Solving the Power Problem and the Climate Problem in one go

A massive project funded by Google exposes the vast untapped geothermal reserves available to the United States’ power infrastructure.

How much energy? you ask. Well, the researchers based their estimates on what current technology is able to extract – not any hypothetical future advances. Even so, it turns out that there is three million megawatts of potential geothermal energy below the surface of the United States. That’s ten times the energy of every coal plant in the United States online today.

The actual report is here. It staggers me that these figures are using current tech, not future tech. That’s done by drilling deep holes — which we can already do quite well, to reach buried oil — then piping water down one hole so we get steam back up another. Steam that powers a turbine, that generates enough electricity for practically free that all we have to do is keep the thing running and free of mineral deposits from then on.

Whenever someone says we have to stop burning fossil fuels (coal, natural gas, gasoline in vehicles, etc) and start using renewable energy, the only options ever proffered are a) wind/solar, b) nuclear, and c) biodiesel. I mention geothermal in pretty much every single such conversation, and nobody bothers pursuing that line of argumentation — the greens-vs-dirtys always talk about nuclear power as being the only realistic solution to the problem, where biodiesel continues pumping CO2 into the atmosphere, and wind and solar are grossly inefficient in current implementations.

Space solar might become a reality if we can figure out how to keep it from getting shot through by every passing bit of space debris we’ve already put into orbit, but it’s not nearly usable in an infrastructure capacity today without massive investment. Which the oil industry has proven unwilling to do, by the by, as long as they’re getting huge subsidies from the government despite record profits.

And nuclear power has its share of issues — we humans have a tendency toward sloppiness with our engineering endeavours, and so accidents like Fukushima happen when we build nuclear power plants without regard for the potential destruction of main power by a natural disaster. The fact that the core was allowed to melt down because they had no way to generate power to pump water to cool the rods, suggests to me that they had no adequate and proper contingency plan for how to deal with a tsunami knocking out their only means of cooling the radioactive materials. The result, more radiation than originally estimated. More than twice the radiation originally estimated.

We could learn from these mistakes and build better, more resilient, more disaster-proof nuclear power plants. But nuclear materials are inherently dangerous, and we are really harnessing the lightning by using them. They also take a long time to build, so if we started one today, it would go on line in ten or so years. The US and China are practically competing to see who can build the most coal plants the fastest, and contributing to the global warming that will inevitably result in a planetary mass extinction event, not to mention the suffering it will cause humans themselves.

And all the while, the solution to our power problem and our climate change problem is right beneath our feet.

I’m not saying that there will not be some potential far-off hazard that necessitates that we switch off of geothermal power — say, for instance, bleeding off the core’s heat slowly, causing a permanent cooling of the inner parts of the planetary core. Call that “global cooling”, I guess. Considering eighty percent of the core’s heat is coming from radioactive materials generating that heat, and the core has stayed as hot as it is for 4.5 billion years so far, I’d say we’re safe for a good long while yet. Certainly long enough to be able to get off this rock and colonize another some time in the distant future, if such a feat is possible, and if we can avoid committing suicide until then. Hilariously, that means geothermal power is technically nuclear power, so the nuclear power advocates were sort of right all along! You know, once we look past the fact that we’re trying to bottle that lightning in our own back yards and are shocked (heh) when it destroys everything around us.

But we humans are terrible at long-term planning, aren’t we? Both with regard to engineering as evidenced with our nuclear power problems, and with regard to the long term survival prospects of our species as a whole.

Solving the Power Problem and the Climate Problem in one go

32 thoughts on “Solving the Power Problem and the Climate Problem in one go

  1. 2

    No no, I meant we’re good at digging really deep holes. We do it now for reaching oil that’s, say, a dozen kilometers below the surface, and the geothermal survey’s only looking at how hot it gets 6.5km down.

    So you dig two holes parallel to each other, then connect them together at the bottom of the hole. Put water down one, steam comes up the other. It powers the turbine, then the condensation can be recycled right back down the first hole.

  2. 3

    The practical way to connect the two holes at their bottom ends is, of course, hydraulic fracturing. That’s the practice that feeds me and the company I work for, and, here lately, is a huge political bugaboo in some of the places (Colorado?) where there is not only oil/gas but also shallow geothermal heat.

    This has been done in the past, but it’s nice to see a possible revival. Recent frac technology will make it work much better than it could have 30 years ago.

  3. 4

    But we humans are terrible at long-term planning, aren’t we?

    They’re probably the exception rather than the norm, but there are many examples where we as humans have shown our capacity to plan and implement massive, long-term public engineering projects.

    Here in the UK, the classic example is the London sewerage system built by the Victorians (esp. Joseph Bazalgette). Not only was it a huge undertaking functionally (which to a shockingly large degree is unchanged and still functional more than 100 years later), they made buildings which essentially pumped human waste look like this.

    The Dutch can likewise point to all their reclaimed land. USAians and Canadians will have their own examples (interstate highways? Panama canal?). Even in the ancient world, the Romans built incredible roads across the known world and massive aquaducts carrying clean water to their parched capital city, and the Chinese built a wall thousands of miles long*.

    Most, if not all, of those examples are the result of a powerful individual or a small group having a vision and the ability to impose that vision on everybody else. Since in democratic countries we (presumably?) don’t want to return to concentrating such power in individuals or small groups, the challenge is to ensure a large enough body of the population has the education, knowledge and understanding to share the vision.

    * OK, that one ultimately wasn’t that useful (at least for for its intended purpose).

  4. 5

    I’m not sure I’d quite knock renewables too much (see, for example, here).

    All the same, this is welcome news. And it can be implemented at all levels, from local on up (my wife & I are watching clips of the Rick Mercer Report and we saw the one where he visited Jack Layton’s house not too long ago – they had a closet space where they were going to install a geothermal system).

  5. 6

    Not buying the numbers there; they literally do not add up.

    Multiple sources give the total geothermal heat flux as being on the order of 45TW, worldwide. Obviously, only a proportion of this passes through the USA, which represents ~2% of the earth’s surface (or 6.7% of its land area). It is not therefore reasonable to suppose that the USA could recover 3TW of energy, given that only a moderate percentage of the total heat flux could ever be converted to useful work.

  6. cmv

    The numbers are hard to follow, as they give production, but no scale to the production. It turns out that the numbers given (taking the 10x coal production figure at face value) are per hour. 3TW/h. The 45TW flux you mentioned is actually annual (I had to follow the cited source on Wikipedia to find that.) The reason that this isn’t actually a problem for geothermal energy production is that that “flux” is the total annual heat loss from the system, not the total amount of heat energy in the system.

  7. 9

    Yeah, 45TW lost at the surface per annum, but regenerated from the core at a rate of 30TW. The total heat content of the planet is evidently 1031 joules, according to Wiki. I should think that’s TONS.

    As for dismissing wind and solar, well, I suppose I’m dismissing them as singular magic bullets. They’re not. But in tandem with other solutions, they can sure as hell make a huge dent in our current crisis and give us enough time to get our acts together and build as many geothermal plants as humanly imaginable.

  8. cmv

    One of the cooler (or warmer) things about geothermal is that there doesn’t actually have to be any conversion to electricity to get a benefit. The most common usage right now is for heating and cooling, which uses the heat energy directly. The cost of a geothermal HVAC system for a house is currently just a bit more than a standard gas or electric furnace, and runs on just a bit of power for the pumps and fans once it’s up and running. Anyone who needs a new furnace should look at converting.

  9. 11

    How many basic physics fails in the responses to my comments?

    First, 45TW is 45TW, it’s already a rate (terajoules per second), there is no “TW/hr” or “annual” or other such nonsense.

    Second, this is a total energy figure, it doesn’t vary by depth until you get deep enough that you’re below a lot of the radioactive decay (and thus below some of the heat production, reducing the available energy), which will be well into the mantle if not the core. The deepest wells we can dig are just surface scratches by comparison.

    The efficiency with which you can extract work from the energy flow varies by depth because the temperature difference increases, and as we all should know, the maximum proportion of energy we can extract from a heat difference is determined by the hot and cold temperatures. Geothermal plants can extract typically 10% to 30% of the heat energy that is brought to the surface by the hot working fluid, depending on temperature.

    The rate of heat production from radioactive decay is irrelevant except to calculate how much the Earth’s core will cool over time; geothermal energy production won’t change that rate detectably. It turns out that the geothermal energy flow will not decrease enough in the next hundred million or so years to matter much.

  10. 12

    Oh, and the 10^31 joules figure is also irrelevant because essentially all of that energy is much too deep to get at; mostly in the core.

    The 45TW figure is the rate at which that energy is currently flowing up through the mantle into the crust and up to the surface. We can drill a few km down, but that’s not going to have any measurable impact on the insulating effect of nearly 3000km of mantle.

  11. 13

    To put our argument another way — you’re suggesting that it’s terribly cold on top of the blankets, so it can’t possibly be warm enough to stay cozy under the covers curled up next to your wife.

  12. 14

    You’re really not having much luck with this physics stuff.

    Imagine we draw a sphere at some depth a bit below the deepest bore we’re likely to drill. The rate of flow of energy across that surface is still 45TW, same as it is at the surface or at twice or three times that depth. (This is just conservation of energy; the only way there could be more energy flow at deeper levels would be if the energy was going somewhere else between that depth and the surface, which it clearly isn’t.)

    Now, you can certainly extract energy from a deep bore at a faster rate than it is being replenished from below – but the problem with this is rather obvious; the volume of rock around the end of the bore will cool down, reducing efficiency. It will NOT significantly increase the rate of heat flow from below, because a few km of borehole is nothing compared to the nearly 3000km of mantle underlying it. (If you have a pile of 300 blankets, then removing the top one isn’t going to make much difference to the rate of heat flow.)

    You seem remarkably confused about basic concepts like heat, energy and temperature.

  13. cmv

    @Andrew -as I read it, the rate quotes as 45TW looked to be an amount of loss, in Tera Watts, not Joules. Look here for a conversion. It is a very different number.
    Since we are talking about the amount of power which can be generated, the unit of measure is watts per hour.

  14. 16

    Another thing that’s important to note, from that paper I linked you, is the fact that the energy output by the core at 860 EJ/yr is double the entire human race’s current energy consumption rate. Even if we can only harness at most 20-30% of that with current tech, that’s a hell of a lot of power that we don’t have to burn fossil fuels to get.

  15. 17

    I shouldn’t have added the “per annum”, having taken the cue from cmv. But what I’ve said is accurate. The planet is losing heat at the surface at a rate of 45TW. That does not describe the total heat in the system, nor the total heat at 6.5km deep, only the rate of loss under “normal” conditions. Tapping into the heat at increased depth will increase the amount of heat lost by the planet, because we’ll basically be converting that heat into work we can harness by making it turn water into steam. By doing this, we’ll be sapping a little bit more of that heat than would dissipate naturally at the surface.

    The deepest hole we’ve ever drilled was 12km – the Kola Superdeep Borehole. This is far deeper than the 6.5km necessary to tap the available heat shown in that map. We can drill 6.5km relatively easily. And at those depths, Google’s numbers are probably quite accurate. So what part of it are you actually trying to refute?

    And while, yes, we couldn’t possibly tap all that power, or we’d be operating at 100% efficiency in cooling the planet’s magma layer and would probably speed this “global cooling” suggested in the original post along at an alarming rate. But we can certainly supplant a hell of a lot of the existing coal plants at 1/8th the cost for CO2 emissions. The US has 77 geothermal plants right now, generating 3,086 MW. Renewables only make up 8% of the power consumption, and I don’t have on hand what percentage geothermal is of that 8%, but I can assure you it’s relatively low. Certainly geothermal should be used far more than it already is. And surely you’d agree with that.

  16. 18

    And you seem remarkably confused about the viability of geothermal energy. Read this paper. The heatsink effect can be compensated for by regulating how fast the heat is extracted from the depths, meaning a power plant can last 250 years well over 30 years with a mean production of 8MW, 250 MWe-years before having to stop production until the heat returns to useable levels as long as you’re not extracting that heat at exorbitant rates. These power plants are still useable. Mostly because one does not need to drill through 3000km of mantle (how the hell would you manage such a thing?) to get deep enough to actually be able to extract enough heat to turn water into steam. And since we’re talking about something that turns to steam at 100°C, we don’t need to get very deep at all according to these charts. You’d want to go a bit higher so it stays steam on the way back up, but what you’re doing in this case is effectively bleeding the mantle of its heat. And that paper I linked shows exactly how much power you can extract without having to stop operations after a time.

    I already conceded that you can’t get 100% efficiency (nor would I ever have argued that that’s the case), but I still fail to understand what you’re actually arguing about, because those numbers show how hot it is at 6.5km deep. That means we can achieve a great deal of efficiency or use very shallow bores in certain specific spots where the temperature is higher.

    Please just explain to me what I’m actually “wrong” or “confused” about, and stop being such a complete and utter asshat. Yes, I’m a layman. I never said otherwise. And I actually understood a good deal of both what you said and what was in that paper before you condescended to shout at me about what a failure I am at physics.

  17. 19

    @cmv: please, please, learn some basic physics.

    “watts per hour” isn’t a unit that makes the slightest sense in this context (it would be a rate of change of power over time). You may be confusing it with “watt-hours” (watts times hours, not watts per hour), which is a quantity of energy (1 watt is one joule per sec, so 1 watt-hour is 3600 joules). Power station capacity is often given in watt-hours per year, which is just watts multiplied by the number of hours in a year. (These units based on watt-hours are used because energy costs are invariably charged to the user in multiples of watt-hours rather than joules.)

    The US total coal-fired capacity is about 340 GW (= 2980 TWhr/year). The article says “10 times” that, and says “3 million megawatts”, so we’re definitely talking about 3 TW, not some figure in Whr/yr or other unit.

    The 45 TW figure is the total rate of heat energy flowing from the interior of the earth to the surface. Note that this is a tiny figure compared to the amount of solar energy absorbed at the surface, which is about 89 PW (i.e. 2000 times larger).

    @Jason: I’m not the slightest bit confused about the viability of geothermal energy. What I’m pointing out is that the US can not sustainably produce 3 TW of usable electric power from geothermal sources as is claimed. If you did in fact build out a nationwide borehole network and actually produce 3 TW of electricity from it (we’ll assume at 20% efficiency, which is probably on the high end), then over time you would end up cooling the areas around the ends of your boreholes, reducing production capacity, which would stabilize at an absolute maximum of approximately 200 GW reflecting the proportion of the global heat flow passing through the US.

    Now, it might be that it would take decades or centuries before production actually dropped that far; I don’t have an estimate for how long it would take. But the long-term conclusion is inescapable: the US can not practically replace its coal-fired capacity with long-term-sustainable geothermal generation, even by covering the entire country with boreholes. Therefore geothermal is confined to the following roles:

    1) Local usage in areas where it happens to be practical, to supplement other sources of power

    2) Use for heat production rather than electricity (avoids the efficiency issue)

    3) Interim production at ultimately unsustainable rates (but which might nonetheless last for long enough to be useful, i.e. many decades, and is probably cleaner than coal – but remember that fracking will usually be involved).

    I’ll try and explain the basic physics again:

    However deep your borehole, imagine a layer of rock well beneath it, and consider that there is an upwelling of heat in that rock at a flux of about 90 milliwatts per square meter. That heat is all there is to replenish the heat that you’re taking to the surface via the borehole. Obviously, the area around the bore, above, below, and sideways, will also cool down as a result of contributing some of the heat extracted, but all of that heat ultimately came from the slow upwelling from below, and can only be replaced by it.

    Now, you may think “but won’t the heat from below come up faster because of the cooler rock around the borehole”, and the answer to that is yes it will, but only by a tiny fraction, since that flow of heat is determined by the core temperature, the surface temperature, and the thermal conductivity of those 3000 km of mantle and crust in between. The borehole has the effect of increasing the conductivity of the top few km of that, which clearly isn’t enough to make much difference to the rate of flow.

    Now, the US has a land area of about 10 million km^2, which at 90 milliwatts/m^2 is 0.9 TW total; call it 1 TW as a round number (the heat flux isn’t entirely even, so it could be rather more or less, but not hugely so). So that 1 TW, times the efficiency of power extraction, is an absolute hard limit on the amount of geothermal energy that can be extracted without ongoing cooling of the upper layers of the crust (which clearly isn’t sustainable in the long term).

    Note that going deeper (within reasonable limits) doesn’t increase the sustainable level of energy, since whatever the depth, that heat flux coming up from the core is the only thing ultimately replenishing whatever heat you take up the borehole. (It does increase the efficiency with which you can convert the heat to electricity or other work, of course, since the temperature is higher.)

    So you can see that generating 3 TW of electricity from geothermal (which means about 15 TW of heat energy, assuming 20% efficiency) is vastly higher (>15x) than the sustainable level and will therefore end up cooling the areas of the crust used (which in turn reduces energy production rate).

  18. 20

    Jason @18: Yes, 45 TW is about double the current world energy usage, but consider this: if you created a borehole network over the entire land area and ocean floor (clearly impractical) and got an overall 20% efficiency in converting the entire heat flux, that’d get you 9 TW of power, or a bit over a third of the demand.

    Contrast that with solar power, which has flux densities on the order of 2000 times the geothermal heat flux. With reasonably conservative estimates on power production, you could generate the entire world energy demand using an area 400km square, or about 2/3rds of the state of Arizona. (If you used a chunk of the Sahara instead, you’d need even less space.)

    Which do you think is easier?

  19. cmv

    @Andrew – You’re right, I did get mixed up with how to calculate Wh of usage, as opposed to what the generation. Thank you for clarifying.
    The difference in the numbers is still (possibly)explicable in the paper that came up with the 45TW of heat flux – it mentions that the heat loss is not even across the globe, but significantly higher on land. I can’t look it up again this morning, but that might get closer to the available energy for the continental US.

  20. 22

    I would be thrilled if there was a peer-reviewed write-up of this work and not just a press release, web site, and Google Earth app. Instant suspicion. I have to land on Andrew’s side on this.

    What they are talking about is the total thermal energy currently present in the earth’s crust under the US (that can be reached by current tech), which is much more than the energy flux.

    In theory you could keep pulling out more energy than the flux until the thermal energy stores are used up (at which point you’d be flux limited). But I imagine this would also destroy the thermal heat gradient which would diminish geothermal energy’s effectiveness, especially when it comes to rate of energy production.

    I’m skeptical. I’d like to believe this will be useful, but I don’t think you will be able to get a significant amount of that stored energy out of the ground on a large scale. Geothermal just isn’t enough energy to provide our needs.

  21. 23

    You know, Iceland has been running geothermal plants for quite some time. Geothermal and hydro provide 80% of Iceland’s electricity requirements.

    They don’t seem to have a problem with localized cooling or rate flow issues.

    If it works, then it works.

  22. 24

    Iceland is one of those places where geothermal is locally favourable. Sitting as it does on the Atlantic rift, and with its extensive volcanism and hydrothermal activity providing natural heat conduits to the surface or nearly so, it is likely that the country has a much larger total heat flux than the worldwide average (though I don’t have figures).

    Iceland also uses a lot of its geothermal energy for direct heating rather than power generation; this is a much more efficient use, especially in a cold climate.

  23. 25

    I’m still trying to wrap my head around Andrew’s actual objections — are they directed solely at the numbers produced by the study, saying that what geothermal energy is underground is not tappable appreciably and thus can’t supplant needs, or just that it can’t supplement our needs?

    I’m well aware there’s more than a bit of hyperbole in saying “look how much energy is available to us, therefore in a perfect world we could extract all of it”. Of course we couldn’t. And even if we could, sustainability demands that we couldn’t extract more than the amount of energy put out by the radioactive elements.

    How I figure things, having plants be put into production in cycles might be the best way to do it, where some plants are given time to allow the heat to accumulate once again. Given Figure 3 in the paper I linked earlier, it looks as though it does not take very long at all for the energy efficiency to reach 95% of what it was originally — run a plant for 30 years, rest it for 10, and you should be pretty well back in business. Rinse, lather and repeat. Four plants covering four different areas could use that cycle to run any three plants at a time, and thus provide constant power, though certainly not nearly as much as the numbers given by this study as “available”.

    But likewise, the amount of energy hitting us from the sun is mostly untappable, so your 89PW figure suffers the exact same problem as the figures given in that study. Current photovoltaic tech is either grossly inefficient, or where they can achieve 40% efficiency, are extremely expensive to produce at the moment. I’m not saying solar isn’t viable, by any stretch of the imagination. Ideally, I’d love to see a) space solar, beaming the power down to Earth in diffuse microwave beams, and b) some sort of tech that would allow us to tap all that sunlight that’s hitting our streets and roads and roofing with minimal wear and tear.

    Regarding geothermal, here’s my question to you physics-savvy folks. How much power COULD we extract reliably and sustainably (e.g., how much of the current power consumption load could we replace with geothermal), how many coal plants COULD we replace (and thus how much CO2 could we keep from entering the atmosphere knowing that geothermal produces 1/8th the CO2 of a coal plant), and would staggering power plants regionally be an option?

  24. 26

    Yeah, Andrew is right about the physics. If you’re going to extract 3 TW you’re going to have to do it by cooling the crust, which is not sustainable in the long term. So, let’s rough out how long “long term” is…

    The US has about 10 million (1e7) km^2 of land area. If we’re extracting heat from the top 6km of that, we’re talking about 1.8e18 g of rock (assuming basalt at 3 g/cm^3. Basalt has a specific heat of slightly less than 1 J/g-K, so let’s just say 1.8e18 J/K is available in the form of heat already in the top layer of rock.

    If you’re generating 3 TW of power at 20% efficiency, you need 15 TW of heat, or 15e12 J/s. Divide by our stored heat of 1.8e18 to get about 8e-6 K/s. That’s your rate of cooling. There’s about π-ty million seconds in a year, so multiply by that to get the temperature drop per year: 258 Kelvins.

    So to get 3 TW of usable energy by cooling the rock, you have to cool the entire land area of the US to a depth of 6 km by 250°C per year. Unless I’ve massively fucked up a factor of 1000 somewhere, this can’t possibly work.

  25. 27

    @ Johnny Vector:

    You’ve dropped some orders of magnitude in the conversion from rock volume to weight.

    60 million cubic kilometers is 6 x 10^7 x (1000^3) = 6e16 m^3, which is 6e22 cm^3. At 3 g/cm^3 that gives 1.8e23 g, five orders of magnitude more than your figure.

  26. 28

    Aside from Andrew G.s observations about the limits of geothermal, isn’t the whole thing rather experimental? SFAIK this deep borehole geothermal hasn’t been done yet & we should have a few successful generating stations before deciding it’s so great that we don’t need to even think about energy source X.

    There are some other possible energy sources, like high altitude (100s to 1000s of meters up) wind power, or ocean current power, where I say, yes build some prototypes, but don’t *count* on it working well.

    Meanwhile, I think you are way too dismissive of nuclear.

    Fukushima I leaked radioactivity (without SFAIK harming anyone), but Fukushima II which was built about 10 years later & about 10 km south along the coast didn’t make the news because it didn’t leak. I would like to see an analysis of why the difference, especially if is is because of improvements in the technology over a decade.

    The fuss about the ‘dangers’ of nuclear power always seems strange to me. It seems to involve ignoring the fact that nothing is perfectly safe & when you compare energy sources (eg: ) nuclear looks very good.

    Nuclear power isn’t perfect, it’s just better than the alternatives.

  27. 29

    Let’s try it with the following assumptions (some of these are different from before, because I’m being more conservative):

    1) we wish to replace coal-fired electricity and further contribute to energy supply, so let’s assume a target generating capacity of 1 TW (about 3x current coal-fired capacity)

    2) net efficiency of 5% including all sources of heat loss in addition to generation efficiency, i.e. we take 20J of heat out of the deep end of the bore for every 1J of electricity delivered at the plant output

    3) Sustainability over at least 150 years (if we don’t have fusion or at least orbital solar by then, something’s gone badly wrong) with a net long-term temperature loss of no more than 20K (possibly overall over many production/regeneration cycles)

    So the total heat consumption is:

    1e12 We / 0.05 (efficiency) = 2e13 Wt

    (We = watts of electricity, Wt = watts of heat)

    2e13 Wt x 150 (years) x 3.2e7 (seconds per year) = 1e23 J (total joules of heat removed from the crust)

    Using Johnny Vector’s figures for heat capacity, 1 m^3 of rock has a heat capacity of around 3e6 J/K, or 6e7 J for a 20K loss of temperature. We neglect the replenishment of heat from the core because it is small compared to our extraction rate (as discussed above).

    So to provide our 1e23 J, we need effective access to 1.7e15 m^3 of rock, which is 1.7e6 km^3.

    If a single plant could produce, say, 200 MWe, then we need 5000 plants to produce 1 TW (that’s about 4x the number of existing coal plants, but we’re targetting 3x the capacity). If each has a design life of 50 years, then there will be 15000 plants over 150 years. So each plant represents about 120 km^3 of rock. That’s actually well within the feasible range – it’s easily consistent with the plant dimensions for the model EGS plant given in that paper. (Allowing for the likely volume of rock around the plant proper which will conduct significant heat into the plant area over the design lifetime.)

    So unless I’m way off in my numbers, this makes geothermal look like a reasonable short to medium term interim replacement for coal-fired electricity (conditional on the practical aspects of actually building the plants). It’s not sustainable at that level over timescales of millennia, as discussed before, and therefore isn’t in the same class as solar, or wind, or tidal power all of which can be sustained indefinitely.

  28. 30

    Jim @29: There are two “enhanced” geothermal plants (using fracking to extract heat from dry rock beds) that I know of: Soultz-sous-Forets, France and Landau, Germany. The Landau plant produces 22MW. There’s also a demo plant in Cooper Basin, Australia which can apparently produce 25MW. It’s in its infancy, but certainly not “experimental”.

    In April 2011, 25 people were dead from Fukushima Daiichi, not including the tsunami victims proper. During the evacuation, a group of elderly patients in a nursing home were not evacuated, and a number of those patients died. Your numbers are therefore way off, even if I don’t have current ones for direct fallout death.

    Lying about nuclear safety seems to be common practice. A trend I’ve noticed is that for coal plants, the numbers always include every single outdoor pollution-related death, assuming every one of them is a direct result of coal. More conservative numbers estimate the coal-attributable deaths at one tenth that. And when a coal plant is hit by a natural disaster, you don’t have to evacuate 210,000 residents surrounding the plant like they did when Fukushima melted down.

    I’m not saying nuclear power is evil or anything, just that there’s a hardcore nuclear apologist movement on the intertubes along the same lines as the Ron Paul supporters — they search for anything remotely anti-nuke and muddy the waters.

  29. 31

    JohnnyVector: I’ve seen estimates of 3-5 years depending on size and location to build a geothermal plant.

    On the question of nuclear safety, see this post for the current safety concerns from the Fukushima meltdown. There are a lot of news items here. Read them all.

  30. 32

    Andrew G. Thanks! Cubed, squared, huh. Makes a difference. So the correct numbers make it look viable from a thermodynamic standpoint. Although a 20 K temperature drop is more than I’d be willing to allow without checking with some geologists. Let’s see if I can at least keep the units straight this time…

    The Encyclopedia Brittanica online says basalt has a CTE of about 5e-6 per Kelvin, so 20 K would shrink the rock by 1e-4. For your 120 km^3 per station, if we’re pulling heat out of a slab 3 km in height, that would be 40 km^2, or a disk 7 km in diameter. Multiply by 1e-4 to get 70 cm of contraction. Maybe 70 cm over that distance is not a problem, if the rock just bends.

    Well, let’s see, the San Andreas fault accumulates something like 3.5 cm/year of motion, so 70 cm in 150 years is much less than that. And it’s limited in extent, so the energy released if something does slip is not going to be anything like an earthquake on the San Andreas. So, probably okay, but there are smarter people who could answer that better.

    So the prime question for me is now how long would it take to deploy a useful amount of it. And I agree the waste heat should be used for heating, if possible. That could be difficult with centralized deep holes, since transporting heat is much harder than transporting electricity or fuel.

Comments are closed.