Floating Cities as a Response to Climate Change

This image is a vision of the concept by the architect Vincent Callebaut.

My subject here is perhaps the most dramatic consequence of global warming -- the rise in sea levels -- and what we should do about them.

In the worst case this rise would reach 200+', erasing a considerable fraction of the planet's surface, submerging many of the world's largest population centers, forcing hundreds of millions of people to migrate, and wiping out all the ecologies associated with that vanished surface.

This scenario is quite uncontroversial as a possibility. The volume of water locked up in the polar ice sheets can be observed and measured directly, from satellites. Even if we couldn't do that, a total meltdown has already happened before, at least once, in the Cretaceous, about 100 million years ago, and we can see where the sea levels were then just by looking at where marine fossils show up now. So our sense of what will happen if everything melts is pretty solid (unlike almost everything else about this issue).

This feature gains even more interest when you consider that there is very little in the literature to discourage us from thinking that at some point we will have to face it. There is a consensus that warming drives melting and that CO2 accumulation drives warming (not a universal opinion in either case, but it is clear where majority opinion lies). The virtually unbroken record of failure at controlling emissions to date suggests that it would be a huge achievement just to reduce the annual increase (which would do no more than push the date and therefore the consequences of any given concentration off for a few years anyway). Terminating emissions altogether, let alone pulling a serious amount of CO2 out of the atmosphere and storing it somewhere, seems totally out of reach. Of course any of these good things might happen -- they aren't physically impossible -- but given the history of the species it doesn't seem to be imprudent to think about what we will need to do if they do not.

Plus it might not matter anyway. The literature is full of "positive feedback cycles," processes that, once triggered, will generate melting independently of marginal changes to CO2 concentrations. One such cycle is the relation of melting to the Earth's albedo: melting converts whiter area, on either land or water, to darker area, which then absorbs more heat, generating more warming, melting more ice, etc. A second is the release of methane from the thawing of the permafrost (methane has 25 times the greenhouse effect of CO2 per unit weight). The more warming, the more thawing, the more thawing, the larger the release; the larger the release, the more warming. (Some sceptics think that a negative feedback cycle might be found in cloud formation -- the more warming, the more evaporation; the more evaporation, the more cloud cover, the more cover, the more sunlight bounces off without reaching the surface, and therefore the less warming. So far they have not won many converts.)

PFCs pop up in ice dynamics as well. One theory is that each chunk of ice that calves off increases the stresses experienced by the remaining, "upstream," ice. When the stress regime passes a certain value, the nature of which will depend on local conditions, upstream ice will start to break off and slide into the sea spontaneously. A second idea is that ice sheets tend to compact the earth under them such that the sheets end up sitting in a depression. When ocean levels reach the outer lip of this "pothole" they pour down into it, accelerating melting from below. (http://tinyurl.com/ya2sy93) Add all these positive feedback cycles to the odds that CO2 concentrations might double or triple, and the prospects of a sea level rise in the range of several dozens of feet seem good enough to be worth thinking about.

Our thoughts on that topic are shaped by two basic constraints. The first is that the number of people likely to be affected by a rise of 10 or 20 feet, let alone 200, is just enormous. Over the last hundred years the planet has experienced a huge migration to coastal cities on all its continents, and coastal cities tend to sit in lowlands, at altitudes of just a few feet. Today fourteen of the world's seventeen "megacities" (= a population of > ten million) are coastal, as are a substantial fraction of the world's major cities (populations > one million but < ten). A significant sea rise therefore threatens the accommodations of at least several hundred million people. Any solution we think of will need to scale to numbers like these.

The second constraint is that we can not know with any confidence the details of how this rise will occur: the rate of the increase, the year-by-year variations in that rate, the frequency and power of the storms that are the first expressions of that rise, or the point at which the rise finally plateaus out. In a perfect world we would be able to count with confidence on a small, steady increase over a long period, something like one centimeter of rise a year for several thousand years. Even here the politics of relocating such enormous populations would be difficult -- in many cases these migrations would have to be across national borders, since many of the most populous nations in the world would vanish almost entirely -- but given enough time one can just barely glimpse of how it might be brought off with fairly conventional policy responses.

But alas our situation is not so accommodating. Sea level rise is controlled largely by ice sheet melting and that is controlled in turn by the relationship of ice sheets to their underlying bedrock. This interface is very hard to instrument, and even if it weren't, ice cracking dynamics are basically chaotic, essentially so much solid weather, which means that any predictions we make will come with very wide margins of error. So we need a policy response that not only scales across many orders of magnitude in population size but over many degrees and kinds of uncertainties.

Only one policy response seems to meet these constraints: floating cities. Unfortunately floating cities have a terrible giggle problem -- they sound like something out of a movie with ray guns and flying saucers -- but it cannot be denied that the basic idea of making new land sounds on its face like an appropriate response to a problem caused by the destruction of old land. Plus they do pass the scaling standard. They scale over population size because there is a huge amount of space (more all the time!) in the oceans. They scale over uncertainty in two different ways. First, residents of a floating city don't care what the sea does: whatever the rise is, whatever the rate of rise is, life will flow placidly onward. Residents will live decoupled from all the uncertainties around sea rise -- degrees, timing, variation -- and they will know that.

Second, floating cities can be built really cheaply, which means they scale over another species of uncertainty, which is the timing of demand. It is possible that at some point we will need to find alternate accommodations for millions of people -- people who in many cases will have very few resources of their own -- in a short period of time. Floating cities speak to this fiscal constraint in several ways. First, they can be built top-down, all at once, to very high degrees of standardization, with all the details built in at the point of manufacture. Very large production volumes, necessary to give the learning curves needed to drive costs low, are an assumption of the exercise.

Finally they can be built out of very cheap materials. The cheapest material in this society is plastic film, and we could use that, generating rigid structures sitting on huge floating mats of industrial strength bubble wrap, made rigid with plastic ribs. (I know; there's that giggle factor again.) Think of it as 3D printing on the largest scale. In the ideal case, assuming a mature technology, the factory would itself be out in the ocean, farming and harvesting huge mats of genetically engineered algae that would generate all the raw material needed through their own photosynthesis (and water and carbon dioxide). In this scenario all the basic resources: surface area, energy, carbon (from CO2), and water would be not just cheap but free, and while the costs of acquiring those resources might start high, there is nothing I can think of that prevents them from dropping arbitrarily close to zero. Almost everything in our society that is cheap today, like candy bars, is cheap because its manufacture is dominated by these three factors: large production volumes, thoroughgoing standardization and therefore process automation, and cheap resources. The same principles would work with floating cities.

Of course low costs have advantages beyond allowing quick and massive crisis response. Human beings are not the only creatures threatened by sea level rise: coastal ecologies almost everywhere will be flooded out. Not all will be able to retreat inland, both because the climate will change and because higher population densities will increase the competition for surface area from buildings and roads. Assuming the right economics, floating city technology might allow us to preserve these ecologies by building floating nature preserves and locating them at the appropriate latitudes. And of course the cheaper this real estate is, the more of it can be donated to populations unable to buy their own way onto these platforms.

A last point to make about floating cities is that they will always have plenty of energy, even assuming a self-imposed constraint of depending only on renewables. They will be able to get power from wind, from waves, from temperature gradients, from the algae (biodiesel), and of course the sun. If they can rotate, they can even use concentrated solar, driving cell efficiencies to 40% and 50%, while locating themselves, if they choose, where sunlight is most powerful and continuous. Renewable energy farms on land are ultimately restricted by land costs; power plants floating in the middle of the ocean have no such limitation. Whatever problems can be solved with lots of energy -- like wave suppression -- will be problems floating cities will not have.

How would all this start?? If and when shelter on floating cities gets cheap, in the ballpark of a few bucks/sf, many people will need no further incentive than that. However, in the beginning, costs will not be low; buyers will need to be found who are willing to pay a premium. Who might these be? Perhaps NOAA, or its equivalent in some other country, would turn itself, in part, into a marine NASA, launching residential colonies that would grow over time from hundreds of residents to thousands on up. Perhaps. However, it is obvious that this line will not be crossed casually. Our whole culture, certainly our political culture, is based on the assumption of fixed geographical referents. What we mean by Boston or Massachusetts or the United States has a lot to do with where they sit on the planet's surface. The idea of living on a city or city- state that could sail anywhere it wanted in the globe is going to be conceptually disruptive. Living in a plastic city will take some getting used to as well. (Though of course residents will start adapting and renovating as soon as they move in.)

But all this might change as countries everywhere get hit by one record-breaking storm after another, as it becomes clear that the huge investment in beachfront property is not threatened but doomed, and as less radical solutions like population transfer are found to be either ineffective or so politically refractory as to be impractical. In those circumstances there may well be a strong political return in being able to show people that governments are in control, that they are not leaving the fates of their constituent populations to the whims of the weather, that if need be there is a certain, stable, effective, if novel, solution at hand. Governments might start thinking that being able to create taxable real estate has some real upsides, given that the real estate from which they had been accustomed to collecting the highest revenues is losing more and more value. In theory there is no reason why only coastal cities should share in this windfall. Inland cities -- Akron, Los Vegas, Dallas -- might generate clones of themselves out in the ocean. Is this too weird to imagine?

The first migrants might be people who love the coastal life, love living a stroll from the ocean, love the island light, like having a sailboat moored within walking distance, and, perhaps because they have had some bad experiences, like knowing their houses are not going to be flooded or washed away. Another, perhaps larger population might be strong environmentalists, people resonating to the argument that floating cities are the only place on the planet where humans can live without interfering with the lives of other species (because the lack of marine iron, an essential nutrient to most forms of life, makes the deep oceans close to deserts). Floating cities, with their huge algae farms, would be CO2 sinks. Residents on them would not have just a small carbon footprint, they would have a negative footprint; they would be enjoying the least intrusive way possible of living on the planet. (Some might use this logic to argue for a migration to floating cities whatever the story is with sea levels.)

In any event, as of 2011 this scenario is full of unknowns. The proposition I am advancing is that a modest research program to assess the engineering and economic issues involved would not be out of place. I hope you agree.