[entropyfails]

I haven’t seen any discussion around here about the new meta-model report on ethanol EROEI put out at Berkley by the name of “ERG Biofuel Analysis Meta-Model.” You can read about the model here http://rael.berkeley.edu/EBAMM/. NPR even did a segment with the author’s of the study so I figure we should take a look at it.

This study takes on Pimentel, Patzek, Shapouri, Graboski, de Oliveira, and Wang. For the low EROEI studies, they added a result from Graboski named “coproduct credits” measured in (MJ/L). Though these coproducts actually come as outputs of the ethanol process they argue that since they displace other energy sources that we should model them as inputs into the systems for the purposes of calculating EROEI. Using this displacement model does seem reasonable, though the values range anywhere from 1.86 as reported by Pimentel to Shapouri’s fantastic 7.97. As a default, if a model didn’t include coproducts then they used the Graboski’s 4.13 value, probably because of its proximity to the mean of the aforementioned range.

Since a lot of people give credit to the Patzek and Pimentel numbers on this site, I’ll report how those fared in this report. Patzek gets blasted pretty hard at UNDERSTATING the amount of energy inputs needed! However, because he doesn’t consider the coproducts, they feel he overstates the energy loss of corn ethanol. His reported NEV (MJ/L) of -5 becomes -1.6 when the “coproducts” gets added back in.

Pimentel gets accused of overstating the amount of energy needed to manufacture Nitrogen and of not adding his 1.86 “coproducts” factor back into the NEV calculations. Reducing the Nitrogen costs and adding back in the coproduct amount turns the -6.1 NEV into a -3.7 one.

So if you feel that Patzek and Pimentel have done the best job in their research on this subject, then nothing has changed for you. Both studies, even after correction, still show a negative Net Energy.

Now how did the “high” NEV studies fare? Shapouri’s NEV of 8.9 gets dropped to 8.0. Graboski’s 3.9 NEV gets dropped to a 3.1. Wang’s 6.9 gets dropped to 6.1. The only exception to this rule comes from de Oliviera’s study which didn’t consider coproducts as they felt that no good data for coproducts existed at the time of their study. EBAMM faithfully adds the Graboski 4.1 MJ/L factor back in to make the de Olivera number go from 1.6 to 4.8.

To put it simply, they raised the 3 lowest NEV studies and lowered the 3 highest studies. The 3 highest studies made errors in the amount of energy inputs needed. The 3 lowest didn’t include the coproducts as energy outputs of the system. We, the poor readers, don’t end up with any clearer idea of if the coproducts should be 1.9, 4.1, or 7.3 MJ/L. If we use the 7.3 value in all of the studies, then every one of the studies shows a positive Net Energy Value. If we use the 1.9 MJ/L factor found in Pimentel, we get no NEV greater than 4 for any of the studies and half of them at or less than 0.

Also, the studies range from 19 to 27 MJ/L of energy inputs for each L of output even after their corrections. Patzek and Pimentel have much higher agricultural energy costs 27,000 MJ/ha and 32,000 MJ/ha respectively. This contrasts greatly with Shapouri’s number of 18,000 MJ/ha. Patzek and Pimentel include labor transportation and Farm Machinery energy costs while Shapouri ignores them. One of these guys has it right, and one has it way wrong.

So wrapping it up, Net Energy Value ends up being in the eye of the beholder. None of the studies agree and all of them overstated their cases. The meta-study makes no real conclusions other than when asked “How much energy do we get from making Corn ethanol?” we can reply “Not much. Maybe none.”

[pup55]

This is the best post I have seen in a long time because in the spreadsheet they give all of the constants and fudge factors they used for the fertilizer inputs, irrigation, yields per acre, etc. so it is really valuable to have around, and in fact, the link should probably be saved on here somewhere for future reference.

The most optimistic one, which was the adjusted Shapouri case, gave a yield of about 3463 liters of fuel per hectare, or about 911 gallons per hectare per year. In light of the usage of 9.5 million barrels per day, times 42 gallons per barrel, etc. etc. to make a long story short, you will need about 1.6 million square km to replace the current gasoline usage in the US. This is about the same as the combined total land area of Texas, California, Montana and New Mexico.

Another way to calculate it is that the entire 9.3 million square km land area of the US is considered about 20% "arable". So, at 1.8 million square km "arable" this means about 80% of the arable land in the country would have to be converted to this switchgrass if the yield figures are correct, and we wanted to replace 100% of our transportation fuel.

Of course, we would not do the whole thing, maybe a goal of 10% would take the edge off of our fuel cravings for a year or two, and we would probably not miss that much space, but the point being that in order for this to make a dent in the current problem, the effort will have to be truly massive, and is going to be really expensive.

There are some other details (water, natural gas availability, investment in refinery capacity, economic issues such as plant size vs. production efficiency vs. collection efficiency etc. etc. that have not been fully explored that may make it completely unworkable on the scale needed to make much of a difference.

But, it looks like a pretty good science fair project for somebody and might be worth trying at a bigger scale to see what happens.

[backstop]

Pup -

There's a further few factors that appear pivotal to regarding Agribusiness Ethanol as a serious large-scale investment prospect.

1/. The declining soil fertility under industrial farming, which can either be offset ( for a while) with increased chem fertilizer inputs, at the cost accelerating the process, or reversed by a return to mixed farming, with crop rotation and manure fertilization, at the cost of substantially lower yields during the recovery period.

2/. The trend of declining rainfall and increasing drought events across much of the grainlands, apparently related to the decline of the hydrocycle through the Amazon rainforest. (The rising GOM sea surface temperatures have also been related to this).

3/. The removal of a significant fraction of US grain yields from food & fodder markets would undoubtedly affect global food prices, meaning not only starvation rising as a consequence, but also US food import costs rising, which of course would hit the poorest hardest, as usual.

regards,

Backstop

[pup55]

Oh, you are quite right, Backstop...

The figures above are of course the really optimistic scenario if everything goes perfectly.

Actual implementation of anything like this would be extremely problematic. Your points are exactly right.

I would add a couple more:

Insect control: Conversion of this much space to a grasslike monoculture crop will cause increase in bug populations (locusts and grasshoppers) which are already developing a resistance to the chemicals we hose them down with.

economics of industrial farming: You can make the argument that because the crop would not be very dense (bales of grass, presumably, that are not free-flowing) the production and transportation of this stuff would be a lot less simple than corn and soybeans, of which the energy-dense portion can be pumped, shipped and barged all over without too much problem. Rail capacity would be a serious question.

Seasonality: How to process that much grass between October and April without having it go bad. How to size the refineries so that the stuff would not sit around in storage and go bad while it was waiting to be refined.

[EnergySpin]

$this->bbcode_second_pass_quote('pup55', 'O')h, you are quite right, Backstop...

The figures above are of course the really optimistic scenario if everything goes perfectly.

Actual implementation of anything like this would be extremely problematic. Your points are exactly right.

I would add a couple more:

Insect control: Conversion of this much space to a grasslike monoculture crop will cause increase in bug populations (locusts and grasshoppers) which are already developing a resistance to the chemicals we hose them down with.

economics of industrial farming: You can make the argument that because the crop would not be very dense (bales of grass, presumably, that are not free-flowing) the production and transportation of this stuff would be a lot less simple than corn and soybeans, of which the energy-dense portion can be pumped, shipped and barged all over without too much problem. Rail capacity would be a serious question.

Seasonality: How to process that much grass between October and April without having it go bad. How to size the refineries so that the stuff would not sit around in storage and go bad while it was waiting to be refined.

[hotsacks]

We have an enormous petroleum lobby.
The corn lobby is no small potatoes either.
Has anyone heard of a switchgrass lobby?
Could be there's a problem there.
[web]http://www.agmrc.org/agmrc/commodity/biomass/switchgrass/[/web]

[entropyfails]

$this->bbcode_second_pass_quote('EnergySpin', 'I')n any case, BFs should NOT be counted upon as a permanent solution; it is a temporal "fix"/bridge over an all electric transportation sector.