Burning Buried Sunshine (2003) | Hacker News Reader

Burning Buried Sunshine (2003)

plus.maths.org

59 pointsdredmorbius23 comments

> to produce one litre of petrol it takes 1.29 kg of oil, of which 85% (1.1 kg) is carbon. And as only 1/10,750 of the carbon remains from the plants that were buried millions of years ago, our one litre of petrol is the result of 1.1 x 10,750 = 11,825 kg of carbon from ancient plants. Finally, as plants are approximately half carbon, that means that 23.65 tonnes of plants were required to make just one litre of the petrol available at your local station

Wow, and this is a lower bound because it doesn't even consider manufacturing and distribution.

My cognitive dissonance at the gas station is approaching an all-time high.

Maybe a pedantic point: I think there is a difference between saying 24 tonnes of plants were "required" and saying 1 L represents 24 tonnes of plants. The same way 1 kg of iron represents some, I imagine incomprehensible amount of fusion. But if you want 1 kg of iron, you don't need to go out and expend that energy, any more than you need to go out and expend 24 tonnes if plants. Both are basically finite, with geological or astrophysical origins, rather than something we can make again on human timescales

The difference being that iron is recyclable and can be re-smelted (if oxidized) without needing to be fused in stars whereas with fossil fuels we aren’t so lucky. (However, synthesizing hydrocarbons IS feasible.)

Recycling carbon is easy, we just use plants.

Soybeans are a common way of capturing carbon from the atmosphere, using solar energy, to produce diesel.

Yeah, its almost as annoying as when people talk about "using" water. The cycle time is 9 days, in 9 days it comes back down as rain, very little water ever gets "used".

Because they typically omit "easily available". There's lot's of water in the ocean, why do people die from lack of water in the middle of big body of water?

Honestly, because they don’t drink it soon enough / slow enough. There’s a pretty cool study on how you can drink seawater and remain hydrated if you do it right.

I meant freshwater cycle time.

Yes there are certainly places that run out of water, I meant more places like Vancouver where the city just doesn’t build a bigger reservoir and instead bemoans an annual water shortage.

Economic and market discussions do a very poor job of distinguishing stocks and flows.

At the risk of a major digression, one sees this constantly in discussions of inequality where net personal wealth or market capitalisation (both stocks) are compared to income or GDP (both flows).

(And at the risk of digressing my own digression, economists aren't even settled on whether wealth is a stock or a flow. Adam Smith's definition is the latter.)

With water, there is is a throughput in a given region. That flow is not uniformly available, and varies with time (further complications). It serves processes and consumers (in both economic and ecological senses), and changes to those flows and availability do have major impacts (whether surpluses or shortages).

There are times when the critical need is getting rid of excess water (via dikes, pumps, and other control structures), others when the critical need is obtaining or preserving water (again: via pumps, reservoirs, cisterns, etc.). The total demand, flow, and impact are high enough that even modest deviations from normal experience or expectations are often devastating --- floods, droughts, and the like.

Transporting water more than a few km / miles is tremendously expensive and typically requires major public infrastructure. It's best moved under gravity flow alone.

If the litre of water you have used comes from an excess, there is no issue. If it won't be replaced for a year, it's the last litre available, and the nearest additional litre is 100km away ... you may be having a problem.

That hasn’t really been my experience with water.

As far as I know it costs about the same as any other good to transport in small volumes and is incredibly cheap to transport in larger volumes, on par with oil.

I do tend to agree with you tho that if you want government to transport it, it gets ridiculous as one would expect from tasking government to do something nature does for free, and the private sector does extremely efficiently.

And how much water is needed?

It's not the unit volume cost, it's the total cost.

The United States consumes about 20 million barrels of oil per day. That's about 840 million gallons.

Water consumption is 322 billion gallons/day, 380 times more. That's about a thousand gallons per person per day.

Note that this isn't individual direct consumption (drinking, cleaning, bathing), but the net total use. If you're looking at replacing all city consumption (residential, commercial, industrial) plus agrigulture, you'd be moving these volumes of water. Even at pennies per gallon, the costs add up.

Keep in mind that the watershed for major Western US cities already extends over hundreds to thousands of miles. A chief reason that the Western US looks far more like an island archipelligo than the richly-settled East is because of water scarcity and prior claims.

https://upload.wikimedia.org/wikipedia/en/4/48/Earthlights_d...

San Francisco (and much of the Penninsula) draws water from the Hetch Hetchy Reservoir, 180 miles east. Los Angeles draws much of its water from the California Aqueduct, streching over 500 miles north, and the Colorado River, whose watershed extends to the eastern front of the Rocky Mountains, 1,500 miles to the east. Both delivery systems operate virtually entirely under gravity flow.

The California aqueduct has exceptions --- there are pumping stations in the Delta which provide a few metres of lift, and the Edmunston Pumping Plant which lifts water over the Tehachapi mountains utilises 14 * 80,000 horsepower pumps (835 combined MW) to provide 600m (1,926 feet) of lift to 2 million gallons/minute (7.5 million litres/minute), 7,500 tonnes of water per minute --- roughly the capacity of the Oroville Dam spillway at peak capacity, for comparison. It has its own dedicated 1 GW capacity generating plant.

Note that when the Oroville reservoir is discharging water it is generating electricity. What you gain in shipping Oroville water you lose shipping Tehachapi water. The opportunity cost is not 835 MW, but 1,670 MW. At retail California rates, that's nearly $200,000 per hour of operation, $4.6 million per day, $1.7 billion/year, to move water 750 km across land and 600 m high ... and that's only considering the Edmunston plant itself (there are numerous other pump stations in the system as a whole).

The California Department of Water Resources annual budget is about a billon dollars (there's considerable variation).

So long as you want to use / access water where it's flowing already, costs are low. Major river systems (the Mississippi-Missouri-Ohio-Red river system in the US is among the largest in the world) provide just that.

But when you start wanting to move water laterally, or worse, through or over mountains, costs increase tremendously.

________________________________

References:

US petroleum use: https://www.eia.gov/todayinenergy/detail.php?id=44556

US water use: https://www.usgs.gov/mission-areas/water-resources/science/w...

Edmunston Plant: https://en.wikipedia.org/wiki/Edmonston_Pumping_Plant

Oroville Dam: https://en.wikipedia.org/wiki/Oroville_Dam_crisis

California Aqueduct: https://en.wikipedia.org/wiki/California_Aqueduct

CA DWR budget: https://lao.ca.gov/Publications/Report/4103

I very strongly recommend reading the underlying article itself. I've submitted that numerous times to HN, unfortunately PDFs seem not to have much uptake, in part because many readers arrive via mobile devices.

https://www-legacy.dge.carnegiescience.edu/DGE/Dukes/Dukes_C...

I'd stumbled across this through my own research, trying to find literature with a detailed description of fossil-fuel formation. I've since found it referenced by numerous other works --- yes, one of my forms of entertainment is looking for familiar citations amongst footnotes. Smil's Energy and Civilization being one work citing Dukes.

It's not just the material conversion that's staggering, there's the time of accumulation. For petroleum, humans burn in one year a quantity which took five million years to accumulate.

> 23.65 tonnes of plants were required to make just one litre of the petrol available at your local station

The correct number would be closer to 3.54 tonnes of plant material. The other 18.91 tonnes aren't spent. They are instead fully recycled and re-enter in the system (or go away out of the oil processing chain).

A pedantic point, but hydrocarbons don’t actually “store energy”. The energy from the combustion actually comes from the oxygen! We just don’t think of it as the “fuel” since it’s so abundant.

A better intuition is this: plants, etc. used energy to create two products, oxygen (which went into the atmosphere) and hydrocarbons (buried in the ground). When we recombine them and add heat, we reconfigure some bonds and release energy in the oxygen bond. The hydrocarbons act as a “sponge” for the oxygen atoms, and the denser the sponge the more energy we can release.

If it was the oxygen buried underground we would probably call it “fuel”!

Whether you consider the energy stored in the oxygen or the hydrocarbon doesn’t particularly matter. The energy is in them being split apart.

I read a proposal to send an aircraft to Titan and bring along oxygen as the energy source to burn with the methane in the atmosphere. (It works… sort of. In a methane and oxygen stoichiometric reaction, the oxygen is 4/5ths the mass so it’s not nearly as good of a deal as it is on Earth.)

Since Titan’s surface is mostly water ice you’d think it would be easier to use electrolysis to generate oxygen locally.

> Whether you consider the energy stored in the oxygen or the hydrocarbon doesn’t particularly matter.

It truly doesn't matter which one you consider it to be, given that both are technically incorrect.

> The energy is in them being split apart.

This is a common misconception (likely) perpetuated by highschool chemistry textbooks. The energy is actually release by the combination of oxygen with hydrogen and carbon. Let me explain.

It takes energy to break chemical bonds. Some bonds take more energy to break, others take less. When a bond is created, an amount of energey (roughly) equal to the energy require to create the bond is released.

Here I will demonstrate the combustion of a simple hydrocarbon:

CH4 + 2 O2 -> CO2 + 2 H2O

When this reaction takes place, there is not energy "stored" in either the Carbon-Hydrogen bonds or the Oxygen-Oxygen bonds that is "given off". Rather, energy is actually required to break the bonds. It goes something like this:

CH4 + 2 O2 + (a little bit of energy) -> CO2 + 2 H2O + (a lot of energy).

Making hydrocarbons (or hydrogen from water) is an energy intensive process because breaking the bonds in CO2 and H2O requires a lot of energy.

Yes, I didn't say the energy is stored in the bonds or is released by breaking bonds. I was referring to the fact that photosynthesis (or electrolysis) is used to separate the oxygen from the carbon/hydrogen in H2O or CO2. That's where the energy is stored (in the separation of the fuel from the oxygen).

Yeah this is correct and I was oversimplifying. Though that begs the question of how that equation balances out - where does (a lot of energy) comes from?

The products on the right side sit at a lower energy level (they hold together “tighter”) overall. On the left, O2 needs less energy to break up while CH4 needs more, so in that sense the energy “comes” from O2 being easier to break apart. Conceptually I imagine these as a bunch of magnets with the two “magnets” in O2 held at a larger gap (so easier to take apart). With a jolt, things reconfigure and snap into a tighter pattern, and there’s an overall energy release.

While it's basically certain as can be that anthracite is fossilized plants, on account of the leaf imprints, locations, and so on, the evidence for oil is far less conclusive. At the end of the day though it makes no difference. Whether or not the Earth's oil reserves are buried sunshine, primordial hydrocarbons with biological contamination, or both doesn't make much difference. More isn't getting made on human timescales.

False.

The paper on which this article is based directly addresses that:

Marine sediments produced the majority (about 86%) of the world’s petroleum, while deltaic and lacustrine sediments yielded smaller amounts (about 11% and 3%, respectively) (Demaison, 1993). The organic matter in carbon-rich marine sediments is primarily composed of the remains of phytoplankton. The percentage of annually fixed organic carbon that accumulates in sediment (the PF) is difficult to estimate because it can vary over several orders of magnitude, depending on the length of time that organic matter experiences oxic conditions, and on other factors (Bralower and Thierstein, 1987; Bordenave, 1993; Canfield, 1994; Gélinas et al., 2001). Two types of marine settings are thought to have been the major environments for the formation of petroleum source rocks: silled basins with oxygen-deficient waters, and highly productive upwelling zones (which generated 78% and 8% of world petroleum stocks, respectively; Demaison, 1993).

https://www-legacy.dge.carnegiescience.edu/DGE/Dukes/Dukes_C...

> Finally, as plants are approximately half carbon, that means that 23.65 tonnes of plants were required to make just one litre of the petrol available at your local station.

Surely plants are majority water by mass? A quick google claims 89-90% water[1]

1: https://www.toppr.com/ask/question/the-percentage-of-water-i...

This was my first thought, too. I wonder if the author considered this?

Cellulose is almost 50% carbon by mass, so perhaps the dry weight is half carbon? That would mean you need to derate the authors numbers by about 75%; it's still a large input for a small output though.

Reply on news.ycombinator.com