Hacker News Reader

Comment by chris_va | Hacker News Reader

I'd be careful about getting too optimistic.

Conventional geothermal is about 4-5 cents/kWh, on par with natural gas in the US. The dominant capital cost is drilling a large well (you need volume), and so geothermal plants are generally only built in areas that require shallow (1km) wells.

Well costs are ~quadratic in depth. Given how much money has already been spent optimizing drilling for the oil&gas industry, along with how cutthroat that market is, I don't see the cost coming down significantly. As a result, deep geothermal will likely be limited to niche regions like Iceland. And you need deep geothermal to scale it past the existing locations.

I would love to be wrong, since geothermal checks all the boxes for renewables and is also suitable for base load power, but I don't see an obvious path forward short of a drilling tech miracle.

(source: Climate and Energy R&D group)

View on HN

Don't bill deep geothermal as any sort of green. There are usually emissions that have health consequences for the population around the power plant. The corrosion of the pipes bringing steam from the drill site to the turbines is massive. Drilling that deep and pumping water into the hole ups the risk of earthquakes.

(source: I live in Reykjavík and have witnessed first-hand the increased stench from the Hellisheiði geothermal plant)

It is, however, extremely low carbon. Wind turbines kill birds and make noise, solar panels need toxic manufacturing processes. But carbon emissions are the overriding problem.

"Wind turbines kill birds"

15,000x fewer than house cats do though.

Cats kill exactly zero eagles per year.

The bird thing is not as big of a deal as some seem to think. Wind turbines kill a couple hundred thousand birds each year [1]. In contrast, cats kill hundreds of millions of birds each year [2]. People still like cats. I also don't think the noise concern is a reason to discount wind turbines.

Of course, dirty alternatives have their own toll on wildlife.

1. http://www.smithsonianmag.com/smart-news/how-many-birds-do-w... 2. https://www.sciencenews.org/article/cats-kill-more-one-billi...

Also, chickens are birds. Not wild birds, but birds. If the concern for birds is some kind of animal rights perspective, chickens are up into the billions per year. If people want to prevent needless bird deaths, that seems like the place to start.

I think the main issue for most people is biodiversity and ecological impact rather than sentimentality. Very few wild animals pass away quietly in their sleep.

FYI, for deep wells you generally have to switch to supercritical CO2 instead of water, for the corrosion issue.

I don't disagree with you—I've been to an 800 MW geothermal facility in northern California and was taken aback by the giant heaps of sulphur everywhere—but other renewables come with their own downsides.

- contemporary photovoltaic panels require all sorts of strange metals to make, which have environmental impacts via mining. This might change with new materials. Solar also tends to have quite high carbon dioxide impacts due to the manufacturing process. These are obviously way less than gas or coal, but about double compared to other renewables and nuclear.

- hydroelectric is notorious for destroying huge ecosystems stretching far above and far below the dam and its reservoir. Hydro is so bad that California doesn't let large projects count in the Renewable Portfolio Standard.

It's tough; there's no perfect energy source.

...contemporary photovoltaic panels require all sorts of strange metals to make...

This is not quite right. A few kinds of contemporary PV require exotic elements such as indium, gallium, and tellurium. But the market is dominated by crystalline silicon cell technology (about 93% market share in 2015): https://www.ise.fraunhofer.de/content/dam/ise/de/documents/p...

"contemporary photovoltaic panels require all sorts of strange metals to make"

Certain kinds do (e.g. CdTe and CIGS), but not the kind that you'd put on your house. If modern solar panels required significant amounts of rare earths they wouldn't be as cheap as they are.

"Solar also tends to have quite high carbon dioxide impacts due to the manufacturing process."

About 30g/kwh of CO2 vs. coal's ~900g/kwh, and a lot of that is simply because it is usually currently made without green power (i.e. it will go down).

"It's tough; there's no perfect energy source."

There's a lot of half truths told about really good energy sources.

Yeah, coal is obviously WAY worse than solar vis a vis carbon dioxide. I'm not disputing that. But compared to other renewables, it doesn't do as well as you'd expect:

- wind: 11 g / kWh [1]

- nuclear: 12 g / kWh [2]

- solar PV: 45 g / kWh [3]

- gas: 450 g / kWh [4]

Of course, 12 g vs 45 g is NOTHING compared to 45 g vs 450 g (and coal is another 2x atop gas!), so I'm making a bit of a mountain out of a molehill.

[1] http://www.nrel.gov/analysis/sustain_lca_wind.html

[2] http://www.nrel.gov/analysis/sustain_lca_nuclear.html

[3] http://www.nrel.gov/analysis/sustain_lca_pv.html

[4] http://www.nrel.gov/analysis/sustain_lca_ngas.html

Gas actually contributes more to global warming even than coal does because of the methane (~100x as powerful a warming gas as CO2) also released during extraction:

http://inhabitat.com/updated-cornell-study-shows-fracking-ca...

On the other hand, methane breaks down fast. It's only an immediate problem; if we didn't release more CO2 than could be absorbed, then we'd have a fairly decent methane budget as well.

As it stands, of course, it makes everything worse.

> Hydro is so bad that California doesn't let large projects count in the Renewable

10 cubic kilometers of water: It's the worldwide-recognized ratio under which hydro is considered renewable, and above which the ecosystem destruction is too important.

DISCLAIMER: I'm not an expert in this and I can be very off concerning this area.

Having said this, I think the risk of man-made earthquakes is exaggerated. I mean, I think it might be possible to trigger small seismic events, but they should not be the destructive kind. I doubt we can trigger massive destructive eathquakes like this. Also, from what I understand, smaller seismic evnts should lower the energy that the bigger ones have.

Tell that to the people of Selfoss and Hveragerði which have now become accustomed to smaller quakes on a more regular basis than usually in the past.

Is it damaging though? That is the question I want answered. I mean, even in my post, I did not question the presence of more frequent small earthquakes, I questioned if these earthquakes are any sort of real issue that produces damages or simply unfounded panic.

Not being able to trust that the ground you stand on won't shake violently? Are you being serious?

I am being totally serious. Why are earthquakes bad? Because they destroy goods, buildings and kill people. If eathquakes do not destroy goods, buildings and kill people are they that terrible anymore or simply an inconvenience at most?

It might sound alien if you have a certain way of looking at things, but really, is my question not logical?

Spoken from the privileged position of not having to worry about it like it is of little concern to those who do. I'm leaving it as an exvercise for you to think of analogies to other aggrieved groups that would not take kindly to being told their concerns are invalid.

Yeah it would seem like frequent small earthquakes are way better than rare devastating ones, so it would even be better to provoke them rather than wait like in SF and Vancouver... If only, it helps rehearsing the response, convincing taxpayers to purchase tge equipment and finding early which buildings are dangerous.

Has the lifetime of deep GeoThermal wells been studied at all? Unlike oil/natural gas drilling, a deep geothermal well should provide energy indefinitely barring collapse or obstruction of the well. I'd venture the cost of shutdown and clearing obstructions is fairly low relative to the initial CapEx of the well.

Yes, most wells are productive for about 30 years, and then you have to wait a bit (another 30) to let the rock heat back up. The NPV past 60 years is pretty tiny, so it is hard to amortize capex that far out.

Worst case, you run them continuously at half capacity, but then your cost per kWh is about 2x.

>Yes, most wells are productive for about 30 years

Wait, I haven't looked at geothermal in a long time so I'm not at all up to speed on the current state of the at. But previously you wrote:

>and so geothermal plants are generally only built in areas that require shallow (1km) wells

That sounds like your "30 years" number there is for shallow wells. Intuitively it doesn't seem like this logic would necessarily be identical for deeper wells: obviously at some depth the rock is always extremely hot (or, in fact, molten). The article seems to indicate that they expect a deeper well to produce higher temperatures and pressures, in turn allowing more energy production, and that seems like it'd make sense. Maybe 1km vs 5km makes no difference in lifetime beyond that, but does your analysis include deep wells already in an equivalent area? Otherwise it doesn't seem reasonable to extrapolate past lifetimes of shallow wells to future lifetimes of deep wells, even ignoring that they say they'll be able to achieve far higher energy generation per well.

> obviously at some depth the rock is always extremely hot

Not if you keep pumping in water to run a steam turbine.

I assume that's how geothermal energy works, and I'm not going to double check.

>Not if you keep pumping in water to run a steam turbine.

Are you being serious? In terms of known capabilities the Kola borehole shows that even with older tech we can drill down to over 12km in depth, and there have been decades [1,2,3] of efforts drill all the way through the Mohorovičić discontinuity into the mantle (20+km minimum from a continental starting point, 5-10+km from ocean). Lithosphere/mantle boundary tends to be defined by temperature, in that around 1800°F is when the weakest common mineral (olivine) begins to exhibit viscosity. Well before that it'd reach the point where the amount of pressurized water we could pump through it wouldn't be able to keep up with new thermal load. So yeah, at some depth the rock is always extremely hot.

In terms of economics though of course there is some set of balance points for a given level of tech and economics between cost of drilling to a given depth, gains in lifetime/power density, maintenance costs, costs/losses of actually moving working fluid from heat point to generator, reliability advantages, etc etc. Presumably experts go through and consider all of this, it should be tractable within some set of risk bars. If an experienced power company is actually putting down millions on doing a project after evaluation, I see no reason to doubt that they've done the math to their/their investors' satisfaction to a level sufficient for a pilot project.

1. http://www.smithsonianmag.com/science-nature/decades-long-qu...

2. https://en.wikipedia.org/wiki/Chiky%C5%AB

3. https://www.extremetech.com/extreme/247370-drill-baby-drill-...

I'm just making the simple point that however hot the rock is, it will cool down when you pour enough water on it.

I don't know what the Mohorovičić discontinuity is, but I'm sure it doesn't change this physics fact.

The rock will radiate heat sufficient enough to boil the water before it can even come close to coming in contact, so no, basic thermodynamics would like a word with your physics fact....

And the water, in boiling, will remove some of that heat that would otherwise radiate back. The entire premise of geothermal energy is removing some heat from deep down in the earth. Whether that amount is big enough to cause a noticeable difference is was is being debated.

Yes, you have a point, though the reality is even more complicated.

There are physical limits to the heat capacity of the rock, and the heat flux into the region. Deeper wells are not necessarily hotter. Going hotter adds other costs (drilling mud, water=>co2, casement, etc), so its a big multivariate optimization.

Dumb question and I'm not trying to be pendantic: if the whole world got their power from geothermal, which must (I assume) extract heat and thus cool the core, at what point is there some effect (on magnetic poles? rotation of earth? tides??)?

I wonder similar things about solar - what effect is there on global weather patterns when enough solar energy to make a difference is extracted from the environment? Like, what would happen globally if the entire state of Arizona was in the shadow of solar arrays?

Think of it this way... the Earth is a rock ~4K miles in radius. Essentially all of it is molten rock. When people try hard we can melt rocks to make metals. But we need special-built furnaces, machines, etc. And these furnaces are maybe 50 feet across.

So the Earth has a heat stored which is (4K miles / 50 feet)^3 times larger than a blast furnace.

i.e. about 7x10^16 times larger.

If everyone on Earth had their own personal blast furnace, we'd be controlling 10 million times less energy than what's in the core of the Earth.

Similar calculations apply to other calculations. A Physics 101 calculation I did (years ago) was how much a tidal plant would slow down the rotation of the Earth. The answer was by less than a second a day, after 100 years of running.

> ke, what would happen globally if the entire state of Arizona was in the shadow of solar arrays?

The plants wouldn't grow. :)

But seriously, the energy has to go somewhere. If it's hitting solar panels instead of the ground, the net effect to the world is essentially zero.

Maybe locally the ground would be cooler because it's in the shade. But that's about it.

Less than a second a day in 100 years is a huge change in earth scale. I'm curious, what kind of tidal plant would even come close? Is it assuming we extract 100% energy from the tidal waves and entirely stop the tides, somehow?

I'm fairly certain stopping the tides would be devastating to the environment since several ecosystems (tidal plains, marshes, riparian zones, etc.) depend upon them.

Doesn't it depend on how much less than a second it is?

You are thinking in linear terms. Humanity's historical energy use definitely doesn't look linear, more like exponential - https://gailtheactuary.files.wordpress.com/2012/03/world-ene...

I imagine that if we were able to tap into deep geothermal energy the continuing exponential growth would eventually pass some sort of geothermal limit point.

This is true, but I know for a fact that US energy use at least has roughly dropped or remained stable for the last decade. Increase in energy use is typically driven by people's increasing incomes and their ability to buy energy for using appliances and such. We can expect energy use to increase significantly globally until everyone is at a base level of income, but after that it should be capped at population growth.

Remember Asimov's city-planet Trantor. It relied in part on geothermal energy (core taps) with huge radiator arrays at the poles.

There's no way that a tidal plant running for 100 years is going to change the day by 1 second. You're many orders of magnitudes off.

That means that if you built 100 of those plants that in 1 year you'd change the length of the day by 1 second.

The 2011 Tōhoku earthquake and tsunami released 3.9×10^22 joules of energy[1] and changed the day by just over a microsecond[2]. A single Tsar Bomba releases 2.1×10^17 joules, so you'd need the energy of around 200 thousand of those to produce the same amount of energy as the earthquake, and then assuming you can direct that energy you need 1000 times more to shift the day by a second instead of a microsecond.

That means that each one of your proposed tidal plants would need to generate the energy of 20,000,000 Tsar Bombas per year.

That's one Tsar Bomba's worth of energy being released every 1.5 seconds at each one of those 100 plants. Buy stock in sunscreen companies now!

Each one of your 100 plants would satisfy all of current yearly world energy production in just 30 minutes[3][4].

I'm too lazy to calculate what that would do to the surrounding area, but I'd wager that they're going to severely depress local property prices due much of the surrounding beach-side property area now being turned into permanent molten lava.

The good news is you could probably build really good wind power plants now, due to constant hurricane winds not seen on Earth since the K–T impact, but I digress.

I didn't check your molten rock calculation but given how off your tidal plant calculation is I wouldn't be surprised if it's way off as well.

1. https://en.wikipedia.org/wiki/2011_T%C5%8Dhoku_earthquake_an...

2. https://www.nasa.gov/topics/earth/features/japanquake/earth2...

3. https://en.wikipedia.org/wiki/World_energy_consumption

4. ((5.67x10^20)/(2.1x10^17))/1.5/60 = 30

> given how off your tidal plant calculation is I wouldn't be surprised if it's way off as well.

he showed his work for his first calculation. Looked reasonable to me.

His numbers look fine, but I think the assumptions are false.

    > And these furnaces are maybe 50 feet across. [...]
    > So the Earth has a heat stored which is
    > (4K miles / 50 feet)^3

This seems to assume that the earth is also cube shaped like a hypothetical 50ft^3 blast furnace, instead of a sphere (or oblate spheroid before anyone out-pedants me).

It also seems to assume that a 50ft^3 blast furnace contains energy equivalent to 50ft^3 of molten mantle. If you look at diagrams of blast furnaces you can see that a relatively tiny part of them is actually molten material.

Not being able to find numbers on the stored joules of industrial blast furnaces, I'm just going to just state that 50ft^3 of mantle has a crapload more potential energy than 50ft^3 of blast furnace, most of which is air, equipment & empty space, not molten material.

He is right that there's at least as much energy stored in the earth as he's saying, but it's going to be a crapload more than that.

The insides of blast furnaces are also significantly colder than the average temperature of the mantle.

The point is that these are "back of the envelope" calculations.

Fermi was known for asking questions like "how many gold balls git into a suitcase". The point isn't to get a perfect answer. The point is to get a reasonable answer.

In this case, the point is to answer the OP question in laymans terms. Using not much more than reasonable assumptions, and grade 5 math.

Or, you can bike shed over details.

> and then assuming you can direct that energy you need 1000 times more to shift the day by a second instead of a microsecond.

Still off by a factor of 1000, a microsecond is one millionth of a second.

Yes you're right. I mixed up micro & milli there. So:

* [...] 20,000,000,000 Tsar Bombas per year

* One Tsar Bomba going off every 1.5ms

* World energy production satisfied in 1.8 seconds

We're also going to need more sunscreen.

> There's no way that a tidal plant running for 100 years is going to change the day by 1 second. You're many orders of magnitudes off.

It was a recollection of a problem from 25 years ago for crying out loud.

Thinking about it some more, it's probably slowing down the rotation of the earth by 1 second over the course of 100 years. Not 1 second per day. Which makes about 4-5 orders of magnitude difference.

> I didn't check your molten rock calculation but given how off your tidal plant calculation is I wouldn't be surprised if it's way off as well.

The point was to get a "back of the envelope" calculation. Not to get a precise answer to 4 decimal places.

What's with the bike shedding? Is the point here to have a discussion, or to prove that you're smarter than everyone else?

The scale at which humans draw energy from the core is negligible compared to the amount. Think about it, earth has been cooling down through radiating from the surface for billions of years. Given the ratio of the geothermal station and the surface area of earth, I don't think there is anything to worry about.

As for solar, the incoming energy would have been mostly converted to heat. When we use the electricity produced, we convert it back to heat again. If the entire state of AZ were covered with solar panels, and all the energy spent in NY, there would be some localized changes but not much in the global scale. For the foreseeable future, we are not even close to covering up just 1% of the land area of AZ.

Earth's radiation scales as a cube of the radius. Our energy usage is growing exponentially. You can apply big-O analysis to this problem ;)

> Our energy usage is growing exponentially.

Well, with efficiency gains in the latest years energy consumption (not only electricity) actually is constant or decreasing in developed countries. Add to that decreasing population fertility and if you expect that developing countries are going to follow the same trends after a certain threshold then that's not a good assumption long-term.

Japan: https://en.wikipedia.org/wiki/Energy_in_Japan#/media/File:Ja...

US: https://en.wikipedia.org/wiki/Energy_in_the_United_States#/m...

Europe: http://ec.europa.eu/eurostat/statistics-explained/images/thu...

Like the others said, the earth core is HUGE.

But I am a bit sceptical to the no impact scenario. We know a lot allready about the inside of the earth, but afaik we do not really understand it. There are many flows underneath etc. and small changes can have big impacts. Look for example at the amount of artifical CO2 from humans. It is tiny in comparison with the rest of the atmosphere. And yet it gives us a lot of troubles.

And about geothermal: well, here in germany there have been allready some minor earthquakes, probably caused by geothermal plants ...

(German) http://m.focus.de/wissen/klima/tid-14230/energie-mythen-myth...

I was wondering if anyone else had made the connection to claims (or maybe that's too weak of a word) that fracking causes earthquakes.

Well according to a professor in canada, fracking definitely causes earthquakes ...

https://mobile.nytimes.com/2016/11/18/science/fracking-earth...

Forage indeed can make earthquakes. An experiment was run in Alsace (France) and as far as I know it was stopped due to this kind of problems.

Even if you don't understand French, you can look at pictures of this article: http://www.lefigaro.fr/actualite-france/2013/11/19/01016-201...

Another experiment was stopped in Switzerland for the same reason: https://www.theguardian.com/world/2009/dec/15/swiss-geotherm...

Its a fun question on a Friday, with two possible avenues; "How much heat drain would it take to freeze the Earth's core?", and "What would freezing the earth's core do to life on the planet?"

For me, the more interesting one, which was the 'bad guy' in the forgettable movie "The Core". The earth's magnetic field is created by the motion of the fluid iron (which is conductive) surrounding the solid core[1]. That it is moving and carries charge creates a dynamic dynamo which determines the strength of the magnetic field. Changes in the fluid are why the magnetic poles move around and sometimes change places.

The number 1 feature of the magnetic field (for me at least) is that it deflects the solar wind (those charged particles racing out from the Sun) which would otherwise collide with the gases surrounding our planet and accelerate them outward. Eventually stripping the planet of most of its atmosphere.

One of the less well known ways the solar system could wipe out humanity.

[1] http://www.physics.org/article-questions.asp?id=64

>Well costs are ~quadratic in depth.

Anyone know as to why this is the case? At first pass it seems like your cutting tools should wear out in proportion to depth. But I suppose as things get warmer tool wear becomes faster. What types of tool materials are used for rock boring? Diamond? Tungsten carbide? Is is more of a "conventional machining" process, or an abrasive / water jet? Or explosive like in hard rock mining? What else has an effect on cost for deep holes? Shaft length gets longer, so the twist becomes more of a problem if your are driving rotating bits from the top? Does it have to do with removing the material from the bottom of the hole to the top? Others?

"Anyone know as to why this is the case?"

Think of a telescopic antenna. That structure is similar to what is done to bore a well. You start with large casing diameter and add progressively smaller casings until you reach the target depth. The deeper you go the bigger the casing you must start with. Offshore deep water stuff can be a meter in diameter with 10cm solid steel walls; monstrously heavy and costly materials. And you use kilometers of the stuff.

There are a lot of reasons why costs scale as they do with depth; casing is just one of the more obvious.

Anyone recommend a good pop-sci book on deep hole boring? Seems like it could be an interesting light engineering read.

For deep wells would the need for casing change depending on the material you're drilling into? once your into hard granite bedrock I'd imagine wells of a certain diameter are stable with minimal to no casing.

Yes, geology is a big factor in engineering a well. Pressure is another. Every well is engineered to the prevailing conditions and requirements. Sometimes it is possible to elide heavy casing which can save large amounts of money.

> deep geothermal will likely be limited to niche regions like Iceland

True, but aren't there plenty of such regions around the world? In the US I can think of at least Hawaii and Yellowstone.

Not the greatest population centers, but power lines have a good range. Also, Iceland is making good use of its comparative advantage by doing power intensive production like aluminum smelting and bitcoin mining.

What about the west coast volcanoes in North America?

https://en.wikipedia.org/wiki/List_of_Cascade_volcanoes

...how deep do you have to go to get to the 400C temperature level on some of those volcanoes??

DuckDuckGo for Cascades Fumaroles. You can hike to vents emitting steam, many through glaciers. http://www.ivm-fund.org/mount-baker-volcano-fumaroles/

Also many hot springs could be potential GeoThermal candidate areas http://www.hotspringsenthusiast.com/USsprings.asp

The world's largest geothermal energy field is in California, 1½ hour north of San Francisco: https://en.wikipedia.org/wiki/The_Geysers

Imagine the political/public outrage around building an alternative energy source in Yellowstone National Park. People on both sides of the aisle would be losing their minds.

That said, geothermal activity isn't confined to the park borders, and there may be a suitable spot outside of the park. But again, politics. Building a geothermal plant in the middle of natural gas country would be a challenge.

Small note about Iceland- My web host (1984) is located there, and the servers run on 100% geothermal. That, in combination with the stellar privacy laws, make Iceland a nice fit for me.

The plants are relatively humble buildings that just emit water vapor, so they're minor eyesores more than a real danger to the environment.

That said, I agree that it would take a lot to get anything like that built.

It can get pretty crazy the resources consumed by a drill rig during the average week drilling. New bits, roughly 1200 or so liters of fuel every 24 hours, all kinds of sacrificial bits of tube if they are core drilling their holes. Any hundreds of liters of drilling fluids depending on down hole conditions. I was blown away by the consumption when I saw it first hand. The fuel and drilling bits were the craziest. Then all of the maintenance consumables that you go through as a result of running 24/7.

I shudder to consider the quantities of hardware consumed to drill 4km+ you would get all kinds of crazy hard rock down there, not to mention the time it would take for rod pulls.

> geothermal checks all the boxes for renewables

Article:

"Although geothermal energy is still preferable to gas, coal and oil, it's not 'completely renewable and without problems... As soon as you start drilling you have issues to it, such as sulphur pollution and CO2 emission...'"

A bit late to the party but i wanted to back this up by linking to the relevant chapter from the fantastic book by David MacKay: http://withouthotair.com/c16/page_96.shtml

The chapter is only 4 pages long and he quickly works out that if done sustainably, even blanketing the entire country with geothermal plants could provide at most 2 kWh/day per person in the UK. Current consumption is about 125 kWh/day per person. So even without the capex argument, it is just not a very efficient source of energy long-term.

Your statement should be qualified with "for the UK region".

> Other places in the world have more promising hot dry rocks, so if you want to know the geothermal answers for other countries, be sure to ask a local. But sadly for Britain, geothermal will only ever play a tiny part.

The additional advantage of Geotermal from an operations standpoint it has 0 marginal cost which guarantees it will be producing on most economically dispatched power grids.

Natural gas use will need to stop because of co2. IOW that "on par" cost of NG unsustainably excludes externalities.

I think the real approach to geo should be more focused on decentralized use, instead of trying to do big arrays of holes, like solar, it's in the combination with localization of other technologies in array themselves that seem to offer the most opportunity.

Wind, solar, geo... they all work together better than as a single unit.

That is sort of true from a grid perspective, but the math doesn't really work out well for geothermal. Geothermal has high capex, low opex, so you need it running 24/7 to amortize the high capital costs.

If you combine it with renewables, you get high fluctuation of demand from the grid, which selects for the exact opposite technology (higher opex, lower capex). Natural gas fits the bill, so that is where the market is going.

Reply on news.ycombinator.com