A world unto itself orbiting Jupiter, Ganymede is the largest moon in the solar system - bigger even than the planet Mercury - and combines many of the same fascinating features as its fellow Galilean moons with fewer of the hazards that would interfere with future exploration and development. It's a world of ice like Europa, but much bigger and with a potential liquid ocean much farther beneath the surface, causing its visible features to be older and more pockmarked by impacts. And yet the things that make it scientifically less interesting than Europa make Ganymede far more likely to become a future abode of human civilization.
The progress of our adventure so far (current in bold):
1. The SunGanymede in true color, as we would naturally see it if we were there:
4. Earth (Vol. 1)
5. Earth (Vol. 2)
6. Earth (Vol. 3)
7. Earth (Vol. 4)
8. Earth (Vol. 5)
9. Earth (Vol. 6)
11. Mars (Vol. 1)
12. Mars (Vol. 2)
13. Mars (Vol. 3)
14. Phobos & Deimos
15. Asteroids (Vol. 1)
16. Asteroids (Vol. 2)
17. Asteroids (Vol. 3)
19. Jupiter (Vol. 1)
20. Jupiter (Vol. 2)
22. Europa (Vol. 1)
23. Europa (Vol. 2)
27. Rings of Saturn
30. Tethys, Dione, and Rhea
33. Minor Moons of Saturn
35. Moons of Uranus
38. The Kuiper Belt & Scattered Disk
40. The Interstellar Neighborhood
Ganymede is the seventh moon of Jupiter, and the third Galilean moon. The size of its orbit varies between 62% and 66% larger than the Earth-Moon system, but the variance is overwhelmingly because of Luna's eccentricity - all of the Galilean moons have very circular orbits. Diagrams of the Jovian system and depth in Jupiter's gravity well:
Something to notice from the two diagrams above is that even though the distance from Jupiter is increasing by a larger margin with each subsequent Galilean moon, the gravitational "climb" is shallowing out, which is based on how gravity fields are shaped: The closer to the origin, the steeper the descent or climb involved in any given change in distance. This is because an inverse-square law applies to gravity based on distance from the center of mass - twice the distance means a quarter of the force, and three times the distance means 1/9th of the force. Since gravitational force from a body is never zero anywhere in the universe, this means that increasing distance approaches zero gravitational force asymptotically.
Relative gravitational depth is a very important factor in determining future patterns of human development on gas giant moons, as I discuss in more detail later. The reason is that it dictates the fuel/energy cost of trade in a multi-world economy, so the most populous and prosperous locations will be those with the optimum combination of available resources, habitability, and accessibility to other high-value locations. Ganymede scores highly on all three compared to most other major bodies in the solar system. A contextual image either rising or setting behind the limb of Jupiter:
As we've done with Io and Europa, we can simulate how big Jupiter would be in the sky of Ganymede's planet-facing side by adding the planet into the black sky of an Apollo photo taken from the Moon. We can do this using a handy NASA tool that tells you how big in angular terms any body is from any other at a given date and time - which wouldn't visibly change in this case because of how circular the orbit is (although Jupiter would still go through phases):
If you want to really imagine it, just picture the surface in the image as rounded hills of dirty ice rather than jagged rocks or sandy plains. The exact size generated by the NASA tool for Jupiter in the Ganymedian sky is 7.6°, which is a lot larger than the little over half a degree of Earth's Moon in the terrestrial sky, though not nearly as impressive as the views from Europa or Io (12.3° and 19.5°, respectively). Of course, you couldn't directly see those views anyway, because the radiation environment on those moons is too extreme to spend much time in front of windows - even thick, leaded ones. Ganymede is another story, so the view of Jupiter seen above could very well be something humans see on a regular basis through windows and spacesuit face plates. And given how bright Jupiter is even in Earth's sky, just imagine it from one of its own moons.
The original location of Ganymede in the gas and dust that formed the Jovian system was the densest outside of Jupiter itself, so it's thought to have formed very quickly (in about 10,000 years) in a process that trapped enormous amounts of heat in its interior that is still gradually bleeding away. As a result this residual heat, Ganymede is thought to have a subsurface ocean 200 km below the ice, although if there is life in it we won't know about it for a very long time given the difficulty of accessing it. However, it will be a lot easier to operate on the Ganymedian surface over extended periods of time than on Europa, so that could somewhat compensate for the greater difficulty of drilling /melting through such thick ice.
1. Orbital and Rotational Features
Ganymede's month is about 7 Earth days, 4 hours, and 42 minutes, and its day is the same because it's tidally locked to Jupiter - i.e., it always presents the same face to the planet, so it only rotates once every time it completes an orbit. It also has the least eccentric orbit of the Galilean moons, which is typical of a body that formed in the densest region of a planetary accretion disk (barring subsequent catastrophe perturbing its orbit). This long day will present some energy challenges to exploring its surface, since the night is three-and-a-half Earth days long, but this is actually more convenient than the two-week night of our own Moon.
However, the Sun is much weaker in the Jovian system, so there is no practical advantage relative to the Moon, all things considered - two weeks of charging batteries and storing heat in the lunar daytime is more than enough to deal with the lunar night because the Sun is so intense. Three-and-a-half days of doing the same on Ganymede wouldn't even be enough to survive the daytime temperatures, let alone provide energy during the night.
All four of the Galilean moons orbit in resonance with each other, perpetuating the intense tidal pressures responsible for Io's volcanism and Europa's theorized ocean. Since Ganymede has the largest mass and orbits much closer to these two than Callisto, it plays the single greatest role in the phenomenon apart from Jupiter itself. A low-angle perspective of the Galilean moon orbital resonance:
From a higher angle:
2. Size and Mass Characteristics
Ganymede is the fourth largest solid body in the solar system after Earth, Venus, and Mars, and larger - though much less massive - than Mercury due to the low density of water ice versus the large iron core that dominates the planet. It's somewhat short (about 15%) of being twice the size of our Moon, and is the largest satellite in the solar system, being slightly larger than Saturn's moon Titan. Despite a relatively low density, it is also the tenth most massive object in the solar system in general, and the fifth most massive solid object. Various size and orbital properties of solar system moons can be compared in the table at the bottom here, and a much wider range of parameters can be compared and organized among all solar system objects here.
Gravity on Ganymede is about 15% of g - slightly less than on the Moon and Io, but comparable. So videos of Apollo astronauts moving around on the lunar surface give an accurate idea of how explorers of the Jovian moon would look in terms of fall rate and momentum. If you weigh 160 lbs on Earth, you weigh 26.6 lbs on the Moon and 24 lbs on Ganymede. Size comparisons to comparable solid bodies in the solar system - mouse over to see the title if you don't recognize something:
The average temperature on Ganymede is about 110 K (-163 °C / -262 °F). Although it exists in vacuum, the spread between minimum and maximum temperatures is mitigated by the dimness of the Sun - it only goes up to 152 K (-121.15 °C / -186.07 °F) in full daylight at the equator, and down to 70 K (-203 °C / -334 °F) at minimum. Typical daylight temperatures on Ganymede are only slightly warmer than nighttime temperatures on the Moon. This is typical of Jovian moons because Jupiter is beyond the Frost Line - the distance from the Sun where H2O endures as ice in full vacuum rather than sublimating into gas.
However, it remains warm enough that most gases do not freeze in most places, at most times. This is why the surface is dominated by water ice rather than the ammonia (NH3) and methane (CH4) ices that begin to appear somewhat beyond Saturn and completely dominate by the time you reach Neptune. These "organic" ices are why many of the most distant objects in the solar system, such as Neptune's moon Triton in the last comparison image, are tinged pink. Bodies dominated by water ice, however, are either white or grey, depending on their histories and whatever kinds of dust accumulate on the surface from the surrounding environment.
4. Internal Features
The intense heat of its formation allowed Ganymede's interior to fully differentiate into layers of differing densities and compositions. Although it is known to have such layers, and more or less what the possibilities are, their exact proportions aren't yet known with any certainty. The Galileo probe had detected a slight magnetic field, so that's thought to be a strong indicator that the moon has a liquid or partly liquid iron core. Above that is rock, probably extending about halfway to the surface - all that would remain if you stripped away all the water ice.
The next layer is the ice mantle, which may or may not have "plumes" of liquid water or viscous slush rising from heat expelled by lower layers in ways similar to a rocky mantle. Convection caused by such movements could contribute to tectonic processes that disrupt the surface in some regions. If I understand the theory correctly - and I make no guarantees on that point, especially since even the experts don't seem to know very much about Ganymede - once the warmed liquid or slush enters the uppermost reaches of the mantle near the crust, it would encounter a layer where the changes in temperature and pressure allow it to flow outward globally to create the ocean. This liquid layer is thought to be about 200 km below the surface, and is believed to be sandwiched between two solid ice layers. A rough diagram:
The possibility of life existing in this ocean would seem to be in the same ballpark as Europa, and if it did exist would probably be similar - very basic microbes. But since the ice is far thicker than on Europa, and thus accessing the Ganymedian ocean much more difficult, there is little exobiological interest in the moon compared to Mars, Europa, and Titan. If there is life there, it's so deep that we won't find it until there are already thriving civilizations on Ganymede.
5. Surface Features
Despite its apparent similarities to our Moon, Ganymede has very little in common with Luna - its surface is water ice, not rock, and is relatively flat compared to the lunar surface. As a result, there are few tall mountains or craggy crater rims, because ice isn't strong enough to hold the weight of large vertical features compared to rock. What it does have that differentiates it from the other icy Galilean moons is a "two-toned" surface divided between older, darker, crater-riddled terrain somewhat like its more distant comrade Calllisto, and younger, lighter-colored, grooved terrain more like Europa. Whereas Europa has almost no craters, and Callisto is almost nothing but craters, the two surface types both occur in large regions on Ganymede. A full simulated rotation, showing the complete surface from an equatorial perspective:
A couple of global views, the second a mosaic:
A high-res partial global view:
The largest dark region and single most prominent feature of Ganymede, known as Galileo Regio, is visible most prominently in the introductory image, the size comparisons, and in the upper right portion of the image directly above. Although it looks like a lunar mare, it has nothing to do with that kind of feature - it did not form by impact, and is not a plain formed by upwelled fluid. Rather, it formed by default of nothing much happening to it while areas around it were tectonically active. It is very old, and pockmarked with craters whereas lunar maria are relatively young and smooth. The parallel linear features running along it are "creases" in the crust due to tectonic activity happening around it.
The broad, light-colored "river" features are actually crustal faults comprised of many smaller grooves where the ice has rifted apart due to interior convection. This is mostly not an ongoing process, but overwhelmingly a result of activity in the distant past, although some level of activity may be continuing. You can see how the rifting of the largest of these faults running North to South has impinged on Galileo Regio and caused its terrain to buckle. The most prominent rifting region - which runs from the top to the bottom right in the image just above - is called Uruk Sulcus and is thought to be the source of the pressure exerted on Galileo Regio. You may also have noticed the smaller dark region to the East of Gaileo that's divided into Northern and Southern parts: This is called Perrine Regio, and the fault that separates it from Galileo is known as Xibalba Sulcus. The East-West fault that cuts Perrine in half is called Nineveh Sulcus, after the ancient capital of Assyria.
As for craters, the most prominent are the massive white impact in the Southern hemisphere to the Southwest of Gaileo Regio called Osiris, and the Northern hemisphere impact seen in the full-globe mosaic to the Southeast of Perrine Regio called Tros. These are not the largest craters measured by rim diameter, but they are the most eye-catching due to their extremely large and bright ejecta blankets. Below are Zooms from the high-res image above, showing various features including Uruk, Galileo, and Osiris:
Here in the zoom of Gaileo Regio, you can see that it is not the smooth plain it appears from a distance, but a wrinkled, pockmarked terrain:
Projection map of Ganymede, divided into longitudinal thirds in order to fit the frame while remaining legible:
The US Geological Survey (USGS) maintains a list of all officially named features on Ganymede here, giving the latitude, longitude, and a description of the namesake. If there are features in the above maps you would like to know about whose labels are unclear, you can identify its latitude and longitude in the map and then find its name in the linked list. There are 184 approved names for Ganymedian features, and also 6 proposed names listed that were dropped or voted down. Nomenclature guidelines established by the International Astronomical Union (IAU) for naming features on Ganymede are as follows:
Craters and Catenae (crater chains): Gods and heroes of ancient Sumeria, Mesopotamia, Phoenicia, and Egypt.And now for some more detailed imagery, beginning with high-altitude regional views. They're all in black and white, as far as I can tell. If I don't say what it is and you want to know, mouse over to see the title:
Facula/Faculae (bright spots): Places from Egyptian mythology.
Fossa/Fossae (long, narrow, shallow ditches): Gods and rulers from ancient Sumeria, Mesopotamia, Phoenicia, and Egypt.
Patera/Paterae (irregular craters): Valleys of the Fertile Crescent.
Regio/Regiones (regions): Astronomers who discovered Jovian satellites.
Sulcus/Sulci (Systems of parallel grooves and ridges): Places from miscellaneous mythologies.
Marius Regio and Nippur Sulcus:
In some of the next images, you'll see a lot of similarities to imagery from Europa, but there are profound differences as well. A good way to tell apart imagery from Europa and Ganymede is that even in relatively smooth regions of the latter, they tend to be peppered with craters while Europa's linear features are overwhelmingly smooth and intact. Mid-altitude views - particularly note the ones depicting boundaries between sulci and regiones:
A couple of comparisons of Europa and Ganymede Sulci terrain with identical resolutions - see if you can identify which is which (answer is in the image title - mouse over to see):
Boundary regions with topographic details:
High-res images - some of the titles state the resolution, if no scale bar is visible:
An ice caldera in Sippar Sulcus - such things form when warm slush or liquid water ooze to the surface, like a water "shield volcano":
A catena from the impact of a comet that had fragmented:
The most awesome image of all: The highest-resolution ever taken of Ganymede, with the smallest details being only 11 meters across. Ice hills:
Daily radiation dose on Ganymede is about 80 mSv (millisieverts) or 0.08 sieverts (Sv) per day. This is 450 times smaller than radiation on Io (36 Sv), and about 68 times smaller than the dosage on Europa (5.4 Sv), and that fact makes a big difference in the practicality of exploring and settling Ganymede. It is still significant - about 28 times the yearly background on Earth's surface - and shielding would be required, but the level of shielding needed is within the realm of the practical. Even unshielded, it would take several weeks of this exposure to suffer from radiation sickness, and months to receive a lethal dose.
All that would be needed to bring the radiation level down to yearly Earth ambient would be 3 feet of water shielding or 2 inches of lead on exposed surfaces, although I'm not sure how being in ice phase affects the calculation for water. But since habitats of any significant size would likely be bored into the ice anyway - with heavy insulation to avoid melting - the shielding requirement is probably trivial. In order to visually compare the radiation environment on Ganymede with Europa and Callisto, we have to vertically extend the bar chart quite a bit:
Again, despite the comparative safety it is still a lot of radiation - equivalent to having 2-3 full-body CT scans every day. So there would need to be limitations on time spent walking around outside in spacesuits, and the fabric would probably need to be leaded to minimize exposure. Electronics would also have to be rad-hardened to function outside of shielded habitats and vehicles. But it is manageable.
V. Modern Relevance to Humanity
Modern interest in Ganymede is exclusively scientific, but in that domain it's almost entirely overshadowed by its inward brethren Io and Europa. It is nowhere near as dynamic as either of them, and its inner position relative to Callisto has made it somewhat less attractive as a setting for science fiction explorations of Jovian system colonization. Most references to Ganymede in SF literature are made in passing rather than playing any central role in the plot. This, I think, is ironic given the profound role I see for it in the real future.
VI. Future Relevance to Humanity
What are the most scientifically interesting places on Earth? You probably wouldn't think of places like Manhattan Island, the Seine river valley, or most other places that have become hives of human activity - and that would have been true even before they were ever settled by humans, because the fact is that boring places with mediocre climates and a middling level of resource availability tend to be easier to settle than "interesting" places. The majority of the most scientifically fascinating areas on Earth - the Amazon, Antarctica, the Arctic, the Himalayas, the Sahara desert, the Andes - to this day have relatively scattered human activity compared to middle-case environments like Italy and Britain, and we can expect a similar pattern to unfold in the settlement of the Jovian system.
Io and Europa command nearly the full attention of today's scientific interest in Jovian satellites, but some day Ganymede and Callisto will be the overwhelming focus of human development and settlement activity in that part of the solar system precisely because they're not nearly as interesting. The radiation levels are much lower, gravity is comparable to the Moon, water ice is just as abundant as on Europa, and they orbit in much shallower regions of Jupiter's gravity well than their more interesting inward brethren, meaning they are easier to reach from elsewhere in the solar system and also easier to leave again.
The distinctions between Ganymede and Callisto on this account are less easy to guess at. I have no idea whether people will end up finding Ganymede or Callisto more attractive - whether the ease of reaching and leaving Callisto and its much lower radiation environment than Ganymede will make the former more populous and economically active than the slightly higher gravity, better views of Jupiter, and more geologically active interior of the latter would attract. There could be any number of advantages to a somewhat more active geology, but it's too close to really call. In either case, I would bet heavily that both moons end up with big populations and very powerful economies.
Primary attractions of settling Ganymede: Reasonable surface gravity; inexhaustible water ice for fuel, ship/habitat shielding, industrial purposes, and drinking; views of Jupiter from the planet-facing side; proximity to Callisto, which is also an attractive settlement destination, as well as the primarily scientific targets Europa and Io; and relative accessibility to the Jupiter Trojan asteroids. It also benefits from Jupiter being on what I call the Fusion Line - the farthest destinations where operating without some form of nuclear power is possible (albeit very difficult), and that would become economically irresistible once fusion technology is achieved. The reason is that energy is the only major obstacle to settling Ganymede and Callisto once travel to and from them is practical.
That isn't to belittle the difficulty of either challenge: Very likely it will be centuries before human travel to and from the Jovian system is safer and more economical than epic voyages of exploration pursued at enormous cost and hazard, and the development of fusion power seems to be demonstrating Hofstadter's Law, so I wouldn't care to bet on which will happen first. But once they do happen - and I consider this pretty much inevitable - there is no reason whatsoever that Ganymede would not turn into a major center of human civilization, and several powerful reasons that it would.
Even before fusion though, ice is very good insulation as a building material, and there is a possible source of energy within the Jovian system in the Io flux tube (see the Io page for more details), so there are options. One could theoretically fly ships full of water ice from Ganymede or Callisto into the flux tube, use the energy present there to separate it into hydrogen and oxygen, and then fly the resulting products back to the two moons as fuel. It could even be an automated system that cycles back and forth, with a continuous chain of cargo ships at every point along the route. This is hardly an ideal energy system, and would be vulnerable to disruptions both natural or man-made, but it could be sufficient to allow settlement before the advent of fusion power.
On a more abstract note, it's important to understand that as a world - rather than, say, an asteroid that might be turned into a city, or an artificial station designed as such - Ganymede and other major targets of colonization would have multiple cultures and possibly multiple political states on them. So when we speak of a human future on these bodies, we're not talking about some 1950s American fantasy consisting exclusively of English-speaking white people, or the utopian ideal of a fusion of terrestrial cultures creating a single, uniform new one for a new world. We're talking about new mixtures of the entire Earth, some of which would settle different parts of the same body and have little to do with each other for generations.
So in imagining a future "Ganymedian" culture, there could be many broad defining features, but it could also end up not necessarily being the most politically important level of organization - e.g., "Latin American" vs. "Colombian." And there's really no way to guess at this point how many cultures and political states would occupy any given body, how unified or fractious they would be, how similar or diverse, etc. We can say with confidence that bigger and more economically prosperous ethnic groups from Earth are more likely to form dominant groups, but beyond that it's unknowable. Another confounding factor is that, in my view, a lot of the movement to the Jovian system would be 2nd-wave colonization coming from people born on Mars, the Moon, and other non-Earth locations, so they too would be products of new mixtures and might see themselves as an entirely separate race from the terrestrial cultures that formed their native society.
That fact in itself could potentially be a source of friction and cultural division going forward: Colonists on a given body that come directly from Earth vs. those from other non-Earth worlds, and thus with more direct cultural roots in space settlement - not to mention whatever effect the new environments have on the way people look, when they live in lower gravity, colder environments to conserve energy, lower pressure, higher mutation rates from radiation, etc. People from Earth of any race will look different than people born on Mars or other places, both because of the new genetic mixtures and the effect of environment on their bodies.
Most likely, I would guess, the bulk of emigration to the Jovian system would be 2nd-wave from Mars and asteroids, so whatever cultures and physical appearances are born out there would be remixed yet again on Ganymede and Callisto. People from Earth would probably come later, once the really hard part was done and these worlds were cruising economically, since by and large migrations from highly developed places to frontiers only happen for two reasons: Being forced to leave home by a natural or political/economic catastrophe, or because expectations have become equalized enough that people just naturally start diffusing into the new environment.
This is the basic reason why it was Spain, France, and Britain, rather than China that colonized the New World continents, despite China's demonstrated capacity to conduct large-scale exploration - the European kingdoms were just plain younger, less culturally evolved civilizations that could transition more easily to the anarchy of frontier life. Whereas an average person in 16th century China might have looked at the New World (if they ever heard of it at all) and thought, "Why the hell would I want to go there? It's dirty, dangerous, and full of people who are nothing like me," a lot of Europeans appear to have reached totally different conclusions from the same data: "It's dirty, dangerous, and full of people who are nothing like me. Sounds like a situation I can exploit!" Of course, since there is no native intelligence on Ganymede, there are no ethical considerations to colonization - just an environment and the possibility of benefiting from it.
It was mainly economic imperative that caused the first waves of Chinese immigration to the Americas, when the underclass who were literally starving in their homelands had an alternative option for the first time in millennia. And now that the US is pretty well developed, East Asian immigration is booming again even though the economy in Asia is also booming - it's just natural diffusion from a high-population area to a lower-population area rather than an economic imperative, at least in most cases. We can apply the same principle to imagining future migrations into (or between) other worlds than Earth, with Earth playing the role of China - culturally developed, dense, ancient, and slow to take interest in new places compared to societies just a few generations removed from colonization themselves (analogous to the United States).
So just for the fun of it, we can make the specific prediction that most of the people who end up colonizing Ganymede and Callisto will be Martians. Or, an even more interesting possibility - maybe the Martians would colonize Callisto, and then the Callistoans would colonize Ganymede, making it a 3rd-wave colony rather than 2nd. Maybe both: Colonization within the Jovian system could happen to a major extent once there is a thriving civilization anywhere in it, so you could have people born on Callisto settle Ganymede from a raw frontier into a more manageable place; people from Mars would then swoop in and turn it from manageable to idyllic; and then people from Earth would just diffuse into it the more civilized it becomes. Any of these scenarios is plausible, as far as I can see.
As to timing, we are most likely talking about the 2nd and 3rd quarters of the current millennium (23rd to 28th centuries): The current quarter would be occupied with exploring, settling, and developing Mars, the Moon, and some of the more attractive nearby asteroids, although there would be human exploratory missions to the Jovian system beginning at some point in the 22nd century - perhaps even late in the 21st. If you're wondering where I'm getting this unit of time - quarter of a millennium - it was the unit most applicable to settlement of the New World: The first quarter millennium of expansion into the New World was concerned with exploration, establishment, and raw survival; the second quarter saw the coalescence of functioning states, independent cultural identities, and thriving economies.
The timeline of colonization is driven by the social and physical limits of human procreation in the context of available resources, and I would say the existence of modern science and technology more than compensates for the much larger scale of global as opposed to continental settlement. So, in simpler terms, I think it's reasonable to assume that the timeline of New World colonization would be applicable to solar system expansion, with a quarter-millennium being the standard unit of time for each step along the way. However, "each step" does not necessarily mean each planet - by the time it becomes practical to live on moons of Saturn, pretty much everything else (short of interstellar) will be too, so the step after Jupiter would see concurrent expansion into the rest of the solar system. So the progression goes like this, with each step taking about 250 years to produce the next:
1. Moon/Mars/Near-Earth Asteroids (NEAs)
2. Main Belt asteroids, Jupiter Trojans, and Jovian moons.
3. Everywhere else in the solar system.
Progress on each would naturally continue even while the next step has begun, so they overlap to some extent, with Moon/Mars/NEA settlements turning into advanced civilizations over the same time period that they begin to first colonize the Main Belt, Jupiter Trojans, and Jovian moons. Then as the 3rd wave colonies become advanced and start striking out even further to the rest of the solar system, the 2nd wave starts to quiet down and become boring and self-involved. Rinse, repeat. And once a given wave is a few steps behind the furthest frontier, it just joins Earth as part of the stodgy Old World that doesn't notice or really care what's going on Out There.
One bizarre possibility is that Ganymede could be terraformed. If you were to pump enough heat into the ground and/or have an ongoing program of sub-surface thermonuclear detonations, you would be able to slowly build up a water vapor atmosphere capable of holding on to the heat. Once the temperature and pressure rose high enough, the result would be an ocean world. It would never be self-sustaining over very long periods of time, like hundreds of thousands of years - the gravity is too low, the Sun far too weak, and the radiation would slowly strip away the hydrogen from vapor in the upper atmosphere. But on human timescales it would be a piece of cake (relatively speaking) to keep the atmosphere warm and pressurized, particularly for a notional civilization with fusion power.
But as noted in previous discussions of terraforming vacuum worlds (as in the Luna diary), the process would be messy and inconvenient for the infrastructure already developed for a frigid moon in vacuum, so there are plenty of reasons colonists might choose to keep it frozen and airless. Terraformation would also be aesthetically ruinous, completely eradicating all surface features - in fact, all surfaces period - and destroy its image from space, turning it into a featureless white ball of uniform cloud. Not to mention that neither Jupiter, nor other moons, nor the stars would ever be visible again from the surface. So while it's fanciful and exciting to imagine great ship-cities on an ocean world, this is not likely an option the future inhabitants of Ganymede would exercise.
However, keeping it frozen in the presence of a major human population would present its own challenges. In particular, cities would have to build massive radiative towers into the sky to dispose of waste heat into space rather than allowing it to be absorbed by the ice. This is a problem that would be common to all colonized icy bodies, so we could expect such radiative towers to become a typical feature of outer solar system civilization. Ganymede, in addition to Callisto, is bound to play a pioneering role in acquainting humans with the unique challenges of managing a global ice shell environment.
VII. Future of Ganymede
The vast quantities of water on Ganymede suggest that H2O would be its principal economic commodity, which implies a distant future - and I do mean distant, like tens of thousands of years - where a solar system teeming with vast, thirsty human activity has drained it dry. It is an unimaginable volume of water - more than all the Earth's oceans combined - but the potential growth rate of human activity in the solar system is also almost unimaginable, particularly due to the main technological applications of hydrogen and oxygen. So one possible future of Ganymede is its reduction to a much smaller rocky moon, which itself might then be mined to extinction by whatever gigantic applications demand it.
Given how humans currently live, this might sound far-fetched, but when you add ubiquitous fusion power into the scenario you make it conceivable that a single future city somewhere could consume more in both energy and material resources than the entire human species on Earth does today. Such a pattern holds true in reverse, when you compare the consumption of a modern city to the whole species 20,000 years ago, so given a vastly expanded resource base it makes sense that the progression could continue. Especially when you take into consideration the enormous fuel/energy requirements of a routinely spacefaring civilization in order to move from planet to planet and moon to moon.
However, Ganymede doesn't have to be drilled and strip mined to extinction. There are plenty of social, political, and other reasons that a future civilization would choose to preserve it at some point, in which case its fate would largely follow the natural course of the Sun's evolution. As the Sun expands, temperatures in the Jovian system increase substantially, and for a time it would become the featureless white ball of cloud above a global ocean I mentioned earlier. However, that would eventually blow away along with Jupiter's atmosphere in the waves of mass being ejected by the Sun, leaving it a bare rock anyway. From there, depending on exactly how things shape up, it would either lose orbital momentum from moving through the gas escaping from Jupiter and crash into its parent planet, or else Jupiter would lose mass quickly enough that its gravitational hold on Ganymede would fade.
In the latter case, it would spiral outward and escape Jupiter into an elliptical orbit around the Sun. However, as the Sun itself lost mass, it would both move farther away due to the shallower gravitational force and because the continuing solar mass ejections would push it away. Eventually it would re-freeze, accreting both water and organic ices on to its surface, and join the vast profusion of icy bodies wandering the darkness between the stars.
VIII. Catalog of Exploration
1. Past & current probes:
Pioneer 10 (USA - 1973 flyby)
Pioneer 11 (USA - 1974 flyby)
Voyager 1 (USA - 1979 flyby)
Voyager 2 (USA - 1979 flyby)
Galileo (USA - 1995 to 2003, Jupiter orbiter / Galilean moon flybys)
Cassini-Huygens (USA and Europe - 2000 flyby)
New Horizons (USA - 2007 flyby)
2. Future probes:
Juno (USA - en route, scheduled Jupiter orbiter to reach system in 2016, will flyby Galilean moons)
Jupiter Icy Moon Explorer (JUICE) (Europe - 2022 launch, 2030 enter Jovian system, all Galilean moon flybys then Ganymede orbit)