oceans of liquid methane. They might also have liquid water deep under the surface, melted by heat from the planet's core. Density would be 1 to 1.3 times that of Earth for these giant rocky planets. More likely, however, is that these massive worlds are ice giants with a density more like 0.3.

Chilly: Planets outside the Goldilocks Zone but with temperatures above 200 Kelvins. Ice is still a rock, but atmospheres of carbon dioxide, methane and ammonia are possible. Densities are about the same as above, but there's more options for atmospheric gases — and some of those are decent greenhouse gases so there's the possibility that a planet Earth-sized or larger in this region might be able to claw its way up to the melting point of water, at least some of the time. At the upper end of the size scale they would simply be "warm Neptune" planets with low density.

Goldilocks Zone: When water isn't a rock, planetary density changes abruptly. Having liquid water on a planet means two things. Yes, you might have life. But it also puts the planet at risk of losing its hydrogen. See, water molecules in the upper atmosphere can get broken into their constituent atoms by energetic particles from space. Some fraction of the free hydrogen can then escape the planet into space before it oxidizes again. Over geological time scales this can allow a planet to lose all its hydrogen. That means no water, and it also has knock-on effects on things like continental plate movement and vulcanism. It seems to have happened to Venus, and on Mars until that world lost so much atmosphere it got too cold for liquid water to exist on the surface. A powerful magnetic field like Earth's keeps the energetic particles at bay.

We don't have a very good understanding of how planetary magnetic fields work, but they seem to be generated in the dense metal core. Bigger planets are more likely to have a dense molten core, but it isn't universal. Venus seems like it ought to have a molten core but it doesn't generate much of a magnetic field at all. This may be related to Venus's weird slow rotation. As always, when science isn't sure, that means writers can indulge themselves. Choose or flip a coin to decide if your planet has a magnetic field as powerful as Earth's.

For small rocky bodies less than 0.1 Earths, the Goldilocks zone means no more water. All of it melted and got blown away. In the 0.1 to 1 Earth range, it all depends on how good a magnetic field the planet has. No planetary magnetic field means no water. Hydrogen might remain in heavy molecules like sulfuric acid. A working magnetic field means at least some surface water if the planet's big enough (more on that later). This does mean a higher density: Mercury's density is about the same as Earth's — 25 percent higher than Mars's density even though Mars has twice Mercury's mass. So simply assume all rocky inner-system bodies with a mass of 0.1 Earth or more have about Earth's density.

Bigger bodies, "super-Earth" and even "mega-Earth" sized can almost certainly hang on to their water even without magnetic protection, as their escape velocity is high enough to retain lighter atoms. For them the danger is runaway greenhouse effect due to all that atmosphere.

Hot: Within the inner edge of the Goldilocks zone, small worlds have all their water and volatiles cooked away, and big planets suffer a runaway greenhouse effect as water vapor is a major greenhouse gas. Expect a lot of airless Mercury-type worlds and a lot of Venusian and "super-Venus" planets, all with Earth-type density.

Above about 400 Kelvins, sulfur is a liquid. It's a reasonably common element, so some hot planets might have molten sulfur seas, especially if they have no hydrogen to bond with the sulfur and turn it into sulfuric acid. This raises some interesting possibilities for complex molecules of carbon, silicon and sulfur in a liquid sulfur medium.

A giant "super-Earth" type in the inner zone might be what's known as a Chthonian planet — the inner core of a gas giant which migrated inward but had all its lighter elements boiled away. Some exoplanets are thought to be this category, but we don't know for sure. A Chthonian world can be extremely dense, up to 1.5 times the Earth's density.

Molten: Above about 1000 Kelvins, rocks start to get gooey and planets basically don't have a solid surface any more. At those temperatures all the lighter elements can escape from any world smaller than a brown dwarf. Assume all bodies in that temperature range are made of dense heavy rock and metal with a density up to 1.5 times that of Earth.

Whew! It took a lot more space than I expected just to get the planetary density. Next time we'll really start putting it all together and figure out size, gravity, and maybe take a swing at atmosphere.

Notes on Worldbuilding, Part 6: Placing Planets

The two defining features for a planet are its mass and its temperature. Those data can pretty much predict its likely size, composition, and how suitable it is for life.

Temperature is a function of the world's distance from its primary star, and how bright that star is. By "bright" I mean the star's luminosity, which is a measure of how much energy it puts out. That is distinct from its visual magnitude, which describes how bright it looks as seen from Earth.

I'm not going to re-type data you can find elsewhere, so if you're using a real star, look up its luminosity. If you're making up a star system, look up the type of star you've got — F3, K9, whatever — and pick a value from the known range of luminosity for that kind of star.

To figure the temperature for each orbit around that star, you can do it the easy way or the hard way.

The hard way is to actually compute the black-body temperature based on the incoming energy flux. Read about it here: https://en.wikipedia.org/wiki/Planetary_equilibrium_temperature.

The easy way is to just compare the values for your fictional star system to the Solar System. Divide the star's luminosity (measured in multiples of the Sun's value) by the distance in AU squared. That gives you a value of how much energy the planet gets compared to Earth. Take the fourth root of that (hit the square root button twice) and multiply by 287, which is Earth's average temperature in degrees Kelvin. Then convert back into Celsius by subtracting 273 from the result. (You have to go through the Kelvin degrees step or it won't work properly.)

Note that this is the average temperature across the entire surface, and assumes that the planet reflects about as much energy as Earth does, retains as much heat as Earth does, and has a rotation rate that's roughly comparable. But it's good enough to make rough estimates of things like whether water can exist.

The Goldilocks Zone: Because water is not just important for life but a whole lot of other things, and because it's so common in the Universe, it's important to figure out the "climate zones" in a star system.

The inner zone is the space around the star where a planet's average temperature is too hot for liquid water to exist. Worlds orbiting in that zone won't have oceans, won't have ice, and unless they are very massive they won't have any atmospheric gases containing hydrogen (no hydrogen, no methane, no ammonia, etc.). Since gas giants and brown dwarfs are basically nothing but hydrogen, you're unlikely to have them in the inner zone — unless there's a "hot Jupiter" situation where a giant planet formed in the outer system and migrated inward over time.

I'm going to define the boundary of the inner zone as the distance at which a planet's temperature is 373 Kelvin — the boiling point of water at 1 atmosphere of pressure. A more dense atmosphere with higher pressure might keep water from boiling off — but such an atmosphere will also have a higher greenhouse effect and retain more heat, so I'm just going to let those factors cancel each other out.

Compute the boundary distance using the equation D (distance in AU) = the square root of (Star luminosity divided by 2.85). Orbits within that boundary can only have rocky worlds, except as noted above.   

The "Goldilocks Zone" (or in astro-speak the Circumstellar Habitable Zone) is the belt with temperatures between boiling and freezing. We've already figured the inner boundary, but the outer edge is a bit more fuzzy. A small world like Mars might be too cold at its current distance from the Sun, but if Mars was bigger and had a more dense atmosphere it might retain enough heat for liquid water. And we've seen that moons like Io and Europa orbiting a giant planet can get heat from tidal forces even if they're far from the Sun. So I'm going to figure the outer edge as the distance at which an Earthlike world would be below freezing, but keep in mind that there may be exceptions.

Compute the outer edge of the Goldilocks Zone using the equation D = square root of (Star luminosity divided by 0.82). Planets orbiting beyond that distance may have plenty of water, but most of it will be solid ice.

Mass: The other critical feature for planets is mass. Again I'm going to use Earth equivalents rather than kilograms, because the numbers are a lot handier.

We don't exactly know what determines how much mass a planet can gather up during early formation. Obviously, if you're using a real star system, just use the estimates for planetary mass that professional planetary scientists have worked so hard to come up with for you.

But for an imaginary star system, you're basically free to assign any value you want. The lower end is around the mass of Pluto (0.002 Earth mass), while the upper end is somewhere in the range of brown dwarf ojects (3000 to 16,000 Earth masses). Above that it's a star. Looking at actual exoplanet data, the biggest known exoplanet is a brown dwarf about 30 times more massive than Jupiter, or about 10,000 Earths.

Among planets, as among stars, it's a good rule of thumb to assume that there are a few big massive ones, a larger number of medium-sized ones, and a whole lot of small ones. If we look at the known multiplanetary systems (courtesy of Wikipedia: https://en.wikipedia.org/wiki/List_of_multiplanetary_systems), one can see that this is pretty accurate — although keep in mind that most of these systems probably have long-period planets we don't know about.

I'm going to lump planets into classes based on mass, and you can put them into your fictional star system as you like. Remember that massive worlds may interdict nearby orbits (see "Failed Planets" in the previous post).

Brown Dwarfs: Really big planets, with a mass above 1000 times that of the Earth. I doubt there will be more than one of these per star system, and they don't appear to be very common. If you're generating the system randomly, generate a random number from 1 to 10, subtract 9, and the result is the number of brown dwarf-sized bodies orbiting the star. The mass is a 10-sided die roll times 1000 Earths.

Gas Giants: Big planets like Jupiter or Saturn, with a mass from about 50 to 1000 times that of the Earth. Our own Solar System has two. If  you're randomly generating a star system, I suggest rolling a 6-sided die and subtracting 2 to get the number of gas giants in a star system. Put them beyond the outer edge of the Goldilocks Zone (unless you want a Hot Jupiter, in which case put it close to the star and eliminate all the planets between its orbit and the outer Goldilocks edge).

Generate the mass of a gas giant by rolling a 10-sided die and multiplying the result by 50 to get the mass of the planet in Earth masses. If the die result is 10, reroll the die and multiply the new result by 100 Earths instead. This method means most gas giants will be between 50 and 450 Earths, with a small number of really big ones.  

Remember to check for "failed planets" in orbits near your gas giant worlds.     

Ice Giants: This is what we now call planets like Uranus and Neptune, with several times the mass of the Earth but low density. I'm giving them a mass from 5 to 50 times Earth. This seems to be a pretty common planetary type, so just roll a 6-sided die to find how many to put in your planetary system. Put them anywhere in the star sytems — astronomers have identified several "hot Neptunes" orbiting other stars. Generate the mass of an Ice Giant by rolling a 6-sided die and multiplying the result by 5 Earths; if the result is a 6 then re-roll and multiply by 10 Earths instead.

Super-Earths: This class of planet has no representatives in the Solar System, unless perhaps Earth itself qualifies as a smalle example. It refers to dense worlds of rock or metal with a mass of 2 to 10 times that of Earth. There may exist "Mega-Earths" with masses of 10 or more Earths, but the observations are in dispute. To get the mass of a Super-Earth generate a result from 1 to 100 using percentile dice or a random number generator, and divide that by 10 (because we're getting into the regime where small differences in mass matter). If you want to allow for Mega-Earths, re-roll any resuilt of 100 and divide by 5 instead.

Rocky Worlds: This is the category that the four inner worlds of the Solar System fit into. Their masses range from 0.1 Earth (Mars) to 1 (Earth, obviously). Use them to fill up any remaining orbits in your star system.

I'm going to assume that rocky worlds can actually meet the lower bound of Super-Earths, so we're actually looking at a range of 0.1 to 2, but it seems likely that the majority of rocky worlds will be small. Generate mass for a rocky world by rolling a 12-sided die and dividing the result by 10. If you roll a 12, re-roll two dice and divide the result by 10.

Dwarf Planets: If you really, really want to spend a lot of time, you can generate dwarf planets for your star system. They have a mass of less than 1/100 Earth, and you can stick them into "failed planet" orbits along with a lot of asteroids, or beyond the outermost planetary orbit in the star's Kuiper Belt.  

Next Time: Planetary Composition!

Notes on Worldbuilding, Part 5: Planetary Systems

Until the boom in exoplanet studies, we really knew nothing about how planetary systems form. In the old days, with only the Solar System as our guide, it looked simple: small rocky worlds near the Sun, big giants in the outer regions. But then we began to observe things like "hot Jupiters" orbiting almost close enough to touch the parent star, giant rocky "super-Earths" and other weirdness. So now we really have no idea if there are any hard and fast rules for how planetary systems form.

Here are some rules of thumb, which aren't laws of nature but are pretty good guidelines.

Scale: The size of a planetary system seems to be roughly in proportion to the mass of the parent star. Our Sun's planets orbit in a range from about 1/3 Astronomical Units for Mercury (45 million kilometers) out to 30 AU (or 4.5 billion kilometers) for Neptune, with small bodies comets extending out to ten thousand times that distance.

An Astronomical Unit is the distance between Earth and the Sun, and is going to come up a lot in this post. It's 148,800,000 kilometers, and for most purposes you can approximate it as 150 million km.

By contrast with our Solar System, the red dwarf TRAPPIST-1 (http://www.trappist.one) has a tenth the mass of the Sun, and its planets are all crowded into a band between 1.5 million and 9 million kilometers (0.01 to 0.06 AU) — roughly the scale of the moons of a planet like Jupiter.

 So if you're creating a planetary system, multiply the mass of the parent star (in Solar masses) by about 35 AU and treat the result as the outer limit for planetary orbits in that star system. In the case of binary or multiple star systems, use that number or 1/3 of the distance between the primary star and its closest companion star, whichever is smaller.

How Many Planets?: The Sun has eight planets plus some dwarf planets and a lot of small stuff. Other stars with known exoplanets have fewer major planets, typically just two or three known. But that's probably an artifact of the methods we use to detect exoplanets: the periodic dimming of the primary star as orbiting planets pass across its disk as seen from Earth is going to be limited by the time astronomers have been watching. Distant, slow-orbiting worlds simply haven't been detected yet. We didn't know about Uranus and Neptune for most of human history.

In the absence of any other data, I'm going to use the Solar System as a baseline for how many planets a given star possesses. If you want to generate the number randomly using 6-sided dice, roll 3 dice and subtract 2 from the total, to get a number from 1 to 16 with the peak between 8 and 9.   

Spacing: The planets of the Solar System follow an interesting rule of spacing known as Bode's Law (https://www.britannica.com/science/Bodes-law), but other known planetary systems don't seem to have the same orbital spacing, and apparently it's just a bit of numerology. You can basically put planets wherever you want in another star system.

 It does seem to be true that the interval between planets increases with distance from the parent star, so my suggestion is to pick the distance for the innermost planet and then work outward as follows.

1. Innermost Orbit: multiply the mass of the parent star (in Solar masses) by a randomly-generated number between 0.1 and 1. My suggestion is to use a 10-sided die and divide the result by 10. The result is the orbital distance of the innermost planet in Astronomical Units.
2. Multiplier: determine a multiplying factor for the spacing of the next orbit. Again, I suggest using a 10-sided die roll divided by 10, and then add that to 1, so you have a number from 1.1 to 2.
3. Second Orbit: Multiply the innermost orbit by the Multiplier to get the distance of the second orbit around the star.
4. Recalculate the Multiplier: Roll a 6-sided die, subtract 1 from the result, and divide that by 10, to get a number from 0 to 0.5. Add that to the existing Multiplier from step 2 to get a new Multiplier. Use that to calculate the third orbit, and so on. Never use a lower Multiplier, and stop when you get a result greater than 2. If the distance exceeds the maximum size of the system, stop there. I tried it a few times and got star systems that look like this (with a parent star equal to the Sun in size). All distances are in AU.

Orbit:                   1         2         3          4          5           6          7          8          9      

Solar System       0.4     0.7       1        1.5        3*          5         9.5       19        30

System A             0.5     0.6     0.7       1.2       2.6        5.4       11.3     23.6  

System B             0.3     0.5     0.8        1.5       3.4        7.7       17.7  

System C             0.5     0.8     1.4        2.4       5.1       10.7      22.5  

*Note: the 3 AU orbit in the Solar System is the Asteroid Belt, which we will consider a "failed" planet as described below.

"Failed" Planets: There's a big gap in the Solar System between Mars and Jupiter — bigger than the distance from Mars to the Sun, in fact. How come? Jupiter itself is the prime suspect. In much of the space between Mars and Jupiter, an orbiting world will "synch up" with Jupiter every few years, and experience Jupiter's distant but powerful gravitational pull. The result (according to current theories) is that much of the matter in that part of the Solar System during the planetary formation era got knocked away by Jupiter, or captured by the giant itself. There simply wasn't enough left to form a planet.

This suggests that a large gas giant — anything bigger than about 100 times Earth's mass — will disrupt the orbits between about 0.4 and 1.6 times its own orbital distance. Leave those empty or make them asteroid belts.

Migrating Planets: One of the most startling exoplanet discoveries was the category of worlds called "Hot Jupiters." These are large, massive gas giants orbiting extremely close to the parent star. Current theories of planetary formation simply don't account for that — close to a star it should be just too hot for a gas giant planet to form.

So the current explanation is that in some systems large worlds form and then migrate inward as a result of interactions with other giants. So if you swap Jupiter and Saturn in the Solar System, the smaller inner world would get kicked into an eccentric orbit, gradually dropping down over millions of years until it orbited close to the Sun.

Needless to say, such a world would probably sling away or obliterate all the worlds whose space it passed through during that period. So if you want a giant world close to the star, pick an orbit in the outer system, beyond the "Goldilocks Zone" where water stops being a rock, and erase everything between that orbit and the one you want to put it in.

Nomenclature: Back in the good old days, science fiction writers had worked out a great convention for naming planets circling other stars. Obviously they'd be numbered, from closest to farthest out. So there would be planets like Tau Ceti III or Fomalhaut IX. Nice, simple, and elegant.

Naturally astronomers had to go and mess it up. Astronomers ruin everything. Their system (which has been around for a long time, actually) is that companion bodies of stars get labeled with letters in order of discovery, beginning with "b" because A is always the primary star. It's good for their purposes, but maddening for everyone else. So Tau Ceti has eight possible planets, and going outward from the star they are b, g, c, h, d, e, f, and i. Inelegant! Infuriating!

Fortunately, even astronomers think this is kind of lame, so the International Astronomical Union has started the Named Exoworlds project to bestow actual names on stars deemed sufficiently interesting, and on their planets. Wikipedia has a list of them here: https://en.wikipedia.org/wiki/List_of_proper_names_of_stars. The IAU retains the final approval authority so we're not going to get any objects named "Planet McPlanetface" any time soon.

So the star 55 Cancri A got renamed Copernicus, and its planets b, c, d, e, and f became Galileo, Brahe, Lipperhey, Janssen, and Harriot. A vast improvement.

For a fictional world, there are several options:

1. Use a real name if it's a real planet.
2. Use a plausible name (without duplicating one of the ones already taken). Something a bit stodgy and uncontroversial.
3. If the planet is going to have native intelligent life, use their name for it, or the closest approximation a human mouth can manage.

Next Time: Placing Planets!

Notes on Worldbuilding, Part 4: Stars!

Now we're going to get into the real nuts-and-bolts part, creating worlds and aliens. Note that you can start this process at either end: begin with the star and work your way through the planet to its life forms and their society, or start with what you need for the story and work backward. Either approach is fine. I'm using the cosmic to domestic progression simply because the steps build on each other more simply.

Real Planets: The simplest way to create a star system or a planet is to let the Universe do the work. Until about 20 years ago we only knew about the worlds of the Solar System, and there was still uncertainty about how common planets actually were. The first exoplanet was cornfirmed in 2002, and since then the floodgates have opened. Nowadays we have basic orbital data and at least an estimate of mass for thousands of planets circling other stars. I know of at least a couple of SF writers who have basically given up on creating their own alien star systems and just pick real ones.

You can find them in a number of places. NASA's Exoplanet Archive is here: https://exoplanetarchive.ipac.caltech.edu, and the European Extrasolar Planets Encyclopaedia is here: http://exoplanet.eu. There's also the Open Exoplanet Catalogue: http://openexoplanetcatalogue.com. And finally Wikipedia's various lists of exoplanets are pretty handy: https://en.wikipedia.org/wiki/Lists_of_exoplanets.

Stars: Unless you want a dark and cold "rogue planet" drifting in interstellar space, your alien world needs a star. We think that planets all form in stellar systems, though some get kicked out by interactions with other planets.* The choice of star is thus very important.

Stars are classified by temperature (color) which roughly correlates to mass, because the bigger a star is, the hotter and brighter it is. The classes range from O (huge and searingly bright blue-white giants) through B, A, F, G (yellow stars like the Sun), K, and M (red dwarfs); plus white dwarfs, brown dwarfs, and black holes.

Note: when I talk about stars here, I'm talking about stars on the "Main Sequence," the time when their energy output is relatively stable, which occupies most of a star's lifespan. Stars leave the Main Sequence as they deplete their fuel, swelling to red giants and then either collapsing or blasting apart in a supernova explosion, depending on how big they are. Those are not good candidates for lifebearing worlds, although one could certainly tell an interesting story of a civilization watching its star enter those last stages.

To have an Earthlike world a star needs to live long enough for planets to form and life to evolve. Big bright stars live fast and die young — which means that most of the stars you've heard of are unlikely to have lifebearing worlds. Star Trek loves to put planets around familiar stars like Rigel, Deneb, Spica, or Canopus — and all of them are bright short-lived stars unlikely to have any planets at all, let alone habitable ones.

Red dwarfs make up the majority of all stars, between 2/3 and 3/4, but they have problems, too: they are so dim that habitable worlds are likely tidally locked, and they tend to be flare stars. These aren't insoluble problems; a large moon orbiting a planet in the habitable zone of a red dwarf would have a day-night cycle, and there have been suggestions that most red dwarf flare activity is along their polar axes, where planets don't typically orbit.

The sweet spot for stars to have worlds capable of supporting life like our own appears to be in the F-G-K range. These stars have masses between 0.5 and 1.5 times that of the Sun, and brightness from about 0.1 to 7 times the Sun's output. They also have a long enough lifespan for solid worlds and complex life to form. Collectively they make up about 20 percent of the stars in the Galaxy.

Note that about a third of all star systems are binary or multi-star systems, with two or more stars orbiting around each other. These range from very close "contact binaries" to stars so distant it's hard to tell if they're actually orbiting or just passing nearby. They can and do have planets, so if you want a Star Wars style double sunset you can have it. For a close binary just treat them as a single star. Distant binaries function as separate star systems which just happen to be near each other.

If you want to use real stars (with or without known exoplanets), the best resource is SolStation.com: http://solstation.com/#sthash.IUQ23ZbY.dpbs. They've got encyclopedic descriptions of stars out to 100 light-years from Earth (http://solstation.com/stars.htm#sthash.htedKjWl.dpbs), plus star maps and very useful articles on planets and the potential for life.

Winchell Chung (his name's going to come up a lot) has not one but three very useful pages. The first is his "Atomic Rockets" discussion of stars and other stellar objects: http://www.projectrho.com/public_html/rocket/spacemaps.php, which includes some extremely useful "delta-V" maps of the Solar System, plus increasing scale maps out to intergalactic scale. His page on "Weird Astronomy" (http://www.projectrho.com/public_html/rocket/weirdastronomy.php) discusses real objects of interest. He also has a separate site for creating star maps, showing where things are in the Solar neighborhood. Very useful if you're creating a game setting or fictional "sandbox" with multiple inhabited star systems. It's at http://www.projectrho.com/public_html/starmaps/, and includes links to other star catalogues I haven't used.

And, again, Wikipedia is surprisingly useful, presumably because there aren't any aliens online engaging in edit wars. Their list of Lists of Stars is a good starting point: https://en.wikipedia.org/wiki/Lists_of_stars.

The most important features of a star are its mass and its luminosity. These are related: the bigger, the brighter. There are any number of sites listing the mass and brightness for different star types. If you want to make up a star rather than use a real one, just pick the type you want.

Brown Dwarfs: In recent years astronomers have gotten better at detecting "brown dwarf" objects — too small to be luminous, but big enough to generate some heat of their own. Basically they're the intermediate step between giant planets like Jupiter and the small end of red dwarf stars.

There's three very interesting things about brown dwarfs. First, there are probably an awful lot of them. In general, stars are more common the smaller they are. I've mentioned that nearly 3/4 of stars are red dwarfs. If this distribution holds true, then there are probably more brown dwarfs in the Galaxy than there are visible stars! Interstellar explorers and colonists might leapfrog from one brown dwarf to another across relatively short distances.

The second is that they do emit heat, and can have moons. A brown dwarf might have icy moons with subsurface liquid oceans, as is suspected of Europa and Enceladus in our own Solar System. That means a lot of places where life might arise.

And third, the atmospheres of smaller brown dwarf objects can be at "Goldilocks" temperatures where liquid water and — potentially — life might exist. As I mentioned, there are probably lots and lots of brown dwarfs, so even if it's unlikely for life to form on one, with enough rolls of the dice the right result might come up.

Wikipedia really has the best list of brown dwarfs I've been able to find, and includes links to full articles on some of them: https://en.wikipedia.org/wiki/List_of_brown_dwarfs.

*Or by the I.A.U.

Next Time: Systems!

Notes on Worldbuilding, Part 3: The Future!

Fictional worlds which differ from our own because of some scientific or pseudo-scientific rationale are basically the definition of science fiction. After all, SF stories have all kinds of plots, all kinds of characters, all kinds of themes, but they all take place in a "science fiction world." Sometimes that's as simple as "the modern day, but with aliens invading," and sometimes it's as complex as Dune's distant future empire.

Future Earths: One keystone of SF worldbuilding is creating plausible future societies. As any study of the history of SF indicates, this is actually very difficult. Pretty much any future society as depicted more than about 20 years ago seems ridiculous to modern audiences. To actually predict the future of Earth would obviously require a complete understanding of contemporary Earth, and nobody has that, either. Even professional analysts who get paid to forecast the next few years for governments and businesses have a mediocre record at best. If you can do better, you're in the wrong line of work writing fiction.

Most future worldbuilding in SF tends to take the approach of assuming one particular current trend will continue or increase, and show the effects of that. Often it is taken to an absurd level deliberately, so that the story can serve as a dire warning about what we shouldn't be doing. This gives you stories like The Space Merchants or Make Room, Make Room! (aka Soylent Green). The author takes a negative trend, extrapolates it into the future with no countervailing influences, and there's your dystopian future!

I think one reason the Cyberpunk school of SF made such a huge splash in the 1980s and 1990s was simply that it wasn't based on dire warnings. William Gibson and other cyberpunk authors seemed genuinely interested in depicting the "real" future rather than writing polemics.

A contrary approach is to look at history as cyclical. We're in a "Progressive" and globalizing era now? Assume 2030 will be conservative and nationalistic. Religion is on the wane in the West? Assume another "Great Awakening" of faith. This is a good way to show some contrast and perspective on current attitudes — but sadly a lot of contemporary readers may view your forecast as advocacy, and attack you for not having those current attitudes.

The big trick to creating future societies isn't actually predicting the future, it's persuading the audience that this is a plausible future. Which is one reason older futures seem so unconvincing: they were written for a different audience. Tastes and assumptions change.

So, how do you do it? How do you make a plausible future? You know what I'm going to say: do your research.

Look at population trends. I'm a little annoyed by the writers and moviemakers who still trot out the "overpopulated dystopian future" trope when in the real world, people who get paid to forecast future trends for government and business are warning about underpopulation in large parts of the developed world. China and Russia are set to halve in population over the next half-century. Japan and Europe are nearly in the same boat. Will those trends continue? We don't know. But any plausible future Earth should include a billion Africans and half-empty ghost cities in Europe and Asia.

Look at economics. One of the greatest accomplishments in human history happened quietly in the 1990s and early 2000s as a billion people in the poorest parts of the world gained middle-class incomes and lifestyles. This will have knock-on effects, including some nobody will predict. But it's definitely a game-changer.

Look at technology. Not just the marketing press releases rewritten as "technology news" about next year's cell phones or self-driving cars, but actual fundamental changes in tech. If Elon Musk's "Starship" can cut orbital launch costs tenfold (he claims a hundredfold, but I doubt it), then all sorts of things which were impossible suddenly become feasible. If cheap and clean power from fusion ever stops being "ten years away" we'll see a very different world.

And when you've done all that research, think it through. Consider how these changes will affect people's lives and attitudes. For example, SF writers predicted videophones and wrist phones or pocket phones since the 1940s — but who envisioned the social effects? Video of an arrest sparking riots, Ukrainians using their phones to track an invasion in real time, a weird culture of alienation and exhibitionism on social media, all from mobile phones with cameras in them.

Obviously you can't really predict the future, and your predictions are probably going to be wrong. (I wrote a whole essay about how lousy SF writers are at predicting things.) But you should at least try, so that your reader will come away with the feeling that your future world is solid, not cardboard.

Here are some useful resources for info and speculation about future politics, economics, and technology:

Army Mad Scientist Blog: this is a great site run by the Army's Training and Doctrine Command. They spend a lot of time thinking about future military issues and technology. https://madsciblog.tradoc.army.mil

Army Table of Future Technologies: taken from the Mad Scientist Blog, this is a footnoted table of future military tech, organized by how far out it is in time. The embedded links lead to other sites of interest. You can download it here: https://community.apan.org/wg/tradoc-g2/mad-scientist/b/weblog/posts/table-of-future-technologies

Imperial College London Table of Disruptive Technologies: a listing of new technologies, organized by how disruptive they may be and how soon they may appear. The notes are a little vague sometimes, but they do give one a direction for research. https://www.imperial.ac.uk/media/imperial-college/administration-and-support-services/enterprise-office/public/Table-of-Disruptive-Technologies.pdf

Isaac Arthur: Mr. Arthur runs an excellent YouTube channel about science and futurism. The main focus is on space exploration and colonization, but over the years he's touched on all sorts of topics. https://www.youtube.com/channel/UCZFipeZtQM5CKUjx6grh54g

NASA Innovative Advanced Concepts: This is a round-up of research NASA's been sponsoring on cool stuff. https://www.nasa.gov/directorates/spacetech/niac/NIAC_funded_studies.html

National Space Society Roadmap to Space Settlement: A step-by-step plan to colonize the Solar System, with nice listings of what technology is needed for each step. https://space.nss.org/nss-roadmap-to-space-settlement-3rd-edition-2018-contents/

Stratfor: this is a for-profit think tank with loads of good material. They do charge for their reports, so you either have to buy a subscription or make friends with someone who already has one. https://worldview.stratfor.com

Winchell Chung: This series I'm writing is going to make a lot of references to Mr. Chung's famous "Atomic Rockets" Web site, which has expanded over the years to cover a staggering variety of topics. Here's his page about future histories: http://www.projectrho.com/public_html/rocket/futurehistory.php#id--Predicting_the_Future

Next time: The Stars.

Notes on Worldbuilding, Part 2: Why?

From here on, science fiction is going to be the primary focus, so when I say "worldbuilding" just assume it's SF. And, yes, I'm eventually getting to the nuts and bolts, but I think it's important to understand what we're doing and why, before we get to the how.

Why? The first thing to keep in mind is that fantasy and science fiction are still FICTION. They are stories, and if they don't work as stories no amount of research and worldbuilding matters. All worldbuilding should be in service to the story.

Fitting the world to the story is kind of the Platonic ideal of worldbuilding from a literary standpoint. If you want a particular thing to happen to particular characters, create a world in which it can happen. I will discuss the opposite approach, fitting the story to the world, below.

The "Cambias Test": This is a concept I came up with long ago as part of a failed roleplaying game project. It's a simple question: "Does this setting allow stories which would be impossible in the real world or in the past?" If the answer is no, then just use one of those instead.

An obvious corollary of the Test is that you only need to do as much worldbuilding as you need to for a particular story. By this I don't mean that you should be slapdash or lazy. Merely that a short story's background can be lightly sketched in, whereas a novel series probably needs a detailed world. Too much "world" can choke a story if you're too enamored of what you've created and try to shove it all in the reader's face.

Stealing: It is very common to use renamed or "reskinned" versions of real-world places, events, and cultures for fiction. It's practically universal in fantasy, and quite common in SF.

There are some good reasons for this. A real-world society has its own internal logic and will hang together. If your horse nomad tribes look and act like Mongols or Comanches, then they make sense because those real-world horse nomads made sense. However, see the Cambias Test above: if your story could be set in medieval Mongolia or early 19th-century Texas, why not just do that?

Fitting the Story to the World: Of course, sometimes there are real-world reasons for why you can't make up a world to fit your story.

One obvious constraint in SF is if you're using an actual planet or celestial body. You can't make up details unless they're consistent with what is already known, so in effect you have to fit the story to the science.

For example, in his classic novel Mission of Gravity, Hal Clement read an article about an exoplanet orbiting Barnard's Star. (Note, Peter van de Kamp's announcement in the 1960s of a giant planet orbiting Barnard's Star turned out to be an experimental error. But in 2018 a planet was discovered circling Barnard's Star, very different from what was described fifty years before.) Anyway, Clement started with what was known: an estimate of the planet's mass and orbital period. Using then-current theories he came up with a plausible world — a huge, high-gravity planet spinning at an insanely fast rate so it's distorted into a fat disk shape, with surface gravity of about 3 times Earth at the equator and something like a hundred times that at the poles. Then he created a biosphere and intelligent beings living on that world — and then he came up with a story.

Fitting the story to the world also becomes important when you're working within a "sandbox" setting you've already created. Or if you're using a published setting invented by someone else.

Okay, so you've got a setting, either a real world or maybe a planet in some other writer's sandbox. How do you create a story for that environment — one which actually makes use of it, not just as windowdressing? There are several things to hang a story on.

Impossible Desires: What can or can't happen in that fictional world which is different from our current environment? This is basically a character motivation version of the Cambias Test. If the setting creates a desire different from the desires of people on contemporary Earth, then there's your story right there. 

Signature Events: Is there an important or characteristic occurrence in that world which doesn't happen here and now? For example, Mercury rotates with a period of 176 days relative to the Sun, and of course it's very close to the Sun so daytime temperatures can get up to 700 Kelvin (about 800 Fahrenheit). That's deadly, but the slow rotation means a walking human could stay ahead of the dawn line. That's a good story hook. (I know it's been done but I can't track down the story or the author.)

Playing Tourist: Exploration of a sufficiently interesting setting can be all the excuse you need for a story. Jules Verne did it, H.G. Wells did it, Larry Niven did it, and I've done it myself. You'll want a goal to pull your characters across the world, and then the places they visit can hinder or help them on their way. It's useful if that goal is an Impossible Desire not available on Earth in 2022, and the characters experience a Signature Event or two along the way, but not necessary. This approach does depend on just how good your setting really is. If you happen to know more than almost anyone else about a real planet or moon, just do this.

Hardness and Handwaving: How realistic should you be? It depends. As I mentioned above, science fiction is fiction. Scientific accuracy is a literary choice for your story.

In particular, very realistic or "hard" SF makes the reader's "suspension of disbelief" easier, because all the elements of the story really could happen. It allows the author to do some pop-science tourism, showing off cool stuff that really might exist. And it supports themes of discovery, gritty realism, knowledge as power, and humanity's place in what appears to be an amoral Universe.

But there are degrees of hardness. You may allow yourself one or a few pieces of "magic" like faster-than-light travel, gravity control, or whatever. This allows the author to put recognizable near-future humans into alien environments on planets of other stars, interacting with alien civilizations or travelling into the past. A solidly "hard SF" story of interstellar travel would have to be set centuries in the future, when humans and human society are almost unrecognizable to a contemporary reader. So sometimes less hard science makes it easier for the reader to accept a story!

And if you want or need something blatantly unrealistic, like prophecies or interbreeding with aliens, then go ahead and do it. As I said, accuracy is a literary choice. However! To sell that unreal element, make the other aspects of the story as plausible and grounded as possible. Weird stuff happening to weird people in a weird setting is probably just too damned weird.

Next Time: The Future!

Notes on Worldbuilding, Part 1: Real and Not-Real Worlds

A couple of years ago I taught a class via Zoom for the Pioneer Valley Writers Workshop, on science fiction worldbuilding. Now I've decided to write up and expand my notes for that course and post that all here in a series of blog posts. My ultimate goal is to keep adding to this series and improving it, making a resource for science fiction writers, game designers, and anybody else who cares.

Let's begin with a definition: Worldbuilding is the art (and science) of creating settings for works of fiction, and conveying knowledge of that setting to the audience. Note that this doesn't restrict worldbuilding to science fiction or fantasy — it's part of all fiction.

Real-ish Settings: "Real world" worldbuilding means creating fictional towns and countries, and integrating them into the real world. Examples include Faulkner's Yoknapatawpha County, Conrad's republic of Costaguana, any basically any fictional setting which isn't in the atlas. This often shades over into alternate history (see below), since the existence of fictional cities and countries would logically change the course of events, though in practice authors and readers simply ignore this.

In this sort of worldbuilding, your fictional locations have to be realistic — they have to look and feel like real places. The easiest way to do this is to base the fictional town on a real one in the same region, of about the same size and demographics. So if I want to write a horror story set in Deerfield, Massachusetts, but don't want to offend my neighbors by depicting them as inbred cannibals, I'll just move the town a few miles upriver (across the river from Northfield and nudging Bernardston to the west a bit), rename it "Kingsfield" and get on with the story. It will feel like a small Western Massachusetts town because it is a small Western Massachusetts town.

Historical Settings: A different set of worldbuilding problems confront writers who want to use real or "real-ish" (see above) settings, but in a different historical era. A subset of historical fiction involves secret history, where the events of the story are supposedly "real," but never made it into the historical record for various reasons.

The great thing about historical settings is that millions of people already did all the work of creating the setting for you. All you have to do is research. That research is your first problem. People who enjoy historical fiction tend to be very interested in, and at least somewhat knowledgeable about history, so you really have to get the details right.

Reading just one book isn't enough. You really have to dig in. Memoirs and travelogues from the period are your best resources. Some big scholarly tome by a 19th-century British writer is always useful. (No matter what the subject, that big scholarly British tome is certain to exist.) Period photos are incredibly useful, although they don't help with subjects before the invention of photography. If at all possible, talk to a historian who specializes in that period — they may be able to suggest other sources for more research.

The second problem when you're using historical settings is simply that the real world has no obligation to be plausible. I once wrote a rather snotty critique of a colleague's story set during World War II because the central event seemed so unrealistic. You guessed it: the story was based on a real incident.

That doesn't matter! You can't personally tell every reader "No! It's totally real!" Which means your task as a writer is to convey, through the story, that this blatantly implausible thing actually existed. This is a question of art, not science, so I can't tell you the infallible way to accomplish it. Suffice to say, the more realistically you depict your historical setting, the better the odds are that you can sell the reader something implausible — whether it's your own invention or a real thing.

Alternate Histories: The alternate history subgenre is lumped in with science fiction mostly for historical reasons: it was SF authors and readers who made alternate history a popular genre, and SF fans seem to like it. Basically the idea is to imagine if history had gone otherwise, and figure out what society would look like.

There are two main approaches to building an alternate history. The first is where you pick a year or an event, change it, and try to extrapolate what happens from there. The Civil War and World War II are the most popular alternatives, with a failed or averted American Revolution and the Roman Empire somehow not collapsing as perennial favorites. You pick a change and then apply your knowledge of how history works in order to see what comes of it.

The second approach is to pick an event to change which leads to the world you want to use for your story. This does mean you need to research and understand why the world didn't wind up that way historically. For example, if you want a story set in a world where the Aztec and Inca empires of the New World never got conquered by the Spanish, you have to address the factors which gave the Spanish an advantage: technology, disease, and endemic political instability in the two empires. This may mean your change point needs to be much earlier: send some Europeans across the Atlantic in the 1200s or so, so that the massive disease outbreaks have burned out before Columbus shows up. Maybe handwave that the aftermath of great plagues leads to social reforms in Aztec and Inca society which reduces their instability.

Whichever method you use, you need a very good understanding of the history around the event you're changing — both before and after. Do the research!

Fantasy Worldbuilding: Fantasy worlds, with magical powers and supernatural creatures, require more building than realistic or historical settings. In fact, one could argue that fantasy needs more worldbuilding than SF, since SF writers operate within the constraints of known (or hypothetical) science while fantasy authors need to imagine whole alternate systems of magic, unnatural biology, and mythologies.  

"Wainscot" Fantasies: The Enclyclopedia of Fantasy uses this term to describe fantasy settings coexisting with our own world, as in the Harry Potter series. In these, most of the worldbuilding is already done, as in realistic fiction. You've got the world outside your window, just with secret magic stuff going on.

The author does have to create the fantastic elements — and, crucially, explain why they don't have more visible effect on the world. It often winds up leading to secret history where the influence of the fantastic is being deliberately concealed. I used that approach in my book The Initiate, and I cleverly added a rule that most supernatural beings are immaterial spirits which affect human perceptions directly, inside the brain. So you may see a terrifying demon, but if you take its picture with your cell phone all you get is a distorted blob.

Secondary Worlds: A "secondary world" is a fantasy which takes place in a setting unconnected to the real world, as in The Lord of the Rings (though even that is supposedly an imaginary prehistory). This is where you can have pretty much anything you can imagine since you don't have to shoehorn it into Earth and its history. However, there is a big nerd question which crops up, and that is simply how are there humans in this alien environment? Usually the answer is, "because there are," and that's all. I'd love to see a secondary-world fantasy which featured nonhuman protagonists.

Creating a fantastic secondary world takes basically all the steps I'm going to outline for making alien planets and civilizations in this series of blog posts. Plus you have to put considerable and serious thought into the role of magic and gods in your world. Good luck!

A Note on Sensitivity, Inclusion and Appropriation: There are plenty of minefields waiting when you do anything involving real-world societies and cultures — and I can personally attest that there are people out there who will try to judge a fictional alien civilization by the standards of this week's conventional Twitter wisdom.

I think most of that is nonsense. The best way to be respectful or sensitive is to know what you're talking about. If you want to write about life on a Navajo reservation in Arizona, you have to have seen it for yourself. Go there, spend some time, talk to people, learn the in-jokes and the things which will make reservation residents nod in familiarity.

There's a common phrase in the fiction field: "Write what you know." I think it is better to reverse it: "Know what you write." Learn, find out, and do your damnedest to get it right. There will still be outraged critics. Nowadays there are people who literally make a living by being outraged. Don't worry about them. If you've gotten it right and told the truth, your conscience is clear. 

Next Time: Why?

The Trip to Iceland, Day 7

And so we say farewell to scenic Iceland. On our final day we had a big breakfast at our hotel, made a few last-minute purchases, and then packed up the car and left the Hotel Borg. We drove south, back to the Reykjanes peninsula, and spent some of the time before our flight doing some sightseeing in the otherworldly landscape.

The most interesting stop was at the "Bridge Between Continents," which spans a chasm which is part of the Mid-Atlantic Rift system. One side is attached to the North American plate, the other side is attached to the European plate. Presumably the bridge itself has some expansion joints to allow for the constant widening of the rift.

As seasoned international travelers, we allowed plenty of time at the airport — half an hour to drop off the car, half an hour to check in, half an hour to pass through security, etc. And, as happens so often, each of those things took about ten minutes, leaving us with an hour to kill in Keflavik International Airport.

Fortunately, it's not a bad airport to hang around in. We blew all our Icelandic currency at the duty-free shop on a bottle of Brinnevin and a box of Hraun. Lunch was one of the best airport meals I've ever eaten: a big Danish-style open-faced sandwich of smoked salmon, with boiled egg and salad greens.

The flight boarded on time and the trip back to Logan felt much quicker than the outward journey, even though I think it actually took a few minutes longer.

We both enjoyed Iceland enormously, and agreed that it would be worth a stopover next time we're flying to Europe.

Sir William Jackson Hooker's departure from Iceland was a little more worrisome than ours, as the crew of the ship which brought him there had more or less accidentally overthrown the government in the course of a dispute with the Danish governor, so when a British warship arrived to straighten things out Sir William had to cut short his lichen-gathering.

". . . and I here learned that the vessel, which we had observed the evening before entering the Bay, was the Talbot sloop of war, commanded by the Honorable Alexander Jones. . . . (T)he captain had proceeded without loss of time to Reikevig harbor, that he might have an opportunity of ascertaining more correctly the facts connected with a revolution in the government, of which he had heard at the former place, but had received only a short and unsatisfactory account. The consequence of these enquiries was his issuing orders, that the persons, principally concerned in bringing about this change of affairs, should with all possible expedition proceed to England, where a full account of all the transactions was to be laid before the British government. From this time, therefore, my researches to Iceland may be regarded as nearly at an end; and, though various circumstances prevented the sailing of our vessel until the twenty-fifth of August, yet the daily, and sometimes hourly, expectation of being called on board, prevented my making any excursion to a distance from Reikevig. (. . . ) My memory no farther enables me to continue my journal in any thing like a regular manner, but, even had this been the case, yet still such would be found the uninteresting nature of the events that happened, except, indeed, those political ones that are more fully detailed in the Appendix A, that they could afford but little amusement."

And on that note, our expedition comes to an end.

The Trip to Iceland, Day 6

My experience of travel to various places has taught me that one should always include a "slack day" in one's travel plans. There are a number of good reasons for having a day with few or no activities planned: you can re-visit someplace you really enjoyed, you can switch around if something is closed or too crowded, you have a cushion in case someone doesn't feel good or is simply too tired, and you've got a day to just hang out in another country. Sometimes that's the most enjoyable part of a trip.

Friday was our slack day. We got swabs stuck up our noses to make sure we'd be allowed to fly home, then got lunch at a conveyor-belt sushi restaurant on the waterfront. It was pretty good — in the same league as sush in Seattle, which was better than any I've had on the East Coast but not as good as in Japan. Evidently nose-swabbing gave us an appetite because we put away a lot of sushi, mostly salmon and tuna.

One aspect of "globalization" which has been consistently disappointing to me is the remarkable lack of culinary crossovers. Iceland is a country where people eat a lot of fish, sushi is a Japanese method of preparing fish, so why didn't I see any codfish sushi? Why don't the taco trucks in Philadelphia have cheesesteak tacos? Why don't po-boy places in New Orleans serve banh mi sandwiches?

I appreciate authenticity — but I'm also aware that most authentic cuisines are a mass of borrowing and hybridization. In this era of creeping global monoculture, it's odd that food lags behind. I think we're missing some potentially good dishes.

Anyway, the sushi was good. After lunch we got our swimsuits and drove down to the Nautholsvik Geothermal Beach, which is a public park at the south end of Reykjavik, just past the municipal airport and Reykjavik University (not to be confused with the University of Iceland, which is on the other side of the airport).

Like the much bigger, much fancier, and much more famous Blue Lagoon spa, Nautholsvik is not a natural geothermal spring. It's the hot water outflow from a geothermal heating system — which means the water temperature can be monitored and controlled to avoid boiling the patrons. Natural springs don't have thermostats.

Sir William Jackson Hooker described the "boiling springs" of Iceland in his day: "I rode, however, one morning, to the hot spring, where I found a tent pitched, and as many Icelandic women and girls as it could possibly hold, sheltering themselves in it from the weather. They had come with their linen, which was brought on horses from the town, to the hot spring, where all the clothes of the people, for many miles round, are washed. Some of them had a few little miserable potatoes, not so large as a full-sized walnut, which they were cooking in the spring for their dinner, and which they offered me. I had carried with me some eider-ducks' eggs, for the purpose of trying how long it would take to boil them hard, and I found they required ten minutes, whilst lying in a part of the water where the thermometer rose to 200°."

You get a sense of what a high-trust society Iceland has in places like the Nautholsvik locker room. There are no lockers, and certainly no locks. Just plastic baskets to stash your clothes — and keys, and wallet, and phone — in while you soak in the hot pool.

The water's about 100 degrees Fahrenheit, very pleasant on a day when the air temperature didn't get above 50. There is a sheltered cove with a sandy beach where crazy people like my science advisor can go and dip themselves in life-threatening cold seawater before retreating up the slope to the hot water of the pool. I didn't indulge in any such foolishness.

Perhaps it was an artifact of the time of day, mid-afternoon, but we were definitely in the "prime demographic" for the Nautholsvik geothermal pool — middle-aged and older adults. After we'd been there about an hour some teenagers arrived, but they were still in the minority when we left. Maybe teens don't feel as much of a need to soak their joints in hot water.

Relaxed almost to a comatose state, we showered off and then had dinner at what may be my favorite restaurant in Iceland: Islenski Barinn. It has local food like fish pie (more like a cross between shepherd's pie and chowder), dung-smoked salmon on brown bread, excellent crepes, and local beer. The decor looks unchanged since Iceland's independence, and it is considerably less expensive than Dill.

After some more strolling about, we finished up our last full day in Reykjavik with a glass of wine at one of the downtown hotels, and then to bed.